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.
82 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
85 JH/02 The smtp transport option "multi_domain" is now expanded.
87 JH/03 The smtp transport now requests PRDR by default, if the server offers
90 JH/04 Certificate name checking on server certificates, when exim is a client,
91 is now done by default. The transport option tls_verify_cert_hostnames
92 can be used to disable this per-host. The build option
93 EXPERIMENTAL_CERTNAMES is withdrawn.
95 JH/05 The value of the tls_verify_certificates smtp transport and main options
96 default to the word "system" to access the system default CA bundle.
97 For GnuTLS, only version 3.0.20 or later.
99 JH/06 Verification of the server certificate for a TLS connection is now tried
100 (but not required) by default. The verification status is now logged by
101 default, for both outbound TLS and client-certificate supplying inbound
104 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
107 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
108 Status Notification (bounce) messages are now MIME format per RFC 3464.
109 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
110 under the control of the dsn_advertise_hosts option, and routers may
111 have a dsn_lasthop option.
113 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
114 default, modifiable by a malware= option. The list separator for
115 the options can now be changed in the usual way. Bug 68.
117 JH/10 The smtp_receive_timeout main option is now expanded before use.
119 JH/11 The incoming_interface log option now also enables logging of the
120 local interface on delivery outgoing connections.
122 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
123 if the interface and destination host and port all match.
125 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
128 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
129 Patch from Andrew Lewis.
131 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
132 now supports optional time-restrictions, weighting, and priority
133 modifiers per server. Patch originally by <rommer@active.by>.
135 JH/16 The spamd_address main option now supports a mixed list of local
136 and remote servers. Remote servers can be IPv6 addresses, and
137 specify a port-range.
139 JH/17 Bug 68: The spamd_address main option now supports an optional
140 timeout value per server.
142 JH/18 Bug 1581: Router and transport options headers_add/remove can
143 now have the list separator specified.
145 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
148 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
151 JH/21 Support for the A6 type of dns record is withdrawn.
153 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
154 rather than the verbs used.
156 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
157 from 255 to 1024 chars.
159 JH/24 Verification callouts now attempt to use TLS by default.
161 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
162 are generic router options now. The defaults didn't change.
164 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
165 Original patch from Alexander Shikoff, worked over by JH.
167 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
168 files if ZCAT_COMMAND is not executable.
170 JH/26 Bug 1539: Add timout/retry options on dnsdb lookups.
172 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
174 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
175 Normally benign, it bites when the pair was led to by a CNAME;
176 modern usage is to not canoicalize the domain to a CNAME target
177 (and we were inconsistent anyway for A-only vs AAAA+A).
179 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
181 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
182 when evaluating $sender_host_dnssec.
184 JH/31 Check the HELO verification lookup for DNSSEC, adding new
185 $sender_helo_dnssec variable.
187 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
189 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
191 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
193 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
194 documented as working, but never had. Support all but $spam_report.
196 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
197 added for tls authenticator.
202 TL/01 When running the test suite, the README says that variables such as
203 no_msglog_check are global and can be placed anywhere in a specific
204 test's script, however it was observed that placement needed to be near
205 the beginning for it to behave that way. Changed the runtest perl
206 script to read through the entire script once to detect and set these
207 variables, reset to the beginning of the script, and then run through
208 the script parsing/test process like normal.
210 TL/02 The BSD's have an arc4random API. One of the functions to induce
211 adding randomness was arc4random_stir(), but it has been removed in
212 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
213 function when detected.
215 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
216 cause callback expansion.
218 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
219 syntax errors in an expansion can be treated as a string instead of
220 logging or causing an error, due to the internal use of bool_lax
221 instead of bool when processing it.
223 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
224 server certificates when making smtp deliveries.
226 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
228 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
230 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
232 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
233 Merged patch from Sebastian Wiedenroth.
235 JH/05 Fix results-pipe from transport process. Several recipients, combined
236 with certificate use, exposed issues where response data items split
237 over buffer boundaries were not parsed properly. This eventually
238 resulted in duplicates being sent. This issue only became common enough
239 to notice due to the introduction of conection certificate information,
240 the item size being so much larger. Found and fixed by Wolfgang Breyha.
242 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
243 size buffer was used, resulting in syntax errors when an expansion
246 JH/07 Add support for directories of certificates when compiled with a GnuTLS
247 version 3.3.6 or later.
249 JH/08 Rename the TPDA expermimental facility to Event Actions. The #ifdef
250 is EXPERIMENTAL_EVENT, the main-configuration and transport options
251 both become "event_action", the variables become $event_name, $event_data
252 and $event_defer_errno. There is a new variable $verify_mode, usable in
253 routers, transports and related events. The tls:cert event is now also
254 raised for inbound connections, if the main configuration event_action
257 TL/06 In test suite, disable OCSP for old versions of openssl which contained
258 early OCSP support, but no stapling (appears to be less than 1.0.0).
260 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
261 server certificate names available under the smtp transport option
262 "tls_verify_cert_hostname" now do not permit multi-component wildcard
265 JH/10 Time-related extraction expansions from certificates now use the main
266 option "timezone" setting for output formatting, and are consistent
267 between OpenSSL and GnuTLS compilations. Bug 1541.
269 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
270 encoded parameter in the incoming message. Bug 1558.
272 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
273 include certificate info, eximon was claiming there were spoolfile
276 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
278 JH/14 Log delivery-related information more consistently, using the sequence
279 "H=<name> [<ip>]" wherever possible.
281 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
282 are problematic for Debian distribution, omit them from the release
285 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
287 JH/16 Fix string representation of time values on 64bit time_t anchitectures.
290 JH/17 Fix a null-indirection in certextract expansions when a nondefault
291 output list separator was used.
296 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
297 checkers that were complaining about end of non-void function with no
300 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
301 This was a regression intruduced in 4.83 by another bugfix.
303 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
305 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
306 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
312 TF/01 Correctly close the server side of TLS when forking for delivery.
314 When a message was received over SMTP with TLS, Exim failed to clear up
315 the incoming connection properly after forking off the child process to
316 deliver the message. In some situations the subsequent outgoing
317 delivery connection happened to have the same fd number as the incoming
318 connection previously had. Exim would try to use TLS and fail, logging
319 a "Bad file descriptor" error.
321 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
322 utilities have not been installed.
324 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
325 temporary space as the ACL may create new global variables.
327 TL/01 LDAP support uses per connection or global context settings, depending
328 upon the detected version of the libraries at build time.
330 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
331 to extract and use the src ip:port in logging and expansions as if it
332 were a direct connection from the outside internet. PPv2 support was
333 updated based on HAProxy spec change in May 2014.
335 JH/02 Add ${listextract {number}{list}{success}{fail}}.
337 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
338 Properly escape header and check for NULL return.
340 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
343 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
345 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
346 characters in header names, implemented as a verify condition.
347 Contributed by Michael Fischer v. Mollard.
349 TL/05 Rename SPF condition results err_perm and err_temp to standardized
350 results permerror and temperror. Previous values are deprecated but
351 still accepted. In a future release, err_perm and err_temp will be
352 completely removed, which will be a backward incompatibility if the
353 ACL tests for either of these two old results. Patch contributed by
354 user bes-internal on the mailing list.
356 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
358 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
359 selectors, in both main and reject logs.
361 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
364 JH/07 Add malware type "sock" for talking to simple daemon.
366 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
368 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
369 routers/transports under cutthrough routing.
371 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
372 numbers. Touch up "bool" conditional to keep the same definition.
374 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
376 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
378 JH/12 Expand items in router/transport headers_add or headers_remove lists
379 individually rather than the list as a whole. Bug 1452.
381 Required for reasonable handling of multiple headers_ options when
382 they may be empty; requires that headers_remove items with embedded
383 colons must have them doubled (or the list-separator changed).
385 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
386 view the policy declared in the DMARC record. Currently, $dmarc_status
387 is a combined value of both the record presence and the result of the
390 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
392 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
393 dnslookup router and the smtp transport (applying to the forward
396 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
397 of ldap servers used for a specific lookup. Patch provided by Heiko
400 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
401 New variable $lookup_dnssec_authenticated for observability.
403 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
404 Patch submitted by Lars Timman.
406 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
408 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
409 Requires trusted mode and valid format message id, aborts otherwise.
410 Patch contributed by Heiko Schlichting.
412 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
413 certextract with support for various fields. Bug 1358.
415 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
416 is requested by default, modifiable by smtp transport option
419 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
420 operate on certificate variables to give certificate fingerprints
421 Also new ${sha256:cert_variable}.
423 JH/23 The PRDR feature is moved from being Experimental into the mainline.
425 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
426 Christian Aistleitner.
428 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
430 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
431 file. Patch from Wolfgang Breyha.
433 JH/25 Expand the coverage of the delivery $host and $host_address to
434 client authenticators run in verify callout. Bug 1476.
436 JH/26 Port service names are now accepted for tls_on_connect_ports, to
437 align with daemon_smtp_ports. Bug 72.
439 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
440 support and error reporting did not work properly.
442 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
443 and is readable. Patch from Andrew Colin Kissa.
445 TL/14 Enhance documentation of ${run expansion and how it parses the
446 commandline after expansion, particularly in the case when an
447 unquoted variable expansion results in an empty value.
449 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
451 PP/02 Fix internal collision of T_APL on systems which support RFC3123
452 by renaming away from it. Addresses GH issue 15, reported by
455 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
457 TL/15 SECURITY: prevent double expansion in math comparison functions
458 (can expand unsanitized data). Not remotely exploitable.
465 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
467 PP/02 Make -n do something, by making it not do something.
468 When combined with -bP, the name of an option is not output.
470 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
473 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
474 $sender_host_name and config options to manage this, and basic check
477 PP/05 DSCP support for outbound connections and control modifier for inbound.
479 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
480 (Only plugin which currently uses this is kerberos4, which nobody should
481 be using, but we should make it available and other future plugins might
482 conceivably use it, even though it would break NAT; stuff *should* be
483 using channel bindings instead).
485 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
486 name; added for Sendmail compatibility; requires admin caller.
487 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
488 just ignore it); requires trusted caller.
489 Also parse but ignore: -Ac -Am -X<logfile>
492 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
494 TL/02 Add +smtp_confirmation as a default logging option.
496 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
497 Patch by Magnus Holmgren from 2007-02-20.
499 TL/04 Bugzilla 1281 - Spec typo.
500 Bugzilla 1283 - Spec typo.
501 Bugzilla 1290 - Spec grammar fixes.
503 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
505 TL/06 Add Experimental DMARC support using libopendmarc libraries.
507 TL/07 Fix an out of order global option causing a segfault. Reported to dev
508 mailing list by by Dmitry Isaikin.
510 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
512 JH/02 Support "G" suffix to numbers in ${if comparisons.
514 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
516 NM/01 Bugzilla 1197 - Spec typo
517 Bugzilla 1196 - Spec examples corrections
519 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
521 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
522 gnutls_enable_pkcs11, but renamed to more accurately indicate its
525 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
526 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
528 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
529 "acl {{name}{arg}...}", and optional args on acl condition
532 JH/05 Permit multiple router/transport headers_add/remove lines.
534 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
536 JH/07 Avoid using a waiting database for a single-message-only transport.
537 Performance patch from Paul Fisher. Bugzilla 1262.
539 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
542 JH/09 Add $headers_added variable, with content from use of ACL modifier
543 add_header (but not yet added to the message). Bugzilla 199.
545 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
546 Pulled from Bugzilla 817 by Wolfgang Breyha.
548 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
550 (nb: this is the same fix as in Exim 4.80.1)
552 JH/11 Add A= logging on delivery lines, and a client_set_id option on
555 JH/12 Add optional authenticated_sender logging to A= and a log_selector
558 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
560 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
561 advertise SMTP AUTH mechanism to us, instead of a generic
562 protocol violation error. Also, make Exim more robust to bad
563 data from the Dovecot auth socket.
565 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
567 When a queue runner is handling a message, Exim first routes the
568 recipient addresses, during which it prunes them based on the retry
569 hints database. After that it attempts to deliver the message to
570 any remaining recipients. It then updates the hints database using
573 So if a recipient address works intermittently, it can get repeatedly
574 deferred at routing time. The retry hints record remains fresh so the
575 address never reaches the final cutoff time.
577 This is a fairly common occurrence when a user is bumping up against
578 their storage quota. Exim had some logic in its local delivery code
579 to deal with this. However it did not apply to per-recipient defers
580 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
582 This change adds a proper retry rule check during routing so that the
583 final cutoff time is checked against the message's age. We only do
584 this check if there is an address retry record and there is not a
585 domain retry record; this implies that previous attempts to handle
586 the address had the retry_use_local_parts option turned on. We use
587 this as an approximation for the destination being like a local
588 delivery, as in LMTP.
590 I suspect this new check makes the old local delivery cutoff check
591 redundant, but I have not verified this so I left the code in place.
593 TF/02 Correct gecos expansion when From: is a prefix of the username.
595 Test 0254 submits a message to Exim with the header
599 When I ran the test suite under the user fanf2, Exim expanded
600 the header to contain my full name, whereas it should have added
601 a Resent-Sender: header. It erroneously treats any prefix of the
602 username as equal to the username.
604 This change corrects that bug.
606 GF/01 DCC debug and logging tidyup
607 Error conditions log to paniclog rather than rejectlog.
608 Debug lines prefixed by "DCC: " to remove any ambiguity.
610 TF/03 Avoid unnecessary rebuilds of lookup-related code.
612 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
613 Bug spotted by Jeremy Harris; was flawed since initial commit.
614 Would have resulted in OCSP responses post-SNI triggering an Exim
615 NULL dereference and crash.
617 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
619 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
620 Bug detection, analysis and fix by Samuel Thibault.
621 Bugzilla 1331, Debian bug #698092.
623 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
625 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
626 Server implementation by Todd Lyons, client by JH.
627 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
628 config variable "prdr_enable" controls whether the server
629 advertises the facility. If the client requests PRDR a new
630 acl_data_smtp_prdr ACL is called once for each recipient, after
631 the body content is received and before the acl_smtp_data ACL.
632 The client is controlled by bolth of: a hosts_try_prdr option
633 on the smtp transport, and the server advertisement.
634 Default client logging of deliveries and rejections involving
635 PRDR are flagged with the string "PRDR".
637 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
638 fclose(). Diagnosis by Todd Lyons.
640 PP/17 Update configure.default to handle IPv6 localhost better.
641 Patch by Alain Williams (plus minor tweaks).
644 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
645 This is now consistent with GnuTLS, and is now documented: the
646 previous undocumented portable approach to treating the option as
647 unset was to force an expansion failure. That still works, and
648 an empty string is now equivalent.
650 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
651 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
652 not performing validation itself.
654 PP/20 Added force_command boolean option to pipe transport.
655 Patch from Nick Koston, of cPanel Inc.
657 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
660 TF/04 Added udpsend ACL modifer and hexquote expansion operator
662 PP/21 Fix eximon continuous updating with timestamped log-files.
663 Broken in a format-string cleanup in 4.80, missed when I repaired the
664 other false fix of the same issue.
665 Report and fix from Heiko Schlichting.
668 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
669 Report from Prashanth Katuri.
671 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
672 It's SecureTransport, so affects any MacOS clients which use the
673 system-integrated TLS libraries, including email clients.
675 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
676 using a MIME ACL for non-SMTP local injection.
677 Report and assistance in diagnosis by Warren Baker.
679 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
681 JH/16 Fix comparisons for 64b. Bugzilla 1385.
683 TL/09 Add expansion variable $authenticated_fail_id to keep track of
684 last id that failed so it may be referenced in subsequent ACL's.
686 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
689 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
690 ldap library initialization, allowing self-signed CA's to be
691 used. Also properly sets require_cert option later in code by
692 using NULL (global ldap config) instead of ldap handle (per
693 session). Bug diagnosis and testing by alxgomz.
695 TL/12 Enhanced documentation in the ratelimit.pl script provided in
696 the src/util/ subdirectory.
698 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
699 renamed to Transport Post Delivery Action by Jeremy Harris, as
702 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
703 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
704 redis_servers = needs to be configured which will be used by the redis
705 lookup. Patch from Warren Baker, of The Packet Hub.
707 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
709 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
710 hostname or reverse DNS when processing a host list. Used suggestions
711 from multiple comments on this bug.
713 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
715 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
716 Missed a few lines, added it to make the runtest require no keyboard
719 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
720 contains upper case chars. Make router use caseful_local_part.
722 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
723 support when GnuTLS has been built with p11-kit.
729 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
731 This, or similar/improved, will also be change PP/11 of 4.82.
737 PP/01 Handle short writes when writing local log-files.
738 In practice, only affects FreeBSD (8 onwards).
739 Bugzilla 1053, with thanks to Dmitry Isaikin.
741 NM/01 Bugzilla 949 - Documentation tweak
743 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
746 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
748 PP/02 Implemented gsasl authenticator.
750 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
752 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
753 `pkg-config foo` for cflags/libs.
755 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
756 with rest of GSASL and with heimdal_gssapi.
758 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
759 `pkg-config foo` for cflags/libs for the TLS implementation.
761 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
762 properties get this fed in as external SSF. A number of robustness
763 and debugging improvements to the cyrus_sasl authenticator.
765 PP/08 cyrus_sasl server now expands the server_realm option.
767 PP/09 Bugzilla 1214 - Log authentication information in reject log.
768 Patch by Jeremy Harris.
770 PP/10 Added dbmjz lookup type.
772 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
774 PP/12 MAIL args handles TAB as well as SP, for better interop with
775 non-compliant senders.
776 Analysis and variant patch by Todd Lyons.
778 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
779 Bug report from Lars Müller <lars@samba.org> (via SUSE),
780 Patch from Dirk Mueller <dmueller@suse.com>
782 PP/13 tls_peerdn now print-escaped for spool files.
783 Observed some $tls_peerdn in wild which contained \n, which resulted
784 in spool file corruption.
786 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
787 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
788 or write after TLS renegotiation, which otherwise led to messages
791 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
792 as a tracking header (ie: a signed header comes before the signature).
793 Patch from Wolfgang Breyha.
795 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
796 comma-sep list; embedded commas doubled.
798 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
800 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
802 Report and patch from Dmitry Banschikov.
804 PP/16 Removed "dont_insert_empty_fragments" fron "openssl_options".
805 Removed SSL_clear() after SSL_new() which led to protocol negotiation
806 failures. We appear to now support TLS1.1+ with Exim.
808 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
809 lets Exim select keys and certificates based upon TLS SNI from client.
810 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
811 before an outbound SMTP session. New log_selector, +tls_sni.
813 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
814 NULL dereference. Report and patch from Alun Jones.
816 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
817 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
818 Not seeing resolver debug output on NetBSD, but suspect this is a
819 resolver implementation change.
821 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
822 Left warnings. Added "eximon gdb" invocation mode.
824 PP/21 Defaulting "accept_8bitmime" to true, not false.
826 PP/22 Added -bw for inetd wait mode support.
828 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
829 locate the relevant includes and libraries. Made this the default.
831 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
832 Bugzilla 1246, report and most of solution from Tomasz Kusy.
834 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
835 This may cause build issues on older platforms.
837 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
838 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
839 gnutls_require_mac & gnutls_require_protocols (no longer supported).
840 Added SNI support via GnuTLS too.
841 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
843 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
845 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
846 Applied second patch from Janne, implementing suggestion to default
847 multiple-strings-in-record handling to match SPF spec.
849 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
851 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
852 read-only, out of scope).
853 Patch from Wolfgang Breyha, report from Stuart Northfield.
855 PP/29 Fix three issues highlighted by clang analyser static analysis.
856 Only crash-plausible issue would require the Cambridge-specific
857 iplookup router and a misconfiguration.
858 Report from Marcin Mirosław.
860 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
862 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
863 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
864 As part of this, removing so much warning spew let me fix some minor
865 real issues in debug logging.
867 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
868 assignment on my part. Fixed.
870 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
871 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
872 Janne Snabb (who went above and beyond: thank you).
874 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
875 string otherwise requires a connection and a bunch more work and it's
876 relatively easy to get wrong. Should also expose TLS library linkage
879 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
880 64-bit ${eval} (JH/03).
882 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
883 GNU libc to support some of the 64-bit stuff, should not lead to
884 conflicts. Defined before os.h is pulled in, so if a given platform
885 needs to override this, it can.
887 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
888 protection layer was required, which is not implemented.
889 Bugzilla 1254, patch from Wolfgang Breyha.
891 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
892 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
893 tls_dhparam take prime identifiers. Also unbreak combination of
894 OpenSSL+DH_params+TLSSNI.
896 PP/39 Disable SSLv2 by default in OpenSSL support.
902 PP/01 Solaris build fix for Oracle's LDAP libraries.
903 Bugzilla 1109, patch from Stephen Usher.
905 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
907 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
910 TF/02 Fix a couple more cases where we did not log the error message
911 when unlink() failed. See also change 4.74-TF/03.
913 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
914 lock up or crash if it happened to be inside a call to libc when it
915 got a SIGUSR1 from exiwhat.
917 The SIGUSR1 handler appends the current process status to the process
918 log which is later printed by exiwhat. It used to use the general
919 purpose logging code to do this, but several functions it calls are
920 not safe for signals.
922 The new output code in the SIGUSR1 handler is specific to the process
923 log, and simple enough that it's easy to inspect for signal safety.
924 Removing some special cases also simplifies the general logging code.
925 Removing the spurious timestamps from the process log simplifies
928 TF/04 Improved ratelimit ACL condition.
930 The /noupdate option has been deprecated in favour of /readonly which
931 has clearer semantics. The /leaky, /strict, and /readonly update modes
932 are mutually exclusive. The update mode is no longer included in the
933 database key; it just determines when the database is updated. (This
934 means that when you upgrde Exim will forget old rate measurements.)
936 Exim now checks that the per_* options are used with an update mode that
937 makes sense for the current ACL. For example, when Exim is processing a
938 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
939 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
940 must specify per_mail/readonly. If you omit the update mode it defaults to
941 /leaky where that makes sense (as before) or /readonly where required.
943 The /noupdate option is now undocumented but still supported for
944 backwards compatibility. It is equivalent to /readonly except that in
945 ACLs where /readonly is required you may specify /leaky/noupdate or
946 /strict/noupdate which are treated the same as /readonly.
948 A useful new feature is the /count= option. This is a generalization
949 of the per_byte option, so that you can measure the throughput of other
950 aggregate values. For example, the per_byte option is now equivalent
951 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
953 The per_rcpt option has been generalized using the /count= mechanism
954 (though it's more complicated than the per_byte equivalence). When it is
955 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
956 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
957 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
958 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
959 Note that using per_rcpt with a non-readonly update mode in more than
960 one ACL will cause the recipients to be double-counted. (The per_mail
961 and per_byte options don't have this problem.)
963 The handling of very low rates has changed slightly. If the computed rate
964 is less than the event's count (usually one) then this event is the first
965 after a long gap. In this case the rate is set to the same as this event's
966 count, so that the first message of a spam run is counted properly.
968 The major new feature is a mechanism for counting the rate of unique
969 events. The new per_addr option counts the number of different
970 recipients that someone has sent messages to in the last time period. It
971 behaves like per_rcpt if all the recipient addresses are different, but
972 duplicate recipient addresses do not increase the measured rate. Like
973 the /count= option this is a general mechanism, so the per_addr option
974 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
975 example, measure the rate that a client uses different sender addresses
976 with the options per_mail/unique=$sender_address. There are further
977 details in the main documentation.
979 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
981 TF/06 Removed a few PCRE remnants.
983 TF/07 Automatically extract Exim's version number from tags in the git
984 repository when doing development or release builds.
986 PP/02 Raise smtp_cmd_buffer_size to 16kB.
987 Bugzilla 879. Patch from Paul Fisher.
989 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
990 Heavily based on revision 40f9a89a from Simon Arlott's tree.
993 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
995 PP/05 Variable $av_failed, true if the AV scanner deferred.
996 Bugzilla 1078. Patch from John Horne.
998 PP/06 Stop make process more reliably on build failure.
999 Bugzilla 1087. Patch from Heiko Schlittermann.
1001 PP/07 Make maildir_use_size_file an _expandable_ boolean.
1002 Bugzilla 1089. Patch from Heiko Schlittermann.
1004 PP/08 Handle ${run} returning more data than OS pipe buffer size.
1005 Bugzilla 1131. Patch from Holger Weiß.
1007 PP/09 Handle IPv6 addresses with SPF.
1008 Bugzilla 860. Patch from Wolfgang Breyha.
1010 PP/10 GnuTLS: support TLS 1.2 & 1.1.
1012 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
1015 PP/11 match_* no longer expand right-hand-side by default.
1016 New compile-time build option, EXPAND_LISTMATCH_RHS.
1017 New expansion conditions, "inlist", "inlisti".
1019 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
1021 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
1023 PP/14 fix log_write() format string regression from TF/03.
1024 Bugzilla 1152. Patch from Dmitry Isaikin.
1030 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
1032 PP/02 Harmonised TLS library version reporting; only show if debugging.
1033 Layout now matches that introduced for other libraries in 4.74 PP/03.
1035 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
1037 PP/04 New "dns_use_edns0" global option.
1039 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
1042 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
1043 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
1045 TK/01 Updated PolarSSL code to 0.14.2.
1046 Bugzilla 1097. Patch from Andreas Metzler.
1048 PP/07 Catch divide-by-zero in ${eval:...}.
1049 Fixes bugzilla 1102.
1051 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
1054 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
1055 format-string attack -- SECURITY: remote arbitrary code execution.
1057 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
1058 time unintentionally subject to list matching rules, letting the header
1059 cause arbitrary Exim lookups (of items which can occur in lists, *not*
1060 arbitrary string expansion). This allowed for information disclosure.
1062 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
1063 INT_MIN/-1 -- value coerced to INT_MAX.
1069 NM/01 Workround for PCRE version dependency in version reporting
1072 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
1073 This fixes portability to compilers other than gcc, notably
1074 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
1076 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
1077 makefiles for portability to HP-UX and POSIX correctness.
1079 PP/01 Permit LOOKUP_foo enabling on the make command-line.
1080 Also via indented variable definition in the Makefile.
1081 (Debugging by Oliver Heesakkers).
1083 PP/02 Restore caching of spamd results with expanded spamd_address.
1084 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
1086 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
1087 Improves build reliability. Fix from: Frank Elsner
1089 NM/02 Fix wide character breakage in the rfc2047 coding
1090 Fixes bug 1064. Patch from Andrey N. Oktyabrski
1092 NM/03 Allow underscore in dnslist lookups
1093 Fixes bug 1026. Patch from Graeme Fowler
1095 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
1096 Code patches from Adam Ciarcinski of NetBSD.
1098 NM/04 Fixed exiqgrep to cope with mailq missing size issue
1101 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
1102 is logged, to avoid truncation. Patch from John Horne.
1104 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
1105 Patch from Jakob Hirsch.
1107 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
1108 SQL string expansion failure details.
1109 Patch from Andrey Oktyabrski.
1111 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
1112 Patch from Simon Arlott.
1114 PP/09 New lookups functionality failed to compile on old gcc which rejects
1115 extern declarations in function scope.
1116 Patch from Oliver Fleischmann
1118 PP/10 Use sig_atomic_t for flags set from signal handlers.
1119 Check getgroups() return and improve debugging.
1120 Fixed developed for diagnosis in bug 927 (which turned out to be
1123 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
1124 Patch from Mark Zealey.
1126 PP/12 Bugzilla 1056: Improved spamd server selection.
1127 Patch from Mark Zealey.
1129 PP/13 Bugzilla 1086: Deal with maildir quota file races.
1130 Based on patch from Heiko Schlittermann.
1132 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
1133 Patch from Uwe Doering, sign-off by Michael Haardt.
1135 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
1136 variable declaration deep within a block. Bug and patch from
1139 PP/15 lookups-Makefile IRIX compatibilty coercion.
1141 PP/16 Make DISABLE_DKIM build knob functional.
1143 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
1144 Patch by Simon Arlott
1146 TF/03 Fix valgrind.h portability to C89 compilers that do not support
1147 variable argument macros. Our copy now differs from upstream.
1153 TF/01 Failure to get a lock on a hints database can have serious
1154 consequences so log it to the panic log.
1156 TF/02 Log LMTP confirmation messages in the same way as SMTP,
1157 controlled using the smtp_confirmation log selector.
1159 TF/03 Include the error message when we fail to unlink a spool file.
1161 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
1162 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
1163 for maintaining out-of-tree patches for some time.
1165 PP/01 Bugzilla 139: Documentation and portability issues.
1166 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
1167 Handle per-OS dynamic-module compilation flags.
1169 PP/02 Let /dev/null have normal permissions.
1170 The 4.73 fixes were a little too stringent and complained about the
1171 permissions on /dev/null. Exempt it from some checks.
1172 Reported by Andreas M. Kirchwitz.
1174 PP/03 Report version information for many libraries, including
1175 Exim version information for dynamically loaded libraries. Created
1176 version.h, now support a version extension string for distributors
1177 who patch heavily. Dynamic module ABI change.
1179 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
1180 privilege escalation vulnerability whereby the Exim run-time user
1181 can cause root to append content of the attacker's choosing to
1184 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
1187 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
1188 If dropping privileges for untrusted macros, we disabled normal logging
1189 on the basis that it would fail; for the Exim run-time user, this is not
1190 the case, and it resulted in successful deliveries going unlogged.
1191 Fixed. Reported by Andreas Metzler.
1197 PP/01 Date: & Message-Id: revert to normally being appended to a message,
1198 only prepend for the Resent-* case. Fixes regression introduced in
1199 Exim 4.70 by NM/22 for Bugzilla 607.
1201 PP/02 Include check_rfc2047_length in configure.default because we're seeing
1202 increasing numbers of administrators be bitten by this.
1204 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
1206 PP/03 Bugzilla 994: added openssl_options main configuration option.
1208 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
1210 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
1212 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
1214 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
1215 without a peer certificate, leading to a segfault because of an
1216 assumption that peers always have certificates. Be a little more
1217 paranoid. Problem reported by Martin Tscholak.
1219 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
1220 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
1221 NB: ClamAV planning to remove STREAM in "middle of 2010".
1222 CL also introduces -bmalware, various -d+acl logging additions and
1223 more caution in buffer sizes.
1225 PP/09 Implemented reverse_ip expansion operator.
1227 PP/10 Bugzilla 937: provide a "debug" ACL control.
1229 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
1231 PP/12 Bugzilla 973: Implement --version.
1233 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
1235 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
1237 PP/15 Bugzilla 816: support multiple condition rules on Routers.
1239 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
1240 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
1241 ignore trailing whitespace.
1243 JJ/02 prevent non-panic DKIM error from being sent to paniclog
1245 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
1248 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
1249 Notification from Dr Andrew Aitchison.
1251 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
1252 ExtendedDetectionInfo response format.
1253 Notification from John Horne.
1255 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
1258 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
1259 XSL and documented dependency on system catalogs, with examples of how
1262 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
1265 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
1266 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
1267 configuration file which is writeable by the Exim user or group.
1269 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
1270 of configuration files to cover files specified with the -C option if
1271 they are going to be used with root privileges, not just the default
1274 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
1275 option (effectively making it always true).
1277 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
1278 files to be used while preserving root privileges.
1280 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
1281 that rogue child processes cannot use them.
1283 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
1284 run-time user, instead of root.
1286 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
1287 Exim run-time user without dropping privileges.
1289 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
1290 result string, instead of calling string_vformat() twice with the same
1293 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
1294 for other users. Others should always drop root privileges if they use
1295 -C on the command line, even for a whitelisted configure file.
1297 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
1299 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
1305 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
1306 $data_path, and $header_path variables; fixed documentation bugs and
1309 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
1310 exipick to access non-standard spools, including the "frozen" queue
1313 NM/01 Bugzilla 965: Support mysql stored procedures.
1314 Patch from Alain Williams
1316 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
1318 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
1319 Patch from Andreas Metzler
1321 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
1322 Patch from Kirill Miazine
1324 NM/05 Bugzilla 671: Added umask to procmail example.
1326 JJ/03 installed exipick 20100323.0, fixing doc bug
1328 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
1329 directory. Notification and patch from Dan Rosenberg.
1331 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
1333 TK/02 Improve log output when DKIM signing operation fails.
1335 MH/01 Treat the transport option dkim_domain as a colon separated
1336 list, not as a single string, and sign the message with each element,
1337 omitting multiple occurences of the same signer.
1339 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
1340 Bugzilla 985, 986. Patch by Simon Arlott
1342 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
1343 Patch by Simon Arlott
1345 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
1346 MBX locking. Notification from Dan Rosenberg.
1352 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
1354 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
1356 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
1358 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
1360 NM/04 Bugzilla 915: Flag broken perl installation during build.
1366 TK/01 Added patch by Johannes Berg that expands the main option
1367 "spamd_address" if it starts with a dollar sign.
1369 TK/02 Write list of recipients to X-Envelope-Sender header when building
1370 the mbox-format spool file for content scanning (suggested by Jakob
1373 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
1374 (http://www.dcc-servers.net/) support via dccifd. Activated by
1375 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
1377 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
1378 by Mark Daniel Reidel <mr@df.eu>.
1380 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
1381 When building exim an external PCRE library is now needed -
1382 PCRE is a system library on the majority of modern systems.
1383 See entry on PCRE_LIBS in EDITME file.
1385 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
1386 conversation. Added nologin parameter to request.
1387 Patch contributed by Kirill Miazine.
1389 TF/01 Do not log submission mode rewrites if they do not change the address.
1391 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
1393 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
1394 log files in place. Contributed by Roberto Lima.
1396 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
1398 TF/03 Bugzilla 615: When checking the local_parts router precondition
1399 after a local_part_suffix or local_part_prefix option, Exim now
1400 does not use the address's named list lookup cache, since this
1401 contains cached lookups for the whole local part.
1403 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
1404 Robert Millan. Documentation is in experimental-spec.txt.
1406 TF/04 Bugzilla 668: Fix parallel build (make -j).
1408 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
1410 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
1411 Patch provided by Jan Srzednicki.
1413 TF/05 Leading white space used to be stripped from $spam_report which
1414 wrecked the formatting. Now it is preserved.
1416 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
1417 that they are available at delivery time.
1419 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
1421 TF/08 TLS error reporting now respects the incoming_interface and
1422 incoming_port log selectors.
1424 TF/09 Produce a more useful error message if an SMTP transport's hosts
1425 setting expands to an empty string.
1427 NM/06 Bugzilla 744: EXPN did not work under TLS.
1428 Patch provided by Phil Pennock.
1430 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
1431 Patch provided by Richard Godbee.
1433 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
1434 acl_smtp_notquit, added index entry.
1436 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
1437 Patch provided by Eugene Bujak.
1439 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
1440 accept(). Patch provided by Maxim Dounin.
1442 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
1443 Patch provided by Phil Pennock.
1445 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
1447 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
1448 Patch provided by Brad "anomie" Jorsch.
1450 NM/14 Bugzilla 622: Added timeout setting to transport filter.
1451 Patch provided by Dean Brooks.
1453 TK/05 Add native DKIM support (does not depend on external libraries).
1455 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
1456 Patch provided by Graeme Fowler.
1458 NM/16 Bugzilla 851: Documentation example syntax fix.
1460 NM/17 Changed NOTICE file to remove references to embedded PCRE.
1462 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
1465 NM/19 Bugzilla 745: TLS version reporting.
1466 Patch provided by Phil Pennock.
1468 NM/20 Bugzilla 167: bool: condition support.
1469 Patch provided by Phil Pennock.
1471 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
1472 clients. Patch provided by Phil Pennock.
1474 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
1475 Patch provided by Brad "anomie" Jorsch.
1477 NM/23 Bugzilla 687: Fix misparses in eximstats.
1478 Patch provided by Heiko Schlittermann.
1480 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
1481 Patch provided by Heiko Schlittermann.
1483 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
1484 plus update to original patch.
1486 NM/26 Bugzilla 799: Documentation correction for ratelimit.
1488 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
1489 Patch provided by David Brownlee.
1491 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
1493 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
1495 NM/30 Bugzilla 888: TLS documentation bugfixes.
1497 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
1499 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
1500 Unlike the original bugzilla I have changed all shell scripts in src tree.
1502 NM/33 Bugzilla 898: Transport filter timeout fix.
1503 Patch by Todd Rinaldo.
1505 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
1506 Patch by Serge Demonchaux.
1508 NM/35 Bugzilla 39: Base64 decode bug fixes.
1509 Patch by Jakob Hirsch.
1511 NM/36 Bugzilla 909: Correct connect() call in dcc code.
1513 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
1515 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
1517 NM/39 Bugzilla 911: Fixed MakeLinks build script.
1523 TK/01 Add preliminary DKIM support. Currently requires a forked version of
1524 ALT-N's libdkim that I have put here:
1525 http://duncanthrax.net/exim-experimental/
1527 Note to Michael Haardt: I had to rename some vars in sieve.c. They
1528 were called 'true' and it seems that C99 defines that as a reserved
1529 keyword to be used with 'bool' variable types. That means you could
1530 not include C99-style headers which use bools without triggering
1531 build errors in sieve.c.
1533 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
1534 as mailq or other aliases. Changed the --help handling significantly
1535 to do whats expected. exim_usage() emits usage/help information.
1537 SC/01 Added the -bylocaldomain option to eximstats.
1539 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
1541 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
1543 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
1549 PH/01 Another patch from the Sieve maintainer.
1551 PH/02 When an IPv6 address is converted to a string for single-key lookup
1552 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
1553 dots are used instead of colons so that keys in lsearch files need not
1554 contain colons. This was done some time before quoting was made available
1555 in lsearch files. However, iplsearch files do require colons in IPv6 keys
1556 (notated using the quote facility) so as to distinguish them from IPv4
1557 keys. This meant that lookups for IP addresses in host lists did not work
1558 for iplsearch lookups.
1560 This has been fixed by arranging for IPv6 addresses to be expressed with
1561 colons if the lookup type is iplsearch. This is not incompatible, because
1562 previously such lookups could never work.
1564 The situation is now rather anomolous, since one *can* have colons in
1565 ordinary lsearch keys. However, making the change in all cases is
1566 incompatible and would probably break a number of configurations.
1568 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
1571 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
1572 conversion specification without a maximum field width, thereby enabling
1573 a rogue spamd server to cause a buffer overflow. While nobody in their
1574 right mind would setup Exim to query an untrusted spamd server, an
1575 attacker that gains access to a server running spamd could potentially
1576 exploit this vulnerability to run arbitrary code as the Exim user.
1578 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
1579 $primary_hostname instead of what libspf2 thinks the hosts name is.
1581 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
1582 a directory entry by the name of the lookup key. Previously, if a
1583 symlink pointed to a non-existing file or a file in a directory that
1584 Exim lacked permissions to read, a lookup for a key matching that
1585 symlink would fail. Now it is enough that a matching directory entry
1586 exists, symlink or not. (Bugzilla 503.)
1588 PH/03 The body_linecount and body_zerocount variables are now exported in the
1591 PH/04 Added the $dnslist_matched variable.
1593 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
1594 This means they are set thereafter only if the connection becomes
1597 PH/06 Added the client_condition to authenticators so that some can be skipped
1598 by clients under certain conditions.
1600 PH/07 The error message for a badly-placed control=no_multiline_responses left
1601 "_responses" off the end of the name.
1603 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
1605 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
1606 (without spaces) instead of just copying the configuration text.
1608 PH/10 Added the /noupdate option to the ratelimit ACL condition.
1610 PH/11 Added $max_received_linelength.
1612 PH/12 Added +ignore_defer and +include_defer to host lists.
1614 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
1615 way in which PCRE > 7.0 is built.
1617 PH/14 Implemented queue_only_load_latch.
1619 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
1620 MAIL command. The effect was to mangle the value on 64-bit systems.
1622 PH/16 Another patch from the Sieve maintainer.
1624 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
1626 PH/18 If a system quota error occurred while trying to create the file for
1627 a maildir delivery, the message "Mailbox is full" was not appended to the
1628 bounce if the delivery eventually timed out. Change 4.67/27 below applied
1629 only to a quota excession during the actual writing of the file.
1631 PH/19 It seems that peer DN values may contain newlines (and other non-printing
1632 characters?) which causes problems in log lines. The DN values are now
1633 passed through string_printing() before being added to log lines.
1635 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
1636 and InterBase are left for another time.)
1638 PH/21 Added message_body_newlines option.
1640 PH/22 Guard against possible overflow in moan_check_errorcopy().
1642 PH/23 POSIX allows open() to be a macro; guard against that.
1644 PH/24 If the recipient of an error message contained an @ in the local part
1645 (suitably quoted, of course), incorrect values were put in $domain and
1646 $local_part during the evaluation of errors_copy.
1652 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
1653 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
1656 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
1657 issue a MAIL command.
1659 PH/02 In an ACL statement such as
1661 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
1663 if a client was not listed at all, or was listed with a value other than
1664 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
1665 the condition was not true (as it should be), so access was not denied.
1666 The bug was that the ! inversion was incorrectly passed on to the second
1667 item. This has been fixed.
1669 PH/03 Added additional dnslists conditions == and =& which are different from
1670 = and & when the dns lookup returns more than one IP address.
1672 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
1673 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
1675 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
1676 FSYNC, which compiles an option called disable_fsync that allows for
1677 bypassing fsync(). The documentation is heavily laced with warnings.
1679 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
1681 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
1682 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
1683 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
1684 including adding "make clean"; (3) Added -fPIC when compiling the test
1685 dynamically loaded module, to get rid of a warning.
1687 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
1688 message fails, move_frozen_messages = true and ignore_bounce_errors_after
1689 = 0s. The bug is otherwise harmless.
1691 PH/07 There was a bug in the dovecot authenticator such that the value of
1692 $auth1 could be overwritten, and so not correctly preserved, after a
1693 successful authentication. This usually meant that the value preserved by
1694 the server_setid option was incorrect.
1696 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
1698 PH/09 Installed PCRE release 7.0.
1700 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
1701 run for batched SMTP input. It is now run at the start of every message
1702 in the batch. While fixing this I discovered that the process information
1703 (output by running exiwhat) was not always getting set for -bs and -bS
1704 input. This is fixed, and it now also says "batched" for BSMTP.
1706 PH/11 Added control=no_pipelining.
1708 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
1709 patch, slightly modified), and move the expansion of helo_data till after
1710 the connection is made in the smtp transport (so it can use these
1713 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
1715 PH/14 Added log_selector = +pid.
1717 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
1719 PH/16 Add ${if forany and ${if forall.
1721 PH/17 Added dsn_from option to vary the From: line in DSNs.
1723 PH/18 Flush SMTP output before performing a callout, unless control =
1724 no_callout_flush is set.
1726 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
1727 was true (the default) a successful delivery failed to delete the retry
1728 item, thus causing premature timeout of the address. The bug is now
1731 PH/20 Added hosts_avoid_pipelining to the smtp transport.
1733 PH/21 Long custom messages for fakedefer and fakereject are now split up
1734 into multiline reponses in the same way that messages for "deny" and
1735 other ACL rejections are.
1737 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
1738 with slight modification.
1740 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
1741 draft, changing the syntax and factoring some duplicate code".
1743 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
1744 for deliveries of the second and subsequent messages over the same SMTP
1747 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
1748 ${reduce, with only minor "tidies".
1750 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
1752 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
1753 expansion side effects.
1755 PH/27 When a message times out after an over-quota error from an Exim-imposed
1756 quota, the bounce message says "mailbox is full". This message was not
1757 being given when it was a system quota that was exceeded. It now should
1760 MH/03 Made $recipients available in local_scan(). local_scan() already has
1761 better access to the recipient list through recipients_list[], but
1762 $recipients can be useful in postmaster-provided expansion strings.
1764 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
1765 in the case of a MAIL command with additional options following the
1766 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
1767 were accidentally chopped off.
1769 PH/29 SMTP synchronization checks are implemented when a command is read -
1770 there is a check that no more input is waiting when there shouldn't be
1771 any. However, for some commands, a delay in an ACL can mean that it is
1772 some time before the response is written. In this time, more input might
1773 arrive, invalidly. So now there are extra checks after an ACL has run for
1774 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
1775 pipelining has not been advertised.
1777 PH/30 MH's patch to allow iscntrl() characters to be list separators.
1779 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
1780 returned in the SMTP response when smtp_return_error_details was false.
1781 This has been fixed.
1783 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
1784 instead of the C I/O that was originally supplied, because problems were
1785 reported on Solaris.
1787 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
1788 Exim which did not show up earlier: it was assuming that a call to
1789 SSL_CTX_set_info_callback() might give an error value. In fact, there is
1790 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
1791 was a macro that became an assignment, so it seemed to work. This has
1792 changed to a proper function call with a void return, hence the compile
1793 error. Exim's code has been fixed.
1795 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
1798 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
1800 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
1802 PH/37 If a message is not accepted after it has had an id assigned (e.g.
1803 because it turns out to be too big or there is a timeout) there is no
1804 "Completed" line in the log. When some messages of this type were
1805 selected by exigrep, they were listed as "not completed". Others were
1806 picked up by some special patterns. I have improved the selection
1807 criteria to be more general.
1809 PH/38 The host_find_failed option in the manualroute router can now be set
1810 to "ignore", to completely ignore a host whose IP address cannot be
1811 found. If all hosts are ignored, the behaviour is controlled by the new
1812 host_all_ignored option.
1814 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
1815 homing or because of multiple MX records with /mx) generated more than
1816 one IP address, and the following item turned out to be the local host,
1817 all the secondary addresses of the first item were incorrectly removed
1818 from the list, along with the local host and any following hosts (which
1819 is what is supposed to happen).
1821 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
1822 whoever called Exim into the -H file. In the case of the daemon it was
1823 behaving confusingly. When first started, it used values for whoever
1824 started the daemon, but after a SIGHUP it used the Exim user (because it
1825 calls itself on a restart). I have changed the code so that it now always
1828 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
1829 message are rejected with the same error (e.g. no authentication or bad
1830 sender address), and a DATA command is nevertheless sent (as can happen
1831 with PIPELINING or a stupid MUA), the error message that was given to the
1832 RCPT commands is included in the rejection of the DATA command. This is
1833 intended to be helpful for MUAs that show only the final error to their
1836 PH/42 Another patch from the Sieve maintainer.
1838 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
1839 Eximstats - Fixed some broken HTML links and added missing column headers
1841 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
1842 columns for Rejects, Temp Rejects, Ham, and Spam rows.
1844 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
1846 PH/43 Yet another patch from the Sieve maintainer.
1848 PH/44 I found a way to check for a TCP/IP connection going away before sending
1849 the response to the final '.' that terminates a message, but only in the
1850 case where the client has not sent further data following the '.'
1851 (unfortunately, this is allowed). However, in many cases there won't be
1852 any further data because there won't be any more messages to send. A call
1853 to select() can be used: if it shows that the input is "ready", there is
1854 either input waiting, or the socket has been closed. An attempt to read
1855 the next input character can distinguish the two cases. Previously, Exim
1856 would have sent an OK response which the client would never have see.
1857 This could lead to message repetition. This fix should cure that, at
1858 least in a lot of common cases.
1860 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
1861 advertised in response to EHLO.
1867 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
1868 fixed by 4.65/MH/01 (is this a record?) are fixed:
1870 (i) An empty string was always treated as zero by the numeric comparison
1871 operators. This behaviour has been restored.
1873 (ii) It is documented that the numeric comparison operators always treat
1874 their arguments as decimal numbers. This was broken in that numbers
1875 starting with 0 were being interpreted as octal.
1877 While fixing these problems I realized that there was another issue that
1878 hadn't been noticed. Values of message_size_limit (both the global option
1879 and the transport option) were treated as octal if they started with 0.
1880 The documentation was vague. These values are now always treated as
1881 decimal, and I will make that clear in the documentation.
1887 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
1888 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
1891 MH/01 Don't check that the operands of numeric comparison operators are
1892 integers when their expansion is in "skipping" mode (fixes bug
1893 introduced by 4.64-PH/07).
1895 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
1896 child addresses, Exim now panics and dies. Previously, because the count
1897 is held in a short int, deliveries were likely to be lost. As such a
1898 large number of recipients for a single message is ridiculous
1899 (performance will be very, very poor), I have chosen to impose a limit
1900 rather than extend the field.
1906 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
1907 leftover -K file (the existence of which was triggered by #402).
1908 While we were at it, introduced process PID as part of the -K
1909 filename. This should rule out race conditions when creating
1912 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
1913 processing considerably. Previous code took too long for large mails,
1914 triggering a timeout which in turn triggers #401.
1916 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
1917 in the DK code in transports.c. sendfile() is not really portable,
1918 hence the _LINUX specificness.
1920 TF/01 In the add_headers option to the mail command in an Exim filter,
1921 there was a bug that Exim would claim a syntax error in any
1922 header after the first one which had an odd number of characters
1925 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
1926 callout verification, Exim cached a "reject" for the entire domain. This
1927 is correct for most verifications, but it is not correct for a recipient
1928 verification with use_sender or use_postmaster set, because in that case
1929 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
1930 case of MAIL FROM:<> rejection from other early rejections (e.g.
1931 rejection of HELO). When verifying a recipient using a non-null MAIL
1932 address, the cache is ignored if it shows MAIL FROM:<> rejection.
1933 Whatever the result of the callout, the value of the domain cache is
1934 left unchanged (for any other kind of callout, getting as far as trying
1935 RCPT means that the domain itself is ok).
1937 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
1940 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
1941 manifest itself as EPIPE rather than ECONNECT. When tidying away a
1942 session, the daemon ignores ECONNECT errors and logs others; it now
1943 ignores EPIPE as well.
1945 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
1946 (quoted-printable decoding).
1948 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
1949 later the small subsequent patch to fix an introduced bug.
1951 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
1953 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
1955 PH/08 An error is now given if message_size_limit is specified negative.
1957 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
1958 to be given (somewhat) arbitrary names.
1960 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
1963 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
1964 miscellaneous code fixes
1966 PH/10 Added the log_reject_target ACL modifier to specify where to log
1969 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
1970 hostname. This is wrong, because it relates to the incoming message (and
1971 probably the interface on which it is arriving) and not to the outgoing
1972 callout (which could be using a different interface). This has been
1973 changed to use the value of the helo_data option from the smtp transport
1974 instead - this is what is used when a message is actually being sent. If
1975 there is no remote transport (possible with a router that sets up host
1976 addresses), $smtp_active_hostname is used.
1978 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
1979 tweaks were necessary in order to get it to work (see also 21 below):
1980 (a) The code assumed that strncpy() returns a negative number on buffer
1981 overflow, which isn't the case. Replaced with Exim's string_format()
1983 (b) There were several signed/unsigned issues. I just did the minimum
1984 hacking in of casts. There is scope for a larger refactoring.
1985 (c) The code used strcasecmp() which is not a standard C function.
1986 Replaced with Exim's strcmpic() function.
1987 (d) The code set only $1; it now sets $auth1 as well.
1988 (e) A simple test gave the error "authentication client didn't specify
1989 service in request". It would seem that Dovecot has changed its
1990 interface. Fortunately there's a specification; I followed it and
1991 changed what the client sends and it appears to be working now.
1993 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
1996 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
1997 address A is aliased to B and C, where B exists and C does not. Without
1998 -v the output is "A verified" because verification stops after a
1999 successful redirection if more than one address is generated. However,
2000 with -v the child addresses are also verified. Exim was outputting "A
2001 failed to verify" and then showing the successful verification for C,
2002 with its parentage. It now outputs "B failed to verify", showing B's
2003 parentage before showing the successful verification of C.
2005 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
2006 look up a TXT record in a specific list after matching in a combined
2009 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
2010 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
2011 they consult the DNS. I had assumed they would set it the way they
2012 wanted; and indeed my experiments on Linux seem to show that in some
2013 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
2014 To be on the safe side, however, I have now made the interface to
2015 host_find_byname() similar to host_find_bydns(), with an argument
2016 containing the DNS resolver options. The host_find_byname() function now
2017 sets these options at its start, just as host_find_bydns() does. The smtp
2018 transport options dns_qualify_single and dns_search_parents are passed to
2019 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
2020 of host_find_byname() use the default settings of RES_DEFNAMES
2021 (qualify_single) but not RES_DNSRCH (search_parents).
2023 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
2024 spool_read_header() do less string testing, by means of a preliminary
2025 switch on the second character of optional "-foo" lines. (This is
2026 overdue, caused by the large number of possibilities that now exist.
2027 Originally there were few.) While I was there, I also converted the
2028 str(n)cmp tests so they don't re-test the leading "-" and the first
2029 character, in the hope this might squeeze out yet more improvement.
2031 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
2032 flag allowing group syntax was set by the header_syntax check but not
2033 turned off, possible causing trouble later; (2) The flag was not being
2034 set at all for the header_verify test, causing "group"-style headers to
2035 be rejected. I have now set it in this case, and also caused header_
2036 verify to ignore an empty address taken from a group. While doing this, I
2037 came across some other cases where the code for allowing group syntax
2038 while scanning a header line wasn't quite right (mostly, not resetting
2039 the flag correctly in the right place). These bugs could have caused
2040 trouble for malformed header lines. I hope it is now all correct.
2042 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
2043 with the "reply" argument non-NULL. The code, however (which originally
2044 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
2045 but it didn't always do it. This confused somebody who was copying the
2046 code for some other use. I have removed all the tests.
2048 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
2049 feature that was used to support insecure browsers during the U.S. crypto
2050 embargo. It requires special client support, and Exim is probably the
2051 only MTA that supported it -- and would never use it because real RSA is
2052 always available. This code has been removed, because it had the bad
2053 effect of slowing Exim down by computing (never used) parameters for the
2054 RSA_EXPORT functionality.
2056 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
2057 authenticator to fail if there's a tab character in the incoming data
2058 (there should never be unless someone is messing about, as it's supposed
2059 to be base64-encoded). Also added, on Timo's advice, the "secured" option
2060 if the connection is using TLS or if the remote IP is the same as the
2061 local IP, and the "valid-client-cert option" if a client certificate has
2064 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
2065 authenticators. This can be used for authorization after authentication
2066 succeeds. (In the case of plaintext, it servers for both authentication
2069 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
2070 if any retry times were supplied.
2072 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
2073 connection, where there is no client IP address to check. In this
2074 situation, the verify now always succeeds.
2076 PH/25 Applied John Jetmore's -Mset patch.
2078 PH/26 Added -bem to be like -Mset, but loading a message from a file.
2080 PH/27 In a string expansion for a processed (not raw) header when multiple
2081 headers of the same name were present, leading whitespace was being
2082 removed from all of them, but trailing whitespace was being removed only
2083 from the last one. Now trailing whitespace is removed from each header
2084 before concatenation. Completely empty headers in a concatenation (as
2085 before) are ignored.
2087 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
2088 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
2090 PH/29 [Removed. This was a change that I later backed out, and forgot to
2091 correct the ChangeLog entry (that I had efficiently created) before
2092 committing the later change.]
2094 PH/30 Exim was sometimes attempting to deliver messages that had suffered
2095 address errors (4xx response to RCPT) over the same connection as other
2096 messages routed to the same hosts. Such deliveries are always "forced",
2097 so retry times are not inspected. This resulted in far too many retries
2098 for the affected addresses. The effect occurred only when there were more
2099 hosts than the hosts_max_try setting in the smtp transport when it had
2100 the 4xx errors. Those hosts that it had tried were not added to the list
2101 of hosts for which the message was waiting, so if all were tried, there
2102 was no problem. Two fixes have been applied:
2104 (i) If there are any address or message errors in an SMTP delivery, none
2105 of the hosts (tried or untried) are now added to the list of hosts
2106 for which the message is waiting, so the message should not be a
2107 candidate for sending over the same connection that was used for a
2108 successful delivery of some other message. This seems entirely
2109 reasonable: after all the message is NOT "waiting for some host".
2110 This is so "obvious" that I'm not sure why it wasn't done
2111 previously. Hope I haven't missed anything, but it can't do any
2112 harm, as the worst effect is to miss an optimization.
2114 (ii) If, despite (i), such a delivery is accidentally attempted, the
2115 routing retry time is respected, so at least it doesn't keep
2116 hammering the server.
2118 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
2119 in ${readsocket because some servers need this prod.
2121 PH/32 Added some extra debug output when updating a wait-xxx database.
2123 PH/33 The hint "could be header name not terminated by colon", which has been
2124 given for certain expansion errors for a long time, was not being given
2125 for the ${if def:h_colon_omitted{... case.
2127 PH/34 The spec says: "With one important exception, whenever a domain list is
2128 being scanned, $domain contains the subject domain." There was at least
2129 one case where this was not true.
2131 PH/35 The error "getsockname() failed: connection reset by peer" was being
2132 written to the panic log as well as the main log, but it isn't really
2133 panic-worthy as it just means the connection died rather early on. I have
2134 removed the panic log writing for the ECONNRESET error when getsockname()
2137 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
2138 runs only) independently of the message's sender address. This meant
2139 that, if the 4xx error was in fact related to the sender, a different
2140 message to the same recipient with a different sender could confuse
2141 things. In particualar, this can happen when sending to a greylisting
2142 server, but other circumstances could also provoke similar problems.
2143 I have changed the default so that the retry time for these errors is now
2144 based a combination of the sender and recipient addresses. This change
2145 can be overridden by setting address_retry_include_sender=false in the
2148 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
2149 remote server are returned as part of bounce messages. This was not
2150 happening for LMTP over a pipe (the lmtp transport), but now it is the
2151 same for both kinds of LMTP.
2153 PH/38 Despite being documented as not happening, Exim was rewriting addresses
2154 in header lines that were in fact CNAMEs. This is no longer the case.
2156 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
2157 and queue runs started by the daemon processed all messages. This has
2158 been fixed so that -R and -S can now usefully be given with -q<time>.
2160 PH/40 Import PCRE release 6.7 (fixes some bugs).
2162 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
2164 PH/42 Give an error if -q is specified more than once.
2166 PH/43 Renamed the variables $interface_address and $interface_port as
2167 $received_ip_address and $received_port, to make it clear that these
2168 values apply to message reception, and not to the outgoing interface when
2169 a message is delivered. (The old names remain recognized, of course.)
2171 PH/44 There was no timeout on the connect() call when using a Unix domain
2172 socket in the ${readsocket expansion. There now is.
2174 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
2175 be meaningful with "accept".
2177 SC/01 Eximstats V1.43
2178 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
2180 SC/02 Eximstats V1.44
2181 Use a glob alias rather than an array ref in the generated
2182 parser. This improves both readability and performance.
2184 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
2185 Collect SpamAssassin and rejection statistics.
2186 Don't display local sender or destination tables unless
2187 there is data to show.
2188 Added average volumes into the top table text output.
2190 SC/04 Eximstats V1.46
2191 Collect data on the number of addresses (recipients)
2192 as well as the number of messages.
2194 SC/05 Eximstats V1.47
2195 Added 'Message too big' to the list of mail rejection
2196 reasons (thanks to Marco Gaiarin).
2198 SC/06 Eximstats V1.48
2199 Mainlog lines which have GMT offsets and are too short to
2200 have a flag are now skipped.
2202 SC/07 Eximstats V1.49 (Alain Williams)
2203 Added the -emptyok flag.
2205 SC/08 Eximstats V1.50
2206 Fixes for obtaining the IP address from reject messages.
2208 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
2209 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
2210 whitesspace changes from 4.64-PH/27
2212 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
2215 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
2216 are found, allow negative numbers in numeric criteria)
2218 JJ/06 exipick.20061117.2, added new $message_body_missing variable
2220 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
2221 to match changes made in 4.64-PH/43
2223 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
2225 PH/47 Put in an explicit test for a DNS lookup of an address record where the
2226 "domain" is actually an IP address, and force a failure. This locks out
2227 those revolvers/nameservers that support "A-for-A" lookups, in
2228 contravention of the specifications.
2230 PH/48 When a host name was looked up from an IP address, and the subsequent
2231 forward lookup of the name timed out, the host name was left in
2232 $sender_host_name, contrary to the specification.
2234 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
2235 restricted to single-key lookups, Exim was not diagnosing an error if
2236 * or *@ was used with a query-style lookup.
2238 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
2240 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
2241 long ago, but noone interested enough thought of it. Let's just say that
2242 the "1.1" means that there are some new functions that weren't there at
2243 some point in the past.
2245 PH/51 Error processing for expansion failure of helo_data from an smtp
2246 transport during callout processing was broken.
2248 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
2249 tested/used via the -bh/-bhc/-bs options.
2251 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
2252 bug, fixed in subsequent PCRE releases).
2254 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
2255 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
2257 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
2263 SC/01 Use a glob alias rather than an array ref in eximstats generated
2264 parser. This improves both readability and performance.
2266 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
2267 Don't display local sender or destination tables in eximstats unless
2268 there is data to show.
2269 Added average volumes into the eximstats top table text output.
2271 SC/03 Collect data on the number of addresses (recipients) as well
2272 as the number of messages in eximstats.
2274 TF/01 Correct an error in the documentation for the redirect router. Exim
2275 does (usually) call initgroups() when daemonizing.
2277 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
2278 with consistent privilege compared to when running as a daemon.
2280 TF/03 Note in the spec that $authenticated_id is not set for local
2281 submissions from trusted users.
2283 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
2284 Thanks to Dean Brooks <dean@iglou.com> for the patch.
2286 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
2287 by adding some example configuration directives to the default
2288 configuration file. A little bit of work is required to uncomment the
2289 directives and define how usernames and passwords are checked, but
2290 there is now a framework to start from.
2292 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
2293 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
2294 without this. I don't know how relevant this is to other LDAP libraries.
2296 PH/02 Add the verb name to the "unknown ACL verb" error.
2298 PH/03 Magnus Holmgren's patch for filter_prepend_home.
2300 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
2302 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
2303 directory not expanded when it should be if an expanded home directory
2304 was set for the address (which is overridden by the transport).
2306 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
2309 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
2310 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
2311 because it is too late at that time, and has no effect.
2313 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
2314 security issue with \' (bugzilla #107). I could not use the
2315 PQescapeStringConn() function, because it needs a PGconn value as one of
2318 PH/08 When testing addresses using -bt, indicate those final addresses that
2319 are duplicates that would not cause an additional delivery. At least one
2320 person was confused, thinking that -bt output corresponded to deliveries.
2321 (Suppressing duplicates isn't a good idea as you lose the information
2322 about possibly different redirections that led to the duplicates.)
2324 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
2325 systems where poll() doesn't work, in particular OS X.
2327 PH/10 Added more information to debugging output for retry time not reached.
2329 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
2330 operations in malware.c.
2332 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
2335 PH/13 If write_rejectlog was set false when logging was sent to syslog with
2336 syslog_duplication set false, log lines that would normally be written
2337 both the the main log and to the reject log were not written to syslog at
2340 PH/14 In the default configuration, change the use of "message" in ACL warn
2341 statements to "add_header".
2343 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
2344 not followed by a command (e.g. "seen endif").
2346 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
2347 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
2350 PH/17 Added extra conditions to the default value of delay_warning_condition
2354 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
2355 { match{$h_precedence:}{(?i)bulk|list|junk} } \
2356 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
2359 The Auto-Submitted: and various List- headers are standardised, whereas I
2360 don't think Precedence: ever was.
2362 PH/18 Refactored debugging code in route_finduser() to show more information,
2363 in particular, the error code if getpwnam() issues one.
2365 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
2366 This is apparently needed in addition to the PH/07 change above to avoid
2367 any possible encoding problems.
2369 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
2370 but not after initializing Perl.
2372 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
2373 output them only if debugging. By default they are written stderr,
2374 apparently, which is not desirable.
2376 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
2379 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
2382 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
2384 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
2385 authenticated or an ident call has been made. Suppress the default
2386 values for $authenticated_id and $authenticated_sender (but permit -oMai
2387 and -oMas) when testing with -bh.
2389 PH/24 Re-jigged the order of the tests in the default configuration so that the
2390 tests for valid domains and recipients precede the DNS black list and CSA
2391 tests, on the grounds that those ones are more expensive.
2393 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
2394 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
2395 This bug exists in every version of Exim that I still have, right back to
2398 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
2399 However, an attempt to turn on case-sensitivity in a regex key by
2400 including (?-i) didn't work because the subject string was already
2401 lowercased, and the effects were non-intuitive. It turns out that a
2402 one-line patch can be used to allow (?-i) to work as expected.
2408 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
2409 other effects) broke the use of negated acl sub-conditions.
2411 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
2414 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
2415 "Deny" causes Exim to reject the incoming connection with a 554 error.
2416 Unfortunately, if there is a major crisis, such as a disk failure,
2417 tcp-wrappers gives "deny", whereas what one would like would be some
2418 kind of temporary error. A kludge has been added to help with this.
2419 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
2420 554 error is used if errno is still zero or contains ENOENT (which occurs
2421 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
2424 PH/03 Add -lutil to the default FreeBSD LIBS setting.
2426 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
2427 errors. Otherwise a message that provokes a temporary error (when other
2428 messages do not) can cause a whole host to time out.
2430 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
2431 addresses were routed directly to files or pipes from a redirect router.
2432 File deliveries just didn't batch; pipe deliveries might have suffered
2435 PH/06 A failure to get a lock for a hints database would erroneously always say
2436 "Failed to get write lock", even when it was really a read lock.
2438 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
2439 of 0600. This has been changed to use the value of the lockfile_mode
2440 option (which defaults to 0600).
2442 PH/08 Applied small patch from the Sieve maintainer.
2444 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
2445 folder from quota calculations, a direct delivery into this folder messed
2446 up the contents of the maildirsize file. This was because the regex was
2447 used only to exclude .Trash (or whatever) when the size of the mailbox
2448 was calculated. There was no check that a delivery was happening into an
2449 excluded directory. This bug has been fixed by ignoring all quota
2450 processing for deliveries into excluded directories.
2452 PH/10 Added the maildirfolder_create_regex option to appendfile.
2458 PH/01 The code for finding all the local interface addresses on a FreeBSD
2459 system running IPv6 was broken. This may well have applied to all BSD
2460 systems, as well as to others that have similar system calls. The broken
2461 code found IPv4 interfaces correctly, but gave incorrect values for the
2462 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
2463 that it would not match correctly against @[] and not recognize the IPv6
2466 PH/02 The ipliteral router was not recognizing addresses of the form user@
2467 [ipv6:....] because it didn't know about the "ipv6:" prefix.
2469 PH/03 Added disable_ipv6.
2471 PH/04 Changed $reply_address to use the raw form of the headers instead of the
2472 decoded form, because it is most often used to construct To: headers
2473 lines in autoreplies, and the decoded form may well be syntactically
2474 invalid. However, $reply_address has leading white space removed, and all
2475 newlines turned into spaces so that the autoreply transport does not
2478 PH/05 If group was specified without a user on a router, and no group or user
2479 was specified on a transport, the group from the router was ignored.
2481 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
2482 for visible compile-time settings that can be used to change these
2483 numbers, for those that want even more. Backwards compatibility with old
2484 spool files has been maintained. However, going back to a previous Exim
2485 release will lost any variables that are in spool files.
2487 PH/07 Two small changes when running in the test harness: increase delay when
2488 passing a TCP/IP connection to a new process, in case the original
2489 process has to generate a bounce, and remove special handling of
2490 127.0.0.2 (sic), which is no longer necessary.
2492 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
2493 be the same on different OS.
2495 PH/09 Moved a debug statement in filter processing to avoid a race problem when
2498 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
2499 whether --show-vars was specified or not
2501 JJ/02 exipick: Added support for new ACL variable spool format introduced
2504 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
2505 syntactically invalid From: or Reply-to: line, and a filter used this to
2506 generate an autoreply, and therefore failed to obtain an address for the
2507 autoreply, Exim could try to deliver to a non-existent relative file
2508 name, causing unrelated and misleading errors. What now happens is that
2509 it logs this as a hard delivery error, but does not attempt to create a
2512 PH/11 The exinext utility has a -C option for testing purposes, but although
2513 the given file was scanned by exinext itself; it wasn't being passed on
2514 when Exim was called.
2516 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
2517 an end-of-file indication when reading a command response.
2519 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
2520 compiled. In many other places in Exim, IPv6 addresses are always
2521 recognized, so I have changed this. It also means that IPv4 domain
2522 literals of the form [IPV4:n.n.n.n] are now always recognized.
2524 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
2525 used if the router is not running as root, for example, when verifying at
2526 ACL time, or when using -bh. The debugging output from this situation was
2527 non-existent - all you got was a failure to exec. I have made two
2530 (a) Failures to set uid/gid, the current directory, or a process leader
2531 in a subprocess such as that created by queryprogram now generate
2532 suitable debugging ouput when -d is set.
2534 (b) The queryprogram router detects when it is not running as root,
2535 outputs suitable debugging information if -d is set, and then runs
2536 the subprocess without attempting to change uid/gid.
2538 PH/15 Minor change to Makefile for building test_host (undocumented testing
2541 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
2542 additional section of a DNS packet that returns MX or SRV records.
2543 Instead, it always explicitly searches for A/AAAA records. This avoids
2544 major problems that occur when a DNS server includes only records of one
2545 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
2546 fixed another bug: if SRV records were looked up and the corresponding
2547 address records were *not* found in the additional section, the port
2548 values from the SRV records were lost.
2550 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
2551 using the correct key (the original address) when searching the retry
2552 rules in order to find which one to use for generating the retry hint.
2554 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
2555 adding the default one. Similarly, if it contains a Reply-To: header, the
2556 errors_reply_to option, if set, is not used.
2558 PH/19 When calculating a retry time, Exim used to measure the "time since
2559 failure" by looking at the "first failed" field in the retry record. Now
2560 it does not use this if it is later than than the arrival time of the
2561 message. Instead it uses the arrival time. This makes for better
2562 behaviour in cases where some deliveries succeed, thus re-setting the
2563 "first failed" field. An example is a quota failure for a huge message
2564 when small messages continue to be delivered. Without this change, the
2565 "time since failure" will always be short, possible causing more frequent
2566 delivery attempts for the huge message than are intended.
2567 [Note: This change was subsequently modified - see PH/04 for 4.62.]
2569 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
2570 $1, $2, $3) because the numerical variables can be reset during some
2571 expansion items (e.g. "match"), thereby losing the authentication data.
2573 PH/21 Make -bV show the size of off_t variables so that the test suite can
2574 decide whether to run tests for quotas > 2G.
2576 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
2577 mailbox_size, and mailbox_filecount in the appendfile transport. If a
2578 filecount value is greater than 2G or if a quota value is greater than 2G
2579 on a system where the size of off_t is not greater than 4, a panic error
2582 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
2583 never match. The debug and -bh output now contains an explicit error
2584 message indicating a malformed IPv4 address or mask.
2586 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
2587 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
2588 PH/23 above applies.
2590 PH/25 Do not write to syslog when running in the test harness. The only
2591 occasion when this arises is a failure to open the main or panic logs
2592 (for which there is an explicit test).
2594 PH/26 Added the /no_tell option to "control=freeze".
2596 PH/27 If a host name lookup failed very early in a connection, for example, if
2597 the IP address matched host_lookup and the reverse lookup yielded a name
2598 that did not have a forward lookup, an error message of the form "no IP
2599 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
2600 could be logged. Now it outputs the IP address instead of "NULL".
2602 PH/28 An enabling patch from MH: add new function child_open_exim2() which
2603 allows the sender and the authenticated sender to be set when
2604 submitting a message from within Exim. Since child_open_exim() is
2605 documented for local_scan(), the new function should be too.
2607 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
2608 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
2609 results in an empty string is now treated as unset.
2611 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
2613 PH/31 Added sender_verify_fail logging option.
2615 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
2616 needed by RFC 822 but not by RFC 2822 was commented out. I have now
2617 tidied the source and removed it altogether.
2619 PH/33 When a queue run was abandoned because the load average was too high, a
2620 log line was always written; now it is written only if the queue_run log
2621 selector is set. In addition, the log line for abandonment now contains
2622 information about the queue run such as the pid. This is always present
2623 in "start" and "stop" lines but was omitted from the "abandon" line.
2625 PH/34 Omit spaces between a header name and the colon in the error message that
2626 is given when verify = headers_syntax fails (if there are lots of them,
2627 the message gets confusing).
2629 PH/35 Change the default for dns_check_names_pattern to allow slashes within
2630 names, as there are now some PTR records that contain slashes. This check
2631 is only to protect against broken name servers that fall over on strange
2632 characters, so the fact that it applies to all lookups doesn't matter.
2634 PH/36 Now that the new test suite is complete, we can remove some of the
2635 special code in Exim that was needed for the old test suite. For example,
2636 sorting DNS records because real resolvers return them in an arbitrary
2637 order. The new test suite's fake resolver always returns records in the
2640 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
2641 bounces) except when queue_only is set, to avoid logging races between
2642 the different processes.
2644 PH/38 Panic-die if .include specifies a non-absolute path.
2646 PH/39 A tweak to the "H" retry rule from its user.
2648 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
2649 a label. They prevented compilation on older perls.
2651 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
2652 a warning to be raised on newish perls.
2654 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
2655 on queue. Changes to match documented behaviour of showing count of
2656 messages matching specified criteria.
2658 PH/40 Changed the default ident timeout from 30s to 5s.
2660 PH/41 Added support for the use of login_cap features, on those BSD systems
2661 that have them, for controlling the resources used by pipe deliveries.
2663 PH/42 The content-scanning code uses fopen() to create files in which to put
2664 message data. Previously it was not paying any attention to the mode of
2665 the files. Exim runs with umask(0) because the rest of the code creates
2666 files with open(), and sets the required mode explicitly. Thus, these
2667 files were ending up world-writeable. This was not a big issue, because,
2668 being within the spool directory, they were not world-accessible. I have
2669 created a function called modefopen, which takes an additional mode
2670 argument. It sets umask(777), creates the file, chmods it to the required
2671 mode, then resets the umask. All the relevant calls to fopen() in the
2672 content scanning code have been changed to use this function.
2674 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
2675 to 24 hours. This avoids potential overflow problems when processing G
2676 and H retry rules. I suspect nobody ever tinkers with this value.
2678 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
2680 PH/45 When the plaintext authenticator is running as a client, the server's
2681 challenges are checked to ensure they are valid base64 strings. By
2682 default, the authentication attempt is cancelled if an invalid string is
2683 received. Setting client_ignore_invalid_base64 true ignores these errors.
2684 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
2685 they are received. Thus, the responses can be made to depend on the
2686 challenges. If an invalid string is ignored, an empty string is placed in
2689 PH/46 Messages that are created by the autoreply transport now contains a
2690 References: header, in accordance with RFCs 2822 and 3834.
2692 PH/47 Added authenticated_sender_force to the smtp transport.
2694 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
2696 PH/49 Installed latest patch from the Sieve maintainer.
2698 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
2699 was also set, the appendfile transport was unnecessarily scanning a
2700 directory of message files (e.g. for maildir delivery) to find the count
2701 of files (along with the size), even though it did not need this
2702 information. It now does the scan only if it needs to find either the
2703 size of the count of files.
2705 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
2707 PH/52 Two bugs concerned with error handling when the smtp transport is
2710 (i) Exim was not creating retry information for temporary errors given
2711 for individual recipients after the DATA command when the smtp transport
2712 was used in LMTP mode. This meant that they could be retried too
2713 frequently, and not timed out correctly.
2715 (ii) Exim was setting the flag that allows error details to be returned
2716 for LMTP errors on RCPT commands, but not for LMTP errors for individual
2717 recipients that were returned after the DATA command.
2719 PH/53 This is related to PH/52, but is more general: for any failing address,
2720 when detailed error information was permitted to be returned to the
2721 sender, but the error was temporary, then after the final timeout, only
2722 "retry timeout exceeded" was returned. Now it returns the full error as
2723 well as "retry timeout exceeded".
2725 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
2726 do this, and (what is worse) MTAs that accept it.
2728 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
2729 will now be deprecated.
2731 PH/56 New os.c-cygwin from the Cygwin maintainer.
2733 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
2734 formats (previously only available in exim formats via -bpr, -bpru,
2735 and -bpra. Now also available in native and exiqgrep formats)
2737 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
2738 with very large, slow to parse queues
2740 JJ/08 exipick: added ! as generic prefix to negate any criteria format
2742 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
2744 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
2745 responses to authentication challenges, though it was showing the
2746 challenges; (ii) I've removed the CR characters from the debug output for
2749 PH/58 Allow for the insertion of a newline as well as a space when a string
2750 is turned into more than one encoded-word during RFC 2047 encoding. The
2751 Sieve code now uses this.
2753 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
2754 data_4xx, lost_connection, tls_required.
2756 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
2757 message was being sent as an SMTP response.
2759 PH/61 Add -l and -k options to exicyclog.
2761 PH/62 When verifying, if an address was redirected to one new address, so that
2762 verification continued, and the new address failed or deferred after
2763 having set something in $address_data, the value of $address_data was not
2764 passed back to the ACL. This was different to the case when no
2765 redirection occurred. The value is now passed back in both cases.
2767 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
2768 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
2769 use login_cap.h, so on its own it isn't the distinguishing feature. The
2770 new name refers directly to the setclassresources() function.
2772 PH/65 Added configuration files for NetBSD3.
2774 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
2776 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
2777 is preferred over IPv4.
2779 PH/68 The bounce_return_message and bounce_return_body options were not being
2780 honoured for bounces generated during the reception of non-SMTP messages.
2781 In particular, this applied to messages rejected by the ACL. This bug has
2782 been fixed. However, if bounce_return_message is true and bounce_return_
2783 body is false, the headers that are returned for a non-SMTP message
2784 include only those that have been read before the error was detected.
2785 (In the case of an ACL rejection, they have all been read.)
2787 PH/69 The HTML version of the specification is now built in a directory called
2788 spec_html instead of spec.html, because the latter looks like a path with
2789 a MIME-type, and this confuses some software.
2791 PH/70 Catch two compiler warnings in sieve.c.
2793 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
2794 function verify_get_ident() calls ip_connect() to connect a socket, but
2795 if the "connect()" function timed out, ip_connect() used to close the
2796 socket. However, verify_get_ident() also closes the socket later, and in
2797 between Exim writes to the log, which may get opened at this point. When
2798 the socket was closed in ip_connect(), the log could get the same file
2799 descriptor number as the socket. This naturally causes chaos. The fix is
2800 not to close the socket in ip_connect(); the socket should be closed by
2801 the function that creates it. There was only one place in the code where
2802 this was missing, in the iplookup router, which I don't think anybody now
2803 uses, but I've fixed it anyway.
2805 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
2806 well as to direct DNS lookups. Otherwise the handling of names in host
2807 lists is inconsistent and therefore confusing.
2813 PH/01 Two changes to the default runtime configuration:
2815 (1) Move the checks for relay_from_hosts and authenticated clients from
2816 after to before the (commented out) DNS black list checks.
2818 (2) Add control=submission to the relay_from_hosts and authenticated
2819 clients checks, on the grounds that messages accepted by these
2820 statements are most likely to be submissions.
2822 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
2824 (1) Generate an error if the third argument for the ${prvs expansion is
2827 (2) Treat a missing third argument of ${prvscheck as if it were an empty
2830 (3) Reset the variables that are obtained from the first argument of
2831 ${prvscheck and used in the second argument before leaving the code,
2832 because their memory is reclaimed, so using them afterwards may do
2835 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
2836 one (it's much easier than Tom thought :-).
2838 (5) Because of (4), we can now allow for the use of $prvscheck_result
2839 inside the third argument.
2841 PH/03 For some reason, the default setting of PATH when running a command from
2842 a pipe transport was just "/usr/bin". I have changed it to
2845 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
2846 anything to be listed in the output from -bV.
2848 PH/05 When a filter generated an autoreply, the entire To: header line was
2849 quoted in the delivery log line, like this:
2851 => >A.N.Other <ano@some.domain> <original@ddress> ...
2853 This has been changed so that it extracts the operative address. There
2854 may be more than one such address. If so, they are comma-separated, like
2857 => >ano@some.domain,ona@other.domain <original@ddress> ...
2859 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
2860 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
2861 not being looked up in the rDNS to get a host name, Exim was showing the
2862 IP address twice in Received: lines, even though the IP addresses were
2863 identical. For example:
2865 Received: from [1.2.3.4] (helo=[1.2.3.4])
2867 However, if the real host name was known, it was omitting the HELO data
2868 if it matched the actual IP address. This has been tidied up so that it
2869 doesn't show the same IP address twice.
2871 PH/07 When both +timestamp and +memory debugging was on, the value given by
2872 $tod_xxx expansions could be wrong, because the tod_stamp() function was
2873 called by the debug printing, thereby overwriting the timestamp buffer.
2874 Debugging no longer uses the tod_stamp() function when +timestamp is set.
2876 PH/08 When the original message was included in an autoreply transport, it
2877 always said "this is a copy of the message, including all the headers",
2878 even if body_only or headers_only was set. It now gives an appropriate
2881 PH/09 Applied a patch from the Sieve maintainer which:
2883 o fixes some comments
2884 o adds the (disabled) notify extension core
2885 o adds some debug output for the result of if/elsif tests
2886 o points to the current vacation draft in the documentation
2887 and documents the missing references header update
2891 o fixes a bug in processing the envelope test (when testing
2892 multiple envelope elements, the last element determinted the
2895 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
2896 Electronic Mail") by including:
2898 Auto-submitted: auto-generated
2900 in the messages that it generates (bounce messages and others, such as
2901 warnings). In the case of bounce messages for non-SMTP mesages, there was
2902 also a typo: it was using "Auto_submitted" (underscore instead of
2903 hyphen). Since every message generated by Exim is necessarily in response
2904 to another message, thes have all been changed to:
2906 Auto-Submitted: auto-replied
2908 in accordance with these statements in the RFC:
2910 The auto-replied keyword:
2912 - SHOULD be used on messages sent in direct response to another
2913 message by an automatic process,
2915 - MUST NOT be used on manually-generated messages,
2917 - MAY be used on Delivery Status Notifications (DSNs) and Message
2918 Disposition Notifications (MDNs),
2920 - MUST NOT be used on messages generated by automatic or periodic
2921 processes, except for messages which are automatic responses to
2924 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
2925 to the default Received: header definition.
2927 PH/12 Added log selector acl_warn_skipped (default on).
2929 PH/13 After a successful wildlsearch lookup, discard the values of numeric
2930 variables because (a) they are in the wrong storage pool and (b) even if
2931 they were copied, it wouldn't work properly because of the caching.
2933 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
2934 checking when decoding. Apparently there are clients that generate
2935 overlong encoded strings. Why am I not surprised?
2937 PH/15 If the first argument of "${if match_address" was not empty, but did not
2938 contain an "@" character, Exim crashed. Now it writes a panic log message
2939 and treats the condition as false.
2941 PH/16 In autoreply, treat an empty string for "once" the same as unset.
2943 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
2944 extension "envelope-auth". The code is finished and in agreement with
2945 other implementations, but there is no documentation so far and in fact,
2946 nobody wrote the draft yet. This extension is currently #undef'ed, thus
2947 not changing the active code.
2949 Print executed "if" and "elsif" statements when debugging is used. This
2950 helps a great deal to understand what a filter does.
2952 Document more things not specified clearly in RFC3028. I had all this
2953 sorted out, when out of a sudden new issues came to my mind. Oops."
2955 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
2958 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
2959 canonical form (as documented). However, after a host name lookup from
2960 the IP address, check_host() was doing a simple string comparison with
2961 addresses acquired from the DNS when checking that the found name did
2962 have the original IP as one of its addresses. Since any found IPv6
2963 addresses are likely to be in abbreviated form, the comparison could
2964 fail. Luckily, there already exists a function for doing the comparison
2965 by converting both addresses to binary, so now that is used instead of
2966 the text comparison.
2968 PH/20 There was another similar case to PH/19, when a complete host name was
2969 given in a host list; looking up its IP address could give an abbreviated
2970 form, whereas the current host's name might or might not be abbreviated.
2971 The same fix has been applied.
2977 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
2978 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
2981 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
2982 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
2984 PH/03 Typo: missing ".o" in src/pcre/Makefile.
2986 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
2987 header line, restrict the check to what is listed in RFCs 2369 and 2929.
2988 Also, for "Auto-Submitted", treat anything other than "no" as
2989 non-personal, in accordance with RFC 3834. (Previously it treated
2990 anything starting "auto-" as non-personal.)
2992 TF/01 The control=submission/name=... option had a problem with syntax
2993 errors if the name included a slash character. The /name= option
2994 now slurps the rest of the string, so it can include any characters
2995 but it must come last in the list of options (after /sender_retain
2998 PH/05 Some modifications to the interface to the fake nameserver for the new
3006 TK/01 Added the "success_on_redirect" address verification option. See
3007 NewStuff for rationale and an example.
3009 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
3011 PH/02 Patch to exigrep to allow it to work on syslog lines.
3013 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
3014 fread() to skip over the body file's header line, because in Cygwin the
3015 header line is locked and is inaccessible.
3017 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
3018 co-exist for some time) to make it clear that it is the Exim ID that is
3019 referenced, not the Message-ID: header line.
3021 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
3022 string_format() function, because snprintf() does not exist on all
3025 PH/06 The use of forbid_filter_existstest now also locks out the use of the
3026 ${stat: expansion item.
3028 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
3029 protocol synchronization error", to keep the pedants happy.
3031 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
3032 well as for IRIX systems, when gcc is being used. See the host.c source
3035 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
3037 PH/10 Named domain lists were not working if used in a queue_smtp_domains
3040 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
3041 transport and to the smtp transport in LMTP mode.
3043 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
3045 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
3046 run a filter in a subprocess. This could lead to confusion in subsequent
3047 lookups in the parent process. There should also be a search_tidyup() at
3048 the end of the subprocess.
3050 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
3051 only if the host matched helo_try_verify_hosts, which caused the
3052 verification to occur when the EHLO/HELO command was issued. The ACL just
3053 tested the remembered result. Now, if a previous verification attempt has
3054 not happened, "verify = helo" does it there and then.
3056 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
3058 TK/03 Fix log output including CR from clamd.
3060 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
3061 exist provoked a memory error which could cause a segfault.
3063 PH/15 Installed PCRE 6.2
3065 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
3067 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
3068 of the problem. Specifically, suggested +O2 rather than +O1 for the
3071 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
3073 PH/20 If a delivery was routed to a non-standard port by means of an SRV
3074 record, the port was not correctly logged when the outgoing_port log
3075 selector was set (it logged the transort's default port).
3077 PH/21 Added support for host-specific ports to manualroute, queryprogram,
3078 fallback_hosts, and "hosts" in the smtp transport.
3080 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
3081 host errors such as "Connection refused".
3083 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
3084 authentication with radiusclient 0.4.9:
3086 - Error returned from rc_read_config was caught wrongly
3087 - Username/password not passed on to radius server due to wrong length.
3089 The presumption is that some radiusclient API changes for 4.51/PH/17
3090 were not taken care of correctly. The code is still untested by me (my
3091 Linux distribution still has 0.3.2 of radiusclient), but it was
3092 contributed by a Radius user.
3094 PH/24 When doing a callout, the value of $domain wasn't set correctly when
3095 expanding the "port" option of the smtp transport.
3097 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
3098 while reading a MIME header. Thanks to Tom Hughes for a patch.
3100 PH/24 Include config.h inside local_scan.h so that configuration settings are
3103 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
3104 that in an ACL for RCPT (for example), you can examine exactly what was
3107 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
3108 commands, but it was not correctly comparing the address with the actual
3109 client host address. Thus, it would show the EHLO address in Received:
3110 header lines when this was not necessary.
3112 PH/27 Added the % operator to ${eval:}.
3114 PH/28 Exim tries to create and chdir to its spool directory when it starts;
3115 it should be ignoring failures (because with -C, for example, it has lost
3116 privilege). It wasn't ignoring creation failures other than "already
3119 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
3122 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
3123 because an input line was too long, either on its own, or by virtue of
3124 too many continuations, the temporary file was not being removed, and the
3125 return code was incorrect.
3127 PH/31 Missing "BOOL" in function definition in filtertest.c.
3129 PH/32 Applied Sieve patches from the maintainer.
3131 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
3133 PH/33 Added "verify = not_blind".
3135 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
3136 Local/Makefile (with some defaults set). These are used in built scripts
3137 such as exicyclog, but they have never been used in the exim_install
3138 script (though there are many overriding facilities there). I have
3139 arranged that the exim_install script now takes note of these two
3142 PH/35 Installed configuration files for Dragonfly.
3144 PH/36 When a locally submitted message by a trusted user did not contain a
3145 From: header, and the sender address was obtained from -f or from an SMTP
3146 MAIL command, and the trusted user did not use -F to supply a sender
3147 name, $originator_name was incorrectly used when constructing a From:
3148 header. Furthermore, $originator_name was used for submission mode
3149 messages from external hosts without From: headers in a similar way,
3150 which is clearly wrong.
3152 PH/37 Added control=suppress_local_fixups.
3154 PH/38 When log_selector = +received_sender was set, and the addition of the
3155 sender made the log line's construction buffer exactly full, or one byte
3156 less than full, an overflow happened when the terminating "\n" was
3159 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
3160 when the result of a list match is failure because a DNS lookup failed.
3162 PH/40 RM_COMMAND is now used in the building process.
3164 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
3165 the "build-* directories that it finds.
3167 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
3168 address in a domain literal was a prefix of an interface address.
3170 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
3171 when verifying a sender address, unless rewrite_headers is false.
3173 PH/44 Wrote a long comment about why errors_to addresses are verified as
3174 recipients, not senders.
3176 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
3177 the ratelimit ACL was added.
3179 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
3181 PH/46 Added extra information about PostgreSQL errors to the error string.
3183 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
3184 avoiding the need to install special zones in a real server. This is
3185 backwards compatible; if it can't find the fake resolver, it drops back.
3186 Thus, both old and new test suites can be run.
3188 TF/02 Added util/ratelimit.pl
3190 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
3191 clock is set back in time.
3193 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
3194 Candler <B.Candler@pobox.com>.
3196 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
3197 OK for addresses that are the result of redirections.
3199 PH/48 A number of further additions for the benefit of the new test suite,
3200 including a fake gethostbyname() that interfaces to the fake DNS resolver
3203 TF/06 The fix for widen_domains has also been applied to qualify_single and
3204 search_parents which are the other dnslookup options that can cause
3207 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
3210 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
3212 TF/07 Exim produced the error message "an SRV record indicated no SMTP
3213 service" if it encountered an MX record with an empty target hostname.
3214 The message is now "an MX or SRV record indicated no SMTP service".
3216 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
3217 if the DNS of the sending site is misconfigured. This is quite a
3218 common situation. This change restores the behaviour of treating a
3219 helo verification defer as a failure.
3221 PH/51 If self=fail was set on a router, the bounce message did not include the
3222 actual error message.
3228 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
3230 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
3231 command itself ended in error, the underlying message about the transport
3232 filter timeout was being overwritten with the pipe command error. Now the
3233 underlying error message should be appended to the second error message.
3235 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
3237 PH/02 Reduce the amount of output that "make" produces by default. Full output
3238 can still be requested.
3240 PH/03 The warning log line about a condition test deferring for a "warn" verb
3241 was being output only once per connection, rather than after each
3242 occurrence (because it was using the same function as for successful
3243 "warn" verbs). This seems wrong, so I have changed it.
3245 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
3246 it should not have, which might have caused a crash in the right
3247 circumstances, but probably never did.
3249 PH/04 Installed a modified version of Tony Finch's patch to make submission
3250 mode fix the return path as well as the Sender: header line, and to
3251 add a /name= option so that you can make the user's friendly name appear
3254 TF/03 Added the control = fakedefer ACL modifier.
3256 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
3257 Mark Lowes for thorough testing.
3259 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
3261 TK/03 Merged latest SRS patch from Miles Wilton.
3263 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
3264 with the definition in sysexits.h (which is #included earlier).
3265 Fortunately, Exim does not actually use EX_OK. The code used to try to
3266 preserve the sysexits.h value, by assumimg that macro definitions were
3267 scanned for macro replacements. I have been disabused of this notion,
3268 so now the code just undefines EX_OK before #including unistd.h.
3270 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
3271 in the smtp transport. When a block could not be written in a single
3272 write() function, the timeout was being re-applied to each part-write.
3273 This seems wrong - if the receiver was accepting one byte at a time it
3274 would take for ever. The timeout is now adjusted when this happens. It
3275 doesn't have to be particularly precise.
3277 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
3278 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
3280 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
3281 without a domain if the check to <postmaster@domain> fails.
3283 SC/01 Eximstats: added -xls and the ability to specify output files
3284 (patch written by Frank Heydlauf).
3286 SC/02 Eximstats: use FileHandles for outputing results.
3288 SC/03 Eximstats: allow any combination of xls, txt, and html output.
3290 SC/04 Eximstats: fixed display of large numbers with -nvr option
3292 SC/05 Eximstats: fixed merging of reports with empty tables.
3294 SC/06 Eximstats: added the -include_original_destination flag
3296 SC/07 Eximstats: removed tabs and trailing whitespace.
3298 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
3300 TK/06 MBOX spool code: Add real "From " MBOX separator line
3301 so the .eml file is really in mbox format (even though
3302 most programs do not really care). Patch from Alex Miller.
3304 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
3305 The latter is generated from $received_to and is only set if the
3306 message has one envelope recipient. SA can use these headers,
3307 obviously out-of-the-box. Patch from Alex Miller.
3309 PH/08 The ${def test on a variable was returning false if the variable's
3310 value was "0", contrary to what the specification has always said!
3311 The result should be true unless the variable is empty.
3313 PH/09 The syntax error of a character other than { following "${if
3314 def:variable_name" (after optional whitespace) was not being diagnosed.
3315 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
3316 accidental colon was present, for example, could give incorrect results.
3318 PH/10 Tidied the code in a number of places where the st_size field of a stat()
3319 result is used (not including appendfile, where other changes are about
3322 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
3323 This involved changing a lot of size variables from int to off_t. It
3324 should work with maildirs and everything.
3326 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
3327 spamd dying while we are connected to it.
3329 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
3332 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
3333 function for BDB 4.3.
3335 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
3337 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
3338 the sources for PCRE have been re-arranged and more files are now
3341 PH/15 The code I had for printing potentially long long variables in PH/11
3342 above was not the best (it lost precision). The length of off_t variables
3343 is now inspected at build time, and an appropriate printing format (%ld
3344 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
3345 to be "long long int" or "long int". This is needed for the internal
3346 formatting function string_vformat().
3348 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
3349 the configuration file to be ":syslog", then the script "guesses" where
3350 the logs files are, rather than using the compiled in default. In our
3351 case the guess is not the same as the compiled default, so the script
3352 suddenly stopped working when I started to use syslog. The patch checks
3353 to see if log_file_path is "". If so, it attempts to read it from exim
3354 with no configuration file to get the compiled in version, before it
3355 falls back to the previous guessing code."
3357 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
3358 implementing BATV in an Exim configuration. See NewStuff for the gory
3361 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
3362 Makefile that are specific to HP-UX.
3364 PH/18 If the "use_postmaster" option was set for a recipient callout together
3365 with the "random" option, the postmaster address was used as the MAIL
3366 FROM address for the random test, but not for the subsequent recipient
3367 test. It is now used for both.
3369 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
3370 patch removes a few documentation additions to RFC 3028, because the
3371 latest draft now contains them. It adds the new en;ascii-case comparator
3372 and a new error check for 8bit text in MIME parts. Comparator and
3373 require names are now matched exactly. I enabled the subaddress
3374 extension, but it is not well tested yet (read: it works for me)."
3376 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
3377 rework some of the code of TK/09 above to avoid the hardwired use of
3378 "%lld" and "long long". Replaced the call to snprintf() with a call to
3381 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
3382 records point to non-existent hosts", "retry timeout exceeded", and
3383 "retry time not reached for any host after a long failure period".
3385 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
3386 experimental DomainKeys support:
3388 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
3389 (2) On an error such as an illegally used "control", the wrong name for
3390 the control was given.
3392 These problems did NOT occur unless DomainKeys support was compiled.
3394 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
3396 PH/24 Added ${if match_ip condition.
3398 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
3399 databases so that it will be absolutely obvious if a crash occurs in the
3400 DB library. This is a regular occurrence (often caused by mis-matched
3403 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
3404 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
3405 picked up on a user's system that detects such things. There doesn't seem
3406 to be a gcc warning option for this - only an attribute that has to be
3407 put on the function's prototype. It seems that in Fedora Core 4 they have
3408 set this on a number of new functions. No doubt there will be more in due
3411 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
3412 specify a transport. However, if an address that was verified by such a
3413 router was the subject of a callout, Exim crashed because it tried to
3414 read the rcpt_include_affixes from the non-existent transport. Now it
3415 just assumes that the setting of that option is false. This bug was
3416 introduced by 4.51/PH/31.
3418 PH/28 Changed -d+all to exclude +memory, because that information is very
3419 rarely of interest, but it makes the output a lot bigger. People tend to
3420 do -d+all out of habit.
3422 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
3423 code in os-type was giving problems when libc.so lives in lib64, like on
3426 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
3427 aren't the modern standard, and it seems that some systems' include files
3428 don't always have them. Exim was already checking for some of the newer
3429 ones like T_AAAA, and defining it itself. I've added checks for all the
3430 record types that Exim uses.
3432 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
3433 not automatically generating a new one, as it is supposed to. This
3434 prevented TLS from working. If the file did exist, but contained invalid
3435 data, a new version was generated, as expected. It was only the case of a
3436 non-existent file that was broken.
3438 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
3439 with a change in libdomainkeys > 0.64.
3441 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
3442 from DNS. If the selector record carries the flag, it now has
3443 precedence over the domain-wide flag.
3445 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
3447 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
3448 the use of a port name that isn't defined in /etc/services) occurred, the
3449 message was deferred as in a normal delivery, and thus remained on the
3450 spool, instead of being failed because of the mua_wrapper setting. This
3451 is now fixed, and I tidied up some of the mua_wrapper messages at the
3454 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
3455 the messages in a hash of arrays rather than using individual hashes.
3456 This is a bit cleaner and results in dramatic memory savings, albeit
3457 at a slight CPU cost.
3459 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
3460 as requested by Marc Sherman.
3462 SC/10 Eximstats: added histograms for user specified patterns as requested
3465 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
3467 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
3468 fopen() in the content-scanning modules that did not already have it.
3474 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
3475 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
3477 TK/02 Fix ACL "control" statement not being available in MIME ACL.
3479 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
3481 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
3482 to test Sieve filters that use "vacation".
3484 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
3485 that changes the way the GnuTLS parameters are stored in the cache file.
3486 The new format can be generated externally. For backward compatibility,
3487 if the data in the cache doesn't make sense, Exim assumes it has read an
3488 old-format file, and it generates new data and writes a new file. This
3489 means that you can't go back to an older release without removing the
3492 PH/03 A redirect router that has both "unseen" and "one_time" set does not
3493 work if there are any delivery delays because "one_time" forces the
3494 parent to be marked "delivered", so its unseen clone is never tried
3495 again. For this reason, Exim now forbids the simultaneous setting of
3498 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
3499 redirected to themselves ("homonym" addresses). Read the long ChangeLog
3500 entry if you want to know the details. The fix, however, neglected to
3501 consider the case when local delivery batching is involved. The test for
3502 "previously delivered" was not happening when checking to see if an
3503 address could be batched with a previous (undelivered) one; under
3504 certain circumstances this could lead to multiple deliveries to the same
3507 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
3508 in its include files, and this causes problems building Exim.
3510 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
3511 header_syntax) but Exim was just ignoring anything given after a slash.
3512 In particular, this caused confusion with an attempt to use "verify =
3513 reverse_host_lookup/defer_ok". An error is now given when options are
3514 supplied for verify items that do not have them. (Maybe reverse_host_
3515 lookup should have a defer_ok option, but that's a different point.)
3517 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
3518 defined by RFC 821) to 2048, because there were problems with some AUTH
3519 commands, and RFC 1869 says the size should be increased for extended
3520 SMTP commands that take arguments.
3522 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
3525 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
3526 "unknown" error; now it says that the functionality isn't in the binary.
3528 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
3529 an address' error message when a string expansion fails (syntax or
3530 whatever). Otherwise the password may appear in the log. Following change
3531 PH/42 below, there is no longer a chance of it appearing in a bounce
3534 PH/11 Installed exipick version 20050225.0 from John Jetmore.
3536 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
3537 of its addresses was ever tried. (Bugzilla bug #2.)
3539 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
3540 the result incorrectly in the debug output. (It correctly added a newline
3541 to what was transported.)
3543 TF/01 Added $received_time.
3545 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
3546 commented out examples of how to interface to a virus scanner and to
3547 SpamAssassin. Also added commented examples of av_scanner and
3548 spamd_address settings.
3550 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
3551 and controls are allowed in which ACLs. There were a couple of minor
3552 errors. Some of the entries in the conditions table (which is a table of
3553 where they are NOT allowed) were getting very unwieldy; rewrote them as a
3554 negation of where the condition IS allowed.
3556 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
3558 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
3559 header file does not have a version number, so I've had to invent a new
3560 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
3561 API. The code is untested by me (my Linux distribution still has 0.3.2 of
3562 radiusclient), but it was contributed by a Radius user.
3564 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
3565 files or directories, for OpenSSL.
3567 PH/19 When an Exim process that is running as root has to create an Exim log
3568 file, it does so in a subprocess that runs as exim:exim so as to get the
3569 ownership right at creation (otherwise, other Exim processes might see
3570 the file with the wrong ownership). There was no test for failure of this
3571 fork() call, which would lead to the process getting stuck as it waited
3572 for a non-existent subprocess. Forks do occasionally fail when resources
3573 run out. I reviewed all the other calls to fork(); they all seem to check
3576 PH/20 When checking for unexpected SMTP input at connect time (before writing
3577 the banner), Exim was not dealing correctly with a non-positive return
3578 from the read() function. If the client had disconnected by this time,
3579 the result was a log entry for a synchronization error with an empty
3580 string after "input=" when read() returned zero. If read() returned -1
3581 (an event I could not check), uninitialized data bytes were printed.
3582 There were reports of junk text (parts of files, etc) appearing after
3585 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
3587 PH/22 Added support for macro redefinition, and (re)definition in between
3588 driver and ACL definitions.
3590 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
3591 forgetting to use the resulting value; it was using the unexpanded value.
3593 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
3594 hadn't been configured. The fix is from Juergen Kreileder, who
3595 understands it better than I do:
3597 "Here's what I see happening with three configured cyrus_sasl
3598 authenticators configured (plain, login, cram-md5):
3600 On startup auth_cyrus_sasl_init() gets called for each of these.
3601 This means three calls to sasl_listmech() without a specified mech_list.
3602 => SASL tests which mechs of all available mechs actually work
3603 => three warnings about OTP not working
3604 => the returned list contains: plain, login, cram-md5, digest-md5, ...
3606 With the patch, sasl_listmech() also gets called three times. But now
3607 SASL's mech_list option is set to the server_mech specified in the the
3608 authenticator. Or in other words, the answer from sasl_listmech()
3609 gets limited to just the mech you're testing for (which is different
3611 => the return list contains just 'plain' or 'login', 'cram-md5' or
3612 nothing depending on the value of ob->server_mech.
3614 I've just tested the patch: Authentication still works fine,
3615 unavailable mechs specified in the exim configuration are still
3616 caught, and the auth.log warnings about OTP are gone."
3618 PH/25 When debugging is enabled, the contents of the command line are added
3619 to the debugging output, even when log_selector=+arguments is not
3622 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
3623 answer is "GNU", and only if the return is "GNU/something" is the answer
3626 PH/27 $acl_verify_message is now set immediately after the failure of a
3627 verification in an ACL, and so is available in subsequent modifiers. In
3628 particular, the message can be preserved by coding like this:
3630 warn !verify = sender
3631 set acl_m0 = $acl_verify_message
3633 Previously, $acl_verify_message was set only while expanding "message"
3634 and "log_message" when a very denied access.
3636 PH/28 Modified OS/os.c-Linux with
3638 -#ifndef OS_LOAD_AVERAGE
3639 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
3641 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
3642 nomenclature these days.)
3644 PH/29 Installed patch from the Sieve maintainer that adds the options
3645 sieve_useraddress and sieve_subaddress to the redirect router.
3647 PH/30 In these circumstances:
3648 . Two addresses routed to the same list of hosts;
3649 . First host does not offer TLS;
3650 . First host accepts first address;
3651 . First host gives temporary error to second address;
3652 . Second host offers TLS and a TLS session is established;
3653 . Second host accepts second address.
3654 Exim incorrectly logged both deliveries with the TLS parameters (cipher
3655 and peerdn, if requested) that were in fact used only for the second
3658 PH/31 When doing a callout as part of verifying an address, Exim was not paying
3659 attention to any local part prefix or suffix that was matched by the
3660 router that accepted the address. It now behaves in the same way as it
3661 does for delivery: the affixes are removed from the local part unless
3662 rcpt_include_affixes is set on the transport.
3664 PH/32 Add the sender address, as F=<...>, to the log line when logging a
3665 timeout during the DATA phase of an incoming message.
3667 PH/33 Sieve envelope tests were broken for match types other than :is. I have
3668 applied a patch sanctioned by the Sieve maintainer.
3670 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
3671 the uid or gid is negative. A case of a negative gid caused this to be
3672 noticed. The fix allows for either to be negative.
3674 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
3675 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
3678 PH/36 The change PH/12 above was broken. Fixed it.
3680 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
3681 the grounds that routing the same address twice would always produce the
3682 same answer. This might have been true once, but it is certainly no
3683 longer true now. Routing a child address may depend on the previous
3684 routing that produced that child. Some complicated redirection strategies
3685 went wrong when messages had multiple recipients, and made Exim's
3686 behaviour dependent on the order in which the addresses were given.
3688 I have moved the duplicate checking until after the routing is complete.
3689 Exim scans the addresses that are assigned to local and remote
3690 transports, and removes any duplicates. This means that more work will be
3691 done, as duplicates will always all be routed, but duplicates are
3692 presumably rare, so I don't expect this is of any significance.
3694 For deliveries to pipes, files, and autoreplies, the duplicate checking
3695 still happens during the routing process, since they are not going to be
3698 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
3699 It corrects a timeout issue with spamd. This is Ian's comment: "The
3700 background is that sometimes spamd either never reads data from a
3701 connection it has accepted, or it never writes response data. The exiscan
3702 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
3703 blindly assumes that writes won't block so it may never time out."
3705 PH/39 Allow G after quota size as well as K and M.
3707 PH/40 The value set for $authenticated_id in an authenticator may not contain
3708 binary zeroes or newlines because the value is written to log lines and
3709 to spool files. There was no check on this. Now the value is run through
3710 the string_printing() function so that such characters are converted to
3711 printable escape sequences.
3713 PH/41 $message_linecount is a new variable that contains the total number of
3714 lines in the message. Compare $body_linecount, which is the count for the
3717 PH/42 Exim no longer gives details of delivery errors for specific addresses in
3718 bounce and delay warning messages, except in certain special cases, which
3721 (a) An SMTP error message from a remote host;
3722 (b) A message specified in a :fail: redirection;
3723 (c) A message specified in a "fail" command in a system filter;
3724 (d) A message specified in a FAIL return from the queryprogram router;
3725 (e) A message specified by the cannot_route_message router option.
3727 In these cases only, Exim does include the error details in bounce and
3728 warning messages. There are also a few cases where bland messages such
3729 as "unrouteable address" or "local delivery error" are given.
3731 PH/43 $value is now also set for the "else" part of a ${run expansion.
3733 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
3734 being worked on, but at least Exim now implements the latest version to
3737 PH/45 In a pipe transport, although a timeout while waiting for the pipe
3738 process to complete was treated as a delivery failure, a timeout while
3739 writing the message to the pipe was logged, but erroneously treated as a
3740 successful delivery. Such timeouts include transport filter timeouts. For
3741 consistency with the overall process timeout, these timeouts are now
3742 treated as errors, giving rise to delivery failures by default. However,
3743 there is now a new Boolean option for the pipe transport called
3744 timeout_defer, which, if set TRUE, converts the failures into defers for
3745 both kinds of timeout. A transport filter timeout is now identified in
3748 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
3749 systems where "make" and "gmake" are different, calling "gmake" at top
3750 level broke things. I've arranged for the value of $(MAKE) to be passed
3751 from the Makefile to this script so that it can call the same version of
3755 A note about Exim versions 4.44 and 4.50
3756 ----------------------------------------
3758 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
3759 changes of various kinds. As a consequence, a big documentation update was
3760 needed. This delayed the release for rather longer than seemed good, especially
3761 in the light of a couple of (minor) security issues. Therefore, the changes
3762 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
3763 release. So 4.44 and 4.50 are in effect two different branches that both start
3766 I have left the 4.50 change log unchanged; it contains all the changes since
3767 4.43. The change log for 4.44 is below; many of its items are identical to
3768 those for 4.50. This seems to be the most sensible way to preserve the
3769 historical information.
3775 1. Minor wording change to the doc/README.SIEVE file.
3777 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
3778 computation of the current number of files was incorrect.
3780 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
3781 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
3784 4. Give more explanation in the error message when the command for a transport
3785 filter fails to execute.
3787 5. There are several places where Exim runs a non-Exim command in a
3788 subprocess. The SIGUSR1 signal should be disabled for these processes. This
3789 was being done only for the command run by the queryprogram router. It is
3790 now done for all such subprocesses. The other cases are: ${run, transport
3791 filters, and the commands run by the lmtp and pipe transports.
3793 6. Added CONFIGURE_GROUP build-time option.
3795 7. Some older OS have a limit of 256 on the maximum number of file
3796 descriptors. Exim was using setrlimit() to set 1000 as a large value
3797 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
3798 systems. I've change it so that if it can't get 1000, it tries for 256.
3800 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
3801 was an oversight, and furthermore, ever since the addition of extra
3802 controls (e.g. 4.43/32), the checks on when to allow different forms of
3803 "control" were broken. There should now be diagnostics for all cases when a
3804 control that does not make sense is encountered.
3806 9. Added the /retain_sender option to "control=submission".
3808 10. $recipients is now available in the predata ACL (oversight).
3810 11. Tidy the search cache before the fork to do a delivery from a message
3811 received from the command line. Otherwise the child will trigger a lookup
3812 failure and thereby defer the delivery if it tries to use (for example) a
3813 cached ldap connection that the parent has called unbind on.
3815 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
3816 of $address_data from the recipient verification was clobbered by the
3817 sender verification.
3819 13. The value of address_data from a sender verification is now available in
3820 $sender_address_data in subsequent conditions in the ACL statement.
3822 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
3824 15. Added a new option "connect=<time>" to callout options, to set a different
3827 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
3828 was its contents. (It was OK if the option was not defined at all.)
3830 17. A "Completed" log line is now written for messages that are removed from
3831 the spool by the -Mrm option.
3833 18. New variables $sender_verify_failure and $recipient_verify_failure contain
3834 information about exactly what failed.
3836 19. Added -dd to debug only the daemon process.
3838 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
3839 handles timeouts, both on the server side and network timeouts. Renamed the
3840 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
3842 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
3843 It is now set to "smtps".
3845 22. $host_address is now set to the target address during the checking of
3846 ignore_target_hosts.
3848 23. When checking ignore_target_hosts for an ipliteral router, no host name was
3849 being passed; this would have caused $sender_host_name to have been used if
3850 matching the list had actually called for a host name (not very likely,
3851 since this list is usually IP addresses). A host name is now passed as
3854 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
3855 code that specifies a non-restarting handler (typically sigaction() in
3856 modern systems) in an attempt to fix a rare and obscure crash bug.
3858 25. Narrowed the window for a race in the daemon that could cause it to ignore
3859 SIGCHLD signals. This is not a major problem, because they are used only to
3860 wake it up if nothing else does.
3862 26. A malformed maildirsize file could cause Exim to calculate negative values
3863 for the mailbox size or file count. Odd effects could occur as a result.
3864 The maildirsize information is now recalculated if the size or filecount
3867 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
3868 support for a long time. Removed HAVE_SYS_VFS_H.
3870 28. Installed the latest version of exipick from John Jetmore.
3872 29. In an address list, if the pattern was not a regular expression, an empty
3873 subject address (from a bounce message) matched only if the pattern was an
3874 empty string. Non-empty patterns were not even tested. This was the wrong
3875 because it is perfectly reasonable to use an empty address as part of a
3876 database query. An empty address is now tested by patterns that are
3877 lookups. However, all the other forms of pattern expect the subject to
3878 contain a local part and a domain, and therefore, for them, an empty
3879 address still always fails if the pattern is not itself empty.
3881 30. Exim went into a mad DNS loop when attempting to do a callout where the
3882 host was specified on an smtp transport, and looking it up yielded more
3883 than one IP address.
3885 31. Re-factored the code for checking spool and log partition space into a
3886 function that finds that data and another that does the check. The former
3887 is then used to implement four new variables: $spool_space, $log_space,
3888 $spool_inodes, and $log_inodes.
3890 32. The RFC2047 encoding function was originally intended for short strings
3891 such as real names; it was not keeping to the 75-character limit for
3892 encoded words that the RFC imposes. It now respects the limit, and
3893 generates multiple encoded words if necessary. To be on the safe side, I
3894 have increased the buffer size for the ${rfc2047: expansion operator from
3897 33. It is now permitted to omit both strings after an "if" condition; if the
3898 condition is true, the result is "true". As before, when the second string
3899 is omitted, a false condition yields an empty string. This makes it less
3900 cumbersome to write custom ACL and router conditions.
3902 34. Failure to deliver a bounce message always caused it to be frozen, even if
3903 there was an errors_to setting on the router. The errors_to setting is now
3906 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
3907 canonical form (fully expanded) before being placed in
3908 $sender_host_address.
3910 36. The table in the code that translates DNS record types into text (T_A to
3911 "A" for instance) was missing entries for NS and CNAME. It is just possible
3912 that this could have caused confusion if both these types were looked up
3913 for the same domain, because the text type is used as part of Exim's
3914 per-process caching. But the chance of anyone hitting this buglet seems
3917 37. The dnsdb lookup has been extended in a number of ways.
3919 (1) There is a new type, "zns", which walks up the domain tree until it
3920 finds some nameserver records. It should be used with care.
3922 (2) There is a new type, "mxh", which is like "mx" except that it returns
3923 just the host names, not the priorities.
3925 (3) It is now possible to give a list of domains (or IP addresses) to be
3926 looked up. The behaviour when one of the lookups defers can be
3927 controlled by a keyword.
3929 (4) It is now possible to specify the separator character for use when
3930 multiple records are returned.
3932 38. The dnslists ACL condition has been extended: it is now possible to supply
3933 a list of IP addresses and/or domains to be looked up in a particular DNS
3936 39. Added log_selector=+queue_time_overall.
3938 40. When running the queue in the test harness, wait just a tad after forking a
3939 delivery process, to get repeatability of debugging output.
3941 41. Include certificate and key file names in error message when GnuTLS fails
3942 to set them up, because the GnuTLS error message doesn't include the name
3943 of the failing file when there is a problem reading it.
3945 42. Allow both -bf and -bF in the same test run.
3947 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
3949 44. The "Exiscan patch" is now merged into the mainline Exim source.
3951 45. Sometimes the final signoff response after QUIT could fail to get
3952 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
3953 < 0 before doing a fflush(). This bug looks as though it goes back to the
3954 introduction of TLS in release 3.20, but "sometimes" must have been rare
3955 because the tests only now provoked it.
3957 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
3958 (this can affect the format of dates).
3960 47. exim_tidydb, when checking for the continued existence of a message for
3961 which it has found a message-specific retry record, was not finding
3962 messages that were in split spool directories. Consequently, it was
3963 deleting retry records that should have stayed in existence.
3965 48. Steve fixed some bugs in eximstats.
3967 49. The SPA authentication driver was not abandoning authentication and moving
3968 on to the next authenticator when an expansion was forced to fail,
3969 contradicting the general specification for all authenticators. Instead it
3970 was generating a temporary error. It now behaves as specified.
3972 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
3973 (the order specifies the preference for clients). The order is now AES256,
3974 AES128, 3DES, ARCFOUR128.
3976 51. Small patch to Sieve code - explicitly set From: when generating an
3979 52. Exim crashed if a remote delivery caused a very long error message to be
3980 recorded - for instance if somebody sent an entire SpamAssassin report back
3981 as a large number of 550 error lines. This bug was coincidentally fixed by
3982 increasing the size of one of Exim's internal buffers (big_buffer) that
3983 happened as part of the Exiscan merge. However, to be on the safe side, I
3984 have made the code more robust (and fixed the comments that describe what
3987 53. Now that there can be additional text after "Completed" in log lines (if
3988 the queue_time_overall log selector is set), a one-byte patch to exigrep
3989 was needed to allow it to recognize "Completed" as not the last thing in
3992 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
3993 patch that reportedly fixes this has been added. I am not expert enough to
3994 create a test for it. This is what the patch creator wrote:
3996 "I found a little strange behaviour of ldap code when working with
3997 Windows 2003 AD Domain, where users was placed in more than one
3998 Organization Units. When I tried to give exim partial DN, the exit code
3999 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
4000 But simultaneously result of request was absolutely normal ldap result,
4001 so I produce this patch..."
4003 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
4004 REFERENCE, so I have modified the code to exclude the patch when that macro
4007 55. Some experimental protocols are using DNS PTR records for new purposes. The
4008 keys for these records are domain names, not reversed IP addresses. The
4009 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4010 leaves it alone. Component reversal etc. now happens only for IP addresses.
4013 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
4015 57. Double the size of the debug message buffer (to 2048) so that more of very
4016 long debug lines gets shown.
4018 58. The exicyclog utility now does better if the number of log files to keep
4019 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
4021 59. Two changes related to the smtp_active_hostname option:
4023 (1) $smtp_active_hostname is now available as a variable.
4024 (2) The default for smtp_banner uses $smtp_active_hostname instead
4025 of $primary_hostname.
4027 60. The host_aton() function is supposed to be passed a string that is known
4028 to be a valid IP address. However, in the case of IPv6 addresses, it was
4029 not checking this. This is a hostage to fortune. Exim now panics and dies
4030 if the condition is not met. A case was found where this could be provoked
4031 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4032 components; fortuitously, this particular loophole had already been fixed
4033 by change 4.50/55 above.
4035 If there are any other similar loopholes, the new check in host_aton()
4036 itself should stop them being exploited. The report I received stated that
4037 data on the command line could provoke the exploit when Exim was running as
4038 exim, but did not say which command line option was involved. All I could
4039 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4040 running as the user.
4043 61. There was a buffer overflow vulnerability in the SPA authentication code
4044 (which came originally from the Samba project). I have added a test to the
4045 spa_base64_to_bits() function which I hope fixes it.
4048 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
4049 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
4051 63. The daemon start-up calls getloadavg() while still root for those OS that
4052 need the first call to be done as root, but it missed one case: when
4053 deliver_queue_load_max is set with deliver_drop_privilege. This is
4054 necessary for the benefit of the queue runner, because there is no re-exec
4055 when deliver_drop_privilege is set.
4057 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
4058 This has been fixed.
4060 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4061 was in use, was not putting the data itself into the right store pool;
4062 consequently, it could be overwritten for a subsequent message in the same
4063 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4066 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
4068 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
4069 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
4070 the function were treating the return as a boolean value, which happened to
4071 work because 0=false and not-0=true, but is not correct code.
4073 68. The host_aton() function was not handling scoped IPv6 addresses (those
4074 with, for example, "%eth0" on the end) correctly.
4076 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
4077 negated items (that is, ~something) in unsigned ints. Some compilers
4078 apparently mutter when there is no cast.
4080 70. If an address verification called from an ACL failed, and did not produce a
4081 user-specific message (i.e. there was only a "system" message), nothing was
4082 put in $acl_verify_message. In this situation, it now puts the system
4085 71. Change 4.23/11 added synchronization checking at the start of an SMTP
4086 session; change 4.31/43 added the unwanted input to the log line - except
4087 that it did not do this in the start of session case. It now does.
4089 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
4090 This is wrong and can cause the other end to generate a synchronization
4091 error if it is another Exim or anything else that does the synchronization
4092 check. A QUIT command is no longer sent after a timeout.
4094 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
4095 during host lookups.
4097 74. The defer_ok option of callout verification was not working if it was used
4098 when verifying addresses in header lines, that is, for this case:
4100 verify = header_sender/callout=defer_ok
4102 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
4103 those file descriptors could be used for SMTP connections. If anything
4104 wrote to stderr (the example that came up was "warn" in embedded Perl), it
4105 could be sent to the SMTP client, causing chaos. The daemon now opens
4106 stdin, stdout, and stderr to /dev/null when it puts itself into the
4109 76. Arrange for output from Perl's "warn" command to be written to Exim's main
4110 log by default. The user can override this with suitable Perl magic.
4112 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
4113 the log message when discard triggers, was not working for the DATA ACL or
4114 for the non-SMTP ACL.
4116 78. Error message wording change in sieve.c.
4118 79. If smtp_accept_max_per_host was set, the number of connections could be
4119 restricted to fewer than expected, because the daemon was trying to set up
4120 a new connection before checking whether the processes handling previous
4121 connections had finished. The check for completed processes is now done
4122 earlier. On busy systems, this bug wouldn't be noticed because something
4123 else would have woken the daemon, and it would have reaped the completed
4126 80. If a message was submitted locally by a user whose login name contained one
4127 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
4128 It caused a spool format error. I have fixed the spool reading code. A
4129 related problem was that the "from" clause in the Received: line became
4130 illegal because of the space(s). It is now covered by ${quote_local_part.
4132 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
4135 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
4137 83. Patch from Sieve maintainer to fix unterminated string problem in
4138 "vacation" handling.
4140 84. Some minor changes to the Linux configuration files to help with other
4141 OS variants using glibc.
4143 85. One more patch for Sieve to update vacation handling to latest spec.
4146 ----------------------------------------------------
4147 See the note above about the 4.44 and 4.50 releases.
4148 ----------------------------------------------------
4154 1. Change 4.43/35 introduced a bug that caused file counts to be
4155 incorrectly computed when quota_filecount was set in an appendfile
4158 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4159 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4162 3. Give more explanation in the error message when the command for a transport
4163 filter fails to execute.
4165 4. There are several places where Exim runs a non-Exim command in a
4166 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4167 was being done only for the command run by the queryprogram router. It is
4168 now done for all such subprocesses. The other cases are: ${run, transport
4169 filters, and the commands run by the lmtp and pipe transports.
4171 5. Some older OS have a limit of 256 on the maximum number of file
4172 descriptors. Exim was using setrlimit() to set 1000 as a large value
4173 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4174 systems. I've change it so that if it can't get 1000, it tries for 256.
4176 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4177 was an oversight, and furthermore, ever since the addition of extra
4178 controls (e.g. 4.43/32), the checks on when to allow different forms of
4179 "control" were broken. There should now be diagnostics for all cases when a
4180 control that does not make sense is encountered.
4182 7. $recipients is now available in the predata ACL (oversight).
4184 8. Tidy the search cache before the fork to do a delivery from a message
4185 received from the command line. Otherwise the child will trigger a lookup
4186 failure and thereby defer the delivery if it tries to use (for example) a
4187 cached ldap connection that the parent has called unbind on.
4189 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4190 of $address_data from the recipient verification was clobbered by the
4191 sender verification.
4193 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4194 was its contents. (It was OK if the option was not defined at all.)
4196 11. A "Completed" log line is now written for messages that are removed from
4197 the spool by the -Mrm option.
4199 12. $host_address is now set to the target address during the checking of
4200 ignore_target_hosts.
4202 13. When checking ignore_target_hosts for an ipliteral router, no host name was
4203 being passed; this would have caused $sender_host_name to have been used if
4204 matching the list had actually called for a host name (not very likely,
4205 since this list is usually IP addresses). A host name is now passed as
4208 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4209 code that specifies a non-restarting handler (typically sigaction() in
4210 modern systems) in an attempt to fix a rare and obscure crash bug.
4212 15. Narrowed the window for a race in the daemon that could cause it to ignore
4213 SIGCHLD signals. This is not a major problem, because they are used only to
4214 wake it up if nothing else does.
4216 16. A malformed maildirsize file could cause Exim to calculate negative values
4217 for the mailbox size or file count. Odd effects could occur as a result.
4218 The maildirsize information is now recalculated if the size or filecount
4221 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4222 support for a long time. Removed HAVE_SYS_VFS_H.
4224 18. Updated exipick to current release from John Jetmore.
4226 19. Allow an empty sender to be matched against a lookup in an address list.
4227 Previously the only cases considered were a regular expression, or an
4230 20. Exim went into a mad DNS lookup loop when doing a callout where the
4231 host was specified on the transport, if the DNS lookup yielded more than
4234 21. The RFC2047 encoding function was originally intended for short strings
4235 such as real names; it was not keeping to the 75-character limit for
4236 encoded words that the RFC imposes. It now respects the limit, and
4237 generates multiple encoded words if necessary. To be on the safe side, I
4238 have increased the buffer size for the ${rfc2047: expansion operator from
4241 22. Failure to deliver a bounce message always caused it to be frozen, even if
4242 there was an errors_to setting on the router. The errors_to setting is now
4245 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4246 canonical form (fully expanded) before being placed in
4247 $sender_host_address.
4249 24. Updated eximstats to version 1.33
4251 25. Include certificate and key file names in error message when GnuTLS fails
4252 to set them up, because the GnuTLS error message doesn't include the name
4253 of the failing file when there is a problem reading it.
4255 26. Expand error message when OpenSSL has problems setting up cert/key files.
4258 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
4259 (this can affect the format of dates).
4261 28. exim_tidydb, when checking for the continued existence of a message for
4262 which it has found a message-specific retry record, was not finding
4263 messages that were in split spool directories. Consequently, it was
4264 deleting retry records that should have stayed in existence.
4266 29. eximstats updated to version 1.35
4267 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
4268 1.35 - bugfix such that pie charts by volume are generated correctly
4270 30. The SPA authentication driver was not abandoning authentication and moving
4271 on to the next authenticator when an expansion was forced to fail,
4272 contradicting the general specification for all authenticators. Instead it
4273 was generating a temporary error. It now behaves as specified.
4275 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
4276 (the order specifies the preference for clients). The order is now AES256,
4277 AES128, 3DES, ARCFOUR128.
4279 31. Small patch to Sieve code - explicitly set From: when generating an
4282 32. Exim crashed if a remote delivery caused a very long error message to be
4283 recorded - for instance if somebody sent an entire SpamAssassin report back
4284 as a large number of 550 error lines. This bug was coincidentally fixed by
4285 increasing the size of one of Exim's internal buffers (big_buffer) that
4286 happened as part of the Exiscan merge. However, to be on the safe side, I
4287 have made the code more robust (and fixed the comments that describe what
4290 33. Some experimental protocols are using DNS PTR records for new purposes. The
4291 keys for these records are domain names, not reversed IP addresses. The
4292 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4293 leaves it alone. Component reversal etc. now happens only for IP addresses.
4296 34. The host_aton() function is supposed to be passed a string that is known
4297 to be a valid IP address. However, in the case of IPv6 addresses, it was
4298 not checking this. This is a hostage to fortune. Exim now panics and dies
4299 if the condition is not met. A case was found where this could be provoked
4300 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4301 components; fortuitously, this particular loophole had already been fixed
4302 by change 4.50/55 or 4.44/33 above.
4304 If there are any other similar loopholes, the new check in host_aton()
4305 itself should stop them being exploited. The report I received stated that
4306 data on the command line could provoke the exploit when Exim was running as
4307 exim, but did not say which command line option was involved. All I could
4308 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4309 running as the user.
4312 35. There was a buffer overflow vulnerability in the SPA authentication code
4313 (which came originally from the Samba project). I have added a test to the
4314 spa_base64_to_bits() function which I hope fixes it.
4317 36. The daemon start-up calls getloadavg() while still root for those OS that
4318 need the first call to be done as root, but it missed one case: when
4319 deliver_queue_load_max is set with deliver_drop_privilege. This is
4320 necessary for the benefit of the queue runner, because there is no re-exec
4321 when deliver_drop_privilege is set.
4323 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4324 was in use, was not putting the data itself into the right store pool;
4325 consequently, it could be overwritten for a subsequent message in the same
4326 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4329 38. Sometimes the final signoff response after QUIT could fail to get
4330 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4331 < 0 before doing a fflush(). This bug looks as though it goes back to the
4332 introduction of TLS in release 3.20, but "sometimes" must have been rare
4333 because the tests only now provoked it.
4339 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
4340 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
4341 Now it returns an integer. A number of calls were still expecting a T/F
4342 return. Fortuitously, in all cases, the tests worked in OK situations,
4343 which is the norm. However, things would have gone wrong on any write
4344 failures on the smtp file descriptor. This function is used when sending
4345 messages over SMTP and also when doing verify callouts.
4347 2. When Exim is called to do synchronous delivery of a locally submitted
4348 message (the -odf or -odi options), it no longer closes stderr before doing
4351 3. Implemented the mua_wrapper option.
4353 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
4355 5. Implemented the functions header_remove(), header_testname(),
4356 header_add_at_position(), and receive_remove_recipient(), and exported them
4359 6. If an ACL "warn" statement specified the addition of headers, Exim already
4360 inserted X-ACL-Warn: at the start if there was no header name. However, it
4361 was not making this test for the second and subsequent header lines if
4362 there were newlines in the string. This meant that an invalid header could
4363 be inserted if Exim was badly configured.
4365 7. Allow an ACL "warn" statement to add header lines at the start or after all
4366 the Received: headers, as well as at the end.
4368 8. Added the rcpt_4xx retry error code.
4370 9. Added postmaster_mailfrom=xxx to callout verification option.
4372 10. Added mailfrom=xxxx to the callout verification option, for verify=
4375 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
4376 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
4378 12. Inserted some casts to stop certain compilers warning when using pointer
4379 differences as field lengths or precisions in printf-type calls (mostly
4380 affecting debugging statements).
4382 13. Added optional readline() support for -be (dynamically loaded).
4384 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
4385 same clock tick as a message's arrival, so that its received time was the
4386 same as the "first fail" time on the retry record, and that message
4387 remained on the queue past the ultimate address timeout, every queue runner
4388 would try a delivery (because it was past the ultimate address timeout) but
4389 after another failure, the ultimate address timeout, which should have then
4390 bounced the address, did not kick in. This was a "< instead of <=" error;
4391 in most cases the first failure would have been in the next clock tick
4392 after the received time, and all would be well.
4394 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
4395 being recognized when the domain list was tested by the match_domain
4396 condition in an expansion string.
4398 16. Added the ${str2b64: operator.
4400 17. Exim was always calling setrlimit() to set a large limit for the number of
4401 processes, without checking whether the existing limit was already
4402 adequate. (It did check for the limit on file descriptors.) Furthermore,
4403 errors from getrlimit() and setrlimit() were being ignored. Now they are
4404 logged to the main and panic logs, but Exim does carry on, to try to do its
4405 job under whatever limits there are.
4407 18. Imported PCRE 5.0.
4409 19. Trivial typo in log message " temporarily refused connection" (the leading
4412 20. If the log selector return_path_on_delivery was set and an address was
4413 redirected to /dev/null, the delivery process crashed because it assumed
4414 that a return path would always be set for a "successful" delivery. In this
4415 case, the whole delivery is bypassed as an optimization, and therefore no
4418 21. Internal re-arrangement: the function for sending a challenge and reading
4419 a response while authentication was assuming a zero-terminated challenge
4420 string. It's now changed to take a pointer and a length, to allow for
4421 binary data in such strings.
4423 22. Added the cyrus_sasl authenticator (code supplied by MBM).
4425 23. Exim was not respecting finduser_retries when seeking the login of the
4426 uid under which it was called; it was always trying 10 times. (The default
4427 setting of finduser_retries is zero.) Also, it was sleeping after the final
4428 failure, which is pointless.
4430 24. Implemented tls_on_connect_ports.
4432 25. Implemented acl_smtp_predata.
4434 26. If the domain in control=submission is set empty, Exim assumes that the
4435 authenticated id is a complete email address when it generates From: or
4436 Sender: header lines.
4438 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
4439 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
4440 chown and chgrp in /bin and hostname in /usr/bin.
4442 28. Exim was keeping the "process log" file open after each use, just as it
4443 does for the main log. This opens the possibility of it remaining open for
4444 long periods when the USR1 signal hits a daemon. Occasional processlog
4445 errors were reported, that could have been caused by this. Anyway, it seems
4446 much more sensible not to leave this file open at all, so that is what now
4449 29. The long-running daemon process does not normally write to the log once it
4450 has entered its main loop, and it closes the log before doing so. This is
4451 so that log files can straightforwardly be renamed and moved. However,
4452 there are a couple of unusual error situations where the daemon does write
4453 log entries, and I had neglected to close the log afterwards.
4455 30. The text of an SMTP error response that was received during a remote
4456 delivery was being truncated at 512 bytes. This is too short for some of
4457 the long messages that one sometimes sees. I've increased the limit to
4460 31. It is now possible to make retry rules that apply only when a message has a
4461 specific sender, in particular, an empty sender.
4463 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
4464 it possible to be selective about when SMTP synchronization is enforced.
4466 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
4468 32. Implemented hosts_connection_nolog.
4470 33. Added an ACL for QUIT.
4472 34. Setting "delay_warning=" to disable warnings was not working; it gave a
4475 35. Added mailbox_size and mailbox_filecount to appendfile.
4477 36. Added control = no_multiline_responses to ACLs.
4479 37. There was a bug in the logic of the code that waits for the clock to tick
4480 in the case where the clock went backwards by a substantial amount such
4481 that the microsecond fraction of "now" was more than the microsecond
4482 fraction of "then" (but the whole seconds number was less).
4484 38. Added support for the libradius Radius client library this is found on
4485 FreeBSD (previously only the radiusclient library was supported).
4491 1. When certain lookups returned multiple values in the form name=value, the
4492 quoting of the values was not always being done properly. Specifically:
4493 (a) If the value started with a double quote, but contained no whitespace,
4495 (b) If the value contained whitespace other than a space character (i.e.
4496 tabs or newlines or carriage returns) it was not quoted.
4497 This fix has been applied to the mysql and pgsql lookups by writing a
4498 separate quoting function and calling it from the lookup code. The fix
4499 should probably also be applied to nisplus, ibase and oracle lookups, but
4500 since I cannot test any of those, I have not disturbed their existing code.
4502 2. A hit in the callout cache for a specific address caused a log line with no
4503 reason for rejecting RCPT. Now it says "Previous (cached) callout
4504 verification failure".
4506 3. There was an off-by-one bug in the queryprogram router. An over-long
4507 return line was truncated at 256 instead of 255 characters, thereby
4508 overflowing its buffer with the terminating zero. As well as fixing this, I
4509 have increased the buffer size to 1024 (and made a note to document this).
4511 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
4512 when Exim is waiting for an SMTP response from a remote server, Exim
4513 restarts its select() call on the socket, thereby resetting its timeout.
4514 This is not a problem when such interrupts are rare. Somebody set up a cron
4515 job to run exiwhat every 2 minutes, which is less than the normal select()
4516 timeout (5 or 10 minutes). This meant that the select() timeout never
4517 kicked in because it was always reset. I have fixed this by comparing the
4518 time when an interrupt arrives with the time at the start of the first call
4519 to select(). If more time than the timeout has elapsed, the interrupt is
4520 treated as a timeout.
4522 5. Some internal re-factoring in preparation for the addition of Sieve
4523 extensions (by MH). In particular, the "personal" test is moved to a
4524 separate function, and given an option for scanning Cc: and Bcc: (which is
4525 not set for Exim filters).
4527 6. When Exim created an email address using the login of the caller as the
4528 local part (e.g. when creating a From: or Sender: header line), it was not
4529 quoting the local part when it contained special characters such as @.
4531 7. Installed new OpenBSD configuration files.
4533 8. Reworded some messages for syntax errors in "and" and "or" conditions to
4534 try to make them clearer.
4536 9. Callout options, other than the timeout value, were being ignored when
4537 verifying sender addresses in header lines. For example, when using
4539 verify = header_sender/callout=no_cache
4541 the cache was (incorrectly) being used.
4543 10. Added a missing instance of ${EXE} to the exim_install script; this affects
4544 only the Cygwin environment.
4546 11. When return_path_on_delivery was set as a log selector, if different remote
4547 addresses in the same message used different return paths and parallel
4548 remote delivery occurred, the wrong values would sometimes be logged.
4549 (Whenever a remote delivery process finished, the return path value from
4550 the most recently started remote delivery process was logged.)
4552 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
4553 lines when AUTH and/or TLS are in use. This is the "received protocol"
4554 field. Exim used to use "asmtp" for authenticated SMTP, without any
4555 indication (in the protocol name) for TLS use. Now it follows the RFC and
4556 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
4557 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
4558 names appear in log lines as well as in Received: header lines.
4560 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
4561 extensions, and comparison tests, and to fix some bugs.
4563 14. Changes to the "personal" filter test:
4565 (1) The test was buggy in that it was just doing the equivalent of
4566 "contains" tests on header lines. For example, if a user's address was
4567 anne@some.where, the "personal" test would incorrectly be true for
4569 To: susanne@some.where
4571 This test is now done by extracting each address from the header in turn,
4572 and checking the entire address. Other tests that are part of "personal"
4573 are now done using regular expressions (for example, to check local parts
4574 of addresses in From: header lines).
4576 (2) The list of non-personal local parts in From: addresses has been
4577 extended to include "listserv", "majordomo", "*-request", and "owner-*",
4578 taken from the Sieve specification recommendations.
4580 (3) If the message contains any header line starting with "List-" it is
4581 treated as non-personal.
4583 (4) The test for "circular" in the Subject: header line has been removed
4584 because it now seems ill-conceived.
4586 15. Minor typos in src/EDITME comments corrected.
4588 16. Installed latest exipick from John Jetmore.
4590 17. If headers_add on a router specified a text string that was too long for
4591 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
4592 of string_sprintf() is now avoided.
4594 18. $message_body_size was not set (it was always zero) when running the DATA
4595 ACL and the local_scan() function.
4597 19. For the "mail" command in an Exim filter, no default was being set for
4598 the once_repeat time, causing a random time value to be used if "once" was
4599 specified. (If the value happened to be <= 0, no repeat happened.) The
4600 default is now 0s, meaning "never repeat". The "vacation" command was OK
4601 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
4602 (I found it when inspecting the code).
4604 20. There is now an overall timeout for performing a callout verification. It
4605 defaults to 4 times the callout timeout, which applies to individual SMTP
4606 commands during the callout. The overall timeout applies when there is more
4607 than one host that can be tried. The timeout is checked before trying the
4608 next host. This prevents very long delays if there are a large number of
4609 hosts and all are timing out (e.g. when the network connections are timing
4610 out). The value of the overall timeout can be changed by specifying an
4611 additional sub-option for "callout", called "maxwait". For example:
4613 verify = sender/callout=5s,maxwait=20s
4615 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
4616 to the end before writing, but this should make it even safer).
4618 22. Exim was forgetting that it had advertised PIPELINING for the second and
4619 subsequent messages on an SMTP connection. It was also not resetting its
4620 memory on STARTTLS and an internal HELO.
4622 23. When Exim logs an SMTP synchronization error within a session, it now
4623 records whether PIPELINING has been advertised or not.
4625 24. Added 3 instances of "(long int)" casts to time_t variables that were being
4626 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
4627 rather than long int.
4629 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
4631 26. Added the never_mail option to autoreply.
4637 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
4638 crash if the getsockname() call failed; this can happen if a connection is
4639 closed very soon after it is established. The problem was simply in the
4640 order in which certain operations were done, causing Exim to try to write
4641 to the SMTP stream before it had set up the file descriptor. The bug has
4642 been fixed by making things happen in the correct order.
4648 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
4649 before the connection was closed, thus losing the rejection response.
4651 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
4652 some early Solaris releases, but causes trouble in current releases where
4653 socklen_t is defined.
4655 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
4658 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
4661 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
4662 that releasing the top of it at the end releases what was used for sub-
4663 expansions (unless the block got too big). However, discard this block if
4664 the first thing is a variable or header, so that we can use its block when
4665 it is dynamic (useful for very large $message_headers, for example).
4667 6. Lookups now cache *every* query, not just the most recent. A new, separate
4668 store pool is used for this. It can be recovered when all lookup caches are
4669 flushed. Lookups now release memory at the end of their result strings.
4670 This has involved some general refactoring of the lookup sources.
4672 7. Some code has been added to the store_xxx() functions to reduce the amount
4673 of flapping under certain conditions.
4675 8. log_incoming_interface used to affect only the <= reception log lines. Now
4676 it causes the local interface and port to be added to several more SMTP log
4677 lines, for example "SMTP connection from", and rejection lines.
4679 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
4681 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
4683 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
4684 could be overwritten at the end of the current message (or the start of a
4685 new message if it was set in a HELO ACL). The value is now preserved for
4686 the duration of the SMTP connection.
4688 12. If a transport had a headers_rewrite setting, and a matching header line
4689 contained an unqualified address, that address was qualified, even if it
4690 did not match any rewriting rules. The underlying bug was that the values
4691 of the flags that permit the existence of unqualified sender and recipient
4692 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
4693 non-local messages, and by -bnq for local messages) were not being
4694 preserved with the message after it was received.
4696 13. When Exim was logging an SMTP synchronization error, it could sometimes log
4697 "next input=" as part of the text comprising the host identity instead of
4698 the correct text. The code was using the same buffer for two different
4699 strings. However, depending on which order the printing function evaluated
4700 its arguments, the bug did not always show up. Under Linux, for example, my
4701 test suite worked just fine.
4703 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
4704 doesn't work with some older versions of Perl. It has been changed to "my",
4705 which in any case is probably the better facility to use.
4707 15. A really picky compiler found some instances of statements for creating
4708 error messages that either had too many or two few arguments for the format
4711 16. The size of the buffer for calls to the DNS resolver has been increased
4712 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
4713 for addresses that have a lot of PTR records. This alleviates a problem; it
4714 does not fully solve it.
4716 17. A dnsdb lookup for PTR records that receives more data than will fit in the
4717 buffer now truncates the list and logs the incident, which is the same
4718 action as happens when Exim is looking up a host name and its aliases.
4719 Previously in this situation something unpredictable would happen;
4720 sometimes it was "internal error: store_reset failed".
4722 18. If a server dropped the connection unexpectedly when an Exim client was
4723 using GnuTLS and trying to read a response, the client delivery process
4724 crashed while trying to generate an error log message.
4726 19. If a "warn" verb in an ACL added multiple headers to a message in a single
4727 string, for example:
4729 warn message = H1: something\nH2: something
4731 the text was added as a single header line from Exim's point of view
4732 though it ended up OK in the delivered message. However, searching for the
4733 second and subsequent header lines using $h_h2: did not work. This has been
4734 fixed. Similarly, if a system filter added multiple headers in this way,
4735 the routers could not see them.
4737 20. Expanded the error message when iplsearch is called with an invalid key to
4738 suggest using net-iplsearch in a host list.
4740 21. When running tests using -bh, any delays imposed by "delay" modifiers in
4741 ACLs are no longer actually imposed (and a message to that effect is
4744 22. If a "gecos" field in a passwd entry contained escaped characters, in
4745 particular, if it contained a \" sequence, Exim got it wrong when building
4746 a From: or a Sender: header from that name. A second bug also caused
4747 incorrect handling when an unquoted " was present following a character
4748 that needed quoting.
4750 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
4751 was not being matched caselessly.
4753 24. Arranged for all hyphens in the exim.8 source to be escaped with
4756 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
4757 properly. Recipient callout cache records were still being keyed to include
4758 the sender, even when use_sender was set false. This led to far more
4759 callouts that were necessary. The sender is no longer included in the key
4760 when use_sender is false.
4762 26. Added "control = submission" modifier to ACLs.
4764 27. Added the ${base62d: operator to decode base 62 numbers.
4766 28. dnsdb lookups can now access SRV records.
4768 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
4769 the configuration file.
4771 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
4772 (-v) mode. This makes the output for a verbose queue run more intelligible.
4774 31. Added a use_postmaster feature to recipient callouts.
4776 32. Added the $body_zerocount variable, containing the number of binary zero
4777 bytes in the message body.
4779 33. The time of last modification of the "new" subdirectory is now used as the
4780 "mailbox time last read" when there is a quota error for a maildir
4783 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
4785 35. Added +ignore_unknown as a special item in host lists.
4787 36. Code for decoding IPv6 addresses in host lists is now included, even if
4788 IPv6 support is not being compiled. This fixes a bug in which an IPv6
4789 address was recognized as an IP address, but was then not correctly decoded
4790 into binary, causing unexpected and incorrect effects when compared with
4797 1. Very minor rewording of debugging text in manualroute to say "list of
4798 hosts" instead of "hostlist".
4800 2. If verify=header_syntax was set, and a header line with an unqualified
4801 address (no domain) and a large number of spaces between the end of the
4802 name and the colon was received, the reception process suffered a buffer
4803 overflow, and (when I tested it) crashed. This was caused by some obsolete
4804 code that should have been removed. The fix is to remove it!
4806 3. When running in the test harness, delay a bit after writing a bounce
4807 message to get a bit more predictability in the log output.
4809 4. Added a call to search_tidyup() just before forking a reception process. In
4810 theory, someone could use a lookup in the expansion of smtp_accept_max_
4811 per_host which, without the tidyup, could leave open a database connection.
4813 5. Added the variables $recipient_data and $sender_data which get set from a
4814 lookup success in an ACL "recipients" or "senders" condition, or a router
4815 "senders" option, similar to $domain_data and $local_part_data.
4817 6. Moved the writing of debug_print from before to after the "senders" test
4820 7. Change 4.31/66 (moving the time when the Received: is generated) caused
4821 problems for message scanning, either using a data ACL, or using
4822 local_scan() because the Received: header was not generated till after they
4823 were called (in order to set the time as the time of reception completion).
4824 I have revised the way this works. The header is now generated after the
4825 body is received, but before the ACL or local_scan() are called. After they
4826 are run, the timestamp in the header is updated.
4832 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
4833 before starting a queue runner without re-exec. This happened only when
4834 deliver_drop_privilege was set or when the Exim user was set to root. The
4835 effect of the bug was that timeouts during subsequent deliveries caused
4836 crashes instead of being properly handled. The handler is now left at its
4837 default (and expected) setting.
4839 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
4840 message, again when deliver_drop_privilege is set or Exim is run as root.
4841 The bug described in (1) was not present in this case, but the tidying up
4842 of the other signals was missing. I have made the two cases consistent.
4844 3. The ignore_target_hosts setting on a manualroute router was being ignored
4845 for hosts that were looked up using the /MX notation.
4847 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
4850 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
4851 operated on the sender address. After changing the $sender_address to <>
4852 for the sender address verify, Exim was re-instated it as the original
4853 (before rewriting) address, but remembering that it had rewritten it, so it
4854 wasn't rewriting it again. This bug also had the effect of breaking the
4855 sender address verification caching when the sender address was rewritten.
4857 6. The ignore_target_hosts option was being ignored by the ipliteral router.
4858 This has been changed so that if the ip literal address matches
4859 ignore_target_hosts, the router declines.
4861 7. Added expansion conditions match_domain, match_address, and match_local_
4862 part (NOT match_host).
4864 8. The placeholder for the Received: header didn't have a length field set.
4866 9. Added code to Exim itself and to exim_lock to test for a specific race
4867 condition that could lead to file corruption when using MBX delivery. The
4868 issue is with the lockfile that is created in /tmp. If this file is removed
4869 after a process has opened it but before that process has acquired a lock,
4870 there is the potential for a second process to recreate the file and also
4871 acquire a lock. This could lead to two Exim processes writing to the file
4872 at the same time. The added code performs the same test as UW imapd; it
4873 checks after acquiring the lock that its file descriptor still refers to
4874 the same named file.
4876 10. The buffer for building added header lines was of fixed size, 8192 bytes.
4877 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
4880 11. Added the smtp_active_hostname option. If used, this will typically be made
4881 to depend on the incoming interface address. Because $interface_address is
4882 not set up until the daemon has forked a reception process, error responses
4883 that can happen earlier (such as "too many connections") no longer contain
4886 12. If an expansion in a condition on a "warn" statement fails because a lookup
4887 defers, the "warn" statement is abandoned, and the next ACL statement is
4888 processed. Previously this caused the whole ACL to be aborted.
4890 13. Added the iplsearch lookup type.
4892 14. Added ident_timeout as a log selector.
4894 15. Added tls_certificate_verified as a log selector.
4896 16. Added a global option tls_require_ciphers (compare the smtp transport
4897 option of the same name). This controls incoming TLS connections.
4899 17. I finally figured out how to make tls_require_ciphers do a similar thing
4900 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
4901 before starting the TLS session.
4903 18. Tabs are now shown as \t in -bP output.
4905 19. If the log selector return_path_on_delivery was set, Exim crashed when
4906 bouncing a message because it had too many Received: header lines.
4908 20. If two routers both had headers_remove settings, and the first one included
4909 a superfluous trailing colon, the final name in the first list and the
4910 first name in the second list were incorrectly joined into one item (with a
4911 colon in the middle).
4917 1. Added -C and -D options to the exinext utility, mainly to make it easier
4918 to include in the automated testing, but these could be helpful when
4919 multiple configurations are in use.
4921 2. The exinext utility was not formatting the output nicely when there was
4922 an alternate port involved in the retry record key, nor when there was a
4923 message id as well (for retries that were specific to a specific message
4924 and a specific host). It was also confused by IPv6 addresses, because of
4925 the additional colons they contain. I have fixed the IPv4 problem, and
4926 patched it up to do a reasonable job for IPv6.
4928 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
4929 delivery, the log line now contains "pipelined" if PIPELINING was used.
4931 4. An SMTP transport process used to panic and die if the bind() call to set
4932 an explicit outgoing interface failed. This has been changed; it is now
4933 treated in the same way as a connect() failure.
4935 5. A reference to $sender_host_name in the part of a conditional expansion
4936 that was being skipped was still causing a DNS lookup. This no longer
4939 6. The def: expansion condition was not recognizing references to header lines
4940 that used bh_ and bheader_.
4942 7. Added the _cache feature to named lists.
4944 8. The code for checking quota_filecount in the appendfile transport was
4945 allowing one more file than it should have been.
4947 9. For compatibility with Sendmail, the command line option
4955 and sets the incoming protocol and host name (for trusted callers). The
4956 host name and its colon can be omitted when only the protocol is to be set.
4957 Note the Exim already has two private options, -pd and -ps, that refer to
4958 embedded Perl. It is therefore impossible to set a protocol value of "d" or
4959 "s", but I don't think that's a major issue.
4961 10. A number of refactoring changes to the code, none of which should affect
4964 (a) The number of logging options was getting close to filling up the
4965 32-bit word that was used as a bit map. I have split them into two classes:
4966 those that are passed in the argument to log_write(), and those that are
4967 only ever tested independently outside of that function. These are now in
4968 separate 32-bit words, so there is plenty of room for expansion again.
4969 There is no change in the user interface or the logging behaviour.
4971 (b) When building, for example, log lines, the code previously used a
4972 macro that called string_cat() twice, in order to add two strings. This is
4973 not really sufficiently general. Furthermore, there was one instance where
4974 it was actually wrong because one of the argument was used twice, and in
4975 one call a function was used. (As it happened, calling the function twice
4976 did not affect the overall behaviour.) The macro has been replaced by a
4977 function that can join an arbitrary number of extra strings onto a growing
4980 (c) The code for expansion conditions now uses a table and a binary chop
4981 instead of a serial search (which was left over from when there were very
4982 few conditions). Also, it now recognizes conditions like "pam" even when
4983 the relevant support is not compiled in: a suitably worded error message is
4984 given if an attempt is made to use such a condition.
4986 11. Added ${time_interval:xxxxx}.
4988 12. A bug was causing one of the ddress fields not to be passed back correctly
4989 from remote delivery subprocesses. The field in question was not being
4990 subsequently used, so this caused to problems in practice.
4992 13. Added new log selectors queue_time and deliver_time.
4994 14. Might have fixed a bug in maildirsizefile handling that threw up
4995 "unexpected character" debug warnings, and recalculated the data
4996 unnecessarily. In any case, I expanded the warning message to give more
4999 15. Added the message "Restricted characters in address" to the statements in
5000 the default ACL that block characters like @ and % in local parts.
5002 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
5003 Three changes have been made:
5005 (a) There was a serious bug; a negative response to MAIL caused the whole
5006 recipient domain to be cached as invalid, thereby blocking all messages
5007 to all local parts at the same domain, from all senders. This bug has
5008 been fixed. The domain is no longer cached after a negative response to
5009 MAIL if the sender used is not empty.
5011 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
5014 (c) A new callout option, "use_sender" has been added for people who want
5015 the modified behaviour.
5021 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
5024 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
5025 indeed breaks things for older releases.
5027 3. Added additional logging to the case where there is a problem reading data
5028 from a filter that is running in a subprocess using a pipe, in order to
5029 try to track down a specific problem.
5031 4. Testing facility fudge: when running in the test harness and attempting
5032 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
5033 getting "No route to host". Convert this to a timeout.
5035 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
5038 6. Some OS don't have socklen_t but use size_t instead. This affects the
5039 fifth argument of getsockopt() amongst other things. This is now
5040 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
5041 can be set for individual OS. I have set it for SunOS5, OSF1, and
5042 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
5043 some earlier ones do not.
5045 7. Change 4.30/15 was not doing the test caselessly.
5047 8. The standard form for an IPv6 address literal was being rejected by address
5048 parsing in, for example, MAIL and RCPT commands. An example of this kind of
5049 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5050 this, as well as the form without the "IPv6" on the front (but only when
5051 address literals are enabled, of course).
5053 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
5055 10. Exim crashed if a message with an empty sender address specified by -f
5056 encountered a router with an errors_to setting. This could be provoked only
5057 by a command such as
5061 where an empty string was supplied; "<>" did not hit this bug.
5063 11. Installed PCRE release 4.5.
5065 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
5066 remained set. It is now erased.
5068 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
5069 times from message ids (which are base 36 rather than the normal 62).
5071 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
5072 were being counted as actual protocol errors, and logged if the log
5073 selector +smtp_protocol_error was set. One cannot be perfect in this test,
5074 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
5075 and DATA following a set of rejected RCPTs do not count as protocol errors.
5076 In other words, Exim assumes they were pipelined, though this may not
5077 actually be the case. Of course, in all cases the client gets an
5078 appropriate error code.
5080 15. If a lookup fails in an ACL condition, a message about the failure may
5081 be available; it is used if testing the ACL cannot continue, because most
5082 such messages specify what the cause of the deferral is. However, some
5083 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
5084 that caused an old message to be retained and used if a later statement
5085 caused a defer, replacing the real cause of the deferral.
5087 16. If an IP address had so many PTR records that the DNS lookup buffer
5088 was not large enough to hold them, Exim could crash while trying to process
5089 the truncated data. It now detects and logs this case.
5091 17. Further to 4.21/58, another change has been made: if (and only if) the
5092 first line of a message (the first header line) ends with CRLF, a bare LF
5093 in a subsequent header line has a space inserted after it, so as not to
5094 terminate the header.
5096 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
5097 unnecessarily, used atoi() instead of strtol(), and didn't check the
5098 termination when getting file sizes from file names by regex.
5100 19. Completely re-implemented the support for maildirsize files, in the light
5101 of a number of problems with the previous contributed implementation
5102 (4.30/29). In particular:
5104 . If the quota is zero, the maildirsize file is maintained, but no quota is
5107 . If the maildir directory does not exist, it is created before any attempt
5108 to write a maildirsize file.
5110 . The quota value in the file is just a cache; if the quota is changed in
5111 the transport, the new value overrides.
5113 . A regular expression is available for excluding directories from the
5116 20. The autoreply transport checks the characters in options that define the
5117 message's headers; it allows continued headers, but it was checking with
5118 isspace() after an embedded newline instead of explicitly looking for a
5121 21. If all the "regular" hosts to which an address was routed had passed their
5122 expiry times, and had not reached their retry times, the address was
5123 bounced, even if fallback hosts were defined. Now Exim should go on to try
5126 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
5127 equivalent code in the SMTP transport. Some hosts send humungous responses
5128 to HELO/EHLO, more than 1024 it seems.
5130 23. Refactoring: code in filter.c used (void *) for "any old type" but this
5131 gives compiler warnings in some environments. I've now done it "properly",
5134 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
5135 (because of problems with the built-in one) was declared to return uschar *
5136 instead of char *, causing compiler failure.
5138 25. Fixed a file descriptor leak when processing alias/forward files.
5140 26. Fixed a minor format string issue in dbfn.c.
5142 27. Typo in exim.c: ("dmbnz" for "dbmnz").
5144 28. If a filter file refered to $h_xxx or $message_headers, and the headers
5145 contained RFC 2047 "words", Exim's memory could, under certain conditions,
5148 29. When a sender address is verified, it is cached, to save repeating the test
5149 when there is more than one recipient in a message. However, when the
5150 verification involves a callout, it is possible for different callout
5151 options to be set for different recipients. It is too complicated to keep
5152 track of this in the cache, so now Exim always runs a verification when a
5153 callout is required, relying on the callout cache for the optimization.
5154 The overhead is duplication of the address routing, but this should not be
5157 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
5158 to be verified with callout=postmaster, and the main callout worked but the
5159 postmaster check failed, the verification correctly failed. However, if a
5160 subsequent RCPT command asked for sender verification *without* the
5161 postmaster check, incorrect caching caused this verification also to fail,
5164 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
5165 it was not caching the DNS options (qualify_single, search_parents) that
5166 were used when the lookup failed. A subsequent lookup with different
5167 options therefore always gave the same answer, though there were cases
5168 where it should not have. (Example: a "domains = !$mx_any" option on a
5169 dnslookup router: the "domains" option is always processed without any
5170 widening, but the router might have qualify_single set.) Now Exim uses the
5171 cached value only when the same options are set.
5173 32. Added John Jetmore's "exipick" utility to the distribution.
5175 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
5176 than a timeout (e.g. a certificate is required, and is not provided), an
5177 Exim server now closes the connection immediately. Previously it waited for
5178 the client to close - but if the client is SSL, it seems that they each
5179 wait for each other, leading to a delay before one of them times out.
5181 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
5182 maintained 0.8.x compatibility because I don't think many are using it, and
5183 it is clearly obsolete.
5185 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
5188 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
5189 client certificate was expired. A simple patch fixes this, though I don't
5190 understand the full logic of why the verify callback is called multiple
5193 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
5194 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
5195 which causes problems with some clients (such as the Certicom SSL Plus
5196 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
5197 disables the coutermeasure allowing Eudora to connect."
5199 38. Exim was not checking that a write() to a log file succeeded. This could
5200 lead to Bad Things if a log got too big, in particular if it hit a file
5201 size limit. Exim now panics and dies if it cannot write to a log file, just
5202 as it does if it cannot open a log file.
5204 39. Modified OS/Makefile-Linux so that it now contains
5206 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
5208 The two -D definitions ensure that Exim is compiled with large file
5209 support, which makes it possible to handle log files that are bigger than
5212 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
5213 instance) a domain was checked against a named list that involved a lookup,
5214 causing $domain_data to be set, then another domain was checked against the
5215 same list, then the first domain was re-checked, the value of $domain_data
5216 after the final check could be wrong. In particular, if the second check
5217 failed, it could be set empty. This bug probably also applied to
5220 41. The strip_trailing_dot option was not being applied to the address given
5221 with the -f command-line option.
5223 42. The code for reading a message's header from the spool was incrementing
5224 $received_count, but never initializing it. This meant that the value was
5225 incorrect (doubled) while delivering a message in the same process in which
5226 it was received. In the most common configuration of Exim, this never
5227 happens - a fresh exec is done - but it can happen when
5228 deliver_drop_privilege is set.
5230 43. When Exim logs an SMTP synchronization error - client data sent too soon -
5231 it now includes up to 150 characters of the unexpected data in the log
5234 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
5235 and building data strings. The size of both of these buffers was 10 000
5236 bytes - far larger than anybody would *ever* want, thought I. Needless to
5237 say, somebody hit the limit. I have increased the maximum line length to
5238 20 000 and the maximum data length of concatenated lines to 100 000. I have
5239 also fixed two bugs, because there was no checking on these buffers. Tsk,
5240 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
5241 buffer is too small.
5243 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
5244 lsearch lookups. Now it does.
5246 46. When parsing a route_list item in a manualroute router, a fixed-length
5247 buffer was used for the list of hosts. I made this 1024 bytes long,
5248 thinking that nobody would ever have a list of hosts that long. Wrong.
5249 Somebody had a whole pile of complicated expansion conditions, and the
5250 string was silently truncated, leading to an expansion error. It turns out
5251 that it is easier to change to an unlimited length (owing to other changes
5252 that have happened since this code was originally written) than to build
5253 structure for giving a limitation error. The length of the item that
5254 expands into the list of hosts is now unlimited.
5256 47. The lsearch lookup could not handle data where the length of text line was
5257 more than 4095 characters. Such lines were truncated, leading to shortened
5258 data being returned. It should now handle lines of any length.
5260 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
5261 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
5264 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
5265 replacing strings such as BIN_DIRECTORY in a source file: the replacement
5266 no longer happens in comment lines. A list of replacements is now placed
5267 at the head of all of the source files, except those whose only change is
5268 to replace PERL_COMMAND in the very first #! line.
5270 50. Replaced the slow insertion sort in queue.c, for sorting the list of
5271 messages on the queue, with a bottom-up merge sort, using code contributed
5272 by Michael Haardt. This should make operations like -bp somewhat faster on
5273 large queues. It won't affect queue runners, except when queue_run_in_order
5276 51. Installed eximstats 1.31 in the distribution.
5278 52. Added support for SRV lookups to the dnslookup router.
5280 53. If an ACL referred to $message_body or $message_body_end, the value was not
5281 reset for any messages that followed in the same SMTP session.
5283 54. The store-handling optimization for building very long strings was not
5284 differentiating between the different store pools. I don't think this
5285 actually made any difference in practice, but I've tidied it.
5287 55. While running the routers to verify a sender address, $sender_address
5288 was still set to the sender address. This is wrong, because when routing to
5289 send a bounce to the sender, it would be empty. Therefore, I have changed
5290 it so that, while verifying a sender address, $sender_address is set to <>.
5291 (There is no change to what happens when verifying a recipient address.)
5293 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
5294 target A or AAAA records (if not already returned) without resetting the
5295 qualify_single or search_parents options of the DNS resolver. These are
5296 inappropriate in this case because the targets of MX and SRV records must
5297 be FQDNs. A broken DNS record could cause trouble if it happened to have a
5298 target that, when qualified, matched something in the local domain. These
5299 two options are now turned off when doing these lookups.
5301 57. It seems that at least some releases of Reiserfs (which does not have the
5302 concept of a fixed number of inodes) returns zero and not -1 for the
5303 number of available inodes. This interacted badly with check_spool_inodes,
5304 which assumed that -1 was the "no such thing" setting. What I have done is
5305 to check that the total number of inodes is greater than zero before doing
5306 the test of how many are available.
5308 58. When a "warn" ACL statement has a log_message modifier, the message is
5309 remembered, and not repeated. This is to avoid a lot of repetition when a
5310 message has many recipients that cause the same warning to be written.
5311 Howewer, Exim was preserving the list of already written lines for an
5312 entire SMTP session, which doesn't seem right. The memory is now reset if a
5313 new message is started.
5315 59. The "rewrite" debugging flag was not showing the result of rewriting in the
5316 debugging output unless log_rewrite was also set.
5318 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
5319 of (int)(handle) when we know that handle contains (void *)(-1).
5321 61. The Exim daemon panic-logs an error return when it closes the incoming
5322 connection. However "connection reset by peer" seems to be common, and
5323 isn't really an error worthy of noting specially, so that particular error
5326 62. When Exim is trying to find all the local interfaces, it used to panic and
5327 die if the ioctl to get the interface flags failed. However, it seems that
5328 on at least one OS (Solaris 9) it is possible to have an interface that is
5329 included in the list of interfaces, but for which you get a failure error
5330 for this call. This happens when the interface is not "plumbed" into a
5331 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
5332 failure of the "get flags" call assumes that the interface is down.
5334 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
5335 makes life easier for people who are doing arithmetic on fields extracted
5336 from dates, where you often get leading zeros that should not be
5337 interpreted as octal.
5339 64. Added qualify_domain to the redirect router, to override the global
5342 65. If a pathologically long header line contained very many addresses (the
5343 report of this problem mentioned 10 000) and each of them was rewritten,
5344 Exim could use up a very large amount of memory. (It kept on making new
5345 copies of the header line as it rewrote, and never released the old ones.)
5346 At the expense of a bit more processing, the header rewriting function has
5347 been changed so that it no longer eats memory in this way.
5349 66. The generation of the Received: header has been moved from the time that a
5350 message starts to be received, to the time that it finishes. The timestamp
5351 in the Received: header should now be very close to that of the <= log
5352 line. There are two side-effects of this change:
5354 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
5355 logged header lines no longer include the local Received: line, because
5356 it has not yet been created. The same applies to a copy of the message
5357 that is returned to a non-SMTP sender when a message is rejected.
5359 (b) When a filter file is tested using -bf, no additional Received: header
5360 is added to the test message. After some thought, I decided that this
5363 This change does not affect the value of $received_for. It is still set
5364 after address rewriting, but before local_scan() is called.
5366 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
5368 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
5369 gave an unhelpful panic error message, and a defer error. I have managed to
5370 change this behaviour so that it now rejects any supplied certificate,
5371 which seems right, as the list of acceptable certificates is empty.
5373 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
5374 gave an unhelpful defer error. I have not managed to make this reject any
5375 supplied certificates, but the error message it gives is "no certificate
5376 supplied", which is not helpful.
5378 70. exigrep's output now also includes lines that are not associated with any
5379 message, but which match the given pattern. Implemented by a patch from
5380 Martin Sluka, which also tidied up the Perl a bit.
5382 71. Recipient callout verification, like sender verification, was using <> in
5383 the MAIL FROM command. This isn't really the right thing, since the actual
5384 sender may affect whether the remote host accepts the recipient or not. I
5385 have changed it to use the actual sender in the callout; this means that
5386 the cache record is now keyed on a recipient/sender pair, not just the
5387 recipient address. There doesn't seem to be a real danger of callout loops,
5388 since a callout by the remote host to check the sender would use <>.
5389 [SEE ABOVE: changed after hitting problems.]
5391 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
5392 temporary errors. However, in the case of such a code being given after
5393 the end of a data transmission (i.e. after ".") Exim was failing to write
5394 a retry record for the message. (Yes, there was some broken host that was
5395 actually sending 8xx at this point.)
5397 73. An unknown lookup type in a host list could cause Exim to panic-die when
5398 the list was checked. (An example that provoked this was putting <; in the
5399 middle of a list instead of at the start.) If this happened during a DATA
5400 ACL check, a -D file could be left lying around. This kind of configuration
5401 error no longer causes Exim to die; instead it causes a defer errror. The
5402 incident is still logged to the main and panic logs.
5404 74. Buglet left over from Exim 3 conversion. The message "too many messages
5405 in one connection" was written to the rejectlog but not the mainlog, except
5406 when address rewriting (yes!) was being logged.
5408 75. Added write_rejectlog option.
5410 76. When a system filter was run not as root (that is, when system_filter_user
5411 was set), the values of the $n variables were not being returned to the
5412 main process; thus, they were not subsequently available in the $sn
5415 77. Added +return_path_on_delivery log selector.
5417 78. A connection timeout was being treated differently from recipients deferred
5418 when testing hosts_max_try with a message that was older than the host's
5419 retry timeout. (The host should not be counted, thus allowing all hosts to
5420 be tried at least once before bouncing.) This may have been the cause of an
5421 occasionally reported bug whereby a message would remain on the queue
5422 longer than the retry timeout, but would be bounced if a delivery was
5423 forced. I say "may" because I never totally pinned down the problem;
5424 setting up timeout/retry tests is difficult. See also the next item.
5426 79. The ultimate address timeout was not being applied to errors that involved
5427 a combination of host plus message (for example, a timeout on a MAIL
5428 command). When an address resolved to a number of possible hosts, and they
5429 were not all tried for each delivery (e.g. because of hosts_max_try), a
5430 message could remain on the queue longer than the retry timeout.
5432 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
5435 81. Fixed an obscure SMTP outgoing bug which required at least the following
5436 conditions: (a) there was another message waiting for the same server;
5437 (b) the server returned 5xx to all RCPT commands in the first message so
5438 that the message was not completed; (c) the server dropped the connection
5439 or gave a negative response to the RSET that Exim sends to abort the
5440 transaction. The observed case was a dropped connection after DATA that had
5441 been sent in pipelining mode. That is, the server had advertised PIPELINING
5442 but was not implementing it correctly. The effect of the bug was incorrect
5443 behaviour, such as trying another host, and this could lead to a crash.
5449 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
5450 and daemon.c were passed as pointers to ints; they should have been
5451 pointers to socklen_t variables (which are typically unsigned ints).
5453 2. Some signed/unsigned type warnings in the os.c file for Linux have been
5456 3. Fixed a really odd bug that affected only the testing scheme; patching a
5457 certain fixed string in the binary changed the value of another string that
5458 happened to be identical to the end of the original first string.
5460 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
5461 name", it returns that address as the IP address. On some operating
5462 systems (e.g. Solaris), it also passes back the IP address string as the
5463 "host name". However, on others (e.g. Linux), it passes back an empty
5464 string. Exim wasn't checking for this, and was changing the host name to an
5465 empty string, assuming it had been canonicized.
5467 5. Although rare, it is permitted to have more than one PTR record for a given
5468 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
5469 all the names associated with an address, because they do in Solaris.
5470 However, it seems that they do not in Linux for data that comes from the
5471 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
5472 I found this out when I moved to a new Linux workstation and tried to run
5473 the Exim test suite.
5475 To get round this problem I have changed the code so that it now does its
5476 own call to the DNS to look up PTR records when searching for a host name.
5477 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
5478 addresses that are only in /etc/hosts are still found.
5480 This behaviour is, however, controlled by an option called host_lookup_
5481 order, which defaults to "bydns:byaddr". If people want to use the other
5482 order, or indeed, just use one or the other means of lookup, they can
5483 specify it in this variable.
5485 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
5486 some operating systems, this comes back from gethostbyaddr() as an empty
5487 string, and this is what Exim used to test for. However, it seems that in
5488 other systems, "." is yielded. Exim now tests for this case too.
5490 7. The values of check_spool_space and check_log_space are now held internally
5491 as a number of kilobytes instead of an absolute number of bytes. If a
5492 numbers is specified without 'K' or 'M', it is rounded up to the nearest
5493 kilobyte. This means that much larger values can be stored.
5495 8. Exim monitor: an attempt to get the action menu when not actually pointing
5496 at a message produces an empty menu entitled "No message selected". This
5497 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
5498 no entries in it ("Shell widget menu has zero width and/or height"). So I
5499 have added a single, blank menu entry in this case.
5501 9. Added ${quote_local_part.
5503 10. MIME decoding is now applied to the contents of Subject: header lines when
5506 11. Now that a reference to $sender_host_address automatically causes a reverse
5507 lookup to occur if necessary (4.13/18), there is no need to arrange for a
5508 host lookup before query-style lookups in lists that might use this
5509 variable. This has therefore been abolished, and the "net-" prefix is no
5510 longer necessary for query-style lookups.
5512 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
5513 have been a typo for LFLAGS, so it has been changed.
5515 13. The install script calls Exim with "-C /dev/null" in order to find the
5516 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
5517 to be output. Howeve, since Exim outputs its version number before the
5518 error, it didn't break the script. It just looked ugly. I fixed this by
5519 always allowing "-C /dev/null" if the caller is root.
5521 14. Ignore overlarge ACL variable number when reading spool file - insurance
5522 against a later release with more variables having written the file.
5524 15. The standard form for an IPv6 address literal was being rejected by EHLO.
5525 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5526 this, as well as the form without the "IPv6" on the front.
5528 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
5529 OS/Makefile-Darwin file.
5531 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
5532 only to LDAP libraries that do not have LDAP_OPT_DEREF.
5534 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
5535 variable. However, some OS use ints for inodes. I've added cast to long int
5536 to get rid of the compiler warning.
5538 19. I had forgotten to lock out "/../" in configuration file names when
5539 ALT_CONFIG_PREFIX was set.
5541 20. Routers used for verification do not need to specify transports. However,
5542 if such a router generated a host list, and callout was configured, Exim
5543 crashed, because it could not find a port number from the (non-existent)
5544 transport. It now assumes port 25 in this circumstance.
5546 21. Added the -t option to exigrep.
5548 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
5549 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
5550 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
5551 them is set, LOOKUP_LSEARCH is forced.
5553 23. "exim -bV" now outputs a list of lookups that are included in the binary.
5555 24. Added sender and host information to the "rejected by local_scan()" log
5556 line; previously there was no indication of these.
5558 25. Added .include_if_exists.
5560 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
5561 Linux. It turns out that not all file systems support this. Apparently some
5562 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
5563 do it.) To cope with this, the error EINVAL, which means that sync-ing is
5564 not supported on the file descriptor, is now ignored when Exim is trying to
5565 sync a directory. This applies only to Linux.
5567 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
5569 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
5570 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
5573 29. Added support for maildirsize files from supplied patch (modified a bit).
5575 30. The use of :fail: followed by an empty string could lead Exim to respond to
5576 sender verification failures with (e.g.):
5578 550 Verification failed for <xxx>
5579 550 Sender verify failed
5581 where the first response line was missing the '-' that indicates it is not
5582 the final line of the response.
5584 31. The loop for finding the name of the user that called Exim had a hardwired
5585 limit of 10; it now uses the value of finduser_retries, which is used for
5586 all other user lookups.
5588 32. Added $received_count variable, available in data and not_smtp ACLs, and at
5591 33. Exim was neglecting to zero errno before one call of strtol() when
5592 expanding a string and expecting an integer value. On some systems this
5593 resulted in spurious "integer overflow" errors. Also, it was casting the
5594 result into an int without checking.
5596 34. Testing for a connection timeout using "timeout_connect" in the retry rules
5597 did not work. The code looks as if it has *never* worked, though it appears
5598 to have been documented since at least releast 1.62. I have made it work.
5600 35. The "timeout_DNS" error in retry rules, also documented since at least
5601 1.62, also never worked. As it isn't clear exactly what this means, and
5602 clearly it isn't a major issue, I have abolished the feature by treating it
5603 as "timeout", and writing a warning to the main and panic logs.
5605 36. The display of retry rules for -brt wasn't always showing the error code
5608 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
5609 timeout_connect_A, timeout_connect_MX.
5611 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
5612 to the empty sender.
5614 39. The daemon was not analysing the content of -oX till after it had closed
5615 stderr and disconnected from the controlling terminal. This meant that any
5616 syntax errors were only noted on the panic log, and the return code from
5617 the command was 0. By re-arranging the code a little, I've made the
5618 decoding happen first, so such errors now appear on stderr, and the return
5619 code is 1. However, the actual setting up of the sockets still happens in
5620 the disconnected process, so errors there are still only recorded on the
5623 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
5624 connections (as happens on some IP stacks) was logged at start up time as
5625 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
5626 it from "IPv6 and IPv4", which means that two separate sockets are being
5629 41. The debug output for gethostbyname2() or getipnodebyname() failures now
5630 says whether AF_INET or AF_INET6 was passed as an argument.
5632 42. Exiwhat output was messed up when time zones were included in log
5635 43. Exiwhat now gives more information about the daemon's listening ports,
5636 and whether -tls-on-connect was used.
5638 44. The "port" option of the smtp transport is now expanded.
5640 45. A "message" modifier in a "warn" statement in a non-message ACL was being
5641 silently ignored. Now an error message is written to the main and panic
5644 46. There's a new ACL modifier called "logwrite" which writes to a log file
5645 as soon as it is encountered.
5647 47. Added $local_user_uid and $local_user_gid at routing time.
5649 48. Exim crashed when trying to verify a sender address that was being
5652 49. Exim was recognizing only a space character after ".include". It now also
5653 recognizes a tab character.
5655 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
5656 extracting the relevant information from the specification. The man page no
5657 longer contains scrambled data for the -d option, and I've added a section
5658 at the front about calling Exim under different names.
5660 51. Added "extra_headers" argument to the "mail" command in filter files.
5662 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
5665 53. Installed eximstats 1.29.
5667 54. Added transport_filter_timeout as a generic transport option.
5669 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
5670 Cc: header lines. This was required by RFC 822, but it not required by RFC
5673 56. Exim used to add From:, Date:, and Message-Id: header lines to any
5674 incoming messages that did not have them. Now it does so only if the
5675 message originates locally, that is, if there is no associated remote host
5676 address. When Resent- header lines are present, this applies to the Resent-
5677 lines rather than the non-Resent- lines.
5679 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
5680 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
5682 58. Messages for configuration errors now include the name of the main
5683 configuration file - useful now that there may be more than one file in a
5684 list (.included file names were always shown).
5686 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
5687 for those rare installations that do not start the daemon as root or run it
5688 setuid root. I've cut out the call to initgroups() if the daemon is not
5691 60. The Exim user and group can now be bound into the binary as text strings
5692 that are looked up at the start of Exim's processing.
5694 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
5696 62. Added $mailstore_basename variable.
5698 63. Installed patch to sieve.c from Michael Haardt.
5700 64. When Exim failed to open the panic log after failing to open the main log,
5701 the original message it was trying to log was written to stderr and debug
5702 output, but if they were not available (the usual case in production), it
5703 was lost. Now it is written to syslog before the two lines that record the
5704 failures to open the logs.
5706 65. Users' Exim filters run in subprocesses under the user's uid. It is
5707 possible for a "deliver" command or an alias in a "personal" command to
5708 provoke an address rewrite. If logging of address rewriting is configured,
5709 this fails because the process is not running as root or exim. There may be
5710 a better way of dealing with this, but for the moment (because 4.30 needs
5711 to be released), I have disabled address rewrite logging when running a
5712 filter in a non-root, non-exim process.
5718 1. The buildconfig auxiliary program wasn't quoting the value set for
5719 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
5720 not defined. This bug was masked in 4.22 by the effect that was fixed in
5723 2. Some messages that were rejected after a message id was allocated were
5724 shown as "incomplete" by exigrep. It no longer does this for messages that
5725 are rejected by local_scan() or the DATA or non-SMTP ACLs.
5727 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
5728 have allow_domain_literals set, the ID did not get logged in the <= line.
5729 Domain literals are now always recognized in Message-ID: header lines.
5731 4. The first argument for a ${extract expansion item is the key name or field
5732 number. Leading and trailing spaces in this item were not being ignored,
5733 causing some misleading effects.
5735 5. When deliver_drop_privilege was set, single queue runner processes started
5736 manually (i.e. by the command "exim -q") or by the daemon (which uses the
5737 same command in the process it spins off) were not dropping privilege.
5739 6. When the daemon running as "exim" started a queue runner, it always
5740 re-executed Exim in the spun-off process. This is a waste of effort when
5741 deliver_drop_privilege is set. The new process now just calls the
5742 queue-runner function directly.
5748 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
5751 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
5752 ignored. Though the use of -oP was forcing the writing of a pid file, it
5753 was always written to the default place.
5755 3. If the message "no IP address found for host xxxx" is generated during
5756 incoming verification, it is now followed by identification of the incoming
5757 connection (so you can more easily find what provoked it).
5759 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
5761 5. Added some features to "harden" Exim a bit more against certain attacks:
5763 (a) There is now a build-time option called FIXED_NEVER_USERS that can
5764 be put in Local/Makefile. This is like the never_users runtime option,
5765 but it cannot be overridden. The default setting is "root".
5767 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
5768 prefix string with which any file named in a -C command line option
5771 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
5772 is retained for -C and -D only if the caller of Exim is root. Without
5773 it, the exim user may also use -C and -D and retain privilege.
5775 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
5776 command line option is disabled.
5778 6. Macro names set by the -D option must start with an upper case letter, just
5779 like macro names defined in the configuration file.
5781 7. Added "dereference=" facility to LDAP.
5783 8. Two instances of the typo "uknown" in the source files are fixed.
5785 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
5786 the Configure-Makefile script screwed up while processing it.
5788 10. Incorporated PCRE 4.4.
5790 11. The SMTP synchronization check was not operating right at the start of an
5791 SMTP session. For example, it could not catch a HELO sent before the client
5792 waited for the greeting. There is now a check for outstanding input at the
5793 point when the greeting is written. Because of the duplex, asynchronous
5794 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
5795 way, but not yet received, when the check is performed.
5797 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
5798 on TCP/IP sockets, because this apparently causes some broken clients to
5801 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
5802 unchanged) from the Cygwin maintainer.
5804 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
5805 was supported instead of testing for mbx. Effectively a typo.
5807 15. The spa authenticator server code was not checking that the input it
5808 received was valid base64.
5810 16. The debug output line for the "set" modifier in ACLs was not showing the
5811 name of the variable that was being set.
5813 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
5815 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
5816 lookup on its own. Something else was needed to trigger the lookup. For
5817 example, a match in host_lookup or the need for a host name in a host list.
5818 Now, if $sender_host_name is referenced and the host name has not yet been
5819 looked up, a lookup is performed. If the lookup fails, the variable remains
5820 empty, and $host_lookup_failed is set to "1".
5822 19. Added "eqi" as a case-independent comparison operator.
5824 20. The saslauthd authentication condition could segfault if neither service
5825 nor realm was specified.
5827 21. If an overflowing value such as "2048M" was set for message_size_limit, the
5828 error message that was logged was misleading, and incoming SMTP
5829 connections were dropped. The message is now more accurate, and temporary
5830 errors are given to SMTP connections.
5832 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
5833 (except RSET) with a 421 error, until QUIT is received. However, it was
5834 failing to send a response to QUIT.
5836 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
5837 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
5838 ACL is now run after the helo_try_verify_hosts code.
5840 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
5841 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
5842 case-independent, so other case variants are also recognized). Apparently
5843 some systems use these upper case variants.
5845 25. If more than two messages were waiting for the same host, and a transport
5846 filter was specified for the transport, Exim sent two messages over the
5847 same TCP/IP connection, and then failed with "socket operation on non-
5848 socket" when it tried to send the third.
5850 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
5852 27. The extern definition of crypt16() in expand.c was not being excluded when
5853 the OS had its own crypt16() function.
5855 28. Added bounce_return_body as a new option, and bounce_return_size_limit
5856 as a preferred synonym for return_size_limit, both as an option and as an
5859 29. Added LIBS=-liconv to OS/Makefile-OSF1.
5861 30. Changed the default configuration ACL to relax the local part checking rule
5862 for addresses that are not in any local domains. For these addresses,
5863 slashes and pipe symbols are allowed within local parts, but the sequence
5864 /../ is explicitly forbidden.
5866 31. SPA server authentication was not clearing the challenge buffer before
5869 32. log_message in a "warn" ACL statement was writing to the reject log as
5870 well as to the main log, which contradicts the documentation and doesn't
5871 seem right (because no rejection is happening). So I have stopped it.
5873 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
5874 However, I am unable to do any testing of this.
5876 34. Fixed an infelicity in the appendfile transport. When checking directories
5877 for a mailbox, to see if any needed to be created, it was accidentally
5878 using path names with one or more superfluous leading slashes; tracing
5879 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
5881 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
5882 contents are added to the log line that is written for every discarded
5883 recipient. (Previously a log_message setting was ignored.)
5885 36. The ${quote: operator now quotes the string if it is empty.
5887 37. The install script runs exim in order to find its version number. If for
5888 some reason other than non-existence or emptiness, which it checks, it
5889 could not run './exim', it was installing it with an empty version number,
5890 i.e. as "exim-". This error state is now caught, and the installation is
5893 38. An argument was missing from the function that creates an error message
5894 when Exim fails to connect to the socket for saslauthd authentication.
5895 This could cause Exim to crash, or give a corrupted message.
5897 39. Added isip, isip4, and isip6 to ${if conditions.
5899 40. The ACL variables $acl_xx are now saved with the message, and can be
5900 accessed later in routers, transports, and filters.
5902 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
5903 strings in the file are not string-expanded.
5905 42. If a MAIL command specified a SIZE value that was too large to fit into an
5906 int variable, the check against message_size_limit failed. Such values are
5907 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
5908 day this will have to be increased, but I don't think I want to be around
5909 when emails are that large.
5916 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
5917 iconv() is not standard in FreeBSD.
5919 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
5920 IPv6 enabled. The observed symptom was a segmentation fault on return from
5921 the function os_common_find_running_interfaces() in src/os.c.
5923 3. In the check_special_case() function in daemon.c I had used "errno" as an
5924 argument name, which causes warnings on some systems. This was basically a
5925 typo, since it was named "eno" in the comments!
5927 4. The code that waits for the clock to tick (at a resolution of some fraction
5928 of a second) so as to ensure message-id uniqueness was always waiting for
5929 at least one whole tick, when it could have waited for less. [This is
5930 almost certainly not relevant at current processor speeds, where it is
5931 unlikely to ever wait at all. But we try to future-proof.]
5933 5. The function that sleeps for a time interval that includes fractions of a
5934 second contained a race. It did not block SIGALRM between setting the
5935 timer, and suspending (a couple of lines later). If the interval was short
5936 and the sigsuspend() was delayed until after it had expired, the suspension
5937 never ended. On busy systems this could lead to processes getting stuck for
5940 6. Some uncommon configurations may cause a lookup to happen in a queue runner
5941 process, before it forks any delivery processes. The open lookup caching
5942 mechanism meant that the open file or database connection was passed into
5943 the delivery process. The problem was that delivery processes always tidy
5944 up cached lookup data. This could cause a problem for the next delivery
5945 process started by the queue runner, because the external queue runner
5946 process does not know about the closure. So the next delivery process
5947 still has data in the lookup cache. In the case of a file lookup, there was
5948 no problem because closing a file descriptor in a subprocess doesn't affect
5949 the parent. However, if the lookup was caching a connection to a database,
5950 the connection was closed, and the second delivery process was likely to
5951 see errors such as "PGSQL: query failed: server closed the connection
5952 unexpectedly". The problem has been fixed by closing all cached lookups
5953 in a queue runner before running a delivery process.
5955 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
5956 seem to have the "const" qualifier which it has on other OS. I've
5959 8. Change 4.21/2 was too strict. It is only if there are two authenticators
5960 *of the same type* (client or server) with the same public name that an
5961 error should be diagnosed.
5963 9. When Exim looked up a host name for an IP address, but failed to find the
5964 original IP address when looking up the host name (a safety check), it
5965 output the message "<ip address> does not match any IP for NULL", which was
5966 confusing, to say the least. The bug was that the host name should have
5967 appeared instead of "NULL".
5969 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
5970 user that is built into the binary, and the -C or -D options is used, root
5971 privilege is dropped before the configuration file is read. In addition,
5972 logging is switched to stderr instead of the normal log files. If the
5973 configuration then re-defines the Exim user, the unprivileged environment
5974 is probably not what is expected, so Exim logs a panic warning message (but
5977 However, if deliver_drop_privilege is set, the unprivileged state may well
5978 be exactly what is intended, so the warning has been cut out in that case,
5979 and Exim is allowed to try to write to its normal log files.
5985 1. smtp_return_error_details was not giving details for temporary sender
5986 or receiver verification errors.
5988 2. Diagnose a configuration error if two authenticators have the same public
5991 3. Exim used not to create the message log file for a message until the first
5992 delivery attempt. This could be confusing when incoming messages were held
5993 for policy or load reasons. The message log file is now created at the time
5994 the message is received, and an initial "Received" line is written to it.
5996 4. The automatically generated man page for command line options had a minor
5997 bug that caused no ill effects; however, a more serious problem was that
5998 the procedure for building the man page automatically didn't always
5999 operate. Consequently, release 4.20 contains an out-of-date version. This
6000 shouldn't happen again.
6002 5. When building Exim with embedded Perl support, the script that builds the
6003 Makefile was calling 'perl' to find its compile-time parameters, ignoring
6004 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
6006 6. The freeze_tell option was not being used for messages that were frozen on
6007 arrival, either by an ACL or by local_scan().
6009 7. Added the smtp_incomplete_transaction log selector.
6011 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
6012 was accepting AUTH without a new EHLO.
6014 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
6015 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
6018 10. Logging of TCP/IP connections (when configured) now happens in the main
6019 daemon process instead of the child process, so that the TCP/IP connection
6020 count is more accurate (but it can never be perfect).
6022 11. The use of "drop" in a nested ACL was not being handled correctly in the
6023 outer ACL. Now, if condition failure induced by the nested "drop" causes
6024 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
6025 or "require"), the connection is dropped.
6027 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
6028 that yield "discard" can now be used with an "accept" or a "discard" verb,
6029 but an error is generated for any others (because I can't see a useful way
6030 to define what should happen).
6032 13. When an ACL is read dynamically from a file (or anywhere else), the lines
6033 are now processed in the same way as lines in the Exim configuration file.
6034 In particular, continuation lines are supported.
6036 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
6038 15. Added -ti meaning -t -i.
6040 16. Check for letters, digits, hyphens, and dots in the names of dnslist
6041 domains, and warn by logging if others are found.
6043 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
6044 returned from GIFCONF when Exim is trying to find the list of interfaces on
6045 a host. The code in os.c has been modified to copy each ifreq to an aligned
6046 structure in all cases.
6048 Also, in some cases, the returned ifreq's were being copied to a 'struct
6049 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
6050 means the last couple of bytes of an IPv6 address could be chopped if the
6051 ifreq contained only a normal sockaddr (14 bytes storage).
6053 18. Named domain lists were not supported in the hosts_treat_as_local option.
6054 An entry such as +xxxx was not recognized, and was treated as a literal
6057 19. Ensure that header lines added by a DATA ACL are included in the reject log
6058 if the ACL subsequently rejects the message.
6060 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
6061 MD5 (which is deprecated).
6063 21. When testing a filter file using -bf, Exim was writing a message when it
6064 took the sender from a "From " line in the message, but it was not doing so
6065 when it took $return_path from a Return-Path: header line. It now does.
6067 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
6068 with a valid header line field name (a series of printing characters
6069 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
6071 23. Changed "disc" in the source to "disk" to conform to the documentation and
6072 the book and for uniformity.
6074 24. Ignore Sendmail's -Ooption=value command line item.
6076 25. When execve() failed while trying to run a command in a pipe transport,
6077 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
6078 could be confused with a return value of 69 from the command itself. This
6079 has been changed to 127, the value the shell returns if it is asked to run
6080 a non-existent command. The wording for the related log line suggests a
6081 non-existent command as the problem.
6083 26. If received_header_text expands to an empty string, do not add a Received:
6084 header line to the message. (Well, it adds a token one on the spool, but
6085 marks it "old" so that it doesn't get used or transmitted.)
6087 27. Installed eximstats 1.28 (addition of -nt option).
6089 28. There was no check for failure on the call to getsockname() in the daemon
6090 code. This can fail if there is a shortage of resources on the system, with
6091 ENOMEM, for example. A temporary error is now given on failure.
6093 29. Contrary to the C standard, it seems that in some environments, the
6094 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
6095 program. Exim now does this explicitly; it affects the formatting of
6096 timestamps using strftime().
6098 30. If exiqsumm was given junk data, it threw up some uninitialized variable
6099 complaints. I've now initialized all the variables, to avoid this.
6101 32. Header lines added by a system filter were not being "seen" during
6102 transport-time rewrites.
6104 33. The info_callback() function passed to OpenSSL is set up with type void
6105 (*)(SSL *, int, int), as described somewhere. However, when calling the
6106 function (actually a macro) that sets it up, the type void(*)() is
6107 expected. I've put in a cast to prevent warnings from picky compilers.
6109 34. If a DNS black list lookup found a CNAME record, but there were no A
6110 records associated with the domain it pointed at, Exim crashed.
6112 35. If a DNS black list lookup returned more than one A record, Exim ignored
6113 all but the first. It now scans all returned addresses if a particular IP
6114 value is being sought. In this situation, the contents of the
6115 $dnslist_value variable are a list of all the addresses, separated by a
6118 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
6119 to accept a host name and all its aliases if the forward lookup for any of
6120 them yielded the IP address of the incoming connection. Now it accepts only
6121 those names whose forward lookup yields the correct IP address. Any other
6122 names are discarded. This closes a loophole whereby a rogue DNS
6123 administrator could create reverse DNS records to break through a
6124 wildcarded host restriction in an ACL.
6126 37. If a user filter or a system filter that ran in a subprocess used any of
6127 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
6128 the wrong values were passed to the pipe command ($thisaddress had the
6129 value of $0, $0 had the value of $1, etc). This bug was introduced by
6130 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
6132 38. Improved the line breaking for long SMTP error messages from ACLs.
6133 Previously, if there was no break point between 40 and 75 characters, Exim
6134 left the rest of the message alone. Two changes have been made: (a) I've
6135 reduced the minimum length to 35 characters; (b) if it can't find a break
6136 point between 35 and 75 characters, it looks ahead and uses the first one
6137 that it finds. This may give the occasional overlong line, but at least the
6138 remaining text gets split now.
6140 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
6141 file descriptors might be low, and that setting 1000 would always raise it.
6142 It turns out that in some environments, the limit is already over 1000 and
6143 that lowering it causes trouble. So now Exim takes care not to decrease it.
6145 40. When delivering a message, the value of $return_path is set to $sender_
6146 address at the start of routing (routers may change the value). By an
6147 oversight, this default was not being set up when an address was tested by
6148 -bt or -bv, which affected the outcome if any router or filter referred to
6151 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
6152 log only when "message" or "log_message" are set. However, if one of the
6153 conditions was an address verification, or a call to a nested ACL, the
6154 messages generated by the underlying test were being passed through. This
6155 no longer happens. The underlying message is available in $acl_verify_
6156 message for both "message" and "log_message" expansions, so it can be
6157 passed through if needed.
6159 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
6160 new expansion $bh_ to give the encoded byte string without charset
6161 translation. Translation happens only if iconv() is available; HAVE_ICONV
6162 indicates this at build time. HEADERS_CHARSET gives the charset to
6163 translate to; headers_charset can change it in the configuration, and
6164 "headers charset" can change it in an individual filter file.
6166 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
6167 that creates RFC 2047 encoded "words" labels them as that charset instead
6168 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
6169 expansion operator; (ii) when Exim creates a From: line for a local
6170 message; (iii) when a header line is rewritten to include a "phrase" part.
6172 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
6173 buggy, causing it to skip the first lines of messages whose message ID
6174 ended in 'D'. This would not have bitten before Exim release 4.14, because
6175 message IDs were unlikely to end in 'D' before then. The effect was to have
6176 incorrect size information for certain domains.
6178 45. #include "config.h" was missing at the start of the crypt16.c module. This
6179 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
6182 46. If there was a timeout during a "random" callout check, Exim treated it as
6183 a failure of the random address, and carried on sending RSET and the real
6184 address. If the delay was just some slowness somewhere, the response to the
6185 original RCPT would be taken as a response to RSET and so on, causing
6186 mayhem of various kinds.
6188 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
6189 when I implemented it. It didn't allow for the fact that some option values
6190 may legitimatetly be negative (e.g. size_addition), and it didn't even do
6191 the right test for positive values.
6193 48. Domain names in DNS records are case-independent. Exim always looks them up
6194 in lower case. Some resolvers return domain names in exactly the case they
6195 appear in the zone file, that is, they may contain uppercase letters. Not
6196 all resolvers do this - some return always lower case. Exim was treating a
6197 change of case by a resolver as a change of domain, similar to a widening
6198 of a domain abbreviation. This triggered its re-routing code and so it was
6199 trying to route what was effectively the same domain again. This normally
6200 caused routing to fail (because the router wouldn't handle the domain
6201 twice). Now Exim checks for this case specially, and just changes the
6202 casing of the domain that it ultimately uses when it transmits the message
6205 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
6208 50. If a filter generated a file delivery with a non-absolute name (possible if
6209 no home directory exists for the router), the forbid_file option was not
6212 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
6213 the existing equality matching.
6215 52. Exim was using ints instead of ino_t variables in some places where it was
6216 dealing with inode numbers.
6218 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
6219 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
6220 called TMPDIR, and if it finds it is different, it changes its value.
6222 54. The smtp_printf() function is now made available to local_scan() so
6223 additional output lines can be written before returning. There is also an
6224 smtp_fflush() function to enable the detection of a dropped connection.
6225 The variables smtp_input and smtp_batched_input are exported to
6228 55. Changed the default runtime configuration: the message "Unknown user"
6229 has been removed from the ACL, and instead placed on the localuser router,
6230 using the cannot_route_message feature. This means that any verification
6231 failures that generate their own messages won't get overridden. Similarly,
6232 the "Unrouteable address" message that was in the ACL for unverifiable
6233 relay addresses has also been removed.
6235 56. Added hosts_avoid_esmtp to the smtp transport.
6237 57. The exicyclog script was not checking for the esoteric option
6238 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
6239 will work only if exicyclog is run under the appropriate euid.
6241 58. Following a discussion on the list, the rules by which Exim recognises line
6242 endings on incoming messages have been changed. The -dropcr and drop_cr
6243 options are now no-ops, retained only for backwards compatibility. The
6244 following line terminators are recognized: LF CRLF CR. However, special
6245 processing applies to CR:
6247 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
6248 nor a local message in the state where . is a terminator.
6250 (ii) If a bare CR is encountered in a header line, an extra space is added
6251 after the line terminator so as not to end the header. The reasoning
6252 behind this is that bare CRs in header lines are most likely either
6253 to be mistakes, or people trying to play silly games.
6255 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
6256 was being incorrectly given as 18 bytes larger than it should have been.
6257 This is a VOB (very old bug).
6259 60. This may never have affected anything current, but just in case it has:
6260 When the local host is found other than at the start of a list of hosts,
6261 the local host, those with the same MX, and any that follow, are discarded.
6262 When the list in question was part of a longer list of hosts, the following
6263 hosts (not currently being processed) were also being discarded. This no
6264 longer happens. I'm not sure if this situation could ever has previously
6267 61. Added the "/MX" feature to lists of hosts in the manualroute and query
6270 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
6271 header. This is something that is recommended in a new Internet Draft, and
6272 is something that is documented as being done by Sendmail. There are two
6273 possible values. For messages generated by the autoreply transport, Exim
6276 Auto-Submitted: auto-replied
6278 whereas for all other generated messages (e.g. bounces) it adds
6280 Auto-Submitted: auto-generated
6282 63. The "personal" condition in filters now includes a test for the
6283 Auto-Submitted: header. If it contains the string "auto-" the message it
6284 not considered personal.
6286 64. Added rcpt_include_affixes as a generic transport option.
6288 65. Added queue_only_override (default true).
6290 66. Added the syslog_duplication option.
6292 67. If what should have been the first header line of a message consisted of
6293 a space followed by a colon, Exim was mis-interpreting it as a header line.
6294 It isn't of course - it is syntactically invalid and should therefore be
6295 treated as the start of the message body. The misbehaviour could have
6296 caused a number of strange effects, including loss of data in subsequent
6297 header lines, and spool format errors.
6299 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
6300 client host had authenticated. This control can now be exercised by an ACL
6301 for more flexibility.
6303 69. By default, callouts do not happen when testing with -bh. There is now a
6304 variant, -bhc, which does actually run the callout code, including
6305 consulting and updating the callout cache.
6307 70. Added support for saslauthd authentication, courtesy of Alexander
6310 71. If statvfs() failed on the spool or log directories while checking their
6311 size for availability, Exim confusingly gave the error "space shortage".
6312 Furthermore, in debugging mode it crashed with a floating point exception.
6313 These checks are done if check_{spool,log}_{space,inodes} are set, and when
6314 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
6315 serious problem, Exim now writes to the main and panic logs when this
6316 happens, with details of the failure. It then refuses to accept the
6317 incoming message, giving the message "spool directory problem" or "log
6318 directory problem" with a 421 code for SMTP messages.
6320 72. When Exim is about to re-exec itself, it ensures that the file descriptors
6321 0, 1, and 2 exist, because some OS complain for execs without them (see
6322 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
6323 descriptors. However, the code omitted to check that the open succeeded,
6324 causing mysterious errors if for some reason the permissions on /dev/null
6325 got screwed. Now Exim writes a message to the main and panic logs, and
6326 bombs out if it can't open /dev/null.
6328 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
6329 interact so that it is all more flexible. It is supposed to remain
6330 backwards compatible. Also added extra_local_interfaces.
6332 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
6333 to bomb out with an assertion failure - to the client this appears as a
6334 connection drop. This problem occurs in the part of the code that was taken
6335 from the Samba project. Fortunately, the assertion is in a very simple
6336 function, so I have fixed this by reproducing the function inline in the
6337 one place where it is called, and arranging for authentication to fail
6338 instead of killing the process with assert().
6340 75. The SPA client code was not working when the server requested OEM rather
6341 than Unicode encoding.
6343 76. Added code to make require_files with a specific uid setting more usable in
6344 the case where statting the file as root fails - usually a non-root-mounted
6345 NFS file system. When this happens and the failure is EACCES, Exim now
6346 forks a subprocess and does the per-uid checking as the relevant uid.
6348 77. Added process_log_path.
6350 78. If log_file_path was not explicitly set, a setting of check_log_space or
6351 check_log_inodes was ignored.
6353 79. If a space check for the spool or log partitions fails, the incident is now
6354 logged. Of course, in the latter case the data may get lost...
6356 80. Added the %p formatting code to string_format() so that it can be used to
6357 print addresses in debug_print(). Adjusted all the address printing in the
6358 debugging in store.c to use %p rather than %d.
6360 81. There was a concern that a line of code in smtp_in.c could overflow a
6361 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
6362 initially expressed, the concern was not well-founded, because trailing
6363 spaces are removed early. However, if the trailing spaces were followed by
6364 a NULL, they did not get removed, so the overflow was possible. Two fixes
6367 (a) I re-wrote the offending code in a cleaner fashion.
6368 (b) If an incoming SMTP command contains a NULL character, it is rejected
6371 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
6372 runs initgroups(), so that if the Exim user is in any additional groups,
6373 they will be used during message reception.
6379 The change log for 4.20 and earlier releases has been archived.
projects / exim.git / blob