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
60 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
63 JH/02 The smtp transport option "multi_domain" is now expanded.
65 JH/03 The smtp transport now requests PRDR by default, if the server offers
68 JH/04 Certificate name checking on server certificates, when exim is a client,
69 is now done by default. The transport option tls_verify_cert_hostnames
70 can be used to disable this per-host. The build option
71 EXPERIMENTAL_CERTNAMES is withdrawn.
73 JH/05 The value of the tls_verify_certificates smtp transport and main options
74 default to the word "system" to access the system default CA bundle.
75 For GnuTLS, only version 3.0.20 or later.
77 JH/06 Verification of the server certificate for a TLS connection is now tried
78 (but not required) by default. The verification status is now logged by
79 default, for both outbound TLS and client-certificate supplying inbound
82 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
85 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
86 Status Notification (bounce) messages are now MIME format per RFC 3464.
87 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
88 under the control of the dsn_advertise_hosts option, and routers may
89 have a dsn_lasthop option.
91 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
92 default, modifiable by a malware= option. The list separator for
93 the options can now be changed in the usual way. Bug 68.
95 JH/10 The smtp_receive_timeout main option is now expanded before use.
97 JH/11 The incoming_interface log option now also enables logging of the
98 local interface on delivery outgoing connections.
100 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
101 if the interface and destination host and port all match.
103 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
106 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
107 Patch from Andrew Lewis.
109 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
110 now supports optional time-restrictions, weighting, and priority
111 modifiers per server. Patch originally by <rommer@active.by>.
113 JH/16 The spamd_address main option now supports a mixed list of local
114 and remote servers. Remote servers can be IPv6 addresses, and
115 specify a port-range.
117 JH/17 Bug 68: The spamd_address main option now supports an optional
118 timeout value per server.
120 JH/18 Bug 1581: Router and transport options headers_add/remove can
121 now have the list separator specified.
123 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
126 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
129 JH/21 Support for the A6 type of dns record is withdrawn.
131 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
132 rather than the verbs used.
134 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
135 from 255 to 1024 chars.
137 JH/24 Verification callouts now attempt to use TLS by default.
139 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
140 are generic router options now. The defaults didn't change.
142 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
143 Original patch from Alexander Shikoff, worked over by JH.
145 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
146 files if ZCAT_COMMAND is not executable.
148 JH/26 Bug 1539: Add timout/retry options on dnsdb lookups.
150 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
152 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
153 Normally benign, it bites when the pair was led to by a CNAME;
154 modern usage is to not canoicalize the domain to a CNAME target
155 (and we were inconsistent anyway for A-only vs AAAA+A).
157 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
159 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
160 when evaluating $sender_host_dnssec.
162 JH/31 Check the HELO verification lookup for DNSSEC, adding new
163 $sender_helo_dnssec variable.
165 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
167 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
169 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
171 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
172 documented as working, but never had. Support all but $spam_report.
174 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
175 added for tls authenticator.
180 TL/01 When running the test suite, the README says that variables such as
181 no_msglog_check are global and can be placed anywhere in a specific
182 test's script, however it was observed that placement needed to be near
183 the beginning for it to behave that way. Changed the runtest perl
184 script to read through the entire script once to detect and set these
185 variables, reset to the beginning of the script, and then run through
186 the script parsing/test process like normal.
188 TL/02 The BSD's have an arc4random API. One of the functions to induce
189 adding randomness was arc4random_stir(), but it has been removed in
190 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
191 function when detected.
193 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
194 cause callback expansion.
196 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
197 syntax errors in an expansion can be treated as a string instead of
198 logging or causing an error, due to the internal use of bool_lax
199 instead of bool when processing it.
201 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
202 server certificates when making smtp deliveries.
204 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
206 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
208 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
210 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
211 Merged patch from Sebastian Wiedenroth.
213 JH/05 Fix results-pipe from transport process. Several recipients, combined
214 with certificate use, exposed issues where response data items split
215 over buffer boundaries were not parsed properly. This eventually
216 resulted in duplicates being sent. This issue only became common enough
217 to notice due to the introduction of conection certificate information,
218 the item size being so much larger. Found and fixed by Wolfgang Breyha.
220 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
221 size buffer was used, resulting in syntax errors when an expansion
224 JH/07 Add support for directories of certificates when compiled with a GnuTLS
225 version 3.3.6 or later.
227 JH/08 Rename the TPDA expermimental facility to Event Actions. The #ifdef
228 is EXPERIMENTAL_EVENT, the main-configuration and transport options
229 both become "event_action", the variables become $event_name, $event_data
230 and $event_defer_errno. There is a new variable $verify_mode, usable in
231 routers, transports and related events. The tls:cert event is now also
232 raised for inbound connections, if the main configuration event_action
235 TL/06 In test suite, disable OCSP for old versions of openssl which contained
236 early OCSP support, but no stapling (appears to be less than 1.0.0).
238 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
239 server certificate names available under the smtp transport option
240 "tls_verify_cert_hostname" now do not permit multi-component wildcard
243 JH/10 Time-related extraction expansions from certificates now use the main
244 option "timezone" setting for output formatting, and are consistent
245 between OpenSSL and GnuTLS compilations. Bug 1541.
247 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
248 encoded parameter in the incoming message. Bug 1558.
250 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
251 include certificate info, eximon was claiming there were spoolfile
254 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
256 JH/14 Log delivery-related information more consistently, using the sequence
257 "H=<name> [<ip>]" wherever possible.
259 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
260 are problematic for Debian distribution, omit them from the release
263 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
265 JH/16 Fix string representation of time values on 64bit time_t anchitectures.
268 JH/17 Fix a null-indirection in certextract expansions when a nondefault
269 output list separator was used.
274 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
275 checkers that were complaining about end of non-void function with no
278 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
279 This was a regression intruduced in 4.83 by another bugfix.
281 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
283 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
284 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
290 TF/01 Correctly close the server side of TLS when forking for delivery.
292 When a message was received over SMTP with TLS, Exim failed to clear up
293 the incoming connection properly after forking off the child process to
294 deliver the message. In some situations the subsequent outgoing
295 delivery connection happened to have the same fd number as the incoming
296 connection previously had. Exim would try to use TLS and fail, logging
297 a "Bad file descriptor" error.
299 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
300 utilities have not been installed.
302 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
303 temporary space as the ACL may create new global variables.
305 TL/01 LDAP support uses per connection or global context settings, depending
306 upon the detected version of the libraries at build time.
308 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
309 to extract and use the src ip:port in logging and expansions as if it
310 were a direct connection from the outside internet. PPv2 support was
311 updated based on HAProxy spec change in May 2014.
313 JH/02 Add ${listextract {number}{list}{success}{fail}}.
315 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
316 Properly escape header and check for NULL return.
318 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
321 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
323 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
324 characters in header names, implemented as a verify condition.
325 Contributed by Michael Fischer v. Mollard.
327 TL/05 Rename SPF condition results err_perm and err_temp to standardized
328 results permerror and temperror. Previous values are deprecated but
329 still accepted. In a future release, err_perm and err_temp will be
330 completely removed, which will be a backward incompatibility if the
331 ACL tests for either of these two old results. Patch contributed by
332 user bes-internal on the mailing list.
334 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
336 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
337 selectors, in both main and reject logs.
339 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
342 JH/07 Add malware type "sock" for talking to simple daemon.
344 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
346 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
347 routers/transports under cutthrough routing.
349 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
350 numbers. Touch up "bool" conditional to keep the same definition.
352 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
354 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
356 JH/12 Expand items in router/transport headers_add or headers_remove lists
357 individually rather than the list as a whole. Bug 1452.
359 Required for reasonable handling of multiple headers_ options when
360 they may be empty; requires that headers_remove items with embedded
361 colons must have them doubled (or the list-separator changed).
363 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
364 view the policy declared in the DMARC record. Currently, $dmarc_status
365 is a combined value of both the record presence and the result of the
368 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
370 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
371 dnslookup router and the smtp transport (applying to the forward
374 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
375 of ldap servers used for a specific lookup. Patch provided by Heiko
378 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
379 New variable $lookup_dnssec_authenticated for observability.
381 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
382 Patch submitted by Lars Timman.
384 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
386 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
387 Requires trusted mode and valid format message id, aborts otherwise.
388 Patch contributed by Heiko Schlichting.
390 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
391 certextract with support for various fields. Bug 1358.
393 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
394 is requested by default, modifiable by smtp transport option
397 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
398 operate on certificate variables to give certificate fingerprints
399 Also new ${sha256:cert_variable}.
401 JH/23 The PRDR feature is moved from being Experimental into the mainline.
403 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
404 Christian Aistleitner.
406 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
408 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
409 file. Patch from Wolfgang Breyha.
411 JH/25 Expand the coverage of the delivery $host and $host_address to
412 client authenticators run in verify callout. Bug 1476.
414 JH/26 Port service names are now accepted for tls_on_connect_ports, to
415 align with daemon_smtp_ports. Bug 72.
417 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
418 support and error reporting did not work properly.
420 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
421 and is readable. Patch from Andrew Colin Kissa.
423 TL/14 Enhance documentation of ${run expansion and how it parses the
424 commandline after expansion, particularly in the case when an
425 unquoted variable expansion results in an empty value.
427 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
429 PP/02 Fix internal collision of T_APL on systems which support RFC3123
430 by renaming away from it. Addresses GH issue 15, reported by
433 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
435 TL/15 SECURITY: prevent double expansion in math comparison functions
436 (can expand unsanitized data). Not remotely exploitable.
443 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
445 PP/02 Make -n do something, by making it not do something.
446 When combined with -bP, the name of an option is not output.
448 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
451 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
452 $sender_host_name and config options to manage this, and basic check
455 PP/05 DSCP support for outbound connections and control modifier for inbound.
457 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
458 (Only plugin which currently uses this is kerberos4, which nobody should
459 be using, but we should make it available and other future plugins might
460 conceivably use it, even though it would break NAT; stuff *should* be
461 using channel bindings instead).
463 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
464 name; added for Sendmail compatibility; requires admin caller.
465 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
466 just ignore it); requires trusted caller.
467 Also parse but ignore: -Ac -Am -X<logfile>
470 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
472 TL/02 Add +smtp_confirmation as a default logging option.
474 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
475 Patch by Magnus Holmgren from 2007-02-20.
477 TL/04 Bugzilla 1281 - Spec typo.
478 Bugzilla 1283 - Spec typo.
479 Bugzilla 1290 - Spec grammar fixes.
481 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
483 TL/06 Add Experimental DMARC support using libopendmarc libraries.
485 TL/07 Fix an out of order global option causing a segfault. Reported to dev
486 mailing list by by Dmitry Isaikin.
488 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
490 JH/02 Support "G" suffix to numbers in ${if comparisons.
492 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
494 NM/01 Bugzilla 1197 - Spec typo
495 Bugzilla 1196 - Spec examples corrections
497 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
499 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
500 gnutls_enable_pkcs11, but renamed to more accurately indicate its
503 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
504 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
506 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
507 "acl {{name}{arg}...}", and optional args on acl condition
510 JH/05 Permit multiple router/transport headers_add/remove lines.
512 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
514 JH/07 Avoid using a waiting database for a single-message-only transport.
515 Performance patch from Paul Fisher. Bugzilla 1262.
517 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
520 JH/09 Add $headers_added variable, with content from use of ACL modifier
521 add_header (but not yet added to the message). Bugzilla 199.
523 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
524 Pulled from Bugzilla 817 by Wolfgang Breyha.
526 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
528 (nb: this is the same fix as in Exim 4.80.1)
530 JH/11 Add A= logging on delivery lines, and a client_set_id option on
533 JH/12 Add optional authenticated_sender logging to A= and a log_selector
536 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
538 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
539 advertise SMTP AUTH mechanism to us, instead of a generic
540 protocol violation error. Also, make Exim more robust to bad
541 data from the Dovecot auth socket.
543 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
545 When a queue runner is handling a message, Exim first routes the
546 recipient addresses, during which it prunes them based on the retry
547 hints database. After that it attempts to deliver the message to
548 any remaining recipients. It then updates the hints database using
551 So if a recipient address works intermittently, it can get repeatedly
552 deferred at routing time. The retry hints record remains fresh so the
553 address never reaches the final cutoff time.
555 This is a fairly common occurrence when a user is bumping up against
556 their storage quota. Exim had some logic in its local delivery code
557 to deal with this. However it did not apply to per-recipient defers
558 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
560 This change adds a proper retry rule check during routing so that the
561 final cutoff time is checked against the message's age. We only do
562 this check if there is an address retry record and there is not a
563 domain retry record; this implies that previous attempts to handle
564 the address had the retry_use_local_parts option turned on. We use
565 this as an approximation for the destination being like a local
566 delivery, as in LMTP.
568 I suspect this new check makes the old local delivery cutoff check
569 redundant, but I have not verified this so I left the code in place.
571 TF/02 Correct gecos expansion when From: is a prefix of the username.
573 Test 0254 submits a message to Exim with the header
577 When I ran the test suite under the user fanf2, Exim expanded
578 the header to contain my full name, whereas it should have added
579 a Resent-Sender: header. It erroneously treats any prefix of the
580 username as equal to the username.
582 This change corrects that bug.
584 GF/01 DCC debug and logging tidyup
585 Error conditions log to paniclog rather than rejectlog.
586 Debug lines prefixed by "DCC: " to remove any ambiguity.
588 TF/03 Avoid unnecessary rebuilds of lookup-related code.
590 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
591 Bug spotted by Jeremy Harris; was flawed since initial commit.
592 Would have resulted in OCSP responses post-SNI triggering an Exim
593 NULL dereference and crash.
595 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
597 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
598 Bug detection, analysis and fix by Samuel Thibault.
599 Bugzilla 1331, Debian bug #698092.
601 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
603 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
604 Server implementation by Todd Lyons, client by JH.
605 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
606 config variable "prdr_enable" controls whether the server
607 advertises the facility. If the client requests PRDR a new
608 acl_data_smtp_prdr ACL is called once for each recipient, after
609 the body content is received and before the acl_smtp_data ACL.
610 The client is controlled by bolth of: a hosts_try_prdr option
611 on the smtp transport, and the server advertisement.
612 Default client logging of deliveries and rejections involving
613 PRDR are flagged with the string "PRDR".
615 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
616 fclose(). Diagnosis by Todd Lyons.
618 PP/17 Update configure.default to handle IPv6 localhost better.
619 Patch by Alain Williams (plus minor tweaks).
622 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
623 This is now consistent with GnuTLS, and is now documented: the
624 previous undocumented portable approach to treating the option as
625 unset was to force an expansion failure. That still works, and
626 an empty string is now equivalent.
628 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
629 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
630 not performing validation itself.
632 PP/20 Added force_command boolean option to pipe transport.
633 Patch from Nick Koston, of cPanel Inc.
635 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
638 TF/04 Added udpsend ACL modifer and hexquote expansion operator
640 PP/21 Fix eximon continuous updating with timestamped log-files.
641 Broken in a format-string cleanup in 4.80, missed when I repaired the
642 other false fix of the same issue.
643 Report and fix from Heiko Schlichting.
646 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
647 Report from Prashanth Katuri.
649 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
650 It's SecureTransport, so affects any MacOS clients which use the
651 system-integrated TLS libraries, including email clients.
653 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
654 using a MIME ACL for non-SMTP local injection.
655 Report and assistance in diagnosis by Warren Baker.
657 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
659 JH/16 Fix comparisons for 64b. Bugzilla 1385.
661 TL/09 Add expansion variable $authenticated_fail_id to keep track of
662 last id that failed so it may be referenced in subsequent ACL's.
664 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
667 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
668 ldap library initialization, allowing self-signed CA's to be
669 used. Also properly sets require_cert option later in code by
670 using NULL (global ldap config) instead of ldap handle (per
671 session). Bug diagnosis and testing by alxgomz.
673 TL/12 Enhanced documentation in the ratelimit.pl script provided in
674 the src/util/ subdirectory.
676 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
677 renamed to Transport Post Delivery Action by Jeremy Harris, as
680 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
681 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
682 redis_servers = needs to be configured which will be used by the redis
683 lookup. Patch from Warren Baker, of The Packet Hub.
685 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
687 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
688 hostname or reverse DNS when processing a host list. Used suggestions
689 from multiple comments on this bug.
691 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
693 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
694 Missed a few lines, added it to make the runtest require no keyboard
697 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
698 contains upper case chars. Make router use caseful_local_part.
700 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
701 support when GnuTLS has been built with p11-kit.
707 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
709 This, or similar/improved, will also be change PP/11 of 4.82.
715 PP/01 Handle short writes when writing local log-files.
716 In practice, only affects FreeBSD (8 onwards).
717 Bugzilla 1053, with thanks to Dmitry Isaikin.
719 NM/01 Bugzilla 949 - Documentation tweak
721 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
724 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
726 PP/02 Implemented gsasl authenticator.
728 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
730 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
731 `pkg-config foo` for cflags/libs.
733 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
734 with rest of GSASL and with heimdal_gssapi.
736 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
737 `pkg-config foo` for cflags/libs for the TLS implementation.
739 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
740 properties get this fed in as external SSF. A number of robustness
741 and debugging improvements to the cyrus_sasl authenticator.
743 PP/08 cyrus_sasl server now expands the server_realm option.
745 PP/09 Bugzilla 1214 - Log authentication information in reject log.
746 Patch by Jeremy Harris.
748 PP/10 Added dbmjz lookup type.
750 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
752 PP/12 MAIL args handles TAB as well as SP, for better interop with
753 non-compliant senders.
754 Analysis and variant patch by Todd Lyons.
756 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
757 Bug report from Lars Müller <lars@samba.org> (via SUSE),
758 Patch from Dirk Mueller <dmueller@suse.com>
760 PP/13 tls_peerdn now print-escaped for spool files.
761 Observed some $tls_peerdn in wild which contained \n, which resulted
762 in spool file corruption.
764 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
765 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
766 or write after TLS renegotiation, which otherwise led to messages
769 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
770 as a tracking header (ie: a signed header comes before the signature).
771 Patch from Wolfgang Breyha.
773 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
774 comma-sep list; embedded commas doubled.
776 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
778 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
780 Report and patch from Dmitry Banschikov.
782 PP/16 Removed "dont_insert_empty_fragments" fron "openssl_options".
783 Removed SSL_clear() after SSL_new() which led to protocol negotiation
784 failures. We appear to now support TLS1.1+ with Exim.
786 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
787 lets Exim select keys and certificates based upon TLS SNI from client.
788 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
789 before an outbound SMTP session. New log_selector, +tls_sni.
791 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
792 NULL dereference. Report and patch from Alun Jones.
794 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
795 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
796 Not seeing resolver debug output on NetBSD, but suspect this is a
797 resolver implementation change.
799 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
800 Left warnings. Added "eximon gdb" invocation mode.
802 PP/21 Defaulting "accept_8bitmime" to true, not false.
804 PP/22 Added -bw for inetd wait mode support.
806 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
807 locate the relevant includes and libraries. Made this the default.
809 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
810 Bugzilla 1246, report and most of solution from Tomasz Kusy.
812 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
813 This may cause build issues on older platforms.
815 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
816 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
817 gnutls_require_mac & gnutls_require_protocols (no longer supported).
818 Added SNI support via GnuTLS too.
819 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
821 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
823 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
824 Applied second patch from Janne, implementing suggestion to default
825 multiple-strings-in-record handling to match SPF spec.
827 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
829 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
830 read-only, out of scope).
831 Patch from Wolfgang Breyha, report from Stuart Northfield.
833 PP/29 Fix three issues highlighted by clang analyser static analysis.
834 Only crash-plausible issue would require the Cambridge-specific
835 iplookup router and a misconfiguration.
836 Report from Marcin Mirosław.
838 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
840 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
841 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
842 As part of this, removing so much warning spew let me fix some minor
843 real issues in debug logging.
845 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
846 assignment on my part. Fixed.
848 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
849 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
850 Janne Snabb (who went above and beyond: thank you).
852 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
853 string otherwise requires a connection and a bunch more work and it's
854 relatively easy to get wrong. Should also expose TLS library linkage
857 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
858 64-bit ${eval} (JH/03).
860 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
861 GNU libc to support some of the 64-bit stuff, should not lead to
862 conflicts. Defined before os.h is pulled in, so if a given platform
863 needs to override this, it can.
865 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
866 protection layer was required, which is not implemented.
867 Bugzilla 1254, patch from Wolfgang Breyha.
869 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
870 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
871 tls_dhparam take prime identifiers. Also unbreak combination of
872 OpenSSL+DH_params+TLSSNI.
874 PP/39 Disable SSLv2 by default in OpenSSL support.
880 PP/01 Solaris build fix for Oracle's LDAP libraries.
881 Bugzilla 1109, patch from Stephen Usher.
883 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
885 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
888 TF/02 Fix a couple more cases where we did not log the error message
889 when unlink() failed. See also change 4.74-TF/03.
891 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
892 lock up or crash if it happened to be inside a call to libc when it
893 got a SIGUSR1 from exiwhat.
895 The SIGUSR1 handler appends the current process status to the process
896 log which is later printed by exiwhat. It used to use the general
897 purpose logging code to do this, but several functions it calls are
898 not safe for signals.
900 The new output code in the SIGUSR1 handler is specific to the process
901 log, and simple enough that it's easy to inspect for signal safety.
902 Removing some special cases also simplifies the general logging code.
903 Removing the spurious timestamps from the process log simplifies
906 TF/04 Improved ratelimit ACL condition.
908 The /noupdate option has been deprecated in favour of /readonly which
909 has clearer semantics. The /leaky, /strict, and /readonly update modes
910 are mutually exclusive. The update mode is no longer included in the
911 database key; it just determines when the database is updated. (This
912 means that when you upgrde Exim will forget old rate measurements.)
914 Exim now checks that the per_* options are used with an update mode that
915 makes sense for the current ACL. For example, when Exim is processing a
916 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
917 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
918 must specify per_mail/readonly. If you omit the update mode it defaults to
919 /leaky where that makes sense (as before) or /readonly where required.
921 The /noupdate option is now undocumented but still supported for
922 backwards compatibility. It is equivalent to /readonly except that in
923 ACLs where /readonly is required you may specify /leaky/noupdate or
924 /strict/noupdate which are treated the same as /readonly.
926 A useful new feature is the /count= option. This is a generalization
927 of the per_byte option, so that you can measure the throughput of other
928 aggregate values. For example, the per_byte option is now equivalent
929 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
931 The per_rcpt option has been generalized using the /count= mechanism
932 (though it's more complicated than the per_byte equivalence). When it is
933 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
934 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
935 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
936 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
937 Note that using per_rcpt with a non-readonly update mode in more than
938 one ACL will cause the recipients to be double-counted. (The per_mail
939 and per_byte options don't have this problem.)
941 The handling of very low rates has changed slightly. If the computed rate
942 is less than the event's count (usually one) then this event is the first
943 after a long gap. In this case the rate is set to the same as this event's
944 count, so that the first message of a spam run is counted properly.
946 The major new feature is a mechanism for counting the rate of unique
947 events. The new per_addr option counts the number of different
948 recipients that someone has sent messages to in the last time period. It
949 behaves like per_rcpt if all the recipient addresses are different, but
950 duplicate recipient addresses do not increase the measured rate. Like
951 the /count= option this is a general mechanism, so the per_addr option
952 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
953 example, measure the rate that a client uses different sender addresses
954 with the options per_mail/unique=$sender_address. There are further
955 details in the main documentation.
957 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
959 TF/06 Removed a few PCRE remnants.
961 TF/07 Automatically extract Exim's version number from tags in the git
962 repository when doing development or release builds.
964 PP/02 Raise smtp_cmd_buffer_size to 16kB.
965 Bugzilla 879. Patch from Paul Fisher.
967 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
968 Heavily based on revision 40f9a89a from Simon Arlott's tree.
971 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
973 PP/05 Variable $av_failed, true if the AV scanner deferred.
974 Bugzilla 1078. Patch from John Horne.
976 PP/06 Stop make process more reliably on build failure.
977 Bugzilla 1087. Patch from Heiko Schlittermann.
979 PP/07 Make maildir_use_size_file an _expandable_ boolean.
980 Bugzilla 1089. Patch from Heiko Schlittermann.
982 PP/08 Handle ${run} returning more data than OS pipe buffer size.
983 Bugzilla 1131. Patch from Holger Weiß.
985 PP/09 Handle IPv6 addresses with SPF.
986 Bugzilla 860. Patch from Wolfgang Breyha.
988 PP/10 GnuTLS: support TLS 1.2 & 1.1.
990 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
993 PP/11 match_* no longer expand right-hand-side by default.
994 New compile-time build option, EXPAND_LISTMATCH_RHS.
995 New expansion conditions, "inlist", "inlisti".
997 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
999 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
1001 PP/14 fix log_write() format string regression from TF/03.
1002 Bugzilla 1152. Patch from Dmitry Isaikin.
1008 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
1010 PP/02 Harmonised TLS library version reporting; only show if debugging.
1011 Layout now matches that introduced for other libraries in 4.74 PP/03.
1013 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
1015 PP/04 New "dns_use_edns0" global option.
1017 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
1020 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
1021 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
1023 TK/01 Updated PolarSSL code to 0.14.2.
1024 Bugzilla 1097. Patch from Andreas Metzler.
1026 PP/07 Catch divide-by-zero in ${eval:...}.
1027 Fixes bugzilla 1102.
1029 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
1032 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
1033 format-string attack -- SECURITY: remote arbitrary code execution.
1035 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
1036 time unintentionally subject to list matching rules, letting the header
1037 cause arbitrary Exim lookups (of items which can occur in lists, *not*
1038 arbitrary string expansion). This allowed for information disclosure.
1040 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
1041 INT_MIN/-1 -- value coerced to INT_MAX.
1047 NM/01 Workround for PCRE version dependency in version reporting
1050 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
1051 This fixes portability to compilers other than gcc, notably
1052 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
1054 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
1055 makefiles for portability to HP-UX and POSIX correctness.
1057 PP/01 Permit LOOKUP_foo enabling on the make command-line.
1058 Also via indented variable definition in the Makefile.
1059 (Debugging by Oliver Heesakkers).
1061 PP/02 Restore caching of spamd results with expanded spamd_address.
1062 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
1064 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
1065 Improves build reliability. Fix from: Frank Elsner
1067 NM/02 Fix wide character breakage in the rfc2047 coding
1068 Fixes bug 1064. Patch from Andrey N. Oktyabrski
1070 NM/03 Allow underscore in dnslist lookups
1071 Fixes bug 1026. Patch from Graeme Fowler
1073 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
1074 Code patches from Adam Ciarcinski of NetBSD.
1076 NM/04 Fixed exiqgrep to cope with mailq missing size issue
1079 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
1080 is logged, to avoid truncation. Patch from John Horne.
1082 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
1083 Patch from Jakob Hirsch.
1085 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
1086 SQL string expansion failure details.
1087 Patch from Andrey Oktyabrski.
1089 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
1090 Patch from Simon Arlott.
1092 PP/09 New lookups functionality failed to compile on old gcc which rejects
1093 extern declarations in function scope.
1094 Patch from Oliver Fleischmann
1096 PP/10 Use sig_atomic_t for flags set from signal handlers.
1097 Check getgroups() return and improve debugging.
1098 Fixed developed for diagnosis in bug 927 (which turned out to be
1101 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
1102 Patch from Mark Zealey.
1104 PP/12 Bugzilla 1056: Improved spamd server selection.
1105 Patch from Mark Zealey.
1107 PP/13 Bugzilla 1086: Deal with maildir quota file races.
1108 Based on patch from Heiko Schlittermann.
1110 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
1111 Patch from Uwe Doering, sign-off by Michael Haardt.
1113 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
1114 variable declaration deep within a block. Bug and patch from
1117 PP/15 lookups-Makefile IRIX compatibilty coercion.
1119 PP/16 Make DISABLE_DKIM build knob functional.
1121 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
1122 Patch by Simon Arlott
1124 TF/03 Fix valgrind.h portability to C89 compilers that do not support
1125 variable argument macros. Our copy now differs from upstream.
1131 TF/01 Failure to get a lock on a hints database can have serious
1132 consequences so log it to the panic log.
1134 TF/02 Log LMTP confirmation messages in the same way as SMTP,
1135 controlled using the smtp_confirmation log selector.
1137 TF/03 Include the error message when we fail to unlink a spool file.
1139 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
1140 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
1141 for maintaining out-of-tree patches for some time.
1143 PP/01 Bugzilla 139: Documentation and portability issues.
1144 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
1145 Handle per-OS dynamic-module compilation flags.
1147 PP/02 Let /dev/null have normal permissions.
1148 The 4.73 fixes were a little too stringent and complained about the
1149 permissions on /dev/null. Exempt it from some checks.
1150 Reported by Andreas M. Kirchwitz.
1152 PP/03 Report version information for many libraries, including
1153 Exim version information for dynamically loaded libraries. Created
1154 version.h, now support a version extension string for distributors
1155 who patch heavily. Dynamic module ABI change.
1157 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
1158 privilege escalation vulnerability whereby the Exim run-time user
1159 can cause root to append content of the attacker's choosing to
1162 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
1165 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
1166 If dropping privileges for untrusted macros, we disabled normal logging
1167 on the basis that it would fail; for the Exim run-time user, this is not
1168 the case, and it resulted in successful deliveries going unlogged.
1169 Fixed. Reported by Andreas Metzler.
1175 PP/01 Date: & Message-Id: revert to normally being appended to a message,
1176 only prepend for the Resent-* case. Fixes regression introduced in
1177 Exim 4.70 by NM/22 for Bugzilla 607.
1179 PP/02 Include check_rfc2047_length in configure.default because we're seeing
1180 increasing numbers of administrators be bitten by this.
1182 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
1184 PP/03 Bugzilla 994: added openssl_options main configuration option.
1186 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
1188 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
1190 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
1192 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
1193 without a peer certificate, leading to a segfault because of an
1194 assumption that peers always have certificates. Be a little more
1195 paranoid. Problem reported by Martin Tscholak.
1197 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
1198 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
1199 NB: ClamAV planning to remove STREAM in "middle of 2010".
1200 CL also introduces -bmalware, various -d+acl logging additions and
1201 more caution in buffer sizes.
1203 PP/09 Implemented reverse_ip expansion operator.
1205 PP/10 Bugzilla 937: provide a "debug" ACL control.
1207 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
1209 PP/12 Bugzilla 973: Implement --version.
1211 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
1213 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
1215 PP/15 Bugzilla 816: support multiple condition rules on Routers.
1217 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
1218 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
1219 ignore trailing whitespace.
1221 JJ/02 prevent non-panic DKIM error from being sent to paniclog
1223 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
1226 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
1227 Notification from Dr Andrew Aitchison.
1229 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
1230 ExtendedDetectionInfo response format.
1231 Notification from John Horne.
1233 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
1236 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
1237 XSL and documented dependency on system catalogs, with examples of how
1240 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
1243 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
1244 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
1245 configuration file which is writeable by the Exim user or group.
1247 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
1248 of configuration files to cover files specified with the -C option if
1249 they are going to be used with root privileges, not just the default
1252 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
1253 option (effectively making it always true).
1255 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
1256 files to be used while preserving root privileges.
1258 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
1259 that rogue child processes cannot use them.
1261 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
1262 run-time user, instead of root.
1264 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
1265 Exim run-time user without dropping privileges.
1267 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
1268 result string, instead of calling string_vformat() twice with the same
1271 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
1272 for other users. Others should always drop root privileges if they use
1273 -C on the command line, even for a whitelisted configure file.
1275 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
1277 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
1283 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
1284 $data_path, and $header_path variables; fixed documentation bugs and
1287 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
1288 exipick to access non-standard spools, including the "frozen" queue
1291 NM/01 Bugzilla 965: Support mysql stored procedures.
1292 Patch from Alain Williams
1294 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
1296 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
1297 Patch from Andreas Metzler
1299 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
1300 Patch from Kirill Miazine
1302 NM/05 Bugzilla 671: Added umask to procmail example.
1304 JJ/03 installed exipick 20100323.0, fixing doc bug
1306 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
1307 directory. Notification and patch from Dan Rosenberg.
1309 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
1311 TK/02 Improve log output when DKIM signing operation fails.
1313 MH/01 Treat the transport option dkim_domain as a colon separated
1314 list, not as a single string, and sign the message with each element,
1315 omitting multiple occurences of the same signer.
1317 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
1318 Bugzilla 985, 986. Patch by Simon Arlott
1320 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
1321 Patch by Simon Arlott
1323 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
1324 MBX locking. Notification from Dan Rosenberg.
1330 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
1332 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
1334 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
1336 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
1338 NM/04 Bugzilla 915: Flag broken perl installation during build.
1344 TK/01 Added patch by Johannes Berg that expands the main option
1345 "spamd_address" if it starts with a dollar sign.
1347 TK/02 Write list of recipients to X-Envelope-Sender header when building
1348 the mbox-format spool file for content scanning (suggested by Jakob
1351 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
1352 (http://www.dcc-servers.net/) support via dccifd. Activated by
1353 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
1355 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
1356 by Mark Daniel Reidel <mr@df.eu>.
1358 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
1359 When building exim an external PCRE library is now needed -
1360 PCRE is a system library on the majority of modern systems.
1361 See entry on PCRE_LIBS in EDITME file.
1363 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
1364 conversation. Added nologin parameter to request.
1365 Patch contributed by Kirill Miazine.
1367 TF/01 Do not log submission mode rewrites if they do not change the address.
1369 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
1371 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
1372 log files in place. Contributed by Roberto Lima.
1374 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
1376 TF/03 Bugzilla 615: When checking the local_parts router precondition
1377 after a local_part_suffix or local_part_prefix option, Exim now
1378 does not use the address's named list lookup cache, since this
1379 contains cached lookups for the whole local part.
1381 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
1382 Robert Millan. Documentation is in experimental-spec.txt.
1384 TF/04 Bugzilla 668: Fix parallel build (make -j).
1386 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
1388 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
1389 Patch provided by Jan Srzednicki.
1391 TF/05 Leading white space used to be stripped from $spam_report which
1392 wrecked the formatting. Now it is preserved.
1394 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
1395 that they are available at delivery time.
1397 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
1399 TF/08 TLS error reporting now respects the incoming_interface and
1400 incoming_port log selectors.
1402 TF/09 Produce a more useful error message if an SMTP transport's hosts
1403 setting expands to an empty string.
1405 NM/06 Bugzilla 744: EXPN did not work under TLS.
1406 Patch provided by Phil Pennock.
1408 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
1409 Patch provided by Richard Godbee.
1411 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
1412 acl_smtp_notquit, added index entry.
1414 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
1415 Patch provided by Eugene Bujak.
1417 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
1418 accept(). Patch provided by Maxim Dounin.
1420 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
1421 Patch provided by Phil Pennock.
1423 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
1425 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
1426 Patch provided by Brad "anomie" Jorsch.
1428 NM/14 Bugzilla 622: Added timeout setting to transport filter.
1429 Patch provided by Dean Brooks.
1431 TK/05 Add native DKIM support (does not depend on external libraries).
1433 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
1434 Patch provided by Graeme Fowler.
1436 NM/16 Bugzilla 851: Documentation example syntax fix.
1438 NM/17 Changed NOTICE file to remove references to embedded PCRE.
1440 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
1443 NM/19 Bugzilla 745: TLS version reporting.
1444 Patch provided by Phil Pennock.
1446 NM/20 Bugzilla 167: bool: condition support.
1447 Patch provided by Phil Pennock.
1449 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
1450 clients. Patch provided by Phil Pennock.
1452 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
1453 Patch provided by Brad "anomie" Jorsch.
1455 NM/23 Bugzilla 687: Fix misparses in eximstats.
1456 Patch provided by Heiko Schlittermann.
1458 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
1459 Patch provided by Heiko Schlittermann.
1461 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
1462 plus update to original patch.
1464 NM/26 Bugzilla 799: Documentation correction for ratelimit.
1466 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
1467 Patch provided by David Brownlee.
1469 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
1471 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
1473 NM/30 Bugzilla 888: TLS documentation bugfixes.
1475 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
1477 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
1478 Unlike the original bugzilla I have changed all shell scripts in src tree.
1480 NM/33 Bugzilla 898: Transport filter timeout fix.
1481 Patch by Todd Rinaldo.
1483 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
1484 Patch by Serge Demonchaux.
1486 NM/35 Bugzilla 39: Base64 decode bug fixes.
1487 Patch by Jakob Hirsch.
1489 NM/36 Bugzilla 909: Correct connect() call in dcc code.
1491 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
1493 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
1495 NM/39 Bugzilla 911: Fixed MakeLinks build script.
1501 TK/01 Add preliminary DKIM support. Currently requires a forked version of
1502 ALT-N's libdkim that I have put here:
1503 http://duncanthrax.net/exim-experimental/
1505 Note to Michael Haardt: I had to rename some vars in sieve.c. They
1506 were called 'true' and it seems that C99 defines that as a reserved
1507 keyword to be used with 'bool' variable types. That means you could
1508 not include C99-style headers which use bools without triggering
1509 build errors in sieve.c.
1511 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
1512 as mailq or other aliases. Changed the --help handling significantly
1513 to do whats expected. exim_usage() emits usage/help information.
1515 SC/01 Added the -bylocaldomain option to eximstats.
1517 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
1519 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
1521 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
1527 PH/01 Another patch from the Sieve maintainer.
1529 PH/02 When an IPv6 address is converted to a string for single-key lookup
1530 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
1531 dots are used instead of colons so that keys in lsearch files need not
1532 contain colons. This was done some time before quoting was made available
1533 in lsearch files. However, iplsearch files do require colons in IPv6 keys
1534 (notated using the quote facility) so as to distinguish them from IPv4
1535 keys. This meant that lookups for IP addresses in host lists did not work
1536 for iplsearch lookups.
1538 This has been fixed by arranging for IPv6 addresses to be expressed with
1539 colons if the lookup type is iplsearch. This is not incompatible, because
1540 previously such lookups could never work.
1542 The situation is now rather anomolous, since one *can* have colons in
1543 ordinary lsearch keys. However, making the change in all cases is
1544 incompatible and would probably break a number of configurations.
1546 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
1549 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
1550 conversion specification without a maximum field width, thereby enabling
1551 a rogue spamd server to cause a buffer overflow. While nobody in their
1552 right mind would setup Exim to query an untrusted spamd server, an
1553 attacker that gains access to a server running spamd could potentially
1554 exploit this vulnerability to run arbitrary code as the Exim user.
1556 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
1557 $primary_hostname instead of what libspf2 thinks the hosts name is.
1559 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
1560 a directory entry by the name of the lookup key. Previously, if a
1561 symlink pointed to a non-existing file or a file in a directory that
1562 Exim lacked permissions to read, a lookup for a key matching that
1563 symlink would fail. Now it is enough that a matching directory entry
1564 exists, symlink or not. (Bugzilla 503.)
1566 PH/03 The body_linecount and body_zerocount variables are now exported in the
1569 PH/04 Added the $dnslist_matched variable.
1571 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
1572 This means they are set thereafter only if the connection becomes
1575 PH/06 Added the client_condition to authenticators so that some can be skipped
1576 by clients under certain conditions.
1578 PH/07 The error message for a badly-placed control=no_multiline_responses left
1579 "_responses" off the end of the name.
1581 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
1583 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
1584 (without spaces) instead of just copying the configuration text.
1586 PH/10 Added the /noupdate option to the ratelimit ACL condition.
1588 PH/11 Added $max_received_linelength.
1590 PH/12 Added +ignore_defer and +include_defer to host lists.
1592 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
1593 way in which PCRE > 7.0 is built.
1595 PH/14 Implemented queue_only_load_latch.
1597 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
1598 MAIL command. The effect was to mangle the value on 64-bit systems.
1600 PH/16 Another patch from the Sieve maintainer.
1602 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
1604 PH/18 If a system quota error occurred while trying to create the file for
1605 a maildir delivery, the message "Mailbox is full" was not appended to the
1606 bounce if the delivery eventually timed out. Change 4.67/27 below applied
1607 only to a quota excession during the actual writing of the file.
1609 PH/19 It seems that peer DN values may contain newlines (and other non-printing
1610 characters?) which causes problems in log lines. The DN values are now
1611 passed through string_printing() before being added to log lines.
1613 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
1614 and InterBase are left for another time.)
1616 PH/21 Added message_body_newlines option.
1618 PH/22 Guard against possible overflow in moan_check_errorcopy().
1620 PH/23 POSIX allows open() to be a macro; guard against that.
1622 PH/24 If the recipient of an error message contained an @ in the local part
1623 (suitably quoted, of course), incorrect values were put in $domain and
1624 $local_part during the evaluation of errors_copy.
1630 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
1631 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
1634 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
1635 issue a MAIL command.
1637 PH/02 In an ACL statement such as
1639 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
1641 if a client was not listed at all, or was listed with a value other than
1642 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
1643 the condition was not true (as it should be), so access was not denied.
1644 The bug was that the ! inversion was incorrectly passed on to the second
1645 item. This has been fixed.
1647 PH/03 Added additional dnslists conditions == and =& which are different from
1648 = and & when the dns lookup returns more than one IP address.
1650 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
1651 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
1653 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
1654 FSYNC, which compiles an option called disable_fsync that allows for
1655 bypassing fsync(). The documentation is heavily laced with warnings.
1657 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
1659 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
1660 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
1661 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
1662 including adding "make clean"; (3) Added -fPIC when compiling the test
1663 dynamically loaded module, to get rid of a warning.
1665 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
1666 message fails, move_frozen_messages = true and ignore_bounce_errors_after
1667 = 0s. The bug is otherwise harmless.
1669 PH/07 There was a bug in the dovecot authenticator such that the value of
1670 $auth1 could be overwritten, and so not correctly preserved, after a
1671 successful authentication. This usually meant that the value preserved by
1672 the server_setid option was incorrect.
1674 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
1676 PH/09 Installed PCRE release 7.0.
1678 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
1679 run for batched SMTP input. It is now run at the start of every message
1680 in the batch. While fixing this I discovered that the process information
1681 (output by running exiwhat) was not always getting set for -bs and -bS
1682 input. This is fixed, and it now also says "batched" for BSMTP.
1684 PH/11 Added control=no_pipelining.
1686 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
1687 patch, slightly modified), and move the expansion of helo_data till after
1688 the connection is made in the smtp transport (so it can use these
1691 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
1693 PH/14 Added log_selector = +pid.
1695 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
1697 PH/16 Add ${if forany and ${if forall.
1699 PH/17 Added dsn_from option to vary the From: line in DSNs.
1701 PH/18 Flush SMTP output before performing a callout, unless control =
1702 no_callout_flush is set.
1704 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
1705 was true (the default) a successful delivery failed to delete the retry
1706 item, thus causing premature timeout of the address. The bug is now
1709 PH/20 Added hosts_avoid_pipelining to the smtp transport.
1711 PH/21 Long custom messages for fakedefer and fakereject are now split up
1712 into multiline reponses in the same way that messages for "deny" and
1713 other ACL rejections are.
1715 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
1716 with slight modification.
1718 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
1719 draft, changing the syntax and factoring some duplicate code".
1721 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
1722 for deliveries of the second and subsequent messages over the same SMTP
1725 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
1726 ${reduce, with only minor "tidies".
1728 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
1730 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
1731 expansion side effects.
1733 PH/27 When a message times out after an over-quota error from an Exim-imposed
1734 quota, the bounce message says "mailbox is full". This message was not
1735 being given when it was a system quota that was exceeded. It now should
1738 MH/03 Made $recipients available in local_scan(). local_scan() already has
1739 better access to the recipient list through recipients_list[], but
1740 $recipients can be useful in postmaster-provided expansion strings.
1742 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
1743 in the case of a MAIL command with additional options following the
1744 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
1745 were accidentally chopped off.
1747 PH/29 SMTP synchronization checks are implemented when a command is read -
1748 there is a check that no more input is waiting when there shouldn't be
1749 any. However, for some commands, a delay in an ACL can mean that it is
1750 some time before the response is written. In this time, more input might
1751 arrive, invalidly. So now there are extra checks after an ACL has run for
1752 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
1753 pipelining has not been advertised.
1755 PH/30 MH's patch to allow iscntrl() characters to be list separators.
1757 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
1758 returned in the SMTP response when smtp_return_error_details was false.
1759 This has been fixed.
1761 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
1762 instead of the C I/O that was originally supplied, because problems were
1763 reported on Solaris.
1765 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
1766 Exim which did not show up earlier: it was assuming that a call to
1767 SSL_CTX_set_info_callback() might give an error value. In fact, there is
1768 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
1769 was a macro that became an assignment, so it seemed to work. This has
1770 changed to a proper function call with a void return, hence the compile
1771 error. Exim's code has been fixed.
1773 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
1776 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
1778 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
1780 PH/37 If a message is not accepted after it has had an id assigned (e.g.
1781 because it turns out to be too big or there is a timeout) there is no
1782 "Completed" line in the log. When some messages of this type were
1783 selected by exigrep, they were listed as "not completed". Others were
1784 picked up by some special patterns. I have improved the selection
1785 criteria to be more general.
1787 PH/38 The host_find_failed option in the manualroute router can now be set
1788 to "ignore", to completely ignore a host whose IP address cannot be
1789 found. If all hosts are ignored, the behaviour is controlled by the new
1790 host_all_ignored option.
1792 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
1793 homing or because of multiple MX records with /mx) generated more than
1794 one IP address, and the following item turned out to be the local host,
1795 all the secondary addresses of the first item were incorrectly removed
1796 from the list, along with the local host and any following hosts (which
1797 is what is supposed to happen).
1799 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
1800 whoever called Exim into the -H file. In the case of the daemon it was
1801 behaving confusingly. When first started, it used values for whoever
1802 started the daemon, but after a SIGHUP it used the Exim user (because it
1803 calls itself on a restart). I have changed the code so that it now always
1806 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
1807 message are rejected with the same error (e.g. no authentication or bad
1808 sender address), and a DATA command is nevertheless sent (as can happen
1809 with PIPELINING or a stupid MUA), the error message that was given to the
1810 RCPT commands is included in the rejection of the DATA command. This is
1811 intended to be helpful for MUAs that show only the final error to their
1814 PH/42 Another patch from the Sieve maintainer.
1816 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
1817 Eximstats - Fixed some broken HTML links and added missing column headers
1819 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
1820 columns for Rejects, Temp Rejects, Ham, and Spam rows.
1822 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
1824 PH/43 Yet another patch from the Sieve maintainer.
1826 PH/44 I found a way to check for a TCP/IP connection going away before sending
1827 the response to the final '.' that terminates a message, but only in the
1828 case where the client has not sent further data following the '.'
1829 (unfortunately, this is allowed). However, in many cases there won't be
1830 any further data because there won't be any more messages to send. A call
1831 to select() can be used: if it shows that the input is "ready", there is
1832 either input waiting, or the socket has been closed. An attempt to read
1833 the next input character can distinguish the two cases. Previously, Exim
1834 would have sent an OK response which the client would never have see.
1835 This could lead to message repetition. This fix should cure that, at
1836 least in a lot of common cases.
1838 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
1839 advertised in response to EHLO.
1845 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
1846 fixed by 4.65/MH/01 (is this a record?) are fixed:
1848 (i) An empty string was always treated as zero by the numeric comparison
1849 operators. This behaviour has been restored.
1851 (ii) It is documented that the numeric comparison operators always treat
1852 their arguments as decimal numbers. This was broken in that numbers
1853 starting with 0 were being interpreted as octal.
1855 While fixing these problems I realized that there was another issue that
1856 hadn't been noticed. Values of message_size_limit (both the global option
1857 and the transport option) were treated as octal if they started with 0.
1858 The documentation was vague. These values are now always treated as
1859 decimal, and I will make that clear in the documentation.
1865 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
1866 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
1869 MH/01 Don't check that the operands of numeric comparison operators are
1870 integers when their expansion is in "skipping" mode (fixes bug
1871 introduced by 4.64-PH/07).
1873 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
1874 child addresses, Exim now panics and dies. Previously, because the count
1875 is held in a short int, deliveries were likely to be lost. As such a
1876 large number of recipients for a single message is ridiculous
1877 (performance will be very, very poor), I have chosen to impose a limit
1878 rather than extend the field.
1884 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
1885 leftover -K file (the existence of which was triggered by #402).
1886 While we were at it, introduced process PID as part of the -K
1887 filename. This should rule out race conditions when creating
1890 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
1891 processing considerably. Previous code took too long for large mails,
1892 triggering a timeout which in turn triggers #401.
1894 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
1895 in the DK code in transports.c. sendfile() is not really portable,
1896 hence the _LINUX specificness.
1898 TF/01 In the add_headers option to the mail command in an Exim filter,
1899 there was a bug that Exim would claim a syntax error in any
1900 header after the first one which had an odd number of characters
1903 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
1904 callout verification, Exim cached a "reject" for the entire domain. This
1905 is correct for most verifications, but it is not correct for a recipient
1906 verification with use_sender or use_postmaster set, because in that case
1907 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
1908 case of MAIL FROM:<> rejection from other early rejections (e.g.
1909 rejection of HELO). When verifying a recipient using a non-null MAIL
1910 address, the cache is ignored if it shows MAIL FROM:<> rejection.
1911 Whatever the result of the callout, the value of the domain cache is
1912 left unchanged (for any other kind of callout, getting as far as trying
1913 RCPT means that the domain itself is ok).
1915 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
1918 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
1919 manifest itself as EPIPE rather than ECONNECT. When tidying away a
1920 session, the daemon ignores ECONNECT errors and logs others; it now
1921 ignores EPIPE as well.
1923 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
1924 (quoted-printable decoding).
1926 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
1927 later the small subsequent patch to fix an introduced bug.
1929 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
1931 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
1933 PH/08 An error is now given if message_size_limit is specified negative.
1935 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
1936 to be given (somewhat) arbitrary names.
1938 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
1941 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
1942 miscellaneous code fixes
1944 PH/10 Added the log_reject_target ACL modifier to specify where to log
1947 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
1948 hostname. This is wrong, because it relates to the incoming message (and
1949 probably the interface on which it is arriving) and not to the outgoing
1950 callout (which could be using a different interface). This has been
1951 changed to use the value of the helo_data option from the smtp transport
1952 instead - this is what is used when a message is actually being sent. If
1953 there is no remote transport (possible with a router that sets up host
1954 addresses), $smtp_active_hostname is used.
1956 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
1957 tweaks were necessary in order to get it to work (see also 21 below):
1958 (a) The code assumed that strncpy() returns a negative number on buffer
1959 overflow, which isn't the case. Replaced with Exim's string_format()
1961 (b) There were several signed/unsigned issues. I just did the minimum
1962 hacking in of casts. There is scope for a larger refactoring.
1963 (c) The code used strcasecmp() which is not a standard C function.
1964 Replaced with Exim's strcmpic() function.
1965 (d) The code set only $1; it now sets $auth1 as well.
1966 (e) A simple test gave the error "authentication client didn't specify
1967 service in request". It would seem that Dovecot has changed its
1968 interface. Fortunately there's a specification; I followed it and
1969 changed what the client sends and it appears to be working now.
1971 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
1974 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
1975 address A is aliased to B and C, where B exists and C does not. Without
1976 -v the output is "A verified" because verification stops after a
1977 successful redirection if more than one address is generated. However,
1978 with -v the child addresses are also verified. Exim was outputting "A
1979 failed to verify" and then showing the successful verification for C,
1980 with its parentage. It now outputs "B failed to verify", showing B's
1981 parentage before showing the successful verification of C.
1983 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
1984 look up a TXT record in a specific list after matching in a combined
1987 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
1988 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
1989 they consult the DNS. I had assumed they would set it the way they
1990 wanted; and indeed my experiments on Linux seem to show that in some
1991 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
1992 To be on the safe side, however, I have now made the interface to
1993 host_find_byname() similar to host_find_bydns(), with an argument
1994 containing the DNS resolver options. The host_find_byname() function now
1995 sets these options at its start, just as host_find_bydns() does. The smtp
1996 transport options dns_qualify_single and dns_search_parents are passed to
1997 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
1998 of host_find_byname() use the default settings of RES_DEFNAMES
1999 (qualify_single) but not RES_DNSRCH (search_parents).
2001 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
2002 spool_read_header() do less string testing, by means of a preliminary
2003 switch on the second character of optional "-foo" lines. (This is
2004 overdue, caused by the large number of possibilities that now exist.
2005 Originally there were few.) While I was there, I also converted the
2006 str(n)cmp tests so they don't re-test the leading "-" and the first
2007 character, in the hope this might squeeze out yet more improvement.
2009 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
2010 flag allowing group syntax was set by the header_syntax check but not
2011 turned off, possible causing trouble later; (2) The flag was not being
2012 set at all for the header_verify test, causing "group"-style headers to
2013 be rejected. I have now set it in this case, and also caused header_
2014 verify to ignore an empty address taken from a group. While doing this, I
2015 came across some other cases where the code for allowing group syntax
2016 while scanning a header line wasn't quite right (mostly, not resetting
2017 the flag correctly in the right place). These bugs could have caused
2018 trouble for malformed header lines. I hope it is now all correct.
2020 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
2021 with the "reply" argument non-NULL. The code, however (which originally
2022 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
2023 but it didn't always do it. This confused somebody who was copying the
2024 code for some other use. I have removed all the tests.
2026 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
2027 feature that was used to support insecure browsers during the U.S. crypto
2028 embargo. It requires special client support, and Exim is probably the
2029 only MTA that supported it -- and would never use it because real RSA is
2030 always available. This code has been removed, because it had the bad
2031 effect of slowing Exim down by computing (never used) parameters for the
2032 RSA_EXPORT functionality.
2034 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
2035 authenticator to fail if there's a tab character in the incoming data
2036 (there should never be unless someone is messing about, as it's supposed
2037 to be base64-encoded). Also added, on Timo's advice, the "secured" option
2038 if the connection is using TLS or if the remote IP is the same as the
2039 local IP, and the "valid-client-cert option" if a client certificate has
2042 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
2043 authenticators. This can be used for authorization after authentication
2044 succeeds. (In the case of plaintext, it servers for both authentication
2047 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
2048 if any retry times were supplied.
2050 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
2051 connection, where there is no client IP address to check. In this
2052 situation, the verify now always succeeds.
2054 PH/25 Applied John Jetmore's -Mset patch.
2056 PH/26 Added -bem to be like -Mset, but loading a message from a file.
2058 PH/27 In a string expansion for a processed (not raw) header when multiple
2059 headers of the same name were present, leading whitespace was being
2060 removed from all of them, but trailing whitespace was being removed only
2061 from the last one. Now trailing whitespace is removed from each header
2062 before concatenation. Completely empty headers in a concatenation (as
2063 before) are ignored.
2065 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
2066 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
2068 PH/29 [Removed. This was a change that I later backed out, and forgot to
2069 correct the ChangeLog entry (that I had efficiently created) before
2070 committing the later change.]
2072 PH/30 Exim was sometimes attempting to deliver messages that had suffered
2073 address errors (4xx response to RCPT) over the same connection as other
2074 messages routed to the same hosts. Such deliveries are always "forced",
2075 so retry times are not inspected. This resulted in far too many retries
2076 for the affected addresses. The effect occurred only when there were more
2077 hosts than the hosts_max_try setting in the smtp transport when it had
2078 the 4xx errors. Those hosts that it had tried were not added to the list
2079 of hosts for which the message was waiting, so if all were tried, there
2080 was no problem. Two fixes have been applied:
2082 (i) If there are any address or message errors in an SMTP delivery, none
2083 of the hosts (tried or untried) are now added to the list of hosts
2084 for which the message is waiting, so the message should not be a
2085 candidate for sending over the same connection that was used for a
2086 successful delivery of some other message. This seems entirely
2087 reasonable: after all the message is NOT "waiting for some host".
2088 This is so "obvious" that I'm not sure why it wasn't done
2089 previously. Hope I haven't missed anything, but it can't do any
2090 harm, as the worst effect is to miss an optimization.
2092 (ii) If, despite (i), such a delivery is accidentally attempted, the
2093 routing retry time is respected, so at least it doesn't keep
2094 hammering the server.
2096 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
2097 in ${readsocket because some servers need this prod.
2099 PH/32 Added some extra debug output when updating a wait-xxx database.
2101 PH/33 The hint "could be header name not terminated by colon", which has been
2102 given for certain expansion errors for a long time, was not being given
2103 for the ${if def:h_colon_omitted{... case.
2105 PH/34 The spec says: "With one important exception, whenever a domain list is
2106 being scanned, $domain contains the subject domain." There was at least
2107 one case where this was not true.
2109 PH/35 The error "getsockname() failed: connection reset by peer" was being
2110 written to the panic log as well as the main log, but it isn't really
2111 panic-worthy as it just means the connection died rather early on. I have
2112 removed the panic log writing for the ECONNRESET error when getsockname()
2115 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
2116 runs only) independently of the message's sender address. This meant
2117 that, if the 4xx error was in fact related to the sender, a different
2118 message to the same recipient with a different sender could confuse
2119 things. In particualar, this can happen when sending to a greylisting
2120 server, but other circumstances could also provoke similar problems.
2121 I have changed the default so that the retry time for these errors is now
2122 based a combination of the sender and recipient addresses. This change
2123 can be overridden by setting address_retry_include_sender=false in the
2126 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
2127 remote server are returned as part of bounce messages. This was not
2128 happening for LMTP over a pipe (the lmtp transport), but now it is the
2129 same for both kinds of LMTP.
2131 PH/38 Despite being documented as not happening, Exim was rewriting addresses
2132 in header lines that were in fact CNAMEs. This is no longer the case.
2134 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
2135 and queue runs started by the daemon processed all messages. This has
2136 been fixed so that -R and -S can now usefully be given with -q<time>.
2138 PH/40 Import PCRE release 6.7 (fixes some bugs).
2140 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
2142 PH/42 Give an error if -q is specified more than once.
2144 PH/43 Renamed the variables $interface_address and $interface_port as
2145 $received_ip_address and $received_port, to make it clear that these
2146 values apply to message reception, and not to the outgoing interface when
2147 a message is delivered. (The old names remain recognized, of course.)
2149 PH/44 There was no timeout on the connect() call when using a Unix domain
2150 socket in the ${readsocket expansion. There now is.
2152 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
2153 be meaningful with "accept".
2155 SC/01 Eximstats V1.43
2156 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
2158 SC/02 Eximstats V1.44
2159 Use a glob alias rather than an array ref in the generated
2160 parser. This improves both readability and performance.
2162 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
2163 Collect SpamAssassin and rejection statistics.
2164 Don't display local sender or destination tables unless
2165 there is data to show.
2166 Added average volumes into the top table text output.
2168 SC/04 Eximstats V1.46
2169 Collect data on the number of addresses (recipients)
2170 as well as the number of messages.
2172 SC/05 Eximstats V1.47
2173 Added 'Message too big' to the list of mail rejection
2174 reasons (thanks to Marco Gaiarin).
2176 SC/06 Eximstats V1.48
2177 Mainlog lines which have GMT offsets and are too short to
2178 have a flag are now skipped.
2180 SC/07 Eximstats V1.49 (Alain Williams)
2181 Added the -emptyok flag.
2183 SC/08 Eximstats V1.50
2184 Fixes for obtaining the IP address from reject messages.
2186 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
2187 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
2188 whitesspace changes from 4.64-PH/27
2190 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
2193 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
2194 are found, allow negative numbers in numeric criteria)
2196 JJ/06 exipick.20061117.2, added new $message_body_missing variable
2198 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
2199 to match changes made in 4.64-PH/43
2201 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
2203 PH/47 Put in an explicit test for a DNS lookup of an address record where the
2204 "domain" is actually an IP address, and force a failure. This locks out
2205 those revolvers/nameservers that support "A-for-A" lookups, in
2206 contravention of the specifications.
2208 PH/48 When a host name was looked up from an IP address, and the subsequent
2209 forward lookup of the name timed out, the host name was left in
2210 $sender_host_name, contrary to the specification.
2212 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
2213 restricted to single-key lookups, Exim was not diagnosing an error if
2214 * or *@ was used with a query-style lookup.
2216 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
2218 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
2219 long ago, but noone interested enough thought of it. Let's just say that
2220 the "1.1" means that there are some new functions that weren't there at
2221 some point in the past.
2223 PH/51 Error processing for expansion failure of helo_data from an smtp
2224 transport during callout processing was broken.
2226 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
2227 tested/used via the -bh/-bhc/-bs options.
2229 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
2230 bug, fixed in subsequent PCRE releases).
2232 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
2233 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
2235 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
2241 SC/01 Use a glob alias rather than an array ref in eximstats generated
2242 parser. This improves both readability and performance.
2244 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
2245 Don't display local sender or destination tables in eximstats unless
2246 there is data to show.
2247 Added average volumes into the eximstats top table text output.
2249 SC/03 Collect data on the number of addresses (recipients) as well
2250 as the number of messages in eximstats.
2252 TF/01 Correct an error in the documentation for the redirect router. Exim
2253 does (usually) call initgroups() when daemonizing.
2255 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
2256 with consistent privilege compared to when running as a daemon.
2258 TF/03 Note in the spec that $authenticated_id is not set for local
2259 submissions from trusted users.
2261 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
2262 Thanks to Dean Brooks <dean@iglou.com> for the patch.
2264 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
2265 by adding some example configuration directives to the default
2266 configuration file. A little bit of work is required to uncomment the
2267 directives and define how usernames and passwords are checked, but
2268 there is now a framework to start from.
2270 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
2271 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
2272 without this. I don't know how relevant this is to other LDAP libraries.
2274 PH/02 Add the verb name to the "unknown ACL verb" error.
2276 PH/03 Magnus Holmgren's patch for filter_prepend_home.
2278 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
2280 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
2281 directory not expanded when it should be if an expanded home directory
2282 was set for the address (which is overridden by the transport).
2284 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
2287 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
2288 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
2289 because it is too late at that time, and has no effect.
2291 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
2292 security issue with \' (bugzilla #107). I could not use the
2293 PQescapeStringConn() function, because it needs a PGconn value as one of
2296 PH/08 When testing addresses using -bt, indicate those final addresses that
2297 are duplicates that would not cause an additional delivery. At least one
2298 person was confused, thinking that -bt output corresponded to deliveries.
2299 (Suppressing duplicates isn't a good idea as you lose the information
2300 about possibly different redirections that led to the duplicates.)
2302 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
2303 systems where poll() doesn't work, in particular OS X.
2305 PH/10 Added more information to debugging output for retry time not reached.
2307 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
2308 operations in malware.c.
2310 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
2313 PH/13 If write_rejectlog was set false when logging was sent to syslog with
2314 syslog_duplication set false, log lines that would normally be written
2315 both the the main log and to the reject log were not written to syslog at
2318 PH/14 In the default configuration, change the use of "message" in ACL warn
2319 statements to "add_header".
2321 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
2322 not followed by a command (e.g. "seen endif").
2324 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
2325 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
2328 PH/17 Added extra conditions to the default value of delay_warning_condition
2332 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
2333 { match{$h_precedence:}{(?i)bulk|list|junk} } \
2334 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
2337 The Auto-Submitted: and various List- headers are standardised, whereas I
2338 don't think Precedence: ever was.
2340 PH/18 Refactored debugging code in route_finduser() to show more information,
2341 in particular, the error code if getpwnam() issues one.
2343 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
2344 This is apparently needed in addition to the PH/07 change above to avoid
2345 any possible encoding problems.
2347 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
2348 but not after initializing Perl.
2350 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
2351 output them only if debugging. By default they are written stderr,
2352 apparently, which is not desirable.
2354 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
2357 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
2360 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
2362 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
2363 authenticated or an ident call has been made. Suppress the default
2364 values for $authenticated_id and $authenticated_sender (but permit -oMai
2365 and -oMas) when testing with -bh.
2367 PH/24 Re-jigged the order of the tests in the default configuration so that the
2368 tests for valid domains and recipients precede the DNS black list and CSA
2369 tests, on the grounds that those ones are more expensive.
2371 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
2372 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
2373 This bug exists in every version of Exim that I still have, right back to
2376 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
2377 However, an attempt to turn on case-sensitivity in a regex key by
2378 including (?-i) didn't work because the subject string was already
2379 lowercased, and the effects were non-intuitive. It turns out that a
2380 one-line patch can be used to allow (?-i) to work as expected.
2386 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
2387 other effects) broke the use of negated acl sub-conditions.
2389 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
2392 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
2393 "Deny" causes Exim to reject the incoming connection with a 554 error.
2394 Unfortunately, if there is a major crisis, such as a disk failure,
2395 tcp-wrappers gives "deny", whereas what one would like would be some
2396 kind of temporary error. A kludge has been added to help with this.
2397 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
2398 554 error is used if errno is still zero or contains ENOENT (which occurs
2399 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
2402 PH/03 Add -lutil to the default FreeBSD LIBS setting.
2404 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
2405 errors. Otherwise a message that provokes a temporary error (when other
2406 messages do not) can cause a whole host to time out.
2408 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
2409 addresses were routed directly to files or pipes from a redirect router.
2410 File deliveries just didn't batch; pipe deliveries might have suffered
2413 PH/06 A failure to get a lock for a hints database would erroneously always say
2414 "Failed to get write lock", even when it was really a read lock.
2416 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
2417 of 0600. This has been changed to use the value of the lockfile_mode
2418 option (which defaults to 0600).
2420 PH/08 Applied small patch from the Sieve maintainer.
2422 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
2423 folder from quota calculations, a direct delivery into this folder messed
2424 up the contents of the maildirsize file. This was because the regex was
2425 used only to exclude .Trash (or whatever) when the size of the mailbox
2426 was calculated. There was no check that a delivery was happening into an
2427 excluded directory. This bug has been fixed by ignoring all quota
2428 processing for deliveries into excluded directories.
2430 PH/10 Added the maildirfolder_create_regex option to appendfile.
2436 PH/01 The code for finding all the local interface addresses on a FreeBSD
2437 system running IPv6 was broken. This may well have applied to all BSD
2438 systems, as well as to others that have similar system calls. The broken
2439 code found IPv4 interfaces correctly, but gave incorrect values for the
2440 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
2441 that it would not match correctly against @[] and not recognize the IPv6
2444 PH/02 The ipliteral router was not recognizing addresses of the form user@
2445 [ipv6:....] because it didn't know about the "ipv6:" prefix.
2447 PH/03 Added disable_ipv6.
2449 PH/04 Changed $reply_address to use the raw form of the headers instead of the
2450 decoded form, because it is most often used to construct To: headers
2451 lines in autoreplies, and the decoded form may well be syntactically
2452 invalid. However, $reply_address has leading white space removed, and all
2453 newlines turned into spaces so that the autoreply transport does not
2456 PH/05 If group was specified without a user on a router, and no group or user
2457 was specified on a transport, the group from the router was ignored.
2459 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
2460 for visible compile-time settings that can be used to change these
2461 numbers, for those that want even more. Backwards compatibility with old
2462 spool files has been maintained. However, going back to a previous Exim
2463 release will lost any variables that are in spool files.
2465 PH/07 Two small changes when running in the test harness: increase delay when
2466 passing a TCP/IP connection to a new process, in case the original
2467 process has to generate a bounce, and remove special handling of
2468 127.0.0.2 (sic), which is no longer necessary.
2470 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
2471 be the same on different OS.
2473 PH/09 Moved a debug statement in filter processing to avoid a race problem when
2476 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
2477 whether --show-vars was specified or not
2479 JJ/02 exipick: Added support for new ACL variable spool format introduced
2482 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
2483 syntactically invalid From: or Reply-to: line, and a filter used this to
2484 generate an autoreply, and therefore failed to obtain an address for the
2485 autoreply, Exim could try to deliver to a non-existent relative file
2486 name, causing unrelated and misleading errors. What now happens is that
2487 it logs this as a hard delivery error, but does not attempt to create a
2490 PH/11 The exinext utility has a -C option for testing purposes, but although
2491 the given file was scanned by exinext itself; it wasn't being passed on
2492 when Exim was called.
2494 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
2495 an end-of-file indication when reading a command response.
2497 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
2498 compiled. In many other places in Exim, IPv6 addresses are always
2499 recognized, so I have changed this. It also means that IPv4 domain
2500 literals of the form [IPV4:n.n.n.n] are now always recognized.
2502 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
2503 used if the router is not running as root, for example, when verifying at
2504 ACL time, or when using -bh. The debugging output from this situation was
2505 non-existent - all you got was a failure to exec. I have made two
2508 (a) Failures to set uid/gid, the current directory, or a process leader
2509 in a subprocess such as that created by queryprogram now generate
2510 suitable debugging ouput when -d is set.
2512 (b) The queryprogram router detects when it is not running as root,
2513 outputs suitable debugging information if -d is set, and then runs
2514 the subprocess without attempting to change uid/gid.
2516 PH/15 Minor change to Makefile for building test_host (undocumented testing
2519 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
2520 additional section of a DNS packet that returns MX or SRV records.
2521 Instead, it always explicitly searches for A/AAAA records. This avoids
2522 major problems that occur when a DNS server includes only records of one
2523 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
2524 fixed another bug: if SRV records were looked up and the corresponding
2525 address records were *not* found in the additional section, the port
2526 values from the SRV records were lost.
2528 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
2529 using the correct key (the original address) when searching the retry
2530 rules in order to find which one to use for generating the retry hint.
2532 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
2533 adding the default one. Similarly, if it contains a Reply-To: header, the
2534 errors_reply_to option, if set, is not used.
2536 PH/19 When calculating a retry time, Exim used to measure the "time since
2537 failure" by looking at the "first failed" field in the retry record. Now
2538 it does not use this if it is later than than the arrival time of the
2539 message. Instead it uses the arrival time. This makes for better
2540 behaviour in cases where some deliveries succeed, thus re-setting the
2541 "first failed" field. An example is a quota failure for a huge message
2542 when small messages continue to be delivered. Without this change, the
2543 "time since failure" will always be short, possible causing more frequent
2544 delivery attempts for the huge message than are intended.
2545 [Note: This change was subsequently modified - see PH/04 for 4.62.]
2547 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
2548 $1, $2, $3) because the numerical variables can be reset during some
2549 expansion items (e.g. "match"), thereby losing the authentication data.
2551 PH/21 Make -bV show the size of off_t variables so that the test suite can
2552 decide whether to run tests for quotas > 2G.
2554 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
2555 mailbox_size, and mailbox_filecount in the appendfile transport. If a
2556 filecount value is greater than 2G or if a quota value is greater than 2G
2557 on a system where the size of off_t is not greater than 4, a panic error
2560 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
2561 never match. The debug and -bh output now contains an explicit error
2562 message indicating a malformed IPv4 address or mask.
2564 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
2565 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
2566 PH/23 above applies.
2568 PH/25 Do not write to syslog when running in the test harness. The only
2569 occasion when this arises is a failure to open the main or panic logs
2570 (for which there is an explicit test).
2572 PH/26 Added the /no_tell option to "control=freeze".
2574 PH/27 If a host name lookup failed very early in a connection, for example, if
2575 the IP address matched host_lookup and the reverse lookup yielded a name
2576 that did not have a forward lookup, an error message of the form "no IP
2577 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
2578 could be logged. Now it outputs the IP address instead of "NULL".
2580 PH/28 An enabling patch from MH: add new function child_open_exim2() which
2581 allows the sender and the authenticated sender to be set when
2582 submitting a message from within Exim. Since child_open_exim() is
2583 documented for local_scan(), the new function should be too.
2585 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
2586 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
2587 results in an empty string is now treated as unset.
2589 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
2591 PH/31 Added sender_verify_fail logging option.
2593 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
2594 needed by RFC 822 but not by RFC 2822 was commented out. I have now
2595 tidied the source and removed it altogether.
2597 PH/33 When a queue run was abandoned because the load average was too high, a
2598 log line was always written; now it is written only if the queue_run log
2599 selector is set. In addition, the log line for abandonment now contains
2600 information about the queue run such as the pid. This is always present
2601 in "start" and "stop" lines but was omitted from the "abandon" line.
2603 PH/34 Omit spaces between a header name and the colon in the error message that
2604 is given when verify = headers_syntax fails (if there are lots of them,
2605 the message gets confusing).
2607 PH/35 Change the default for dns_check_names_pattern to allow slashes within
2608 names, as there are now some PTR records that contain slashes. This check
2609 is only to protect against broken name servers that fall over on strange
2610 characters, so the fact that it applies to all lookups doesn't matter.
2612 PH/36 Now that the new test suite is complete, we can remove some of the
2613 special code in Exim that was needed for the old test suite. For example,
2614 sorting DNS records because real resolvers return them in an arbitrary
2615 order. The new test suite's fake resolver always returns records in the
2618 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
2619 bounces) except when queue_only is set, to avoid logging races between
2620 the different processes.
2622 PH/38 Panic-die if .include specifies a non-absolute path.
2624 PH/39 A tweak to the "H" retry rule from its user.
2626 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
2627 a label. They prevented compilation on older perls.
2629 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
2630 a warning to be raised on newish perls.
2632 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
2633 on queue. Changes to match documented behaviour of showing count of
2634 messages matching specified criteria.
2636 PH/40 Changed the default ident timeout from 30s to 5s.
2638 PH/41 Added support for the use of login_cap features, on those BSD systems
2639 that have them, for controlling the resources used by pipe deliveries.
2641 PH/42 The content-scanning code uses fopen() to create files in which to put
2642 message data. Previously it was not paying any attention to the mode of
2643 the files. Exim runs with umask(0) because the rest of the code creates
2644 files with open(), and sets the required mode explicitly. Thus, these
2645 files were ending up world-writeable. This was not a big issue, because,
2646 being within the spool directory, they were not world-accessible. I have
2647 created a function called modefopen, which takes an additional mode
2648 argument. It sets umask(777), creates the file, chmods it to the required
2649 mode, then resets the umask. All the relevant calls to fopen() in the
2650 content scanning code have been changed to use this function.
2652 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
2653 to 24 hours. This avoids potential overflow problems when processing G
2654 and H retry rules. I suspect nobody ever tinkers with this value.
2656 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
2658 PH/45 When the plaintext authenticator is running as a client, the server's
2659 challenges are checked to ensure they are valid base64 strings. By
2660 default, the authentication attempt is cancelled if an invalid string is
2661 received. Setting client_ignore_invalid_base64 true ignores these errors.
2662 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
2663 they are received. Thus, the responses can be made to depend on the
2664 challenges. If an invalid string is ignored, an empty string is placed in
2667 PH/46 Messages that are created by the autoreply transport now contains a
2668 References: header, in accordance with RFCs 2822 and 3834.
2670 PH/47 Added authenticated_sender_force to the smtp transport.
2672 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
2674 PH/49 Installed latest patch from the Sieve maintainer.
2676 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
2677 was also set, the appendfile transport was unnecessarily scanning a
2678 directory of message files (e.g. for maildir delivery) to find the count
2679 of files (along with the size), even though it did not need this
2680 information. It now does the scan only if it needs to find either the
2681 size of the count of files.
2683 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
2685 PH/52 Two bugs concerned with error handling when the smtp transport is
2688 (i) Exim was not creating retry information for temporary errors given
2689 for individual recipients after the DATA command when the smtp transport
2690 was used in LMTP mode. This meant that they could be retried too
2691 frequently, and not timed out correctly.
2693 (ii) Exim was setting the flag that allows error details to be returned
2694 for LMTP errors on RCPT commands, but not for LMTP errors for individual
2695 recipients that were returned after the DATA command.
2697 PH/53 This is related to PH/52, but is more general: for any failing address,
2698 when detailed error information was permitted to be returned to the
2699 sender, but the error was temporary, then after the final timeout, only
2700 "retry timeout exceeded" was returned. Now it returns the full error as
2701 well as "retry timeout exceeded".
2703 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
2704 do this, and (what is worse) MTAs that accept it.
2706 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
2707 will now be deprecated.
2709 PH/56 New os.c-cygwin from the Cygwin maintainer.
2711 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
2712 formats (previously only available in exim formats via -bpr, -bpru,
2713 and -bpra. Now also available in native and exiqgrep formats)
2715 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
2716 with very large, slow to parse queues
2718 JJ/08 exipick: added ! as generic prefix to negate any criteria format
2720 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
2722 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
2723 responses to authentication challenges, though it was showing the
2724 challenges; (ii) I've removed the CR characters from the debug output for
2727 PH/58 Allow for the insertion of a newline as well as a space when a string
2728 is turned into more than one encoded-word during RFC 2047 encoding. The
2729 Sieve code now uses this.
2731 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
2732 data_4xx, lost_connection, tls_required.
2734 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
2735 message was being sent as an SMTP response.
2737 PH/61 Add -l and -k options to exicyclog.
2739 PH/62 When verifying, if an address was redirected to one new address, so that
2740 verification continued, and the new address failed or deferred after
2741 having set something in $address_data, the value of $address_data was not
2742 passed back to the ACL. This was different to the case when no
2743 redirection occurred. The value is now passed back in both cases.
2745 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
2746 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
2747 use login_cap.h, so on its own it isn't the distinguishing feature. The
2748 new name refers directly to the setclassresources() function.
2750 PH/65 Added configuration files for NetBSD3.
2752 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
2754 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
2755 is preferred over IPv4.
2757 PH/68 The bounce_return_message and bounce_return_body options were not being
2758 honoured for bounces generated during the reception of non-SMTP messages.
2759 In particular, this applied to messages rejected by the ACL. This bug has
2760 been fixed. However, if bounce_return_message is true and bounce_return_
2761 body is false, the headers that are returned for a non-SMTP message
2762 include only those that have been read before the error was detected.
2763 (In the case of an ACL rejection, they have all been read.)
2765 PH/69 The HTML version of the specification is now built in a directory called
2766 spec_html instead of spec.html, because the latter looks like a path with
2767 a MIME-type, and this confuses some software.
2769 PH/70 Catch two compiler warnings in sieve.c.
2771 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
2772 function verify_get_ident() calls ip_connect() to connect a socket, but
2773 if the "connect()" function timed out, ip_connect() used to close the
2774 socket. However, verify_get_ident() also closes the socket later, and in
2775 between Exim writes to the log, which may get opened at this point. When
2776 the socket was closed in ip_connect(), the log could get the same file
2777 descriptor number as the socket. This naturally causes chaos. The fix is
2778 not to close the socket in ip_connect(); the socket should be closed by
2779 the function that creates it. There was only one place in the code where
2780 this was missing, in the iplookup router, which I don't think anybody now
2781 uses, but I've fixed it anyway.
2783 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
2784 well as to direct DNS lookups. Otherwise the handling of names in host
2785 lists is inconsistent and therefore confusing.
2791 PH/01 Two changes to the default runtime configuration:
2793 (1) Move the checks for relay_from_hosts and authenticated clients from
2794 after to before the (commented out) DNS black list checks.
2796 (2) Add control=submission to the relay_from_hosts and authenticated
2797 clients checks, on the grounds that messages accepted by these
2798 statements are most likely to be submissions.
2800 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
2802 (1) Generate an error if the third argument for the ${prvs expansion is
2805 (2) Treat a missing third argument of ${prvscheck as if it were an empty
2808 (3) Reset the variables that are obtained from the first argument of
2809 ${prvscheck and used in the second argument before leaving the code,
2810 because their memory is reclaimed, so using them afterwards may do
2813 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
2814 one (it's much easier than Tom thought :-).
2816 (5) Because of (4), we can now allow for the use of $prvscheck_result
2817 inside the third argument.
2819 PH/03 For some reason, the default setting of PATH when running a command from
2820 a pipe transport was just "/usr/bin". I have changed it to
2823 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
2824 anything to be listed in the output from -bV.
2826 PH/05 When a filter generated an autoreply, the entire To: header line was
2827 quoted in the delivery log line, like this:
2829 => >A.N.Other <ano@some.domain> <original@ddress> ...
2831 This has been changed so that it extracts the operative address. There
2832 may be more than one such address. If so, they are comma-separated, like
2835 => >ano@some.domain,ona@other.domain <original@ddress> ...
2837 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
2838 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
2839 not being looked up in the rDNS to get a host name, Exim was showing the
2840 IP address twice in Received: lines, even though the IP addresses were
2841 identical. For example:
2843 Received: from [1.2.3.4] (helo=[1.2.3.4])
2845 However, if the real host name was known, it was omitting the HELO data
2846 if it matched the actual IP address. This has been tidied up so that it
2847 doesn't show the same IP address twice.
2849 PH/07 When both +timestamp and +memory debugging was on, the value given by
2850 $tod_xxx expansions could be wrong, because the tod_stamp() function was
2851 called by the debug printing, thereby overwriting the timestamp buffer.
2852 Debugging no longer uses the tod_stamp() function when +timestamp is set.
2854 PH/08 When the original message was included in an autoreply transport, it
2855 always said "this is a copy of the message, including all the headers",
2856 even if body_only or headers_only was set. It now gives an appropriate
2859 PH/09 Applied a patch from the Sieve maintainer which:
2861 o fixes some comments
2862 o adds the (disabled) notify extension core
2863 o adds some debug output for the result of if/elsif tests
2864 o points to the current vacation draft in the documentation
2865 and documents the missing references header update
2869 o fixes a bug in processing the envelope test (when testing
2870 multiple envelope elements, the last element determinted the
2873 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
2874 Electronic Mail") by including:
2876 Auto-submitted: auto-generated
2878 in the messages that it generates (bounce messages and others, such as
2879 warnings). In the case of bounce messages for non-SMTP mesages, there was
2880 also a typo: it was using "Auto_submitted" (underscore instead of
2881 hyphen). Since every message generated by Exim is necessarily in response
2882 to another message, thes have all been changed to:
2884 Auto-Submitted: auto-replied
2886 in accordance with these statements in the RFC:
2888 The auto-replied keyword:
2890 - SHOULD be used on messages sent in direct response to another
2891 message by an automatic process,
2893 - MUST NOT be used on manually-generated messages,
2895 - MAY be used on Delivery Status Notifications (DSNs) and Message
2896 Disposition Notifications (MDNs),
2898 - MUST NOT be used on messages generated by automatic or periodic
2899 processes, except for messages which are automatic responses to
2902 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
2903 to the default Received: header definition.
2905 PH/12 Added log selector acl_warn_skipped (default on).
2907 PH/13 After a successful wildlsearch lookup, discard the values of numeric
2908 variables because (a) they are in the wrong storage pool and (b) even if
2909 they were copied, it wouldn't work properly because of the caching.
2911 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
2912 checking when decoding. Apparently there are clients that generate
2913 overlong encoded strings. Why am I not surprised?
2915 PH/15 If the first argument of "${if match_address" was not empty, but did not
2916 contain an "@" character, Exim crashed. Now it writes a panic log message
2917 and treats the condition as false.
2919 PH/16 In autoreply, treat an empty string for "once" the same as unset.
2921 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
2922 extension "envelope-auth". The code is finished and in agreement with
2923 other implementations, but there is no documentation so far and in fact,
2924 nobody wrote the draft yet. This extension is currently #undef'ed, thus
2925 not changing the active code.
2927 Print executed "if" and "elsif" statements when debugging is used. This
2928 helps a great deal to understand what a filter does.
2930 Document more things not specified clearly in RFC3028. I had all this
2931 sorted out, when out of a sudden new issues came to my mind. Oops."
2933 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
2936 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
2937 canonical form (as documented). However, after a host name lookup from
2938 the IP address, check_host() was doing a simple string comparison with
2939 addresses acquired from the DNS when checking that the found name did
2940 have the original IP as one of its addresses. Since any found IPv6
2941 addresses are likely to be in abbreviated form, the comparison could
2942 fail. Luckily, there already exists a function for doing the comparison
2943 by converting both addresses to binary, so now that is used instead of
2944 the text comparison.
2946 PH/20 There was another similar case to PH/19, when a complete host name was
2947 given in a host list; looking up its IP address could give an abbreviated
2948 form, whereas the current host's name might or might not be abbreviated.
2949 The same fix has been applied.
2955 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
2956 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
2959 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
2960 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
2962 PH/03 Typo: missing ".o" in src/pcre/Makefile.
2964 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
2965 header line, restrict the check to what is listed in RFCs 2369 and 2929.
2966 Also, for "Auto-Submitted", treat anything other than "no" as
2967 non-personal, in accordance with RFC 3834. (Previously it treated
2968 anything starting "auto-" as non-personal.)
2970 TF/01 The control=submission/name=... option had a problem with syntax
2971 errors if the name included a slash character. The /name= option
2972 now slurps the rest of the string, so it can include any characters
2973 but it must come last in the list of options (after /sender_retain
2976 PH/05 Some modifications to the interface to the fake nameserver for the new
2984 TK/01 Added the "success_on_redirect" address verification option. See
2985 NewStuff for rationale and an example.
2987 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
2989 PH/02 Patch to exigrep to allow it to work on syslog lines.
2991 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
2992 fread() to skip over the body file's header line, because in Cygwin the
2993 header line is locked and is inaccessible.
2995 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
2996 co-exist for some time) to make it clear that it is the Exim ID that is
2997 referenced, not the Message-ID: header line.
2999 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
3000 string_format() function, because snprintf() does not exist on all
3003 PH/06 The use of forbid_filter_existstest now also locks out the use of the
3004 ${stat: expansion item.
3006 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
3007 protocol synchronization error", to keep the pedants happy.
3009 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
3010 well as for IRIX systems, when gcc is being used. See the host.c source
3013 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
3015 PH/10 Named domain lists were not working if used in a queue_smtp_domains
3018 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
3019 transport and to the smtp transport in LMTP mode.
3021 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
3023 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
3024 run a filter in a subprocess. This could lead to confusion in subsequent
3025 lookups in the parent process. There should also be a search_tidyup() at
3026 the end of the subprocess.
3028 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
3029 only if the host matched helo_try_verify_hosts, which caused the
3030 verification to occur when the EHLO/HELO command was issued. The ACL just
3031 tested the remembered result. Now, if a previous verification attempt has
3032 not happened, "verify = helo" does it there and then.
3034 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
3036 TK/03 Fix log output including CR from clamd.
3038 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
3039 exist provoked a memory error which could cause a segfault.
3041 PH/15 Installed PCRE 6.2
3043 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
3045 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
3046 of the problem. Specifically, suggested +O2 rather than +O1 for the
3049 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
3051 PH/20 If a delivery was routed to a non-standard port by means of an SRV
3052 record, the port was not correctly logged when the outgoing_port log
3053 selector was set (it logged the transort's default port).
3055 PH/21 Added support for host-specific ports to manualroute, queryprogram,
3056 fallback_hosts, and "hosts" in the smtp transport.
3058 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
3059 host errors such as "Connection refused".
3061 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
3062 authentication with radiusclient 0.4.9:
3064 - Error returned from rc_read_config was caught wrongly
3065 - Username/password not passed on to radius server due to wrong length.
3067 The presumption is that some radiusclient API changes for 4.51/PH/17
3068 were not taken care of correctly. The code is still untested by me (my
3069 Linux distribution still has 0.3.2 of radiusclient), but it was
3070 contributed by a Radius user.
3072 PH/24 When doing a callout, the value of $domain wasn't set correctly when
3073 expanding the "port" option of the smtp transport.
3075 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
3076 while reading a MIME header. Thanks to Tom Hughes for a patch.
3078 PH/24 Include config.h inside local_scan.h so that configuration settings are
3081 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
3082 that in an ACL for RCPT (for example), you can examine exactly what was
3085 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
3086 commands, but it was not correctly comparing the address with the actual
3087 client host address. Thus, it would show the EHLO address in Received:
3088 header lines when this was not necessary.
3090 PH/27 Added the % operator to ${eval:}.
3092 PH/28 Exim tries to create and chdir to its spool directory when it starts;
3093 it should be ignoring failures (because with -C, for example, it has lost
3094 privilege). It wasn't ignoring creation failures other than "already
3097 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
3100 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
3101 because an input line was too long, either on its own, or by virtue of
3102 too many continuations, the temporary file was not being removed, and the
3103 return code was incorrect.
3105 PH/31 Missing "BOOL" in function definition in filtertest.c.
3107 PH/32 Applied Sieve patches from the maintainer.
3109 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
3111 PH/33 Added "verify = not_blind".
3113 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
3114 Local/Makefile (with some defaults set). These are used in built scripts
3115 such as exicyclog, but they have never been used in the exim_install
3116 script (though there are many overriding facilities there). I have
3117 arranged that the exim_install script now takes note of these two
3120 PH/35 Installed configuration files for Dragonfly.
3122 PH/36 When a locally submitted message by a trusted user did not contain a
3123 From: header, and the sender address was obtained from -f or from an SMTP
3124 MAIL command, and the trusted user did not use -F to supply a sender
3125 name, $originator_name was incorrectly used when constructing a From:
3126 header. Furthermore, $originator_name was used for submission mode
3127 messages from external hosts without From: headers in a similar way,
3128 which is clearly wrong.
3130 PH/37 Added control=suppress_local_fixups.
3132 PH/38 When log_selector = +received_sender was set, and the addition of the
3133 sender made the log line's construction buffer exactly full, or one byte
3134 less than full, an overflow happened when the terminating "\n" was
3137 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
3138 when the result of a list match is failure because a DNS lookup failed.
3140 PH/40 RM_COMMAND is now used in the building process.
3142 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
3143 the "build-* directories that it finds.
3145 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
3146 address in a domain literal was a prefix of an interface address.
3148 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
3149 when verifying a sender address, unless rewrite_headers is false.
3151 PH/44 Wrote a long comment about why errors_to addresses are verified as
3152 recipients, not senders.
3154 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
3155 the ratelimit ACL was added.
3157 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
3159 PH/46 Added extra information about PostgreSQL errors to the error string.
3161 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
3162 avoiding the need to install special zones in a real server. This is
3163 backwards compatible; if it can't find the fake resolver, it drops back.
3164 Thus, both old and new test suites can be run.
3166 TF/02 Added util/ratelimit.pl
3168 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
3169 clock is set back in time.
3171 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
3172 Candler <B.Candler@pobox.com>.
3174 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
3175 OK for addresses that are the result of redirections.
3177 PH/48 A number of further additions for the benefit of the new test suite,
3178 including a fake gethostbyname() that interfaces to the fake DNS resolver
3181 TF/06 The fix for widen_domains has also been applied to qualify_single and
3182 search_parents which are the other dnslookup options that can cause
3185 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
3188 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
3190 TF/07 Exim produced the error message "an SRV record indicated no SMTP
3191 service" if it encountered an MX record with an empty target hostname.
3192 The message is now "an MX or SRV record indicated no SMTP service".
3194 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
3195 if the DNS of the sending site is misconfigured. This is quite a
3196 common situation. This change restores the behaviour of treating a
3197 helo verification defer as a failure.
3199 PH/51 If self=fail was set on a router, the bounce message did not include the
3200 actual error message.
3206 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
3208 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
3209 command itself ended in error, the underlying message about the transport
3210 filter timeout was being overwritten with the pipe command error. Now the
3211 underlying error message should be appended to the second error message.
3213 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
3215 PH/02 Reduce the amount of output that "make" produces by default. Full output
3216 can still be requested.
3218 PH/03 The warning log line about a condition test deferring for a "warn" verb
3219 was being output only once per connection, rather than after each
3220 occurrence (because it was using the same function as for successful
3221 "warn" verbs). This seems wrong, so I have changed it.
3223 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
3224 it should not have, which might have caused a crash in the right
3225 circumstances, but probably never did.
3227 PH/04 Installed a modified version of Tony Finch's patch to make submission
3228 mode fix the return path as well as the Sender: header line, and to
3229 add a /name= option so that you can make the user's friendly name appear
3232 TF/03 Added the control = fakedefer ACL modifier.
3234 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
3235 Mark Lowes for thorough testing.
3237 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
3239 TK/03 Merged latest SRS patch from Miles Wilton.
3241 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
3242 with the definition in sysexits.h (which is #included earlier).
3243 Fortunately, Exim does not actually use EX_OK. The code used to try to
3244 preserve the sysexits.h value, by assumimg that macro definitions were
3245 scanned for macro replacements. I have been disabused of this notion,
3246 so now the code just undefines EX_OK before #including unistd.h.
3248 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
3249 in the smtp transport. When a block could not be written in a single
3250 write() function, the timeout was being re-applied to each part-write.
3251 This seems wrong - if the receiver was accepting one byte at a time it
3252 would take for ever. The timeout is now adjusted when this happens. It
3253 doesn't have to be particularly precise.
3255 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
3256 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
3258 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
3259 without a domain if the check to <postmaster@domain> fails.
3261 SC/01 Eximstats: added -xls and the ability to specify output files
3262 (patch written by Frank Heydlauf).
3264 SC/02 Eximstats: use FileHandles for outputing results.
3266 SC/03 Eximstats: allow any combination of xls, txt, and html output.
3268 SC/04 Eximstats: fixed display of large numbers with -nvr option
3270 SC/05 Eximstats: fixed merging of reports with empty tables.
3272 SC/06 Eximstats: added the -include_original_destination flag
3274 SC/07 Eximstats: removed tabs and trailing whitespace.
3276 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
3278 TK/06 MBOX spool code: Add real "From " MBOX separator line
3279 so the .eml file is really in mbox format (even though
3280 most programs do not really care). Patch from Alex Miller.
3282 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
3283 The latter is generated from $received_to and is only set if the
3284 message has one envelope recipient. SA can use these headers,
3285 obviously out-of-the-box. Patch from Alex Miller.
3287 PH/08 The ${def test on a variable was returning false if the variable's
3288 value was "0", contrary to what the specification has always said!
3289 The result should be true unless the variable is empty.
3291 PH/09 The syntax error of a character other than { following "${if
3292 def:variable_name" (after optional whitespace) was not being diagnosed.
3293 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
3294 accidental colon was present, for example, could give incorrect results.
3296 PH/10 Tidied the code in a number of places where the st_size field of a stat()
3297 result is used (not including appendfile, where other changes are about
3300 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
3301 This involved changing a lot of size variables from int to off_t. It
3302 should work with maildirs and everything.
3304 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
3305 spamd dying while we are connected to it.
3307 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
3310 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
3311 function for BDB 4.3.
3313 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
3315 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
3316 the sources for PCRE have been re-arranged and more files are now
3319 PH/15 The code I had for printing potentially long long variables in PH/11
3320 above was not the best (it lost precision). The length of off_t variables
3321 is now inspected at build time, and an appropriate printing format (%ld
3322 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
3323 to be "long long int" or "long int". This is needed for the internal
3324 formatting function string_vformat().
3326 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
3327 the configuration file to be ":syslog", then the script "guesses" where
3328 the logs files are, rather than using the compiled in default. In our
3329 case the guess is not the same as the compiled default, so the script
3330 suddenly stopped working when I started to use syslog. The patch checks
3331 to see if log_file_path is "". If so, it attempts to read it from exim
3332 with no configuration file to get the compiled in version, before it
3333 falls back to the previous guessing code."
3335 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
3336 implementing BATV in an Exim configuration. See NewStuff for the gory
3339 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
3340 Makefile that are specific to HP-UX.
3342 PH/18 If the "use_postmaster" option was set for a recipient callout together
3343 with the "random" option, the postmaster address was used as the MAIL
3344 FROM address for the random test, but not for the subsequent recipient
3345 test. It is now used for both.
3347 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
3348 patch removes a few documentation additions to RFC 3028, because the
3349 latest draft now contains them. It adds the new en;ascii-case comparator
3350 and a new error check for 8bit text in MIME parts. Comparator and
3351 require names are now matched exactly. I enabled the subaddress
3352 extension, but it is not well tested yet (read: it works for me)."
3354 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
3355 rework some of the code of TK/09 above to avoid the hardwired use of
3356 "%lld" and "long long". Replaced the call to snprintf() with a call to
3359 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
3360 records point to non-existent hosts", "retry timeout exceeded", and
3361 "retry time not reached for any host after a long failure period".
3363 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
3364 experimental DomainKeys support:
3366 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
3367 (2) On an error such as an illegally used "control", the wrong name for
3368 the control was given.
3370 These problems did NOT occur unless DomainKeys support was compiled.
3372 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
3374 PH/24 Added ${if match_ip condition.
3376 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
3377 databases so that it will be absolutely obvious if a crash occurs in the
3378 DB library. This is a regular occurrence (often caused by mis-matched
3381 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
3382 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
3383 picked up on a user's system that detects such things. There doesn't seem
3384 to be a gcc warning option for this - only an attribute that has to be
3385 put on the function's prototype. It seems that in Fedora Core 4 they have
3386 set this on a number of new functions. No doubt there will be more in due
3389 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
3390 specify a transport. However, if an address that was verified by such a
3391 router was the subject of a callout, Exim crashed because it tried to
3392 read the rcpt_include_affixes from the non-existent transport. Now it
3393 just assumes that the setting of that option is false. This bug was
3394 introduced by 4.51/PH/31.
3396 PH/28 Changed -d+all to exclude +memory, because that information is very
3397 rarely of interest, but it makes the output a lot bigger. People tend to
3398 do -d+all out of habit.
3400 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
3401 code in os-type was giving problems when libc.so lives in lib64, like on
3404 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
3405 aren't the modern standard, and it seems that some systems' include files
3406 don't always have them. Exim was already checking for some of the newer
3407 ones like T_AAAA, and defining it itself. I've added checks for all the
3408 record types that Exim uses.
3410 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
3411 not automatically generating a new one, as it is supposed to. This
3412 prevented TLS from working. If the file did exist, but contained invalid
3413 data, a new version was generated, as expected. It was only the case of a
3414 non-existent file that was broken.
3416 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
3417 with a change in libdomainkeys > 0.64.
3419 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
3420 from DNS. If the selector record carries the flag, it now has
3421 precedence over the domain-wide flag.
3423 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
3425 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
3426 the use of a port name that isn't defined in /etc/services) occurred, the
3427 message was deferred as in a normal delivery, and thus remained on the
3428 spool, instead of being failed because of the mua_wrapper setting. This
3429 is now fixed, and I tidied up some of the mua_wrapper messages at the
3432 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
3433 the messages in a hash of arrays rather than using individual hashes.
3434 This is a bit cleaner and results in dramatic memory savings, albeit
3435 at a slight CPU cost.
3437 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
3438 as requested by Marc Sherman.
3440 SC/10 Eximstats: added histograms for user specified patterns as requested
3443 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
3445 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
3446 fopen() in the content-scanning modules that did not already have it.
3452 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
3453 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
3455 TK/02 Fix ACL "control" statement not being available in MIME ACL.
3457 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
3459 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
3460 to test Sieve filters that use "vacation".
3462 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
3463 that changes the way the GnuTLS parameters are stored in the cache file.
3464 The new format can be generated externally. For backward compatibility,
3465 if the data in the cache doesn't make sense, Exim assumes it has read an
3466 old-format file, and it generates new data and writes a new file. This
3467 means that you can't go back to an older release without removing the
3470 PH/03 A redirect router that has both "unseen" and "one_time" set does not
3471 work if there are any delivery delays because "one_time" forces the
3472 parent to be marked "delivered", so its unseen clone is never tried
3473 again. For this reason, Exim now forbids the simultaneous setting of
3476 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
3477 redirected to themselves ("homonym" addresses). Read the long ChangeLog
3478 entry if you want to know the details. The fix, however, neglected to
3479 consider the case when local delivery batching is involved. The test for
3480 "previously delivered" was not happening when checking to see if an
3481 address could be batched with a previous (undelivered) one; under
3482 certain circumstances this could lead to multiple deliveries to the same
3485 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
3486 in its include files, and this causes problems building Exim.
3488 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
3489 header_syntax) but Exim was just ignoring anything given after a slash.
3490 In particular, this caused confusion with an attempt to use "verify =
3491 reverse_host_lookup/defer_ok". An error is now given when options are
3492 supplied for verify items that do not have them. (Maybe reverse_host_
3493 lookup should have a defer_ok option, but that's a different point.)
3495 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
3496 defined by RFC 821) to 2048, because there were problems with some AUTH
3497 commands, and RFC 1869 says the size should be increased for extended
3498 SMTP commands that take arguments.
3500 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
3503 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
3504 "unknown" error; now it says that the functionality isn't in the binary.
3506 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
3507 an address' error message when a string expansion fails (syntax or
3508 whatever). Otherwise the password may appear in the log. Following change
3509 PH/42 below, there is no longer a chance of it appearing in a bounce
3512 PH/11 Installed exipick version 20050225.0 from John Jetmore.
3514 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
3515 of its addresses was ever tried. (Bugzilla bug #2.)
3517 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
3518 the result incorrectly in the debug output. (It correctly added a newline
3519 to what was transported.)
3521 TF/01 Added $received_time.
3523 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
3524 commented out examples of how to interface to a virus scanner and to
3525 SpamAssassin. Also added commented examples of av_scanner and
3526 spamd_address settings.
3528 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
3529 and controls are allowed in which ACLs. There were a couple of minor
3530 errors. Some of the entries in the conditions table (which is a table of
3531 where they are NOT allowed) were getting very unwieldy; rewrote them as a
3532 negation of where the condition IS allowed.
3534 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
3536 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
3537 header file does not have a version number, so I've had to invent a new
3538 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
3539 API. The code is untested by me (my Linux distribution still has 0.3.2 of
3540 radiusclient), but it was contributed by a Radius user.
3542 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
3543 files or directories, for OpenSSL.
3545 PH/19 When an Exim process that is running as root has to create an Exim log
3546 file, it does so in a subprocess that runs as exim:exim so as to get the
3547 ownership right at creation (otherwise, other Exim processes might see
3548 the file with the wrong ownership). There was no test for failure of this
3549 fork() call, which would lead to the process getting stuck as it waited
3550 for a non-existent subprocess. Forks do occasionally fail when resources
3551 run out. I reviewed all the other calls to fork(); they all seem to check
3554 PH/20 When checking for unexpected SMTP input at connect time (before writing
3555 the banner), Exim was not dealing correctly with a non-positive return
3556 from the read() function. If the client had disconnected by this time,
3557 the result was a log entry for a synchronization error with an empty
3558 string after "input=" when read() returned zero. If read() returned -1
3559 (an event I could not check), uninitialized data bytes were printed.
3560 There were reports of junk text (parts of files, etc) appearing after
3563 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
3565 PH/22 Added support for macro redefinition, and (re)definition in between
3566 driver and ACL definitions.
3568 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
3569 forgetting to use the resulting value; it was using the unexpanded value.
3571 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
3572 hadn't been configured. The fix is from Juergen Kreileder, who
3573 understands it better than I do:
3575 "Here's what I see happening with three configured cyrus_sasl
3576 authenticators configured (plain, login, cram-md5):
3578 On startup auth_cyrus_sasl_init() gets called for each of these.
3579 This means three calls to sasl_listmech() without a specified mech_list.
3580 => SASL tests which mechs of all available mechs actually work
3581 => three warnings about OTP not working
3582 => the returned list contains: plain, login, cram-md5, digest-md5, ...
3584 With the patch, sasl_listmech() also gets called three times. But now
3585 SASL's mech_list option is set to the server_mech specified in the the
3586 authenticator. Or in other words, the answer from sasl_listmech()
3587 gets limited to just the mech you're testing for (which is different
3589 => the return list contains just 'plain' or 'login', 'cram-md5' or
3590 nothing depending on the value of ob->server_mech.
3592 I've just tested the patch: Authentication still works fine,
3593 unavailable mechs specified in the exim configuration are still
3594 caught, and the auth.log warnings about OTP are gone."
3596 PH/25 When debugging is enabled, the contents of the command line are added
3597 to the debugging output, even when log_selector=+arguments is not
3600 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
3601 answer is "GNU", and only if the return is "GNU/something" is the answer
3604 PH/27 $acl_verify_message is now set immediately after the failure of a
3605 verification in an ACL, and so is available in subsequent modifiers. In
3606 particular, the message can be preserved by coding like this:
3608 warn !verify = sender
3609 set acl_m0 = $acl_verify_message
3611 Previously, $acl_verify_message was set only while expanding "message"
3612 and "log_message" when a very denied access.
3614 PH/28 Modified OS/os.c-Linux with
3616 -#ifndef OS_LOAD_AVERAGE
3617 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
3619 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
3620 nomenclature these days.)
3622 PH/29 Installed patch from the Sieve maintainer that adds the options
3623 sieve_useraddress and sieve_subaddress to the redirect router.
3625 PH/30 In these circumstances:
3626 . Two addresses routed to the same list of hosts;
3627 . First host does not offer TLS;
3628 . First host accepts first address;
3629 . First host gives temporary error to second address;
3630 . Second host offers TLS and a TLS session is established;
3631 . Second host accepts second address.
3632 Exim incorrectly logged both deliveries with the TLS parameters (cipher
3633 and peerdn, if requested) that were in fact used only for the second
3636 PH/31 When doing a callout as part of verifying an address, Exim was not paying
3637 attention to any local part prefix or suffix that was matched by the
3638 router that accepted the address. It now behaves in the same way as it
3639 does for delivery: the affixes are removed from the local part unless
3640 rcpt_include_affixes is set on the transport.
3642 PH/32 Add the sender address, as F=<...>, to the log line when logging a
3643 timeout during the DATA phase of an incoming message.
3645 PH/33 Sieve envelope tests were broken for match types other than :is. I have
3646 applied a patch sanctioned by the Sieve maintainer.
3648 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
3649 the uid or gid is negative. A case of a negative gid caused this to be
3650 noticed. The fix allows for either to be negative.
3652 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
3653 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
3656 PH/36 The change PH/12 above was broken. Fixed it.
3658 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
3659 the grounds that routing the same address twice would always produce the
3660 same answer. This might have been true once, but it is certainly no
3661 longer true now. Routing a child address may depend on the previous
3662 routing that produced that child. Some complicated redirection strategies
3663 went wrong when messages had multiple recipients, and made Exim's
3664 behaviour dependent on the order in which the addresses were given.
3666 I have moved the duplicate checking until after the routing is complete.
3667 Exim scans the addresses that are assigned to local and remote
3668 transports, and removes any duplicates. This means that more work will be
3669 done, as duplicates will always all be routed, but duplicates are
3670 presumably rare, so I don't expect this is of any significance.
3672 For deliveries to pipes, files, and autoreplies, the duplicate checking
3673 still happens during the routing process, since they are not going to be
3676 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
3677 It corrects a timeout issue with spamd. This is Ian's comment: "The
3678 background is that sometimes spamd either never reads data from a
3679 connection it has accepted, or it never writes response data. The exiscan
3680 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
3681 blindly assumes that writes won't block so it may never time out."
3683 PH/39 Allow G after quota size as well as K and M.
3685 PH/40 The value set for $authenticated_id in an authenticator may not contain
3686 binary zeroes or newlines because the value is written to log lines and
3687 to spool files. There was no check on this. Now the value is run through
3688 the string_printing() function so that such characters are converted to
3689 printable escape sequences.
3691 PH/41 $message_linecount is a new variable that contains the total number of
3692 lines in the message. Compare $body_linecount, which is the count for the
3695 PH/42 Exim no longer gives details of delivery errors for specific addresses in
3696 bounce and delay warning messages, except in certain special cases, which
3699 (a) An SMTP error message from a remote host;
3700 (b) A message specified in a :fail: redirection;
3701 (c) A message specified in a "fail" command in a system filter;
3702 (d) A message specified in a FAIL return from the queryprogram router;
3703 (e) A message specified by the cannot_route_message router option.
3705 In these cases only, Exim does include the error details in bounce and
3706 warning messages. There are also a few cases where bland messages such
3707 as "unrouteable address" or "local delivery error" are given.
3709 PH/43 $value is now also set for the "else" part of a ${run expansion.
3711 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
3712 being worked on, but at least Exim now implements the latest version to
3715 PH/45 In a pipe transport, although a timeout while waiting for the pipe
3716 process to complete was treated as a delivery failure, a timeout while
3717 writing the message to the pipe was logged, but erroneously treated as a
3718 successful delivery. Such timeouts include transport filter timeouts. For
3719 consistency with the overall process timeout, these timeouts are now
3720 treated as errors, giving rise to delivery failures by default. However,
3721 there is now a new Boolean option for the pipe transport called
3722 timeout_defer, which, if set TRUE, converts the failures into defers for
3723 both kinds of timeout. A transport filter timeout is now identified in
3726 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
3727 systems where "make" and "gmake" are different, calling "gmake" at top
3728 level broke things. I've arranged for the value of $(MAKE) to be passed
3729 from the Makefile to this script so that it can call the same version of
3733 A note about Exim versions 4.44 and 4.50
3734 ----------------------------------------
3736 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
3737 changes of various kinds. As a consequence, a big documentation update was
3738 needed. This delayed the release for rather longer than seemed good, especially
3739 in the light of a couple of (minor) security issues. Therefore, the changes
3740 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
3741 release. So 4.44 and 4.50 are in effect two different branches that both start
3744 I have left the 4.50 change log unchanged; it contains all the changes since
3745 4.43. The change log for 4.44 is below; many of its items are identical to
3746 those for 4.50. This seems to be the most sensible way to preserve the
3747 historical information.
3753 1. Minor wording change to the doc/README.SIEVE file.
3755 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
3756 computation of the current number of files was incorrect.
3758 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
3759 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
3762 4. Give more explanation in the error message when the command for a transport
3763 filter fails to execute.
3765 5. There are several places where Exim runs a non-Exim command in a
3766 subprocess. The SIGUSR1 signal should be disabled for these processes. This
3767 was being done only for the command run by the queryprogram router. It is
3768 now done for all such subprocesses. The other cases are: ${run, transport
3769 filters, and the commands run by the lmtp and pipe transports.
3771 6. Added CONFIGURE_GROUP build-time option.
3773 7. Some older OS have a limit of 256 on the maximum number of file
3774 descriptors. Exim was using setrlimit() to set 1000 as a large value
3775 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
3776 systems. I've change it so that if it can't get 1000, it tries for 256.
3778 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
3779 was an oversight, and furthermore, ever since the addition of extra
3780 controls (e.g. 4.43/32), the checks on when to allow different forms of
3781 "control" were broken. There should now be diagnostics for all cases when a
3782 control that does not make sense is encountered.
3784 9. Added the /retain_sender option to "control=submission".
3786 10. $recipients is now available in the predata ACL (oversight).
3788 11. Tidy the search cache before the fork to do a delivery from a message
3789 received from the command line. Otherwise the child will trigger a lookup
3790 failure and thereby defer the delivery if it tries to use (for example) a
3791 cached ldap connection that the parent has called unbind on.
3793 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
3794 of $address_data from the recipient verification was clobbered by the
3795 sender verification.
3797 13. The value of address_data from a sender verification is now available in
3798 $sender_address_data in subsequent conditions in the ACL statement.
3800 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
3802 15. Added a new option "connect=<time>" to callout options, to set a different
3805 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
3806 was its contents. (It was OK if the option was not defined at all.)
3808 17. A "Completed" log line is now written for messages that are removed from
3809 the spool by the -Mrm option.
3811 18. New variables $sender_verify_failure and $recipient_verify_failure contain
3812 information about exactly what failed.
3814 19. Added -dd to debug only the daemon process.
3816 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
3817 handles timeouts, both on the server side and network timeouts. Renamed the
3818 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
3820 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
3821 It is now set to "smtps".
3823 22. $host_address is now set to the target address during the checking of
3824 ignore_target_hosts.
3826 23. When checking ignore_target_hosts for an ipliteral router, no host name was
3827 being passed; this would have caused $sender_host_name to have been used if
3828 matching the list had actually called for a host name (not very likely,
3829 since this list is usually IP addresses). A host name is now passed as
3832 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
3833 code that specifies a non-restarting handler (typically sigaction() in
3834 modern systems) in an attempt to fix a rare and obscure crash bug.
3836 25. Narrowed the window for a race in the daemon that could cause it to ignore
3837 SIGCHLD signals. This is not a major problem, because they are used only to
3838 wake it up if nothing else does.
3840 26. A malformed maildirsize file could cause Exim to calculate negative values
3841 for the mailbox size or file count. Odd effects could occur as a result.
3842 The maildirsize information is now recalculated if the size or filecount
3845 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
3846 support for a long time. Removed HAVE_SYS_VFS_H.
3848 28. Installed the latest version of exipick from John Jetmore.
3850 29. In an address list, if the pattern was not a regular expression, an empty
3851 subject address (from a bounce message) matched only if the pattern was an
3852 empty string. Non-empty patterns were not even tested. This was the wrong
3853 because it is perfectly reasonable to use an empty address as part of a
3854 database query. An empty address is now tested by patterns that are
3855 lookups. However, all the other forms of pattern expect the subject to
3856 contain a local part and a domain, and therefore, for them, an empty
3857 address still always fails if the pattern is not itself empty.
3859 30. Exim went into a mad DNS loop when attempting to do a callout where the
3860 host was specified on an smtp transport, and looking it up yielded more
3861 than one IP address.
3863 31. Re-factored the code for checking spool and log partition space into a
3864 function that finds that data and another that does the check. The former
3865 is then used to implement four new variables: $spool_space, $log_space,
3866 $spool_inodes, and $log_inodes.
3868 32. The RFC2047 encoding function was originally intended for short strings
3869 such as real names; it was not keeping to the 75-character limit for
3870 encoded words that the RFC imposes. It now respects the limit, and
3871 generates multiple encoded words if necessary. To be on the safe side, I
3872 have increased the buffer size for the ${rfc2047: expansion operator from
3875 33. It is now permitted to omit both strings after an "if" condition; if the
3876 condition is true, the result is "true". As before, when the second string
3877 is omitted, a false condition yields an empty string. This makes it less
3878 cumbersome to write custom ACL and router conditions.
3880 34. Failure to deliver a bounce message always caused it to be frozen, even if
3881 there was an errors_to setting on the router. The errors_to setting is now
3884 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
3885 canonical form (fully expanded) before being placed in
3886 $sender_host_address.
3888 36. The table in the code that translates DNS record types into text (T_A to
3889 "A" for instance) was missing entries for NS and CNAME. It is just possible
3890 that this could have caused confusion if both these types were looked up
3891 for the same domain, because the text type is used as part of Exim's
3892 per-process caching. But the chance of anyone hitting this buglet seems
3895 37. The dnsdb lookup has been extended in a number of ways.
3897 (1) There is a new type, "zns", which walks up the domain tree until it
3898 finds some nameserver records. It should be used with care.
3900 (2) There is a new type, "mxh", which is like "mx" except that it returns
3901 just the host names, not the priorities.
3903 (3) It is now possible to give a list of domains (or IP addresses) to be
3904 looked up. The behaviour when one of the lookups defers can be
3905 controlled by a keyword.
3907 (4) It is now possible to specify the separator character for use when
3908 multiple records are returned.
3910 38. The dnslists ACL condition has been extended: it is now possible to supply
3911 a list of IP addresses and/or domains to be looked up in a particular DNS
3914 39. Added log_selector=+queue_time_overall.
3916 40. When running the queue in the test harness, wait just a tad after forking a
3917 delivery process, to get repeatability of debugging output.
3919 41. Include certificate and key file names in error message when GnuTLS fails
3920 to set them up, because the GnuTLS error message doesn't include the name
3921 of the failing file when there is a problem reading it.
3923 42. Allow both -bf and -bF in the same test run.
3925 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
3927 44. The "Exiscan patch" is now merged into the mainline Exim source.
3929 45. Sometimes the final signoff response after QUIT could fail to get
3930 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
3931 < 0 before doing a fflush(). This bug looks as though it goes back to the
3932 introduction of TLS in release 3.20, but "sometimes" must have been rare
3933 because the tests only now provoked it.
3935 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
3936 (this can affect the format of dates).
3938 47. exim_tidydb, when checking for the continued existence of a message for
3939 which it has found a message-specific retry record, was not finding
3940 messages that were in split spool directories. Consequently, it was
3941 deleting retry records that should have stayed in existence.
3943 48. Steve fixed some bugs in eximstats.
3945 49. The SPA authentication driver was not abandoning authentication and moving
3946 on to the next authenticator when an expansion was forced to fail,
3947 contradicting the general specification for all authenticators. Instead it
3948 was generating a temporary error. It now behaves as specified.
3950 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
3951 (the order specifies the preference for clients). The order is now AES256,
3952 AES128, 3DES, ARCFOUR128.
3954 51. Small patch to Sieve code - explicitly set From: when generating an
3957 52. Exim crashed if a remote delivery caused a very long error message to be
3958 recorded - for instance if somebody sent an entire SpamAssassin report back
3959 as a large number of 550 error lines. This bug was coincidentally fixed by
3960 increasing the size of one of Exim's internal buffers (big_buffer) that
3961 happened as part of the Exiscan merge. However, to be on the safe side, I
3962 have made the code more robust (and fixed the comments that describe what
3965 53. Now that there can be additional text after "Completed" in log lines (if
3966 the queue_time_overall log selector is set), a one-byte patch to exigrep
3967 was needed to allow it to recognize "Completed" as not the last thing in
3970 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
3971 patch that reportedly fixes this has been added. I am not expert enough to
3972 create a test for it. This is what the patch creator wrote:
3974 "I found a little strange behaviour of ldap code when working with
3975 Windows 2003 AD Domain, where users was placed in more than one
3976 Organization Units. When I tried to give exim partial DN, the exit code
3977 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
3978 But simultaneously result of request was absolutely normal ldap result,
3979 so I produce this patch..."
3981 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
3982 REFERENCE, so I have modified the code to exclude the patch when that macro
3985 55. Some experimental protocols are using DNS PTR records for new purposes. The
3986 keys for these records are domain names, not reversed IP addresses. The
3987 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
3988 leaves it alone. Component reversal etc. now happens only for IP addresses.
3991 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
3993 57. Double the size of the debug message buffer (to 2048) so that more of very
3994 long debug lines gets shown.
3996 58. The exicyclog utility now does better if the number of log files to keep
3997 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
3999 59. Two changes related to the smtp_active_hostname option:
4001 (1) $smtp_active_hostname is now available as a variable.
4002 (2) The default for smtp_banner uses $smtp_active_hostname instead
4003 of $primary_hostname.
4005 60. The host_aton() function is supposed to be passed a string that is known
4006 to be a valid IP address. However, in the case of IPv6 addresses, it was
4007 not checking this. This is a hostage to fortune. Exim now panics and dies
4008 if the condition is not met. A case was found where this could be provoked
4009 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4010 components; fortuitously, this particular loophole had already been fixed
4011 by change 4.50/55 above.
4013 If there are any other similar loopholes, the new check in host_aton()
4014 itself should stop them being exploited. The report I received stated that
4015 data on the command line could provoke the exploit when Exim was running as
4016 exim, but did not say which command line option was involved. All I could
4017 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4018 running as the user.
4021 61. There was a buffer overflow vulnerability in the SPA authentication code
4022 (which came originally from the Samba project). I have added a test to the
4023 spa_base64_to_bits() function which I hope fixes it.
4026 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
4027 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
4029 63. The daemon start-up calls getloadavg() while still root for those OS that
4030 need the first call to be done as root, but it missed one case: when
4031 deliver_queue_load_max is set with deliver_drop_privilege. This is
4032 necessary for the benefit of the queue runner, because there is no re-exec
4033 when deliver_drop_privilege is set.
4035 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
4036 This has been fixed.
4038 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4039 was in use, was not putting the data itself into the right store pool;
4040 consequently, it could be overwritten for a subsequent message in the same
4041 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4044 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
4046 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
4047 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
4048 the function were treating the return as a boolean value, which happened to
4049 work because 0=false and not-0=true, but is not correct code.
4051 68. The host_aton() function was not handling scoped IPv6 addresses (those
4052 with, for example, "%eth0" on the end) correctly.
4054 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
4055 negated items (that is, ~something) in unsigned ints. Some compilers
4056 apparently mutter when there is no cast.
4058 70. If an address verification called from an ACL failed, and did not produce a
4059 user-specific message (i.e. there was only a "system" message), nothing was
4060 put in $acl_verify_message. In this situation, it now puts the system
4063 71. Change 4.23/11 added synchronization checking at the start of an SMTP
4064 session; change 4.31/43 added the unwanted input to the log line - except
4065 that it did not do this in the start of session case. It now does.
4067 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
4068 This is wrong and can cause the other end to generate a synchronization
4069 error if it is another Exim or anything else that does the synchronization
4070 check. A QUIT command is no longer sent after a timeout.
4072 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
4073 during host lookups.
4075 74. The defer_ok option of callout verification was not working if it was used
4076 when verifying addresses in header lines, that is, for this case:
4078 verify = header_sender/callout=defer_ok
4080 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
4081 those file descriptors could be used for SMTP connections. If anything
4082 wrote to stderr (the example that came up was "warn" in embedded Perl), it
4083 could be sent to the SMTP client, causing chaos. The daemon now opens
4084 stdin, stdout, and stderr to /dev/null when it puts itself into the
4087 76. Arrange for output from Perl's "warn" command to be written to Exim's main
4088 log by default. The user can override this with suitable Perl magic.
4090 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
4091 the log message when discard triggers, was not working for the DATA ACL or
4092 for the non-SMTP ACL.
4094 78. Error message wording change in sieve.c.
4096 79. If smtp_accept_max_per_host was set, the number of connections could be
4097 restricted to fewer than expected, because the daemon was trying to set up
4098 a new connection before checking whether the processes handling previous
4099 connections had finished. The check for completed processes is now done
4100 earlier. On busy systems, this bug wouldn't be noticed because something
4101 else would have woken the daemon, and it would have reaped the completed
4104 80. If a message was submitted locally by a user whose login name contained one
4105 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
4106 It caused a spool format error. I have fixed the spool reading code. A
4107 related problem was that the "from" clause in the Received: line became
4108 illegal because of the space(s). It is now covered by ${quote_local_part.
4110 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
4113 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
4115 83. Patch from Sieve maintainer to fix unterminated string problem in
4116 "vacation" handling.
4118 84. Some minor changes to the Linux configuration files to help with other
4119 OS variants using glibc.
4121 85. One more patch for Sieve to update vacation handling to latest spec.
4124 ----------------------------------------------------
4125 See the note above about the 4.44 and 4.50 releases.
4126 ----------------------------------------------------
4132 1. Change 4.43/35 introduced a bug that caused file counts to be
4133 incorrectly computed when quota_filecount was set in an appendfile
4136 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4137 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4140 3. Give more explanation in the error message when the command for a transport
4141 filter fails to execute.
4143 4. There are several places where Exim runs a non-Exim command in a
4144 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4145 was being done only for the command run by the queryprogram router. It is
4146 now done for all such subprocesses. The other cases are: ${run, transport
4147 filters, and the commands run by the lmtp and pipe transports.
4149 5. Some older OS have a limit of 256 on the maximum number of file
4150 descriptors. Exim was using setrlimit() to set 1000 as a large value
4151 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4152 systems. I've change it so that if it can't get 1000, it tries for 256.
4154 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4155 was an oversight, and furthermore, ever since the addition of extra
4156 controls (e.g. 4.43/32), the checks on when to allow different forms of
4157 "control" were broken. There should now be diagnostics for all cases when a
4158 control that does not make sense is encountered.
4160 7. $recipients is now available in the predata ACL (oversight).
4162 8. Tidy the search cache before the fork to do a delivery from a message
4163 received from the command line. Otherwise the child will trigger a lookup
4164 failure and thereby defer the delivery if it tries to use (for example) a
4165 cached ldap connection that the parent has called unbind on.
4167 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4168 of $address_data from the recipient verification was clobbered by the
4169 sender verification.
4171 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4172 was its contents. (It was OK if the option was not defined at all.)
4174 11. A "Completed" log line is now written for messages that are removed from
4175 the spool by the -Mrm option.
4177 12. $host_address is now set to the target address during the checking of
4178 ignore_target_hosts.
4180 13. When checking ignore_target_hosts for an ipliteral router, no host name was
4181 being passed; this would have caused $sender_host_name to have been used if
4182 matching the list had actually called for a host name (not very likely,
4183 since this list is usually IP addresses). A host name is now passed as
4186 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4187 code that specifies a non-restarting handler (typically sigaction() in
4188 modern systems) in an attempt to fix a rare and obscure crash bug.
4190 15. Narrowed the window for a race in the daemon that could cause it to ignore
4191 SIGCHLD signals. This is not a major problem, because they are used only to
4192 wake it up if nothing else does.
4194 16. A malformed maildirsize file could cause Exim to calculate negative values
4195 for the mailbox size or file count. Odd effects could occur as a result.
4196 The maildirsize information is now recalculated if the size or filecount
4199 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4200 support for a long time. Removed HAVE_SYS_VFS_H.
4202 18. Updated exipick to current release from John Jetmore.
4204 19. Allow an empty sender to be matched against a lookup in an address list.
4205 Previously the only cases considered were a regular expression, or an
4208 20. Exim went into a mad DNS lookup loop when doing a callout where the
4209 host was specified on the transport, if the DNS lookup yielded more than
4212 21. The RFC2047 encoding function was originally intended for short strings
4213 such as real names; it was not keeping to the 75-character limit for
4214 encoded words that the RFC imposes. It now respects the limit, and
4215 generates multiple encoded words if necessary. To be on the safe side, I
4216 have increased the buffer size for the ${rfc2047: expansion operator from
4219 22. Failure to deliver a bounce message always caused it to be frozen, even if
4220 there was an errors_to setting on the router. The errors_to setting is now
4223 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4224 canonical form (fully expanded) before being placed in
4225 $sender_host_address.
4227 24. Updated eximstats to version 1.33
4229 25. Include certificate and key file names in error message when GnuTLS fails
4230 to set them up, because the GnuTLS error message doesn't include the name
4231 of the failing file when there is a problem reading it.
4233 26. Expand error message when OpenSSL has problems setting up cert/key files.
4236 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
4237 (this can affect the format of dates).
4239 28. exim_tidydb, when checking for the continued existence of a message for
4240 which it has found a message-specific retry record, was not finding
4241 messages that were in split spool directories. Consequently, it was
4242 deleting retry records that should have stayed in existence.
4244 29. eximstats updated to version 1.35
4245 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
4246 1.35 - bugfix such that pie charts by volume are generated correctly
4248 30. The SPA authentication driver was not abandoning authentication and moving
4249 on to the next authenticator when an expansion was forced to fail,
4250 contradicting the general specification for all authenticators. Instead it
4251 was generating a temporary error. It now behaves as specified.
4253 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
4254 (the order specifies the preference for clients). The order is now AES256,
4255 AES128, 3DES, ARCFOUR128.
4257 31. Small patch to Sieve code - explicitly set From: when generating an
4260 32. Exim crashed if a remote delivery caused a very long error message to be
4261 recorded - for instance if somebody sent an entire SpamAssassin report back
4262 as a large number of 550 error lines. This bug was coincidentally fixed by
4263 increasing the size of one of Exim's internal buffers (big_buffer) that
4264 happened as part of the Exiscan merge. However, to be on the safe side, I
4265 have made the code more robust (and fixed the comments that describe what
4268 33. Some experimental protocols are using DNS PTR records for new purposes. The
4269 keys for these records are domain names, not reversed IP addresses. The
4270 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4271 leaves it alone. Component reversal etc. now happens only for IP addresses.
4274 34. The host_aton() function is supposed to be passed a string that is known
4275 to be a valid IP address. However, in the case of IPv6 addresses, it was
4276 not checking this. This is a hostage to fortune. Exim now panics and dies
4277 if the condition is not met. A case was found where this could be provoked
4278 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4279 components; fortuitously, this particular loophole had already been fixed
4280 by change 4.50/55 or 4.44/33 above.
4282 If there are any other similar loopholes, the new check in host_aton()
4283 itself should stop them being exploited. The report I received stated that
4284 data on the command line could provoke the exploit when Exim was running as
4285 exim, but did not say which command line option was involved. All I could
4286 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4287 running as the user.
4290 35. There was a buffer overflow vulnerability in the SPA authentication code
4291 (which came originally from the Samba project). I have added a test to the
4292 spa_base64_to_bits() function which I hope fixes it.
4295 36. The daemon start-up calls getloadavg() while still root for those OS that
4296 need the first call to be done as root, but it missed one case: when
4297 deliver_queue_load_max is set with deliver_drop_privilege. This is
4298 necessary for the benefit of the queue runner, because there is no re-exec
4299 when deliver_drop_privilege is set.
4301 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4302 was in use, was not putting the data itself into the right store pool;
4303 consequently, it could be overwritten for a subsequent message in the same
4304 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4307 38. Sometimes the final signoff response after QUIT could fail to get
4308 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4309 < 0 before doing a fflush(). This bug looks as though it goes back to the
4310 introduction of TLS in release 3.20, but "sometimes" must have been rare
4311 because the tests only now provoked it.
4317 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
4318 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
4319 Now it returns an integer. A number of calls were still expecting a T/F
4320 return. Fortuitously, in all cases, the tests worked in OK situations,
4321 which is the norm. However, things would have gone wrong on any write
4322 failures on the smtp file descriptor. This function is used when sending
4323 messages over SMTP and also when doing verify callouts.
4325 2. When Exim is called to do synchronous delivery of a locally submitted
4326 message (the -odf or -odi options), it no longer closes stderr before doing
4329 3. Implemented the mua_wrapper option.
4331 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
4333 5. Implemented the functions header_remove(), header_testname(),
4334 header_add_at_position(), and receive_remove_recipient(), and exported them
4337 6. If an ACL "warn" statement specified the addition of headers, Exim already
4338 inserted X-ACL-Warn: at the start if there was no header name. However, it
4339 was not making this test for the second and subsequent header lines if
4340 there were newlines in the string. This meant that an invalid header could
4341 be inserted if Exim was badly configured.
4343 7. Allow an ACL "warn" statement to add header lines at the start or after all
4344 the Received: headers, as well as at the end.
4346 8. Added the rcpt_4xx retry error code.
4348 9. Added postmaster_mailfrom=xxx to callout verification option.
4350 10. Added mailfrom=xxxx to the callout verification option, for verify=
4353 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
4354 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
4356 12. Inserted some casts to stop certain compilers warning when using pointer
4357 differences as field lengths or precisions in printf-type calls (mostly
4358 affecting debugging statements).
4360 13. Added optional readline() support for -be (dynamically loaded).
4362 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
4363 same clock tick as a message's arrival, so that its received time was the
4364 same as the "first fail" time on the retry record, and that message
4365 remained on the queue past the ultimate address timeout, every queue runner
4366 would try a delivery (because it was past the ultimate address timeout) but
4367 after another failure, the ultimate address timeout, which should have then
4368 bounced the address, did not kick in. This was a "< instead of <=" error;
4369 in most cases the first failure would have been in the next clock tick
4370 after the received time, and all would be well.
4372 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
4373 being recognized when the domain list was tested by the match_domain
4374 condition in an expansion string.
4376 16. Added the ${str2b64: operator.
4378 17. Exim was always calling setrlimit() to set a large limit for the number of
4379 processes, without checking whether the existing limit was already
4380 adequate. (It did check for the limit on file descriptors.) Furthermore,
4381 errors from getrlimit() and setrlimit() were being ignored. Now they are
4382 logged to the main and panic logs, but Exim does carry on, to try to do its
4383 job under whatever limits there are.
4385 18. Imported PCRE 5.0.
4387 19. Trivial typo in log message " temporarily refused connection" (the leading
4390 20. If the log selector return_path_on_delivery was set and an address was
4391 redirected to /dev/null, the delivery process crashed because it assumed
4392 that a return path would always be set for a "successful" delivery. In this
4393 case, the whole delivery is bypassed as an optimization, and therefore no
4396 21. Internal re-arrangement: the function for sending a challenge and reading
4397 a response while authentication was assuming a zero-terminated challenge
4398 string. It's now changed to take a pointer and a length, to allow for
4399 binary data in such strings.
4401 22. Added the cyrus_sasl authenticator (code supplied by MBM).
4403 23. Exim was not respecting finduser_retries when seeking the login of the
4404 uid under which it was called; it was always trying 10 times. (The default
4405 setting of finduser_retries is zero.) Also, it was sleeping after the final
4406 failure, which is pointless.
4408 24. Implemented tls_on_connect_ports.
4410 25. Implemented acl_smtp_predata.
4412 26. If the domain in control=submission is set empty, Exim assumes that the
4413 authenticated id is a complete email address when it generates From: or
4414 Sender: header lines.
4416 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
4417 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
4418 chown and chgrp in /bin and hostname in /usr/bin.
4420 28. Exim was keeping the "process log" file open after each use, just as it
4421 does for the main log. This opens the possibility of it remaining open for
4422 long periods when the USR1 signal hits a daemon. Occasional processlog
4423 errors were reported, that could have been caused by this. Anyway, it seems
4424 much more sensible not to leave this file open at all, so that is what now
4427 29. The long-running daemon process does not normally write to the log once it
4428 has entered its main loop, and it closes the log before doing so. This is
4429 so that log files can straightforwardly be renamed and moved. However,
4430 there are a couple of unusual error situations where the daemon does write
4431 log entries, and I had neglected to close the log afterwards.
4433 30. The text of an SMTP error response that was received during a remote
4434 delivery was being truncated at 512 bytes. This is too short for some of
4435 the long messages that one sometimes sees. I've increased the limit to
4438 31. It is now possible to make retry rules that apply only when a message has a
4439 specific sender, in particular, an empty sender.
4441 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
4442 it possible to be selective about when SMTP synchronization is enforced.
4444 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
4446 32. Implemented hosts_connection_nolog.
4448 33. Added an ACL for QUIT.
4450 34. Setting "delay_warning=" to disable warnings was not working; it gave a
4453 35. Added mailbox_size and mailbox_filecount to appendfile.
4455 36. Added control = no_multiline_responses to ACLs.
4457 37. There was a bug in the logic of the code that waits for the clock to tick
4458 in the case where the clock went backwards by a substantial amount such
4459 that the microsecond fraction of "now" was more than the microsecond
4460 fraction of "then" (but the whole seconds number was less).
4462 38. Added support for the libradius Radius client library this is found on
4463 FreeBSD (previously only the radiusclient library was supported).
4469 1. When certain lookups returned multiple values in the form name=value, the
4470 quoting of the values was not always being done properly. Specifically:
4471 (a) If the value started with a double quote, but contained no whitespace,
4473 (b) If the value contained whitespace other than a space character (i.e.
4474 tabs or newlines or carriage returns) it was not quoted.
4475 This fix has been applied to the mysql and pgsql lookups by writing a
4476 separate quoting function and calling it from the lookup code. The fix
4477 should probably also be applied to nisplus, ibase and oracle lookups, but
4478 since I cannot test any of those, I have not disturbed their existing code.
4480 2. A hit in the callout cache for a specific address caused a log line with no
4481 reason for rejecting RCPT. Now it says "Previous (cached) callout
4482 verification failure".
4484 3. There was an off-by-one bug in the queryprogram router. An over-long
4485 return line was truncated at 256 instead of 255 characters, thereby
4486 overflowing its buffer with the terminating zero. As well as fixing this, I
4487 have increased the buffer size to 1024 (and made a note to document this).
4489 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
4490 when Exim is waiting for an SMTP response from a remote server, Exim
4491 restarts its select() call on the socket, thereby resetting its timeout.
4492 This is not a problem when such interrupts are rare. Somebody set up a cron
4493 job to run exiwhat every 2 minutes, which is less than the normal select()
4494 timeout (5 or 10 minutes). This meant that the select() timeout never
4495 kicked in because it was always reset. I have fixed this by comparing the
4496 time when an interrupt arrives with the time at the start of the first call
4497 to select(). If more time than the timeout has elapsed, the interrupt is
4498 treated as a timeout.
4500 5. Some internal re-factoring in preparation for the addition of Sieve
4501 extensions (by MH). In particular, the "personal" test is moved to a
4502 separate function, and given an option for scanning Cc: and Bcc: (which is
4503 not set for Exim filters).
4505 6. When Exim created an email address using the login of the caller as the
4506 local part (e.g. when creating a From: or Sender: header line), it was not
4507 quoting the local part when it contained special characters such as @.
4509 7. Installed new OpenBSD configuration files.
4511 8. Reworded some messages for syntax errors in "and" and "or" conditions to
4512 try to make them clearer.
4514 9. Callout options, other than the timeout value, were being ignored when
4515 verifying sender addresses in header lines. For example, when using
4517 verify = header_sender/callout=no_cache
4519 the cache was (incorrectly) being used.
4521 10. Added a missing instance of ${EXE} to the exim_install script; this affects
4522 only the Cygwin environment.
4524 11. When return_path_on_delivery was set as a log selector, if different remote
4525 addresses in the same message used different return paths and parallel
4526 remote delivery occurred, the wrong values would sometimes be logged.
4527 (Whenever a remote delivery process finished, the return path value from
4528 the most recently started remote delivery process was logged.)
4530 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
4531 lines when AUTH and/or TLS are in use. This is the "received protocol"
4532 field. Exim used to use "asmtp" for authenticated SMTP, without any
4533 indication (in the protocol name) for TLS use. Now it follows the RFC and
4534 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
4535 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
4536 names appear in log lines as well as in Received: header lines.
4538 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
4539 extensions, and comparison tests, and to fix some bugs.
4541 14. Changes to the "personal" filter test:
4543 (1) The test was buggy in that it was just doing the equivalent of
4544 "contains" tests on header lines. For example, if a user's address was
4545 anne@some.where, the "personal" test would incorrectly be true for
4547 To: susanne@some.where
4549 This test is now done by extracting each address from the header in turn,
4550 and checking the entire address. Other tests that are part of "personal"
4551 are now done using regular expressions (for example, to check local parts
4552 of addresses in From: header lines).
4554 (2) The list of non-personal local parts in From: addresses has been
4555 extended to include "listserv", "majordomo", "*-request", and "owner-*",
4556 taken from the Sieve specification recommendations.
4558 (3) If the message contains any header line starting with "List-" it is
4559 treated as non-personal.
4561 (4) The test for "circular" in the Subject: header line has been removed
4562 because it now seems ill-conceived.
4564 15. Minor typos in src/EDITME comments corrected.
4566 16. Installed latest exipick from John Jetmore.
4568 17. If headers_add on a router specified a text string that was too long for
4569 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
4570 of string_sprintf() is now avoided.
4572 18. $message_body_size was not set (it was always zero) when running the DATA
4573 ACL and the local_scan() function.
4575 19. For the "mail" command in an Exim filter, no default was being set for
4576 the once_repeat time, causing a random time value to be used if "once" was
4577 specified. (If the value happened to be <= 0, no repeat happened.) The
4578 default is now 0s, meaning "never repeat". The "vacation" command was OK
4579 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
4580 (I found it when inspecting the code).
4582 20. There is now an overall timeout for performing a callout verification. It
4583 defaults to 4 times the callout timeout, which applies to individual SMTP
4584 commands during the callout. The overall timeout applies when there is more
4585 than one host that can be tried. The timeout is checked before trying the
4586 next host. This prevents very long delays if there are a large number of
4587 hosts and all are timing out (e.g. when the network connections are timing
4588 out). The value of the overall timeout can be changed by specifying an
4589 additional sub-option for "callout", called "maxwait". For example:
4591 verify = sender/callout=5s,maxwait=20s
4593 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
4594 to the end before writing, but this should make it even safer).
4596 22. Exim was forgetting that it had advertised PIPELINING for the second and
4597 subsequent messages on an SMTP connection. It was also not resetting its
4598 memory on STARTTLS and an internal HELO.
4600 23. When Exim logs an SMTP synchronization error within a session, it now
4601 records whether PIPELINING has been advertised or not.
4603 24. Added 3 instances of "(long int)" casts to time_t variables that were being
4604 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
4605 rather than long int.
4607 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
4609 26. Added the never_mail option to autoreply.
4615 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
4616 crash if the getsockname() call failed; this can happen if a connection is
4617 closed very soon after it is established. The problem was simply in the
4618 order in which certain operations were done, causing Exim to try to write
4619 to the SMTP stream before it had set up the file descriptor. The bug has
4620 been fixed by making things happen in the correct order.
4626 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
4627 before the connection was closed, thus losing the rejection response.
4629 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
4630 some early Solaris releases, but causes trouble in current releases where
4631 socklen_t is defined.
4633 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
4636 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
4639 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
4640 that releasing the top of it at the end releases what was used for sub-
4641 expansions (unless the block got too big). However, discard this block if
4642 the first thing is a variable or header, so that we can use its block when
4643 it is dynamic (useful for very large $message_headers, for example).
4645 6. Lookups now cache *every* query, not just the most recent. A new, separate
4646 store pool is used for this. It can be recovered when all lookup caches are
4647 flushed. Lookups now release memory at the end of their result strings.
4648 This has involved some general refactoring of the lookup sources.
4650 7. Some code has been added to the store_xxx() functions to reduce the amount
4651 of flapping under certain conditions.
4653 8. log_incoming_interface used to affect only the <= reception log lines. Now
4654 it causes the local interface and port to be added to several more SMTP log
4655 lines, for example "SMTP connection from", and rejection lines.
4657 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
4659 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
4661 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
4662 could be overwritten at the end of the current message (or the start of a
4663 new message if it was set in a HELO ACL). The value is now preserved for
4664 the duration of the SMTP connection.
4666 12. If a transport had a headers_rewrite setting, and a matching header line
4667 contained an unqualified address, that address was qualified, even if it
4668 did not match any rewriting rules. The underlying bug was that the values
4669 of the flags that permit the existence of unqualified sender and recipient
4670 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
4671 non-local messages, and by -bnq for local messages) were not being
4672 preserved with the message after it was received.
4674 13. When Exim was logging an SMTP synchronization error, it could sometimes log
4675 "next input=" as part of the text comprising the host identity instead of
4676 the correct text. The code was using the same buffer for two different
4677 strings. However, depending on which order the printing function evaluated
4678 its arguments, the bug did not always show up. Under Linux, for example, my
4679 test suite worked just fine.
4681 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
4682 doesn't work with some older versions of Perl. It has been changed to "my",
4683 which in any case is probably the better facility to use.
4685 15. A really picky compiler found some instances of statements for creating
4686 error messages that either had too many or two few arguments for the format
4689 16. The size of the buffer for calls to the DNS resolver has been increased
4690 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
4691 for addresses that have a lot of PTR records. This alleviates a problem; it
4692 does not fully solve it.
4694 17. A dnsdb lookup for PTR records that receives more data than will fit in the
4695 buffer now truncates the list and logs the incident, which is the same
4696 action as happens when Exim is looking up a host name and its aliases.
4697 Previously in this situation something unpredictable would happen;
4698 sometimes it was "internal error: store_reset failed".
4700 18. If a server dropped the connection unexpectedly when an Exim client was
4701 using GnuTLS and trying to read a response, the client delivery process
4702 crashed while trying to generate an error log message.
4704 19. If a "warn" verb in an ACL added multiple headers to a message in a single
4705 string, for example:
4707 warn message = H1: something\nH2: something
4709 the text was added as a single header line from Exim's point of view
4710 though it ended up OK in the delivered message. However, searching for the
4711 second and subsequent header lines using $h_h2: did not work. This has been
4712 fixed. Similarly, if a system filter added multiple headers in this way,
4713 the routers could not see them.
4715 20. Expanded the error message when iplsearch is called with an invalid key to
4716 suggest using net-iplsearch in a host list.
4718 21. When running tests using -bh, any delays imposed by "delay" modifiers in
4719 ACLs are no longer actually imposed (and a message to that effect is
4722 22. If a "gecos" field in a passwd entry contained escaped characters, in
4723 particular, if it contained a \" sequence, Exim got it wrong when building
4724 a From: or a Sender: header from that name. A second bug also caused
4725 incorrect handling when an unquoted " was present following a character
4726 that needed quoting.
4728 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
4729 was not being matched caselessly.
4731 24. Arranged for all hyphens in the exim.8 source to be escaped with
4734 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
4735 properly. Recipient callout cache records were still being keyed to include
4736 the sender, even when use_sender was set false. This led to far more
4737 callouts that were necessary. The sender is no longer included in the key
4738 when use_sender is false.
4740 26. Added "control = submission" modifier to ACLs.
4742 27. Added the ${base62d: operator to decode base 62 numbers.
4744 28. dnsdb lookups can now access SRV records.
4746 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
4747 the configuration file.
4749 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
4750 (-v) mode. This makes the output for a verbose queue run more intelligible.
4752 31. Added a use_postmaster feature to recipient callouts.
4754 32. Added the $body_zerocount variable, containing the number of binary zero
4755 bytes in the message body.
4757 33. The time of last modification of the "new" subdirectory is now used as the
4758 "mailbox time last read" when there is a quota error for a maildir
4761 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
4763 35. Added +ignore_unknown as a special item in host lists.
4765 36. Code for decoding IPv6 addresses in host lists is now included, even if
4766 IPv6 support is not being compiled. This fixes a bug in which an IPv6
4767 address was recognized as an IP address, but was then not correctly decoded
4768 into binary, causing unexpected and incorrect effects when compared with
4775 1. Very minor rewording of debugging text in manualroute to say "list of
4776 hosts" instead of "hostlist".
4778 2. If verify=header_syntax was set, and a header line with an unqualified
4779 address (no domain) and a large number of spaces between the end of the
4780 name and the colon was received, the reception process suffered a buffer
4781 overflow, and (when I tested it) crashed. This was caused by some obsolete
4782 code that should have been removed. The fix is to remove it!
4784 3. When running in the test harness, delay a bit after writing a bounce
4785 message to get a bit more predictability in the log output.
4787 4. Added a call to search_tidyup() just before forking a reception process. In
4788 theory, someone could use a lookup in the expansion of smtp_accept_max_
4789 per_host which, without the tidyup, could leave open a database connection.
4791 5. Added the variables $recipient_data and $sender_data which get set from a
4792 lookup success in an ACL "recipients" or "senders" condition, or a router
4793 "senders" option, similar to $domain_data and $local_part_data.
4795 6. Moved the writing of debug_print from before to after the "senders" test
4798 7. Change 4.31/66 (moving the time when the Received: is generated) caused
4799 problems for message scanning, either using a data ACL, or using
4800 local_scan() because the Received: header was not generated till after they
4801 were called (in order to set the time as the time of reception completion).
4802 I have revised the way this works. The header is now generated after the
4803 body is received, but before the ACL or local_scan() are called. After they
4804 are run, the timestamp in the header is updated.
4810 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
4811 before starting a queue runner without re-exec. This happened only when
4812 deliver_drop_privilege was set or when the Exim user was set to root. The
4813 effect of the bug was that timeouts during subsequent deliveries caused
4814 crashes instead of being properly handled. The handler is now left at its
4815 default (and expected) setting.
4817 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
4818 message, again when deliver_drop_privilege is set or Exim is run as root.
4819 The bug described in (1) was not present in this case, but the tidying up
4820 of the other signals was missing. I have made the two cases consistent.
4822 3. The ignore_target_hosts setting on a manualroute router was being ignored
4823 for hosts that were looked up using the /MX notation.
4825 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
4828 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
4829 operated on the sender address. After changing the $sender_address to <>
4830 for the sender address verify, Exim was re-instated it as the original
4831 (before rewriting) address, but remembering that it had rewritten it, so it
4832 wasn't rewriting it again. This bug also had the effect of breaking the
4833 sender address verification caching when the sender address was rewritten.
4835 6. The ignore_target_hosts option was being ignored by the ipliteral router.
4836 This has been changed so that if the ip literal address matches
4837 ignore_target_hosts, the router declines.
4839 7. Added expansion conditions match_domain, match_address, and match_local_
4840 part (NOT match_host).
4842 8. The placeholder for the Received: header didn't have a length field set.
4844 9. Added code to Exim itself and to exim_lock to test for a specific race
4845 condition that could lead to file corruption when using MBX delivery. The
4846 issue is with the lockfile that is created in /tmp. If this file is removed
4847 after a process has opened it but before that process has acquired a lock,
4848 there is the potential for a second process to recreate the file and also
4849 acquire a lock. This could lead to two Exim processes writing to the file
4850 at the same time. The added code performs the same test as UW imapd; it
4851 checks after acquiring the lock that its file descriptor still refers to
4852 the same named file.
4854 10. The buffer for building added header lines was of fixed size, 8192 bytes.
4855 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
4858 11. Added the smtp_active_hostname option. If used, this will typically be made
4859 to depend on the incoming interface address. Because $interface_address is
4860 not set up until the daemon has forked a reception process, error responses
4861 that can happen earlier (such as "too many connections") no longer contain
4864 12. If an expansion in a condition on a "warn" statement fails because a lookup
4865 defers, the "warn" statement is abandoned, and the next ACL statement is
4866 processed. Previously this caused the whole ACL to be aborted.
4868 13. Added the iplsearch lookup type.
4870 14. Added ident_timeout as a log selector.
4872 15. Added tls_certificate_verified as a log selector.
4874 16. Added a global option tls_require_ciphers (compare the smtp transport
4875 option of the same name). This controls incoming TLS connections.
4877 17. I finally figured out how to make tls_require_ciphers do a similar thing
4878 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
4879 before starting the TLS session.
4881 18. Tabs are now shown as \t in -bP output.
4883 19. If the log selector return_path_on_delivery was set, Exim crashed when
4884 bouncing a message because it had too many Received: header lines.
4886 20. If two routers both had headers_remove settings, and the first one included
4887 a superfluous trailing colon, the final name in the first list and the
4888 first name in the second list were incorrectly joined into one item (with a
4889 colon in the middle).
4895 1. Added -C and -D options to the exinext utility, mainly to make it easier
4896 to include in the automated testing, but these could be helpful when
4897 multiple configurations are in use.
4899 2. The exinext utility was not formatting the output nicely when there was
4900 an alternate port involved in the retry record key, nor when there was a
4901 message id as well (for retries that were specific to a specific message
4902 and a specific host). It was also confused by IPv6 addresses, because of
4903 the additional colons they contain. I have fixed the IPv4 problem, and
4904 patched it up to do a reasonable job for IPv6.
4906 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
4907 delivery, the log line now contains "pipelined" if PIPELINING was used.
4909 4. An SMTP transport process used to panic and die if the bind() call to set
4910 an explicit outgoing interface failed. This has been changed; it is now
4911 treated in the same way as a connect() failure.
4913 5. A reference to $sender_host_name in the part of a conditional expansion
4914 that was being skipped was still causing a DNS lookup. This no longer
4917 6. The def: expansion condition was not recognizing references to header lines
4918 that used bh_ and bheader_.
4920 7. Added the _cache feature to named lists.
4922 8. The code for checking quota_filecount in the appendfile transport was
4923 allowing one more file than it should have been.
4925 9. For compatibility with Sendmail, the command line option
4933 and sets the incoming protocol and host name (for trusted callers). The
4934 host name and its colon can be omitted when only the protocol is to be set.
4935 Note the Exim already has two private options, -pd and -ps, that refer to
4936 embedded Perl. It is therefore impossible to set a protocol value of "d" or
4937 "s", but I don't think that's a major issue.
4939 10. A number of refactoring changes to the code, none of which should affect
4942 (a) The number of logging options was getting close to filling up the
4943 32-bit word that was used as a bit map. I have split them into two classes:
4944 those that are passed in the argument to log_write(), and those that are
4945 only ever tested independently outside of that function. These are now in
4946 separate 32-bit words, so there is plenty of room for expansion again.
4947 There is no change in the user interface or the logging behaviour.
4949 (b) When building, for example, log lines, the code previously used a
4950 macro that called string_cat() twice, in order to add two strings. This is
4951 not really sufficiently general. Furthermore, there was one instance where
4952 it was actually wrong because one of the argument was used twice, and in
4953 one call a function was used. (As it happened, calling the function twice
4954 did not affect the overall behaviour.) The macro has been replaced by a
4955 function that can join an arbitrary number of extra strings onto a growing
4958 (c) The code for expansion conditions now uses a table and a binary chop
4959 instead of a serial search (which was left over from when there were very
4960 few conditions). Also, it now recognizes conditions like "pam" even when
4961 the relevant support is not compiled in: a suitably worded error message is
4962 given if an attempt is made to use such a condition.
4964 11. Added ${time_interval:xxxxx}.
4966 12. A bug was causing one of the ddress fields not to be passed back correctly
4967 from remote delivery subprocesses. The field in question was not being
4968 subsequently used, so this caused to problems in practice.
4970 13. Added new log selectors queue_time and deliver_time.
4972 14. Might have fixed a bug in maildirsizefile handling that threw up
4973 "unexpected character" debug warnings, and recalculated the data
4974 unnecessarily. In any case, I expanded the warning message to give more
4977 15. Added the message "Restricted characters in address" to the statements in
4978 the default ACL that block characters like @ and % in local parts.
4980 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
4981 Three changes have been made:
4983 (a) There was a serious bug; a negative response to MAIL caused the whole
4984 recipient domain to be cached as invalid, thereby blocking all messages
4985 to all local parts at the same domain, from all senders. This bug has
4986 been fixed. The domain is no longer cached after a negative response to
4987 MAIL if the sender used is not empty.
4989 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
4992 (c) A new callout option, "use_sender" has been added for people who want
4993 the modified behaviour.
4999 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
5002 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
5003 indeed breaks things for older releases.
5005 3. Added additional logging to the case where there is a problem reading data
5006 from a filter that is running in a subprocess using a pipe, in order to
5007 try to track down a specific problem.
5009 4. Testing facility fudge: when running in the test harness and attempting
5010 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
5011 getting "No route to host". Convert this to a timeout.
5013 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
5016 6. Some OS don't have socklen_t but use size_t instead. This affects the
5017 fifth argument of getsockopt() amongst other things. This is now
5018 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
5019 can be set for individual OS. I have set it for SunOS5, OSF1, and
5020 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
5021 some earlier ones do not.
5023 7. Change 4.30/15 was not doing the test caselessly.
5025 8. The standard form for an IPv6 address literal was being rejected by address
5026 parsing in, for example, MAIL and RCPT commands. An example of this kind of
5027 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5028 this, as well as the form without the "IPv6" on the front (but only when
5029 address literals are enabled, of course).
5031 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
5033 10. Exim crashed if a message with an empty sender address specified by -f
5034 encountered a router with an errors_to setting. This could be provoked only
5035 by a command such as
5039 where an empty string was supplied; "<>" did not hit this bug.
5041 11. Installed PCRE release 4.5.
5043 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
5044 remained set. It is now erased.
5046 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
5047 times from message ids (which are base 36 rather than the normal 62).
5049 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
5050 were being counted as actual protocol errors, and logged if the log
5051 selector +smtp_protocol_error was set. One cannot be perfect in this test,
5052 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
5053 and DATA following a set of rejected RCPTs do not count as protocol errors.
5054 In other words, Exim assumes they were pipelined, though this may not
5055 actually be the case. Of course, in all cases the client gets an
5056 appropriate error code.
5058 15. If a lookup fails in an ACL condition, a message about the failure may
5059 be available; it is used if testing the ACL cannot continue, because most
5060 such messages specify what the cause of the deferral is. However, some
5061 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
5062 that caused an old message to be retained and used if a later statement
5063 caused a defer, replacing the real cause of the deferral.
5065 16. If an IP address had so many PTR records that the DNS lookup buffer
5066 was not large enough to hold them, Exim could crash while trying to process
5067 the truncated data. It now detects and logs this case.
5069 17. Further to 4.21/58, another change has been made: if (and only if) the
5070 first line of a message (the first header line) ends with CRLF, a bare LF
5071 in a subsequent header line has a space inserted after it, so as not to
5072 terminate the header.
5074 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
5075 unnecessarily, used atoi() instead of strtol(), and didn't check the
5076 termination when getting file sizes from file names by regex.
5078 19. Completely re-implemented the support for maildirsize files, in the light
5079 of a number of problems with the previous contributed implementation
5080 (4.30/29). In particular:
5082 . If the quota is zero, the maildirsize file is maintained, but no quota is
5085 . If the maildir directory does not exist, it is created before any attempt
5086 to write a maildirsize file.
5088 . The quota value in the file is just a cache; if the quota is changed in
5089 the transport, the new value overrides.
5091 . A regular expression is available for excluding directories from the
5094 20. The autoreply transport checks the characters in options that define the
5095 message's headers; it allows continued headers, but it was checking with
5096 isspace() after an embedded newline instead of explicitly looking for a
5099 21. If all the "regular" hosts to which an address was routed had passed their
5100 expiry times, and had not reached their retry times, the address was
5101 bounced, even if fallback hosts were defined. Now Exim should go on to try
5104 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
5105 equivalent code in the SMTP transport. Some hosts send humungous responses
5106 to HELO/EHLO, more than 1024 it seems.
5108 23. Refactoring: code in filter.c used (void *) for "any old type" but this
5109 gives compiler warnings in some environments. I've now done it "properly",
5112 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
5113 (because of problems with the built-in one) was declared to return uschar *
5114 instead of char *, causing compiler failure.
5116 25. Fixed a file descriptor leak when processing alias/forward files.
5118 26. Fixed a minor format string issue in dbfn.c.
5120 27. Typo in exim.c: ("dmbnz" for "dbmnz").
5122 28. If a filter file refered to $h_xxx or $message_headers, and the headers
5123 contained RFC 2047 "words", Exim's memory could, under certain conditions,
5126 29. When a sender address is verified, it is cached, to save repeating the test
5127 when there is more than one recipient in a message. However, when the
5128 verification involves a callout, it is possible for different callout
5129 options to be set for different recipients. It is too complicated to keep
5130 track of this in the cache, so now Exim always runs a verification when a
5131 callout is required, relying on the callout cache for the optimization.
5132 The overhead is duplication of the address routing, but this should not be
5135 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
5136 to be verified with callout=postmaster, and the main callout worked but the
5137 postmaster check failed, the verification correctly failed. However, if a
5138 subsequent RCPT command asked for sender verification *without* the
5139 postmaster check, incorrect caching caused this verification also to fail,
5142 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
5143 it was not caching the DNS options (qualify_single, search_parents) that
5144 were used when the lookup failed. A subsequent lookup with different
5145 options therefore always gave the same answer, though there were cases
5146 where it should not have. (Example: a "domains = !$mx_any" option on a
5147 dnslookup router: the "domains" option is always processed without any
5148 widening, but the router might have qualify_single set.) Now Exim uses the
5149 cached value only when the same options are set.
5151 32. Added John Jetmore's "exipick" utility to the distribution.
5153 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
5154 than a timeout (e.g. a certificate is required, and is not provided), an
5155 Exim server now closes the connection immediately. Previously it waited for
5156 the client to close - but if the client is SSL, it seems that they each
5157 wait for each other, leading to a delay before one of them times out.
5159 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
5160 maintained 0.8.x compatibility because I don't think many are using it, and
5161 it is clearly obsolete.
5163 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
5166 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
5167 client certificate was expired. A simple patch fixes this, though I don't
5168 understand the full logic of why the verify callback is called multiple
5171 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
5172 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
5173 which causes problems with some clients (such as the Certicom SSL Plus
5174 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
5175 disables the coutermeasure allowing Eudora to connect."
5177 38. Exim was not checking that a write() to a log file succeeded. This could
5178 lead to Bad Things if a log got too big, in particular if it hit a file
5179 size limit. Exim now panics and dies if it cannot write to a log file, just
5180 as it does if it cannot open a log file.
5182 39. Modified OS/Makefile-Linux so that it now contains
5184 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
5186 The two -D definitions ensure that Exim is compiled with large file
5187 support, which makes it possible to handle log files that are bigger than
5190 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
5191 instance) a domain was checked against a named list that involved a lookup,
5192 causing $domain_data to be set, then another domain was checked against the
5193 same list, then the first domain was re-checked, the value of $domain_data
5194 after the final check could be wrong. In particular, if the second check
5195 failed, it could be set empty. This bug probably also applied to
5198 41. The strip_trailing_dot option was not being applied to the address given
5199 with the -f command-line option.
5201 42. The code for reading a message's header from the spool was incrementing
5202 $received_count, but never initializing it. This meant that the value was
5203 incorrect (doubled) while delivering a message in the same process in which
5204 it was received. In the most common configuration of Exim, this never
5205 happens - a fresh exec is done - but it can happen when
5206 deliver_drop_privilege is set.
5208 43. When Exim logs an SMTP synchronization error - client data sent too soon -
5209 it now includes up to 150 characters of the unexpected data in the log
5212 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
5213 and building data strings. The size of both of these buffers was 10 000
5214 bytes - far larger than anybody would *ever* want, thought I. Needless to
5215 say, somebody hit the limit. I have increased the maximum line length to
5216 20 000 and the maximum data length of concatenated lines to 100 000. I have
5217 also fixed two bugs, because there was no checking on these buffers. Tsk,
5218 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
5219 buffer is too small.
5221 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
5222 lsearch lookups. Now it does.
5224 46. When parsing a route_list item in a manualroute router, a fixed-length
5225 buffer was used for the list of hosts. I made this 1024 bytes long,
5226 thinking that nobody would ever have a list of hosts that long. Wrong.
5227 Somebody had a whole pile of complicated expansion conditions, and the
5228 string was silently truncated, leading to an expansion error. It turns out
5229 that it is easier to change to an unlimited length (owing to other changes
5230 that have happened since this code was originally written) than to build
5231 structure for giving a limitation error. The length of the item that
5232 expands into the list of hosts is now unlimited.
5234 47. The lsearch lookup could not handle data where the length of text line was
5235 more than 4095 characters. Such lines were truncated, leading to shortened
5236 data being returned. It should now handle lines of any length.
5238 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
5239 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
5242 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
5243 replacing strings such as BIN_DIRECTORY in a source file: the replacement
5244 no longer happens in comment lines. A list of replacements is now placed
5245 at the head of all of the source files, except those whose only change is
5246 to replace PERL_COMMAND in the very first #! line.
5248 50. Replaced the slow insertion sort in queue.c, for sorting the list of
5249 messages on the queue, with a bottom-up merge sort, using code contributed
5250 by Michael Haardt. This should make operations like -bp somewhat faster on
5251 large queues. It won't affect queue runners, except when queue_run_in_order
5254 51. Installed eximstats 1.31 in the distribution.
5256 52. Added support for SRV lookups to the dnslookup router.
5258 53. If an ACL referred to $message_body or $message_body_end, the value was not
5259 reset for any messages that followed in the same SMTP session.
5261 54. The store-handling optimization for building very long strings was not
5262 differentiating between the different store pools. I don't think this
5263 actually made any difference in practice, but I've tidied it.
5265 55. While running the routers to verify a sender address, $sender_address
5266 was still set to the sender address. This is wrong, because when routing to
5267 send a bounce to the sender, it would be empty. Therefore, I have changed
5268 it so that, while verifying a sender address, $sender_address is set to <>.
5269 (There is no change to what happens when verifying a recipient address.)
5271 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
5272 target A or AAAA records (if not already returned) without resetting the
5273 qualify_single or search_parents options of the DNS resolver. These are
5274 inappropriate in this case because the targets of MX and SRV records must
5275 be FQDNs. A broken DNS record could cause trouble if it happened to have a
5276 target that, when qualified, matched something in the local domain. These
5277 two options are now turned off when doing these lookups.
5279 57. It seems that at least some releases of Reiserfs (which does not have the
5280 concept of a fixed number of inodes) returns zero and not -1 for the
5281 number of available inodes. This interacted badly with check_spool_inodes,
5282 which assumed that -1 was the "no such thing" setting. What I have done is
5283 to check that the total number of inodes is greater than zero before doing
5284 the test of how many are available.
5286 58. When a "warn" ACL statement has a log_message modifier, the message is
5287 remembered, and not repeated. This is to avoid a lot of repetition when a
5288 message has many recipients that cause the same warning to be written.
5289 Howewer, Exim was preserving the list of already written lines for an
5290 entire SMTP session, which doesn't seem right. The memory is now reset if a
5291 new message is started.
5293 59. The "rewrite" debugging flag was not showing the result of rewriting in the
5294 debugging output unless log_rewrite was also set.
5296 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
5297 of (int)(handle) when we know that handle contains (void *)(-1).
5299 61. The Exim daemon panic-logs an error return when it closes the incoming
5300 connection. However "connection reset by peer" seems to be common, and
5301 isn't really an error worthy of noting specially, so that particular error
5304 62. When Exim is trying to find all the local interfaces, it used to panic and
5305 die if the ioctl to get the interface flags failed. However, it seems that
5306 on at least one OS (Solaris 9) it is possible to have an interface that is
5307 included in the list of interfaces, but for which you get a failure error
5308 for this call. This happens when the interface is not "plumbed" into a
5309 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
5310 failure of the "get flags" call assumes that the interface is down.
5312 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
5313 makes life easier for people who are doing arithmetic on fields extracted
5314 from dates, where you often get leading zeros that should not be
5315 interpreted as octal.
5317 64. Added qualify_domain to the redirect router, to override the global
5320 65. If a pathologically long header line contained very many addresses (the
5321 report of this problem mentioned 10 000) and each of them was rewritten,
5322 Exim could use up a very large amount of memory. (It kept on making new
5323 copies of the header line as it rewrote, and never released the old ones.)
5324 At the expense of a bit more processing, the header rewriting function has
5325 been changed so that it no longer eats memory in this way.
5327 66. The generation of the Received: header has been moved from the time that a
5328 message starts to be received, to the time that it finishes. The timestamp
5329 in the Received: header should now be very close to that of the <= log
5330 line. There are two side-effects of this change:
5332 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
5333 logged header lines no longer include the local Received: line, because
5334 it has not yet been created. The same applies to a copy of the message
5335 that is returned to a non-SMTP sender when a message is rejected.
5337 (b) When a filter file is tested using -bf, no additional Received: header
5338 is added to the test message. After some thought, I decided that this
5341 This change does not affect the value of $received_for. It is still set
5342 after address rewriting, but before local_scan() is called.
5344 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
5346 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
5347 gave an unhelpful panic error message, and a defer error. I have managed to
5348 change this behaviour so that it now rejects any supplied certificate,
5349 which seems right, as the list of acceptable certificates is empty.
5351 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
5352 gave an unhelpful defer error. I have not managed to make this reject any
5353 supplied certificates, but the error message it gives is "no certificate
5354 supplied", which is not helpful.
5356 70. exigrep's output now also includes lines that are not associated with any
5357 message, but which match the given pattern. Implemented by a patch from
5358 Martin Sluka, which also tidied up the Perl a bit.
5360 71. Recipient callout verification, like sender verification, was using <> in
5361 the MAIL FROM command. This isn't really the right thing, since the actual
5362 sender may affect whether the remote host accepts the recipient or not. I
5363 have changed it to use the actual sender in the callout; this means that
5364 the cache record is now keyed on a recipient/sender pair, not just the
5365 recipient address. There doesn't seem to be a real danger of callout loops,
5366 since a callout by the remote host to check the sender would use <>.
5367 [SEE ABOVE: changed after hitting problems.]
5369 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
5370 temporary errors. However, in the case of such a code being given after
5371 the end of a data transmission (i.e. after ".") Exim was failing to write
5372 a retry record for the message. (Yes, there was some broken host that was
5373 actually sending 8xx at this point.)
5375 73. An unknown lookup type in a host list could cause Exim to panic-die when
5376 the list was checked. (An example that provoked this was putting <; in the
5377 middle of a list instead of at the start.) If this happened during a DATA
5378 ACL check, a -D file could be left lying around. This kind of configuration
5379 error no longer causes Exim to die; instead it causes a defer errror. The
5380 incident is still logged to the main and panic logs.
5382 74. Buglet left over from Exim 3 conversion. The message "too many messages
5383 in one connection" was written to the rejectlog but not the mainlog, except
5384 when address rewriting (yes!) was being logged.
5386 75. Added write_rejectlog option.
5388 76. When a system filter was run not as root (that is, when system_filter_user
5389 was set), the values of the $n variables were not being returned to the
5390 main process; thus, they were not subsequently available in the $sn
5393 77. Added +return_path_on_delivery log selector.
5395 78. A connection timeout was being treated differently from recipients deferred
5396 when testing hosts_max_try with a message that was older than the host's
5397 retry timeout. (The host should not be counted, thus allowing all hosts to
5398 be tried at least once before bouncing.) This may have been the cause of an
5399 occasionally reported bug whereby a message would remain on the queue
5400 longer than the retry timeout, but would be bounced if a delivery was
5401 forced. I say "may" because I never totally pinned down the problem;
5402 setting up timeout/retry tests is difficult. See also the next item.
5404 79. The ultimate address timeout was not being applied to errors that involved
5405 a combination of host plus message (for example, a timeout on a MAIL
5406 command). When an address resolved to a number of possible hosts, and they
5407 were not all tried for each delivery (e.g. because of hosts_max_try), a
5408 message could remain on the queue longer than the retry timeout.
5410 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
5413 81. Fixed an obscure SMTP outgoing bug which required at least the following
5414 conditions: (a) there was another message waiting for the same server;
5415 (b) the server returned 5xx to all RCPT commands in the first message so
5416 that the message was not completed; (c) the server dropped the connection
5417 or gave a negative response to the RSET that Exim sends to abort the
5418 transaction. The observed case was a dropped connection after DATA that had
5419 been sent in pipelining mode. That is, the server had advertised PIPELINING
5420 but was not implementing it correctly. The effect of the bug was incorrect
5421 behaviour, such as trying another host, and this could lead to a crash.
5427 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
5428 and daemon.c were passed as pointers to ints; they should have been
5429 pointers to socklen_t variables (which are typically unsigned ints).
5431 2. Some signed/unsigned type warnings in the os.c file for Linux have been
5434 3. Fixed a really odd bug that affected only the testing scheme; patching a
5435 certain fixed string in the binary changed the value of another string that
5436 happened to be identical to the end of the original first string.
5438 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
5439 name", it returns that address as the IP address. On some operating
5440 systems (e.g. Solaris), it also passes back the IP address string as the
5441 "host name". However, on others (e.g. Linux), it passes back an empty
5442 string. Exim wasn't checking for this, and was changing the host name to an
5443 empty string, assuming it had been canonicized.
5445 5. Although rare, it is permitted to have more than one PTR record for a given
5446 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
5447 all the names associated with an address, because they do in Solaris.
5448 However, it seems that they do not in Linux for data that comes from the
5449 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
5450 I found this out when I moved to a new Linux workstation and tried to run
5451 the Exim test suite.
5453 To get round this problem I have changed the code so that it now does its
5454 own call to the DNS to look up PTR records when searching for a host name.
5455 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
5456 addresses that are only in /etc/hosts are still found.
5458 This behaviour is, however, controlled by an option called host_lookup_
5459 order, which defaults to "bydns:byaddr". If people want to use the other
5460 order, or indeed, just use one or the other means of lookup, they can
5461 specify it in this variable.
5463 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
5464 some operating systems, this comes back from gethostbyaddr() as an empty
5465 string, and this is what Exim used to test for. However, it seems that in
5466 other systems, "." is yielded. Exim now tests for this case too.
5468 7. The values of check_spool_space and check_log_space are now held internally
5469 as a number of kilobytes instead of an absolute number of bytes. If a
5470 numbers is specified without 'K' or 'M', it is rounded up to the nearest
5471 kilobyte. This means that much larger values can be stored.
5473 8. Exim monitor: an attempt to get the action menu when not actually pointing
5474 at a message produces an empty menu entitled "No message selected". This
5475 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
5476 no entries in it ("Shell widget menu has zero width and/or height"). So I
5477 have added a single, blank menu entry in this case.
5479 9. Added ${quote_local_part.
5481 10. MIME decoding is now applied to the contents of Subject: header lines when
5484 11. Now that a reference to $sender_host_address automatically causes a reverse
5485 lookup to occur if necessary (4.13/18), there is no need to arrange for a
5486 host lookup before query-style lookups in lists that might use this
5487 variable. This has therefore been abolished, and the "net-" prefix is no
5488 longer necessary for query-style lookups.
5490 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
5491 have been a typo for LFLAGS, so it has been changed.
5493 13. The install script calls Exim with "-C /dev/null" in order to find the
5494 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
5495 to be output. Howeve, since Exim outputs its version number before the
5496 error, it didn't break the script. It just looked ugly. I fixed this by
5497 always allowing "-C /dev/null" if the caller is root.
5499 14. Ignore overlarge ACL variable number when reading spool file - insurance
5500 against a later release with more variables having written the file.
5502 15. The standard form for an IPv6 address literal was being rejected by EHLO.
5503 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5504 this, as well as the form without the "IPv6" on the front.
5506 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
5507 OS/Makefile-Darwin file.
5509 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
5510 only to LDAP libraries that do not have LDAP_OPT_DEREF.
5512 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
5513 variable. However, some OS use ints for inodes. I've added cast to long int
5514 to get rid of the compiler warning.
5516 19. I had forgotten to lock out "/../" in configuration file names when
5517 ALT_CONFIG_PREFIX was set.
5519 20. Routers used for verification do not need to specify transports. However,
5520 if such a router generated a host list, and callout was configured, Exim
5521 crashed, because it could not find a port number from the (non-existent)
5522 transport. It now assumes port 25 in this circumstance.
5524 21. Added the -t option to exigrep.
5526 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
5527 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
5528 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
5529 them is set, LOOKUP_LSEARCH is forced.
5531 23. "exim -bV" now outputs a list of lookups that are included in the binary.
5533 24. Added sender and host information to the "rejected by local_scan()" log
5534 line; previously there was no indication of these.
5536 25. Added .include_if_exists.
5538 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
5539 Linux. It turns out that not all file systems support this. Apparently some
5540 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
5541 do it.) To cope with this, the error EINVAL, which means that sync-ing is
5542 not supported on the file descriptor, is now ignored when Exim is trying to
5543 sync a directory. This applies only to Linux.
5545 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
5547 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
5548 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
5551 29. Added support for maildirsize files from supplied patch (modified a bit).
5553 30. The use of :fail: followed by an empty string could lead Exim to respond to
5554 sender verification failures with (e.g.):
5556 550 Verification failed for <xxx>
5557 550 Sender verify failed
5559 where the first response line was missing the '-' that indicates it is not
5560 the final line of the response.
5562 31. The loop for finding the name of the user that called Exim had a hardwired
5563 limit of 10; it now uses the value of finduser_retries, which is used for
5564 all other user lookups.
5566 32. Added $received_count variable, available in data and not_smtp ACLs, and at
5569 33. Exim was neglecting to zero errno before one call of strtol() when
5570 expanding a string and expecting an integer value. On some systems this
5571 resulted in spurious "integer overflow" errors. Also, it was casting the
5572 result into an int without checking.
5574 34. Testing for a connection timeout using "timeout_connect" in the retry rules
5575 did not work. The code looks as if it has *never* worked, though it appears
5576 to have been documented since at least releast 1.62. I have made it work.
5578 35. The "timeout_DNS" error in retry rules, also documented since at least
5579 1.62, also never worked. As it isn't clear exactly what this means, and
5580 clearly it isn't a major issue, I have abolished the feature by treating it
5581 as "timeout", and writing a warning to the main and panic logs.
5583 36. The display of retry rules for -brt wasn't always showing the error code
5586 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
5587 timeout_connect_A, timeout_connect_MX.
5589 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
5590 to the empty sender.
5592 39. The daemon was not analysing the content of -oX till after it had closed
5593 stderr and disconnected from the controlling terminal. This meant that any
5594 syntax errors were only noted on the panic log, and the return code from
5595 the command was 0. By re-arranging the code a little, I've made the
5596 decoding happen first, so such errors now appear on stderr, and the return
5597 code is 1. However, the actual setting up of the sockets still happens in
5598 the disconnected process, so errors there are still only recorded on the
5601 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
5602 connections (as happens on some IP stacks) was logged at start up time as
5603 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
5604 it from "IPv6 and IPv4", which means that two separate sockets are being
5607 41. The debug output for gethostbyname2() or getipnodebyname() failures now
5608 says whether AF_INET or AF_INET6 was passed as an argument.
5610 42. Exiwhat output was messed up when time zones were included in log
5613 43. Exiwhat now gives more information about the daemon's listening ports,
5614 and whether -tls-on-connect was used.
5616 44. The "port" option of the smtp transport is now expanded.
5618 45. A "message" modifier in a "warn" statement in a non-message ACL was being
5619 silently ignored. Now an error message is written to the main and panic
5622 46. There's a new ACL modifier called "logwrite" which writes to a log file
5623 as soon as it is encountered.
5625 47. Added $local_user_uid and $local_user_gid at routing time.
5627 48. Exim crashed when trying to verify a sender address that was being
5630 49. Exim was recognizing only a space character after ".include". It now also
5631 recognizes a tab character.
5633 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
5634 extracting the relevant information from the specification. The man page no
5635 longer contains scrambled data for the -d option, and I've added a section
5636 at the front about calling Exim under different names.
5638 51. Added "extra_headers" argument to the "mail" command in filter files.
5640 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
5643 53. Installed eximstats 1.29.
5645 54. Added transport_filter_timeout as a generic transport option.
5647 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
5648 Cc: header lines. This was required by RFC 822, but it not required by RFC
5651 56. Exim used to add From:, Date:, and Message-Id: header lines to any
5652 incoming messages that did not have them. Now it does so only if the
5653 message originates locally, that is, if there is no associated remote host
5654 address. When Resent- header lines are present, this applies to the Resent-
5655 lines rather than the non-Resent- lines.
5657 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
5658 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
5660 58. Messages for configuration errors now include the name of the main
5661 configuration file - useful now that there may be more than one file in a
5662 list (.included file names were always shown).
5664 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
5665 for those rare installations that do not start the daemon as root or run it
5666 setuid root. I've cut out the call to initgroups() if the daemon is not
5669 60. The Exim user and group can now be bound into the binary as text strings
5670 that are looked up at the start of Exim's processing.
5672 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
5674 62. Added $mailstore_basename variable.
5676 63. Installed patch to sieve.c from Michael Haardt.
5678 64. When Exim failed to open the panic log after failing to open the main log,
5679 the original message it was trying to log was written to stderr and debug
5680 output, but if they were not available (the usual case in production), it
5681 was lost. Now it is written to syslog before the two lines that record the
5682 failures to open the logs.
5684 65. Users' Exim filters run in subprocesses under the user's uid. It is
5685 possible for a "deliver" command or an alias in a "personal" command to
5686 provoke an address rewrite. If logging of address rewriting is configured,
5687 this fails because the process is not running as root or exim. There may be
5688 a better way of dealing with this, but for the moment (because 4.30 needs
5689 to be released), I have disabled address rewrite logging when running a
5690 filter in a non-root, non-exim process.
5696 1. The buildconfig auxiliary program wasn't quoting the value set for
5697 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
5698 not defined. This bug was masked in 4.22 by the effect that was fixed in
5701 2. Some messages that were rejected after a message id was allocated were
5702 shown as "incomplete" by exigrep. It no longer does this for messages that
5703 are rejected by local_scan() or the DATA or non-SMTP ACLs.
5705 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
5706 have allow_domain_literals set, the ID did not get logged in the <= line.
5707 Domain literals are now always recognized in Message-ID: header lines.
5709 4. The first argument for a ${extract expansion item is the key name or field
5710 number. Leading and trailing spaces in this item were not being ignored,
5711 causing some misleading effects.
5713 5. When deliver_drop_privilege was set, single queue runner processes started
5714 manually (i.e. by the command "exim -q") or by the daemon (which uses the
5715 same command in the process it spins off) were not dropping privilege.
5717 6. When the daemon running as "exim" started a queue runner, it always
5718 re-executed Exim in the spun-off process. This is a waste of effort when
5719 deliver_drop_privilege is set. The new process now just calls the
5720 queue-runner function directly.
5726 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
5729 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
5730 ignored. Though the use of -oP was forcing the writing of a pid file, it
5731 was always written to the default place.
5733 3. If the message "no IP address found for host xxxx" is generated during
5734 incoming verification, it is now followed by identification of the incoming
5735 connection (so you can more easily find what provoked it).
5737 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
5739 5. Added some features to "harden" Exim a bit more against certain attacks:
5741 (a) There is now a build-time option called FIXED_NEVER_USERS that can
5742 be put in Local/Makefile. This is like the never_users runtime option,
5743 but it cannot be overridden. The default setting is "root".
5745 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
5746 prefix string with which any file named in a -C command line option
5749 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
5750 is retained for -C and -D only if the caller of Exim is root. Without
5751 it, the exim user may also use -C and -D and retain privilege.
5753 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
5754 command line option is disabled.
5756 6. Macro names set by the -D option must start with an upper case letter, just
5757 like macro names defined in the configuration file.
5759 7. Added "dereference=" facility to LDAP.
5761 8. Two instances of the typo "uknown" in the source files are fixed.
5763 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
5764 the Configure-Makefile script screwed up while processing it.
5766 10. Incorporated PCRE 4.4.
5768 11. The SMTP synchronization check was not operating right at the start of an
5769 SMTP session. For example, it could not catch a HELO sent before the client
5770 waited for the greeting. There is now a check for outstanding input at the
5771 point when the greeting is written. Because of the duplex, asynchronous
5772 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
5773 way, but not yet received, when the check is performed.
5775 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
5776 on TCP/IP sockets, because this apparently causes some broken clients to
5779 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
5780 unchanged) from the Cygwin maintainer.
5782 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
5783 was supported instead of testing for mbx. Effectively a typo.
5785 15. The spa authenticator server code was not checking that the input it
5786 received was valid base64.
5788 16. The debug output line for the "set" modifier in ACLs was not showing the
5789 name of the variable that was being set.
5791 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
5793 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
5794 lookup on its own. Something else was needed to trigger the lookup. For
5795 example, a match in host_lookup or the need for a host name in a host list.
5796 Now, if $sender_host_name is referenced and the host name has not yet been
5797 looked up, a lookup is performed. If the lookup fails, the variable remains
5798 empty, and $host_lookup_failed is set to "1".
5800 19. Added "eqi" as a case-independent comparison operator.
5802 20. The saslauthd authentication condition could segfault if neither service
5803 nor realm was specified.
5805 21. If an overflowing value such as "2048M" was set for message_size_limit, the
5806 error message that was logged was misleading, and incoming SMTP
5807 connections were dropped. The message is now more accurate, and temporary
5808 errors are given to SMTP connections.
5810 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
5811 (except RSET) with a 421 error, until QUIT is received. However, it was
5812 failing to send a response to QUIT.
5814 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
5815 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
5816 ACL is now run after the helo_try_verify_hosts code.
5818 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
5819 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
5820 case-independent, so other case variants are also recognized). Apparently
5821 some systems use these upper case variants.
5823 25. If more than two messages were waiting for the same host, and a transport
5824 filter was specified for the transport, Exim sent two messages over the
5825 same TCP/IP connection, and then failed with "socket operation on non-
5826 socket" when it tried to send the third.
5828 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
5830 27. The extern definition of crypt16() in expand.c was not being excluded when
5831 the OS had its own crypt16() function.
5833 28. Added bounce_return_body as a new option, and bounce_return_size_limit
5834 as a preferred synonym for return_size_limit, both as an option and as an
5837 29. Added LIBS=-liconv to OS/Makefile-OSF1.
5839 30. Changed the default configuration ACL to relax the local part checking rule
5840 for addresses that are not in any local domains. For these addresses,
5841 slashes and pipe symbols are allowed within local parts, but the sequence
5842 /../ is explicitly forbidden.
5844 31. SPA server authentication was not clearing the challenge buffer before
5847 32. log_message in a "warn" ACL statement was writing to the reject log as
5848 well as to the main log, which contradicts the documentation and doesn't
5849 seem right (because no rejection is happening). So I have stopped it.
5851 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
5852 However, I am unable to do any testing of this.
5854 34. Fixed an infelicity in the appendfile transport. When checking directories
5855 for a mailbox, to see if any needed to be created, it was accidentally
5856 using path names with one or more superfluous leading slashes; tracing
5857 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
5859 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
5860 contents are added to the log line that is written for every discarded
5861 recipient. (Previously a log_message setting was ignored.)
5863 36. The ${quote: operator now quotes the string if it is empty.
5865 37. The install script runs exim in order to find its version number. If for
5866 some reason other than non-existence or emptiness, which it checks, it
5867 could not run './exim', it was installing it with an empty version number,
5868 i.e. as "exim-". This error state is now caught, and the installation is
5871 38. An argument was missing from the function that creates an error message
5872 when Exim fails to connect to the socket for saslauthd authentication.
5873 This could cause Exim to crash, or give a corrupted message.
5875 39. Added isip, isip4, and isip6 to ${if conditions.
5877 40. The ACL variables $acl_xx are now saved with the message, and can be
5878 accessed later in routers, transports, and filters.
5880 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
5881 strings in the file are not string-expanded.
5883 42. If a MAIL command specified a SIZE value that was too large to fit into an
5884 int variable, the check against message_size_limit failed. Such values are
5885 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
5886 day this will have to be increased, but I don't think I want to be around
5887 when emails are that large.
5894 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
5895 iconv() is not standard in FreeBSD.
5897 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
5898 IPv6 enabled. The observed symptom was a segmentation fault on return from
5899 the function os_common_find_running_interfaces() in src/os.c.
5901 3. In the check_special_case() function in daemon.c I had used "errno" as an
5902 argument name, which causes warnings on some systems. This was basically a
5903 typo, since it was named "eno" in the comments!
5905 4. The code that waits for the clock to tick (at a resolution of some fraction
5906 of a second) so as to ensure message-id uniqueness was always waiting for
5907 at least one whole tick, when it could have waited for less. [This is
5908 almost certainly not relevant at current processor speeds, where it is
5909 unlikely to ever wait at all. But we try to future-proof.]
5911 5. The function that sleeps for a time interval that includes fractions of a
5912 second contained a race. It did not block SIGALRM between setting the
5913 timer, and suspending (a couple of lines later). If the interval was short
5914 and the sigsuspend() was delayed until after it had expired, the suspension
5915 never ended. On busy systems this could lead to processes getting stuck for
5918 6. Some uncommon configurations may cause a lookup to happen in a queue runner
5919 process, before it forks any delivery processes. The open lookup caching
5920 mechanism meant that the open file or database connection was passed into
5921 the delivery process. The problem was that delivery processes always tidy
5922 up cached lookup data. This could cause a problem for the next delivery
5923 process started by the queue runner, because the external queue runner
5924 process does not know about the closure. So the next delivery process
5925 still has data in the lookup cache. In the case of a file lookup, there was
5926 no problem because closing a file descriptor in a subprocess doesn't affect
5927 the parent. However, if the lookup was caching a connection to a database,
5928 the connection was closed, and the second delivery process was likely to
5929 see errors such as "PGSQL: query failed: server closed the connection
5930 unexpectedly". The problem has been fixed by closing all cached lookups
5931 in a queue runner before running a delivery process.
5933 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
5934 seem to have the "const" qualifier which it has on other OS. I've
5937 8. Change 4.21/2 was too strict. It is only if there are two authenticators
5938 *of the same type* (client or server) with the same public name that an
5939 error should be diagnosed.
5941 9. When Exim looked up a host name for an IP address, but failed to find the
5942 original IP address when looking up the host name (a safety check), it
5943 output the message "<ip address> does not match any IP for NULL", which was
5944 confusing, to say the least. The bug was that the host name should have
5945 appeared instead of "NULL".
5947 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
5948 user that is built into the binary, and the -C or -D options is used, root
5949 privilege is dropped before the configuration file is read. In addition,
5950 logging is switched to stderr instead of the normal log files. If the
5951 configuration then re-defines the Exim user, the unprivileged environment
5952 is probably not what is expected, so Exim logs a panic warning message (but
5955 However, if deliver_drop_privilege is set, the unprivileged state may well
5956 be exactly what is intended, so the warning has been cut out in that case,
5957 and Exim is allowed to try to write to its normal log files.
5963 1. smtp_return_error_details was not giving details for temporary sender
5964 or receiver verification errors.
5966 2. Diagnose a configuration error if two authenticators have the same public
5969 3. Exim used not to create the message log file for a message until the first
5970 delivery attempt. This could be confusing when incoming messages were held
5971 for policy or load reasons. The message log file is now created at the time
5972 the message is received, and an initial "Received" line is written to it.
5974 4. The automatically generated man page for command line options had a minor
5975 bug that caused no ill effects; however, a more serious problem was that
5976 the procedure for building the man page automatically didn't always
5977 operate. Consequently, release 4.20 contains an out-of-date version. This
5978 shouldn't happen again.
5980 5. When building Exim with embedded Perl support, the script that builds the
5981 Makefile was calling 'perl' to find its compile-time parameters, ignoring
5982 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
5984 6. The freeze_tell option was not being used for messages that were frozen on
5985 arrival, either by an ACL or by local_scan().
5987 7. Added the smtp_incomplete_transaction log selector.
5989 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
5990 was accepting AUTH without a new EHLO.
5992 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
5993 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
5996 10. Logging of TCP/IP connections (when configured) now happens in the main
5997 daemon process instead of the child process, so that the TCP/IP connection
5998 count is more accurate (but it can never be perfect).
6000 11. The use of "drop" in a nested ACL was not being handled correctly in the
6001 outer ACL. Now, if condition failure induced by the nested "drop" causes
6002 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
6003 or "require"), the connection is dropped.
6005 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
6006 that yield "discard" can now be used with an "accept" or a "discard" verb,
6007 but an error is generated for any others (because I can't see a useful way
6008 to define what should happen).
6010 13. When an ACL is read dynamically from a file (or anywhere else), the lines
6011 are now processed in the same way as lines in the Exim configuration file.
6012 In particular, continuation lines are supported.
6014 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
6016 15. Added -ti meaning -t -i.
6018 16. Check for letters, digits, hyphens, and dots in the names of dnslist
6019 domains, and warn by logging if others are found.
6021 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
6022 returned from GIFCONF when Exim is trying to find the list of interfaces on
6023 a host. The code in os.c has been modified to copy each ifreq to an aligned
6024 structure in all cases.
6026 Also, in some cases, the returned ifreq's were being copied to a 'struct
6027 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
6028 means the last couple of bytes of an IPv6 address could be chopped if the
6029 ifreq contained only a normal sockaddr (14 bytes storage).
6031 18. Named domain lists were not supported in the hosts_treat_as_local option.
6032 An entry such as +xxxx was not recognized, and was treated as a literal
6035 19. Ensure that header lines added by a DATA ACL are included in the reject log
6036 if the ACL subsequently rejects the message.
6038 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
6039 MD5 (which is deprecated).
6041 21. When testing a filter file using -bf, Exim was writing a message when it
6042 took the sender from a "From " line in the message, but it was not doing so
6043 when it took $return_path from a Return-Path: header line. It now does.
6045 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
6046 with a valid header line field name (a series of printing characters
6047 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
6049 23. Changed "disc" in the source to "disk" to conform to the documentation and
6050 the book and for uniformity.
6052 24. Ignore Sendmail's -Ooption=value command line item.
6054 25. When execve() failed while trying to run a command in a pipe transport,
6055 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
6056 could be confused with a return value of 69 from the command itself. This
6057 has been changed to 127, the value the shell returns if it is asked to run
6058 a non-existent command. The wording for the related log line suggests a
6059 non-existent command as the problem.
6061 26. If received_header_text expands to an empty string, do not add a Received:
6062 header line to the message. (Well, it adds a token one on the spool, but
6063 marks it "old" so that it doesn't get used or transmitted.)
6065 27. Installed eximstats 1.28 (addition of -nt option).
6067 28. There was no check for failure on the call to getsockname() in the daemon
6068 code. This can fail if there is a shortage of resources on the system, with
6069 ENOMEM, for example. A temporary error is now given on failure.
6071 29. Contrary to the C standard, it seems that in some environments, the
6072 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
6073 program. Exim now does this explicitly; it affects the formatting of
6074 timestamps using strftime().
6076 30. If exiqsumm was given junk data, it threw up some uninitialized variable
6077 complaints. I've now initialized all the variables, to avoid this.
6079 32. Header lines added by a system filter were not being "seen" during
6080 transport-time rewrites.
6082 33. The info_callback() function passed to OpenSSL is set up with type void
6083 (*)(SSL *, int, int), as described somewhere. However, when calling the
6084 function (actually a macro) that sets it up, the type void(*)() is
6085 expected. I've put in a cast to prevent warnings from picky compilers.
6087 34. If a DNS black list lookup found a CNAME record, but there were no A
6088 records associated with the domain it pointed at, Exim crashed.
6090 35. If a DNS black list lookup returned more than one A record, Exim ignored
6091 all but the first. It now scans all returned addresses if a particular IP
6092 value is being sought. In this situation, the contents of the
6093 $dnslist_value variable are a list of all the addresses, separated by a
6096 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
6097 to accept a host name and all its aliases if the forward lookup for any of
6098 them yielded the IP address of the incoming connection. Now it accepts only
6099 those names whose forward lookup yields the correct IP address. Any other
6100 names are discarded. This closes a loophole whereby a rogue DNS
6101 administrator could create reverse DNS records to break through a
6102 wildcarded host restriction in an ACL.
6104 37. If a user filter or a system filter that ran in a subprocess used any of
6105 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
6106 the wrong values were passed to the pipe command ($thisaddress had the
6107 value of $0, $0 had the value of $1, etc). This bug was introduced by
6108 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
6110 38. Improved the line breaking for long SMTP error messages from ACLs.
6111 Previously, if there was no break point between 40 and 75 characters, Exim
6112 left the rest of the message alone. Two changes have been made: (a) I've
6113 reduced the minimum length to 35 characters; (b) if it can't find a break
6114 point between 35 and 75 characters, it looks ahead and uses the first one
6115 that it finds. This may give the occasional overlong line, but at least the
6116 remaining text gets split now.
6118 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
6119 file descriptors might be low, and that setting 1000 would always raise it.
6120 It turns out that in some environments, the limit is already over 1000 and
6121 that lowering it causes trouble. So now Exim takes care not to decrease it.
6123 40. When delivering a message, the value of $return_path is set to $sender_
6124 address at the start of routing (routers may change the value). By an
6125 oversight, this default was not being set up when an address was tested by
6126 -bt or -bv, which affected the outcome if any router or filter referred to
6129 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
6130 log only when "message" or "log_message" are set. However, if one of the
6131 conditions was an address verification, or a call to a nested ACL, the
6132 messages generated by the underlying test were being passed through. This
6133 no longer happens. The underlying message is available in $acl_verify_
6134 message for both "message" and "log_message" expansions, so it can be
6135 passed through if needed.
6137 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
6138 new expansion $bh_ to give the encoded byte string without charset
6139 translation. Translation happens only if iconv() is available; HAVE_ICONV
6140 indicates this at build time. HEADERS_CHARSET gives the charset to
6141 translate to; headers_charset can change it in the configuration, and
6142 "headers charset" can change it in an individual filter file.
6144 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
6145 that creates RFC 2047 encoded "words" labels them as that charset instead
6146 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
6147 expansion operator; (ii) when Exim creates a From: line for a local
6148 message; (iii) when a header line is rewritten to include a "phrase" part.
6150 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
6151 buggy, causing it to skip the first lines of messages whose message ID
6152 ended in 'D'. This would not have bitten before Exim release 4.14, because
6153 message IDs were unlikely to end in 'D' before then. The effect was to have
6154 incorrect size information for certain domains.
6156 45. #include "config.h" was missing at the start of the crypt16.c module. This
6157 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
6160 46. If there was a timeout during a "random" callout check, Exim treated it as
6161 a failure of the random address, and carried on sending RSET and the real
6162 address. If the delay was just some slowness somewhere, the response to the
6163 original RCPT would be taken as a response to RSET and so on, causing
6164 mayhem of various kinds.
6166 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
6167 when I implemented it. It didn't allow for the fact that some option values
6168 may legitimatetly be negative (e.g. size_addition), and it didn't even do
6169 the right test for positive values.
6171 48. Domain names in DNS records are case-independent. Exim always looks them up
6172 in lower case. Some resolvers return domain names in exactly the case they
6173 appear in the zone file, that is, they may contain uppercase letters. Not
6174 all resolvers do this - some return always lower case. Exim was treating a
6175 change of case by a resolver as a change of domain, similar to a widening
6176 of a domain abbreviation. This triggered its re-routing code and so it was
6177 trying to route what was effectively the same domain again. This normally
6178 caused routing to fail (because the router wouldn't handle the domain
6179 twice). Now Exim checks for this case specially, and just changes the
6180 casing of the domain that it ultimately uses when it transmits the message
6183 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
6186 50. If a filter generated a file delivery with a non-absolute name (possible if
6187 no home directory exists for the router), the forbid_file option was not
6190 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
6191 the existing equality matching.
6193 52. Exim was using ints instead of ino_t variables in some places where it was
6194 dealing with inode numbers.
6196 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
6197 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
6198 called TMPDIR, and if it finds it is different, it changes its value.
6200 54. The smtp_printf() function is now made available to local_scan() so
6201 additional output lines can be written before returning. There is also an
6202 smtp_fflush() function to enable the detection of a dropped connection.
6203 The variables smtp_input and smtp_batched_input are exported to
6206 55. Changed the default runtime configuration: the message "Unknown user"
6207 has been removed from the ACL, and instead placed on the localuser router,
6208 using the cannot_route_message feature. This means that any verification
6209 failures that generate their own messages won't get overridden. Similarly,
6210 the "Unrouteable address" message that was in the ACL for unverifiable
6211 relay addresses has also been removed.
6213 56. Added hosts_avoid_esmtp to the smtp transport.
6215 57. The exicyclog script was not checking for the esoteric option
6216 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
6217 will work only if exicyclog is run under the appropriate euid.
6219 58. Following a discussion on the list, the rules by which Exim recognises line
6220 endings on incoming messages have been changed. The -dropcr and drop_cr
6221 options are now no-ops, retained only for backwards compatibility. The
6222 following line terminators are recognized: LF CRLF CR. However, special
6223 processing applies to CR:
6225 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
6226 nor a local message in the state where . is a terminator.
6228 (ii) If a bare CR is encountered in a header line, an extra space is added
6229 after the line terminator so as not to end the header. The reasoning
6230 behind this is that bare CRs in header lines are most likely either
6231 to be mistakes, or people trying to play silly games.
6233 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
6234 was being incorrectly given as 18 bytes larger than it should have been.
6235 This is a VOB (very old bug).
6237 60. This may never have affected anything current, but just in case it has:
6238 When the local host is found other than at the start of a list of hosts,
6239 the local host, those with the same MX, and any that follow, are discarded.
6240 When the list in question was part of a longer list of hosts, the following
6241 hosts (not currently being processed) were also being discarded. This no
6242 longer happens. I'm not sure if this situation could ever has previously
6245 61. Added the "/MX" feature to lists of hosts in the manualroute and query
6248 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
6249 header. This is something that is recommended in a new Internet Draft, and
6250 is something that is documented as being done by Sendmail. There are two
6251 possible values. For messages generated by the autoreply transport, Exim
6254 Auto-Submitted: auto-replied
6256 whereas for all other generated messages (e.g. bounces) it adds
6258 Auto-Submitted: auto-generated
6260 63. The "personal" condition in filters now includes a test for the
6261 Auto-Submitted: header. If it contains the string "auto-" the message it
6262 not considered personal.
6264 64. Added rcpt_include_affixes as a generic transport option.
6266 65. Added queue_only_override (default true).
6268 66. Added the syslog_duplication option.
6270 67. If what should have been the first header line of a message consisted of
6271 a space followed by a colon, Exim was mis-interpreting it as a header line.
6272 It isn't of course - it is syntactically invalid and should therefore be
6273 treated as the start of the message body. The misbehaviour could have
6274 caused a number of strange effects, including loss of data in subsequent
6275 header lines, and spool format errors.
6277 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
6278 client host had authenticated. This control can now be exercised by an ACL
6279 for more flexibility.
6281 69. By default, callouts do not happen when testing with -bh. There is now a
6282 variant, -bhc, which does actually run the callout code, including
6283 consulting and updating the callout cache.
6285 70. Added support for saslauthd authentication, courtesy of Alexander
6288 71. If statvfs() failed on the spool or log directories while checking their
6289 size for availability, Exim confusingly gave the error "space shortage".
6290 Furthermore, in debugging mode it crashed with a floating point exception.
6291 These checks are done if check_{spool,log}_{space,inodes} are set, and when
6292 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
6293 serious problem, Exim now writes to the main and panic logs when this
6294 happens, with details of the failure. It then refuses to accept the
6295 incoming message, giving the message "spool directory problem" or "log
6296 directory problem" with a 421 code for SMTP messages.
6298 72. When Exim is about to re-exec itself, it ensures that the file descriptors
6299 0, 1, and 2 exist, because some OS complain for execs without them (see
6300 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
6301 descriptors. However, the code omitted to check that the open succeeded,
6302 causing mysterious errors if for some reason the permissions on /dev/null
6303 got screwed. Now Exim writes a message to the main and panic logs, and
6304 bombs out if it can't open /dev/null.
6306 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
6307 interact so that it is all more flexible. It is supposed to remain
6308 backwards compatible. Also added extra_local_interfaces.
6310 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
6311 to bomb out with an assertion failure - to the client this appears as a
6312 connection drop. This problem occurs in the part of the code that was taken
6313 from the Samba project. Fortunately, the assertion is in a very simple
6314 function, so I have fixed this by reproducing the function inline in the
6315 one place where it is called, and arranging for authentication to fail
6316 instead of killing the process with assert().
6318 75. The SPA client code was not working when the server requested OEM rather
6319 than Unicode encoding.
6321 76. Added code to make require_files with a specific uid setting more usable in
6322 the case where statting the file as root fails - usually a non-root-mounted
6323 NFS file system. When this happens and the failure is EACCES, Exim now
6324 forks a subprocess and does the per-uid checking as the relevant uid.
6326 77. Added process_log_path.
6328 78. If log_file_path was not explicitly set, a setting of check_log_space or
6329 check_log_inodes was ignored.
6331 79. If a space check for the spool or log partitions fails, the incident is now
6332 logged. Of course, in the latter case the data may get lost...
6334 80. Added the %p formatting code to string_format() so that it can be used to
6335 print addresses in debug_print(). Adjusted all the address printing in the
6336 debugging in store.c to use %p rather than %d.
6338 81. There was a concern that a line of code in smtp_in.c could overflow a
6339 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
6340 initially expressed, the concern was not well-founded, because trailing
6341 spaces are removed early. However, if the trailing spaces were followed by
6342 a NULL, they did not get removed, so the overflow was possible. Two fixes
6345 (a) I re-wrote the offending code in a cleaner fashion.
6346 (b) If an incoming SMTP command contains a NULL character, it is rejected
6349 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
6350 runs initgroups(), so that if the Exim user is in any additional groups,
6351 they will be used during message reception.
6357 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / users / jgh / exim.git / blob