1 Change log file for Exim from version 4.21
2 -------------------------------------------
7 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
8 and 3.4.4 - once the server is enabled to respond to an OCSP request
9 it does even when not requested, resulting in a stapling non-aware
10 client dropping the TLS connection.
12 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
13 support variable-length bit vectors. No functional change.
15 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
16 The I= interface field on outgoing lines is now after the H= remote
17 host field, same as incoming lines. There is a separate
18 outgoing_interface log selector which allows you to disable the
21 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
22 If not running log_selector +smtp_connection the mainlog would be held
23 open indefinitely after a "too many connections" event, including to a
24 deleted file after a log rotate. Leave the per net connection logging
25 leaving it open for efficiency as that will be quickly detected by the
26 check on the next write.
28 HS/01 Bug 1671: Fix post transport crash.
29 Processing the wait-<transport> messages could crash the delivery
30 process if the message IDs didn't exist for some reason. When
31 using 'split_spool_directory=yes' the construction of the spool
32 file name failed already, exposing the same netto behaviour.
34 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
35 mime_regex ACL conditions.
37 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
38 to DSN fail messages (bounces): remote IP, remote greeting, remote response
39 to HELO, local diagnostic string.
41 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
42 log line to debug. Even when configured with a tls authenticator many
43 client connections are expected to not authenticate in this way, so
44 an authenticate fail is not an error.
46 HS/02 Add the Exim version string to the process info. This way exiwhat
47 gives some more detail about the running daemon.
49 JH/06 Bug 1395: time-limit cacheing of DNS lookups, to the TTL value. This may
50 matter for fast-change records such as DNSBLs.
52 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
53 retry record, even if constant. There may be multiple transports with
54 different interface settings and the retry behaviour needs to be kept
57 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
59 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
61 JH/10 Bug 840: fix log_defer_output option of pipe transport
63 JH/11 Bug 830: use same host for all RCPTS of a message, even under
64 hosts_randomize. This matters a lot when combined with mua_wrapper.
66 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
67 ${quote_pgsql:<string>} operator.
69 JH/13 Bug 1708: avoid misaligned access in cached lookup.
71 JH/14 Change header file name for freeradius-client. Relevant if compiling
72 with Radius support; from the Gentoo tree and checked under Fedora.
74 JH/15 Bug 1712: Introduce $prdr_requested flag variable
76 JH/16 Bug 1714: Permit an empty string as expansion result for transport
77 option transport_filter, meaning no filtering.
79 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
81 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
82 defaults to "*" (all hosts). The variable is now available when not built
83 with TLS, default unset, mainly to enable keeping the testuite sane.
84 If a server certificate is not supplied (via tls_certificate) an error is
85 logged, and clients will find TLS connections fail on startup. Presumably
86 they will retry in-clear.
87 Packagers of Exim are strongly encouraged to create a server certificate
90 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
91 with the $config_file variable.
93 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
94 in transport context, after the attempt, and per-recipient. The latter type
95 is per host attempted. The event data is the error message, and the errno
96 information encodes the lookup type (A vs. MX) used for the (first) host,
97 and the trailing two digits of the smtp 4xx reponse.
99 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
100 to write to mainlog (or rejectlog, paniclog) in the window between file
101 creation and permissions/ownership being changed. Particularly affects
102 installations where exicyclog is run as root, rather than exim user;
103 result is that the running daemon panics and dies.
105 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
107 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
108 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
109 "pri" and "weight". Note that the previous implicit priority given by the
110 list order is no longer honoured.
112 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalisation
115 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
116 by defining SUPPORT_SOCKS.
118 JH/26 Move PROXY support from Experimental to mainline, enabled for a build
119 by defining SUPPORT_PROXY. Note that the proxy_required_hosts option
120 is renamed to hosts_proxy, and the proxy_{host,target}_{address,port}.
121 variables are renamed to proxy_{local,external}_{address,port}.
123 JH/27 Move Internationalisation support from Experimental to mainline, enabled
124 for a build by defining SUPPORT_I18N
126 JH/28 Bug 1745: Fix redis lookups to handle (quoted) spaces embedded in parts
127 of the query string, and make ${quote_redis:} do that quoting.
129 JH/29 Move Events support from Experimental to mainline, enabled by default
130 and removable for a build by defining DISABLE_EVENT.
135 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
138 JH/02 The smtp transport option "multi_domain" is now expanded.
140 JH/03 The smtp transport now requests PRDR by default, if the server offers
143 JH/04 Certificate name checking on server certificates, when exim is a client,
144 is now done by default. The transport option tls_verify_cert_hostnames
145 can be used to disable this per-host. The build option
146 EXPERIMENTAL_CERTNAMES is withdrawn.
148 JH/05 The value of the tls_verify_certificates smtp transport and main options
149 default to the word "system" to access the system default CA bundle.
150 For GnuTLS, only version 3.0.20 or later.
152 JH/06 Verification of the server certificate for a TLS connection is now tried
153 (but not required) by default. The verification status is now logged by
154 default, for both outbound TLS and client-certificate supplying inbound
157 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
160 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
161 Status Notification (bounce) messages are now MIME format per RFC 3464.
162 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
163 under the control of the dsn_advertise_hosts option, and routers may
164 have a dsn_lasthop option.
166 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
167 default, modifiable by a malware= option. The list separator for
168 the options can now be changed in the usual way. Bug 68.
170 JH/10 The smtp_receive_timeout main option is now expanded before use.
172 JH/11 The incoming_interface log option now also enables logging of the
173 local interface on delivery outgoing connections.
175 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
176 if the interface and destination host and port all match.
178 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
181 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
182 Patch from Andrew Lewis.
184 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
185 now supports optional time-restrictions, weighting, and priority
186 modifiers per server. Patch originally by <rommer@active.by>.
188 JH/16 The spamd_address main option now supports a mixed list of local
189 and remote servers. Remote servers can be IPv6 addresses, and
190 specify a port-range.
192 JH/17 Bug 68: The spamd_address main option now supports an optional
193 timeout value per server.
195 JH/18 Bug 1581: Router and transport options headers_add/remove can
196 now have the list separator specified.
198 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
201 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
204 JH/21 Support for the A6 type of dns record is withdrawn.
206 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
207 rather than the verbs used.
209 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
210 from 255 to 1024 chars.
212 JH/24 Verification callouts now attempt to use TLS by default.
214 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
215 are generic router options now. The defaults didn't change.
217 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
218 Original patch from Alexander Shikoff, worked over by JH.
220 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
221 files if ZCAT_COMMAND is not executable.
223 JH/26 Bug 1539: Add timout/retry options on dnsdb lookups.
225 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
227 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
228 Normally benign, it bites when the pair was led to by a CNAME;
229 modern usage is to not canoicalize the domain to a CNAME target
230 (and we were inconsistent anyway for A-only vs AAAA+A).
232 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
234 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
235 when evaluating $sender_host_dnssec.
237 JH/31 Check the HELO verification lookup for DNSSEC, adding new
238 $sender_helo_dnssec variable.
240 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
242 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
244 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
246 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
247 documented as working, but never had. Support all but $spam_report.
249 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
250 added for tls authenticator.
255 TL/01 When running the test suite, the README says that variables such as
256 no_msglog_check are global and can be placed anywhere in a specific
257 test's script, however it was observed that placement needed to be near
258 the beginning for it to behave that way. Changed the runtest perl
259 script to read through the entire script once to detect and set these
260 variables, reset to the beginning of the script, and then run through
261 the script parsing/test process like normal.
263 TL/02 The BSD's have an arc4random API. One of the functions to induce
264 adding randomness was arc4random_stir(), but it has been removed in
265 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
266 function when detected.
268 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
269 cause callback expansion.
271 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
272 syntax errors in an expansion can be treated as a string instead of
273 logging or causing an error, due to the internal use of bool_lax
274 instead of bool when processing it.
276 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
277 server certificates when making smtp deliveries.
279 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
281 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
283 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
285 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
286 Merged patch from Sebastian Wiedenroth.
288 JH/05 Fix results-pipe from transport process. Several recipients, combined
289 with certificate use, exposed issues where response data items split
290 over buffer boundaries were not parsed properly. This eventually
291 resulted in duplicates being sent. This issue only became common enough
292 to notice due to the introduction of conection certificate information,
293 the item size being so much larger. Found and fixed by Wolfgang Breyha.
295 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
296 size buffer was used, resulting in syntax errors when an expansion
299 JH/07 Add support for directories of certificates when compiled with a GnuTLS
300 version 3.3.6 or later.
302 JH/08 Rename the TPDA expermimental facility to Event Actions. The #ifdef
303 is EXPERIMENTAL_EVENT, the main-configuration and transport options
304 both become "event_action", the variables become $event_name, $event_data
305 and $event_defer_errno. There is a new variable $verify_mode, usable in
306 routers, transports and related events. The tls:cert event is now also
307 raised for inbound connections, if the main configuration event_action
310 TL/06 In test suite, disable OCSP for old versions of openssl which contained
311 early OCSP support, but no stapling (appears to be less than 1.0.0).
313 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
314 server certificate names available under the smtp transport option
315 "tls_verify_cert_hostname" now do not permit multi-component wildcard
318 JH/10 Time-related extraction expansions from certificates now use the main
319 option "timezone" setting for output formatting, and are consistent
320 between OpenSSL and GnuTLS compilations. Bug 1541.
322 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
323 encoded parameter in the incoming message. Bug 1558.
325 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
326 include certificate info, eximon was claiming there were spoolfile
329 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
331 JH/14 Log delivery-related information more consistently, using the sequence
332 "H=<name> [<ip>]" wherever possible.
334 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
335 are problematic for Debian distribution, omit them from the release
338 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
340 JH/16 Fix string representation of time values on 64bit time_t anchitectures.
343 JH/17 Fix a null-indirection in certextract expansions when a nondefault
344 output list separator was used.
349 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
350 checkers that were complaining about end of non-void function with no
353 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
354 This was a regression intruduced in 4.83 by another bugfix.
356 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
358 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
359 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
365 TF/01 Correctly close the server side of TLS when forking for delivery.
367 When a message was received over SMTP with TLS, Exim failed to clear up
368 the incoming connection properly after forking off the child process to
369 deliver the message. In some situations the subsequent outgoing
370 delivery connection happened to have the same fd number as the incoming
371 connection previously had. Exim would try to use TLS and fail, logging
372 a "Bad file descriptor" error.
374 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
375 utilities have not been installed.
377 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
378 temporary space as the ACL may create new global variables.
380 TL/01 LDAP support uses per connection or global context settings, depending
381 upon the detected version of the libraries at build time.
383 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
384 to extract and use the src ip:port in logging and expansions as if it
385 were a direct connection from the outside internet. PPv2 support was
386 updated based on HAProxy spec change in May 2014.
388 JH/02 Add ${listextract {number}{list}{success}{fail}}.
390 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
391 Properly escape header and check for NULL return.
393 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
396 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
398 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
399 characters in header names, implemented as a verify condition.
400 Contributed by Michael Fischer v. Mollard.
402 TL/05 Rename SPF condition results err_perm and err_temp to standardized
403 results permerror and temperror. Previous values are deprecated but
404 still accepted. In a future release, err_perm and err_temp will be
405 completely removed, which will be a backward incompatibility if the
406 ACL tests for either of these two old results. Patch contributed by
407 user bes-internal on the mailing list.
409 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
411 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
412 selectors, in both main and reject logs.
414 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
417 JH/07 Add malware type "sock" for talking to simple daemon.
419 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
421 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
422 routers/transports under cutthrough routing.
424 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
425 numbers. Touch up "bool" conditional to keep the same definition.
427 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
429 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
431 JH/12 Expand items in router/transport headers_add or headers_remove lists
432 individually rather than the list as a whole. Bug 1452.
434 Required for reasonable handling of multiple headers_ options when
435 they may be empty; requires that headers_remove items with embedded
436 colons must have them doubled (or the list-separator changed).
438 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
439 view the policy declared in the DMARC record. Currently, $dmarc_status
440 is a combined value of both the record presence and the result of the
443 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
445 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
446 dnslookup router and the smtp transport (applying to the forward
449 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
450 of ldap servers used for a specific lookup. Patch provided by Heiko
453 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
454 New variable $lookup_dnssec_authenticated for observability.
456 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
457 Patch submitted by Lars Timman.
459 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
461 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
462 Requires trusted mode and valid format message id, aborts otherwise.
463 Patch contributed by Heiko Schlichting.
465 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
466 certextract with support for various fields. Bug 1358.
468 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
469 is requested by default, modifiable by smtp transport option
472 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
473 operate on certificate variables to give certificate fingerprints
474 Also new ${sha256:cert_variable}.
476 JH/23 The PRDR feature is moved from being Experimental into the mainline.
478 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
479 Christian Aistleitner.
481 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
483 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
484 file. Patch from Wolfgang Breyha.
486 JH/25 Expand the coverage of the delivery $host and $host_address to
487 client authenticators run in verify callout. Bug 1476.
489 JH/26 Port service names are now accepted for tls_on_connect_ports, to
490 align with daemon_smtp_ports. Bug 72.
492 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
493 support and error reporting did not work properly.
495 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
496 and is readable. Patch from Andrew Colin Kissa.
498 TL/14 Enhance documentation of ${run expansion and how it parses the
499 commandline after expansion, particularly in the case when an
500 unquoted variable expansion results in an empty value.
502 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
504 PP/02 Fix internal collision of T_APL on systems which support RFC3123
505 by renaming away from it. Addresses GH issue 15, reported by
508 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
510 TL/15 SECURITY: prevent double expansion in math comparison functions
511 (can expand unsanitized data). Not remotely exploitable.
518 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
520 PP/02 Make -n do something, by making it not do something.
521 When combined with -bP, the name of an option is not output.
523 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
526 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
527 $sender_host_name and config options to manage this, and basic check
530 PP/05 DSCP support for outbound connections and control modifier for inbound.
532 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
533 (Only plugin which currently uses this is kerberos4, which nobody should
534 be using, but we should make it available and other future plugins might
535 conceivably use it, even though it would break NAT; stuff *should* be
536 using channel bindings instead).
538 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
539 name; added for Sendmail compatibility; requires admin caller.
540 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
541 just ignore it); requires trusted caller.
542 Also parse but ignore: -Ac -Am -X<logfile>
545 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
547 TL/02 Add +smtp_confirmation as a default logging option.
549 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
550 Patch by Magnus Holmgren from 2007-02-20.
552 TL/04 Bugzilla 1281 - Spec typo.
553 Bugzilla 1283 - Spec typo.
554 Bugzilla 1290 - Spec grammar fixes.
556 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
558 TL/06 Add Experimental DMARC support using libopendmarc libraries.
560 TL/07 Fix an out of order global option causing a segfault. Reported to dev
561 mailing list by by Dmitry Isaikin.
563 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
565 JH/02 Support "G" suffix to numbers in ${if comparisons.
567 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
569 NM/01 Bugzilla 1197 - Spec typo
570 Bugzilla 1196 - Spec examples corrections
572 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
574 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
575 gnutls_enable_pkcs11, but renamed to more accurately indicate its
578 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
579 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
581 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
582 "acl {{name}{arg}...}", and optional args on acl condition
585 JH/05 Permit multiple router/transport headers_add/remove lines.
587 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
589 JH/07 Avoid using a waiting database for a single-message-only transport.
590 Performance patch from Paul Fisher. Bugzilla 1262.
592 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
595 JH/09 Add $headers_added variable, with content from use of ACL modifier
596 add_header (but not yet added to the message). Bugzilla 199.
598 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
599 Pulled from Bugzilla 817 by Wolfgang Breyha.
601 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
603 (nb: this is the same fix as in Exim 4.80.1)
605 JH/11 Add A= logging on delivery lines, and a client_set_id option on
608 JH/12 Add optional authenticated_sender logging to A= and a log_selector
611 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
613 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
614 advertise SMTP AUTH mechanism to us, instead of a generic
615 protocol violation error. Also, make Exim more robust to bad
616 data from the Dovecot auth socket.
618 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
620 When a queue runner is handling a message, Exim first routes the
621 recipient addresses, during which it prunes them based on the retry
622 hints database. After that it attempts to deliver the message to
623 any remaining recipients. It then updates the hints database using
626 So if a recipient address works intermittently, it can get repeatedly
627 deferred at routing time. The retry hints record remains fresh so the
628 address never reaches the final cutoff time.
630 This is a fairly common occurrence when a user is bumping up against
631 their storage quota. Exim had some logic in its local delivery code
632 to deal with this. However it did not apply to per-recipient defers
633 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
635 This change adds a proper retry rule check during routing so that the
636 final cutoff time is checked against the message's age. We only do
637 this check if there is an address retry record and there is not a
638 domain retry record; this implies that previous attempts to handle
639 the address had the retry_use_local_parts option turned on. We use
640 this as an approximation for the destination being like a local
641 delivery, as in LMTP.
643 I suspect this new check makes the old local delivery cutoff check
644 redundant, but I have not verified this so I left the code in place.
646 TF/02 Correct gecos expansion when From: is a prefix of the username.
648 Test 0254 submits a message to Exim with the header
652 When I ran the test suite under the user fanf2, Exim expanded
653 the header to contain my full name, whereas it should have added
654 a Resent-Sender: header. It erroneously treats any prefix of the
655 username as equal to the username.
657 This change corrects that bug.
659 GF/01 DCC debug and logging tidyup
660 Error conditions log to paniclog rather than rejectlog.
661 Debug lines prefixed by "DCC: " to remove any ambiguity.
663 TF/03 Avoid unnecessary rebuilds of lookup-related code.
665 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
666 Bug spotted by Jeremy Harris; was flawed since initial commit.
667 Would have resulted in OCSP responses post-SNI triggering an Exim
668 NULL dereference and crash.
670 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
672 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
673 Bug detection, analysis and fix by Samuel Thibault.
674 Bugzilla 1331, Debian bug #698092.
676 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
678 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
679 Server implementation by Todd Lyons, client by JH.
680 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
681 config variable "prdr_enable" controls whether the server
682 advertises the facility. If the client requests PRDR a new
683 acl_data_smtp_prdr ACL is called once for each recipient, after
684 the body content is received and before the acl_smtp_data ACL.
685 The client is controlled by bolth of: a hosts_try_prdr option
686 on the smtp transport, and the server advertisement.
687 Default client logging of deliveries and rejections involving
688 PRDR are flagged with the string "PRDR".
690 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
691 fclose(). Diagnosis by Todd Lyons.
693 PP/17 Update configure.default to handle IPv6 localhost better.
694 Patch by Alain Williams (plus minor tweaks).
697 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
698 This is now consistent with GnuTLS, and is now documented: the
699 previous undocumented portable approach to treating the option as
700 unset was to force an expansion failure. That still works, and
701 an empty string is now equivalent.
703 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
704 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
705 not performing validation itself.
707 PP/20 Added force_command boolean option to pipe transport.
708 Patch from Nick Koston, of cPanel Inc.
710 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
713 TF/04 Added udpsend ACL modifer and hexquote expansion operator
715 PP/21 Fix eximon continuous updating with timestamped log-files.
716 Broken in a format-string cleanup in 4.80, missed when I repaired the
717 other false fix of the same issue.
718 Report and fix from Heiko Schlichting.
721 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
722 Report from Prashanth Katuri.
724 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
725 It's SecureTransport, so affects any MacOS clients which use the
726 system-integrated TLS libraries, including email clients.
728 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
729 using a MIME ACL for non-SMTP local injection.
730 Report and assistance in diagnosis by Warren Baker.
732 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
734 JH/16 Fix comparisons for 64b. Bugzilla 1385.
736 TL/09 Add expansion variable $authenticated_fail_id to keep track of
737 last id that failed so it may be referenced in subsequent ACL's.
739 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
742 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
743 ldap library initialization, allowing self-signed CA's to be
744 used. Also properly sets require_cert option later in code by
745 using NULL (global ldap config) instead of ldap handle (per
746 session). Bug diagnosis and testing by alxgomz.
748 TL/12 Enhanced documentation in the ratelimit.pl script provided in
749 the src/util/ subdirectory.
751 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
752 renamed to Transport Post Delivery Action by Jeremy Harris, as
755 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
756 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
757 redis_servers = needs to be configured which will be used by the redis
758 lookup. Patch from Warren Baker, of The Packet Hub.
760 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
762 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
763 hostname or reverse DNS when processing a host list. Used suggestions
764 from multiple comments on this bug.
766 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
768 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
769 Missed a few lines, added it to make the runtest require no keyboard
772 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
773 contains upper case chars. Make router use caseful_local_part.
775 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
776 support when GnuTLS has been built with p11-kit.
782 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
784 This, or similar/improved, will also be change PP/11 of 4.82.
790 PP/01 Handle short writes when writing local log-files.
791 In practice, only affects FreeBSD (8 onwards).
792 Bugzilla 1053, with thanks to Dmitry Isaikin.
794 NM/01 Bugzilla 949 - Documentation tweak
796 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
799 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
801 PP/02 Implemented gsasl authenticator.
803 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
805 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
806 `pkg-config foo` for cflags/libs.
808 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
809 with rest of GSASL and with heimdal_gssapi.
811 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
812 `pkg-config foo` for cflags/libs for the TLS implementation.
814 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
815 properties get this fed in as external SSF. A number of robustness
816 and debugging improvements to the cyrus_sasl authenticator.
818 PP/08 cyrus_sasl server now expands the server_realm option.
820 PP/09 Bugzilla 1214 - Log authentication information in reject log.
821 Patch by Jeremy Harris.
823 PP/10 Added dbmjz lookup type.
825 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
827 PP/12 MAIL args handles TAB as well as SP, for better interop with
828 non-compliant senders.
829 Analysis and variant patch by Todd Lyons.
831 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
832 Bug report from Lars Müller <lars@samba.org> (via SUSE),
833 Patch from Dirk Mueller <dmueller@suse.com>
835 PP/13 tls_peerdn now print-escaped for spool files.
836 Observed some $tls_peerdn in wild which contained \n, which resulted
837 in spool file corruption.
839 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
840 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
841 or write after TLS renegotiation, which otherwise led to messages
844 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
845 as a tracking header (ie: a signed header comes before the signature).
846 Patch from Wolfgang Breyha.
848 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
849 comma-sep list; embedded commas doubled.
851 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
853 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
855 Report and patch from Dmitry Banschikov.
857 PP/16 Removed "dont_insert_empty_fragments" fron "openssl_options".
858 Removed SSL_clear() after SSL_new() which led to protocol negotiation
859 failures. We appear to now support TLS1.1+ with Exim.
861 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
862 lets Exim select keys and certificates based upon TLS SNI from client.
863 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
864 before an outbound SMTP session. New log_selector, +tls_sni.
866 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
867 NULL dereference. Report and patch from Alun Jones.
869 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
870 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
871 Not seeing resolver debug output on NetBSD, but suspect this is a
872 resolver implementation change.
874 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
875 Left warnings. Added "eximon gdb" invocation mode.
877 PP/21 Defaulting "accept_8bitmime" to true, not false.
879 PP/22 Added -bw for inetd wait mode support.
881 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
882 locate the relevant includes and libraries. Made this the default.
884 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
885 Bugzilla 1246, report and most of solution from Tomasz Kusy.
887 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
888 This may cause build issues on older platforms.
890 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
891 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
892 gnutls_require_mac & gnutls_require_protocols (no longer supported).
893 Added SNI support via GnuTLS too.
894 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
896 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
898 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
899 Applied second patch from Janne, implementing suggestion to default
900 multiple-strings-in-record handling to match SPF spec.
902 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
904 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
905 read-only, out of scope).
906 Patch from Wolfgang Breyha, report from Stuart Northfield.
908 PP/29 Fix three issues highlighted by clang analyser static analysis.
909 Only crash-plausible issue would require the Cambridge-specific
910 iplookup router and a misconfiguration.
911 Report from Marcin Mirosław.
913 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
915 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
916 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
917 As part of this, removing so much warning spew let me fix some minor
918 real issues in debug logging.
920 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
921 assignment on my part. Fixed.
923 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
924 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
925 Janne Snabb (who went above and beyond: thank you).
927 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
928 string otherwise requires a connection and a bunch more work and it's
929 relatively easy to get wrong. Should also expose TLS library linkage
932 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
933 64-bit ${eval} (JH/03).
935 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
936 GNU libc to support some of the 64-bit stuff, should not lead to
937 conflicts. Defined before os.h is pulled in, so if a given platform
938 needs to override this, it can.
940 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
941 protection layer was required, which is not implemented.
942 Bugzilla 1254, patch from Wolfgang Breyha.
944 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
945 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
946 tls_dhparam take prime identifiers. Also unbreak combination of
947 OpenSSL+DH_params+TLSSNI.
949 PP/39 Disable SSLv2 by default in OpenSSL support.
955 PP/01 Solaris build fix for Oracle's LDAP libraries.
956 Bugzilla 1109, patch from Stephen Usher.
958 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
960 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
963 TF/02 Fix a couple more cases where we did not log the error message
964 when unlink() failed. See also change 4.74-TF/03.
966 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
967 lock up or crash if it happened to be inside a call to libc when it
968 got a SIGUSR1 from exiwhat.
970 The SIGUSR1 handler appends the current process status to the process
971 log which is later printed by exiwhat. It used to use the general
972 purpose logging code to do this, but several functions it calls are
973 not safe for signals.
975 The new output code in the SIGUSR1 handler is specific to the process
976 log, and simple enough that it's easy to inspect for signal safety.
977 Removing some special cases also simplifies the general logging code.
978 Removing the spurious timestamps from the process log simplifies
981 TF/04 Improved ratelimit ACL condition.
983 The /noupdate option has been deprecated in favour of /readonly which
984 has clearer semantics. The /leaky, /strict, and /readonly update modes
985 are mutually exclusive. The update mode is no longer included in the
986 database key; it just determines when the database is updated. (This
987 means that when you upgrde Exim will forget old rate measurements.)
989 Exim now checks that the per_* options are used with an update mode that
990 makes sense for the current ACL. For example, when Exim is processing a
991 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
992 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
993 must specify per_mail/readonly. If you omit the update mode it defaults to
994 /leaky where that makes sense (as before) or /readonly where required.
996 The /noupdate option is now undocumented but still supported for
997 backwards compatibility. It is equivalent to /readonly except that in
998 ACLs where /readonly is required you may specify /leaky/noupdate or
999 /strict/noupdate which are treated the same as /readonly.
1001 A useful new feature is the /count= option. This is a generalization
1002 of the per_byte option, so that you can measure the throughput of other
1003 aggregate values. For example, the per_byte option is now equivalent
1004 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
1006 The per_rcpt option has been generalized using the /count= mechanism
1007 (though it's more complicated than the per_byte equivalence). When it is
1008 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
1009 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
1010 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
1011 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
1012 Note that using per_rcpt with a non-readonly update mode in more than
1013 one ACL will cause the recipients to be double-counted. (The per_mail
1014 and per_byte options don't have this problem.)
1016 The handling of very low rates has changed slightly. If the computed rate
1017 is less than the event's count (usually one) then this event is the first
1018 after a long gap. In this case the rate is set to the same as this event's
1019 count, so that the first message of a spam run is counted properly.
1021 The major new feature is a mechanism for counting the rate of unique
1022 events. The new per_addr option counts the number of different
1023 recipients that someone has sent messages to in the last time period. It
1024 behaves like per_rcpt if all the recipient addresses are different, but
1025 duplicate recipient addresses do not increase the measured rate. Like
1026 the /count= option this is a general mechanism, so the per_addr option
1027 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
1028 example, measure the rate that a client uses different sender addresses
1029 with the options per_mail/unique=$sender_address. There are further
1030 details in the main documentation.
1032 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
1034 TF/06 Removed a few PCRE remnants.
1036 TF/07 Automatically extract Exim's version number from tags in the git
1037 repository when doing development or release builds.
1039 PP/02 Raise smtp_cmd_buffer_size to 16kB.
1040 Bugzilla 879. Patch from Paul Fisher.
1042 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
1043 Heavily based on revision 40f9a89a from Simon Arlott's tree.
1046 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
1048 PP/05 Variable $av_failed, true if the AV scanner deferred.
1049 Bugzilla 1078. Patch from John Horne.
1051 PP/06 Stop make process more reliably on build failure.
1052 Bugzilla 1087. Patch from Heiko Schlittermann.
1054 PP/07 Make maildir_use_size_file an _expandable_ boolean.
1055 Bugzilla 1089. Patch from Heiko Schlittermann.
1057 PP/08 Handle ${run} returning more data than OS pipe buffer size.
1058 Bugzilla 1131. Patch from Holger Weiß.
1060 PP/09 Handle IPv6 addresses with SPF.
1061 Bugzilla 860. Patch from Wolfgang Breyha.
1063 PP/10 GnuTLS: support TLS 1.2 & 1.1.
1065 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
1068 PP/11 match_* no longer expand right-hand-side by default.
1069 New compile-time build option, EXPAND_LISTMATCH_RHS.
1070 New expansion conditions, "inlist", "inlisti".
1072 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
1074 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
1076 PP/14 fix log_write() format string regression from TF/03.
1077 Bugzilla 1152. Patch from Dmitry Isaikin.
1083 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
1085 PP/02 Harmonised TLS library version reporting; only show if debugging.
1086 Layout now matches that introduced for other libraries in 4.74 PP/03.
1088 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
1090 PP/04 New "dns_use_edns0" global option.
1092 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
1095 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
1096 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
1098 TK/01 Updated PolarSSL code to 0.14.2.
1099 Bugzilla 1097. Patch from Andreas Metzler.
1101 PP/07 Catch divide-by-zero in ${eval:...}.
1102 Fixes bugzilla 1102.
1104 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
1107 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
1108 format-string attack -- SECURITY: remote arbitrary code execution.
1110 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
1111 time unintentionally subject to list matching rules, letting the header
1112 cause arbitrary Exim lookups (of items which can occur in lists, *not*
1113 arbitrary string expansion). This allowed for information disclosure.
1115 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
1116 INT_MIN/-1 -- value coerced to INT_MAX.
1122 NM/01 Workround for PCRE version dependency in version reporting
1125 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
1126 This fixes portability to compilers other than gcc, notably
1127 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
1129 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
1130 makefiles for portability to HP-UX and POSIX correctness.
1132 PP/01 Permit LOOKUP_foo enabling on the make command-line.
1133 Also via indented variable definition in the Makefile.
1134 (Debugging by Oliver Heesakkers).
1136 PP/02 Restore caching of spamd results with expanded spamd_address.
1137 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
1139 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
1140 Improves build reliability. Fix from: Frank Elsner
1142 NM/02 Fix wide character breakage in the rfc2047 coding
1143 Fixes bug 1064. Patch from Andrey N. Oktyabrski
1145 NM/03 Allow underscore in dnslist lookups
1146 Fixes bug 1026. Patch from Graeme Fowler
1148 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
1149 Code patches from Adam Ciarcinski of NetBSD.
1151 NM/04 Fixed exiqgrep to cope with mailq missing size issue
1154 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
1155 is logged, to avoid truncation. Patch from John Horne.
1157 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
1158 Patch from Jakob Hirsch.
1160 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
1161 SQL string expansion failure details.
1162 Patch from Andrey Oktyabrski.
1164 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
1165 Patch from Simon Arlott.
1167 PP/09 New lookups functionality failed to compile on old gcc which rejects
1168 extern declarations in function scope.
1169 Patch from Oliver Fleischmann
1171 PP/10 Use sig_atomic_t for flags set from signal handlers.
1172 Check getgroups() return and improve debugging.
1173 Fixed developed for diagnosis in bug 927 (which turned out to be
1176 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
1177 Patch from Mark Zealey.
1179 PP/12 Bugzilla 1056: Improved spamd server selection.
1180 Patch from Mark Zealey.
1182 PP/13 Bugzilla 1086: Deal with maildir quota file races.
1183 Based on patch from Heiko Schlittermann.
1185 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
1186 Patch from Uwe Doering, sign-off by Michael Haardt.
1188 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
1189 variable declaration deep within a block. Bug and patch from
1192 PP/15 lookups-Makefile IRIX compatibilty coercion.
1194 PP/16 Make DISABLE_DKIM build knob functional.
1196 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
1197 Patch by Simon Arlott
1199 TF/03 Fix valgrind.h portability to C89 compilers that do not support
1200 variable argument macros. Our copy now differs from upstream.
1206 TF/01 Failure to get a lock on a hints database can have serious
1207 consequences so log it to the panic log.
1209 TF/02 Log LMTP confirmation messages in the same way as SMTP,
1210 controlled using the smtp_confirmation log selector.
1212 TF/03 Include the error message when we fail to unlink a spool file.
1214 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
1215 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
1216 for maintaining out-of-tree patches for some time.
1218 PP/01 Bugzilla 139: Documentation and portability issues.
1219 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
1220 Handle per-OS dynamic-module compilation flags.
1222 PP/02 Let /dev/null have normal permissions.
1223 The 4.73 fixes were a little too stringent and complained about the
1224 permissions on /dev/null. Exempt it from some checks.
1225 Reported by Andreas M. Kirchwitz.
1227 PP/03 Report version information for many libraries, including
1228 Exim version information for dynamically loaded libraries. Created
1229 version.h, now support a version extension string for distributors
1230 who patch heavily. Dynamic module ABI change.
1232 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
1233 privilege escalation vulnerability whereby the Exim run-time user
1234 can cause root to append content of the attacker's choosing to
1237 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
1240 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
1241 If dropping privileges for untrusted macros, we disabled normal logging
1242 on the basis that it would fail; for the Exim run-time user, this is not
1243 the case, and it resulted in successful deliveries going unlogged.
1244 Fixed. Reported by Andreas Metzler.
1250 PP/01 Date: & Message-Id: revert to normally being appended to a message,
1251 only prepend for the Resent-* case. Fixes regression introduced in
1252 Exim 4.70 by NM/22 for Bugzilla 607.
1254 PP/02 Include check_rfc2047_length in configure.default because we're seeing
1255 increasing numbers of administrators be bitten by this.
1257 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
1259 PP/03 Bugzilla 994: added openssl_options main configuration option.
1261 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
1263 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
1265 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
1267 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
1268 without a peer certificate, leading to a segfault because of an
1269 assumption that peers always have certificates. Be a little more
1270 paranoid. Problem reported by Martin Tscholak.
1272 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
1273 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
1274 NB: ClamAV planning to remove STREAM in "middle of 2010".
1275 CL also introduces -bmalware, various -d+acl logging additions and
1276 more caution in buffer sizes.
1278 PP/09 Implemented reverse_ip expansion operator.
1280 PP/10 Bugzilla 937: provide a "debug" ACL control.
1282 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
1284 PP/12 Bugzilla 973: Implement --version.
1286 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
1288 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
1290 PP/15 Bugzilla 816: support multiple condition rules on Routers.
1292 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
1293 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
1294 ignore trailing whitespace.
1296 JJ/02 prevent non-panic DKIM error from being sent to paniclog
1298 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
1301 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
1302 Notification from Dr Andrew Aitchison.
1304 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
1305 ExtendedDetectionInfo response format.
1306 Notification from John Horne.
1308 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
1311 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
1312 XSL and documented dependency on system catalogs, with examples of how
1315 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
1318 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
1319 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
1320 configuration file which is writeable by the Exim user or group.
1322 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
1323 of configuration files to cover files specified with the -C option if
1324 they are going to be used with root privileges, not just the default
1327 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
1328 option (effectively making it always true).
1330 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
1331 files to be used while preserving root privileges.
1333 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
1334 that rogue child processes cannot use them.
1336 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
1337 run-time user, instead of root.
1339 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
1340 Exim run-time user without dropping privileges.
1342 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
1343 result string, instead of calling string_vformat() twice with the same
1346 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
1347 for other users. Others should always drop root privileges if they use
1348 -C on the command line, even for a whitelisted configure file.
1350 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
1352 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
1358 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
1359 $data_path, and $header_path variables; fixed documentation bugs and
1362 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
1363 exipick to access non-standard spools, including the "frozen" queue
1366 NM/01 Bugzilla 965: Support mysql stored procedures.
1367 Patch from Alain Williams
1369 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
1371 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
1372 Patch from Andreas Metzler
1374 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
1375 Patch from Kirill Miazine
1377 NM/05 Bugzilla 671: Added umask to procmail example.
1379 JJ/03 installed exipick 20100323.0, fixing doc bug
1381 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
1382 directory. Notification and patch from Dan Rosenberg.
1384 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
1386 TK/02 Improve log output when DKIM signing operation fails.
1388 MH/01 Treat the transport option dkim_domain as a colon separated
1389 list, not as a single string, and sign the message with each element,
1390 omitting multiple occurences of the same signer.
1392 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
1393 Bugzilla 985, 986. Patch by Simon Arlott
1395 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
1396 Patch by Simon Arlott
1398 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
1399 MBX locking. Notification from Dan Rosenberg.
1405 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
1407 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
1409 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
1411 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
1413 NM/04 Bugzilla 915: Flag broken perl installation during build.
1419 TK/01 Added patch by Johannes Berg that expands the main option
1420 "spamd_address" if it starts with a dollar sign.
1422 TK/02 Write list of recipients to X-Envelope-Sender header when building
1423 the mbox-format spool file for content scanning (suggested by Jakob
1426 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
1427 (http://www.dcc-servers.net/) support via dccifd. Activated by
1428 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
1430 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
1431 by Mark Daniel Reidel <mr@df.eu>.
1433 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
1434 When building exim an external PCRE library is now needed -
1435 PCRE is a system library on the majority of modern systems.
1436 See entry on PCRE_LIBS in EDITME file.
1438 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
1439 conversation. Added nologin parameter to request.
1440 Patch contributed by Kirill Miazine.
1442 TF/01 Do not log submission mode rewrites if they do not change the address.
1444 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
1446 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
1447 log files in place. Contributed by Roberto Lima.
1449 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
1451 TF/03 Bugzilla 615: When checking the local_parts router precondition
1452 after a local_part_suffix or local_part_prefix option, Exim now
1453 does not use the address's named list lookup cache, since this
1454 contains cached lookups for the whole local part.
1456 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
1457 Robert Millan. Documentation is in experimental-spec.txt.
1459 TF/04 Bugzilla 668: Fix parallel build (make -j).
1461 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
1463 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
1464 Patch provided by Jan Srzednicki.
1466 TF/05 Leading white space used to be stripped from $spam_report which
1467 wrecked the formatting. Now it is preserved.
1469 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
1470 that they are available at delivery time.
1472 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
1474 TF/08 TLS error reporting now respects the incoming_interface and
1475 incoming_port log selectors.
1477 TF/09 Produce a more useful error message if an SMTP transport's hosts
1478 setting expands to an empty string.
1480 NM/06 Bugzilla 744: EXPN did not work under TLS.
1481 Patch provided by Phil Pennock.
1483 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
1484 Patch provided by Richard Godbee.
1486 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
1487 acl_smtp_notquit, added index entry.
1489 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
1490 Patch provided by Eugene Bujak.
1492 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
1493 accept(). Patch provided by Maxim Dounin.
1495 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
1496 Patch provided by Phil Pennock.
1498 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
1500 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
1501 Patch provided by Brad "anomie" Jorsch.
1503 NM/14 Bugzilla 622: Added timeout setting to transport filter.
1504 Patch provided by Dean Brooks.
1506 TK/05 Add native DKIM support (does not depend on external libraries).
1508 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
1509 Patch provided by Graeme Fowler.
1511 NM/16 Bugzilla 851: Documentation example syntax fix.
1513 NM/17 Changed NOTICE file to remove references to embedded PCRE.
1515 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
1518 NM/19 Bugzilla 745: TLS version reporting.
1519 Patch provided by Phil Pennock.
1521 NM/20 Bugzilla 167: bool: condition support.
1522 Patch provided by Phil Pennock.
1524 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
1525 clients. Patch provided by Phil Pennock.
1527 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
1528 Patch provided by Brad "anomie" Jorsch.
1530 NM/23 Bugzilla 687: Fix misparses in eximstats.
1531 Patch provided by Heiko Schlittermann.
1533 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
1534 Patch provided by Heiko Schlittermann.
1536 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
1537 plus update to original patch.
1539 NM/26 Bugzilla 799: Documentation correction for ratelimit.
1541 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
1542 Patch provided by David Brownlee.
1544 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
1546 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
1548 NM/30 Bugzilla 888: TLS documentation bugfixes.
1550 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
1552 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
1553 Unlike the original bugzilla I have changed all shell scripts in src tree.
1555 NM/33 Bugzilla 898: Transport filter timeout fix.
1556 Patch by Todd Rinaldo.
1558 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
1559 Patch by Serge Demonchaux.
1561 NM/35 Bugzilla 39: Base64 decode bug fixes.
1562 Patch by Jakob Hirsch.
1564 NM/36 Bugzilla 909: Correct connect() call in dcc code.
1566 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
1568 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
1570 NM/39 Bugzilla 911: Fixed MakeLinks build script.
1576 TK/01 Add preliminary DKIM support. Currently requires a forked version of
1577 ALT-N's libdkim that I have put here:
1578 http://duncanthrax.net/exim-experimental/
1580 Note to Michael Haardt: I had to rename some vars in sieve.c. They
1581 were called 'true' and it seems that C99 defines that as a reserved
1582 keyword to be used with 'bool' variable types. That means you could
1583 not include C99-style headers which use bools without triggering
1584 build errors in sieve.c.
1586 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
1587 as mailq or other aliases. Changed the --help handling significantly
1588 to do whats expected. exim_usage() emits usage/help information.
1590 SC/01 Added the -bylocaldomain option to eximstats.
1592 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
1594 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
1596 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
1602 PH/01 Another patch from the Sieve maintainer.
1604 PH/02 When an IPv6 address is converted to a string for single-key lookup
1605 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
1606 dots are used instead of colons so that keys in lsearch files need not
1607 contain colons. This was done some time before quoting was made available
1608 in lsearch files. However, iplsearch files do require colons in IPv6 keys
1609 (notated using the quote facility) so as to distinguish them from IPv4
1610 keys. This meant that lookups for IP addresses in host lists did not work
1611 for iplsearch lookups.
1613 This has been fixed by arranging for IPv6 addresses to be expressed with
1614 colons if the lookup type is iplsearch. This is not incompatible, because
1615 previously such lookups could never work.
1617 The situation is now rather anomolous, since one *can* have colons in
1618 ordinary lsearch keys. However, making the change in all cases is
1619 incompatible and would probably break a number of configurations.
1621 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
1624 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
1625 conversion specification without a maximum field width, thereby enabling
1626 a rogue spamd server to cause a buffer overflow. While nobody in their
1627 right mind would setup Exim to query an untrusted spamd server, an
1628 attacker that gains access to a server running spamd could potentially
1629 exploit this vulnerability to run arbitrary code as the Exim user.
1631 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
1632 $primary_hostname instead of what libspf2 thinks the hosts name is.
1634 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
1635 a directory entry by the name of the lookup key. Previously, if a
1636 symlink pointed to a non-existing file or a file in a directory that
1637 Exim lacked permissions to read, a lookup for a key matching that
1638 symlink would fail. Now it is enough that a matching directory entry
1639 exists, symlink or not. (Bugzilla 503.)
1641 PH/03 The body_linecount and body_zerocount variables are now exported in the
1644 PH/04 Added the $dnslist_matched variable.
1646 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
1647 This means they are set thereafter only if the connection becomes
1650 PH/06 Added the client_condition to authenticators so that some can be skipped
1651 by clients under certain conditions.
1653 PH/07 The error message for a badly-placed control=no_multiline_responses left
1654 "_responses" off the end of the name.
1656 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
1658 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
1659 (without spaces) instead of just copying the configuration text.
1661 PH/10 Added the /noupdate option to the ratelimit ACL condition.
1663 PH/11 Added $max_received_linelength.
1665 PH/12 Added +ignore_defer and +include_defer to host lists.
1667 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
1668 way in which PCRE > 7.0 is built.
1670 PH/14 Implemented queue_only_load_latch.
1672 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
1673 MAIL command. The effect was to mangle the value on 64-bit systems.
1675 PH/16 Another patch from the Sieve maintainer.
1677 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
1679 PH/18 If a system quota error occurred while trying to create the file for
1680 a maildir delivery, the message "Mailbox is full" was not appended to the
1681 bounce if the delivery eventually timed out. Change 4.67/27 below applied
1682 only to a quota excession during the actual writing of the file.
1684 PH/19 It seems that peer DN values may contain newlines (and other non-printing
1685 characters?) which causes problems in log lines. The DN values are now
1686 passed through string_printing() before being added to log lines.
1688 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
1689 and InterBase are left for another time.)
1691 PH/21 Added message_body_newlines option.
1693 PH/22 Guard against possible overflow in moan_check_errorcopy().
1695 PH/23 POSIX allows open() to be a macro; guard against that.
1697 PH/24 If the recipient of an error message contained an @ in the local part
1698 (suitably quoted, of course), incorrect values were put in $domain and
1699 $local_part during the evaluation of errors_copy.
1705 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
1706 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
1709 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
1710 issue a MAIL command.
1712 PH/02 In an ACL statement such as
1714 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
1716 if a client was not listed at all, or was listed with a value other than
1717 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
1718 the condition was not true (as it should be), so access was not denied.
1719 The bug was that the ! inversion was incorrectly passed on to the second
1720 item. This has been fixed.
1722 PH/03 Added additional dnslists conditions == and =& which are different from
1723 = and & when the dns lookup returns more than one IP address.
1725 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
1726 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
1728 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
1729 FSYNC, which compiles an option called disable_fsync that allows for
1730 bypassing fsync(). The documentation is heavily laced with warnings.
1732 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
1734 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
1735 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
1736 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
1737 including adding "make clean"; (3) Added -fPIC when compiling the test
1738 dynamically loaded module, to get rid of a warning.
1740 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
1741 message fails, move_frozen_messages = true and ignore_bounce_errors_after
1742 = 0s. The bug is otherwise harmless.
1744 PH/07 There was a bug in the dovecot authenticator such that the value of
1745 $auth1 could be overwritten, and so not correctly preserved, after a
1746 successful authentication. This usually meant that the value preserved by
1747 the server_setid option was incorrect.
1749 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
1751 PH/09 Installed PCRE release 7.0.
1753 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
1754 run for batched SMTP input. It is now run at the start of every message
1755 in the batch. While fixing this I discovered that the process information
1756 (output by running exiwhat) was not always getting set for -bs and -bS
1757 input. This is fixed, and it now also says "batched" for BSMTP.
1759 PH/11 Added control=no_pipelining.
1761 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
1762 patch, slightly modified), and move the expansion of helo_data till after
1763 the connection is made in the smtp transport (so it can use these
1766 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
1768 PH/14 Added log_selector = +pid.
1770 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
1772 PH/16 Add ${if forany and ${if forall.
1774 PH/17 Added dsn_from option to vary the From: line in DSNs.
1776 PH/18 Flush SMTP output before performing a callout, unless control =
1777 no_callout_flush is set.
1779 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
1780 was true (the default) a successful delivery failed to delete the retry
1781 item, thus causing premature timeout of the address. The bug is now
1784 PH/20 Added hosts_avoid_pipelining to the smtp transport.
1786 PH/21 Long custom messages for fakedefer and fakereject are now split up
1787 into multiline reponses in the same way that messages for "deny" and
1788 other ACL rejections are.
1790 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
1791 with slight modification.
1793 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
1794 draft, changing the syntax and factoring some duplicate code".
1796 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
1797 for deliveries of the second and subsequent messages over the same SMTP
1800 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
1801 ${reduce, with only minor "tidies".
1803 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
1805 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
1806 expansion side effects.
1808 PH/27 When a message times out after an over-quota error from an Exim-imposed
1809 quota, the bounce message says "mailbox is full". This message was not
1810 being given when it was a system quota that was exceeded. It now should
1813 MH/03 Made $recipients available in local_scan(). local_scan() already has
1814 better access to the recipient list through recipients_list[], but
1815 $recipients can be useful in postmaster-provided expansion strings.
1817 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
1818 in the case of a MAIL command with additional options following the
1819 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
1820 were accidentally chopped off.
1822 PH/29 SMTP synchronization checks are implemented when a command is read -
1823 there is a check that no more input is waiting when there shouldn't be
1824 any. However, for some commands, a delay in an ACL can mean that it is
1825 some time before the response is written. In this time, more input might
1826 arrive, invalidly. So now there are extra checks after an ACL has run for
1827 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
1828 pipelining has not been advertised.
1830 PH/30 MH's patch to allow iscntrl() characters to be list separators.
1832 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
1833 returned in the SMTP response when smtp_return_error_details was false.
1834 This has been fixed.
1836 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
1837 instead of the C I/O that was originally supplied, because problems were
1838 reported on Solaris.
1840 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
1841 Exim which did not show up earlier: it was assuming that a call to
1842 SSL_CTX_set_info_callback() might give an error value. In fact, there is
1843 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
1844 was a macro that became an assignment, so it seemed to work. This has
1845 changed to a proper function call with a void return, hence the compile
1846 error. Exim's code has been fixed.
1848 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
1851 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
1853 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
1855 PH/37 If a message is not accepted after it has had an id assigned (e.g.
1856 because it turns out to be too big or there is a timeout) there is no
1857 "Completed" line in the log. When some messages of this type were
1858 selected by exigrep, they were listed as "not completed". Others were
1859 picked up by some special patterns. I have improved the selection
1860 criteria to be more general.
1862 PH/38 The host_find_failed option in the manualroute router can now be set
1863 to "ignore", to completely ignore a host whose IP address cannot be
1864 found. If all hosts are ignored, the behaviour is controlled by the new
1865 host_all_ignored option.
1867 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
1868 homing or because of multiple MX records with /mx) generated more than
1869 one IP address, and the following item turned out to be the local host,
1870 all the secondary addresses of the first item were incorrectly removed
1871 from the list, along with the local host and any following hosts (which
1872 is what is supposed to happen).
1874 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
1875 whoever called Exim into the -H file. In the case of the daemon it was
1876 behaving confusingly. When first started, it used values for whoever
1877 started the daemon, but after a SIGHUP it used the Exim user (because it
1878 calls itself on a restart). I have changed the code so that it now always
1881 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
1882 message are rejected with the same error (e.g. no authentication or bad
1883 sender address), and a DATA command is nevertheless sent (as can happen
1884 with PIPELINING or a stupid MUA), the error message that was given to the
1885 RCPT commands is included in the rejection of the DATA command. This is
1886 intended to be helpful for MUAs that show only the final error to their
1889 PH/42 Another patch from the Sieve maintainer.
1891 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
1892 Eximstats - Fixed some broken HTML links and added missing column headers
1894 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
1895 columns for Rejects, Temp Rejects, Ham, and Spam rows.
1897 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
1899 PH/43 Yet another patch from the Sieve maintainer.
1901 PH/44 I found a way to check for a TCP/IP connection going away before sending
1902 the response to the final '.' that terminates a message, but only in the
1903 case where the client has not sent further data following the '.'
1904 (unfortunately, this is allowed). However, in many cases there won't be
1905 any further data because there won't be any more messages to send. A call
1906 to select() can be used: if it shows that the input is "ready", there is
1907 either input waiting, or the socket has been closed. An attempt to read
1908 the next input character can distinguish the two cases. Previously, Exim
1909 would have sent an OK response which the client would never have see.
1910 This could lead to message repetition. This fix should cure that, at
1911 least in a lot of common cases.
1913 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
1914 advertised in response to EHLO.
1920 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
1921 fixed by 4.65/MH/01 (is this a record?) are fixed:
1923 (i) An empty string was always treated as zero by the numeric comparison
1924 operators. This behaviour has been restored.
1926 (ii) It is documented that the numeric comparison operators always treat
1927 their arguments as decimal numbers. This was broken in that numbers
1928 starting with 0 were being interpreted as octal.
1930 While fixing these problems I realized that there was another issue that
1931 hadn't been noticed. Values of message_size_limit (both the global option
1932 and the transport option) were treated as octal if they started with 0.
1933 The documentation was vague. These values are now always treated as
1934 decimal, and I will make that clear in the documentation.
1940 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
1941 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
1944 MH/01 Don't check that the operands of numeric comparison operators are
1945 integers when their expansion is in "skipping" mode (fixes bug
1946 introduced by 4.64-PH/07).
1948 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
1949 child addresses, Exim now panics and dies. Previously, because the count
1950 is held in a short int, deliveries were likely to be lost. As such a
1951 large number of recipients for a single message is ridiculous
1952 (performance will be very, very poor), I have chosen to impose a limit
1953 rather than extend the field.
1959 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
1960 leftover -K file (the existence of which was triggered by #402).
1961 While we were at it, introduced process PID as part of the -K
1962 filename. This should rule out race conditions when creating
1965 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
1966 processing considerably. Previous code took too long for large mails,
1967 triggering a timeout which in turn triggers #401.
1969 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
1970 in the DK code in transports.c. sendfile() is not really portable,
1971 hence the _LINUX specificness.
1973 TF/01 In the add_headers option to the mail command in an Exim filter,
1974 there was a bug that Exim would claim a syntax error in any
1975 header after the first one which had an odd number of characters
1978 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
1979 callout verification, Exim cached a "reject" for the entire domain. This
1980 is correct for most verifications, but it is not correct for a recipient
1981 verification with use_sender or use_postmaster set, because in that case
1982 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
1983 case of MAIL FROM:<> rejection from other early rejections (e.g.
1984 rejection of HELO). When verifying a recipient using a non-null MAIL
1985 address, the cache is ignored if it shows MAIL FROM:<> rejection.
1986 Whatever the result of the callout, the value of the domain cache is
1987 left unchanged (for any other kind of callout, getting as far as trying
1988 RCPT means that the domain itself is ok).
1990 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
1993 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
1994 manifest itself as EPIPE rather than ECONNECT. When tidying away a
1995 session, the daemon ignores ECONNECT errors and logs others; it now
1996 ignores EPIPE as well.
1998 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
1999 (quoted-printable decoding).
2001 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
2002 later the small subsequent patch to fix an introduced bug.
2004 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
2006 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
2008 PH/08 An error is now given if message_size_limit is specified negative.
2010 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
2011 to be given (somewhat) arbitrary names.
2013 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
2016 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
2017 miscellaneous code fixes
2019 PH/10 Added the log_reject_target ACL modifier to specify where to log
2022 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
2023 hostname. This is wrong, because it relates to the incoming message (and
2024 probably the interface on which it is arriving) and not to the outgoing
2025 callout (which could be using a different interface). This has been
2026 changed to use the value of the helo_data option from the smtp transport
2027 instead - this is what is used when a message is actually being sent. If
2028 there is no remote transport (possible with a router that sets up host
2029 addresses), $smtp_active_hostname is used.
2031 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
2032 tweaks were necessary in order to get it to work (see also 21 below):
2033 (a) The code assumed that strncpy() returns a negative number on buffer
2034 overflow, which isn't the case. Replaced with Exim's string_format()
2036 (b) There were several signed/unsigned issues. I just did the minimum
2037 hacking in of casts. There is scope for a larger refactoring.
2038 (c) The code used strcasecmp() which is not a standard C function.
2039 Replaced with Exim's strcmpic() function.
2040 (d) The code set only $1; it now sets $auth1 as well.
2041 (e) A simple test gave the error "authentication client didn't specify
2042 service in request". It would seem that Dovecot has changed its
2043 interface. Fortunately there's a specification; I followed it and
2044 changed what the client sends and it appears to be working now.
2046 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
2049 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
2050 address A is aliased to B and C, where B exists and C does not. Without
2051 -v the output is "A verified" because verification stops after a
2052 successful redirection if more than one address is generated. However,
2053 with -v the child addresses are also verified. Exim was outputting "A
2054 failed to verify" and then showing the successful verification for C,
2055 with its parentage. It now outputs "B failed to verify", showing B's
2056 parentage before showing the successful verification of C.
2058 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
2059 look up a TXT record in a specific list after matching in a combined
2062 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
2063 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
2064 they consult the DNS. I had assumed they would set it the way they
2065 wanted; and indeed my experiments on Linux seem to show that in some
2066 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
2067 To be on the safe side, however, I have now made the interface to
2068 host_find_byname() similar to host_find_bydns(), with an argument
2069 containing the DNS resolver options. The host_find_byname() function now
2070 sets these options at its start, just as host_find_bydns() does. The smtp
2071 transport options dns_qualify_single and dns_search_parents are passed to
2072 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
2073 of host_find_byname() use the default settings of RES_DEFNAMES
2074 (qualify_single) but not RES_DNSRCH (search_parents).
2076 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
2077 spool_read_header() do less string testing, by means of a preliminary
2078 switch on the second character of optional "-foo" lines. (This is
2079 overdue, caused by the large number of possibilities that now exist.
2080 Originally there were few.) While I was there, I also converted the
2081 str(n)cmp tests so they don't re-test the leading "-" and the first
2082 character, in the hope this might squeeze out yet more improvement.
2084 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
2085 flag allowing group syntax was set by the header_syntax check but not
2086 turned off, possible causing trouble later; (2) The flag was not being
2087 set at all for the header_verify test, causing "group"-style headers to
2088 be rejected. I have now set it in this case, and also caused header_
2089 verify to ignore an empty address taken from a group. While doing this, I
2090 came across some other cases where the code for allowing group syntax
2091 while scanning a header line wasn't quite right (mostly, not resetting
2092 the flag correctly in the right place). These bugs could have caused
2093 trouble for malformed header lines. I hope it is now all correct.
2095 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
2096 with the "reply" argument non-NULL. The code, however (which originally
2097 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
2098 but it didn't always do it. This confused somebody who was copying the
2099 code for some other use. I have removed all the tests.
2101 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
2102 feature that was used to support insecure browsers during the U.S. crypto
2103 embargo. It requires special client support, and Exim is probably the
2104 only MTA that supported it -- and would never use it because real RSA is
2105 always available. This code has been removed, because it had the bad
2106 effect of slowing Exim down by computing (never used) parameters for the
2107 RSA_EXPORT functionality.
2109 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
2110 authenticator to fail if there's a tab character in the incoming data
2111 (there should never be unless someone is messing about, as it's supposed
2112 to be base64-encoded). Also added, on Timo's advice, the "secured" option
2113 if the connection is using TLS or if the remote IP is the same as the
2114 local IP, and the "valid-client-cert option" if a client certificate has
2117 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
2118 authenticators. This can be used for authorization after authentication
2119 succeeds. (In the case of plaintext, it servers for both authentication
2122 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
2123 if any retry times were supplied.
2125 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
2126 connection, where there is no client IP address to check. In this
2127 situation, the verify now always succeeds.
2129 PH/25 Applied John Jetmore's -Mset patch.
2131 PH/26 Added -bem to be like -Mset, but loading a message from a file.
2133 PH/27 In a string expansion for a processed (not raw) header when multiple
2134 headers of the same name were present, leading whitespace was being
2135 removed from all of them, but trailing whitespace was being removed only
2136 from the last one. Now trailing whitespace is removed from each header
2137 before concatenation. Completely empty headers in a concatenation (as
2138 before) are ignored.
2140 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
2141 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
2143 PH/29 [Removed. This was a change that I later backed out, and forgot to
2144 correct the ChangeLog entry (that I had efficiently created) before
2145 committing the later change.]
2147 PH/30 Exim was sometimes attempting to deliver messages that had suffered
2148 address errors (4xx response to RCPT) over the same connection as other
2149 messages routed to the same hosts. Such deliveries are always "forced",
2150 so retry times are not inspected. This resulted in far too many retries
2151 for the affected addresses. The effect occurred only when there were more
2152 hosts than the hosts_max_try setting in the smtp transport when it had
2153 the 4xx errors. Those hosts that it had tried were not added to the list
2154 of hosts for which the message was waiting, so if all were tried, there
2155 was no problem. Two fixes have been applied:
2157 (i) If there are any address or message errors in an SMTP delivery, none
2158 of the hosts (tried or untried) are now added to the list of hosts
2159 for which the message is waiting, so the message should not be a
2160 candidate for sending over the same connection that was used for a
2161 successful delivery of some other message. This seems entirely
2162 reasonable: after all the message is NOT "waiting for some host".
2163 This is so "obvious" that I'm not sure why it wasn't done
2164 previously. Hope I haven't missed anything, but it can't do any
2165 harm, as the worst effect is to miss an optimization.
2167 (ii) If, despite (i), such a delivery is accidentally attempted, the
2168 routing retry time is respected, so at least it doesn't keep
2169 hammering the server.
2171 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
2172 in ${readsocket because some servers need this prod.
2174 PH/32 Added some extra debug output when updating a wait-xxx database.
2176 PH/33 The hint "could be header name not terminated by colon", which has been
2177 given for certain expansion errors for a long time, was not being given
2178 for the ${if def:h_colon_omitted{... case.
2180 PH/34 The spec says: "With one important exception, whenever a domain list is
2181 being scanned, $domain contains the subject domain." There was at least
2182 one case where this was not true.
2184 PH/35 The error "getsockname() failed: connection reset by peer" was being
2185 written to the panic log as well as the main log, but it isn't really
2186 panic-worthy as it just means the connection died rather early on. I have
2187 removed the panic log writing for the ECONNRESET error when getsockname()
2190 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
2191 runs only) independently of the message's sender address. This meant
2192 that, if the 4xx error was in fact related to the sender, a different
2193 message to the same recipient with a different sender could confuse
2194 things. In particualar, this can happen when sending to a greylisting
2195 server, but other circumstances could also provoke similar problems.
2196 I have changed the default so that the retry time for these errors is now
2197 based a combination of the sender and recipient addresses. This change
2198 can be overridden by setting address_retry_include_sender=false in the
2201 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
2202 remote server are returned as part of bounce messages. This was not
2203 happening for LMTP over a pipe (the lmtp transport), but now it is the
2204 same for both kinds of LMTP.
2206 PH/38 Despite being documented as not happening, Exim was rewriting addresses
2207 in header lines that were in fact CNAMEs. This is no longer the case.
2209 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
2210 and queue runs started by the daemon processed all messages. This has
2211 been fixed so that -R and -S can now usefully be given with -q<time>.
2213 PH/40 Import PCRE release 6.7 (fixes some bugs).
2215 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
2217 PH/42 Give an error if -q is specified more than once.
2219 PH/43 Renamed the variables $interface_address and $interface_port as
2220 $received_ip_address and $received_port, to make it clear that these
2221 values apply to message reception, and not to the outgoing interface when
2222 a message is delivered. (The old names remain recognized, of course.)
2224 PH/44 There was no timeout on the connect() call when using a Unix domain
2225 socket in the ${readsocket expansion. There now is.
2227 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
2228 be meaningful with "accept".
2230 SC/01 Eximstats V1.43
2231 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
2233 SC/02 Eximstats V1.44
2234 Use a glob alias rather than an array ref in the generated
2235 parser. This improves both readability and performance.
2237 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
2238 Collect SpamAssassin and rejection statistics.
2239 Don't display local sender or destination tables unless
2240 there is data to show.
2241 Added average volumes into the top table text output.
2243 SC/04 Eximstats V1.46
2244 Collect data on the number of addresses (recipients)
2245 as well as the number of messages.
2247 SC/05 Eximstats V1.47
2248 Added 'Message too big' to the list of mail rejection
2249 reasons (thanks to Marco Gaiarin).
2251 SC/06 Eximstats V1.48
2252 Mainlog lines which have GMT offsets and are too short to
2253 have a flag are now skipped.
2255 SC/07 Eximstats V1.49 (Alain Williams)
2256 Added the -emptyok flag.
2258 SC/08 Eximstats V1.50
2259 Fixes for obtaining the IP address from reject messages.
2261 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
2262 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
2263 whitesspace changes from 4.64-PH/27
2265 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
2268 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
2269 are found, allow negative numbers in numeric criteria)
2271 JJ/06 exipick.20061117.2, added new $message_body_missing variable
2273 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
2274 to match changes made in 4.64-PH/43
2276 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
2278 PH/47 Put in an explicit test for a DNS lookup of an address record where the
2279 "domain" is actually an IP address, and force a failure. This locks out
2280 those revolvers/nameservers that support "A-for-A" lookups, in
2281 contravention of the specifications.
2283 PH/48 When a host name was looked up from an IP address, and the subsequent
2284 forward lookup of the name timed out, the host name was left in
2285 $sender_host_name, contrary to the specification.
2287 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
2288 restricted to single-key lookups, Exim was not diagnosing an error if
2289 * or *@ was used with a query-style lookup.
2291 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
2293 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
2294 long ago, but noone interested enough thought of it. Let's just say that
2295 the "1.1" means that there are some new functions that weren't there at
2296 some point in the past.
2298 PH/51 Error processing for expansion failure of helo_data from an smtp
2299 transport during callout processing was broken.
2301 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
2302 tested/used via the -bh/-bhc/-bs options.
2304 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
2305 bug, fixed in subsequent PCRE releases).
2307 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
2308 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
2310 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
2316 SC/01 Use a glob alias rather than an array ref in eximstats generated
2317 parser. This improves both readability and performance.
2319 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
2320 Don't display local sender or destination tables in eximstats unless
2321 there is data to show.
2322 Added average volumes into the eximstats top table text output.
2324 SC/03 Collect data on the number of addresses (recipients) as well
2325 as the number of messages in eximstats.
2327 TF/01 Correct an error in the documentation for the redirect router. Exim
2328 does (usually) call initgroups() when daemonizing.
2330 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
2331 with consistent privilege compared to when running as a daemon.
2333 TF/03 Note in the spec that $authenticated_id is not set for local
2334 submissions from trusted users.
2336 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
2337 Thanks to Dean Brooks <dean@iglou.com> for the patch.
2339 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
2340 by adding some example configuration directives to the default
2341 configuration file. A little bit of work is required to uncomment the
2342 directives and define how usernames and passwords are checked, but
2343 there is now a framework to start from.
2345 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
2346 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
2347 without this. I don't know how relevant this is to other LDAP libraries.
2349 PH/02 Add the verb name to the "unknown ACL verb" error.
2351 PH/03 Magnus Holmgren's patch for filter_prepend_home.
2353 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
2355 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
2356 directory not expanded when it should be if an expanded home directory
2357 was set for the address (which is overridden by the transport).
2359 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
2362 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
2363 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
2364 because it is too late at that time, and has no effect.
2366 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
2367 security issue with \' (bugzilla #107). I could not use the
2368 PQescapeStringConn() function, because it needs a PGconn value as one of
2371 PH/08 When testing addresses using -bt, indicate those final addresses that
2372 are duplicates that would not cause an additional delivery. At least one
2373 person was confused, thinking that -bt output corresponded to deliveries.
2374 (Suppressing duplicates isn't a good idea as you lose the information
2375 about possibly different redirections that led to the duplicates.)
2377 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
2378 systems where poll() doesn't work, in particular OS X.
2380 PH/10 Added more information to debugging output for retry time not reached.
2382 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
2383 operations in malware.c.
2385 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
2388 PH/13 If write_rejectlog was set false when logging was sent to syslog with
2389 syslog_duplication set false, log lines that would normally be written
2390 both the the main log and to the reject log were not written to syslog at
2393 PH/14 In the default configuration, change the use of "message" in ACL warn
2394 statements to "add_header".
2396 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
2397 not followed by a command (e.g. "seen endif").
2399 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
2400 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
2403 PH/17 Added extra conditions to the default value of delay_warning_condition
2407 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
2408 { match{$h_precedence:}{(?i)bulk|list|junk} } \
2409 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
2412 The Auto-Submitted: and various List- headers are standardised, whereas I
2413 don't think Precedence: ever was.
2415 PH/18 Refactored debugging code in route_finduser() to show more information,
2416 in particular, the error code if getpwnam() issues one.
2418 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
2419 This is apparently needed in addition to the PH/07 change above to avoid
2420 any possible encoding problems.
2422 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
2423 but not after initializing Perl.
2425 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
2426 output them only if debugging. By default they are written stderr,
2427 apparently, which is not desirable.
2429 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
2432 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
2435 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
2437 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
2438 authenticated or an ident call has been made. Suppress the default
2439 values for $authenticated_id and $authenticated_sender (but permit -oMai
2440 and -oMas) when testing with -bh.
2442 PH/24 Re-jigged the order of the tests in the default configuration so that the
2443 tests for valid domains and recipients precede the DNS black list and CSA
2444 tests, on the grounds that those ones are more expensive.
2446 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
2447 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
2448 This bug exists in every version of Exim that I still have, right back to
2451 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
2452 However, an attempt to turn on case-sensitivity in a regex key by
2453 including (?-i) didn't work because the subject string was already
2454 lowercased, and the effects were non-intuitive. It turns out that a
2455 one-line patch can be used to allow (?-i) to work as expected.
2461 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
2462 other effects) broke the use of negated acl sub-conditions.
2464 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
2467 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
2468 "Deny" causes Exim to reject the incoming connection with a 554 error.
2469 Unfortunately, if there is a major crisis, such as a disk failure,
2470 tcp-wrappers gives "deny", whereas what one would like would be some
2471 kind of temporary error. A kludge has been added to help with this.
2472 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
2473 554 error is used if errno is still zero or contains ENOENT (which occurs
2474 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
2477 PH/03 Add -lutil to the default FreeBSD LIBS setting.
2479 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
2480 errors. Otherwise a message that provokes a temporary error (when other
2481 messages do not) can cause a whole host to time out.
2483 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
2484 addresses were routed directly to files or pipes from a redirect router.
2485 File deliveries just didn't batch; pipe deliveries might have suffered
2488 PH/06 A failure to get a lock for a hints database would erroneously always say
2489 "Failed to get write lock", even when it was really a read lock.
2491 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
2492 of 0600. This has been changed to use the value of the lockfile_mode
2493 option (which defaults to 0600).
2495 PH/08 Applied small patch from the Sieve maintainer.
2497 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
2498 folder from quota calculations, a direct delivery into this folder messed
2499 up the contents of the maildirsize file. This was because the regex was
2500 used only to exclude .Trash (or whatever) when the size of the mailbox
2501 was calculated. There was no check that a delivery was happening into an
2502 excluded directory. This bug has been fixed by ignoring all quota
2503 processing for deliveries into excluded directories.
2505 PH/10 Added the maildirfolder_create_regex option to appendfile.
2511 PH/01 The code for finding all the local interface addresses on a FreeBSD
2512 system running IPv6 was broken. This may well have applied to all BSD
2513 systems, as well as to others that have similar system calls. The broken
2514 code found IPv4 interfaces correctly, but gave incorrect values for the
2515 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
2516 that it would not match correctly against @[] and not recognize the IPv6
2519 PH/02 The ipliteral router was not recognizing addresses of the form user@
2520 [ipv6:....] because it didn't know about the "ipv6:" prefix.
2522 PH/03 Added disable_ipv6.
2524 PH/04 Changed $reply_address to use the raw form of the headers instead of the
2525 decoded form, because it is most often used to construct To: headers
2526 lines in autoreplies, and the decoded form may well be syntactically
2527 invalid. However, $reply_address has leading white space removed, and all
2528 newlines turned into spaces so that the autoreply transport does not
2531 PH/05 If group was specified without a user on a router, and no group or user
2532 was specified on a transport, the group from the router was ignored.
2534 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
2535 for visible compile-time settings that can be used to change these
2536 numbers, for those that want even more. Backwards compatibility with old
2537 spool files has been maintained. However, going back to a previous Exim
2538 release will lost any variables that are in spool files.
2540 PH/07 Two small changes when running in the test harness: increase delay when
2541 passing a TCP/IP connection to a new process, in case the original
2542 process has to generate a bounce, and remove special handling of
2543 127.0.0.2 (sic), which is no longer necessary.
2545 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
2546 be the same on different OS.
2548 PH/09 Moved a debug statement in filter processing to avoid a race problem when
2551 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
2552 whether --show-vars was specified or not
2554 JJ/02 exipick: Added support for new ACL variable spool format introduced
2557 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
2558 syntactically invalid From: or Reply-to: line, and a filter used this to
2559 generate an autoreply, and therefore failed to obtain an address for the
2560 autoreply, Exim could try to deliver to a non-existent relative file
2561 name, causing unrelated and misleading errors. What now happens is that
2562 it logs this as a hard delivery error, but does not attempt to create a
2565 PH/11 The exinext utility has a -C option for testing purposes, but although
2566 the given file was scanned by exinext itself; it wasn't being passed on
2567 when Exim was called.
2569 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
2570 an end-of-file indication when reading a command response.
2572 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
2573 compiled. In many other places in Exim, IPv6 addresses are always
2574 recognized, so I have changed this. It also means that IPv4 domain
2575 literals of the form [IPV4:n.n.n.n] are now always recognized.
2577 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
2578 used if the router is not running as root, for example, when verifying at
2579 ACL time, or when using -bh. The debugging output from this situation was
2580 non-existent - all you got was a failure to exec. I have made two
2583 (a) Failures to set uid/gid, the current directory, or a process leader
2584 in a subprocess such as that created by queryprogram now generate
2585 suitable debugging ouput when -d is set.
2587 (b) The queryprogram router detects when it is not running as root,
2588 outputs suitable debugging information if -d is set, and then runs
2589 the subprocess without attempting to change uid/gid.
2591 PH/15 Minor change to Makefile for building test_host (undocumented testing
2594 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
2595 additional section of a DNS packet that returns MX or SRV records.
2596 Instead, it always explicitly searches for A/AAAA records. This avoids
2597 major problems that occur when a DNS server includes only records of one
2598 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
2599 fixed another bug: if SRV records were looked up and the corresponding
2600 address records were *not* found in the additional section, the port
2601 values from the SRV records were lost.
2603 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
2604 using the correct key (the original address) when searching the retry
2605 rules in order to find which one to use for generating the retry hint.
2607 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
2608 adding the default one. Similarly, if it contains a Reply-To: header, the
2609 errors_reply_to option, if set, is not used.
2611 PH/19 When calculating a retry time, Exim used to measure the "time since
2612 failure" by looking at the "first failed" field in the retry record. Now
2613 it does not use this if it is later than than the arrival time of the
2614 message. Instead it uses the arrival time. This makes for better
2615 behaviour in cases where some deliveries succeed, thus re-setting the
2616 "first failed" field. An example is a quota failure for a huge message
2617 when small messages continue to be delivered. Without this change, the
2618 "time since failure" will always be short, possible causing more frequent
2619 delivery attempts for the huge message than are intended.
2620 [Note: This change was subsequently modified - see PH/04 for 4.62.]
2622 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
2623 $1, $2, $3) because the numerical variables can be reset during some
2624 expansion items (e.g. "match"), thereby losing the authentication data.
2626 PH/21 Make -bV show the size of off_t variables so that the test suite can
2627 decide whether to run tests for quotas > 2G.
2629 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
2630 mailbox_size, and mailbox_filecount in the appendfile transport. If a
2631 filecount value is greater than 2G or if a quota value is greater than 2G
2632 on a system where the size of off_t is not greater than 4, a panic error
2635 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
2636 never match. The debug and -bh output now contains an explicit error
2637 message indicating a malformed IPv4 address or mask.
2639 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
2640 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
2641 PH/23 above applies.
2643 PH/25 Do not write to syslog when running in the test harness. The only
2644 occasion when this arises is a failure to open the main or panic logs
2645 (for which there is an explicit test).
2647 PH/26 Added the /no_tell option to "control=freeze".
2649 PH/27 If a host name lookup failed very early in a connection, for example, if
2650 the IP address matched host_lookup and the reverse lookup yielded a name
2651 that did not have a forward lookup, an error message of the form "no IP
2652 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
2653 could be logged. Now it outputs the IP address instead of "NULL".
2655 PH/28 An enabling patch from MH: add new function child_open_exim2() which
2656 allows the sender and the authenticated sender to be set when
2657 submitting a message from within Exim. Since child_open_exim() is
2658 documented for local_scan(), the new function should be too.
2660 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
2661 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
2662 results in an empty string is now treated as unset.
2664 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
2666 PH/31 Added sender_verify_fail logging option.
2668 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
2669 needed by RFC 822 but not by RFC 2822 was commented out. I have now
2670 tidied the source and removed it altogether.
2672 PH/33 When a queue run was abandoned because the load average was too high, a
2673 log line was always written; now it is written only if the queue_run log
2674 selector is set. In addition, the log line for abandonment now contains
2675 information about the queue run such as the pid. This is always present
2676 in "start" and "stop" lines but was omitted from the "abandon" line.
2678 PH/34 Omit spaces between a header name and the colon in the error message that
2679 is given when verify = headers_syntax fails (if there are lots of them,
2680 the message gets confusing).
2682 PH/35 Change the default for dns_check_names_pattern to allow slashes within
2683 names, as there are now some PTR records that contain slashes. This check
2684 is only to protect against broken name servers that fall over on strange
2685 characters, so the fact that it applies to all lookups doesn't matter.
2687 PH/36 Now that the new test suite is complete, we can remove some of the
2688 special code in Exim that was needed for the old test suite. For example,
2689 sorting DNS records because real resolvers return them in an arbitrary
2690 order. The new test suite's fake resolver always returns records in the
2693 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
2694 bounces) except when queue_only is set, to avoid logging races between
2695 the different processes.
2697 PH/38 Panic-die if .include specifies a non-absolute path.
2699 PH/39 A tweak to the "H" retry rule from its user.
2701 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
2702 a label. They prevented compilation on older perls.
2704 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
2705 a warning to be raised on newish perls.
2707 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
2708 on queue. Changes to match documented behaviour of showing count of
2709 messages matching specified criteria.
2711 PH/40 Changed the default ident timeout from 30s to 5s.
2713 PH/41 Added support for the use of login_cap features, on those BSD systems
2714 that have them, for controlling the resources used by pipe deliveries.
2716 PH/42 The content-scanning code uses fopen() to create files in which to put
2717 message data. Previously it was not paying any attention to the mode of
2718 the files. Exim runs with umask(0) because the rest of the code creates
2719 files with open(), and sets the required mode explicitly. Thus, these
2720 files were ending up world-writeable. This was not a big issue, because,
2721 being within the spool directory, they were not world-accessible. I have
2722 created a function called modefopen, which takes an additional mode
2723 argument. It sets umask(777), creates the file, chmods it to the required
2724 mode, then resets the umask. All the relevant calls to fopen() in the
2725 content scanning code have been changed to use this function.
2727 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
2728 to 24 hours. This avoids potential overflow problems when processing G
2729 and H retry rules. I suspect nobody ever tinkers with this value.
2731 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
2733 PH/45 When the plaintext authenticator is running as a client, the server's
2734 challenges are checked to ensure they are valid base64 strings. By
2735 default, the authentication attempt is cancelled if an invalid string is
2736 received. Setting client_ignore_invalid_base64 true ignores these errors.
2737 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
2738 they are received. Thus, the responses can be made to depend on the
2739 challenges. If an invalid string is ignored, an empty string is placed in
2742 PH/46 Messages that are created by the autoreply transport now contains a
2743 References: header, in accordance with RFCs 2822 and 3834.
2745 PH/47 Added authenticated_sender_force to the smtp transport.
2747 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
2749 PH/49 Installed latest patch from the Sieve maintainer.
2751 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
2752 was also set, the appendfile transport was unnecessarily scanning a
2753 directory of message files (e.g. for maildir delivery) to find the count
2754 of files (along with the size), even though it did not need this
2755 information. It now does the scan only if it needs to find either the
2756 size of the count of files.
2758 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
2760 PH/52 Two bugs concerned with error handling when the smtp transport is
2763 (i) Exim was not creating retry information for temporary errors given
2764 for individual recipients after the DATA command when the smtp transport
2765 was used in LMTP mode. This meant that they could be retried too
2766 frequently, and not timed out correctly.
2768 (ii) Exim was setting the flag that allows error details to be returned
2769 for LMTP errors on RCPT commands, but not for LMTP errors for individual
2770 recipients that were returned after the DATA command.
2772 PH/53 This is related to PH/52, but is more general: for any failing address,
2773 when detailed error information was permitted to be returned to the
2774 sender, but the error was temporary, then after the final timeout, only
2775 "retry timeout exceeded" was returned. Now it returns the full error as
2776 well as "retry timeout exceeded".
2778 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
2779 do this, and (what is worse) MTAs that accept it.
2781 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
2782 will now be deprecated.
2784 PH/56 New os.c-cygwin from the Cygwin maintainer.
2786 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
2787 formats (previously only available in exim formats via -bpr, -bpru,
2788 and -bpra. Now also available in native and exiqgrep formats)
2790 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
2791 with very large, slow to parse queues
2793 JJ/08 exipick: added ! as generic prefix to negate any criteria format
2795 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
2797 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
2798 responses to authentication challenges, though it was showing the
2799 challenges; (ii) I've removed the CR characters from the debug output for
2802 PH/58 Allow for the insertion of a newline as well as a space when a string
2803 is turned into more than one encoded-word during RFC 2047 encoding. The
2804 Sieve code now uses this.
2806 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
2807 data_4xx, lost_connection, tls_required.
2809 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
2810 message was being sent as an SMTP response.
2812 PH/61 Add -l and -k options to exicyclog.
2814 PH/62 When verifying, if an address was redirected to one new address, so that
2815 verification continued, and the new address failed or deferred after
2816 having set something in $address_data, the value of $address_data was not
2817 passed back to the ACL. This was different to the case when no
2818 redirection occurred. The value is now passed back in both cases.
2820 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
2821 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
2822 use login_cap.h, so on its own it isn't the distinguishing feature. The
2823 new name refers directly to the setclassresources() function.
2825 PH/65 Added configuration files for NetBSD3.
2827 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
2829 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
2830 is preferred over IPv4.
2832 PH/68 The bounce_return_message and bounce_return_body options were not being
2833 honoured for bounces generated during the reception of non-SMTP messages.
2834 In particular, this applied to messages rejected by the ACL. This bug has
2835 been fixed. However, if bounce_return_message is true and bounce_return_
2836 body is false, the headers that are returned for a non-SMTP message
2837 include only those that have been read before the error was detected.
2838 (In the case of an ACL rejection, they have all been read.)
2840 PH/69 The HTML version of the specification is now built in a directory called
2841 spec_html instead of spec.html, because the latter looks like a path with
2842 a MIME-type, and this confuses some software.
2844 PH/70 Catch two compiler warnings in sieve.c.
2846 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
2847 function verify_get_ident() calls ip_connect() to connect a socket, but
2848 if the "connect()" function timed out, ip_connect() used to close the
2849 socket. However, verify_get_ident() also closes the socket later, and in
2850 between Exim writes to the log, which may get opened at this point. When
2851 the socket was closed in ip_connect(), the log could get the same file
2852 descriptor number as the socket. This naturally causes chaos. The fix is
2853 not to close the socket in ip_connect(); the socket should be closed by
2854 the function that creates it. There was only one place in the code where
2855 this was missing, in the iplookup router, which I don't think anybody now
2856 uses, but I've fixed it anyway.
2858 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
2859 well as to direct DNS lookups. Otherwise the handling of names in host
2860 lists is inconsistent and therefore confusing.
2866 PH/01 Two changes to the default runtime configuration:
2868 (1) Move the checks for relay_from_hosts and authenticated clients from
2869 after to before the (commented out) DNS black list checks.
2871 (2) Add control=submission to the relay_from_hosts and authenticated
2872 clients checks, on the grounds that messages accepted by these
2873 statements are most likely to be submissions.
2875 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
2877 (1) Generate an error if the third argument for the ${prvs expansion is
2880 (2) Treat a missing third argument of ${prvscheck as if it were an empty
2883 (3) Reset the variables that are obtained from the first argument of
2884 ${prvscheck and used in the second argument before leaving the code,
2885 because their memory is reclaimed, so using them afterwards may do
2888 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
2889 one (it's much easier than Tom thought :-).
2891 (5) Because of (4), we can now allow for the use of $prvscheck_result
2892 inside the third argument.
2894 PH/03 For some reason, the default setting of PATH when running a command from
2895 a pipe transport was just "/usr/bin". I have changed it to
2898 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
2899 anything to be listed in the output from -bV.
2901 PH/05 When a filter generated an autoreply, the entire To: header line was
2902 quoted in the delivery log line, like this:
2904 => >A.N.Other <ano@some.domain> <original@ddress> ...
2906 This has been changed so that it extracts the operative address. There
2907 may be more than one such address. If so, they are comma-separated, like
2910 => >ano@some.domain,ona@other.domain <original@ddress> ...
2912 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
2913 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
2914 not being looked up in the rDNS to get a host name, Exim was showing the
2915 IP address twice in Received: lines, even though the IP addresses were
2916 identical. For example:
2918 Received: from [1.2.3.4] (helo=[1.2.3.4])
2920 However, if the real host name was known, it was omitting the HELO data
2921 if it matched the actual IP address. This has been tidied up so that it
2922 doesn't show the same IP address twice.
2924 PH/07 When both +timestamp and +memory debugging was on, the value given by
2925 $tod_xxx expansions could be wrong, because the tod_stamp() function was
2926 called by the debug printing, thereby overwriting the timestamp buffer.
2927 Debugging no longer uses the tod_stamp() function when +timestamp is set.
2929 PH/08 When the original message was included in an autoreply transport, it
2930 always said "this is a copy of the message, including all the headers",
2931 even if body_only or headers_only was set. It now gives an appropriate
2934 PH/09 Applied a patch from the Sieve maintainer which:
2936 o fixes some comments
2937 o adds the (disabled) notify extension core
2938 o adds some debug output for the result of if/elsif tests
2939 o points to the current vacation draft in the documentation
2940 and documents the missing references header update
2944 o fixes a bug in processing the envelope test (when testing
2945 multiple envelope elements, the last element determinted the
2948 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
2949 Electronic Mail") by including:
2951 Auto-submitted: auto-generated
2953 in the messages that it generates (bounce messages and others, such as
2954 warnings). In the case of bounce messages for non-SMTP mesages, there was
2955 also a typo: it was using "Auto_submitted" (underscore instead of
2956 hyphen). Since every message generated by Exim is necessarily in response
2957 to another message, thes have all been changed to:
2959 Auto-Submitted: auto-replied
2961 in accordance with these statements in the RFC:
2963 The auto-replied keyword:
2965 - SHOULD be used on messages sent in direct response to another
2966 message by an automatic process,
2968 - MUST NOT be used on manually-generated messages,
2970 - MAY be used on Delivery Status Notifications (DSNs) and Message
2971 Disposition Notifications (MDNs),
2973 - MUST NOT be used on messages generated by automatic or periodic
2974 processes, except for messages which are automatic responses to
2977 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
2978 to the default Received: header definition.
2980 PH/12 Added log selector acl_warn_skipped (default on).
2982 PH/13 After a successful wildlsearch lookup, discard the values of numeric
2983 variables because (a) they are in the wrong storage pool and (b) even if
2984 they were copied, it wouldn't work properly because of the caching.
2986 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
2987 checking when decoding. Apparently there are clients that generate
2988 overlong encoded strings. Why am I not surprised?
2990 PH/15 If the first argument of "${if match_address" was not empty, but did not
2991 contain an "@" character, Exim crashed. Now it writes a panic log message
2992 and treats the condition as false.
2994 PH/16 In autoreply, treat an empty string for "once" the same as unset.
2996 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
2997 extension "envelope-auth". The code is finished and in agreement with
2998 other implementations, but there is no documentation so far and in fact,
2999 nobody wrote the draft yet. This extension is currently #undef'ed, thus
3000 not changing the active code.
3002 Print executed "if" and "elsif" statements when debugging is used. This
3003 helps a great deal to understand what a filter does.
3005 Document more things not specified clearly in RFC3028. I had all this
3006 sorted out, when out of a sudden new issues came to my mind. Oops."
3008 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
3011 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
3012 canonical form (as documented). However, after a host name lookup from
3013 the IP address, check_host() was doing a simple string comparison with
3014 addresses acquired from the DNS when checking that the found name did
3015 have the original IP as one of its addresses. Since any found IPv6
3016 addresses are likely to be in abbreviated form, the comparison could
3017 fail. Luckily, there already exists a function for doing the comparison
3018 by converting both addresses to binary, so now that is used instead of
3019 the text comparison.
3021 PH/20 There was another similar case to PH/19, when a complete host name was
3022 given in a host list; looking up its IP address could give an abbreviated
3023 form, whereas the current host's name might or might not be abbreviated.
3024 The same fix has been applied.
3030 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
3031 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
3034 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
3035 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
3037 PH/03 Typo: missing ".o" in src/pcre/Makefile.
3039 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
3040 header line, restrict the check to what is listed in RFCs 2369 and 2929.
3041 Also, for "Auto-Submitted", treat anything other than "no" as
3042 non-personal, in accordance with RFC 3834. (Previously it treated
3043 anything starting "auto-" as non-personal.)
3045 TF/01 The control=submission/name=... option had a problem with syntax
3046 errors if the name included a slash character. The /name= option
3047 now slurps the rest of the string, so it can include any characters
3048 but it must come last in the list of options (after /sender_retain
3051 PH/05 Some modifications to the interface to the fake nameserver for the new
3059 TK/01 Added the "success_on_redirect" address verification option. See
3060 NewStuff for rationale and an example.
3062 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
3064 PH/02 Patch to exigrep to allow it to work on syslog lines.
3066 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
3067 fread() to skip over the body file's header line, because in Cygwin the
3068 header line is locked and is inaccessible.
3070 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
3071 co-exist for some time) to make it clear that it is the Exim ID that is
3072 referenced, not the Message-ID: header line.
3074 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
3075 string_format() function, because snprintf() does not exist on all
3078 PH/06 The use of forbid_filter_existstest now also locks out the use of the
3079 ${stat: expansion item.
3081 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
3082 protocol synchronization error", to keep the pedants happy.
3084 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
3085 well as for IRIX systems, when gcc is being used. See the host.c source
3088 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
3090 PH/10 Named domain lists were not working if used in a queue_smtp_domains
3093 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
3094 transport and to the smtp transport in LMTP mode.
3096 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
3098 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
3099 run a filter in a subprocess. This could lead to confusion in subsequent
3100 lookups in the parent process. There should also be a search_tidyup() at
3101 the end of the subprocess.
3103 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
3104 only if the host matched helo_try_verify_hosts, which caused the
3105 verification to occur when the EHLO/HELO command was issued. The ACL just
3106 tested the remembered result. Now, if a previous verification attempt has
3107 not happened, "verify = helo" does it there and then.
3109 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
3111 TK/03 Fix log output including CR from clamd.
3113 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
3114 exist provoked a memory error which could cause a segfault.
3116 PH/15 Installed PCRE 6.2
3118 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
3120 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
3121 of the problem. Specifically, suggested +O2 rather than +O1 for the
3124 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
3126 PH/20 If a delivery was routed to a non-standard port by means of an SRV
3127 record, the port was not correctly logged when the outgoing_port log
3128 selector was set (it logged the transort's default port).
3130 PH/21 Added support for host-specific ports to manualroute, queryprogram,
3131 fallback_hosts, and "hosts" in the smtp transport.
3133 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
3134 host errors such as "Connection refused".
3136 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
3137 authentication with radiusclient 0.4.9:
3139 - Error returned from rc_read_config was caught wrongly
3140 - Username/password not passed on to radius server due to wrong length.
3142 The presumption is that some radiusclient API changes for 4.51/PH/17
3143 were not taken care of correctly. The code is still untested by me (my
3144 Linux distribution still has 0.3.2 of radiusclient), but it was
3145 contributed by a Radius user.
3147 PH/24 When doing a callout, the value of $domain wasn't set correctly when
3148 expanding the "port" option of the smtp transport.
3150 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
3151 while reading a MIME header. Thanks to Tom Hughes for a patch.
3153 PH/24 Include config.h inside local_scan.h so that configuration settings are
3156 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
3157 that in an ACL for RCPT (for example), you can examine exactly what was
3160 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
3161 commands, but it was not correctly comparing the address with the actual
3162 client host address. Thus, it would show the EHLO address in Received:
3163 header lines when this was not necessary.
3165 PH/27 Added the % operator to ${eval:}.
3167 PH/28 Exim tries to create and chdir to its spool directory when it starts;
3168 it should be ignoring failures (because with -C, for example, it has lost
3169 privilege). It wasn't ignoring creation failures other than "already
3172 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
3175 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
3176 because an input line was too long, either on its own, or by virtue of
3177 too many continuations, the temporary file was not being removed, and the
3178 return code was incorrect.
3180 PH/31 Missing "BOOL" in function definition in filtertest.c.
3182 PH/32 Applied Sieve patches from the maintainer.
3184 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
3186 PH/33 Added "verify = not_blind".
3188 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
3189 Local/Makefile (with some defaults set). These are used in built scripts
3190 such as exicyclog, but they have never been used in the exim_install
3191 script (though there are many overriding facilities there). I have
3192 arranged that the exim_install script now takes note of these two
3195 PH/35 Installed configuration files for Dragonfly.
3197 PH/36 When a locally submitted message by a trusted user did not contain a
3198 From: header, and the sender address was obtained from -f or from an SMTP
3199 MAIL command, and the trusted user did not use -F to supply a sender
3200 name, $originator_name was incorrectly used when constructing a From:
3201 header. Furthermore, $originator_name was used for submission mode
3202 messages from external hosts without From: headers in a similar way,
3203 which is clearly wrong.
3205 PH/37 Added control=suppress_local_fixups.
3207 PH/38 When log_selector = +received_sender was set, and the addition of the
3208 sender made the log line's construction buffer exactly full, or one byte
3209 less than full, an overflow happened when the terminating "\n" was
3212 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
3213 when the result of a list match is failure because a DNS lookup failed.
3215 PH/40 RM_COMMAND is now used in the building process.
3217 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
3218 the "build-* directories that it finds.
3220 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
3221 address in a domain literal was a prefix of an interface address.
3223 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
3224 when verifying a sender address, unless rewrite_headers is false.
3226 PH/44 Wrote a long comment about why errors_to addresses are verified as
3227 recipients, not senders.
3229 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
3230 the ratelimit ACL was added.
3232 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
3234 PH/46 Added extra information about PostgreSQL errors to the error string.
3236 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
3237 avoiding the need to install special zones in a real server. This is
3238 backwards compatible; if it can't find the fake resolver, it drops back.
3239 Thus, both old and new test suites can be run.
3241 TF/02 Added util/ratelimit.pl
3243 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
3244 clock is set back in time.
3246 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
3247 Candler <B.Candler@pobox.com>.
3249 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
3250 OK for addresses that are the result of redirections.
3252 PH/48 A number of further additions for the benefit of the new test suite,
3253 including a fake gethostbyname() that interfaces to the fake DNS resolver
3256 TF/06 The fix for widen_domains has also been applied to qualify_single and
3257 search_parents which are the other dnslookup options that can cause
3260 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
3263 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
3265 TF/07 Exim produced the error message "an SRV record indicated no SMTP
3266 service" if it encountered an MX record with an empty target hostname.
3267 The message is now "an MX or SRV record indicated no SMTP service".
3269 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
3270 if the DNS of the sending site is misconfigured. This is quite a
3271 common situation. This change restores the behaviour of treating a
3272 helo verification defer as a failure.
3274 PH/51 If self=fail was set on a router, the bounce message did not include the
3275 actual error message.
3281 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
3283 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
3284 command itself ended in error, the underlying message about the transport
3285 filter timeout was being overwritten with the pipe command error. Now the
3286 underlying error message should be appended to the second error message.
3288 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
3290 PH/02 Reduce the amount of output that "make" produces by default. Full output
3291 can still be requested.
3293 PH/03 The warning log line about a condition test deferring for a "warn" verb
3294 was being output only once per connection, rather than after each
3295 occurrence (because it was using the same function as for successful
3296 "warn" verbs). This seems wrong, so I have changed it.
3298 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
3299 it should not have, which might have caused a crash in the right
3300 circumstances, but probably never did.
3302 PH/04 Installed a modified version of Tony Finch's patch to make submission
3303 mode fix the return path as well as the Sender: header line, and to
3304 add a /name= option so that you can make the user's friendly name appear
3307 TF/03 Added the control = fakedefer ACL modifier.
3309 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
3310 Mark Lowes for thorough testing.
3312 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
3314 TK/03 Merged latest SRS patch from Miles Wilton.
3316 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
3317 with the definition in sysexits.h (which is #included earlier).
3318 Fortunately, Exim does not actually use EX_OK. The code used to try to
3319 preserve the sysexits.h value, by assumimg that macro definitions were
3320 scanned for macro replacements. I have been disabused of this notion,
3321 so now the code just undefines EX_OK before #including unistd.h.
3323 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
3324 in the smtp transport. When a block could not be written in a single
3325 write() function, the timeout was being re-applied to each part-write.
3326 This seems wrong - if the receiver was accepting one byte at a time it
3327 would take for ever. The timeout is now adjusted when this happens. It
3328 doesn't have to be particularly precise.
3330 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
3331 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
3333 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
3334 without a domain if the check to <postmaster@domain> fails.
3336 SC/01 Eximstats: added -xls and the ability to specify output files
3337 (patch written by Frank Heydlauf).
3339 SC/02 Eximstats: use FileHandles for outputing results.
3341 SC/03 Eximstats: allow any combination of xls, txt, and html output.
3343 SC/04 Eximstats: fixed display of large numbers with -nvr option
3345 SC/05 Eximstats: fixed merging of reports with empty tables.
3347 SC/06 Eximstats: added the -include_original_destination flag
3349 SC/07 Eximstats: removed tabs and trailing whitespace.
3351 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
3353 TK/06 MBOX spool code: Add real "From " MBOX separator line
3354 so the .eml file is really in mbox format (even though
3355 most programs do not really care). Patch from Alex Miller.
3357 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
3358 The latter is generated from $received_to and is only set if the
3359 message has one envelope recipient. SA can use these headers,
3360 obviously out-of-the-box. Patch from Alex Miller.
3362 PH/08 The ${def test on a variable was returning false if the variable's
3363 value was "0", contrary to what the specification has always said!
3364 The result should be true unless the variable is empty.
3366 PH/09 The syntax error of a character other than { following "${if
3367 def:variable_name" (after optional whitespace) was not being diagnosed.
3368 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
3369 accidental colon was present, for example, could give incorrect results.
3371 PH/10 Tidied the code in a number of places where the st_size field of a stat()
3372 result is used (not including appendfile, where other changes are about
3375 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
3376 This involved changing a lot of size variables from int to off_t. It
3377 should work with maildirs and everything.
3379 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
3380 spamd dying while we are connected to it.
3382 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
3385 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
3386 function for BDB 4.3.
3388 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
3390 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
3391 the sources for PCRE have been re-arranged and more files are now
3394 PH/15 The code I had for printing potentially long long variables in PH/11
3395 above was not the best (it lost precision). The length of off_t variables
3396 is now inspected at build time, and an appropriate printing format (%ld
3397 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
3398 to be "long long int" or "long int". This is needed for the internal
3399 formatting function string_vformat().
3401 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
3402 the configuration file to be ":syslog", then the script "guesses" where
3403 the logs files are, rather than using the compiled in default. In our
3404 case the guess is not the same as the compiled default, so the script
3405 suddenly stopped working when I started to use syslog. The patch checks
3406 to see if log_file_path is "". If so, it attempts to read it from exim
3407 with no configuration file to get the compiled in version, before it
3408 falls back to the previous guessing code."
3410 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
3411 implementing BATV in an Exim configuration. See NewStuff for the gory
3414 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
3415 Makefile that are specific to HP-UX.
3417 PH/18 If the "use_postmaster" option was set for a recipient callout together
3418 with the "random" option, the postmaster address was used as the MAIL
3419 FROM address for the random test, but not for the subsequent recipient
3420 test. It is now used for both.
3422 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
3423 patch removes a few documentation additions to RFC 3028, because the
3424 latest draft now contains them. It adds the new en;ascii-case comparator
3425 and a new error check for 8bit text in MIME parts. Comparator and
3426 require names are now matched exactly. I enabled the subaddress
3427 extension, but it is not well tested yet (read: it works for me)."
3429 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
3430 rework some of the code of TK/09 above to avoid the hardwired use of
3431 "%lld" and "long long". Replaced the call to snprintf() with a call to
3434 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
3435 records point to non-existent hosts", "retry timeout exceeded", and
3436 "retry time not reached for any host after a long failure period".
3438 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
3439 experimental DomainKeys support:
3441 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
3442 (2) On an error such as an illegally used "control", the wrong name for
3443 the control was given.
3445 These problems did NOT occur unless DomainKeys support was compiled.
3447 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
3449 PH/24 Added ${if match_ip condition.
3451 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
3452 databases so that it will be absolutely obvious if a crash occurs in the
3453 DB library. This is a regular occurrence (often caused by mis-matched
3456 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
3457 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
3458 picked up on a user's system that detects such things. There doesn't seem
3459 to be a gcc warning option for this - only an attribute that has to be
3460 put on the function's prototype. It seems that in Fedora Core 4 they have
3461 set this on a number of new functions. No doubt there will be more in due
3464 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
3465 specify a transport. However, if an address that was verified by such a
3466 router was the subject of a callout, Exim crashed because it tried to
3467 read the rcpt_include_affixes from the non-existent transport. Now it
3468 just assumes that the setting of that option is false. This bug was
3469 introduced by 4.51/PH/31.
3471 PH/28 Changed -d+all to exclude +memory, because that information is very
3472 rarely of interest, but it makes the output a lot bigger. People tend to
3473 do -d+all out of habit.
3475 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
3476 code in os-type was giving problems when libc.so lives in lib64, like on
3479 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
3480 aren't the modern standard, and it seems that some systems' include files
3481 don't always have them. Exim was already checking for some of the newer
3482 ones like T_AAAA, and defining it itself. I've added checks for all the
3483 record types that Exim uses.
3485 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
3486 not automatically generating a new one, as it is supposed to. This
3487 prevented TLS from working. If the file did exist, but contained invalid
3488 data, a new version was generated, as expected. It was only the case of a
3489 non-existent file that was broken.
3491 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
3492 with a change in libdomainkeys > 0.64.
3494 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
3495 from DNS. If the selector record carries the flag, it now has
3496 precedence over the domain-wide flag.
3498 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
3500 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
3501 the use of a port name that isn't defined in /etc/services) occurred, the
3502 message was deferred as in a normal delivery, and thus remained on the
3503 spool, instead of being failed because of the mua_wrapper setting. This
3504 is now fixed, and I tidied up some of the mua_wrapper messages at the
3507 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
3508 the messages in a hash of arrays rather than using individual hashes.
3509 This is a bit cleaner and results in dramatic memory savings, albeit
3510 at a slight CPU cost.
3512 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
3513 as requested by Marc Sherman.
3515 SC/10 Eximstats: added histograms for user specified patterns as requested
3518 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
3520 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
3521 fopen() in the content-scanning modules that did not already have it.
3527 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
3528 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
3530 TK/02 Fix ACL "control" statement not being available in MIME ACL.
3532 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
3534 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
3535 to test Sieve filters that use "vacation".
3537 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
3538 that changes the way the GnuTLS parameters are stored in the cache file.
3539 The new format can be generated externally. For backward compatibility,
3540 if the data in the cache doesn't make sense, Exim assumes it has read an
3541 old-format file, and it generates new data and writes a new file. This
3542 means that you can't go back to an older release without removing the
3545 PH/03 A redirect router that has both "unseen" and "one_time" set does not
3546 work if there are any delivery delays because "one_time" forces the
3547 parent to be marked "delivered", so its unseen clone is never tried
3548 again. For this reason, Exim now forbids the simultaneous setting of
3551 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
3552 redirected to themselves ("homonym" addresses). Read the long ChangeLog
3553 entry if you want to know the details. The fix, however, neglected to
3554 consider the case when local delivery batching is involved. The test for
3555 "previously delivered" was not happening when checking to see if an
3556 address could be batched with a previous (undelivered) one; under
3557 certain circumstances this could lead to multiple deliveries to the same
3560 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
3561 in its include files, and this causes problems building Exim.
3563 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
3564 header_syntax) but Exim was just ignoring anything given after a slash.
3565 In particular, this caused confusion with an attempt to use "verify =
3566 reverse_host_lookup/defer_ok". An error is now given when options are
3567 supplied for verify items that do not have them. (Maybe reverse_host_
3568 lookup should have a defer_ok option, but that's a different point.)
3570 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
3571 defined by RFC 821) to 2048, because there were problems with some AUTH
3572 commands, and RFC 1869 says the size should be increased for extended
3573 SMTP commands that take arguments.
3575 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
3578 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
3579 "unknown" error; now it says that the functionality isn't in the binary.
3581 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
3582 an address' error message when a string expansion fails (syntax or
3583 whatever). Otherwise the password may appear in the log. Following change
3584 PH/42 below, there is no longer a chance of it appearing in a bounce
3587 PH/11 Installed exipick version 20050225.0 from John Jetmore.
3589 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
3590 of its addresses was ever tried. (Bugzilla bug #2.)
3592 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
3593 the result incorrectly in the debug output. (It correctly added a newline
3594 to what was transported.)
3596 TF/01 Added $received_time.
3598 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
3599 commented out examples of how to interface to a virus scanner and to
3600 SpamAssassin. Also added commented examples of av_scanner and
3601 spamd_address settings.
3603 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
3604 and controls are allowed in which ACLs. There were a couple of minor
3605 errors. Some of the entries in the conditions table (which is a table of
3606 where they are NOT allowed) were getting very unwieldy; rewrote them as a
3607 negation of where the condition IS allowed.
3609 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
3611 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
3612 header file does not have a version number, so I've had to invent a new
3613 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
3614 API. The code is untested by me (my Linux distribution still has 0.3.2 of
3615 radiusclient), but it was contributed by a Radius user.
3617 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
3618 files or directories, for OpenSSL.
3620 PH/19 When an Exim process that is running as root has to create an Exim log
3621 file, it does so in a subprocess that runs as exim:exim so as to get the
3622 ownership right at creation (otherwise, other Exim processes might see
3623 the file with the wrong ownership). There was no test for failure of this
3624 fork() call, which would lead to the process getting stuck as it waited
3625 for a non-existent subprocess. Forks do occasionally fail when resources
3626 run out. I reviewed all the other calls to fork(); they all seem to check
3629 PH/20 When checking for unexpected SMTP input at connect time (before writing
3630 the banner), Exim was not dealing correctly with a non-positive return
3631 from the read() function. If the client had disconnected by this time,
3632 the result was a log entry for a synchronization error with an empty
3633 string after "input=" when read() returned zero. If read() returned -1
3634 (an event I could not check), uninitialized data bytes were printed.
3635 There were reports of junk text (parts of files, etc) appearing after
3638 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
3640 PH/22 Added support for macro redefinition, and (re)definition in between
3641 driver and ACL definitions.
3643 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
3644 forgetting to use the resulting value; it was using the unexpanded value.
3646 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
3647 hadn't been configured. The fix is from Juergen Kreileder, who
3648 understands it better than I do:
3650 "Here's what I see happening with three configured cyrus_sasl
3651 authenticators configured (plain, login, cram-md5):
3653 On startup auth_cyrus_sasl_init() gets called for each of these.
3654 This means three calls to sasl_listmech() without a specified mech_list.
3655 => SASL tests which mechs of all available mechs actually work
3656 => three warnings about OTP not working
3657 => the returned list contains: plain, login, cram-md5, digest-md5, ...
3659 With the patch, sasl_listmech() also gets called three times. But now
3660 SASL's mech_list option is set to the server_mech specified in the the
3661 authenticator. Or in other words, the answer from sasl_listmech()
3662 gets limited to just the mech you're testing for (which is different
3664 => the return list contains just 'plain' or 'login', 'cram-md5' or
3665 nothing depending on the value of ob->server_mech.
3667 I've just tested the patch: Authentication still works fine,
3668 unavailable mechs specified in the exim configuration are still
3669 caught, and the auth.log warnings about OTP are gone."
3671 PH/25 When debugging is enabled, the contents of the command line are added
3672 to the debugging output, even when log_selector=+arguments is not
3675 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
3676 answer is "GNU", and only if the return is "GNU/something" is the answer
3679 PH/27 $acl_verify_message is now set immediately after the failure of a
3680 verification in an ACL, and so is available in subsequent modifiers. In
3681 particular, the message can be preserved by coding like this:
3683 warn !verify = sender
3684 set acl_m0 = $acl_verify_message
3686 Previously, $acl_verify_message was set only while expanding "message"
3687 and "log_message" when a very denied access.
3689 PH/28 Modified OS/os.c-Linux with
3691 -#ifndef OS_LOAD_AVERAGE
3692 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
3694 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
3695 nomenclature these days.)
3697 PH/29 Installed patch from the Sieve maintainer that adds the options
3698 sieve_useraddress and sieve_subaddress to the redirect router.
3700 PH/30 In these circumstances:
3701 . Two addresses routed to the same list of hosts;
3702 . First host does not offer TLS;
3703 . First host accepts first address;
3704 . First host gives temporary error to second address;
3705 . Second host offers TLS and a TLS session is established;
3706 . Second host accepts second address.
3707 Exim incorrectly logged both deliveries with the TLS parameters (cipher
3708 and peerdn, if requested) that were in fact used only for the second
3711 PH/31 When doing a callout as part of verifying an address, Exim was not paying
3712 attention to any local part prefix or suffix that was matched by the
3713 router that accepted the address. It now behaves in the same way as it
3714 does for delivery: the affixes are removed from the local part unless
3715 rcpt_include_affixes is set on the transport.
3717 PH/32 Add the sender address, as F=<...>, to the log line when logging a
3718 timeout during the DATA phase of an incoming message.
3720 PH/33 Sieve envelope tests were broken for match types other than :is. I have
3721 applied a patch sanctioned by the Sieve maintainer.
3723 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
3724 the uid or gid is negative. A case of a negative gid caused this to be
3725 noticed. The fix allows for either to be negative.
3727 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
3728 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
3731 PH/36 The change PH/12 above was broken. Fixed it.
3733 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
3734 the grounds that routing the same address twice would always produce the
3735 same answer. This might have been true once, but it is certainly no
3736 longer true now. Routing a child address may depend on the previous
3737 routing that produced that child. Some complicated redirection strategies
3738 went wrong when messages had multiple recipients, and made Exim's
3739 behaviour dependent on the order in which the addresses were given.
3741 I have moved the duplicate checking until after the routing is complete.
3742 Exim scans the addresses that are assigned to local and remote
3743 transports, and removes any duplicates. This means that more work will be
3744 done, as duplicates will always all be routed, but duplicates are
3745 presumably rare, so I don't expect this is of any significance.
3747 For deliveries to pipes, files, and autoreplies, the duplicate checking
3748 still happens during the routing process, since they are not going to be
3751 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
3752 It corrects a timeout issue with spamd. This is Ian's comment: "The
3753 background is that sometimes spamd either never reads data from a
3754 connection it has accepted, or it never writes response data. The exiscan
3755 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
3756 blindly assumes that writes won't block so it may never time out."
3758 PH/39 Allow G after quota size as well as K and M.
3760 PH/40 The value set for $authenticated_id in an authenticator may not contain
3761 binary zeroes or newlines because the value is written to log lines and
3762 to spool files. There was no check on this. Now the value is run through
3763 the string_printing() function so that such characters are converted to
3764 printable escape sequences.
3766 PH/41 $message_linecount is a new variable that contains the total number of
3767 lines in the message. Compare $body_linecount, which is the count for the
3770 PH/42 Exim no longer gives details of delivery errors for specific addresses in
3771 bounce and delay warning messages, except in certain special cases, which
3774 (a) An SMTP error message from a remote host;
3775 (b) A message specified in a :fail: redirection;
3776 (c) A message specified in a "fail" command in a system filter;
3777 (d) A message specified in a FAIL return from the queryprogram router;
3778 (e) A message specified by the cannot_route_message router option.
3780 In these cases only, Exim does include the error details in bounce and
3781 warning messages. There are also a few cases where bland messages such
3782 as "unrouteable address" or "local delivery error" are given.
3784 PH/43 $value is now also set for the "else" part of a ${run expansion.
3786 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
3787 being worked on, but at least Exim now implements the latest version to
3790 PH/45 In a pipe transport, although a timeout while waiting for the pipe
3791 process to complete was treated as a delivery failure, a timeout while
3792 writing the message to the pipe was logged, but erroneously treated as a
3793 successful delivery. Such timeouts include transport filter timeouts. For
3794 consistency with the overall process timeout, these timeouts are now
3795 treated as errors, giving rise to delivery failures by default. However,
3796 there is now a new Boolean option for the pipe transport called
3797 timeout_defer, which, if set TRUE, converts the failures into defers for
3798 both kinds of timeout. A transport filter timeout is now identified in
3801 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
3802 systems where "make" and "gmake" are different, calling "gmake" at top
3803 level broke things. I've arranged for the value of $(MAKE) to be passed
3804 from the Makefile to this script so that it can call the same version of
3808 A note about Exim versions 4.44 and 4.50
3809 ----------------------------------------
3811 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
3812 changes of various kinds. As a consequence, a big documentation update was
3813 needed. This delayed the release for rather longer than seemed good, especially
3814 in the light of a couple of (minor) security issues. Therefore, the changes
3815 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
3816 release. So 4.44 and 4.50 are in effect two different branches that both start
3819 I have left the 4.50 change log unchanged; it contains all the changes since
3820 4.43. The change log for 4.44 is below; many of its items are identical to
3821 those for 4.50. This seems to be the most sensible way to preserve the
3822 historical information.
3828 1. Minor wording change to the doc/README.SIEVE file.
3830 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
3831 computation of the current number of files was incorrect.
3833 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
3834 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
3837 4. Give more explanation in the error message when the command for a transport
3838 filter fails to execute.
3840 5. There are several places where Exim runs a non-Exim command in a
3841 subprocess. The SIGUSR1 signal should be disabled for these processes. This
3842 was being done only for the command run by the queryprogram router. It is
3843 now done for all such subprocesses. The other cases are: ${run, transport
3844 filters, and the commands run by the lmtp and pipe transports.
3846 6. Added CONFIGURE_GROUP build-time option.
3848 7. Some older OS have a limit of 256 on the maximum number of file
3849 descriptors. Exim was using setrlimit() to set 1000 as a large value
3850 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
3851 systems. I've change it so that if it can't get 1000, it tries for 256.
3853 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
3854 was an oversight, and furthermore, ever since the addition of extra
3855 controls (e.g. 4.43/32), the checks on when to allow different forms of
3856 "control" were broken. There should now be diagnostics for all cases when a
3857 control that does not make sense is encountered.
3859 9. Added the /retain_sender option to "control=submission".
3861 10. $recipients is now available in the predata ACL (oversight).
3863 11. Tidy the search cache before the fork to do a delivery from a message
3864 received from the command line. Otherwise the child will trigger a lookup
3865 failure and thereby defer the delivery if it tries to use (for example) a
3866 cached ldap connection that the parent has called unbind on.
3868 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
3869 of $address_data from the recipient verification was clobbered by the
3870 sender verification.
3872 13. The value of address_data from a sender verification is now available in
3873 $sender_address_data in subsequent conditions in the ACL statement.
3875 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
3877 15. Added a new option "connect=<time>" to callout options, to set a different
3880 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
3881 was its contents. (It was OK if the option was not defined at all.)
3883 17. A "Completed" log line is now written for messages that are removed from
3884 the spool by the -Mrm option.
3886 18. New variables $sender_verify_failure and $recipient_verify_failure contain
3887 information about exactly what failed.
3889 19. Added -dd to debug only the daemon process.
3891 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
3892 handles timeouts, both on the server side and network timeouts. Renamed the
3893 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
3895 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
3896 It is now set to "smtps".
3898 22. $host_address is now set to the target address during the checking of
3899 ignore_target_hosts.
3901 23. When checking ignore_target_hosts for an ipliteral router, no host name was
3902 being passed; this would have caused $sender_host_name to have been used if
3903 matching the list had actually called for a host name (not very likely,
3904 since this list is usually IP addresses). A host name is now passed as
3907 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
3908 code that specifies a non-restarting handler (typically sigaction() in
3909 modern systems) in an attempt to fix a rare and obscure crash bug.
3911 25. Narrowed the window for a race in the daemon that could cause it to ignore
3912 SIGCHLD signals. This is not a major problem, because they are used only to
3913 wake it up if nothing else does.
3915 26. A malformed maildirsize file could cause Exim to calculate negative values
3916 for the mailbox size or file count. Odd effects could occur as a result.
3917 The maildirsize information is now recalculated if the size or filecount
3920 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
3921 support for a long time. Removed HAVE_SYS_VFS_H.
3923 28. Installed the latest version of exipick from John Jetmore.
3925 29. In an address list, if the pattern was not a regular expression, an empty
3926 subject address (from a bounce message) matched only if the pattern was an
3927 empty string. Non-empty patterns were not even tested. This was the wrong
3928 because it is perfectly reasonable to use an empty address as part of a
3929 database query. An empty address is now tested by patterns that are
3930 lookups. However, all the other forms of pattern expect the subject to
3931 contain a local part and a domain, and therefore, for them, an empty
3932 address still always fails if the pattern is not itself empty.
3934 30. Exim went into a mad DNS loop when attempting to do a callout where the
3935 host was specified on an smtp transport, and looking it up yielded more
3936 than one IP address.
3938 31. Re-factored the code for checking spool and log partition space into a
3939 function that finds that data and another that does the check. The former
3940 is then used to implement four new variables: $spool_space, $log_space,
3941 $spool_inodes, and $log_inodes.
3943 32. The RFC2047 encoding function was originally intended for short strings
3944 such as real names; it was not keeping to the 75-character limit for
3945 encoded words that the RFC imposes. It now respects the limit, and
3946 generates multiple encoded words if necessary. To be on the safe side, I
3947 have increased the buffer size for the ${rfc2047: expansion operator from
3950 33. It is now permitted to omit both strings after an "if" condition; if the
3951 condition is true, the result is "true". As before, when the second string
3952 is omitted, a false condition yields an empty string. This makes it less
3953 cumbersome to write custom ACL and router conditions.
3955 34. Failure to deliver a bounce message always caused it to be frozen, even if
3956 there was an errors_to setting on the router. The errors_to setting is now
3959 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
3960 canonical form (fully expanded) before being placed in
3961 $sender_host_address.
3963 36. The table in the code that translates DNS record types into text (T_A to
3964 "A" for instance) was missing entries for NS and CNAME. It is just possible
3965 that this could have caused confusion if both these types were looked up
3966 for the same domain, because the text type is used as part of Exim's
3967 per-process caching. But the chance of anyone hitting this buglet seems
3970 37. The dnsdb lookup has been extended in a number of ways.
3972 (1) There is a new type, "zns", which walks up the domain tree until it
3973 finds some nameserver records. It should be used with care.
3975 (2) There is a new type, "mxh", which is like "mx" except that it returns
3976 just the host names, not the priorities.
3978 (3) It is now possible to give a list of domains (or IP addresses) to be
3979 looked up. The behaviour when one of the lookups defers can be
3980 controlled by a keyword.
3982 (4) It is now possible to specify the separator character for use when
3983 multiple records are returned.
3985 38. The dnslists ACL condition has been extended: it is now possible to supply
3986 a list of IP addresses and/or domains to be looked up in a particular DNS
3989 39. Added log_selector=+queue_time_overall.
3991 40. When running the queue in the test harness, wait just a tad after forking a
3992 delivery process, to get repeatability of debugging output.
3994 41. Include certificate and key file names in error message when GnuTLS fails
3995 to set them up, because the GnuTLS error message doesn't include the name
3996 of the failing file when there is a problem reading it.
3998 42. Allow both -bf and -bF in the same test run.
4000 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
4002 44. The "Exiscan patch" is now merged into the mainline Exim source.
4004 45. Sometimes the final signoff response after QUIT could fail to get
4005 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4006 < 0 before doing a fflush(). This bug looks as though it goes back to the
4007 introduction of TLS in release 3.20, but "sometimes" must have been rare
4008 because the tests only now provoked it.
4010 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
4011 (this can affect the format of dates).
4013 47. exim_tidydb, when checking for the continued existence of a message for
4014 which it has found a message-specific retry record, was not finding
4015 messages that were in split spool directories. Consequently, it was
4016 deleting retry records that should have stayed in existence.
4018 48. Steve fixed some bugs in eximstats.
4020 49. The SPA authentication driver was not abandoning authentication and moving
4021 on to the next authenticator when an expansion was forced to fail,
4022 contradicting the general specification for all authenticators. Instead it
4023 was generating a temporary error. It now behaves as specified.
4025 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
4026 (the order specifies the preference for clients). The order is now AES256,
4027 AES128, 3DES, ARCFOUR128.
4029 51. Small patch to Sieve code - explicitly set From: when generating an
4032 52. Exim crashed if a remote delivery caused a very long error message to be
4033 recorded - for instance if somebody sent an entire SpamAssassin report back
4034 as a large number of 550 error lines. This bug was coincidentally fixed by
4035 increasing the size of one of Exim's internal buffers (big_buffer) that
4036 happened as part of the Exiscan merge. However, to be on the safe side, I
4037 have made the code more robust (and fixed the comments that describe what
4040 53. Now that there can be additional text after "Completed" in log lines (if
4041 the queue_time_overall log selector is set), a one-byte patch to exigrep
4042 was needed to allow it to recognize "Completed" as not the last thing in
4045 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
4046 patch that reportedly fixes this has been added. I am not expert enough to
4047 create a test for it. This is what the patch creator wrote:
4049 "I found a little strange behaviour of ldap code when working with
4050 Windows 2003 AD Domain, where users was placed in more than one
4051 Organization Units. When I tried to give exim partial DN, the exit code
4052 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
4053 But simultaneously result of request was absolutely normal ldap result,
4054 so I produce this patch..."
4056 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
4057 REFERENCE, so I have modified the code to exclude the patch when that macro
4060 55. Some experimental protocols are using DNS PTR records for new purposes. The
4061 keys for these records are domain names, not reversed IP addresses. The
4062 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4063 leaves it alone. Component reversal etc. now happens only for IP addresses.
4066 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
4068 57. Double the size of the debug message buffer (to 2048) so that more of very
4069 long debug lines gets shown.
4071 58. The exicyclog utility now does better if the number of log files to keep
4072 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
4074 59. Two changes related to the smtp_active_hostname option:
4076 (1) $smtp_active_hostname is now available as a variable.
4077 (2) The default for smtp_banner uses $smtp_active_hostname instead
4078 of $primary_hostname.
4080 60. The host_aton() function is supposed to be passed a string that is known
4081 to be a valid IP address. However, in the case of IPv6 addresses, it was
4082 not checking this. This is a hostage to fortune. Exim now panics and dies
4083 if the condition is not met. A case was found where this could be provoked
4084 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4085 components; fortuitously, this particular loophole had already been fixed
4086 by change 4.50/55 above.
4088 If there are any other similar loopholes, the new check in host_aton()
4089 itself should stop them being exploited. The report I received stated that
4090 data on the command line could provoke the exploit when Exim was running as
4091 exim, but did not say which command line option was involved. All I could
4092 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4093 running as the user.
4096 61. There was a buffer overflow vulnerability in the SPA authentication code
4097 (which came originally from the Samba project). I have added a test to the
4098 spa_base64_to_bits() function which I hope fixes it.
4101 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
4102 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
4104 63. The daemon start-up calls getloadavg() while still root for those OS that
4105 need the first call to be done as root, but it missed one case: when
4106 deliver_queue_load_max is set with deliver_drop_privilege. This is
4107 necessary for the benefit of the queue runner, because there is no re-exec
4108 when deliver_drop_privilege is set.
4110 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
4111 This has been fixed.
4113 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4114 was in use, was not putting the data itself into the right store pool;
4115 consequently, it could be overwritten for a subsequent message in the same
4116 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4119 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
4121 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
4122 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
4123 the function were treating the return as a boolean value, which happened to
4124 work because 0=false and not-0=true, but is not correct code.
4126 68. The host_aton() function was not handling scoped IPv6 addresses (those
4127 with, for example, "%eth0" on the end) correctly.
4129 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
4130 negated items (that is, ~something) in unsigned ints. Some compilers
4131 apparently mutter when there is no cast.
4133 70. If an address verification called from an ACL failed, and did not produce a
4134 user-specific message (i.e. there was only a "system" message), nothing was
4135 put in $acl_verify_message. In this situation, it now puts the system
4138 71. Change 4.23/11 added synchronization checking at the start of an SMTP
4139 session; change 4.31/43 added the unwanted input to the log line - except
4140 that it did not do this in the start of session case. It now does.
4142 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
4143 This is wrong and can cause the other end to generate a synchronization
4144 error if it is another Exim or anything else that does the synchronization
4145 check. A QUIT command is no longer sent after a timeout.
4147 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
4148 during host lookups.
4150 74. The defer_ok option of callout verification was not working if it was used
4151 when verifying addresses in header lines, that is, for this case:
4153 verify = header_sender/callout=defer_ok
4155 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
4156 those file descriptors could be used for SMTP connections. If anything
4157 wrote to stderr (the example that came up was "warn" in embedded Perl), it
4158 could be sent to the SMTP client, causing chaos. The daemon now opens
4159 stdin, stdout, and stderr to /dev/null when it puts itself into the
4162 76. Arrange for output from Perl's "warn" command to be written to Exim's main
4163 log by default. The user can override this with suitable Perl magic.
4165 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
4166 the log message when discard triggers, was not working for the DATA ACL or
4167 for the non-SMTP ACL.
4169 78. Error message wording change in sieve.c.
4171 79. If smtp_accept_max_per_host was set, the number of connections could be
4172 restricted to fewer than expected, because the daemon was trying to set up
4173 a new connection before checking whether the processes handling previous
4174 connections had finished. The check for completed processes is now done
4175 earlier. On busy systems, this bug wouldn't be noticed because something
4176 else would have woken the daemon, and it would have reaped the completed
4179 80. If a message was submitted locally by a user whose login name contained one
4180 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
4181 It caused a spool format error. I have fixed the spool reading code. A
4182 related problem was that the "from" clause in the Received: line became
4183 illegal because of the space(s). It is now covered by ${quote_local_part.
4185 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
4188 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
4190 83. Patch from Sieve maintainer to fix unterminated string problem in
4191 "vacation" handling.
4193 84. Some minor changes to the Linux configuration files to help with other
4194 OS variants using glibc.
4196 85. One more patch for Sieve to update vacation handling to latest spec.
4199 ----------------------------------------------------
4200 See the note above about the 4.44 and 4.50 releases.
4201 ----------------------------------------------------
4207 1. Change 4.43/35 introduced a bug that caused file counts to be
4208 incorrectly computed when quota_filecount was set in an appendfile
4211 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4212 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4215 3. Give more explanation in the error message when the command for a transport
4216 filter fails to execute.
4218 4. There are several places where Exim runs a non-Exim command in a
4219 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4220 was being done only for the command run by the queryprogram router. It is
4221 now done for all such subprocesses. The other cases are: ${run, transport
4222 filters, and the commands run by the lmtp and pipe transports.
4224 5. Some older OS have a limit of 256 on the maximum number of file
4225 descriptors. Exim was using setrlimit() to set 1000 as a large value
4226 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4227 systems. I've change it so that if it can't get 1000, it tries for 256.
4229 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4230 was an oversight, and furthermore, ever since the addition of extra
4231 controls (e.g. 4.43/32), the checks on when to allow different forms of
4232 "control" were broken. There should now be diagnostics for all cases when a
4233 control that does not make sense is encountered.
4235 7. $recipients is now available in the predata ACL (oversight).
4237 8. Tidy the search cache before the fork to do a delivery from a message
4238 received from the command line. Otherwise the child will trigger a lookup
4239 failure and thereby defer the delivery if it tries to use (for example) a
4240 cached ldap connection that the parent has called unbind on.
4242 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4243 of $address_data from the recipient verification was clobbered by the
4244 sender verification.
4246 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4247 was its contents. (It was OK if the option was not defined at all.)
4249 11. A "Completed" log line is now written for messages that are removed from
4250 the spool by the -Mrm option.
4252 12. $host_address is now set to the target address during the checking of
4253 ignore_target_hosts.
4255 13. When checking ignore_target_hosts for an ipliteral router, no host name was
4256 being passed; this would have caused $sender_host_name to have been used if
4257 matching the list had actually called for a host name (not very likely,
4258 since this list is usually IP addresses). A host name is now passed as
4261 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4262 code that specifies a non-restarting handler (typically sigaction() in
4263 modern systems) in an attempt to fix a rare and obscure crash bug.
4265 15. Narrowed the window for a race in the daemon that could cause it to ignore
4266 SIGCHLD signals. This is not a major problem, because they are used only to
4267 wake it up if nothing else does.
4269 16. A malformed maildirsize file could cause Exim to calculate negative values
4270 for the mailbox size or file count. Odd effects could occur as a result.
4271 The maildirsize information is now recalculated if the size or filecount
4274 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4275 support for a long time. Removed HAVE_SYS_VFS_H.
4277 18. Updated exipick to current release from John Jetmore.
4279 19. Allow an empty sender to be matched against a lookup in an address list.
4280 Previously the only cases considered were a regular expression, or an
4283 20. Exim went into a mad DNS lookup loop when doing a callout where the
4284 host was specified on the transport, if the DNS lookup yielded more than
4287 21. The RFC2047 encoding function was originally intended for short strings
4288 such as real names; it was not keeping to the 75-character limit for
4289 encoded words that the RFC imposes. It now respects the limit, and
4290 generates multiple encoded words if necessary. To be on the safe side, I
4291 have increased the buffer size for the ${rfc2047: expansion operator from
4294 22. Failure to deliver a bounce message always caused it to be frozen, even if
4295 there was an errors_to setting on the router. The errors_to setting is now
4298 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4299 canonical form (fully expanded) before being placed in
4300 $sender_host_address.
4302 24. Updated eximstats to version 1.33
4304 25. Include certificate and key file names in error message when GnuTLS fails
4305 to set them up, because the GnuTLS error message doesn't include the name
4306 of the failing file when there is a problem reading it.
4308 26. Expand error message when OpenSSL has problems setting up cert/key files.
4311 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
4312 (this can affect the format of dates).
4314 28. exim_tidydb, when checking for the continued existence of a message for
4315 which it has found a message-specific retry record, was not finding
4316 messages that were in split spool directories. Consequently, it was
4317 deleting retry records that should have stayed in existence.
4319 29. eximstats updated to version 1.35
4320 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
4321 1.35 - bugfix such that pie charts by volume are generated correctly
4323 30. The SPA authentication driver was not abandoning authentication and moving
4324 on to the next authenticator when an expansion was forced to fail,
4325 contradicting the general specification for all authenticators. Instead it
4326 was generating a temporary error. It now behaves as specified.
4328 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
4329 (the order specifies the preference for clients). The order is now AES256,
4330 AES128, 3DES, ARCFOUR128.
4332 31. Small patch to Sieve code - explicitly set From: when generating an
4335 32. Exim crashed if a remote delivery caused a very long error message to be
4336 recorded - for instance if somebody sent an entire SpamAssassin report back
4337 as a large number of 550 error lines. This bug was coincidentally fixed by
4338 increasing the size of one of Exim's internal buffers (big_buffer) that
4339 happened as part of the Exiscan merge. However, to be on the safe side, I
4340 have made the code more robust (and fixed the comments that describe what
4343 33. Some experimental protocols are using DNS PTR records for new purposes. The
4344 keys for these records are domain names, not reversed IP addresses. The
4345 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4346 leaves it alone. Component reversal etc. now happens only for IP addresses.
4349 34. The host_aton() function is supposed to be passed a string that is known
4350 to be a valid IP address. However, in the case of IPv6 addresses, it was
4351 not checking this. This is a hostage to fortune. Exim now panics and dies
4352 if the condition is not met. A case was found where this could be provoked
4353 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4354 components; fortuitously, this particular loophole had already been fixed
4355 by change 4.50/55 or 4.44/33 above.
4357 If there are any other similar loopholes, the new check in host_aton()
4358 itself should stop them being exploited. The report I received stated that
4359 data on the command line could provoke the exploit when Exim was running as
4360 exim, but did not say which command line option was involved. All I could
4361 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4362 running as the user.
4365 35. There was a buffer overflow vulnerability in the SPA authentication code
4366 (which came originally from the Samba project). I have added a test to the
4367 spa_base64_to_bits() function which I hope fixes it.
4370 36. The daemon start-up calls getloadavg() while still root for those OS that
4371 need the first call to be done as root, but it missed one case: when
4372 deliver_queue_load_max is set with deliver_drop_privilege. This is
4373 necessary for the benefit of the queue runner, because there is no re-exec
4374 when deliver_drop_privilege is set.
4376 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4377 was in use, was not putting the data itself into the right store pool;
4378 consequently, it could be overwritten for a subsequent message in the same
4379 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4382 38. Sometimes the final signoff response after QUIT could fail to get
4383 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4384 < 0 before doing a fflush(). This bug looks as though it goes back to the
4385 introduction of TLS in release 3.20, but "sometimes" must have been rare
4386 because the tests only now provoked it.
4392 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
4393 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
4394 Now it returns an integer. A number of calls were still expecting a T/F
4395 return. Fortuitously, in all cases, the tests worked in OK situations,
4396 which is the norm. However, things would have gone wrong on any write
4397 failures on the smtp file descriptor. This function is used when sending
4398 messages over SMTP and also when doing verify callouts.
4400 2. When Exim is called to do synchronous delivery of a locally submitted
4401 message (the -odf or -odi options), it no longer closes stderr before doing
4404 3. Implemented the mua_wrapper option.
4406 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
4408 5. Implemented the functions header_remove(), header_testname(),
4409 header_add_at_position(), and receive_remove_recipient(), and exported them
4412 6. If an ACL "warn" statement specified the addition of headers, Exim already
4413 inserted X-ACL-Warn: at the start if there was no header name. However, it
4414 was not making this test for the second and subsequent header lines if
4415 there were newlines in the string. This meant that an invalid header could
4416 be inserted if Exim was badly configured.
4418 7. Allow an ACL "warn" statement to add header lines at the start or after all
4419 the Received: headers, as well as at the end.
4421 8. Added the rcpt_4xx retry error code.
4423 9. Added postmaster_mailfrom=xxx to callout verification option.
4425 10. Added mailfrom=xxxx to the callout verification option, for verify=
4428 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
4429 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
4431 12. Inserted some casts to stop certain compilers warning when using pointer
4432 differences as field lengths or precisions in printf-type calls (mostly
4433 affecting debugging statements).
4435 13. Added optional readline() support for -be (dynamically loaded).
4437 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
4438 same clock tick as a message's arrival, so that its received time was the
4439 same as the "first fail" time on the retry record, and that message
4440 remained on the queue past the ultimate address timeout, every queue runner
4441 would try a delivery (because it was past the ultimate address timeout) but
4442 after another failure, the ultimate address timeout, which should have then
4443 bounced the address, did not kick in. This was a "< instead of <=" error;
4444 in most cases the first failure would have been in the next clock tick
4445 after the received time, and all would be well.
4447 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
4448 being recognized when the domain list was tested by the match_domain
4449 condition in an expansion string.
4451 16. Added the ${str2b64: operator.
4453 17. Exim was always calling setrlimit() to set a large limit for the number of
4454 processes, without checking whether the existing limit was already
4455 adequate. (It did check for the limit on file descriptors.) Furthermore,
4456 errors from getrlimit() and setrlimit() were being ignored. Now they are
4457 logged to the main and panic logs, but Exim does carry on, to try to do its
4458 job under whatever limits there are.
4460 18. Imported PCRE 5.0.
4462 19. Trivial typo in log message " temporarily refused connection" (the leading
4465 20. If the log selector return_path_on_delivery was set and an address was
4466 redirected to /dev/null, the delivery process crashed because it assumed
4467 that a return path would always be set for a "successful" delivery. In this
4468 case, the whole delivery is bypassed as an optimization, and therefore no
4471 21. Internal re-arrangement: the function for sending a challenge and reading
4472 a response while authentication was assuming a zero-terminated challenge
4473 string. It's now changed to take a pointer and a length, to allow for
4474 binary data in such strings.
4476 22. Added the cyrus_sasl authenticator (code supplied by MBM).
4478 23. Exim was not respecting finduser_retries when seeking the login of the
4479 uid under which it was called; it was always trying 10 times. (The default
4480 setting of finduser_retries is zero.) Also, it was sleeping after the final
4481 failure, which is pointless.
4483 24. Implemented tls_on_connect_ports.
4485 25. Implemented acl_smtp_predata.
4487 26. If the domain in control=submission is set empty, Exim assumes that the
4488 authenticated id is a complete email address when it generates From: or
4489 Sender: header lines.
4491 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
4492 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
4493 chown and chgrp in /bin and hostname in /usr/bin.
4495 28. Exim was keeping the "process log" file open after each use, just as it
4496 does for the main log. This opens the possibility of it remaining open for
4497 long periods when the USR1 signal hits a daemon. Occasional processlog
4498 errors were reported, that could have been caused by this. Anyway, it seems
4499 much more sensible not to leave this file open at all, so that is what now
4502 29. The long-running daemon process does not normally write to the log once it
4503 has entered its main loop, and it closes the log before doing so. This is
4504 so that log files can straightforwardly be renamed and moved. However,
4505 there are a couple of unusual error situations where the daemon does write
4506 log entries, and I had neglected to close the log afterwards.
4508 30. The text of an SMTP error response that was received during a remote
4509 delivery was being truncated at 512 bytes. This is too short for some of
4510 the long messages that one sometimes sees. I've increased the limit to
4513 31. It is now possible to make retry rules that apply only when a message has a
4514 specific sender, in particular, an empty sender.
4516 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
4517 it possible to be selective about when SMTP synchronization is enforced.
4519 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
4521 32. Implemented hosts_connection_nolog.
4523 33. Added an ACL for QUIT.
4525 34. Setting "delay_warning=" to disable warnings was not working; it gave a
4528 35. Added mailbox_size and mailbox_filecount to appendfile.
4530 36. Added control = no_multiline_responses to ACLs.
4532 37. There was a bug in the logic of the code that waits for the clock to tick
4533 in the case where the clock went backwards by a substantial amount such
4534 that the microsecond fraction of "now" was more than the microsecond
4535 fraction of "then" (but the whole seconds number was less).
4537 38. Added support for the libradius Radius client library this is found on
4538 FreeBSD (previously only the radiusclient library was supported).
4544 1. When certain lookups returned multiple values in the form name=value, the
4545 quoting of the values was not always being done properly. Specifically:
4546 (a) If the value started with a double quote, but contained no whitespace,
4548 (b) If the value contained whitespace other than a space character (i.e.
4549 tabs or newlines or carriage returns) it was not quoted.
4550 This fix has been applied to the mysql and pgsql lookups by writing a
4551 separate quoting function and calling it from the lookup code. The fix
4552 should probably also be applied to nisplus, ibase and oracle lookups, but
4553 since I cannot test any of those, I have not disturbed their existing code.
4555 2. A hit in the callout cache for a specific address caused a log line with no
4556 reason for rejecting RCPT. Now it says "Previous (cached) callout
4557 verification failure".
4559 3. There was an off-by-one bug in the queryprogram router. An over-long
4560 return line was truncated at 256 instead of 255 characters, thereby
4561 overflowing its buffer with the terminating zero. As well as fixing this, I
4562 have increased the buffer size to 1024 (and made a note to document this).
4564 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
4565 when Exim is waiting for an SMTP response from a remote server, Exim
4566 restarts its select() call on the socket, thereby resetting its timeout.
4567 This is not a problem when such interrupts are rare. Somebody set up a cron
4568 job to run exiwhat every 2 minutes, which is less than the normal select()
4569 timeout (5 or 10 minutes). This meant that the select() timeout never
4570 kicked in because it was always reset. I have fixed this by comparing the
4571 time when an interrupt arrives with the time at the start of the first call
4572 to select(). If more time than the timeout has elapsed, the interrupt is
4573 treated as a timeout.
4575 5. Some internal re-factoring in preparation for the addition of Sieve
4576 extensions (by MH). In particular, the "personal" test is moved to a
4577 separate function, and given an option for scanning Cc: and Bcc: (which is
4578 not set for Exim filters).
4580 6. When Exim created an email address using the login of the caller as the
4581 local part (e.g. when creating a From: or Sender: header line), it was not
4582 quoting the local part when it contained special characters such as @.
4584 7. Installed new OpenBSD configuration files.
4586 8. Reworded some messages for syntax errors in "and" and "or" conditions to
4587 try to make them clearer.
4589 9. Callout options, other than the timeout value, were being ignored when
4590 verifying sender addresses in header lines. For example, when using
4592 verify = header_sender/callout=no_cache
4594 the cache was (incorrectly) being used.
4596 10. Added a missing instance of ${EXE} to the exim_install script; this affects
4597 only the Cygwin environment.
4599 11. When return_path_on_delivery was set as a log selector, if different remote
4600 addresses in the same message used different return paths and parallel
4601 remote delivery occurred, the wrong values would sometimes be logged.
4602 (Whenever a remote delivery process finished, the return path value from
4603 the most recently started remote delivery process was logged.)
4605 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
4606 lines when AUTH and/or TLS are in use. This is the "received protocol"
4607 field. Exim used to use "asmtp" for authenticated SMTP, without any
4608 indication (in the protocol name) for TLS use. Now it follows the RFC and
4609 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
4610 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
4611 names appear in log lines as well as in Received: header lines.
4613 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
4614 extensions, and comparison tests, and to fix some bugs.
4616 14. Changes to the "personal" filter test:
4618 (1) The test was buggy in that it was just doing the equivalent of
4619 "contains" tests on header lines. For example, if a user's address was
4620 anne@some.where, the "personal" test would incorrectly be true for
4622 To: susanne@some.where
4624 This test is now done by extracting each address from the header in turn,
4625 and checking the entire address. Other tests that are part of "personal"
4626 are now done using regular expressions (for example, to check local parts
4627 of addresses in From: header lines).
4629 (2) The list of non-personal local parts in From: addresses has been
4630 extended to include "listserv", "majordomo", "*-request", and "owner-*",
4631 taken from the Sieve specification recommendations.
4633 (3) If the message contains any header line starting with "List-" it is
4634 treated as non-personal.
4636 (4) The test for "circular" in the Subject: header line has been removed
4637 because it now seems ill-conceived.
4639 15. Minor typos in src/EDITME comments corrected.
4641 16. Installed latest exipick from John Jetmore.
4643 17. If headers_add on a router specified a text string that was too long for
4644 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
4645 of string_sprintf() is now avoided.
4647 18. $message_body_size was not set (it was always zero) when running the DATA
4648 ACL and the local_scan() function.
4650 19. For the "mail" command in an Exim filter, no default was being set for
4651 the once_repeat time, causing a random time value to be used if "once" was
4652 specified. (If the value happened to be <= 0, no repeat happened.) The
4653 default is now 0s, meaning "never repeat". The "vacation" command was OK
4654 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
4655 (I found it when inspecting the code).
4657 20. There is now an overall timeout for performing a callout verification. It
4658 defaults to 4 times the callout timeout, which applies to individual SMTP
4659 commands during the callout. The overall timeout applies when there is more
4660 than one host that can be tried. The timeout is checked before trying the
4661 next host. This prevents very long delays if there are a large number of
4662 hosts and all are timing out (e.g. when the network connections are timing
4663 out). The value of the overall timeout can be changed by specifying an
4664 additional sub-option for "callout", called "maxwait". For example:
4666 verify = sender/callout=5s,maxwait=20s
4668 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
4669 to the end before writing, but this should make it even safer).
4671 22. Exim was forgetting that it had advertised PIPELINING for the second and
4672 subsequent messages on an SMTP connection. It was also not resetting its
4673 memory on STARTTLS and an internal HELO.
4675 23. When Exim logs an SMTP synchronization error within a session, it now
4676 records whether PIPELINING has been advertised or not.
4678 24. Added 3 instances of "(long int)" casts to time_t variables that were being
4679 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
4680 rather than long int.
4682 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
4684 26. Added the never_mail option to autoreply.
4690 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
4691 crash if the getsockname() call failed; this can happen if a connection is
4692 closed very soon after it is established. The problem was simply in the
4693 order in which certain operations were done, causing Exim to try to write
4694 to the SMTP stream before it had set up the file descriptor. The bug has
4695 been fixed by making things happen in the correct order.
4701 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
4702 before the connection was closed, thus losing the rejection response.
4704 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
4705 some early Solaris releases, but causes trouble in current releases where
4706 socklen_t is defined.
4708 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
4711 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
4714 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
4715 that releasing the top of it at the end releases what was used for sub-
4716 expansions (unless the block got too big). However, discard this block if
4717 the first thing is a variable or header, so that we can use its block when
4718 it is dynamic (useful for very large $message_headers, for example).
4720 6. Lookups now cache *every* query, not just the most recent. A new, separate
4721 store pool is used for this. It can be recovered when all lookup caches are
4722 flushed. Lookups now release memory at the end of their result strings.
4723 This has involved some general refactoring of the lookup sources.
4725 7. Some code has been added to the store_xxx() functions to reduce the amount
4726 of flapping under certain conditions.
4728 8. log_incoming_interface used to affect only the <= reception log lines. Now
4729 it causes the local interface and port to be added to several more SMTP log
4730 lines, for example "SMTP connection from", and rejection lines.
4732 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
4734 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
4736 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
4737 could be overwritten at the end of the current message (or the start of a
4738 new message if it was set in a HELO ACL). The value is now preserved for
4739 the duration of the SMTP connection.
4741 12. If a transport had a headers_rewrite setting, and a matching header line
4742 contained an unqualified address, that address was qualified, even if it
4743 did not match any rewriting rules. The underlying bug was that the values
4744 of the flags that permit the existence of unqualified sender and recipient
4745 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
4746 non-local messages, and by -bnq for local messages) were not being
4747 preserved with the message after it was received.
4749 13. When Exim was logging an SMTP synchronization error, it could sometimes log
4750 "next input=" as part of the text comprising the host identity instead of
4751 the correct text. The code was using the same buffer for two different
4752 strings. However, depending on which order the printing function evaluated
4753 its arguments, the bug did not always show up. Under Linux, for example, my
4754 test suite worked just fine.
4756 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
4757 doesn't work with some older versions of Perl. It has been changed to "my",
4758 which in any case is probably the better facility to use.
4760 15. A really picky compiler found some instances of statements for creating
4761 error messages that either had too many or two few arguments for the format
4764 16. The size of the buffer for calls to the DNS resolver has been increased
4765 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
4766 for addresses that have a lot of PTR records. This alleviates a problem; it
4767 does not fully solve it.
4769 17. A dnsdb lookup for PTR records that receives more data than will fit in the
4770 buffer now truncates the list and logs the incident, which is the same
4771 action as happens when Exim is looking up a host name and its aliases.
4772 Previously in this situation something unpredictable would happen;
4773 sometimes it was "internal error: store_reset failed".
4775 18. If a server dropped the connection unexpectedly when an Exim client was
4776 using GnuTLS and trying to read a response, the client delivery process
4777 crashed while trying to generate an error log message.
4779 19. If a "warn" verb in an ACL added multiple headers to a message in a single
4780 string, for example:
4782 warn message = H1: something\nH2: something
4784 the text was added as a single header line from Exim's point of view
4785 though it ended up OK in the delivered message. However, searching for the
4786 second and subsequent header lines using $h_h2: did not work. This has been
4787 fixed. Similarly, if a system filter added multiple headers in this way,
4788 the routers could not see them.
4790 20. Expanded the error message when iplsearch is called with an invalid key to
4791 suggest using net-iplsearch in a host list.
4793 21. When running tests using -bh, any delays imposed by "delay" modifiers in
4794 ACLs are no longer actually imposed (and a message to that effect is
4797 22. If a "gecos" field in a passwd entry contained escaped characters, in
4798 particular, if it contained a \" sequence, Exim got it wrong when building
4799 a From: or a Sender: header from that name. A second bug also caused
4800 incorrect handling when an unquoted " was present following a character
4801 that needed quoting.
4803 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
4804 was not being matched caselessly.
4806 24. Arranged for all hyphens in the exim.8 source to be escaped with
4809 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
4810 properly. Recipient callout cache records were still being keyed to include
4811 the sender, even when use_sender was set false. This led to far more
4812 callouts that were necessary. The sender is no longer included in the key
4813 when use_sender is false.
4815 26. Added "control = submission" modifier to ACLs.
4817 27. Added the ${base62d: operator to decode base 62 numbers.
4819 28. dnsdb lookups can now access SRV records.
4821 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
4822 the configuration file.
4824 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
4825 (-v) mode. This makes the output for a verbose queue run more intelligible.
4827 31. Added a use_postmaster feature to recipient callouts.
4829 32. Added the $body_zerocount variable, containing the number of binary zero
4830 bytes in the message body.
4832 33. The time of last modification of the "new" subdirectory is now used as the
4833 "mailbox time last read" when there is a quota error for a maildir
4836 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
4838 35. Added +ignore_unknown as a special item in host lists.
4840 36. Code for decoding IPv6 addresses in host lists is now included, even if
4841 IPv6 support is not being compiled. This fixes a bug in which an IPv6
4842 address was recognized as an IP address, but was then not correctly decoded
4843 into binary, causing unexpected and incorrect effects when compared with
4850 1. Very minor rewording of debugging text in manualroute to say "list of
4851 hosts" instead of "hostlist".
4853 2. If verify=header_syntax was set, and a header line with an unqualified
4854 address (no domain) and a large number of spaces between the end of the
4855 name and the colon was received, the reception process suffered a buffer
4856 overflow, and (when I tested it) crashed. This was caused by some obsolete
4857 code that should have been removed. The fix is to remove it!
4859 3. When running in the test harness, delay a bit after writing a bounce
4860 message to get a bit more predictability in the log output.
4862 4. Added a call to search_tidyup() just before forking a reception process. In
4863 theory, someone could use a lookup in the expansion of smtp_accept_max_
4864 per_host which, without the tidyup, could leave open a database connection.
4866 5. Added the variables $recipient_data and $sender_data which get set from a
4867 lookup success in an ACL "recipients" or "senders" condition, or a router
4868 "senders" option, similar to $domain_data and $local_part_data.
4870 6. Moved the writing of debug_print from before to after the "senders" test
4873 7. Change 4.31/66 (moving the time when the Received: is generated) caused
4874 problems for message scanning, either using a data ACL, or using
4875 local_scan() because the Received: header was not generated till after they
4876 were called (in order to set the time as the time of reception completion).
4877 I have revised the way this works. The header is now generated after the
4878 body is received, but before the ACL or local_scan() are called. After they
4879 are run, the timestamp in the header is updated.
4885 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
4886 before starting a queue runner without re-exec. This happened only when
4887 deliver_drop_privilege was set or when the Exim user was set to root. The
4888 effect of the bug was that timeouts during subsequent deliveries caused
4889 crashes instead of being properly handled. The handler is now left at its
4890 default (and expected) setting.
4892 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
4893 message, again when deliver_drop_privilege is set or Exim is run as root.
4894 The bug described in (1) was not present in this case, but the tidying up
4895 of the other signals was missing. I have made the two cases consistent.
4897 3. The ignore_target_hosts setting on a manualroute router was being ignored
4898 for hosts that were looked up using the /MX notation.
4900 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
4903 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
4904 operated on the sender address. After changing the $sender_address to <>
4905 for the sender address verify, Exim was re-instated it as the original
4906 (before rewriting) address, but remembering that it had rewritten it, so it
4907 wasn't rewriting it again. This bug also had the effect of breaking the
4908 sender address verification caching when the sender address was rewritten.
4910 6. The ignore_target_hosts option was being ignored by the ipliteral router.
4911 This has been changed so that if the ip literal address matches
4912 ignore_target_hosts, the router declines.
4914 7. Added expansion conditions match_domain, match_address, and match_local_
4915 part (NOT match_host).
4917 8. The placeholder for the Received: header didn't have a length field set.
4919 9. Added code to Exim itself and to exim_lock to test for a specific race
4920 condition that could lead to file corruption when using MBX delivery. The
4921 issue is with the lockfile that is created in /tmp. If this file is removed
4922 after a process has opened it but before that process has acquired a lock,
4923 there is the potential for a second process to recreate the file and also
4924 acquire a lock. This could lead to two Exim processes writing to the file
4925 at the same time. The added code performs the same test as UW imapd; it
4926 checks after acquiring the lock that its file descriptor still refers to
4927 the same named file.
4929 10. The buffer for building added header lines was of fixed size, 8192 bytes.
4930 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
4933 11. Added the smtp_active_hostname option. If used, this will typically be made
4934 to depend on the incoming interface address. Because $interface_address is
4935 not set up until the daemon has forked a reception process, error responses
4936 that can happen earlier (such as "too many connections") no longer contain
4939 12. If an expansion in a condition on a "warn" statement fails because a lookup
4940 defers, the "warn" statement is abandoned, and the next ACL statement is
4941 processed. Previously this caused the whole ACL to be aborted.
4943 13. Added the iplsearch lookup type.
4945 14. Added ident_timeout as a log selector.
4947 15. Added tls_certificate_verified as a log selector.
4949 16. Added a global option tls_require_ciphers (compare the smtp transport
4950 option of the same name). This controls incoming TLS connections.
4952 17. I finally figured out how to make tls_require_ciphers do a similar thing
4953 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
4954 before starting the TLS session.
4956 18. Tabs are now shown as \t in -bP output.
4958 19. If the log selector return_path_on_delivery was set, Exim crashed when
4959 bouncing a message because it had too many Received: header lines.
4961 20. If two routers both had headers_remove settings, and the first one included
4962 a superfluous trailing colon, the final name in the first list and the
4963 first name in the second list were incorrectly joined into one item (with a
4964 colon in the middle).
4970 1. Added -C and -D options to the exinext utility, mainly to make it easier
4971 to include in the automated testing, but these could be helpful when
4972 multiple configurations are in use.
4974 2. The exinext utility was not formatting the output nicely when there was
4975 an alternate port involved in the retry record key, nor when there was a
4976 message id as well (for retries that were specific to a specific message
4977 and a specific host). It was also confused by IPv6 addresses, because of
4978 the additional colons they contain. I have fixed the IPv4 problem, and
4979 patched it up to do a reasonable job for IPv6.
4981 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
4982 delivery, the log line now contains "pipelined" if PIPELINING was used.
4984 4. An SMTP transport process used to panic and die if the bind() call to set
4985 an explicit outgoing interface failed. This has been changed; it is now
4986 treated in the same way as a connect() failure.
4988 5. A reference to $sender_host_name in the part of a conditional expansion
4989 that was being skipped was still causing a DNS lookup. This no longer
4992 6. The def: expansion condition was not recognizing references to header lines
4993 that used bh_ and bheader_.
4995 7. Added the _cache feature to named lists.
4997 8. The code for checking quota_filecount in the appendfile transport was
4998 allowing one more file than it should have been.
5000 9. For compatibility with Sendmail, the command line option
5008 and sets the incoming protocol and host name (for trusted callers). The
5009 host name and its colon can be omitted when only the protocol is to be set.
5010 Note the Exim already has two private options, -pd and -ps, that refer to
5011 embedded Perl. It is therefore impossible to set a protocol value of "d" or
5012 "s", but I don't think that's a major issue.
5014 10. A number of refactoring changes to the code, none of which should affect
5017 (a) The number of logging options was getting close to filling up the
5018 32-bit word that was used as a bit map. I have split them into two classes:
5019 those that are passed in the argument to log_write(), and those that are
5020 only ever tested independently outside of that function. These are now in
5021 separate 32-bit words, so there is plenty of room for expansion again.
5022 There is no change in the user interface or the logging behaviour.
5024 (b) When building, for example, log lines, the code previously used a
5025 macro that called string_cat() twice, in order to add two strings. This is
5026 not really sufficiently general. Furthermore, there was one instance where
5027 it was actually wrong because one of the argument was used twice, and in
5028 one call a function was used. (As it happened, calling the function twice
5029 did not affect the overall behaviour.) The macro has been replaced by a
5030 function that can join an arbitrary number of extra strings onto a growing
5033 (c) The code for expansion conditions now uses a table and a binary chop
5034 instead of a serial search (which was left over from when there were very
5035 few conditions). Also, it now recognizes conditions like "pam" even when
5036 the relevant support is not compiled in: a suitably worded error message is
5037 given if an attempt is made to use such a condition.
5039 11. Added ${time_interval:xxxxx}.
5041 12. A bug was causing one of the ddress fields not to be passed back correctly
5042 from remote delivery subprocesses. The field in question was not being
5043 subsequently used, so this caused to problems in practice.
5045 13. Added new log selectors queue_time and deliver_time.
5047 14. Might have fixed a bug in maildirsizefile handling that threw up
5048 "unexpected character" debug warnings, and recalculated the data
5049 unnecessarily. In any case, I expanded the warning message to give more
5052 15. Added the message "Restricted characters in address" to the statements in
5053 the default ACL that block characters like @ and % in local parts.
5055 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
5056 Three changes have been made:
5058 (a) There was a serious bug; a negative response to MAIL caused the whole
5059 recipient domain to be cached as invalid, thereby blocking all messages
5060 to all local parts at the same domain, from all senders. This bug has
5061 been fixed. The domain is no longer cached after a negative response to
5062 MAIL if the sender used is not empty.
5064 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
5067 (c) A new callout option, "use_sender" has been added for people who want
5068 the modified behaviour.
5074 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
5077 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
5078 indeed breaks things for older releases.
5080 3. Added additional logging to the case where there is a problem reading data
5081 from a filter that is running in a subprocess using a pipe, in order to
5082 try to track down a specific problem.
5084 4. Testing facility fudge: when running in the test harness and attempting
5085 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
5086 getting "No route to host". Convert this to a timeout.
5088 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
5091 6. Some OS don't have socklen_t but use size_t instead. This affects the
5092 fifth argument of getsockopt() amongst other things. This is now
5093 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
5094 can be set for individual OS. I have set it for SunOS5, OSF1, and
5095 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
5096 some earlier ones do not.
5098 7. Change 4.30/15 was not doing the test caselessly.
5100 8. The standard form for an IPv6 address literal was being rejected by address
5101 parsing in, for example, MAIL and RCPT commands. An example of this kind of
5102 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5103 this, as well as the form without the "IPv6" on the front (but only when
5104 address literals are enabled, of course).
5106 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
5108 10. Exim crashed if a message with an empty sender address specified by -f
5109 encountered a router with an errors_to setting. This could be provoked only
5110 by a command such as
5114 where an empty string was supplied; "<>" did not hit this bug.
5116 11. Installed PCRE release 4.5.
5118 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
5119 remained set. It is now erased.
5121 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
5122 times from message ids (which are base 36 rather than the normal 62).
5124 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
5125 were being counted as actual protocol errors, and logged if the log
5126 selector +smtp_protocol_error was set. One cannot be perfect in this test,
5127 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
5128 and DATA following a set of rejected RCPTs do not count as protocol errors.
5129 In other words, Exim assumes they were pipelined, though this may not
5130 actually be the case. Of course, in all cases the client gets an
5131 appropriate error code.
5133 15. If a lookup fails in an ACL condition, a message about the failure may
5134 be available; it is used if testing the ACL cannot continue, because most
5135 such messages specify what the cause of the deferral is. However, some
5136 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
5137 that caused an old message to be retained and used if a later statement
5138 caused a defer, replacing the real cause of the deferral.
5140 16. If an IP address had so many PTR records that the DNS lookup buffer
5141 was not large enough to hold them, Exim could crash while trying to process
5142 the truncated data. It now detects and logs this case.
5144 17. Further to 4.21/58, another change has been made: if (and only if) the
5145 first line of a message (the first header line) ends with CRLF, a bare LF
5146 in a subsequent header line has a space inserted after it, so as not to
5147 terminate the header.
5149 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
5150 unnecessarily, used atoi() instead of strtol(), and didn't check the
5151 termination when getting file sizes from file names by regex.
5153 19. Completely re-implemented the support for maildirsize files, in the light
5154 of a number of problems with the previous contributed implementation
5155 (4.30/29). In particular:
5157 . If the quota is zero, the maildirsize file is maintained, but no quota is
5160 . If the maildir directory does not exist, it is created before any attempt
5161 to write a maildirsize file.
5163 . The quota value in the file is just a cache; if the quota is changed in
5164 the transport, the new value overrides.
5166 . A regular expression is available for excluding directories from the
5169 20. The autoreply transport checks the characters in options that define the
5170 message's headers; it allows continued headers, but it was checking with
5171 isspace() after an embedded newline instead of explicitly looking for a
5174 21. If all the "regular" hosts to which an address was routed had passed their
5175 expiry times, and had not reached their retry times, the address was
5176 bounced, even if fallback hosts were defined. Now Exim should go on to try
5179 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
5180 equivalent code in the SMTP transport. Some hosts send humungous responses
5181 to HELO/EHLO, more than 1024 it seems.
5183 23. Refactoring: code in filter.c used (void *) for "any old type" but this
5184 gives compiler warnings in some environments. I've now done it "properly",
5187 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
5188 (because of problems with the built-in one) was declared to return uschar *
5189 instead of char *, causing compiler failure.
5191 25. Fixed a file descriptor leak when processing alias/forward files.
5193 26. Fixed a minor format string issue in dbfn.c.
5195 27. Typo in exim.c: ("dmbnz" for "dbmnz").
5197 28. If a filter file refered to $h_xxx or $message_headers, and the headers
5198 contained RFC 2047 "words", Exim's memory could, under certain conditions,
5201 29. When a sender address is verified, it is cached, to save repeating the test
5202 when there is more than one recipient in a message. However, when the
5203 verification involves a callout, it is possible for different callout
5204 options to be set for different recipients. It is too complicated to keep
5205 track of this in the cache, so now Exim always runs a verification when a
5206 callout is required, relying on the callout cache for the optimization.
5207 The overhead is duplication of the address routing, but this should not be
5210 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
5211 to be verified with callout=postmaster, and the main callout worked but the
5212 postmaster check failed, the verification correctly failed. However, if a
5213 subsequent RCPT command asked for sender verification *without* the
5214 postmaster check, incorrect caching caused this verification also to fail,
5217 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
5218 it was not caching the DNS options (qualify_single, search_parents) that
5219 were used when the lookup failed. A subsequent lookup with different
5220 options therefore always gave the same answer, though there were cases
5221 where it should not have. (Example: a "domains = !$mx_any" option on a
5222 dnslookup router: the "domains" option is always processed without any
5223 widening, but the router might have qualify_single set.) Now Exim uses the
5224 cached value only when the same options are set.
5226 32. Added John Jetmore's "exipick" utility to the distribution.
5228 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
5229 than a timeout (e.g. a certificate is required, and is not provided), an
5230 Exim server now closes the connection immediately. Previously it waited for
5231 the client to close - but if the client is SSL, it seems that they each
5232 wait for each other, leading to a delay before one of them times out.
5234 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
5235 maintained 0.8.x compatibility because I don't think many are using it, and
5236 it is clearly obsolete.
5238 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
5241 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
5242 client certificate was expired. A simple patch fixes this, though I don't
5243 understand the full logic of why the verify callback is called multiple
5246 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
5247 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
5248 which causes problems with some clients (such as the Certicom SSL Plus
5249 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
5250 disables the coutermeasure allowing Eudora to connect."
5252 38. Exim was not checking that a write() to a log file succeeded. This could
5253 lead to Bad Things if a log got too big, in particular if it hit a file
5254 size limit. Exim now panics and dies if it cannot write to a log file, just
5255 as it does if it cannot open a log file.
5257 39. Modified OS/Makefile-Linux so that it now contains
5259 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
5261 The two -D definitions ensure that Exim is compiled with large file
5262 support, which makes it possible to handle log files that are bigger than
5265 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
5266 instance) a domain was checked against a named list that involved a lookup,
5267 causing $domain_data to be set, then another domain was checked against the
5268 same list, then the first domain was re-checked, the value of $domain_data
5269 after the final check could be wrong. In particular, if the second check
5270 failed, it could be set empty. This bug probably also applied to
5273 41. The strip_trailing_dot option was not being applied to the address given
5274 with the -f command-line option.
5276 42. The code for reading a message's header from the spool was incrementing
5277 $received_count, but never initializing it. This meant that the value was
5278 incorrect (doubled) while delivering a message in the same process in which
5279 it was received. In the most common configuration of Exim, this never
5280 happens - a fresh exec is done - but it can happen when
5281 deliver_drop_privilege is set.
5283 43. When Exim logs an SMTP synchronization error - client data sent too soon -
5284 it now includes up to 150 characters of the unexpected data in the log
5287 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
5288 and building data strings. The size of both of these buffers was 10 000
5289 bytes - far larger than anybody would *ever* want, thought I. Needless to
5290 say, somebody hit the limit. I have increased the maximum line length to
5291 20 000 and the maximum data length of concatenated lines to 100 000. I have
5292 also fixed two bugs, because there was no checking on these buffers. Tsk,
5293 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
5294 buffer is too small.
5296 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
5297 lsearch lookups. Now it does.
5299 46. When parsing a route_list item in a manualroute router, a fixed-length
5300 buffer was used for the list of hosts. I made this 1024 bytes long,
5301 thinking that nobody would ever have a list of hosts that long. Wrong.
5302 Somebody had a whole pile of complicated expansion conditions, and the
5303 string was silently truncated, leading to an expansion error. It turns out
5304 that it is easier to change to an unlimited length (owing to other changes
5305 that have happened since this code was originally written) than to build
5306 structure for giving a limitation error. The length of the item that
5307 expands into the list of hosts is now unlimited.
5309 47. The lsearch lookup could not handle data where the length of text line was
5310 more than 4095 characters. Such lines were truncated, leading to shortened
5311 data being returned. It should now handle lines of any length.
5313 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
5314 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
5317 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
5318 replacing strings such as BIN_DIRECTORY in a source file: the replacement
5319 no longer happens in comment lines. A list of replacements is now placed
5320 at the head of all of the source files, except those whose only change is
5321 to replace PERL_COMMAND in the very first #! line.
5323 50. Replaced the slow insertion sort in queue.c, for sorting the list of
5324 messages on the queue, with a bottom-up merge sort, using code contributed
5325 by Michael Haardt. This should make operations like -bp somewhat faster on
5326 large queues. It won't affect queue runners, except when queue_run_in_order
5329 51. Installed eximstats 1.31 in the distribution.
5331 52. Added support for SRV lookups to the dnslookup router.
5333 53. If an ACL referred to $message_body or $message_body_end, the value was not
5334 reset for any messages that followed in the same SMTP session.
5336 54. The store-handling optimization for building very long strings was not
5337 differentiating between the different store pools. I don't think this
5338 actually made any difference in practice, but I've tidied it.
5340 55. While running the routers to verify a sender address, $sender_address
5341 was still set to the sender address. This is wrong, because when routing to
5342 send a bounce to the sender, it would be empty. Therefore, I have changed
5343 it so that, while verifying a sender address, $sender_address is set to <>.
5344 (There is no change to what happens when verifying a recipient address.)
5346 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
5347 target A or AAAA records (if not already returned) without resetting the
5348 qualify_single or search_parents options of the DNS resolver. These are
5349 inappropriate in this case because the targets of MX and SRV records must
5350 be FQDNs. A broken DNS record could cause trouble if it happened to have a
5351 target that, when qualified, matched something in the local domain. These
5352 two options are now turned off when doing these lookups.
5354 57. It seems that at least some releases of Reiserfs (which does not have the
5355 concept of a fixed number of inodes) returns zero and not -1 for the
5356 number of available inodes. This interacted badly with check_spool_inodes,
5357 which assumed that -1 was the "no such thing" setting. What I have done is
5358 to check that the total number of inodes is greater than zero before doing
5359 the test of how many are available.
5361 58. When a "warn" ACL statement has a log_message modifier, the message is
5362 remembered, and not repeated. This is to avoid a lot of repetition when a
5363 message has many recipients that cause the same warning to be written.
5364 Howewer, Exim was preserving the list of already written lines for an
5365 entire SMTP session, which doesn't seem right. The memory is now reset if a
5366 new message is started.
5368 59. The "rewrite" debugging flag was not showing the result of rewriting in the
5369 debugging output unless log_rewrite was also set.
5371 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
5372 of (int)(handle) when we know that handle contains (void *)(-1).
5374 61. The Exim daemon panic-logs an error return when it closes the incoming
5375 connection. However "connection reset by peer" seems to be common, and
5376 isn't really an error worthy of noting specially, so that particular error
5379 62. When Exim is trying to find all the local interfaces, it used to panic and
5380 die if the ioctl to get the interface flags failed. However, it seems that
5381 on at least one OS (Solaris 9) it is possible to have an interface that is
5382 included in the list of interfaces, but for which you get a failure error
5383 for this call. This happens when the interface is not "plumbed" into a
5384 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
5385 failure of the "get flags" call assumes that the interface is down.
5387 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
5388 makes life easier for people who are doing arithmetic on fields extracted
5389 from dates, where you often get leading zeros that should not be
5390 interpreted as octal.
5392 64. Added qualify_domain to the redirect router, to override the global
5395 65. If a pathologically long header line contained very many addresses (the
5396 report of this problem mentioned 10 000) and each of them was rewritten,
5397 Exim could use up a very large amount of memory. (It kept on making new
5398 copies of the header line as it rewrote, and never released the old ones.)
5399 At the expense of a bit more processing, the header rewriting function has
5400 been changed so that it no longer eats memory in this way.
5402 66. The generation of the Received: header has been moved from the time that a
5403 message starts to be received, to the time that it finishes. The timestamp
5404 in the Received: header should now be very close to that of the <= log
5405 line. There are two side-effects of this change:
5407 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
5408 logged header lines no longer include the local Received: line, because
5409 it has not yet been created. The same applies to a copy of the message
5410 that is returned to a non-SMTP sender when a message is rejected.
5412 (b) When a filter file is tested using -bf, no additional Received: header
5413 is added to the test message. After some thought, I decided that this
5416 This change does not affect the value of $received_for. It is still set
5417 after address rewriting, but before local_scan() is called.
5419 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
5421 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
5422 gave an unhelpful panic error message, and a defer error. I have managed to
5423 change this behaviour so that it now rejects any supplied certificate,
5424 which seems right, as the list of acceptable certificates is empty.
5426 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
5427 gave an unhelpful defer error. I have not managed to make this reject any
5428 supplied certificates, but the error message it gives is "no certificate
5429 supplied", which is not helpful.
5431 70. exigrep's output now also includes lines that are not associated with any
5432 message, but which match the given pattern. Implemented by a patch from
5433 Martin Sluka, which also tidied up the Perl a bit.
5435 71. Recipient callout verification, like sender verification, was using <> in
5436 the MAIL FROM command. This isn't really the right thing, since the actual
5437 sender may affect whether the remote host accepts the recipient or not. I
5438 have changed it to use the actual sender in the callout; this means that
5439 the cache record is now keyed on a recipient/sender pair, not just the
5440 recipient address. There doesn't seem to be a real danger of callout loops,
5441 since a callout by the remote host to check the sender would use <>.
5442 [SEE ABOVE: changed after hitting problems.]
5444 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
5445 temporary errors. However, in the case of such a code being given after
5446 the end of a data transmission (i.e. after ".") Exim was failing to write
5447 a retry record for the message. (Yes, there was some broken host that was
5448 actually sending 8xx at this point.)
5450 73. An unknown lookup type in a host list could cause Exim to panic-die when
5451 the list was checked. (An example that provoked this was putting <; in the
5452 middle of a list instead of at the start.) If this happened during a DATA
5453 ACL check, a -D file could be left lying around. This kind of configuration
5454 error no longer causes Exim to die; instead it causes a defer errror. The
5455 incident is still logged to the main and panic logs.
5457 74. Buglet left over from Exim 3 conversion. The message "too many messages
5458 in one connection" was written to the rejectlog but not the mainlog, except
5459 when address rewriting (yes!) was being logged.
5461 75. Added write_rejectlog option.
5463 76. When a system filter was run not as root (that is, when system_filter_user
5464 was set), the values of the $n variables were not being returned to the
5465 main process; thus, they were not subsequently available in the $sn
5468 77. Added +return_path_on_delivery log selector.
5470 78. A connection timeout was being treated differently from recipients deferred
5471 when testing hosts_max_try with a message that was older than the host's
5472 retry timeout. (The host should not be counted, thus allowing all hosts to
5473 be tried at least once before bouncing.) This may have been the cause of an
5474 occasionally reported bug whereby a message would remain on the queue
5475 longer than the retry timeout, but would be bounced if a delivery was
5476 forced. I say "may" because I never totally pinned down the problem;
5477 setting up timeout/retry tests is difficult. See also the next item.
5479 79. The ultimate address timeout was not being applied to errors that involved
5480 a combination of host plus message (for example, a timeout on a MAIL
5481 command). When an address resolved to a number of possible hosts, and they
5482 were not all tried for each delivery (e.g. because of hosts_max_try), a
5483 message could remain on the queue longer than the retry timeout.
5485 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
5488 81. Fixed an obscure SMTP outgoing bug which required at least the following
5489 conditions: (a) there was another message waiting for the same server;
5490 (b) the server returned 5xx to all RCPT commands in the first message so
5491 that the message was not completed; (c) the server dropped the connection
5492 or gave a negative response to the RSET that Exim sends to abort the
5493 transaction. The observed case was a dropped connection after DATA that had
5494 been sent in pipelining mode. That is, the server had advertised PIPELINING
5495 but was not implementing it correctly. The effect of the bug was incorrect
5496 behaviour, such as trying another host, and this could lead to a crash.
5502 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
5503 and daemon.c were passed as pointers to ints; they should have been
5504 pointers to socklen_t variables (which are typically unsigned ints).
5506 2. Some signed/unsigned type warnings in the os.c file for Linux have been
5509 3. Fixed a really odd bug that affected only the testing scheme; patching a
5510 certain fixed string in the binary changed the value of another string that
5511 happened to be identical to the end of the original first string.
5513 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
5514 name", it returns that address as the IP address. On some operating
5515 systems (e.g. Solaris), it also passes back the IP address string as the
5516 "host name". However, on others (e.g. Linux), it passes back an empty
5517 string. Exim wasn't checking for this, and was changing the host name to an
5518 empty string, assuming it had been canonicized.
5520 5. Although rare, it is permitted to have more than one PTR record for a given
5521 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
5522 all the names associated with an address, because they do in Solaris.
5523 However, it seems that they do not in Linux for data that comes from the
5524 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
5525 I found this out when I moved to a new Linux workstation and tried to run
5526 the Exim test suite.
5528 To get round this problem I have changed the code so that it now does its
5529 own call to the DNS to look up PTR records when searching for a host name.
5530 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
5531 addresses that are only in /etc/hosts are still found.
5533 This behaviour is, however, controlled by an option called host_lookup_
5534 order, which defaults to "bydns:byaddr". If people want to use the other
5535 order, or indeed, just use one or the other means of lookup, they can
5536 specify it in this variable.
5538 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
5539 some operating systems, this comes back from gethostbyaddr() as an empty
5540 string, and this is what Exim used to test for. However, it seems that in
5541 other systems, "." is yielded. Exim now tests for this case too.
5543 7. The values of check_spool_space and check_log_space are now held internally
5544 as a number of kilobytes instead of an absolute number of bytes. If a
5545 numbers is specified without 'K' or 'M', it is rounded up to the nearest
5546 kilobyte. This means that much larger values can be stored.
5548 8. Exim monitor: an attempt to get the action menu when not actually pointing
5549 at a message produces an empty menu entitled "No message selected". This
5550 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
5551 no entries in it ("Shell widget menu has zero width and/or height"). So I
5552 have added a single, blank menu entry in this case.
5554 9. Added ${quote_local_part.
5556 10. MIME decoding is now applied to the contents of Subject: header lines when
5559 11. Now that a reference to $sender_host_address automatically causes a reverse
5560 lookup to occur if necessary (4.13/18), there is no need to arrange for a
5561 host lookup before query-style lookups in lists that might use this
5562 variable. This has therefore been abolished, and the "net-" prefix is no
5563 longer necessary for query-style lookups.
5565 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
5566 have been a typo for LFLAGS, so it has been changed.
5568 13. The install script calls Exim with "-C /dev/null" in order to find the
5569 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
5570 to be output. Howeve, since Exim outputs its version number before the
5571 error, it didn't break the script. It just looked ugly. I fixed this by
5572 always allowing "-C /dev/null" if the caller is root.
5574 14. Ignore overlarge ACL variable number when reading spool file - insurance
5575 against a later release with more variables having written the file.
5577 15. The standard form for an IPv6 address literal was being rejected by EHLO.
5578 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5579 this, as well as the form without the "IPv6" on the front.
5581 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
5582 OS/Makefile-Darwin file.
5584 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
5585 only to LDAP libraries that do not have LDAP_OPT_DEREF.
5587 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
5588 variable. However, some OS use ints for inodes. I've added cast to long int
5589 to get rid of the compiler warning.
5591 19. I had forgotten to lock out "/../" in configuration file names when
5592 ALT_CONFIG_PREFIX was set.
5594 20. Routers used for verification do not need to specify transports. However,
5595 if such a router generated a host list, and callout was configured, Exim
5596 crashed, because it could not find a port number from the (non-existent)
5597 transport. It now assumes port 25 in this circumstance.
5599 21. Added the -t option to exigrep.
5601 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
5602 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
5603 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
5604 them is set, LOOKUP_LSEARCH is forced.
5606 23. "exim -bV" now outputs a list of lookups that are included in the binary.
5608 24. Added sender and host information to the "rejected by local_scan()" log
5609 line; previously there was no indication of these.
5611 25. Added .include_if_exists.
5613 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
5614 Linux. It turns out that not all file systems support this. Apparently some
5615 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
5616 do it.) To cope with this, the error EINVAL, which means that sync-ing is
5617 not supported on the file descriptor, is now ignored when Exim is trying to
5618 sync a directory. This applies only to Linux.
5620 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
5622 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
5623 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
5626 29. Added support for maildirsize files from supplied patch (modified a bit).
5628 30. The use of :fail: followed by an empty string could lead Exim to respond to
5629 sender verification failures with (e.g.):
5631 550 Verification failed for <xxx>
5632 550 Sender verify failed
5634 where the first response line was missing the '-' that indicates it is not
5635 the final line of the response.
5637 31. The loop for finding the name of the user that called Exim had a hardwired
5638 limit of 10; it now uses the value of finduser_retries, which is used for
5639 all other user lookups.
5641 32. Added $received_count variable, available in data and not_smtp ACLs, and at
5644 33. Exim was neglecting to zero errno before one call of strtol() when
5645 expanding a string and expecting an integer value. On some systems this
5646 resulted in spurious "integer overflow" errors. Also, it was casting the
5647 result into an int without checking.
5649 34. Testing for a connection timeout using "timeout_connect" in the retry rules
5650 did not work. The code looks as if it has *never* worked, though it appears
5651 to have been documented since at least releast 1.62. I have made it work.
5653 35. The "timeout_DNS" error in retry rules, also documented since at least
5654 1.62, also never worked. As it isn't clear exactly what this means, and
5655 clearly it isn't a major issue, I have abolished the feature by treating it
5656 as "timeout", and writing a warning to the main and panic logs.
5658 36. The display of retry rules for -brt wasn't always showing the error code
5661 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
5662 timeout_connect_A, timeout_connect_MX.
5664 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
5665 to the empty sender.
5667 39. The daemon was not analysing the content of -oX till after it had closed
5668 stderr and disconnected from the controlling terminal. This meant that any
5669 syntax errors were only noted on the panic log, and the return code from
5670 the command was 0. By re-arranging the code a little, I've made the
5671 decoding happen first, so such errors now appear on stderr, and the return
5672 code is 1. However, the actual setting up of the sockets still happens in
5673 the disconnected process, so errors there are still only recorded on the
5676 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
5677 connections (as happens on some IP stacks) was logged at start up time as
5678 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
5679 it from "IPv6 and IPv4", which means that two separate sockets are being
5682 41. The debug output for gethostbyname2() or getipnodebyname() failures now
5683 says whether AF_INET or AF_INET6 was passed as an argument.
5685 42. Exiwhat output was messed up when time zones were included in log
5688 43. Exiwhat now gives more information about the daemon's listening ports,
5689 and whether -tls-on-connect was used.
5691 44. The "port" option of the smtp transport is now expanded.
5693 45. A "message" modifier in a "warn" statement in a non-message ACL was being
5694 silently ignored. Now an error message is written to the main and panic
5697 46. There's a new ACL modifier called "logwrite" which writes to a log file
5698 as soon as it is encountered.
5700 47. Added $local_user_uid and $local_user_gid at routing time.
5702 48. Exim crashed when trying to verify a sender address that was being
5705 49. Exim was recognizing only a space character after ".include". It now also
5706 recognizes a tab character.
5708 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
5709 extracting the relevant information from the specification. The man page no
5710 longer contains scrambled data for the -d option, and I've added a section
5711 at the front about calling Exim under different names.
5713 51. Added "extra_headers" argument to the "mail" command in filter files.
5715 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
5718 53. Installed eximstats 1.29.
5720 54. Added transport_filter_timeout as a generic transport option.
5722 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
5723 Cc: header lines. This was required by RFC 822, but it not required by RFC
5726 56. Exim used to add From:, Date:, and Message-Id: header lines to any
5727 incoming messages that did not have them. Now it does so only if the
5728 message originates locally, that is, if there is no associated remote host
5729 address. When Resent- header lines are present, this applies to the Resent-
5730 lines rather than the non-Resent- lines.
5732 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
5733 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
5735 58. Messages for configuration errors now include the name of the main
5736 configuration file - useful now that there may be more than one file in a
5737 list (.included file names were always shown).
5739 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
5740 for those rare installations that do not start the daemon as root or run it
5741 setuid root. I've cut out the call to initgroups() if the daemon is not
5744 60. The Exim user and group can now be bound into the binary as text strings
5745 that are looked up at the start of Exim's processing.
5747 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
5749 62. Added $mailstore_basename variable.
5751 63. Installed patch to sieve.c from Michael Haardt.
5753 64. When Exim failed to open the panic log after failing to open the main log,
5754 the original message it was trying to log was written to stderr and debug
5755 output, but if they were not available (the usual case in production), it
5756 was lost. Now it is written to syslog before the two lines that record the
5757 failures to open the logs.
5759 65. Users' Exim filters run in subprocesses under the user's uid. It is
5760 possible for a "deliver" command or an alias in a "personal" command to
5761 provoke an address rewrite. If logging of address rewriting is configured,
5762 this fails because the process is not running as root or exim. There may be
5763 a better way of dealing with this, but for the moment (because 4.30 needs
5764 to be released), I have disabled address rewrite logging when running a
5765 filter in a non-root, non-exim process.
5771 1. The buildconfig auxiliary program wasn't quoting the value set for
5772 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
5773 not defined. This bug was masked in 4.22 by the effect that was fixed in
5776 2. Some messages that were rejected after a message id was allocated were
5777 shown as "incomplete" by exigrep. It no longer does this for messages that
5778 are rejected by local_scan() or the DATA or non-SMTP ACLs.
5780 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
5781 have allow_domain_literals set, the ID did not get logged in the <= line.
5782 Domain literals are now always recognized in Message-ID: header lines.
5784 4. The first argument for a ${extract expansion item is the key name or field
5785 number. Leading and trailing spaces in this item were not being ignored,
5786 causing some misleading effects.
5788 5. When deliver_drop_privilege was set, single queue runner processes started
5789 manually (i.e. by the command "exim -q") or by the daemon (which uses the
5790 same command in the process it spins off) were not dropping privilege.
5792 6. When the daemon running as "exim" started a queue runner, it always
5793 re-executed Exim in the spun-off process. This is a waste of effort when
5794 deliver_drop_privilege is set. The new process now just calls the
5795 queue-runner function directly.
5801 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
5804 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
5805 ignored. Though the use of -oP was forcing the writing of a pid file, it
5806 was always written to the default place.
5808 3. If the message "no IP address found for host xxxx" is generated during
5809 incoming verification, it is now followed by identification of the incoming
5810 connection (so you can more easily find what provoked it).
5812 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
5814 5. Added some features to "harden" Exim a bit more against certain attacks:
5816 (a) There is now a build-time option called FIXED_NEVER_USERS that can
5817 be put in Local/Makefile. This is like the never_users runtime option,
5818 but it cannot be overridden. The default setting is "root".
5820 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
5821 prefix string with which any file named in a -C command line option
5824 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
5825 is retained for -C and -D only if the caller of Exim is root. Without
5826 it, the exim user may also use -C and -D and retain privilege.
5828 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
5829 command line option is disabled.
5831 6. Macro names set by the -D option must start with an upper case letter, just
5832 like macro names defined in the configuration file.
5834 7. Added "dereference=" facility to LDAP.
5836 8. Two instances of the typo "uknown" in the source files are fixed.
5838 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
5839 the Configure-Makefile script screwed up while processing it.
5841 10. Incorporated PCRE 4.4.
5843 11. The SMTP synchronization check was not operating right at the start of an
5844 SMTP session. For example, it could not catch a HELO sent before the client
5845 waited for the greeting. There is now a check for outstanding input at the
5846 point when the greeting is written. Because of the duplex, asynchronous
5847 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
5848 way, but not yet received, when the check is performed.
5850 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
5851 on TCP/IP sockets, because this apparently causes some broken clients to
5854 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
5855 unchanged) from the Cygwin maintainer.
5857 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
5858 was supported instead of testing for mbx. Effectively a typo.
5860 15. The spa authenticator server code was not checking that the input it
5861 received was valid base64.
5863 16. The debug output line for the "set" modifier in ACLs was not showing the
5864 name of the variable that was being set.
5866 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
5868 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
5869 lookup on its own. Something else was needed to trigger the lookup. For
5870 example, a match in host_lookup or the need for a host name in a host list.
5871 Now, if $sender_host_name is referenced and the host name has not yet been
5872 looked up, a lookup is performed. If the lookup fails, the variable remains
5873 empty, and $host_lookup_failed is set to "1".
5875 19. Added "eqi" as a case-independent comparison operator.
5877 20. The saslauthd authentication condition could segfault if neither service
5878 nor realm was specified.
5880 21. If an overflowing value such as "2048M" was set for message_size_limit, the
5881 error message that was logged was misleading, and incoming SMTP
5882 connections were dropped. The message is now more accurate, and temporary
5883 errors are given to SMTP connections.
5885 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
5886 (except RSET) with a 421 error, until QUIT is received. However, it was
5887 failing to send a response to QUIT.
5889 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
5890 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
5891 ACL is now run after the helo_try_verify_hosts code.
5893 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
5894 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
5895 case-independent, so other case variants are also recognized). Apparently
5896 some systems use these upper case variants.
5898 25. If more than two messages were waiting for the same host, and a transport
5899 filter was specified for the transport, Exim sent two messages over the
5900 same TCP/IP connection, and then failed with "socket operation on non-
5901 socket" when it tried to send the third.
5903 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
5905 27. The extern definition of crypt16() in expand.c was not being excluded when
5906 the OS had its own crypt16() function.
5908 28. Added bounce_return_body as a new option, and bounce_return_size_limit
5909 as a preferred synonym for return_size_limit, both as an option and as an
5912 29. Added LIBS=-liconv to OS/Makefile-OSF1.
5914 30. Changed the default configuration ACL to relax the local part checking rule
5915 for addresses that are not in any local domains. For these addresses,
5916 slashes and pipe symbols are allowed within local parts, but the sequence
5917 /../ is explicitly forbidden.
5919 31. SPA server authentication was not clearing the challenge buffer before
5922 32. log_message in a "warn" ACL statement was writing to the reject log as
5923 well as to the main log, which contradicts the documentation and doesn't
5924 seem right (because no rejection is happening). So I have stopped it.
5926 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
5927 However, I am unable to do any testing of this.
5929 34. Fixed an infelicity in the appendfile transport. When checking directories
5930 for a mailbox, to see if any needed to be created, it was accidentally
5931 using path names with one or more superfluous leading slashes; tracing
5932 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
5934 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
5935 contents are added to the log line that is written for every discarded
5936 recipient. (Previously a log_message setting was ignored.)
5938 36. The ${quote: operator now quotes the string if it is empty.
5940 37. The install script runs exim in order to find its version number. If for
5941 some reason other than non-existence or emptiness, which it checks, it
5942 could not run './exim', it was installing it with an empty version number,
5943 i.e. as "exim-". This error state is now caught, and the installation is
5946 38. An argument was missing from the function that creates an error message
5947 when Exim fails to connect to the socket for saslauthd authentication.
5948 This could cause Exim to crash, or give a corrupted message.
5950 39. Added isip, isip4, and isip6 to ${if conditions.
5952 40. The ACL variables $acl_xx are now saved with the message, and can be
5953 accessed later in routers, transports, and filters.
5955 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
5956 strings in the file are not string-expanded.
5958 42. If a MAIL command specified a SIZE value that was too large to fit into an
5959 int variable, the check against message_size_limit failed. Such values are
5960 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
5961 day this will have to be increased, but I don't think I want to be around
5962 when emails are that large.
5969 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
5970 iconv() is not standard in FreeBSD.
5972 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
5973 IPv6 enabled. The observed symptom was a segmentation fault on return from
5974 the function os_common_find_running_interfaces() in src/os.c.
5976 3. In the check_special_case() function in daemon.c I had used "errno" as an
5977 argument name, which causes warnings on some systems. This was basically a
5978 typo, since it was named "eno" in the comments!
5980 4. The code that waits for the clock to tick (at a resolution of some fraction
5981 of a second) so as to ensure message-id uniqueness was always waiting for
5982 at least one whole tick, when it could have waited for less. [This is
5983 almost certainly not relevant at current processor speeds, where it is
5984 unlikely to ever wait at all. But we try to future-proof.]
5986 5. The function that sleeps for a time interval that includes fractions of a
5987 second contained a race. It did not block SIGALRM between setting the
5988 timer, and suspending (a couple of lines later). If the interval was short
5989 and the sigsuspend() was delayed until after it had expired, the suspension
5990 never ended. On busy systems this could lead to processes getting stuck for
5993 6. Some uncommon configurations may cause a lookup to happen in a queue runner
5994 process, before it forks any delivery processes. The open lookup caching
5995 mechanism meant that the open file or database connection was passed into
5996 the delivery process. The problem was that delivery processes always tidy
5997 up cached lookup data. This could cause a problem for the next delivery
5998 process started by the queue runner, because the external queue runner
5999 process does not know about the closure. So the next delivery process
6000 still has data in the lookup cache. In the case of a file lookup, there was
6001 no problem because closing a file descriptor in a subprocess doesn't affect
6002 the parent. However, if the lookup was caching a connection to a database,
6003 the connection was closed, and the second delivery process was likely to
6004 see errors such as "PGSQL: query failed: server closed the connection
6005 unexpectedly". The problem has been fixed by closing all cached lookups
6006 in a queue runner before running a delivery process.
6008 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
6009 seem to have the "const" qualifier which it has on other OS. I've
6012 8. Change 4.21/2 was too strict. It is only if there are two authenticators
6013 *of the same type* (client or server) with the same public name that an
6014 error should be diagnosed.
6016 9. When Exim looked up a host name for an IP address, but failed to find the
6017 original IP address when looking up the host name (a safety check), it
6018 output the message "<ip address> does not match any IP for NULL", which was
6019 confusing, to say the least. The bug was that the host name should have
6020 appeared instead of "NULL".
6022 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
6023 user that is built into the binary, and the -C or -D options is used, root
6024 privilege is dropped before the configuration file is read. In addition,
6025 logging is switched to stderr instead of the normal log files. If the
6026 configuration then re-defines the Exim user, the unprivileged environment
6027 is probably not what is expected, so Exim logs a panic warning message (but
6030 However, if deliver_drop_privilege is set, the unprivileged state may well
6031 be exactly what is intended, so the warning has been cut out in that case,
6032 and Exim is allowed to try to write to its normal log files.
6038 1. smtp_return_error_details was not giving details for temporary sender
6039 or receiver verification errors.
6041 2. Diagnose a configuration error if two authenticators have the same public
6044 3. Exim used not to create the message log file for a message until the first
6045 delivery attempt. This could be confusing when incoming messages were held
6046 for policy or load reasons. The message log file is now created at the time
6047 the message is received, and an initial "Received" line is written to it.
6049 4. The automatically generated man page for command line options had a minor
6050 bug that caused no ill effects; however, a more serious problem was that
6051 the procedure for building the man page automatically didn't always
6052 operate. Consequently, release 4.20 contains an out-of-date version. This
6053 shouldn't happen again.
6055 5. When building Exim with embedded Perl support, the script that builds the
6056 Makefile was calling 'perl' to find its compile-time parameters, ignoring
6057 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
6059 6. The freeze_tell option was not being used for messages that were frozen on
6060 arrival, either by an ACL or by local_scan().
6062 7. Added the smtp_incomplete_transaction log selector.
6064 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
6065 was accepting AUTH without a new EHLO.
6067 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
6068 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
6071 10. Logging of TCP/IP connections (when configured) now happens in the main
6072 daemon process instead of the child process, so that the TCP/IP connection
6073 count is more accurate (but it can never be perfect).
6075 11. The use of "drop" in a nested ACL was not being handled correctly in the
6076 outer ACL. Now, if condition failure induced by the nested "drop" causes
6077 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
6078 or "require"), the connection is dropped.
6080 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
6081 that yield "discard" can now be used with an "accept" or a "discard" verb,
6082 but an error is generated for any others (because I can't see a useful way
6083 to define what should happen).
6085 13. When an ACL is read dynamically from a file (or anywhere else), the lines
6086 are now processed in the same way as lines in the Exim configuration file.
6087 In particular, continuation lines are supported.
6089 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
6091 15. Added -ti meaning -t -i.
6093 16. Check for letters, digits, hyphens, and dots in the names of dnslist
6094 domains, and warn by logging if others are found.
6096 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
6097 returned from GIFCONF when Exim is trying to find the list of interfaces on
6098 a host. The code in os.c has been modified to copy each ifreq to an aligned
6099 structure in all cases.
6101 Also, in some cases, the returned ifreq's were being copied to a 'struct
6102 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
6103 means the last couple of bytes of an IPv6 address could be chopped if the
6104 ifreq contained only a normal sockaddr (14 bytes storage).
6106 18. Named domain lists were not supported in the hosts_treat_as_local option.
6107 An entry such as +xxxx was not recognized, and was treated as a literal
6110 19. Ensure that header lines added by a DATA ACL are included in the reject log
6111 if the ACL subsequently rejects the message.
6113 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
6114 MD5 (which is deprecated).
6116 21. When testing a filter file using -bf, Exim was writing a message when it
6117 took the sender from a "From " line in the message, but it was not doing so
6118 when it took $return_path from a Return-Path: header line. It now does.
6120 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
6121 with a valid header line field name (a series of printing characters
6122 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
6124 23. Changed "disc" in the source to "disk" to conform to the documentation and
6125 the book and for uniformity.
6127 24. Ignore Sendmail's -Ooption=value command line item.
6129 25. When execve() failed while trying to run a command in a pipe transport,
6130 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
6131 could be confused with a return value of 69 from the command itself. This
6132 has been changed to 127, the value the shell returns if it is asked to run
6133 a non-existent command. The wording for the related log line suggests a
6134 non-existent command as the problem.
6136 26. If received_header_text expands to an empty string, do not add a Received:
6137 header line to the message. (Well, it adds a token one on the spool, but
6138 marks it "old" so that it doesn't get used or transmitted.)
6140 27. Installed eximstats 1.28 (addition of -nt option).
6142 28. There was no check for failure on the call to getsockname() in the daemon
6143 code. This can fail if there is a shortage of resources on the system, with
6144 ENOMEM, for example. A temporary error is now given on failure.
6146 29. Contrary to the C standard, it seems that in some environments, the
6147 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
6148 program. Exim now does this explicitly; it affects the formatting of
6149 timestamps using strftime().
6151 30. If exiqsumm was given junk data, it threw up some uninitialized variable
6152 complaints. I've now initialized all the variables, to avoid this.
6154 32. Header lines added by a system filter were not being "seen" during
6155 transport-time rewrites.
6157 33. The info_callback() function passed to OpenSSL is set up with type void
6158 (*)(SSL *, int, int), as described somewhere. However, when calling the
6159 function (actually a macro) that sets it up, the type void(*)() is
6160 expected. I've put in a cast to prevent warnings from picky compilers.
6162 34. If a DNS black list lookup found a CNAME record, but there were no A
6163 records associated with the domain it pointed at, Exim crashed.
6165 35. If a DNS black list lookup returned more than one A record, Exim ignored
6166 all but the first. It now scans all returned addresses if a particular IP
6167 value is being sought. In this situation, the contents of the
6168 $dnslist_value variable are a list of all the addresses, separated by a
6171 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
6172 to accept a host name and all its aliases if the forward lookup for any of
6173 them yielded the IP address of the incoming connection. Now it accepts only
6174 those names whose forward lookup yields the correct IP address. Any other
6175 names are discarded. This closes a loophole whereby a rogue DNS
6176 administrator could create reverse DNS records to break through a
6177 wildcarded host restriction in an ACL.
6179 37. If a user filter or a system filter that ran in a subprocess used any of
6180 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
6181 the wrong values were passed to the pipe command ($thisaddress had the
6182 value of $0, $0 had the value of $1, etc). This bug was introduced by
6183 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
6185 38. Improved the line breaking for long SMTP error messages from ACLs.
6186 Previously, if there was no break point between 40 and 75 characters, Exim
6187 left the rest of the message alone. Two changes have been made: (a) I've
6188 reduced the minimum length to 35 characters; (b) if it can't find a break
6189 point between 35 and 75 characters, it looks ahead and uses the first one
6190 that it finds. This may give the occasional overlong line, but at least the
6191 remaining text gets split now.
6193 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
6194 file descriptors might be low, and that setting 1000 would always raise it.
6195 It turns out that in some environments, the limit is already over 1000 and
6196 that lowering it causes trouble. So now Exim takes care not to decrease it.
6198 40. When delivering a message, the value of $return_path is set to $sender_
6199 address at the start of routing (routers may change the value). By an
6200 oversight, this default was not being set up when an address was tested by
6201 -bt or -bv, which affected the outcome if any router or filter referred to
6204 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
6205 log only when "message" or "log_message" are set. However, if one of the
6206 conditions was an address verification, or a call to a nested ACL, the
6207 messages generated by the underlying test were being passed through. This
6208 no longer happens. The underlying message is available in $acl_verify_
6209 message for both "message" and "log_message" expansions, so it can be
6210 passed through if needed.
6212 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
6213 new expansion $bh_ to give the encoded byte string without charset
6214 translation. Translation happens only if iconv() is available; HAVE_ICONV
6215 indicates this at build time. HEADERS_CHARSET gives the charset to
6216 translate to; headers_charset can change it in the configuration, and
6217 "headers charset" can change it in an individual filter file.
6219 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
6220 that creates RFC 2047 encoded "words" labels them as that charset instead
6221 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
6222 expansion operator; (ii) when Exim creates a From: line for a local
6223 message; (iii) when a header line is rewritten to include a "phrase" part.
6225 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
6226 buggy, causing it to skip the first lines of messages whose message ID
6227 ended in 'D'. This would not have bitten before Exim release 4.14, because
6228 message IDs were unlikely to end in 'D' before then. The effect was to have
6229 incorrect size information for certain domains.
6231 45. #include "config.h" was missing at the start of the crypt16.c module. This
6232 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
6235 46. If there was a timeout during a "random" callout check, Exim treated it as
6236 a failure of the random address, and carried on sending RSET and the real
6237 address. If the delay was just some slowness somewhere, the response to the
6238 original RCPT would be taken as a response to RSET and so on, causing
6239 mayhem of various kinds.
6241 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
6242 when I implemented it. It didn't allow for the fact that some option values
6243 may legitimatetly be negative (e.g. size_addition), and it didn't even do
6244 the right test for positive values.
6246 48. Domain names in DNS records are case-independent. Exim always looks them up
6247 in lower case. Some resolvers return domain names in exactly the case they
6248 appear in the zone file, that is, they may contain uppercase letters. Not
6249 all resolvers do this - some return always lower case. Exim was treating a
6250 change of case by a resolver as a change of domain, similar to a widening
6251 of a domain abbreviation. This triggered its re-routing code and so it was
6252 trying to route what was effectively the same domain again. This normally
6253 caused routing to fail (because the router wouldn't handle the domain
6254 twice). Now Exim checks for this case specially, and just changes the
6255 casing of the domain that it ultimately uses when it transmits the message
6258 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
6261 50. If a filter generated a file delivery with a non-absolute name (possible if
6262 no home directory exists for the router), the forbid_file option was not
6265 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
6266 the existing equality matching.
6268 52. Exim was using ints instead of ino_t variables in some places where it was
6269 dealing with inode numbers.
6271 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
6272 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
6273 called TMPDIR, and if it finds it is different, it changes its value.
6275 54. The smtp_printf() function is now made available to local_scan() so
6276 additional output lines can be written before returning. There is also an
6277 smtp_fflush() function to enable the detection of a dropped connection.
6278 The variables smtp_input and smtp_batched_input are exported to
6281 55. Changed the default runtime configuration: the message "Unknown user"
6282 has been removed from the ACL, and instead placed on the localuser router,
6283 using the cannot_route_message feature. This means that any verification
6284 failures that generate their own messages won't get overridden. Similarly,
6285 the "Unrouteable address" message that was in the ACL for unverifiable
6286 relay addresses has also been removed.
6288 56. Added hosts_avoid_esmtp to the smtp transport.
6290 57. The exicyclog script was not checking for the esoteric option
6291 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
6292 will work only if exicyclog is run under the appropriate euid.
6294 58. Following a discussion on the list, the rules by which Exim recognises line
6295 endings on incoming messages have been changed. The -dropcr and drop_cr
6296 options are now no-ops, retained only for backwards compatibility. The
6297 following line terminators are recognized: LF CRLF CR. However, special
6298 processing applies to CR:
6300 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
6301 nor a local message in the state where . is a terminator.
6303 (ii) If a bare CR is encountered in a header line, an extra space is added
6304 after the line terminator so as not to end the header. The reasoning
6305 behind this is that bare CRs in header lines are most likely either
6306 to be mistakes, or people trying to play silly games.
6308 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
6309 was being incorrectly given as 18 bytes larger than it should have been.
6310 This is a VOB (very old bug).
6312 60. This may never have affected anything current, but just in case it has:
6313 When the local host is found other than at the start of a list of hosts,
6314 the local host, those with the same MX, and any that follow, are discarded.
6315 When the list in question was part of a longer list of hosts, the following
6316 hosts (not currently being processed) were also being discarded. This no
6317 longer happens. I'm not sure if this situation could ever has previously
6320 61. Added the "/MX" feature to lists of hosts in the manualroute and query
6323 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
6324 header. This is something that is recommended in a new Internet Draft, and
6325 is something that is documented as being done by Sendmail. There are two
6326 possible values. For messages generated by the autoreply transport, Exim
6329 Auto-Submitted: auto-replied
6331 whereas for all other generated messages (e.g. bounces) it adds
6333 Auto-Submitted: auto-generated
6335 63. The "personal" condition in filters now includes a test for the
6336 Auto-Submitted: header. If it contains the string "auto-" the message it
6337 not considered personal.
6339 64. Added rcpt_include_affixes as a generic transport option.
6341 65. Added queue_only_override (default true).
6343 66. Added the syslog_duplication option.
6345 67. If what should have been the first header line of a message consisted of
6346 a space followed by a colon, Exim was mis-interpreting it as a header line.
6347 It isn't of course - it is syntactically invalid and should therefore be
6348 treated as the start of the message body. The misbehaviour could have
6349 caused a number of strange effects, including loss of data in subsequent
6350 header lines, and spool format errors.
6352 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
6353 client host had authenticated. This control can now be exercised by an ACL
6354 for more flexibility.
6356 69. By default, callouts do not happen when testing with -bh. There is now a
6357 variant, -bhc, which does actually run the callout code, including
6358 consulting and updating the callout cache.
6360 70. Added support for saslauthd authentication, courtesy of Alexander
6363 71. If statvfs() failed on the spool or log directories while checking their
6364 size for availability, Exim confusingly gave the error "space shortage".
6365 Furthermore, in debugging mode it crashed with a floating point exception.
6366 These checks are done if check_{spool,log}_{space,inodes} are set, and when
6367 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
6368 serious problem, Exim now writes to the main and panic logs when this
6369 happens, with details of the failure. It then refuses to accept the
6370 incoming message, giving the message "spool directory problem" or "log
6371 directory problem" with a 421 code for SMTP messages.
6373 72. When Exim is about to re-exec itself, it ensures that the file descriptors
6374 0, 1, and 2 exist, because some OS complain for execs without them (see
6375 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
6376 descriptors. However, the code omitted to check that the open succeeded,
6377 causing mysterious errors if for some reason the permissions on /dev/null
6378 got screwed. Now Exim writes a message to the main and panic logs, and
6379 bombs out if it can't open /dev/null.
6381 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
6382 interact so that it is all more flexible. It is supposed to remain
6383 backwards compatible. Also added extra_local_interfaces.
6385 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
6386 to bomb out with an assertion failure - to the client this appears as a
6387 connection drop. This problem occurs in the part of the code that was taken
6388 from the Samba project. Fortunately, the assertion is in a very simple
6389 function, so I have fixed this by reproducing the function inline in the
6390 one place where it is called, and arranging for authentication to fail
6391 instead of killing the process with assert().
6393 75. The SPA client code was not working when the server requested OEM rather
6394 than Unicode encoding.
6396 76. Added code to make require_files with a specific uid setting more usable in
6397 the case where statting the file as root fails - usually a non-root-mounted
6398 NFS file system. When this happens and the failure is EACCES, Exim now
6399 forks a subprocess and does the per-uid checking as the relevant uid.
6401 77. Added process_log_path.
6403 78. If log_file_path was not explicitly set, a setting of check_log_space or
6404 check_log_inodes was ignored.
6406 79. If a space check for the spool or log partitions fails, the incident is now
6407 logged. Of course, in the latter case the data may get lost...
6409 80. Added the %p formatting code to string_format() so that it can be used to
6410 print addresses in debug_print(). Adjusted all the address printing in the
6411 debugging in store.c to use %p rather than %d.
6413 81. There was a concern that a line of code in smtp_in.c could overflow a
6414 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
6415 initially expressed, the concern was not well-founded, because trailing
6416 spaces are removed early. However, if the trailing spaces were followed by
6417 a NULL, they did not get removed, so the overflow was possible. Two fixes
6420 (a) I re-wrote the offending code in a cleaner fashion.
6421 (b) If an incoming SMTP command contains a NULL character, it is rejected
6424 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
6425 runs initgroups(), so that if the Exim user is in any additional groups,
6426 they will be used during message reception.
6432 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob