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.
132 JH/30 Updated DANE implementation code to current from Viktor Dukhovni.
137 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
140 JH/02 The smtp transport option "multi_domain" is now expanded.
142 JH/03 The smtp transport now requests PRDR by default, if the server offers
145 JH/04 Certificate name checking on server certificates, when exim is a client,
146 is now done by default. The transport option tls_verify_cert_hostnames
147 can be used to disable this per-host. The build option
148 EXPERIMENTAL_CERTNAMES is withdrawn.
150 JH/05 The value of the tls_verify_certificates smtp transport and main options
151 default to the word "system" to access the system default CA bundle.
152 For GnuTLS, only version 3.0.20 or later.
154 JH/06 Verification of the server certificate for a TLS connection is now tried
155 (but not required) by default. The verification status is now logged by
156 default, for both outbound TLS and client-certificate supplying inbound
159 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
162 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
163 Status Notification (bounce) messages are now MIME format per RFC 3464.
164 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
165 under the control of the dsn_advertise_hosts option, and routers may
166 have a dsn_lasthop option.
168 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
169 default, modifiable by a malware= option. The list separator for
170 the options can now be changed in the usual way. Bug 68.
172 JH/10 The smtp_receive_timeout main option is now expanded before use.
174 JH/11 The incoming_interface log option now also enables logging of the
175 local interface on delivery outgoing connections.
177 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
178 if the interface and destination host and port all match.
180 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
183 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
184 Patch from Andrew Lewis.
186 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
187 now supports optional time-restrictions, weighting, and priority
188 modifiers per server. Patch originally by <rommer@active.by>.
190 JH/16 The spamd_address main option now supports a mixed list of local
191 and remote servers. Remote servers can be IPv6 addresses, and
192 specify a port-range.
194 JH/17 Bug 68: The spamd_address main option now supports an optional
195 timeout value per server.
197 JH/18 Bug 1581: Router and transport options headers_add/remove can
198 now have the list separator specified.
200 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
203 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
206 JH/21 Support for the A6 type of dns record is withdrawn.
208 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
209 rather than the verbs used.
211 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
212 from 255 to 1024 chars.
214 JH/24 Verification callouts now attempt to use TLS by default.
216 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
217 are generic router options now. The defaults didn't change.
219 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
220 Original patch from Alexander Shikoff, worked over by JH.
222 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
223 files if ZCAT_COMMAND is not executable.
225 JH/26 Bug 1539: Add timout/retry options on dnsdb lookups.
227 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
229 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
230 Normally benign, it bites when the pair was led to by a CNAME;
231 modern usage is to not canoicalize the domain to a CNAME target
232 (and we were inconsistent anyway for A-only vs AAAA+A).
234 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
236 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
237 when evaluating $sender_host_dnssec.
239 JH/31 Check the HELO verification lookup for DNSSEC, adding new
240 $sender_helo_dnssec variable.
242 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
244 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
246 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
248 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
249 documented as working, but never had. Support all but $spam_report.
251 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
252 added for tls authenticator.
257 TL/01 When running the test suite, the README says that variables such as
258 no_msglog_check are global and can be placed anywhere in a specific
259 test's script, however it was observed that placement needed to be near
260 the beginning for it to behave that way. Changed the runtest perl
261 script to read through the entire script once to detect and set these
262 variables, reset to the beginning of the script, and then run through
263 the script parsing/test process like normal.
265 TL/02 The BSD's have an arc4random API. One of the functions to induce
266 adding randomness was arc4random_stir(), but it has been removed in
267 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
268 function when detected.
270 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
271 cause callback expansion.
273 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
274 syntax errors in an expansion can be treated as a string instead of
275 logging or causing an error, due to the internal use of bool_lax
276 instead of bool when processing it.
278 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
279 server certificates when making smtp deliveries.
281 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
283 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
285 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
287 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
288 Merged patch from Sebastian Wiedenroth.
290 JH/05 Fix results-pipe from transport process. Several recipients, combined
291 with certificate use, exposed issues where response data items split
292 over buffer boundaries were not parsed properly. This eventually
293 resulted in duplicates being sent. This issue only became common enough
294 to notice due to the introduction of conection certificate information,
295 the item size being so much larger. Found and fixed by Wolfgang Breyha.
297 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
298 size buffer was used, resulting in syntax errors when an expansion
301 JH/07 Add support for directories of certificates when compiled with a GnuTLS
302 version 3.3.6 or later.
304 JH/08 Rename the TPDA expermimental facility to Event Actions. The #ifdef
305 is EXPERIMENTAL_EVENT, the main-configuration and transport options
306 both become "event_action", the variables become $event_name, $event_data
307 and $event_defer_errno. There is a new variable $verify_mode, usable in
308 routers, transports and related events. The tls:cert event is now also
309 raised for inbound connections, if the main configuration event_action
312 TL/06 In test suite, disable OCSP for old versions of openssl which contained
313 early OCSP support, but no stapling (appears to be less than 1.0.0).
315 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
316 server certificate names available under the smtp transport option
317 "tls_verify_cert_hostname" now do not permit multi-component wildcard
320 JH/10 Time-related extraction expansions from certificates now use the main
321 option "timezone" setting for output formatting, and are consistent
322 between OpenSSL and GnuTLS compilations. Bug 1541.
324 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
325 encoded parameter in the incoming message. Bug 1558.
327 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
328 include certificate info, eximon was claiming there were spoolfile
331 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
333 JH/14 Log delivery-related information more consistently, using the sequence
334 "H=<name> [<ip>]" wherever possible.
336 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
337 are problematic for Debian distribution, omit them from the release
340 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
342 JH/16 Fix string representation of time values on 64bit time_t anchitectures.
345 JH/17 Fix a null-indirection in certextract expansions when a nondefault
346 output list separator was used.
351 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
352 checkers that were complaining about end of non-void function with no
355 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
356 This was a regression intruduced in 4.83 by another bugfix.
358 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
360 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
361 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
367 TF/01 Correctly close the server side of TLS when forking for delivery.
369 When a message was received over SMTP with TLS, Exim failed to clear up
370 the incoming connection properly after forking off the child process to
371 deliver the message. In some situations the subsequent outgoing
372 delivery connection happened to have the same fd number as the incoming
373 connection previously had. Exim would try to use TLS and fail, logging
374 a "Bad file descriptor" error.
376 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
377 utilities have not been installed.
379 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
380 temporary space as the ACL may create new global variables.
382 TL/01 LDAP support uses per connection or global context settings, depending
383 upon the detected version of the libraries at build time.
385 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
386 to extract and use the src ip:port in logging and expansions as if it
387 were a direct connection from the outside internet. PPv2 support was
388 updated based on HAProxy spec change in May 2014.
390 JH/02 Add ${listextract {number}{list}{success}{fail}}.
392 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
393 Properly escape header and check for NULL return.
395 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
398 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
400 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
401 characters in header names, implemented as a verify condition.
402 Contributed by Michael Fischer v. Mollard.
404 TL/05 Rename SPF condition results err_perm and err_temp to standardized
405 results permerror and temperror. Previous values are deprecated but
406 still accepted. In a future release, err_perm and err_temp will be
407 completely removed, which will be a backward incompatibility if the
408 ACL tests for either of these two old results. Patch contributed by
409 user bes-internal on the mailing list.
411 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
413 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
414 selectors, in both main and reject logs.
416 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
419 JH/07 Add malware type "sock" for talking to simple daemon.
421 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
423 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
424 routers/transports under cutthrough routing.
426 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
427 numbers. Touch up "bool" conditional to keep the same definition.
429 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
431 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
433 JH/12 Expand items in router/transport headers_add or headers_remove lists
434 individually rather than the list as a whole. Bug 1452.
436 Required for reasonable handling of multiple headers_ options when
437 they may be empty; requires that headers_remove items with embedded
438 colons must have them doubled (or the list-separator changed).
440 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
441 view the policy declared in the DMARC record. Currently, $dmarc_status
442 is a combined value of both the record presence and the result of the
445 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
447 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
448 dnslookup router and the smtp transport (applying to the forward
451 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
452 of ldap servers used for a specific lookup. Patch provided by Heiko
455 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
456 New variable $lookup_dnssec_authenticated for observability.
458 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
459 Patch submitted by Lars Timman.
461 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
463 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
464 Requires trusted mode and valid format message id, aborts otherwise.
465 Patch contributed by Heiko Schlichting.
467 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
468 certextract with support for various fields. Bug 1358.
470 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
471 is requested by default, modifiable by smtp transport option
474 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
475 operate on certificate variables to give certificate fingerprints
476 Also new ${sha256:cert_variable}.
478 JH/23 The PRDR feature is moved from being Experimental into the mainline.
480 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
481 Christian Aistleitner.
483 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
485 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
486 file. Patch from Wolfgang Breyha.
488 JH/25 Expand the coverage of the delivery $host and $host_address to
489 client authenticators run in verify callout. Bug 1476.
491 JH/26 Port service names are now accepted for tls_on_connect_ports, to
492 align with daemon_smtp_ports. Bug 72.
494 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
495 support and error reporting did not work properly.
497 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
498 and is readable. Patch from Andrew Colin Kissa.
500 TL/14 Enhance documentation of ${run expansion and how it parses the
501 commandline after expansion, particularly in the case when an
502 unquoted variable expansion results in an empty value.
504 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
506 PP/02 Fix internal collision of T_APL on systems which support RFC3123
507 by renaming away from it. Addresses GH issue 15, reported by
510 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
512 TL/15 SECURITY: prevent double expansion in math comparison functions
513 (can expand unsanitized data). Not remotely exploitable.
520 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
522 PP/02 Make -n do something, by making it not do something.
523 When combined with -bP, the name of an option is not output.
525 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
528 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
529 $sender_host_name and config options to manage this, and basic check
532 PP/05 DSCP support for outbound connections and control modifier for inbound.
534 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
535 (Only plugin which currently uses this is kerberos4, which nobody should
536 be using, but we should make it available and other future plugins might
537 conceivably use it, even though it would break NAT; stuff *should* be
538 using channel bindings instead).
540 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
541 name; added for Sendmail compatibility; requires admin caller.
542 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
543 just ignore it); requires trusted caller.
544 Also parse but ignore: -Ac -Am -X<logfile>
547 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
549 TL/02 Add +smtp_confirmation as a default logging option.
551 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
552 Patch by Magnus Holmgren from 2007-02-20.
554 TL/04 Bugzilla 1281 - Spec typo.
555 Bugzilla 1283 - Spec typo.
556 Bugzilla 1290 - Spec grammar fixes.
558 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
560 TL/06 Add Experimental DMARC support using libopendmarc libraries.
562 TL/07 Fix an out of order global option causing a segfault. Reported to dev
563 mailing list by by Dmitry Isaikin.
565 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
567 JH/02 Support "G" suffix to numbers in ${if comparisons.
569 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
571 NM/01 Bugzilla 1197 - Spec typo
572 Bugzilla 1196 - Spec examples corrections
574 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
576 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
577 gnutls_enable_pkcs11, but renamed to more accurately indicate its
580 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
581 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
583 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
584 "acl {{name}{arg}...}", and optional args on acl condition
587 JH/05 Permit multiple router/transport headers_add/remove lines.
589 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
591 JH/07 Avoid using a waiting database for a single-message-only transport.
592 Performance patch from Paul Fisher. Bugzilla 1262.
594 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
597 JH/09 Add $headers_added variable, with content from use of ACL modifier
598 add_header (but not yet added to the message). Bugzilla 199.
600 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
601 Pulled from Bugzilla 817 by Wolfgang Breyha.
603 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
605 (nb: this is the same fix as in Exim 4.80.1)
607 JH/11 Add A= logging on delivery lines, and a client_set_id option on
610 JH/12 Add optional authenticated_sender logging to A= and a log_selector
613 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
615 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
616 advertise SMTP AUTH mechanism to us, instead of a generic
617 protocol violation error. Also, make Exim more robust to bad
618 data from the Dovecot auth socket.
620 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
622 When a queue runner is handling a message, Exim first routes the
623 recipient addresses, during which it prunes them based on the retry
624 hints database. After that it attempts to deliver the message to
625 any remaining recipients. It then updates the hints database using
628 So if a recipient address works intermittently, it can get repeatedly
629 deferred at routing time. The retry hints record remains fresh so the
630 address never reaches the final cutoff time.
632 This is a fairly common occurrence when a user is bumping up against
633 their storage quota. Exim had some logic in its local delivery code
634 to deal with this. However it did not apply to per-recipient defers
635 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
637 This change adds a proper retry rule check during routing so that the
638 final cutoff time is checked against the message's age. We only do
639 this check if there is an address retry record and there is not a
640 domain retry record; this implies that previous attempts to handle
641 the address had the retry_use_local_parts option turned on. We use
642 this as an approximation for the destination being like a local
643 delivery, as in LMTP.
645 I suspect this new check makes the old local delivery cutoff check
646 redundant, but I have not verified this so I left the code in place.
648 TF/02 Correct gecos expansion when From: is a prefix of the username.
650 Test 0254 submits a message to Exim with the header
654 When I ran the test suite under the user fanf2, Exim expanded
655 the header to contain my full name, whereas it should have added
656 a Resent-Sender: header. It erroneously treats any prefix of the
657 username as equal to the username.
659 This change corrects that bug.
661 GF/01 DCC debug and logging tidyup
662 Error conditions log to paniclog rather than rejectlog.
663 Debug lines prefixed by "DCC: " to remove any ambiguity.
665 TF/03 Avoid unnecessary rebuilds of lookup-related code.
667 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
668 Bug spotted by Jeremy Harris; was flawed since initial commit.
669 Would have resulted in OCSP responses post-SNI triggering an Exim
670 NULL dereference and crash.
672 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
674 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
675 Bug detection, analysis and fix by Samuel Thibault.
676 Bugzilla 1331, Debian bug #698092.
678 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
680 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
681 Server implementation by Todd Lyons, client by JH.
682 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
683 config variable "prdr_enable" controls whether the server
684 advertises the facility. If the client requests PRDR a new
685 acl_data_smtp_prdr ACL is called once for each recipient, after
686 the body content is received and before the acl_smtp_data ACL.
687 The client is controlled by bolth of: a hosts_try_prdr option
688 on the smtp transport, and the server advertisement.
689 Default client logging of deliveries and rejections involving
690 PRDR are flagged with the string "PRDR".
692 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
693 fclose(). Diagnosis by Todd Lyons.
695 PP/17 Update configure.default to handle IPv6 localhost better.
696 Patch by Alain Williams (plus minor tweaks).
699 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
700 This is now consistent with GnuTLS, and is now documented: the
701 previous undocumented portable approach to treating the option as
702 unset was to force an expansion failure. That still works, and
703 an empty string is now equivalent.
705 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
706 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
707 not performing validation itself.
709 PP/20 Added force_command boolean option to pipe transport.
710 Patch from Nick Koston, of cPanel Inc.
712 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
715 TF/04 Added udpsend ACL modifer and hexquote expansion operator
717 PP/21 Fix eximon continuous updating with timestamped log-files.
718 Broken in a format-string cleanup in 4.80, missed when I repaired the
719 other false fix of the same issue.
720 Report and fix from Heiko Schlichting.
723 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
724 Report from Prashanth Katuri.
726 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
727 It's SecureTransport, so affects any MacOS clients which use the
728 system-integrated TLS libraries, including email clients.
730 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
731 using a MIME ACL for non-SMTP local injection.
732 Report and assistance in diagnosis by Warren Baker.
734 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
736 JH/16 Fix comparisons for 64b. Bugzilla 1385.
738 TL/09 Add expansion variable $authenticated_fail_id to keep track of
739 last id that failed so it may be referenced in subsequent ACL's.
741 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
744 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
745 ldap library initialization, allowing self-signed CA's to be
746 used. Also properly sets require_cert option later in code by
747 using NULL (global ldap config) instead of ldap handle (per
748 session). Bug diagnosis and testing by alxgomz.
750 TL/12 Enhanced documentation in the ratelimit.pl script provided in
751 the src/util/ subdirectory.
753 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
754 renamed to Transport Post Delivery Action by Jeremy Harris, as
757 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
758 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
759 redis_servers = needs to be configured which will be used by the redis
760 lookup. Patch from Warren Baker, of The Packet Hub.
762 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
764 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
765 hostname or reverse DNS when processing a host list. Used suggestions
766 from multiple comments on this bug.
768 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
770 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
771 Missed a few lines, added it to make the runtest require no keyboard
774 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
775 contains upper case chars. Make router use caseful_local_part.
777 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
778 support when GnuTLS has been built with p11-kit.
784 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
786 This, or similar/improved, will also be change PP/11 of 4.82.
792 PP/01 Handle short writes when writing local log-files.
793 In practice, only affects FreeBSD (8 onwards).
794 Bugzilla 1053, with thanks to Dmitry Isaikin.
796 NM/01 Bugzilla 949 - Documentation tweak
798 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
801 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
803 PP/02 Implemented gsasl authenticator.
805 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
807 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
808 `pkg-config foo` for cflags/libs.
810 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
811 with rest of GSASL and with heimdal_gssapi.
813 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
814 `pkg-config foo` for cflags/libs for the TLS implementation.
816 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
817 properties get this fed in as external SSF. A number of robustness
818 and debugging improvements to the cyrus_sasl authenticator.
820 PP/08 cyrus_sasl server now expands the server_realm option.
822 PP/09 Bugzilla 1214 - Log authentication information in reject log.
823 Patch by Jeremy Harris.
825 PP/10 Added dbmjz lookup type.
827 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
829 PP/12 MAIL args handles TAB as well as SP, for better interop with
830 non-compliant senders.
831 Analysis and variant patch by Todd Lyons.
833 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
834 Bug report from Lars Müller <lars@samba.org> (via SUSE),
835 Patch from Dirk Mueller <dmueller@suse.com>
837 PP/13 tls_peerdn now print-escaped for spool files.
838 Observed some $tls_peerdn in wild which contained \n, which resulted
839 in spool file corruption.
841 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
842 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
843 or write after TLS renegotiation, which otherwise led to messages
846 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
847 as a tracking header (ie: a signed header comes before the signature).
848 Patch from Wolfgang Breyha.
850 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
851 comma-sep list; embedded commas doubled.
853 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
855 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
857 Report and patch from Dmitry Banschikov.
859 PP/16 Removed "dont_insert_empty_fragments" fron "openssl_options".
860 Removed SSL_clear() after SSL_new() which led to protocol negotiation
861 failures. We appear to now support TLS1.1+ with Exim.
863 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
864 lets Exim select keys and certificates based upon TLS SNI from client.
865 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
866 before an outbound SMTP session. New log_selector, +tls_sni.
868 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
869 NULL dereference. Report and patch from Alun Jones.
871 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
872 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
873 Not seeing resolver debug output on NetBSD, but suspect this is a
874 resolver implementation change.
876 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
877 Left warnings. Added "eximon gdb" invocation mode.
879 PP/21 Defaulting "accept_8bitmime" to true, not false.
881 PP/22 Added -bw for inetd wait mode support.
883 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
884 locate the relevant includes and libraries. Made this the default.
886 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
887 Bugzilla 1246, report and most of solution from Tomasz Kusy.
889 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
890 This may cause build issues on older platforms.
892 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
893 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
894 gnutls_require_mac & gnutls_require_protocols (no longer supported).
895 Added SNI support via GnuTLS too.
896 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
898 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
900 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
901 Applied second patch from Janne, implementing suggestion to default
902 multiple-strings-in-record handling to match SPF spec.
904 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
906 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
907 read-only, out of scope).
908 Patch from Wolfgang Breyha, report from Stuart Northfield.
910 PP/29 Fix three issues highlighted by clang analyser static analysis.
911 Only crash-plausible issue would require the Cambridge-specific
912 iplookup router and a misconfiguration.
913 Report from Marcin Mirosław.
915 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
917 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
918 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
919 As part of this, removing so much warning spew let me fix some minor
920 real issues in debug logging.
922 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
923 assignment on my part. Fixed.
925 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
926 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
927 Janne Snabb (who went above and beyond: thank you).
929 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
930 string otherwise requires a connection and a bunch more work and it's
931 relatively easy to get wrong. Should also expose TLS library linkage
934 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
935 64-bit ${eval} (JH/03).
937 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
938 GNU libc to support some of the 64-bit stuff, should not lead to
939 conflicts. Defined before os.h is pulled in, so if a given platform
940 needs to override this, it can.
942 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
943 protection layer was required, which is not implemented.
944 Bugzilla 1254, patch from Wolfgang Breyha.
946 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
947 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
948 tls_dhparam take prime identifiers. Also unbreak combination of
949 OpenSSL+DH_params+TLSSNI.
951 PP/39 Disable SSLv2 by default in OpenSSL support.
957 PP/01 Solaris build fix for Oracle's LDAP libraries.
958 Bugzilla 1109, patch from Stephen Usher.
960 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
962 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
965 TF/02 Fix a couple more cases where we did not log the error message
966 when unlink() failed. See also change 4.74-TF/03.
968 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
969 lock up or crash if it happened to be inside a call to libc when it
970 got a SIGUSR1 from exiwhat.
972 The SIGUSR1 handler appends the current process status to the process
973 log which is later printed by exiwhat. It used to use the general
974 purpose logging code to do this, but several functions it calls are
975 not safe for signals.
977 The new output code in the SIGUSR1 handler is specific to the process
978 log, and simple enough that it's easy to inspect for signal safety.
979 Removing some special cases also simplifies the general logging code.
980 Removing the spurious timestamps from the process log simplifies
983 TF/04 Improved ratelimit ACL condition.
985 The /noupdate option has been deprecated in favour of /readonly which
986 has clearer semantics. The /leaky, /strict, and /readonly update modes
987 are mutually exclusive. The update mode is no longer included in the
988 database key; it just determines when the database is updated. (This
989 means that when you upgrde Exim will forget old rate measurements.)
991 Exim now checks that the per_* options are used with an update mode that
992 makes sense for the current ACL. For example, when Exim is processing a
993 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
994 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
995 must specify per_mail/readonly. If you omit the update mode it defaults to
996 /leaky where that makes sense (as before) or /readonly where required.
998 The /noupdate option is now undocumented but still supported for
999 backwards compatibility. It is equivalent to /readonly except that in
1000 ACLs where /readonly is required you may specify /leaky/noupdate or
1001 /strict/noupdate which are treated the same as /readonly.
1003 A useful new feature is the /count= option. This is a generalization
1004 of the per_byte option, so that you can measure the throughput of other
1005 aggregate values. For example, the per_byte option is now equivalent
1006 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
1008 The per_rcpt option has been generalized using the /count= mechanism
1009 (though it's more complicated than the per_byte equivalence). When it is
1010 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
1011 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
1012 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
1013 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
1014 Note that using per_rcpt with a non-readonly update mode in more than
1015 one ACL will cause the recipients to be double-counted. (The per_mail
1016 and per_byte options don't have this problem.)
1018 The handling of very low rates has changed slightly. If the computed rate
1019 is less than the event's count (usually one) then this event is the first
1020 after a long gap. In this case the rate is set to the same as this event's
1021 count, so that the first message of a spam run is counted properly.
1023 The major new feature is a mechanism for counting the rate of unique
1024 events. The new per_addr option counts the number of different
1025 recipients that someone has sent messages to in the last time period. It
1026 behaves like per_rcpt if all the recipient addresses are different, but
1027 duplicate recipient addresses do not increase the measured rate. Like
1028 the /count= option this is a general mechanism, so the per_addr option
1029 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
1030 example, measure the rate that a client uses different sender addresses
1031 with the options per_mail/unique=$sender_address. There are further
1032 details in the main documentation.
1034 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
1036 TF/06 Removed a few PCRE remnants.
1038 TF/07 Automatically extract Exim's version number from tags in the git
1039 repository when doing development or release builds.
1041 PP/02 Raise smtp_cmd_buffer_size to 16kB.
1042 Bugzilla 879. Patch from Paul Fisher.
1044 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
1045 Heavily based on revision 40f9a89a from Simon Arlott's tree.
1048 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
1050 PP/05 Variable $av_failed, true if the AV scanner deferred.
1051 Bugzilla 1078. Patch from John Horne.
1053 PP/06 Stop make process more reliably on build failure.
1054 Bugzilla 1087. Patch from Heiko Schlittermann.
1056 PP/07 Make maildir_use_size_file an _expandable_ boolean.
1057 Bugzilla 1089. Patch from Heiko Schlittermann.
1059 PP/08 Handle ${run} returning more data than OS pipe buffer size.
1060 Bugzilla 1131. Patch from Holger Weiß.
1062 PP/09 Handle IPv6 addresses with SPF.
1063 Bugzilla 860. Patch from Wolfgang Breyha.
1065 PP/10 GnuTLS: support TLS 1.2 & 1.1.
1067 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
1070 PP/11 match_* no longer expand right-hand-side by default.
1071 New compile-time build option, EXPAND_LISTMATCH_RHS.
1072 New expansion conditions, "inlist", "inlisti".
1074 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
1076 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
1078 PP/14 fix log_write() format string regression from TF/03.
1079 Bugzilla 1152. Patch from Dmitry Isaikin.
1085 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
1087 PP/02 Harmonised TLS library version reporting; only show if debugging.
1088 Layout now matches that introduced for other libraries in 4.74 PP/03.
1090 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
1092 PP/04 New "dns_use_edns0" global option.
1094 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
1097 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
1098 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
1100 TK/01 Updated PolarSSL code to 0.14.2.
1101 Bugzilla 1097. Patch from Andreas Metzler.
1103 PP/07 Catch divide-by-zero in ${eval:...}.
1104 Fixes bugzilla 1102.
1106 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
1109 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
1110 format-string attack -- SECURITY: remote arbitrary code execution.
1112 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
1113 time unintentionally subject to list matching rules, letting the header
1114 cause arbitrary Exim lookups (of items which can occur in lists, *not*
1115 arbitrary string expansion). This allowed for information disclosure.
1117 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
1118 INT_MIN/-1 -- value coerced to INT_MAX.
1124 NM/01 Workround for PCRE version dependency in version reporting
1127 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
1128 This fixes portability to compilers other than gcc, notably
1129 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
1131 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
1132 makefiles for portability to HP-UX and POSIX correctness.
1134 PP/01 Permit LOOKUP_foo enabling on the make command-line.
1135 Also via indented variable definition in the Makefile.
1136 (Debugging by Oliver Heesakkers).
1138 PP/02 Restore caching of spamd results with expanded spamd_address.
1139 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
1141 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
1142 Improves build reliability. Fix from: Frank Elsner
1144 NM/02 Fix wide character breakage in the rfc2047 coding
1145 Fixes bug 1064. Patch from Andrey N. Oktyabrski
1147 NM/03 Allow underscore in dnslist lookups
1148 Fixes bug 1026. Patch from Graeme Fowler
1150 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
1151 Code patches from Adam Ciarcinski of NetBSD.
1153 NM/04 Fixed exiqgrep to cope with mailq missing size issue
1156 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
1157 is logged, to avoid truncation. Patch from John Horne.
1159 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
1160 Patch from Jakob Hirsch.
1162 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
1163 SQL string expansion failure details.
1164 Patch from Andrey Oktyabrski.
1166 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
1167 Patch from Simon Arlott.
1169 PP/09 New lookups functionality failed to compile on old gcc which rejects
1170 extern declarations in function scope.
1171 Patch from Oliver Fleischmann
1173 PP/10 Use sig_atomic_t for flags set from signal handlers.
1174 Check getgroups() return and improve debugging.
1175 Fixed developed for diagnosis in bug 927 (which turned out to be
1178 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
1179 Patch from Mark Zealey.
1181 PP/12 Bugzilla 1056: Improved spamd server selection.
1182 Patch from Mark Zealey.
1184 PP/13 Bugzilla 1086: Deal with maildir quota file races.
1185 Based on patch from Heiko Schlittermann.
1187 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
1188 Patch from Uwe Doering, sign-off by Michael Haardt.
1190 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
1191 variable declaration deep within a block. Bug and patch from
1194 PP/15 lookups-Makefile IRIX compatibilty coercion.
1196 PP/16 Make DISABLE_DKIM build knob functional.
1198 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
1199 Patch by Simon Arlott
1201 TF/03 Fix valgrind.h portability to C89 compilers that do not support
1202 variable argument macros. Our copy now differs from upstream.
1208 TF/01 Failure to get a lock on a hints database can have serious
1209 consequences so log it to the panic log.
1211 TF/02 Log LMTP confirmation messages in the same way as SMTP,
1212 controlled using the smtp_confirmation log selector.
1214 TF/03 Include the error message when we fail to unlink a spool file.
1216 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
1217 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
1218 for maintaining out-of-tree patches for some time.
1220 PP/01 Bugzilla 139: Documentation and portability issues.
1221 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
1222 Handle per-OS dynamic-module compilation flags.
1224 PP/02 Let /dev/null have normal permissions.
1225 The 4.73 fixes were a little too stringent and complained about the
1226 permissions on /dev/null. Exempt it from some checks.
1227 Reported by Andreas M. Kirchwitz.
1229 PP/03 Report version information for many libraries, including
1230 Exim version information for dynamically loaded libraries. Created
1231 version.h, now support a version extension string for distributors
1232 who patch heavily. Dynamic module ABI change.
1234 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
1235 privilege escalation vulnerability whereby the Exim run-time user
1236 can cause root to append content of the attacker's choosing to
1239 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
1242 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
1243 If dropping privileges for untrusted macros, we disabled normal logging
1244 on the basis that it would fail; for the Exim run-time user, this is not
1245 the case, and it resulted in successful deliveries going unlogged.
1246 Fixed. Reported by Andreas Metzler.
1252 PP/01 Date: & Message-Id: revert to normally being appended to a message,
1253 only prepend for the Resent-* case. Fixes regression introduced in
1254 Exim 4.70 by NM/22 for Bugzilla 607.
1256 PP/02 Include check_rfc2047_length in configure.default because we're seeing
1257 increasing numbers of administrators be bitten by this.
1259 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
1261 PP/03 Bugzilla 994: added openssl_options main configuration option.
1263 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
1265 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
1267 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
1269 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
1270 without a peer certificate, leading to a segfault because of an
1271 assumption that peers always have certificates. Be a little more
1272 paranoid. Problem reported by Martin Tscholak.
1274 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
1275 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
1276 NB: ClamAV planning to remove STREAM in "middle of 2010".
1277 CL also introduces -bmalware, various -d+acl logging additions and
1278 more caution in buffer sizes.
1280 PP/09 Implemented reverse_ip expansion operator.
1282 PP/10 Bugzilla 937: provide a "debug" ACL control.
1284 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
1286 PP/12 Bugzilla 973: Implement --version.
1288 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
1290 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
1292 PP/15 Bugzilla 816: support multiple condition rules on Routers.
1294 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
1295 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
1296 ignore trailing whitespace.
1298 JJ/02 prevent non-panic DKIM error from being sent to paniclog
1300 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
1303 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
1304 Notification from Dr Andrew Aitchison.
1306 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
1307 ExtendedDetectionInfo response format.
1308 Notification from John Horne.
1310 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
1313 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
1314 XSL and documented dependency on system catalogs, with examples of how
1317 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
1320 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
1321 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
1322 configuration file which is writeable by the Exim user or group.
1324 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
1325 of configuration files to cover files specified with the -C option if
1326 they are going to be used with root privileges, not just the default
1329 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
1330 option (effectively making it always true).
1332 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
1333 files to be used while preserving root privileges.
1335 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
1336 that rogue child processes cannot use them.
1338 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
1339 run-time user, instead of root.
1341 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
1342 Exim run-time user without dropping privileges.
1344 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
1345 result string, instead of calling string_vformat() twice with the same
1348 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
1349 for other users. Others should always drop root privileges if they use
1350 -C on the command line, even for a whitelisted configure file.
1352 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
1354 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
1360 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
1361 $data_path, and $header_path variables; fixed documentation bugs and
1364 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
1365 exipick to access non-standard spools, including the "frozen" queue
1368 NM/01 Bugzilla 965: Support mysql stored procedures.
1369 Patch from Alain Williams
1371 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
1373 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
1374 Patch from Andreas Metzler
1376 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
1377 Patch from Kirill Miazine
1379 NM/05 Bugzilla 671: Added umask to procmail example.
1381 JJ/03 installed exipick 20100323.0, fixing doc bug
1383 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
1384 directory. Notification and patch from Dan Rosenberg.
1386 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
1388 TK/02 Improve log output when DKIM signing operation fails.
1390 MH/01 Treat the transport option dkim_domain as a colon separated
1391 list, not as a single string, and sign the message with each element,
1392 omitting multiple occurences of the same signer.
1394 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
1395 Bugzilla 985, 986. Patch by Simon Arlott
1397 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
1398 Patch by Simon Arlott
1400 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
1401 MBX locking. Notification from Dan Rosenberg.
1407 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
1409 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
1411 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
1413 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
1415 NM/04 Bugzilla 915: Flag broken perl installation during build.
1421 TK/01 Added patch by Johannes Berg that expands the main option
1422 "spamd_address" if it starts with a dollar sign.
1424 TK/02 Write list of recipients to X-Envelope-Sender header when building
1425 the mbox-format spool file for content scanning (suggested by Jakob
1428 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
1429 (http://www.dcc-servers.net/) support via dccifd. Activated by
1430 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
1432 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
1433 by Mark Daniel Reidel <mr@df.eu>.
1435 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
1436 When building exim an external PCRE library is now needed -
1437 PCRE is a system library on the majority of modern systems.
1438 See entry on PCRE_LIBS in EDITME file.
1440 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
1441 conversation. Added nologin parameter to request.
1442 Patch contributed by Kirill Miazine.
1444 TF/01 Do not log submission mode rewrites if they do not change the address.
1446 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
1448 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
1449 log files in place. Contributed by Roberto Lima.
1451 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
1453 TF/03 Bugzilla 615: When checking the local_parts router precondition
1454 after a local_part_suffix or local_part_prefix option, Exim now
1455 does not use the address's named list lookup cache, since this
1456 contains cached lookups for the whole local part.
1458 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
1459 Robert Millan. Documentation is in experimental-spec.txt.
1461 TF/04 Bugzilla 668: Fix parallel build (make -j).
1463 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
1465 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
1466 Patch provided by Jan Srzednicki.
1468 TF/05 Leading white space used to be stripped from $spam_report which
1469 wrecked the formatting. Now it is preserved.
1471 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
1472 that they are available at delivery time.
1474 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
1476 TF/08 TLS error reporting now respects the incoming_interface and
1477 incoming_port log selectors.
1479 TF/09 Produce a more useful error message if an SMTP transport's hosts
1480 setting expands to an empty string.
1482 NM/06 Bugzilla 744: EXPN did not work under TLS.
1483 Patch provided by Phil Pennock.
1485 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
1486 Patch provided by Richard Godbee.
1488 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
1489 acl_smtp_notquit, added index entry.
1491 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
1492 Patch provided by Eugene Bujak.
1494 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
1495 accept(). Patch provided by Maxim Dounin.
1497 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
1498 Patch provided by Phil Pennock.
1500 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
1502 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
1503 Patch provided by Brad "anomie" Jorsch.
1505 NM/14 Bugzilla 622: Added timeout setting to transport filter.
1506 Patch provided by Dean Brooks.
1508 TK/05 Add native DKIM support (does not depend on external libraries).
1510 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
1511 Patch provided by Graeme Fowler.
1513 NM/16 Bugzilla 851: Documentation example syntax fix.
1515 NM/17 Changed NOTICE file to remove references to embedded PCRE.
1517 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
1520 NM/19 Bugzilla 745: TLS version reporting.
1521 Patch provided by Phil Pennock.
1523 NM/20 Bugzilla 167: bool: condition support.
1524 Patch provided by Phil Pennock.
1526 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
1527 clients. Patch provided by Phil Pennock.
1529 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
1530 Patch provided by Brad "anomie" Jorsch.
1532 NM/23 Bugzilla 687: Fix misparses in eximstats.
1533 Patch provided by Heiko Schlittermann.
1535 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
1536 Patch provided by Heiko Schlittermann.
1538 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
1539 plus update to original patch.
1541 NM/26 Bugzilla 799: Documentation correction for ratelimit.
1543 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
1544 Patch provided by David Brownlee.
1546 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
1548 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
1550 NM/30 Bugzilla 888: TLS documentation bugfixes.
1552 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
1554 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
1555 Unlike the original bugzilla I have changed all shell scripts in src tree.
1557 NM/33 Bugzilla 898: Transport filter timeout fix.
1558 Patch by Todd Rinaldo.
1560 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
1561 Patch by Serge Demonchaux.
1563 NM/35 Bugzilla 39: Base64 decode bug fixes.
1564 Patch by Jakob Hirsch.
1566 NM/36 Bugzilla 909: Correct connect() call in dcc code.
1568 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
1570 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
1572 NM/39 Bugzilla 911: Fixed MakeLinks build script.
1578 TK/01 Add preliminary DKIM support. Currently requires a forked version of
1579 ALT-N's libdkim that I have put here:
1580 http://duncanthrax.net/exim-experimental/
1582 Note to Michael Haardt: I had to rename some vars in sieve.c. They
1583 were called 'true' and it seems that C99 defines that as a reserved
1584 keyword to be used with 'bool' variable types. That means you could
1585 not include C99-style headers which use bools without triggering
1586 build errors in sieve.c.
1588 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
1589 as mailq or other aliases. Changed the --help handling significantly
1590 to do whats expected. exim_usage() emits usage/help information.
1592 SC/01 Added the -bylocaldomain option to eximstats.
1594 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
1596 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
1598 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
1604 PH/01 Another patch from the Sieve maintainer.
1606 PH/02 When an IPv6 address is converted to a string for single-key lookup
1607 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
1608 dots are used instead of colons so that keys in lsearch files need not
1609 contain colons. This was done some time before quoting was made available
1610 in lsearch files. However, iplsearch files do require colons in IPv6 keys
1611 (notated using the quote facility) so as to distinguish them from IPv4
1612 keys. This meant that lookups for IP addresses in host lists did not work
1613 for iplsearch lookups.
1615 This has been fixed by arranging for IPv6 addresses to be expressed with
1616 colons if the lookup type is iplsearch. This is not incompatible, because
1617 previously such lookups could never work.
1619 The situation is now rather anomolous, since one *can* have colons in
1620 ordinary lsearch keys. However, making the change in all cases is
1621 incompatible and would probably break a number of configurations.
1623 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
1626 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
1627 conversion specification without a maximum field width, thereby enabling
1628 a rogue spamd server to cause a buffer overflow. While nobody in their
1629 right mind would setup Exim to query an untrusted spamd server, an
1630 attacker that gains access to a server running spamd could potentially
1631 exploit this vulnerability to run arbitrary code as the Exim user.
1633 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
1634 $primary_hostname instead of what libspf2 thinks the hosts name is.
1636 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
1637 a directory entry by the name of the lookup key. Previously, if a
1638 symlink pointed to a non-existing file or a file in a directory that
1639 Exim lacked permissions to read, a lookup for a key matching that
1640 symlink would fail. Now it is enough that a matching directory entry
1641 exists, symlink or not. (Bugzilla 503.)
1643 PH/03 The body_linecount and body_zerocount variables are now exported in the
1646 PH/04 Added the $dnslist_matched variable.
1648 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
1649 This means they are set thereafter only if the connection becomes
1652 PH/06 Added the client_condition to authenticators so that some can be skipped
1653 by clients under certain conditions.
1655 PH/07 The error message for a badly-placed control=no_multiline_responses left
1656 "_responses" off the end of the name.
1658 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
1660 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
1661 (without spaces) instead of just copying the configuration text.
1663 PH/10 Added the /noupdate option to the ratelimit ACL condition.
1665 PH/11 Added $max_received_linelength.
1667 PH/12 Added +ignore_defer and +include_defer to host lists.
1669 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
1670 way in which PCRE > 7.0 is built.
1672 PH/14 Implemented queue_only_load_latch.
1674 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
1675 MAIL command. The effect was to mangle the value on 64-bit systems.
1677 PH/16 Another patch from the Sieve maintainer.
1679 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
1681 PH/18 If a system quota error occurred while trying to create the file for
1682 a maildir delivery, the message "Mailbox is full" was not appended to the
1683 bounce if the delivery eventually timed out. Change 4.67/27 below applied
1684 only to a quota excession during the actual writing of the file.
1686 PH/19 It seems that peer DN values may contain newlines (and other non-printing
1687 characters?) which causes problems in log lines. The DN values are now
1688 passed through string_printing() before being added to log lines.
1690 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
1691 and InterBase are left for another time.)
1693 PH/21 Added message_body_newlines option.
1695 PH/22 Guard against possible overflow in moan_check_errorcopy().
1697 PH/23 POSIX allows open() to be a macro; guard against that.
1699 PH/24 If the recipient of an error message contained an @ in the local part
1700 (suitably quoted, of course), incorrect values were put in $domain and
1701 $local_part during the evaluation of errors_copy.
1707 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
1708 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
1711 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
1712 issue a MAIL command.
1714 PH/02 In an ACL statement such as
1716 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
1718 if a client was not listed at all, or was listed with a value other than
1719 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
1720 the condition was not true (as it should be), so access was not denied.
1721 The bug was that the ! inversion was incorrectly passed on to the second
1722 item. This has been fixed.
1724 PH/03 Added additional dnslists conditions == and =& which are different from
1725 = and & when the dns lookup returns more than one IP address.
1727 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
1728 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
1730 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
1731 FSYNC, which compiles an option called disable_fsync that allows for
1732 bypassing fsync(). The documentation is heavily laced with warnings.
1734 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
1736 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
1737 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
1738 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
1739 including adding "make clean"; (3) Added -fPIC when compiling the test
1740 dynamically loaded module, to get rid of a warning.
1742 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
1743 message fails, move_frozen_messages = true and ignore_bounce_errors_after
1744 = 0s. The bug is otherwise harmless.
1746 PH/07 There was a bug in the dovecot authenticator such that the value of
1747 $auth1 could be overwritten, and so not correctly preserved, after a
1748 successful authentication. This usually meant that the value preserved by
1749 the server_setid option was incorrect.
1751 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
1753 PH/09 Installed PCRE release 7.0.
1755 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
1756 run for batched SMTP input. It is now run at the start of every message
1757 in the batch. While fixing this I discovered that the process information
1758 (output by running exiwhat) was not always getting set for -bs and -bS
1759 input. This is fixed, and it now also says "batched" for BSMTP.
1761 PH/11 Added control=no_pipelining.
1763 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
1764 patch, slightly modified), and move the expansion of helo_data till after
1765 the connection is made in the smtp transport (so it can use these
1768 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
1770 PH/14 Added log_selector = +pid.
1772 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
1774 PH/16 Add ${if forany and ${if forall.
1776 PH/17 Added dsn_from option to vary the From: line in DSNs.
1778 PH/18 Flush SMTP output before performing a callout, unless control =
1779 no_callout_flush is set.
1781 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
1782 was true (the default) a successful delivery failed to delete the retry
1783 item, thus causing premature timeout of the address. The bug is now
1786 PH/20 Added hosts_avoid_pipelining to the smtp transport.
1788 PH/21 Long custom messages for fakedefer and fakereject are now split up
1789 into multiline reponses in the same way that messages for "deny" and
1790 other ACL rejections are.
1792 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
1793 with slight modification.
1795 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
1796 draft, changing the syntax and factoring some duplicate code".
1798 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
1799 for deliveries of the second and subsequent messages over the same SMTP
1802 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
1803 ${reduce, with only minor "tidies".
1805 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
1807 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
1808 expansion side effects.
1810 PH/27 When a message times out after an over-quota error from an Exim-imposed
1811 quota, the bounce message says "mailbox is full". This message was not
1812 being given when it was a system quota that was exceeded. It now should
1815 MH/03 Made $recipients available in local_scan(). local_scan() already has
1816 better access to the recipient list through recipients_list[], but
1817 $recipients can be useful in postmaster-provided expansion strings.
1819 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
1820 in the case of a MAIL command with additional options following the
1821 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
1822 were accidentally chopped off.
1824 PH/29 SMTP synchronization checks are implemented when a command is read -
1825 there is a check that no more input is waiting when there shouldn't be
1826 any. However, for some commands, a delay in an ACL can mean that it is
1827 some time before the response is written. In this time, more input might
1828 arrive, invalidly. So now there are extra checks after an ACL has run for
1829 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
1830 pipelining has not been advertised.
1832 PH/30 MH's patch to allow iscntrl() characters to be list separators.
1834 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
1835 returned in the SMTP response when smtp_return_error_details was false.
1836 This has been fixed.
1838 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
1839 instead of the C I/O that was originally supplied, because problems were
1840 reported on Solaris.
1842 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
1843 Exim which did not show up earlier: it was assuming that a call to
1844 SSL_CTX_set_info_callback() might give an error value. In fact, there is
1845 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
1846 was a macro that became an assignment, so it seemed to work. This has
1847 changed to a proper function call with a void return, hence the compile
1848 error. Exim's code has been fixed.
1850 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
1853 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
1855 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
1857 PH/37 If a message is not accepted after it has had an id assigned (e.g.
1858 because it turns out to be too big or there is a timeout) there is no
1859 "Completed" line in the log. When some messages of this type were
1860 selected by exigrep, they were listed as "not completed". Others were
1861 picked up by some special patterns. I have improved the selection
1862 criteria to be more general.
1864 PH/38 The host_find_failed option in the manualroute router can now be set
1865 to "ignore", to completely ignore a host whose IP address cannot be
1866 found. If all hosts are ignored, the behaviour is controlled by the new
1867 host_all_ignored option.
1869 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
1870 homing or because of multiple MX records with /mx) generated more than
1871 one IP address, and the following item turned out to be the local host,
1872 all the secondary addresses of the first item were incorrectly removed
1873 from the list, along with the local host and any following hosts (which
1874 is what is supposed to happen).
1876 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
1877 whoever called Exim into the -H file. In the case of the daemon it was
1878 behaving confusingly. When first started, it used values for whoever
1879 started the daemon, but after a SIGHUP it used the Exim user (because it
1880 calls itself on a restart). I have changed the code so that it now always
1883 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
1884 message are rejected with the same error (e.g. no authentication or bad
1885 sender address), and a DATA command is nevertheless sent (as can happen
1886 with PIPELINING or a stupid MUA), the error message that was given to the
1887 RCPT commands is included in the rejection of the DATA command. This is
1888 intended to be helpful for MUAs that show only the final error to their
1891 PH/42 Another patch from the Sieve maintainer.
1893 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
1894 Eximstats - Fixed some broken HTML links and added missing column headers
1896 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
1897 columns for Rejects, Temp Rejects, Ham, and Spam rows.
1899 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
1901 PH/43 Yet another patch from the Sieve maintainer.
1903 PH/44 I found a way to check for a TCP/IP connection going away before sending
1904 the response to the final '.' that terminates a message, but only in the
1905 case where the client has not sent further data following the '.'
1906 (unfortunately, this is allowed). However, in many cases there won't be
1907 any further data because there won't be any more messages to send. A call
1908 to select() can be used: if it shows that the input is "ready", there is
1909 either input waiting, or the socket has been closed. An attempt to read
1910 the next input character can distinguish the two cases. Previously, Exim
1911 would have sent an OK response which the client would never have see.
1912 This could lead to message repetition. This fix should cure that, at
1913 least in a lot of common cases.
1915 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
1916 advertised in response to EHLO.
1922 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
1923 fixed by 4.65/MH/01 (is this a record?) are fixed:
1925 (i) An empty string was always treated as zero by the numeric comparison
1926 operators. This behaviour has been restored.
1928 (ii) It is documented that the numeric comparison operators always treat
1929 their arguments as decimal numbers. This was broken in that numbers
1930 starting with 0 were being interpreted as octal.
1932 While fixing these problems I realized that there was another issue that
1933 hadn't been noticed. Values of message_size_limit (both the global option
1934 and the transport option) were treated as octal if they started with 0.
1935 The documentation was vague. These values are now always treated as
1936 decimal, and I will make that clear in the documentation.
1942 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
1943 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
1946 MH/01 Don't check that the operands of numeric comparison operators are
1947 integers when their expansion is in "skipping" mode (fixes bug
1948 introduced by 4.64-PH/07).
1950 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
1951 child addresses, Exim now panics and dies. Previously, because the count
1952 is held in a short int, deliveries were likely to be lost. As such a
1953 large number of recipients for a single message is ridiculous
1954 (performance will be very, very poor), I have chosen to impose a limit
1955 rather than extend the field.
1961 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
1962 leftover -K file (the existence of which was triggered by #402).
1963 While we were at it, introduced process PID as part of the -K
1964 filename. This should rule out race conditions when creating
1967 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
1968 processing considerably. Previous code took too long for large mails,
1969 triggering a timeout which in turn triggers #401.
1971 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
1972 in the DK code in transports.c. sendfile() is not really portable,
1973 hence the _LINUX specificness.
1975 TF/01 In the add_headers option to the mail command in an Exim filter,
1976 there was a bug that Exim would claim a syntax error in any
1977 header after the first one which had an odd number of characters
1980 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
1981 callout verification, Exim cached a "reject" for the entire domain. This
1982 is correct for most verifications, but it is not correct for a recipient
1983 verification with use_sender or use_postmaster set, because in that case
1984 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
1985 case of MAIL FROM:<> rejection from other early rejections (e.g.
1986 rejection of HELO). When verifying a recipient using a non-null MAIL
1987 address, the cache is ignored if it shows MAIL FROM:<> rejection.
1988 Whatever the result of the callout, the value of the domain cache is
1989 left unchanged (for any other kind of callout, getting as far as trying
1990 RCPT means that the domain itself is ok).
1992 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
1995 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
1996 manifest itself as EPIPE rather than ECONNECT. When tidying away a
1997 session, the daemon ignores ECONNECT errors and logs others; it now
1998 ignores EPIPE as well.
2000 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
2001 (quoted-printable decoding).
2003 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
2004 later the small subsequent patch to fix an introduced bug.
2006 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
2008 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
2010 PH/08 An error is now given if message_size_limit is specified negative.
2012 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
2013 to be given (somewhat) arbitrary names.
2015 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
2018 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
2019 miscellaneous code fixes
2021 PH/10 Added the log_reject_target ACL modifier to specify where to log
2024 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
2025 hostname. This is wrong, because it relates to the incoming message (and
2026 probably the interface on which it is arriving) and not to the outgoing
2027 callout (which could be using a different interface). This has been
2028 changed to use the value of the helo_data option from the smtp transport
2029 instead - this is what is used when a message is actually being sent. If
2030 there is no remote transport (possible with a router that sets up host
2031 addresses), $smtp_active_hostname is used.
2033 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
2034 tweaks were necessary in order to get it to work (see also 21 below):
2035 (a) The code assumed that strncpy() returns a negative number on buffer
2036 overflow, which isn't the case. Replaced with Exim's string_format()
2038 (b) There were several signed/unsigned issues. I just did the minimum
2039 hacking in of casts. There is scope for a larger refactoring.
2040 (c) The code used strcasecmp() which is not a standard C function.
2041 Replaced with Exim's strcmpic() function.
2042 (d) The code set only $1; it now sets $auth1 as well.
2043 (e) A simple test gave the error "authentication client didn't specify
2044 service in request". It would seem that Dovecot has changed its
2045 interface. Fortunately there's a specification; I followed it and
2046 changed what the client sends and it appears to be working now.
2048 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
2051 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
2052 address A is aliased to B and C, where B exists and C does not. Without
2053 -v the output is "A verified" because verification stops after a
2054 successful redirection if more than one address is generated. However,
2055 with -v the child addresses are also verified. Exim was outputting "A
2056 failed to verify" and then showing the successful verification for C,
2057 with its parentage. It now outputs "B failed to verify", showing B's
2058 parentage before showing the successful verification of C.
2060 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
2061 look up a TXT record in a specific list after matching in a combined
2064 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
2065 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
2066 they consult the DNS. I had assumed they would set it the way they
2067 wanted; and indeed my experiments on Linux seem to show that in some
2068 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
2069 To be on the safe side, however, I have now made the interface to
2070 host_find_byname() similar to host_find_bydns(), with an argument
2071 containing the DNS resolver options. The host_find_byname() function now
2072 sets these options at its start, just as host_find_bydns() does. The smtp
2073 transport options dns_qualify_single and dns_search_parents are passed to
2074 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
2075 of host_find_byname() use the default settings of RES_DEFNAMES
2076 (qualify_single) but not RES_DNSRCH (search_parents).
2078 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
2079 spool_read_header() do less string testing, by means of a preliminary
2080 switch on the second character of optional "-foo" lines. (This is
2081 overdue, caused by the large number of possibilities that now exist.
2082 Originally there were few.) While I was there, I also converted the
2083 str(n)cmp tests so they don't re-test the leading "-" and the first
2084 character, in the hope this might squeeze out yet more improvement.
2086 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
2087 flag allowing group syntax was set by the header_syntax check but not
2088 turned off, possible causing trouble later; (2) The flag was not being
2089 set at all for the header_verify test, causing "group"-style headers to
2090 be rejected. I have now set it in this case, and also caused header_
2091 verify to ignore an empty address taken from a group. While doing this, I
2092 came across some other cases where the code for allowing group syntax
2093 while scanning a header line wasn't quite right (mostly, not resetting
2094 the flag correctly in the right place). These bugs could have caused
2095 trouble for malformed header lines. I hope it is now all correct.
2097 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
2098 with the "reply" argument non-NULL. The code, however (which originally
2099 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
2100 but it didn't always do it. This confused somebody who was copying the
2101 code for some other use. I have removed all the tests.
2103 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
2104 feature that was used to support insecure browsers during the U.S. crypto
2105 embargo. It requires special client support, and Exim is probably the
2106 only MTA that supported it -- and would never use it because real RSA is
2107 always available. This code has been removed, because it had the bad
2108 effect of slowing Exim down by computing (never used) parameters for the
2109 RSA_EXPORT functionality.
2111 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
2112 authenticator to fail if there's a tab character in the incoming data
2113 (there should never be unless someone is messing about, as it's supposed
2114 to be base64-encoded). Also added, on Timo's advice, the "secured" option
2115 if the connection is using TLS or if the remote IP is the same as the
2116 local IP, and the "valid-client-cert option" if a client certificate has
2119 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
2120 authenticators. This can be used for authorization after authentication
2121 succeeds. (In the case of plaintext, it servers for both authentication
2124 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
2125 if any retry times were supplied.
2127 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
2128 connection, where there is no client IP address to check. In this
2129 situation, the verify now always succeeds.
2131 PH/25 Applied John Jetmore's -Mset patch.
2133 PH/26 Added -bem to be like -Mset, but loading a message from a file.
2135 PH/27 In a string expansion for a processed (not raw) header when multiple
2136 headers of the same name were present, leading whitespace was being
2137 removed from all of them, but trailing whitespace was being removed only
2138 from the last one. Now trailing whitespace is removed from each header
2139 before concatenation. Completely empty headers in a concatenation (as
2140 before) are ignored.
2142 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
2143 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
2145 PH/29 [Removed. This was a change that I later backed out, and forgot to
2146 correct the ChangeLog entry (that I had efficiently created) before
2147 committing the later change.]
2149 PH/30 Exim was sometimes attempting to deliver messages that had suffered
2150 address errors (4xx response to RCPT) over the same connection as other
2151 messages routed to the same hosts. Such deliveries are always "forced",
2152 so retry times are not inspected. This resulted in far too many retries
2153 for the affected addresses. The effect occurred only when there were more
2154 hosts than the hosts_max_try setting in the smtp transport when it had
2155 the 4xx errors. Those hosts that it had tried were not added to the list
2156 of hosts for which the message was waiting, so if all were tried, there
2157 was no problem. Two fixes have been applied:
2159 (i) If there are any address or message errors in an SMTP delivery, none
2160 of the hosts (tried or untried) are now added to the list of hosts
2161 for which the message is waiting, so the message should not be a
2162 candidate for sending over the same connection that was used for a
2163 successful delivery of some other message. This seems entirely
2164 reasonable: after all the message is NOT "waiting for some host".
2165 This is so "obvious" that I'm not sure why it wasn't done
2166 previously. Hope I haven't missed anything, but it can't do any
2167 harm, as the worst effect is to miss an optimization.
2169 (ii) If, despite (i), such a delivery is accidentally attempted, the
2170 routing retry time is respected, so at least it doesn't keep
2171 hammering the server.
2173 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
2174 in ${readsocket because some servers need this prod.
2176 PH/32 Added some extra debug output when updating a wait-xxx database.
2178 PH/33 The hint "could be header name not terminated by colon", which has been
2179 given for certain expansion errors for a long time, was not being given
2180 for the ${if def:h_colon_omitted{... case.
2182 PH/34 The spec says: "With one important exception, whenever a domain list is
2183 being scanned, $domain contains the subject domain." There was at least
2184 one case where this was not true.
2186 PH/35 The error "getsockname() failed: connection reset by peer" was being
2187 written to the panic log as well as the main log, but it isn't really
2188 panic-worthy as it just means the connection died rather early on. I have
2189 removed the panic log writing for the ECONNRESET error when getsockname()
2192 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
2193 runs only) independently of the message's sender address. This meant
2194 that, if the 4xx error was in fact related to the sender, a different
2195 message to the same recipient with a different sender could confuse
2196 things. In particualar, this can happen when sending to a greylisting
2197 server, but other circumstances could also provoke similar problems.
2198 I have changed the default so that the retry time for these errors is now
2199 based a combination of the sender and recipient addresses. This change
2200 can be overridden by setting address_retry_include_sender=false in the
2203 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
2204 remote server are returned as part of bounce messages. This was not
2205 happening for LMTP over a pipe (the lmtp transport), but now it is the
2206 same for both kinds of LMTP.
2208 PH/38 Despite being documented as not happening, Exim was rewriting addresses
2209 in header lines that were in fact CNAMEs. This is no longer the case.
2211 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
2212 and queue runs started by the daemon processed all messages. This has
2213 been fixed so that -R and -S can now usefully be given with -q<time>.
2215 PH/40 Import PCRE release 6.7 (fixes some bugs).
2217 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
2219 PH/42 Give an error if -q is specified more than once.
2221 PH/43 Renamed the variables $interface_address and $interface_port as
2222 $received_ip_address and $received_port, to make it clear that these
2223 values apply to message reception, and not to the outgoing interface when
2224 a message is delivered. (The old names remain recognized, of course.)
2226 PH/44 There was no timeout on the connect() call when using a Unix domain
2227 socket in the ${readsocket expansion. There now is.
2229 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
2230 be meaningful with "accept".
2232 SC/01 Eximstats V1.43
2233 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
2235 SC/02 Eximstats V1.44
2236 Use a glob alias rather than an array ref in the generated
2237 parser. This improves both readability and performance.
2239 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
2240 Collect SpamAssassin and rejection statistics.
2241 Don't display local sender or destination tables unless
2242 there is data to show.
2243 Added average volumes into the top table text output.
2245 SC/04 Eximstats V1.46
2246 Collect data on the number of addresses (recipients)
2247 as well as the number of messages.
2249 SC/05 Eximstats V1.47
2250 Added 'Message too big' to the list of mail rejection
2251 reasons (thanks to Marco Gaiarin).
2253 SC/06 Eximstats V1.48
2254 Mainlog lines which have GMT offsets and are too short to
2255 have a flag are now skipped.
2257 SC/07 Eximstats V1.49 (Alain Williams)
2258 Added the -emptyok flag.
2260 SC/08 Eximstats V1.50
2261 Fixes for obtaining the IP address from reject messages.
2263 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
2264 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
2265 whitesspace changes from 4.64-PH/27
2267 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
2270 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
2271 are found, allow negative numbers in numeric criteria)
2273 JJ/06 exipick.20061117.2, added new $message_body_missing variable
2275 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
2276 to match changes made in 4.64-PH/43
2278 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
2280 PH/47 Put in an explicit test for a DNS lookup of an address record where the
2281 "domain" is actually an IP address, and force a failure. This locks out
2282 those revolvers/nameservers that support "A-for-A" lookups, in
2283 contravention of the specifications.
2285 PH/48 When a host name was looked up from an IP address, and the subsequent
2286 forward lookup of the name timed out, the host name was left in
2287 $sender_host_name, contrary to the specification.
2289 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
2290 restricted to single-key lookups, Exim was not diagnosing an error if
2291 * or *@ was used with a query-style lookup.
2293 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
2295 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
2296 long ago, but noone interested enough thought of it. Let's just say that
2297 the "1.1" means that there are some new functions that weren't there at
2298 some point in the past.
2300 PH/51 Error processing for expansion failure of helo_data from an smtp
2301 transport during callout processing was broken.
2303 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
2304 tested/used via the -bh/-bhc/-bs options.
2306 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
2307 bug, fixed in subsequent PCRE releases).
2309 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
2310 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
2312 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
2318 SC/01 Use a glob alias rather than an array ref in eximstats generated
2319 parser. This improves both readability and performance.
2321 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
2322 Don't display local sender or destination tables in eximstats unless
2323 there is data to show.
2324 Added average volumes into the eximstats top table text output.
2326 SC/03 Collect data on the number of addresses (recipients) as well
2327 as the number of messages in eximstats.
2329 TF/01 Correct an error in the documentation for the redirect router. Exim
2330 does (usually) call initgroups() when daemonizing.
2332 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
2333 with consistent privilege compared to when running as a daemon.
2335 TF/03 Note in the spec that $authenticated_id is not set for local
2336 submissions from trusted users.
2338 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
2339 Thanks to Dean Brooks <dean@iglou.com> for the patch.
2341 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
2342 by adding some example configuration directives to the default
2343 configuration file. A little bit of work is required to uncomment the
2344 directives and define how usernames and passwords are checked, but
2345 there is now a framework to start from.
2347 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
2348 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
2349 without this. I don't know how relevant this is to other LDAP libraries.
2351 PH/02 Add the verb name to the "unknown ACL verb" error.
2353 PH/03 Magnus Holmgren's patch for filter_prepend_home.
2355 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
2357 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
2358 directory not expanded when it should be if an expanded home directory
2359 was set for the address (which is overridden by the transport).
2361 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
2364 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
2365 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
2366 because it is too late at that time, and has no effect.
2368 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
2369 security issue with \' (bugzilla #107). I could not use the
2370 PQescapeStringConn() function, because it needs a PGconn value as one of
2373 PH/08 When testing addresses using -bt, indicate those final addresses that
2374 are duplicates that would not cause an additional delivery. At least one
2375 person was confused, thinking that -bt output corresponded to deliveries.
2376 (Suppressing duplicates isn't a good idea as you lose the information
2377 about possibly different redirections that led to the duplicates.)
2379 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
2380 systems where poll() doesn't work, in particular OS X.
2382 PH/10 Added more information to debugging output for retry time not reached.
2384 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
2385 operations in malware.c.
2387 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
2390 PH/13 If write_rejectlog was set false when logging was sent to syslog with
2391 syslog_duplication set false, log lines that would normally be written
2392 both the the main log and to the reject log were not written to syslog at
2395 PH/14 In the default configuration, change the use of "message" in ACL warn
2396 statements to "add_header".
2398 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
2399 not followed by a command (e.g. "seen endif").
2401 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
2402 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
2405 PH/17 Added extra conditions to the default value of delay_warning_condition
2409 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
2410 { match{$h_precedence:}{(?i)bulk|list|junk} } \
2411 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
2414 The Auto-Submitted: and various List- headers are standardised, whereas I
2415 don't think Precedence: ever was.
2417 PH/18 Refactored debugging code in route_finduser() to show more information,
2418 in particular, the error code if getpwnam() issues one.
2420 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
2421 This is apparently needed in addition to the PH/07 change above to avoid
2422 any possible encoding problems.
2424 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
2425 but not after initializing Perl.
2427 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
2428 output them only if debugging. By default they are written stderr,
2429 apparently, which is not desirable.
2431 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
2434 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
2437 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
2439 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
2440 authenticated or an ident call has been made. Suppress the default
2441 values for $authenticated_id and $authenticated_sender (but permit -oMai
2442 and -oMas) when testing with -bh.
2444 PH/24 Re-jigged the order of the tests in the default configuration so that the
2445 tests for valid domains and recipients precede the DNS black list and CSA
2446 tests, on the grounds that those ones are more expensive.
2448 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
2449 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
2450 This bug exists in every version of Exim that I still have, right back to
2453 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
2454 However, an attempt to turn on case-sensitivity in a regex key by
2455 including (?-i) didn't work because the subject string was already
2456 lowercased, and the effects were non-intuitive. It turns out that a
2457 one-line patch can be used to allow (?-i) to work as expected.
2463 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
2464 other effects) broke the use of negated acl sub-conditions.
2466 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
2469 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
2470 "Deny" causes Exim to reject the incoming connection with a 554 error.
2471 Unfortunately, if there is a major crisis, such as a disk failure,
2472 tcp-wrappers gives "deny", whereas what one would like would be some
2473 kind of temporary error. A kludge has been added to help with this.
2474 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
2475 554 error is used if errno is still zero or contains ENOENT (which occurs
2476 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
2479 PH/03 Add -lutil to the default FreeBSD LIBS setting.
2481 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
2482 errors. Otherwise a message that provokes a temporary error (when other
2483 messages do not) can cause a whole host to time out.
2485 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
2486 addresses were routed directly to files or pipes from a redirect router.
2487 File deliveries just didn't batch; pipe deliveries might have suffered
2490 PH/06 A failure to get a lock for a hints database would erroneously always say
2491 "Failed to get write lock", even when it was really a read lock.
2493 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
2494 of 0600. This has been changed to use the value of the lockfile_mode
2495 option (which defaults to 0600).
2497 PH/08 Applied small patch from the Sieve maintainer.
2499 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
2500 folder from quota calculations, a direct delivery into this folder messed
2501 up the contents of the maildirsize file. This was because the regex was
2502 used only to exclude .Trash (or whatever) when the size of the mailbox
2503 was calculated. There was no check that a delivery was happening into an
2504 excluded directory. This bug has been fixed by ignoring all quota
2505 processing for deliveries into excluded directories.
2507 PH/10 Added the maildirfolder_create_regex option to appendfile.
2513 PH/01 The code for finding all the local interface addresses on a FreeBSD
2514 system running IPv6 was broken. This may well have applied to all BSD
2515 systems, as well as to others that have similar system calls. The broken
2516 code found IPv4 interfaces correctly, but gave incorrect values for the
2517 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
2518 that it would not match correctly against @[] and not recognize the IPv6
2521 PH/02 The ipliteral router was not recognizing addresses of the form user@
2522 [ipv6:....] because it didn't know about the "ipv6:" prefix.
2524 PH/03 Added disable_ipv6.
2526 PH/04 Changed $reply_address to use the raw form of the headers instead of the
2527 decoded form, because it is most often used to construct To: headers
2528 lines in autoreplies, and the decoded form may well be syntactically
2529 invalid. However, $reply_address has leading white space removed, and all
2530 newlines turned into spaces so that the autoreply transport does not
2533 PH/05 If group was specified without a user on a router, and no group or user
2534 was specified on a transport, the group from the router was ignored.
2536 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
2537 for visible compile-time settings that can be used to change these
2538 numbers, for those that want even more. Backwards compatibility with old
2539 spool files has been maintained. However, going back to a previous Exim
2540 release will lost any variables that are in spool files.
2542 PH/07 Two small changes when running in the test harness: increase delay when
2543 passing a TCP/IP connection to a new process, in case the original
2544 process has to generate a bounce, and remove special handling of
2545 127.0.0.2 (sic), which is no longer necessary.
2547 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
2548 be the same on different OS.
2550 PH/09 Moved a debug statement in filter processing to avoid a race problem when
2553 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
2554 whether --show-vars was specified or not
2556 JJ/02 exipick: Added support for new ACL variable spool format introduced
2559 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
2560 syntactically invalid From: or Reply-to: line, and a filter used this to
2561 generate an autoreply, and therefore failed to obtain an address for the
2562 autoreply, Exim could try to deliver to a non-existent relative file
2563 name, causing unrelated and misleading errors. What now happens is that
2564 it logs this as a hard delivery error, but does not attempt to create a
2567 PH/11 The exinext utility has a -C option for testing purposes, but although
2568 the given file was scanned by exinext itself; it wasn't being passed on
2569 when Exim was called.
2571 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
2572 an end-of-file indication when reading a command response.
2574 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
2575 compiled. In many other places in Exim, IPv6 addresses are always
2576 recognized, so I have changed this. It also means that IPv4 domain
2577 literals of the form [IPV4:n.n.n.n] are now always recognized.
2579 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
2580 used if the router is not running as root, for example, when verifying at
2581 ACL time, or when using -bh. The debugging output from this situation was
2582 non-existent - all you got was a failure to exec. I have made two
2585 (a) Failures to set uid/gid, the current directory, or a process leader
2586 in a subprocess such as that created by queryprogram now generate
2587 suitable debugging ouput when -d is set.
2589 (b) The queryprogram router detects when it is not running as root,
2590 outputs suitable debugging information if -d is set, and then runs
2591 the subprocess without attempting to change uid/gid.
2593 PH/15 Minor change to Makefile for building test_host (undocumented testing
2596 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
2597 additional section of a DNS packet that returns MX or SRV records.
2598 Instead, it always explicitly searches for A/AAAA records. This avoids
2599 major problems that occur when a DNS server includes only records of one
2600 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
2601 fixed another bug: if SRV records were looked up and the corresponding
2602 address records were *not* found in the additional section, the port
2603 values from the SRV records were lost.
2605 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
2606 using the correct key (the original address) when searching the retry
2607 rules in order to find which one to use for generating the retry hint.
2609 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
2610 adding the default one. Similarly, if it contains a Reply-To: header, the
2611 errors_reply_to option, if set, is not used.
2613 PH/19 When calculating a retry time, Exim used to measure the "time since
2614 failure" by looking at the "first failed" field in the retry record. Now
2615 it does not use this if it is later than than the arrival time of the
2616 message. Instead it uses the arrival time. This makes for better
2617 behaviour in cases where some deliveries succeed, thus re-setting the
2618 "first failed" field. An example is a quota failure for a huge message
2619 when small messages continue to be delivered. Without this change, the
2620 "time since failure" will always be short, possible causing more frequent
2621 delivery attempts for the huge message than are intended.
2622 [Note: This change was subsequently modified - see PH/04 for 4.62.]
2624 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
2625 $1, $2, $3) because the numerical variables can be reset during some
2626 expansion items (e.g. "match"), thereby losing the authentication data.
2628 PH/21 Make -bV show the size of off_t variables so that the test suite can
2629 decide whether to run tests for quotas > 2G.
2631 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
2632 mailbox_size, and mailbox_filecount in the appendfile transport. If a
2633 filecount value is greater than 2G or if a quota value is greater than 2G
2634 on a system where the size of off_t is not greater than 4, a panic error
2637 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
2638 never match. The debug and -bh output now contains an explicit error
2639 message indicating a malformed IPv4 address or mask.
2641 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
2642 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
2643 PH/23 above applies.
2645 PH/25 Do not write to syslog when running in the test harness. The only
2646 occasion when this arises is a failure to open the main or panic logs
2647 (for which there is an explicit test).
2649 PH/26 Added the /no_tell option to "control=freeze".
2651 PH/27 If a host name lookup failed very early in a connection, for example, if
2652 the IP address matched host_lookup and the reverse lookup yielded a name
2653 that did not have a forward lookup, an error message of the form "no IP
2654 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
2655 could be logged. Now it outputs the IP address instead of "NULL".
2657 PH/28 An enabling patch from MH: add new function child_open_exim2() which
2658 allows the sender and the authenticated sender to be set when
2659 submitting a message from within Exim. Since child_open_exim() is
2660 documented for local_scan(), the new function should be too.
2662 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
2663 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
2664 results in an empty string is now treated as unset.
2666 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
2668 PH/31 Added sender_verify_fail logging option.
2670 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
2671 needed by RFC 822 but not by RFC 2822 was commented out. I have now
2672 tidied the source and removed it altogether.
2674 PH/33 When a queue run was abandoned because the load average was too high, a
2675 log line was always written; now it is written only if the queue_run log
2676 selector is set. In addition, the log line for abandonment now contains
2677 information about the queue run such as the pid. This is always present
2678 in "start" and "stop" lines but was omitted from the "abandon" line.
2680 PH/34 Omit spaces between a header name and the colon in the error message that
2681 is given when verify = headers_syntax fails (if there are lots of them,
2682 the message gets confusing).
2684 PH/35 Change the default for dns_check_names_pattern to allow slashes within
2685 names, as there are now some PTR records that contain slashes. This check
2686 is only to protect against broken name servers that fall over on strange
2687 characters, so the fact that it applies to all lookups doesn't matter.
2689 PH/36 Now that the new test suite is complete, we can remove some of the
2690 special code in Exim that was needed for the old test suite. For example,
2691 sorting DNS records because real resolvers return them in an arbitrary
2692 order. The new test suite's fake resolver always returns records in the
2695 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
2696 bounces) except when queue_only is set, to avoid logging races between
2697 the different processes.
2699 PH/38 Panic-die if .include specifies a non-absolute path.
2701 PH/39 A tweak to the "H" retry rule from its user.
2703 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
2704 a label. They prevented compilation on older perls.
2706 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
2707 a warning to be raised on newish perls.
2709 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
2710 on queue. Changes to match documented behaviour of showing count of
2711 messages matching specified criteria.
2713 PH/40 Changed the default ident timeout from 30s to 5s.
2715 PH/41 Added support for the use of login_cap features, on those BSD systems
2716 that have them, for controlling the resources used by pipe deliveries.
2718 PH/42 The content-scanning code uses fopen() to create files in which to put
2719 message data. Previously it was not paying any attention to the mode of
2720 the files. Exim runs with umask(0) because the rest of the code creates
2721 files with open(), and sets the required mode explicitly. Thus, these
2722 files were ending up world-writeable. This was not a big issue, because,
2723 being within the spool directory, they were not world-accessible. I have
2724 created a function called modefopen, which takes an additional mode
2725 argument. It sets umask(777), creates the file, chmods it to the required
2726 mode, then resets the umask. All the relevant calls to fopen() in the
2727 content scanning code have been changed to use this function.
2729 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
2730 to 24 hours. This avoids potential overflow problems when processing G
2731 and H retry rules. I suspect nobody ever tinkers with this value.
2733 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
2735 PH/45 When the plaintext authenticator is running as a client, the server's
2736 challenges are checked to ensure they are valid base64 strings. By
2737 default, the authentication attempt is cancelled if an invalid string is
2738 received. Setting client_ignore_invalid_base64 true ignores these errors.
2739 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
2740 they are received. Thus, the responses can be made to depend on the
2741 challenges. If an invalid string is ignored, an empty string is placed in
2744 PH/46 Messages that are created by the autoreply transport now contains a
2745 References: header, in accordance with RFCs 2822 and 3834.
2747 PH/47 Added authenticated_sender_force to the smtp transport.
2749 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
2751 PH/49 Installed latest patch from the Sieve maintainer.
2753 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
2754 was also set, the appendfile transport was unnecessarily scanning a
2755 directory of message files (e.g. for maildir delivery) to find the count
2756 of files (along with the size), even though it did not need this
2757 information. It now does the scan only if it needs to find either the
2758 size of the count of files.
2760 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
2762 PH/52 Two bugs concerned with error handling when the smtp transport is
2765 (i) Exim was not creating retry information for temporary errors given
2766 for individual recipients after the DATA command when the smtp transport
2767 was used in LMTP mode. This meant that they could be retried too
2768 frequently, and not timed out correctly.
2770 (ii) Exim was setting the flag that allows error details to be returned
2771 for LMTP errors on RCPT commands, but not for LMTP errors for individual
2772 recipients that were returned after the DATA command.
2774 PH/53 This is related to PH/52, but is more general: for any failing address,
2775 when detailed error information was permitted to be returned to the
2776 sender, but the error was temporary, then after the final timeout, only
2777 "retry timeout exceeded" was returned. Now it returns the full error as
2778 well as "retry timeout exceeded".
2780 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
2781 do this, and (what is worse) MTAs that accept it.
2783 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
2784 will now be deprecated.
2786 PH/56 New os.c-cygwin from the Cygwin maintainer.
2788 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
2789 formats (previously only available in exim formats via -bpr, -bpru,
2790 and -bpra. Now also available in native and exiqgrep formats)
2792 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
2793 with very large, slow to parse queues
2795 JJ/08 exipick: added ! as generic prefix to negate any criteria format
2797 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
2799 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
2800 responses to authentication challenges, though it was showing the
2801 challenges; (ii) I've removed the CR characters from the debug output for
2804 PH/58 Allow for the insertion of a newline as well as a space when a string
2805 is turned into more than one encoded-word during RFC 2047 encoding. The
2806 Sieve code now uses this.
2808 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
2809 data_4xx, lost_connection, tls_required.
2811 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
2812 message was being sent as an SMTP response.
2814 PH/61 Add -l and -k options to exicyclog.
2816 PH/62 When verifying, if an address was redirected to one new address, so that
2817 verification continued, and the new address failed or deferred after
2818 having set something in $address_data, the value of $address_data was not
2819 passed back to the ACL. This was different to the case when no
2820 redirection occurred. The value is now passed back in both cases.
2822 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
2823 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
2824 use login_cap.h, so on its own it isn't the distinguishing feature. The
2825 new name refers directly to the setclassresources() function.
2827 PH/65 Added configuration files for NetBSD3.
2829 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
2831 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
2832 is preferred over IPv4.
2834 PH/68 The bounce_return_message and bounce_return_body options were not being
2835 honoured for bounces generated during the reception of non-SMTP messages.
2836 In particular, this applied to messages rejected by the ACL. This bug has
2837 been fixed. However, if bounce_return_message is true and bounce_return_
2838 body is false, the headers that are returned for a non-SMTP message
2839 include only those that have been read before the error was detected.
2840 (In the case of an ACL rejection, they have all been read.)
2842 PH/69 The HTML version of the specification is now built in a directory called
2843 spec_html instead of spec.html, because the latter looks like a path with
2844 a MIME-type, and this confuses some software.
2846 PH/70 Catch two compiler warnings in sieve.c.
2848 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
2849 function verify_get_ident() calls ip_connect() to connect a socket, but
2850 if the "connect()" function timed out, ip_connect() used to close the
2851 socket. However, verify_get_ident() also closes the socket later, and in
2852 between Exim writes to the log, which may get opened at this point. When
2853 the socket was closed in ip_connect(), the log could get the same file
2854 descriptor number as the socket. This naturally causes chaos. The fix is
2855 not to close the socket in ip_connect(); the socket should be closed by
2856 the function that creates it. There was only one place in the code where
2857 this was missing, in the iplookup router, which I don't think anybody now
2858 uses, but I've fixed it anyway.
2860 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
2861 well as to direct DNS lookups. Otherwise the handling of names in host
2862 lists is inconsistent and therefore confusing.
2868 PH/01 Two changes to the default runtime configuration:
2870 (1) Move the checks for relay_from_hosts and authenticated clients from
2871 after to before the (commented out) DNS black list checks.
2873 (2) Add control=submission to the relay_from_hosts and authenticated
2874 clients checks, on the grounds that messages accepted by these
2875 statements are most likely to be submissions.
2877 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
2879 (1) Generate an error if the third argument for the ${prvs expansion is
2882 (2) Treat a missing third argument of ${prvscheck as if it were an empty
2885 (3) Reset the variables that are obtained from the first argument of
2886 ${prvscheck and used in the second argument before leaving the code,
2887 because their memory is reclaimed, so using them afterwards may do
2890 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
2891 one (it's much easier than Tom thought :-).
2893 (5) Because of (4), we can now allow for the use of $prvscheck_result
2894 inside the third argument.
2896 PH/03 For some reason, the default setting of PATH when running a command from
2897 a pipe transport was just "/usr/bin". I have changed it to
2900 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
2901 anything to be listed in the output from -bV.
2903 PH/05 When a filter generated an autoreply, the entire To: header line was
2904 quoted in the delivery log line, like this:
2906 => >A.N.Other <ano@some.domain> <original@ddress> ...
2908 This has been changed so that it extracts the operative address. There
2909 may be more than one such address. If so, they are comma-separated, like
2912 => >ano@some.domain,ona@other.domain <original@ddress> ...
2914 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
2915 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
2916 not being looked up in the rDNS to get a host name, Exim was showing the
2917 IP address twice in Received: lines, even though the IP addresses were
2918 identical. For example:
2920 Received: from [1.2.3.4] (helo=[1.2.3.4])
2922 However, if the real host name was known, it was omitting the HELO data
2923 if it matched the actual IP address. This has been tidied up so that it
2924 doesn't show the same IP address twice.
2926 PH/07 When both +timestamp and +memory debugging was on, the value given by
2927 $tod_xxx expansions could be wrong, because the tod_stamp() function was
2928 called by the debug printing, thereby overwriting the timestamp buffer.
2929 Debugging no longer uses the tod_stamp() function when +timestamp is set.
2931 PH/08 When the original message was included in an autoreply transport, it
2932 always said "this is a copy of the message, including all the headers",
2933 even if body_only or headers_only was set. It now gives an appropriate
2936 PH/09 Applied a patch from the Sieve maintainer which:
2938 o fixes some comments
2939 o adds the (disabled) notify extension core
2940 o adds some debug output for the result of if/elsif tests
2941 o points to the current vacation draft in the documentation
2942 and documents the missing references header update
2946 o fixes a bug in processing the envelope test (when testing
2947 multiple envelope elements, the last element determinted the
2950 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
2951 Electronic Mail") by including:
2953 Auto-submitted: auto-generated
2955 in the messages that it generates (bounce messages and others, such as
2956 warnings). In the case of bounce messages for non-SMTP mesages, there was
2957 also a typo: it was using "Auto_submitted" (underscore instead of
2958 hyphen). Since every message generated by Exim is necessarily in response
2959 to another message, thes have all been changed to:
2961 Auto-Submitted: auto-replied
2963 in accordance with these statements in the RFC:
2965 The auto-replied keyword:
2967 - SHOULD be used on messages sent in direct response to another
2968 message by an automatic process,
2970 - MUST NOT be used on manually-generated messages,
2972 - MAY be used on Delivery Status Notifications (DSNs) and Message
2973 Disposition Notifications (MDNs),
2975 - MUST NOT be used on messages generated by automatic or periodic
2976 processes, except for messages which are automatic responses to
2979 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
2980 to the default Received: header definition.
2982 PH/12 Added log selector acl_warn_skipped (default on).
2984 PH/13 After a successful wildlsearch lookup, discard the values of numeric
2985 variables because (a) they are in the wrong storage pool and (b) even if
2986 they were copied, it wouldn't work properly because of the caching.
2988 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
2989 checking when decoding. Apparently there are clients that generate
2990 overlong encoded strings. Why am I not surprised?
2992 PH/15 If the first argument of "${if match_address" was not empty, but did not
2993 contain an "@" character, Exim crashed. Now it writes a panic log message
2994 and treats the condition as false.
2996 PH/16 In autoreply, treat an empty string for "once" the same as unset.
2998 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
2999 extension "envelope-auth". The code is finished and in agreement with
3000 other implementations, but there is no documentation so far and in fact,
3001 nobody wrote the draft yet. This extension is currently #undef'ed, thus
3002 not changing the active code.
3004 Print executed "if" and "elsif" statements when debugging is used. This
3005 helps a great deal to understand what a filter does.
3007 Document more things not specified clearly in RFC3028. I had all this
3008 sorted out, when out of a sudden new issues came to my mind. Oops."
3010 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
3013 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
3014 canonical form (as documented). However, after a host name lookup from
3015 the IP address, check_host() was doing a simple string comparison with
3016 addresses acquired from the DNS when checking that the found name did
3017 have the original IP as one of its addresses. Since any found IPv6
3018 addresses are likely to be in abbreviated form, the comparison could
3019 fail. Luckily, there already exists a function for doing the comparison
3020 by converting both addresses to binary, so now that is used instead of
3021 the text comparison.
3023 PH/20 There was another similar case to PH/19, when a complete host name was
3024 given in a host list; looking up its IP address could give an abbreviated
3025 form, whereas the current host's name might or might not be abbreviated.
3026 The same fix has been applied.
3032 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
3033 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
3036 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
3037 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
3039 PH/03 Typo: missing ".o" in src/pcre/Makefile.
3041 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
3042 header line, restrict the check to what is listed in RFCs 2369 and 2929.
3043 Also, for "Auto-Submitted", treat anything other than "no" as
3044 non-personal, in accordance with RFC 3834. (Previously it treated
3045 anything starting "auto-" as non-personal.)
3047 TF/01 The control=submission/name=... option had a problem with syntax
3048 errors if the name included a slash character. The /name= option
3049 now slurps the rest of the string, so it can include any characters
3050 but it must come last in the list of options (after /sender_retain
3053 PH/05 Some modifications to the interface to the fake nameserver for the new
3061 TK/01 Added the "success_on_redirect" address verification option. See
3062 NewStuff for rationale and an example.
3064 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
3066 PH/02 Patch to exigrep to allow it to work on syslog lines.
3068 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
3069 fread() to skip over the body file's header line, because in Cygwin the
3070 header line is locked and is inaccessible.
3072 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
3073 co-exist for some time) to make it clear that it is the Exim ID that is
3074 referenced, not the Message-ID: header line.
3076 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
3077 string_format() function, because snprintf() does not exist on all
3080 PH/06 The use of forbid_filter_existstest now also locks out the use of the
3081 ${stat: expansion item.
3083 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
3084 protocol synchronization error", to keep the pedants happy.
3086 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
3087 well as for IRIX systems, when gcc is being used. See the host.c source
3090 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
3092 PH/10 Named domain lists were not working if used in a queue_smtp_domains
3095 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
3096 transport and to the smtp transport in LMTP mode.
3098 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
3100 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
3101 run a filter in a subprocess. This could lead to confusion in subsequent
3102 lookups in the parent process. There should also be a search_tidyup() at
3103 the end of the subprocess.
3105 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
3106 only if the host matched helo_try_verify_hosts, which caused the
3107 verification to occur when the EHLO/HELO command was issued. The ACL just
3108 tested the remembered result. Now, if a previous verification attempt has
3109 not happened, "verify = helo" does it there and then.
3111 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
3113 TK/03 Fix log output including CR from clamd.
3115 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
3116 exist provoked a memory error which could cause a segfault.
3118 PH/15 Installed PCRE 6.2
3120 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
3122 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
3123 of the problem. Specifically, suggested +O2 rather than +O1 for the
3126 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
3128 PH/20 If a delivery was routed to a non-standard port by means of an SRV
3129 record, the port was not correctly logged when the outgoing_port log
3130 selector was set (it logged the transort's default port).
3132 PH/21 Added support for host-specific ports to manualroute, queryprogram,
3133 fallback_hosts, and "hosts" in the smtp transport.
3135 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
3136 host errors such as "Connection refused".
3138 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
3139 authentication with radiusclient 0.4.9:
3141 - Error returned from rc_read_config was caught wrongly
3142 - Username/password not passed on to radius server due to wrong length.
3144 The presumption is that some radiusclient API changes for 4.51/PH/17
3145 were not taken care of correctly. The code is still untested by me (my
3146 Linux distribution still has 0.3.2 of radiusclient), but it was
3147 contributed by a Radius user.
3149 PH/24 When doing a callout, the value of $domain wasn't set correctly when
3150 expanding the "port" option of the smtp transport.
3152 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
3153 while reading a MIME header. Thanks to Tom Hughes for a patch.
3155 PH/24 Include config.h inside local_scan.h so that configuration settings are
3158 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
3159 that in an ACL for RCPT (for example), you can examine exactly what was
3162 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
3163 commands, but it was not correctly comparing the address with the actual
3164 client host address. Thus, it would show the EHLO address in Received:
3165 header lines when this was not necessary.
3167 PH/27 Added the % operator to ${eval:}.
3169 PH/28 Exim tries to create and chdir to its spool directory when it starts;
3170 it should be ignoring failures (because with -C, for example, it has lost
3171 privilege). It wasn't ignoring creation failures other than "already
3174 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
3177 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
3178 because an input line was too long, either on its own, or by virtue of
3179 too many continuations, the temporary file was not being removed, and the
3180 return code was incorrect.
3182 PH/31 Missing "BOOL" in function definition in filtertest.c.
3184 PH/32 Applied Sieve patches from the maintainer.
3186 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
3188 PH/33 Added "verify = not_blind".
3190 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
3191 Local/Makefile (with some defaults set). These are used in built scripts
3192 such as exicyclog, but they have never been used in the exim_install
3193 script (though there are many overriding facilities there). I have
3194 arranged that the exim_install script now takes note of these two
3197 PH/35 Installed configuration files for Dragonfly.
3199 PH/36 When a locally submitted message by a trusted user did not contain a
3200 From: header, and the sender address was obtained from -f or from an SMTP
3201 MAIL command, and the trusted user did not use -F to supply a sender
3202 name, $originator_name was incorrectly used when constructing a From:
3203 header. Furthermore, $originator_name was used for submission mode
3204 messages from external hosts without From: headers in a similar way,
3205 which is clearly wrong.
3207 PH/37 Added control=suppress_local_fixups.
3209 PH/38 When log_selector = +received_sender was set, and the addition of the
3210 sender made the log line's construction buffer exactly full, or one byte
3211 less than full, an overflow happened when the terminating "\n" was
3214 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
3215 when the result of a list match is failure because a DNS lookup failed.
3217 PH/40 RM_COMMAND is now used in the building process.
3219 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
3220 the "build-* directories that it finds.
3222 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
3223 address in a domain literal was a prefix of an interface address.
3225 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
3226 when verifying a sender address, unless rewrite_headers is false.
3228 PH/44 Wrote a long comment about why errors_to addresses are verified as
3229 recipients, not senders.
3231 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
3232 the ratelimit ACL was added.
3234 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
3236 PH/46 Added extra information about PostgreSQL errors to the error string.
3238 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
3239 avoiding the need to install special zones in a real server. This is
3240 backwards compatible; if it can't find the fake resolver, it drops back.
3241 Thus, both old and new test suites can be run.
3243 TF/02 Added util/ratelimit.pl
3245 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
3246 clock is set back in time.
3248 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
3249 Candler <B.Candler@pobox.com>.
3251 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
3252 OK for addresses that are the result of redirections.
3254 PH/48 A number of further additions for the benefit of the new test suite,
3255 including a fake gethostbyname() that interfaces to the fake DNS resolver
3258 TF/06 The fix for widen_domains has also been applied to qualify_single and
3259 search_parents which are the other dnslookup options that can cause
3262 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
3265 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
3267 TF/07 Exim produced the error message "an SRV record indicated no SMTP
3268 service" if it encountered an MX record with an empty target hostname.
3269 The message is now "an MX or SRV record indicated no SMTP service".
3271 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
3272 if the DNS of the sending site is misconfigured. This is quite a
3273 common situation. This change restores the behaviour of treating a
3274 helo verification defer as a failure.
3276 PH/51 If self=fail was set on a router, the bounce message did not include the
3277 actual error message.
3283 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
3285 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
3286 command itself ended in error, the underlying message about the transport
3287 filter timeout was being overwritten with the pipe command error. Now the
3288 underlying error message should be appended to the second error message.
3290 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
3292 PH/02 Reduce the amount of output that "make" produces by default. Full output
3293 can still be requested.
3295 PH/03 The warning log line about a condition test deferring for a "warn" verb
3296 was being output only once per connection, rather than after each
3297 occurrence (because it was using the same function as for successful
3298 "warn" verbs). This seems wrong, so I have changed it.
3300 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
3301 it should not have, which might have caused a crash in the right
3302 circumstances, but probably never did.
3304 PH/04 Installed a modified version of Tony Finch's patch to make submission
3305 mode fix the return path as well as the Sender: header line, and to
3306 add a /name= option so that you can make the user's friendly name appear
3309 TF/03 Added the control = fakedefer ACL modifier.
3311 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
3312 Mark Lowes for thorough testing.
3314 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
3316 TK/03 Merged latest SRS patch from Miles Wilton.
3318 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
3319 with the definition in sysexits.h (which is #included earlier).
3320 Fortunately, Exim does not actually use EX_OK. The code used to try to
3321 preserve the sysexits.h value, by assumimg that macro definitions were
3322 scanned for macro replacements. I have been disabused of this notion,
3323 so now the code just undefines EX_OK before #including unistd.h.
3325 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
3326 in the smtp transport. When a block could not be written in a single
3327 write() function, the timeout was being re-applied to each part-write.
3328 This seems wrong - if the receiver was accepting one byte at a time it
3329 would take for ever. The timeout is now adjusted when this happens. It
3330 doesn't have to be particularly precise.
3332 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
3333 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
3335 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
3336 without a domain if the check to <postmaster@domain> fails.
3338 SC/01 Eximstats: added -xls and the ability to specify output files
3339 (patch written by Frank Heydlauf).
3341 SC/02 Eximstats: use FileHandles for outputing results.
3343 SC/03 Eximstats: allow any combination of xls, txt, and html output.
3345 SC/04 Eximstats: fixed display of large numbers with -nvr option
3347 SC/05 Eximstats: fixed merging of reports with empty tables.
3349 SC/06 Eximstats: added the -include_original_destination flag
3351 SC/07 Eximstats: removed tabs and trailing whitespace.
3353 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
3355 TK/06 MBOX spool code: Add real "From " MBOX separator line
3356 so the .eml file is really in mbox format (even though
3357 most programs do not really care). Patch from Alex Miller.
3359 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
3360 The latter is generated from $received_to and is only set if the
3361 message has one envelope recipient. SA can use these headers,
3362 obviously out-of-the-box. Patch from Alex Miller.
3364 PH/08 The ${def test on a variable was returning false if the variable's
3365 value was "0", contrary to what the specification has always said!
3366 The result should be true unless the variable is empty.
3368 PH/09 The syntax error of a character other than { following "${if
3369 def:variable_name" (after optional whitespace) was not being diagnosed.
3370 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
3371 accidental colon was present, for example, could give incorrect results.
3373 PH/10 Tidied the code in a number of places where the st_size field of a stat()
3374 result is used (not including appendfile, where other changes are about
3377 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
3378 This involved changing a lot of size variables from int to off_t. It
3379 should work with maildirs and everything.
3381 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
3382 spamd dying while we are connected to it.
3384 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
3387 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
3388 function for BDB 4.3.
3390 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
3392 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
3393 the sources for PCRE have been re-arranged and more files are now
3396 PH/15 The code I had for printing potentially long long variables in PH/11
3397 above was not the best (it lost precision). The length of off_t variables
3398 is now inspected at build time, and an appropriate printing format (%ld
3399 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
3400 to be "long long int" or "long int". This is needed for the internal
3401 formatting function string_vformat().
3403 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
3404 the configuration file to be ":syslog", then the script "guesses" where
3405 the logs files are, rather than using the compiled in default. In our
3406 case the guess is not the same as the compiled default, so the script
3407 suddenly stopped working when I started to use syslog. The patch checks
3408 to see if log_file_path is "". If so, it attempts to read it from exim
3409 with no configuration file to get the compiled in version, before it
3410 falls back to the previous guessing code."
3412 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
3413 implementing BATV in an Exim configuration. See NewStuff for the gory
3416 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
3417 Makefile that are specific to HP-UX.
3419 PH/18 If the "use_postmaster" option was set for a recipient callout together
3420 with the "random" option, the postmaster address was used as the MAIL
3421 FROM address for the random test, but not for the subsequent recipient
3422 test. It is now used for both.
3424 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
3425 patch removes a few documentation additions to RFC 3028, because the
3426 latest draft now contains them. It adds the new en;ascii-case comparator
3427 and a new error check for 8bit text in MIME parts. Comparator and
3428 require names are now matched exactly. I enabled the subaddress
3429 extension, but it is not well tested yet (read: it works for me)."
3431 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
3432 rework some of the code of TK/09 above to avoid the hardwired use of
3433 "%lld" and "long long". Replaced the call to snprintf() with a call to
3436 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
3437 records point to non-existent hosts", "retry timeout exceeded", and
3438 "retry time not reached for any host after a long failure period".
3440 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
3441 experimental DomainKeys support:
3443 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
3444 (2) On an error such as an illegally used "control", the wrong name for
3445 the control was given.
3447 These problems did NOT occur unless DomainKeys support was compiled.
3449 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
3451 PH/24 Added ${if match_ip condition.
3453 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
3454 databases so that it will be absolutely obvious if a crash occurs in the
3455 DB library. This is a regular occurrence (often caused by mis-matched
3458 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
3459 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
3460 picked up on a user's system that detects such things. There doesn't seem
3461 to be a gcc warning option for this - only an attribute that has to be
3462 put on the function's prototype. It seems that in Fedora Core 4 they have
3463 set this on a number of new functions. No doubt there will be more in due
3466 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
3467 specify a transport. However, if an address that was verified by such a
3468 router was the subject of a callout, Exim crashed because it tried to
3469 read the rcpt_include_affixes from the non-existent transport. Now it
3470 just assumes that the setting of that option is false. This bug was
3471 introduced by 4.51/PH/31.
3473 PH/28 Changed -d+all to exclude +memory, because that information is very
3474 rarely of interest, but it makes the output a lot bigger. People tend to
3475 do -d+all out of habit.
3477 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
3478 code in os-type was giving problems when libc.so lives in lib64, like on
3481 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
3482 aren't the modern standard, and it seems that some systems' include files
3483 don't always have them. Exim was already checking for some of the newer
3484 ones like T_AAAA, and defining it itself. I've added checks for all the
3485 record types that Exim uses.
3487 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
3488 not automatically generating a new one, as it is supposed to. This
3489 prevented TLS from working. If the file did exist, but contained invalid
3490 data, a new version was generated, as expected. It was only the case of a
3491 non-existent file that was broken.
3493 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
3494 with a change in libdomainkeys > 0.64.
3496 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
3497 from DNS. If the selector record carries the flag, it now has
3498 precedence over the domain-wide flag.
3500 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
3502 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
3503 the use of a port name that isn't defined in /etc/services) occurred, the
3504 message was deferred as in a normal delivery, and thus remained on the
3505 spool, instead of being failed because of the mua_wrapper setting. This
3506 is now fixed, and I tidied up some of the mua_wrapper messages at the
3509 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
3510 the messages in a hash of arrays rather than using individual hashes.
3511 This is a bit cleaner and results in dramatic memory savings, albeit
3512 at a slight CPU cost.
3514 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
3515 as requested by Marc Sherman.
3517 SC/10 Eximstats: added histograms for user specified patterns as requested
3520 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
3522 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
3523 fopen() in the content-scanning modules that did not already have it.
3529 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
3530 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
3532 TK/02 Fix ACL "control" statement not being available in MIME ACL.
3534 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
3536 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
3537 to test Sieve filters that use "vacation".
3539 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
3540 that changes the way the GnuTLS parameters are stored in the cache file.
3541 The new format can be generated externally. For backward compatibility,
3542 if the data in the cache doesn't make sense, Exim assumes it has read an
3543 old-format file, and it generates new data and writes a new file. This
3544 means that you can't go back to an older release without removing the
3547 PH/03 A redirect router that has both "unseen" and "one_time" set does not
3548 work if there are any delivery delays because "one_time" forces the
3549 parent to be marked "delivered", so its unseen clone is never tried
3550 again. For this reason, Exim now forbids the simultaneous setting of
3553 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
3554 redirected to themselves ("homonym" addresses). Read the long ChangeLog
3555 entry if you want to know the details. The fix, however, neglected to
3556 consider the case when local delivery batching is involved. The test for
3557 "previously delivered" was not happening when checking to see if an
3558 address could be batched with a previous (undelivered) one; under
3559 certain circumstances this could lead to multiple deliveries to the same
3562 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
3563 in its include files, and this causes problems building Exim.
3565 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
3566 header_syntax) but Exim was just ignoring anything given after a slash.
3567 In particular, this caused confusion with an attempt to use "verify =
3568 reverse_host_lookup/defer_ok". An error is now given when options are
3569 supplied for verify items that do not have them. (Maybe reverse_host_
3570 lookup should have a defer_ok option, but that's a different point.)
3572 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
3573 defined by RFC 821) to 2048, because there were problems with some AUTH
3574 commands, and RFC 1869 says the size should be increased for extended
3575 SMTP commands that take arguments.
3577 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
3580 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
3581 "unknown" error; now it says that the functionality isn't in the binary.
3583 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
3584 an address' error message when a string expansion fails (syntax or
3585 whatever). Otherwise the password may appear in the log. Following change
3586 PH/42 below, there is no longer a chance of it appearing in a bounce
3589 PH/11 Installed exipick version 20050225.0 from John Jetmore.
3591 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
3592 of its addresses was ever tried. (Bugzilla bug #2.)
3594 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
3595 the result incorrectly in the debug output. (It correctly added a newline
3596 to what was transported.)
3598 TF/01 Added $received_time.
3600 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
3601 commented out examples of how to interface to a virus scanner and to
3602 SpamAssassin. Also added commented examples of av_scanner and
3603 spamd_address settings.
3605 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
3606 and controls are allowed in which ACLs. There were a couple of minor
3607 errors. Some of the entries in the conditions table (which is a table of
3608 where they are NOT allowed) were getting very unwieldy; rewrote them as a
3609 negation of where the condition IS allowed.
3611 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
3613 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
3614 header file does not have a version number, so I've had to invent a new
3615 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
3616 API. The code is untested by me (my Linux distribution still has 0.3.2 of
3617 radiusclient), but it was contributed by a Radius user.
3619 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
3620 files or directories, for OpenSSL.
3622 PH/19 When an Exim process that is running as root has to create an Exim log
3623 file, it does so in a subprocess that runs as exim:exim so as to get the
3624 ownership right at creation (otherwise, other Exim processes might see
3625 the file with the wrong ownership). There was no test for failure of this
3626 fork() call, which would lead to the process getting stuck as it waited
3627 for a non-existent subprocess. Forks do occasionally fail when resources
3628 run out. I reviewed all the other calls to fork(); they all seem to check
3631 PH/20 When checking for unexpected SMTP input at connect time (before writing
3632 the banner), Exim was not dealing correctly with a non-positive return
3633 from the read() function. If the client had disconnected by this time,
3634 the result was a log entry for a synchronization error with an empty
3635 string after "input=" when read() returned zero. If read() returned -1
3636 (an event I could not check), uninitialized data bytes were printed.
3637 There were reports of junk text (parts of files, etc) appearing after
3640 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
3642 PH/22 Added support for macro redefinition, and (re)definition in between
3643 driver and ACL definitions.
3645 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
3646 forgetting to use the resulting value; it was using the unexpanded value.
3648 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
3649 hadn't been configured. The fix is from Juergen Kreileder, who
3650 understands it better than I do:
3652 "Here's what I see happening with three configured cyrus_sasl
3653 authenticators configured (plain, login, cram-md5):
3655 On startup auth_cyrus_sasl_init() gets called for each of these.
3656 This means three calls to sasl_listmech() without a specified mech_list.
3657 => SASL tests which mechs of all available mechs actually work
3658 => three warnings about OTP not working
3659 => the returned list contains: plain, login, cram-md5, digest-md5, ...
3661 With the patch, sasl_listmech() also gets called three times. But now
3662 SASL's mech_list option is set to the server_mech specified in the the
3663 authenticator. Or in other words, the answer from sasl_listmech()
3664 gets limited to just the mech you're testing for (which is different
3666 => the return list contains just 'plain' or 'login', 'cram-md5' or
3667 nothing depending on the value of ob->server_mech.
3669 I've just tested the patch: Authentication still works fine,
3670 unavailable mechs specified in the exim configuration are still
3671 caught, and the auth.log warnings about OTP are gone."
3673 PH/25 When debugging is enabled, the contents of the command line are added
3674 to the debugging output, even when log_selector=+arguments is not
3677 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
3678 answer is "GNU", and only if the return is "GNU/something" is the answer
3681 PH/27 $acl_verify_message is now set immediately after the failure of a
3682 verification in an ACL, and so is available in subsequent modifiers. In
3683 particular, the message can be preserved by coding like this:
3685 warn !verify = sender
3686 set acl_m0 = $acl_verify_message
3688 Previously, $acl_verify_message was set only while expanding "message"
3689 and "log_message" when a very denied access.
3691 PH/28 Modified OS/os.c-Linux with
3693 -#ifndef OS_LOAD_AVERAGE
3694 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
3696 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
3697 nomenclature these days.)
3699 PH/29 Installed patch from the Sieve maintainer that adds the options
3700 sieve_useraddress and sieve_subaddress to the redirect router.
3702 PH/30 In these circumstances:
3703 . Two addresses routed to the same list of hosts;
3704 . First host does not offer TLS;
3705 . First host accepts first address;
3706 . First host gives temporary error to second address;
3707 . Second host offers TLS and a TLS session is established;
3708 . Second host accepts second address.
3709 Exim incorrectly logged both deliveries with the TLS parameters (cipher
3710 and peerdn, if requested) that were in fact used only for the second
3713 PH/31 When doing a callout as part of verifying an address, Exim was not paying
3714 attention to any local part prefix or suffix that was matched by the
3715 router that accepted the address. It now behaves in the same way as it
3716 does for delivery: the affixes are removed from the local part unless
3717 rcpt_include_affixes is set on the transport.
3719 PH/32 Add the sender address, as F=<...>, to the log line when logging a
3720 timeout during the DATA phase of an incoming message.
3722 PH/33 Sieve envelope tests were broken for match types other than :is. I have
3723 applied a patch sanctioned by the Sieve maintainer.
3725 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
3726 the uid or gid is negative. A case of a negative gid caused this to be
3727 noticed. The fix allows for either to be negative.
3729 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
3730 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
3733 PH/36 The change PH/12 above was broken. Fixed it.
3735 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
3736 the grounds that routing the same address twice would always produce the
3737 same answer. This might have been true once, but it is certainly no
3738 longer true now. Routing a child address may depend on the previous
3739 routing that produced that child. Some complicated redirection strategies
3740 went wrong when messages had multiple recipients, and made Exim's
3741 behaviour dependent on the order in which the addresses were given.
3743 I have moved the duplicate checking until after the routing is complete.
3744 Exim scans the addresses that are assigned to local and remote
3745 transports, and removes any duplicates. This means that more work will be
3746 done, as duplicates will always all be routed, but duplicates are
3747 presumably rare, so I don't expect this is of any significance.
3749 For deliveries to pipes, files, and autoreplies, the duplicate checking
3750 still happens during the routing process, since they are not going to be
3753 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
3754 It corrects a timeout issue with spamd. This is Ian's comment: "The
3755 background is that sometimes spamd either never reads data from a
3756 connection it has accepted, or it never writes response data. The exiscan
3757 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
3758 blindly assumes that writes won't block so it may never time out."
3760 PH/39 Allow G after quota size as well as K and M.
3762 PH/40 The value set for $authenticated_id in an authenticator may not contain
3763 binary zeroes or newlines because the value is written to log lines and
3764 to spool files. There was no check on this. Now the value is run through
3765 the string_printing() function so that such characters are converted to
3766 printable escape sequences.
3768 PH/41 $message_linecount is a new variable that contains the total number of
3769 lines in the message. Compare $body_linecount, which is the count for the
3772 PH/42 Exim no longer gives details of delivery errors for specific addresses in
3773 bounce and delay warning messages, except in certain special cases, which
3776 (a) An SMTP error message from a remote host;
3777 (b) A message specified in a :fail: redirection;
3778 (c) A message specified in a "fail" command in a system filter;
3779 (d) A message specified in a FAIL return from the queryprogram router;
3780 (e) A message specified by the cannot_route_message router option.
3782 In these cases only, Exim does include the error details in bounce and
3783 warning messages. There are also a few cases where bland messages such
3784 as "unrouteable address" or "local delivery error" are given.
3786 PH/43 $value is now also set for the "else" part of a ${run expansion.
3788 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
3789 being worked on, but at least Exim now implements the latest version to
3792 PH/45 In a pipe transport, although a timeout while waiting for the pipe
3793 process to complete was treated as a delivery failure, a timeout while
3794 writing the message to the pipe was logged, but erroneously treated as a
3795 successful delivery. Such timeouts include transport filter timeouts. For
3796 consistency with the overall process timeout, these timeouts are now
3797 treated as errors, giving rise to delivery failures by default. However,
3798 there is now a new Boolean option for the pipe transport called
3799 timeout_defer, which, if set TRUE, converts the failures into defers for
3800 both kinds of timeout. A transport filter timeout is now identified in
3803 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
3804 systems where "make" and "gmake" are different, calling "gmake" at top
3805 level broke things. I've arranged for the value of $(MAKE) to be passed
3806 from the Makefile to this script so that it can call the same version of
3810 A note about Exim versions 4.44 and 4.50
3811 ----------------------------------------
3813 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
3814 changes of various kinds. As a consequence, a big documentation update was
3815 needed. This delayed the release for rather longer than seemed good, especially
3816 in the light of a couple of (minor) security issues. Therefore, the changes
3817 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
3818 release. So 4.44 and 4.50 are in effect two different branches that both start
3821 I have left the 4.50 change log unchanged; it contains all the changes since
3822 4.43. The change log for 4.44 is below; many of its items are identical to
3823 those for 4.50. This seems to be the most sensible way to preserve the
3824 historical information.
3830 1. Minor wording change to the doc/README.SIEVE file.
3832 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
3833 computation of the current number of files was incorrect.
3835 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
3836 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
3839 4. Give more explanation in the error message when the command for a transport
3840 filter fails to execute.
3842 5. There are several places where Exim runs a non-Exim command in a
3843 subprocess. The SIGUSR1 signal should be disabled for these processes. This
3844 was being done only for the command run by the queryprogram router. It is
3845 now done for all such subprocesses. The other cases are: ${run, transport
3846 filters, and the commands run by the lmtp and pipe transports.
3848 6. Added CONFIGURE_GROUP build-time option.
3850 7. Some older OS have a limit of 256 on the maximum number of file
3851 descriptors. Exim was using setrlimit() to set 1000 as a large value
3852 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
3853 systems. I've change it so that if it can't get 1000, it tries for 256.
3855 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
3856 was an oversight, and furthermore, ever since the addition of extra
3857 controls (e.g. 4.43/32), the checks on when to allow different forms of
3858 "control" were broken. There should now be diagnostics for all cases when a
3859 control that does not make sense is encountered.
3861 9. Added the /retain_sender option to "control=submission".
3863 10. $recipients is now available in the predata ACL (oversight).
3865 11. Tidy the search cache before the fork to do a delivery from a message
3866 received from the command line. Otherwise the child will trigger a lookup
3867 failure and thereby defer the delivery if it tries to use (for example) a
3868 cached ldap connection that the parent has called unbind on.
3870 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
3871 of $address_data from the recipient verification was clobbered by the
3872 sender verification.
3874 13. The value of address_data from a sender verification is now available in
3875 $sender_address_data in subsequent conditions in the ACL statement.
3877 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
3879 15. Added a new option "connect=<time>" to callout options, to set a different
3882 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
3883 was its contents. (It was OK if the option was not defined at all.)
3885 17. A "Completed" log line is now written for messages that are removed from
3886 the spool by the -Mrm option.
3888 18. New variables $sender_verify_failure and $recipient_verify_failure contain
3889 information about exactly what failed.
3891 19. Added -dd to debug only the daemon process.
3893 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
3894 handles timeouts, both on the server side and network timeouts. Renamed the
3895 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
3897 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
3898 It is now set to "smtps".
3900 22. $host_address is now set to the target address during the checking of
3901 ignore_target_hosts.
3903 23. When checking ignore_target_hosts for an ipliteral router, no host name was
3904 being passed; this would have caused $sender_host_name to have been used if
3905 matching the list had actually called for a host name (not very likely,
3906 since this list is usually IP addresses). A host name is now passed as
3909 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
3910 code that specifies a non-restarting handler (typically sigaction() in
3911 modern systems) in an attempt to fix a rare and obscure crash bug.
3913 25. Narrowed the window for a race in the daemon that could cause it to ignore
3914 SIGCHLD signals. This is not a major problem, because they are used only to
3915 wake it up if nothing else does.
3917 26. A malformed maildirsize file could cause Exim to calculate negative values
3918 for the mailbox size or file count. Odd effects could occur as a result.
3919 The maildirsize information is now recalculated if the size or filecount
3922 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
3923 support for a long time. Removed HAVE_SYS_VFS_H.
3925 28. Installed the latest version of exipick from John Jetmore.
3927 29. In an address list, if the pattern was not a regular expression, an empty
3928 subject address (from a bounce message) matched only if the pattern was an
3929 empty string. Non-empty patterns were not even tested. This was the wrong
3930 because it is perfectly reasonable to use an empty address as part of a
3931 database query. An empty address is now tested by patterns that are
3932 lookups. However, all the other forms of pattern expect the subject to
3933 contain a local part and a domain, and therefore, for them, an empty
3934 address still always fails if the pattern is not itself empty.
3936 30. Exim went into a mad DNS loop when attempting to do a callout where the
3937 host was specified on an smtp transport, and looking it up yielded more
3938 than one IP address.
3940 31. Re-factored the code for checking spool and log partition space into a
3941 function that finds that data and another that does the check. The former
3942 is then used to implement four new variables: $spool_space, $log_space,
3943 $spool_inodes, and $log_inodes.
3945 32. The RFC2047 encoding function was originally intended for short strings
3946 such as real names; it was not keeping to the 75-character limit for
3947 encoded words that the RFC imposes. It now respects the limit, and
3948 generates multiple encoded words if necessary. To be on the safe side, I
3949 have increased the buffer size for the ${rfc2047: expansion operator from
3952 33. It is now permitted to omit both strings after an "if" condition; if the
3953 condition is true, the result is "true". As before, when the second string
3954 is omitted, a false condition yields an empty string. This makes it less
3955 cumbersome to write custom ACL and router conditions.
3957 34. Failure to deliver a bounce message always caused it to be frozen, even if
3958 there was an errors_to setting on the router. The errors_to setting is now
3961 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
3962 canonical form (fully expanded) before being placed in
3963 $sender_host_address.
3965 36. The table in the code that translates DNS record types into text (T_A to
3966 "A" for instance) was missing entries for NS and CNAME. It is just possible
3967 that this could have caused confusion if both these types were looked up
3968 for the same domain, because the text type is used as part of Exim's
3969 per-process caching. But the chance of anyone hitting this buglet seems
3972 37. The dnsdb lookup has been extended in a number of ways.
3974 (1) There is a new type, "zns", which walks up the domain tree until it
3975 finds some nameserver records. It should be used with care.
3977 (2) There is a new type, "mxh", which is like "mx" except that it returns
3978 just the host names, not the priorities.
3980 (3) It is now possible to give a list of domains (or IP addresses) to be
3981 looked up. The behaviour when one of the lookups defers can be
3982 controlled by a keyword.
3984 (4) It is now possible to specify the separator character for use when
3985 multiple records are returned.
3987 38. The dnslists ACL condition has been extended: it is now possible to supply
3988 a list of IP addresses and/or domains to be looked up in a particular DNS
3991 39. Added log_selector=+queue_time_overall.
3993 40. When running the queue in the test harness, wait just a tad after forking a
3994 delivery process, to get repeatability of debugging output.
3996 41. Include certificate and key file names in error message when GnuTLS fails
3997 to set them up, because the GnuTLS error message doesn't include the name
3998 of the failing file when there is a problem reading it.
4000 42. Allow both -bf and -bF in the same test run.
4002 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
4004 44. The "Exiscan patch" is now merged into the mainline Exim source.
4006 45. Sometimes the final signoff response after QUIT could fail to get
4007 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4008 < 0 before doing a fflush(). This bug looks as though it goes back to the
4009 introduction of TLS in release 3.20, but "sometimes" must have been rare
4010 because the tests only now provoked it.
4012 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
4013 (this can affect the format of dates).
4015 47. exim_tidydb, when checking for the continued existence of a message for
4016 which it has found a message-specific retry record, was not finding
4017 messages that were in split spool directories. Consequently, it was
4018 deleting retry records that should have stayed in existence.
4020 48. Steve fixed some bugs in eximstats.
4022 49. The SPA authentication driver was not abandoning authentication and moving
4023 on to the next authenticator when an expansion was forced to fail,
4024 contradicting the general specification for all authenticators. Instead it
4025 was generating a temporary error. It now behaves as specified.
4027 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
4028 (the order specifies the preference for clients). The order is now AES256,
4029 AES128, 3DES, ARCFOUR128.
4031 51. Small patch to Sieve code - explicitly set From: when generating an
4034 52. Exim crashed if a remote delivery caused a very long error message to be
4035 recorded - for instance if somebody sent an entire SpamAssassin report back
4036 as a large number of 550 error lines. This bug was coincidentally fixed by
4037 increasing the size of one of Exim's internal buffers (big_buffer) that
4038 happened as part of the Exiscan merge. However, to be on the safe side, I
4039 have made the code more robust (and fixed the comments that describe what
4042 53. Now that there can be additional text after "Completed" in log lines (if
4043 the queue_time_overall log selector is set), a one-byte patch to exigrep
4044 was needed to allow it to recognize "Completed" as not the last thing in
4047 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
4048 patch that reportedly fixes this has been added. I am not expert enough to
4049 create a test for it. This is what the patch creator wrote:
4051 "I found a little strange behaviour of ldap code when working with
4052 Windows 2003 AD Domain, where users was placed in more than one
4053 Organization Units. When I tried to give exim partial DN, the exit code
4054 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
4055 But simultaneously result of request was absolutely normal ldap result,
4056 so I produce this patch..."
4058 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
4059 REFERENCE, so I have modified the code to exclude the patch when that macro
4062 55. Some experimental protocols are using DNS PTR records for new purposes. The
4063 keys for these records are domain names, not reversed IP addresses. The
4064 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4065 leaves it alone. Component reversal etc. now happens only for IP addresses.
4068 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
4070 57. Double the size of the debug message buffer (to 2048) so that more of very
4071 long debug lines gets shown.
4073 58. The exicyclog utility now does better if the number of log files to keep
4074 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
4076 59. Two changes related to the smtp_active_hostname option:
4078 (1) $smtp_active_hostname is now available as a variable.
4079 (2) The default for smtp_banner uses $smtp_active_hostname instead
4080 of $primary_hostname.
4082 60. The host_aton() function is supposed to be passed a string that is known
4083 to be a valid IP address. However, in the case of IPv6 addresses, it was
4084 not checking this. This is a hostage to fortune. Exim now panics and dies
4085 if the condition is not met. A case was found where this could be provoked
4086 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4087 components; fortuitously, this particular loophole had already been fixed
4088 by change 4.50/55 above.
4090 If there are any other similar loopholes, the new check in host_aton()
4091 itself should stop them being exploited. The report I received stated that
4092 data on the command line could provoke the exploit when Exim was running as
4093 exim, but did not say which command line option was involved. All I could
4094 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4095 running as the user.
4098 61. There was a buffer overflow vulnerability in the SPA authentication code
4099 (which came originally from the Samba project). I have added a test to the
4100 spa_base64_to_bits() function which I hope fixes it.
4103 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
4104 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
4106 63. The daemon start-up calls getloadavg() while still root for those OS that
4107 need the first call to be done as root, but it missed one case: when
4108 deliver_queue_load_max is set with deliver_drop_privilege. This is
4109 necessary for the benefit of the queue runner, because there is no re-exec
4110 when deliver_drop_privilege is set.
4112 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
4113 This has been fixed.
4115 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4116 was in use, was not putting the data itself into the right store pool;
4117 consequently, it could be overwritten for a subsequent message in the same
4118 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4121 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
4123 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
4124 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
4125 the function were treating the return as a boolean value, which happened to
4126 work because 0=false and not-0=true, but is not correct code.
4128 68. The host_aton() function was not handling scoped IPv6 addresses (those
4129 with, for example, "%eth0" on the end) correctly.
4131 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
4132 negated items (that is, ~something) in unsigned ints. Some compilers
4133 apparently mutter when there is no cast.
4135 70. If an address verification called from an ACL failed, and did not produce a
4136 user-specific message (i.e. there was only a "system" message), nothing was
4137 put in $acl_verify_message. In this situation, it now puts the system
4140 71. Change 4.23/11 added synchronization checking at the start of an SMTP
4141 session; change 4.31/43 added the unwanted input to the log line - except
4142 that it did not do this in the start of session case. It now does.
4144 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
4145 This is wrong and can cause the other end to generate a synchronization
4146 error if it is another Exim or anything else that does the synchronization
4147 check. A QUIT command is no longer sent after a timeout.
4149 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
4150 during host lookups.
4152 74. The defer_ok option of callout verification was not working if it was used
4153 when verifying addresses in header lines, that is, for this case:
4155 verify = header_sender/callout=defer_ok
4157 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
4158 those file descriptors could be used for SMTP connections. If anything
4159 wrote to stderr (the example that came up was "warn" in embedded Perl), it
4160 could be sent to the SMTP client, causing chaos. The daemon now opens
4161 stdin, stdout, and stderr to /dev/null when it puts itself into the
4164 76. Arrange for output from Perl's "warn" command to be written to Exim's main
4165 log by default. The user can override this with suitable Perl magic.
4167 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
4168 the log message when discard triggers, was not working for the DATA ACL or
4169 for the non-SMTP ACL.
4171 78. Error message wording change in sieve.c.
4173 79. If smtp_accept_max_per_host was set, the number of connections could be
4174 restricted to fewer than expected, because the daemon was trying to set up
4175 a new connection before checking whether the processes handling previous
4176 connections had finished. The check for completed processes is now done
4177 earlier. On busy systems, this bug wouldn't be noticed because something
4178 else would have woken the daemon, and it would have reaped the completed
4181 80. If a message was submitted locally by a user whose login name contained one
4182 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
4183 It caused a spool format error. I have fixed the spool reading code. A
4184 related problem was that the "from" clause in the Received: line became
4185 illegal because of the space(s). It is now covered by ${quote_local_part.
4187 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
4190 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
4192 83. Patch from Sieve maintainer to fix unterminated string problem in
4193 "vacation" handling.
4195 84. Some minor changes to the Linux configuration files to help with other
4196 OS variants using glibc.
4198 85. One more patch for Sieve to update vacation handling to latest spec.
4201 ----------------------------------------------------
4202 See the note above about the 4.44 and 4.50 releases.
4203 ----------------------------------------------------
4209 1. Change 4.43/35 introduced a bug that caused file counts to be
4210 incorrectly computed when quota_filecount was set in an appendfile
4213 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4214 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4217 3. Give more explanation in the error message when the command for a transport
4218 filter fails to execute.
4220 4. There are several places where Exim runs a non-Exim command in a
4221 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4222 was being done only for the command run by the queryprogram router. It is
4223 now done for all such subprocesses. The other cases are: ${run, transport
4224 filters, and the commands run by the lmtp and pipe transports.
4226 5. Some older OS have a limit of 256 on the maximum number of file
4227 descriptors. Exim was using setrlimit() to set 1000 as a large value
4228 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4229 systems. I've change it so that if it can't get 1000, it tries for 256.
4231 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4232 was an oversight, and furthermore, ever since the addition of extra
4233 controls (e.g. 4.43/32), the checks on when to allow different forms of
4234 "control" were broken. There should now be diagnostics for all cases when a
4235 control that does not make sense is encountered.
4237 7. $recipients is now available in the predata ACL (oversight).
4239 8. Tidy the search cache before the fork to do a delivery from a message
4240 received from the command line. Otherwise the child will trigger a lookup
4241 failure and thereby defer the delivery if it tries to use (for example) a
4242 cached ldap connection that the parent has called unbind on.
4244 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4245 of $address_data from the recipient verification was clobbered by the
4246 sender verification.
4248 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4249 was its contents. (It was OK if the option was not defined at all.)
4251 11. A "Completed" log line is now written for messages that are removed from
4252 the spool by the -Mrm option.
4254 12. $host_address is now set to the target address during the checking of
4255 ignore_target_hosts.
4257 13. When checking ignore_target_hosts for an ipliteral router, no host name was
4258 being passed; this would have caused $sender_host_name to have been used if
4259 matching the list had actually called for a host name (not very likely,
4260 since this list is usually IP addresses). A host name is now passed as
4263 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4264 code that specifies a non-restarting handler (typically sigaction() in
4265 modern systems) in an attempt to fix a rare and obscure crash bug.
4267 15. Narrowed the window for a race in the daemon that could cause it to ignore
4268 SIGCHLD signals. This is not a major problem, because they are used only to
4269 wake it up if nothing else does.
4271 16. A malformed maildirsize file could cause Exim to calculate negative values
4272 for the mailbox size or file count. Odd effects could occur as a result.
4273 The maildirsize information is now recalculated if the size or filecount
4276 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4277 support for a long time. Removed HAVE_SYS_VFS_H.
4279 18. Updated exipick to current release from John Jetmore.
4281 19. Allow an empty sender to be matched against a lookup in an address list.
4282 Previously the only cases considered were a regular expression, or an
4285 20. Exim went into a mad DNS lookup loop when doing a callout where the
4286 host was specified on the transport, if the DNS lookup yielded more than
4289 21. The RFC2047 encoding function was originally intended for short strings
4290 such as real names; it was not keeping to the 75-character limit for
4291 encoded words that the RFC imposes. It now respects the limit, and
4292 generates multiple encoded words if necessary. To be on the safe side, I
4293 have increased the buffer size for the ${rfc2047: expansion operator from
4296 22. Failure to deliver a bounce message always caused it to be frozen, even if
4297 there was an errors_to setting on the router. The errors_to setting is now
4300 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4301 canonical form (fully expanded) before being placed in
4302 $sender_host_address.
4304 24. Updated eximstats to version 1.33
4306 25. Include certificate and key file names in error message when GnuTLS fails
4307 to set them up, because the GnuTLS error message doesn't include the name
4308 of the failing file when there is a problem reading it.
4310 26. Expand error message when OpenSSL has problems setting up cert/key files.
4313 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
4314 (this can affect the format of dates).
4316 28. exim_tidydb, when checking for the continued existence of a message for
4317 which it has found a message-specific retry record, was not finding
4318 messages that were in split spool directories. Consequently, it was
4319 deleting retry records that should have stayed in existence.
4321 29. eximstats updated to version 1.35
4322 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
4323 1.35 - bugfix such that pie charts by volume are generated correctly
4325 30. The SPA authentication driver was not abandoning authentication and moving
4326 on to the next authenticator when an expansion was forced to fail,
4327 contradicting the general specification for all authenticators. Instead it
4328 was generating a temporary error. It now behaves as specified.
4330 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
4331 (the order specifies the preference for clients). The order is now AES256,
4332 AES128, 3DES, ARCFOUR128.
4334 31. Small patch to Sieve code - explicitly set From: when generating an
4337 32. Exim crashed if a remote delivery caused a very long error message to be
4338 recorded - for instance if somebody sent an entire SpamAssassin report back
4339 as a large number of 550 error lines. This bug was coincidentally fixed by
4340 increasing the size of one of Exim's internal buffers (big_buffer) that
4341 happened as part of the Exiscan merge. However, to be on the safe side, I
4342 have made the code more robust (and fixed the comments that describe what
4345 33. Some experimental protocols are using DNS PTR records for new purposes. The
4346 keys for these records are domain names, not reversed IP addresses. The
4347 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4348 leaves it alone. Component reversal etc. now happens only for IP addresses.
4351 34. The host_aton() function is supposed to be passed a string that is known
4352 to be a valid IP address. However, in the case of IPv6 addresses, it was
4353 not checking this. This is a hostage to fortune. Exim now panics and dies
4354 if the condition is not met. A case was found where this could be provoked
4355 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4356 components; fortuitously, this particular loophole had already been fixed
4357 by change 4.50/55 or 4.44/33 above.
4359 If there are any other similar loopholes, the new check in host_aton()
4360 itself should stop them being exploited. The report I received stated that
4361 data on the command line could provoke the exploit when Exim was running as
4362 exim, but did not say which command line option was involved. All I could
4363 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4364 running as the user.
4367 35. There was a buffer overflow vulnerability in the SPA authentication code
4368 (which came originally from the Samba project). I have added a test to the
4369 spa_base64_to_bits() function which I hope fixes it.
4372 36. The daemon start-up calls getloadavg() while still root for those OS that
4373 need the first call to be done as root, but it missed one case: when
4374 deliver_queue_load_max is set with deliver_drop_privilege. This is
4375 necessary for the benefit of the queue runner, because there is no re-exec
4376 when deliver_drop_privilege is set.
4378 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4379 was in use, was not putting the data itself into the right store pool;
4380 consequently, it could be overwritten for a subsequent message in the same
4381 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4384 38. Sometimes the final signoff response after QUIT could fail to get
4385 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4386 < 0 before doing a fflush(). This bug looks as though it goes back to the
4387 introduction of TLS in release 3.20, but "sometimes" must have been rare
4388 because the tests only now provoked it.
4394 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
4395 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
4396 Now it returns an integer. A number of calls were still expecting a T/F
4397 return. Fortuitously, in all cases, the tests worked in OK situations,
4398 which is the norm. However, things would have gone wrong on any write
4399 failures on the smtp file descriptor. This function is used when sending
4400 messages over SMTP and also when doing verify callouts.
4402 2. When Exim is called to do synchronous delivery of a locally submitted
4403 message (the -odf or -odi options), it no longer closes stderr before doing
4406 3. Implemented the mua_wrapper option.
4408 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
4410 5. Implemented the functions header_remove(), header_testname(),
4411 header_add_at_position(), and receive_remove_recipient(), and exported them
4414 6. If an ACL "warn" statement specified the addition of headers, Exim already
4415 inserted X-ACL-Warn: at the start if there was no header name. However, it
4416 was not making this test for the second and subsequent header lines if
4417 there were newlines in the string. This meant that an invalid header could
4418 be inserted if Exim was badly configured.
4420 7. Allow an ACL "warn" statement to add header lines at the start or after all
4421 the Received: headers, as well as at the end.
4423 8. Added the rcpt_4xx retry error code.
4425 9. Added postmaster_mailfrom=xxx to callout verification option.
4427 10. Added mailfrom=xxxx to the callout verification option, for verify=
4430 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
4431 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
4433 12. Inserted some casts to stop certain compilers warning when using pointer
4434 differences as field lengths or precisions in printf-type calls (mostly
4435 affecting debugging statements).
4437 13. Added optional readline() support for -be (dynamically loaded).
4439 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
4440 same clock tick as a message's arrival, so that its received time was the
4441 same as the "first fail" time on the retry record, and that message
4442 remained on the queue past the ultimate address timeout, every queue runner
4443 would try a delivery (because it was past the ultimate address timeout) but
4444 after another failure, the ultimate address timeout, which should have then
4445 bounced the address, did not kick in. This was a "< instead of <=" error;
4446 in most cases the first failure would have been in the next clock tick
4447 after the received time, and all would be well.
4449 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
4450 being recognized when the domain list was tested by the match_domain
4451 condition in an expansion string.
4453 16. Added the ${str2b64: operator.
4455 17. Exim was always calling setrlimit() to set a large limit for the number of
4456 processes, without checking whether the existing limit was already
4457 adequate. (It did check for the limit on file descriptors.) Furthermore,
4458 errors from getrlimit() and setrlimit() were being ignored. Now they are
4459 logged to the main and panic logs, but Exim does carry on, to try to do its
4460 job under whatever limits there are.
4462 18. Imported PCRE 5.0.
4464 19. Trivial typo in log message " temporarily refused connection" (the leading
4467 20. If the log selector return_path_on_delivery was set and an address was
4468 redirected to /dev/null, the delivery process crashed because it assumed
4469 that a return path would always be set for a "successful" delivery. In this
4470 case, the whole delivery is bypassed as an optimization, and therefore no
4473 21. Internal re-arrangement: the function for sending a challenge and reading
4474 a response while authentication was assuming a zero-terminated challenge
4475 string. It's now changed to take a pointer and a length, to allow for
4476 binary data in such strings.
4478 22. Added the cyrus_sasl authenticator (code supplied by MBM).
4480 23. Exim was not respecting finduser_retries when seeking the login of the
4481 uid under which it was called; it was always trying 10 times. (The default
4482 setting of finduser_retries is zero.) Also, it was sleeping after the final
4483 failure, which is pointless.
4485 24. Implemented tls_on_connect_ports.
4487 25. Implemented acl_smtp_predata.
4489 26. If the domain in control=submission is set empty, Exim assumes that the
4490 authenticated id is a complete email address when it generates From: or
4491 Sender: header lines.
4493 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
4494 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
4495 chown and chgrp in /bin and hostname in /usr/bin.
4497 28. Exim was keeping the "process log" file open after each use, just as it
4498 does for the main log. This opens the possibility of it remaining open for
4499 long periods when the USR1 signal hits a daemon. Occasional processlog
4500 errors were reported, that could have been caused by this. Anyway, it seems
4501 much more sensible not to leave this file open at all, so that is what now
4504 29. The long-running daemon process does not normally write to the log once it
4505 has entered its main loop, and it closes the log before doing so. This is
4506 so that log files can straightforwardly be renamed and moved. However,
4507 there are a couple of unusual error situations where the daemon does write
4508 log entries, and I had neglected to close the log afterwards.
4510 30. The text of an SMTP error response that was received during a remote
4511 delivery was being truncated at 512 bytes. This is too short for some of
4512 the long messages that one sometimes sees. I've increased the limit to
4515 31. It is now possible to make retry rules that apply only when a message has a
4516 specific sender, in particular, an empty sender.
4518 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
4519 it possible to be selective about when SMTP synchronization is enforced.
4521 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
4523 32. Implemented hosts_connection_nolog.
4525 33. Added an ACL for QUIT.
4527 34. Setting "delay_warning=" to disable warnings was not working; it gave a
4530 35. Added mailbox_size and mailbox_filecount to appendfile.
4532 36. Added control = no_multiline_responses to ACLs.
4534 37. There was a bug in the logic of the code that waits for the clock to tick
4535 in the case where the clock went backwards by a substantial amount such
4536 that the microsecond fraction of "now" was more than the microsecond
4537 fraction of "then" (but the whole seconds number was less).
4539 38. Added support for the libradius Radius client library this is found on
4540 FreeBSD (previously only the radiusclient library was supported).
4546 1. When certain lookups returned multiple values in the form name=value, the
4547 quoting of the values was not always being done properly. Specifically:
4548 (a) If the value started with a double quote, but contained no whitespace,
4550 (b) If the value contained whitespace other than a space character (i.e.
4551 tabs or newlines or carriage returns) it was not quoted.
4552 This fix has been applied to the mysql and pgsql lookups by writing a
4553 separate quoting function and calling it from the lookup code. The fix
4554 should probably also be applied to nisplus, ibase and oracle lookups, but
4555 since I cannot test any of those, I have not disturbed their existing code.
4557 2. A hit in the callout cache for a specific address caused a log line with no
4558 reason for rejecting RCPT. Now it says "Previous (cached) callout
4559 verification failure".
4561 3. There was an off-by-one bug in the queryprogram router. An over-long
4562 return line was truncated at 256 instead of 255 characters, thereby
4563 overflowing its buffer with the terminating zero. As well as fixing this, I
4564 have increased the buffer size to 1024 (and made a note to document this).
4566 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
4567 when Exim is waiting for an SMTP response from a remote server, Exim
4568 restarts its select() call on the socket, thereby resetting its timeout.
4569 This is not a problem when such interrupts are rare. Somebody set up a cron
4570 job to run exiwhat every 2 minutes, which is less than the normal select()
4571 timeout (5 or 10 minutes). This meant that the select() timeout never
4572 kicked in because it was always reset. I have fixed this by comparing the
4573 time when an interrupt arrives with the time at the start of the first call
4574 to select(). If more time than the timeout has elapsed, the interrupt is
4575 treated as a timeout.
4577 5. Some internal re-factoring in preparation for the addition of Sieve
4578 extensions (by MH). In particular, the "personal" test is moved to a
4579 separate function, and given an option for scanning Cc: and Bcc: (which is
4580 not set for Exim filters).
4582 6. When Exim created an email address using the login of the caller as the
4583 local part (e.g. when creating a From: or Sender: header line), it was not
4584 quoting the local part when it contained special characters such as @.
4586 7. Installed new OpenBSD configuration files.
4588 8. Reworded some messages for syntax errors in "and" and "or" conditions to
4589 try to make them clearer.
4591 9. Callout options, other than the timeout value, were being ignored when
4592 verifying sender addresses in header lines. For example, when using
4594 verify = header_sender/callout=no_cache
4596 the cache was (incorrectly) being used.
4598 10. Added a missing instance of ${EXE} to the exim_install script; this affects
4599 only the Cygwin environment.
4601 11. When return_path_on_delivery was set as a log selector, if different remote
4602 addresses in the same message used different return paths and parallel
4603 remote delivery occurred, the wrong values would sometimes be logged.
4604 (Whenever a remote delivery process finished, the return path value from
4605 the most recently started remote delivery process was logged.)
4607 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
4608 lines when AUTH and/or TLS are in use. This is the "received protocol"
4609 field. Exim used to use "asmtp" for authenticated SMTP, without any
4610 indication (in the protocol name) for TLS use. Now it follows the RFC and
4611 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
4612 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
4613 names appear in log lines as well as in Received: header lines.
4615 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
4616 extensions, and comparison tests, and to fix some bugs.
4618 14. Changes to the "personal" filter test:
4620 (1) The test was buggy in that it was just doing the equivalent of
4621 "contains" tests on header lines. For example, if a user's address was
4622 anne@some.where, the "personal" test would incorrectly be true for
4624 To: susanne@some.where
4626 This test is now done by extracting each address from the header in turn,
4627 and checking the entire address. Other tests that are part of "personal"
4628 are now done using regular expressions (for example, to check local parts
4629 of addresses in From: header lines).
4631 (2) The list of non-personal local parts in From: addresses has been
4632 extended to include "listserv", "majordomo", "*-request", and "owner-*",
4633 taken from the Sieve specification recommendations.
4635 (3) If the message contains any header line starting with "List-" it is
4636 treated as non-personal.
4638 (4) The test for "circular" in the Subject: header line has been removed
4639 because it now seems ill-conceived.
4641 15. Minor typos in src/EDITME comments corrected.
4643 16. Installed latest exipick from John Jetmore.
4645 17. If headers_add on a router specified a text string that was too long for
4646 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
4647 of string_sprintf() is now avoided.
4649 18. $message_body_size was not set (it was always zero) when running the DATA
4650 ACL and the local_scan() function.
4652 19. For the "mail" command in an Exim filter, no default was being set for
4653 the once_repeat time, causing a random time value to be used if "once" was
4654 specified. (If the value happened to be <= 0, no repeat happened.) The
4655 default is now 0s, meaning "never repeat". The "vacation" command was OK
4656 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
4657 (I found it when inspecting the code).
4659 20. There is now an overall timeout for performing a callout verification. It
4660 defaults to 4 times the callout timeout, which applies to individual SMTP
4661 commands during the callout. The overall timeout applies when there is more
4662 than one host that can be tried. The timeout is checked before trying the
4663 next host. This prevents very long delays if there are a large number of
4664 hosts and all are timing out (e.g. when the network connections are timing
4665 out). The value of the overall timeout can be changed by specifying an
4666 additional sub-option for "callout", called "maxwait". For example:
4668 verify = sender/callout=5s,maxwait=20s
4670 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
4671 to the end before writing, but this should make it even safer).
4673 22. Exim was forgetting that it had advertised PIPELINING for the second and
4674 subsequent messages on an SMTP connection. It was also not resetting its
4675 memory on STARTTLS and an internal HELO.
4677 23. When Exim logs an SMTP synchronization error within a session, it now
4678 records whether PIPELINING has been advertised or not.
4680 24. Added 3 instances of "(long int)" casts to time_t variables that were being
4681 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
4682 rather than long int.
4684 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
4686 26. Added the never_mail option to autoreply.
4692 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
4693 crash if the getsockname() call failed; this can happen if a connection is
4694 closed very soon after it is established. The problem was simply in the
4695 order in which certain operations were done, causing Exim to try to write
4696 to the SMTP stream before it had set up the file descriptor. The bug has
4697 been fixed by making things happen in the correct order.
4703 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
4704 before the connection was closed, thus losing the rejection response.
4706 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
4707 some early Solaris releases, but causes trouble in current releases where
4708 socklen_t is defined.
4710 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
4713 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
4716 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
4717 that releasing the top of it at the end releases what was used for sub-
4718 expansions (unless the block got too big). However, discard this block if
4719 the first thing is a variable or header, so that we can use its block when
4720 it is dynamic (useful for very large $message_headers, for example).
4722 6. Lookups now cache *every* query, not just the most recent. A new, separate
4723 store pool is used for this. It can be recovered when all lookup caches are
4724 flushed. Lookups now release memory at the end of their result strings.
4725 This has involved some general refactoring of the lookup sources.
4727 7. Some code has been added to the store_xxx() functions to reduce the amount
4728 of flapping under certain conditions.
4730 8. log_incoming_interface used to affect only the <= reception log lines. Now
4731 it causes the local interface and port to be added to several more SMTP log
4732 lines, for example "SMTP connection from", and rejection lines.
4734 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
4736 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
4738 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
4739 could be overwritten at the end of the current message (or the start of a
4740 new message if it was set in a HELO ACL). The value is now preserved for
4741 the duration of the SMTP connection.
4743 12. If a transport had a headers_rewrite setting, and a matching header line
4744 contained an unqualified address, that address was qualified, even if it
4745 did not match any rewriting rules. The underlying bug was that the values
4746 of the flags that permit the existence of unqualified sender and recipient
4747 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
4748 non-local messages, and by -bnq for local messages) were not being
4749 preserved with the message after it was received.
4751 13. When Exim was logging an SMTP synchronization error, it could sometimes log
4752 "next input=" as part of the text comprising the host identity instead of
4753 the correct text. The code was using the same buffer for two different
4754 strings. However, depending on which order the printing function evaluated
4755 its arguments, the bug did not always show up. Under Linux, for example, my
4756 test suite worked just fine.
4758 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
4759 doesn't work with some older versions of Perl. It has been changed to "my",
4760 which in any case is probably the better facility to use.
4762 15. A really picky compiler found some instances of statements for creating
4763 error messages that either had too many or two few arguments for the format
4766 16. The size of the buffer for calls to the DNS resolver has been increased
4767 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
4768 for addresses that have a lot of PTR records. This alleviates a problem; it
4769 does not fully solve it.
4771 17. A dnsdb lookup for PTR records that receives more data than will fit in the
4772 buffer now truncates the list and logs the incident, which is the same
4773 action as happens when Exim is looking up a host name and its aliases.
4774 Previously in this situation something unpredictable would happen;
4775 sometimes it was "internal error: store_reset failed".
4777 18. If a server dropped the connection unexpectedly when an Exim client was
4778 using GnuTLS and trying to read a response, the client delivery process
4779 crashed while trying to generate an error log message.
4781 19. If a "warn" verb in an ACL added multiple headers to a message in a single
4782 string, for example:
4784 warn message = H1: something\nH2: something
4786 the text was added as a single header line from Exim's point of view
4787 though it ended up OK in the delivered message. However, searching for the
4788 second and subsequent header lines using $h_h2: did not work. This has been
4789 fixed. Similarly, if a system filter added multiple headers in this way,
4790 the routers could not see them.
4792 20. Expanded the error message when iplsearch is called with an invalid key to
4793 suggest using net-iplsearch in a host list.
4795 21. When running tests using -bh, any delays imposed by "delay" modifiers in
4796 ACLs are no longer actually imposed (and a message to that effect is
4799 22. If a "gecos" field in a passwd entry contained escaped characters, in
4800 particular, if it contained a \" sequence, Exim got it wrong when building
4801 a From: or a Sender: header from that name. A second bug also caused
4802 incorrect handling when an unquoted " was present following a character
4803 that needed quoting.
4805 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
4806 was not being matched caselessly.
4808 24. Arranged for all hyphens in the exim.8 source to be escaped with
4811 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
4812 properly. Recipient callout cache records were still being keyed to include
4813 the sender, even when use_sender was set false. This led to far more
4814 callouts that were necessary. The sender is no longer included in the key
4815 when use_sender is false.
4817 26. Added "control = submission" modifier to ACLs.
4819 27. Added the ${base62d: operator to decode base 62 numbers.
4821 28. dnsdb lookups can now access SRV records.
4823 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
4824 the configuration file.
4826 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
4827 (-v) mode. This makes the output for a verbose queue run more intelligible.
4829 31. Added a use_postmaster feature to recipient callouts.
4831 32. Added the $body_zerocount variable, containing the number of binary zero
4832 bytes in the message body.
4834 33. The time of last modification of the "new" subdirectory is now used as the
4835 "mailbox time last read" when there is a quota error for a maildir
4838 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
4840 35. Added +ignore_unknown as a special item in host lists.
4842 36. Code for decoding IPv6 addresses in host lists is now included, even if
4843 IPv6 support is not being compiled. This fixes a bug in which an IPv6
4844 address was recognized as an IP address, but was then not correctly decoded
4845 into binary, causing unexpected and incorrect effects when compared with
4852 1. Very minor rewording of debugging text in manualroute to say "list of
4853 hosts" instead of "hostlist".
4855 2. If verify=header_syntax was set, and a header line with an unqualified
4856 address (no domain) and a large number of spaces between the end of the
4857 name and the colon was received, the reception process suffered a buffer
4858 overflow, and (when I tested it) crashed. This was caused by some obsolete
4859 code that should have been removed. The fix is to remove it!
4861 3. When running in the test harness, delay a bit after writing a bounce
4862 message to get a bit more predictability in the log output.
4864 4. Added a call to search_tidyup() just before forking a reception process. In
4865 theory, someone could use a lookup in the expansion of smtp_accept_max_
4866 per_host which, without the tidyup, could leave open a database connection.
4868 5. Added the variables $recipient_data and $sender_data which get set from a
4869 lookup success in an ACL "recipients" or "senders" condition, or a router
4870 "senders" option, similar to $domain_data and $local_part_data.
4872 6. Moved the writing of debug_print from before to after the "senders" test
4875 7. Change 4.31/66 (moving the time when the Received: is generated) caused
4876 problems for message scanning, either using a data ACL, or using
4877 local_scan() because the Received: header was not generated till after they
4878 were called (in order to set the time as the time of reception completion).
4879 I have revised the way this works. The header is now generated after the
4880 body is received, but before the ACL or local_scan() are called. After they
4881 are run, the timestamp in the header is updated.
4887 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
4888 before starting a queue runner without re-exec. This happened only when
4889 deliver_drop_privilege was set or when the Exim user was set to root. The
4890 effect of the bug was that timeouts during subsequent deliveries caused
4891 crashes instead of being properly handled. The handler is now left at its
4892 default (and expected) setting.
4894 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
4895 message, again when deliver_drop_privilege is set or Exim is run as root.
4896 The bug described in (1) was not present in this case, but the tidying up
4897 of the other signals was missing. I have made the two cases consistent.
4899 3. The ignore_target_hosts setting on a manualroute router was being ignored
4900 for hosts that were looked up using the /MX notation.
4902 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
4905 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
4906 operated on the sender address. After changing the $sender_address to <>
4907 for the sender address verify, Exim was re-instated it as the original
4908 (before rewriting) address, but remembering that it had rewritten it, so it
4909 wasn't rewriting it again. This bug also had the effect of breaking the
4910 sender address verification caching when the sender address was rewritten.
4912 6. The ignore_target_hosts option was being ignored by the ipliteral router.
4913 This has been changed so that if the ip literal address matches
4914 ignore_target_hosts, the router declines.
4916 7. Added expansion conditions match_domain, match_address, and match_local_
4917 part (NOT match_host).
4919 8. The placeholder for the Received: header didn't have a length field set.
4921 9. Added code to Exim itself and to exim_lock to test for a specific race
4922 condition that could lead to file corruption when using MBX delivery. The
4923 issue is with the lockfile that is created in /tmp. If this file is removed
4924 after a process has opened it but before that process has acquired a lock,
4925 there is the potential for a second process to recreate the file and also
4926 acquire a lock. This could lead to two Exim processes writing to the file
4927 at the same time. The added code performs the same test as UW imapd; it
4928 checks after acquiring the lock that its file descriptor still refers to
4929 the same named file.
4931 10. The buffer for building added header lines was of fixed size, 8192 bytes.
4932 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
4935 11. Added the smtp_active_hostname option. If used, this will typically be made
4936 to depend on the incoming interface address. Because $interface_address is
4937 not set up until the daemon has forked a reception process, error responses
4938 that can happen earlier (such as "too many connections") no longer contain
4941 12. If an expansion in a condition on a "warn" statement fails because a lookup
4942 defers, the "warn" statement is abandoned, and the next ACL statement is
4943 processed. Previously this caused the whole ACL to be aborted.
4945 13. Added the iplsearch lookup type.
4947 14. Added ident_timeout as a log selector.
4949 15. Added tls_certificate_verified as a log selector.
4951 16. Added a global option tls_require_ciphers (compare the smtp transport
4952 option of the same name). This controls incoming TLS connections.
4954 17. I finally figured out how to make tls_require_ciphers do a similar thing
4955 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
4956 before starting the TLS session.
4958 18. Tabs are now shown as \t in -bP output.
4960 19. If the log selector return_path_on_delivery was set, Exim crashed when
4961 bouncing a message because it had too many Received: header lines.
4963 20. If two routers both had headers_remove settings, and the first one included
4964 a superfluous trailing colon, the final name in the first list and the
4965 first name in the second list were incorrectly joined into one item (with a
4966 colon in the middle).
4972 1. Added -C and -D options to the exinext utility, mainly to make it easier
4973 to include in the automated testing, but these could be helpful when
4974 multiple configurations are in use.
4976 2. The exinext utility was not formatting the output nicely when there was
4977 an alternate port involved in the retry record key, nor when there was a
4978 message id as well (for retries that were specific to a specific message
4979 and a specific host). It was also confused by IPv6 addresses, because of
4980 the additional colons they contain. I have fixed the IPv4 problem, and
4981 patched it up to do a reasonable job for IPv6.
4983 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
4984 delivery, the log line now contains "pipelined" if PIPELINING was used.
4986 4. An SMTP transport process used to panic and die if the bind() call to set
4987 an explicit outgoing interface failed. This has been changed; it is now
4988 treated in the same way as a connect() failure.
4990 5. A reference to $sender_host_name in the part of a conditional expansion
4991 that was being skipped was still causing a DNS lookup. This no longer
4994 6. The def: expansion condition was not recognizing references to header lines
4995 that used bh_ and bheader_.
4997 7. Added the _cache feature to named lists.
4999 8. The code for checking quota_filecount in the appendfile transport was
5000 allowing one more file than it should have been.
5002 9. For compatibility with Sendmail, the command line option
5010 and sets the incoming protocol and host name (for trusted callers). The
5011 host name and its colon can be omitted when only the protocol is to be set.
5012 Note the Exim already has two private options, -pd and -ps, that refer to
5013 embedded Perl. It is therefore impossible to set a protocol value of "d" or
5014 "s", but I don't think that's a major issue.
5016 10. A number of refactoring changes to the code, none of which should affect
5019 (a) The number of logging options was getting close to filling up the
5020 32-bit word that was used as a bit map. I have split them into two classes:
5021 those that are passed in the argument to log_write(), and those that are
5022 only ever tested independently outside of that function. These are now in
5023 separate 32-bit words, so there is plenty of room for expansion again.
5024 There is no change in the user interface or the logging behaviour.
5026 (b) When building, for example, log lines, the code previously used a
5027 macro that called string_cat() twice, in order to add two strings. This is
5028 not really sufficiently general. Furthermore, there was one instance where
5029 it was actually wrong because one of the argument was used twice, and in
5030 one call a function was used. (As it happened, calling the function twice
5031 did not affect the overall behaviour.) The macro has been replaced by a
5032 function that can join an arbitrary number of extra strings onto a growing
5035 (c) The code for expansion conditions now uses a table and a binary chop
5036 instead of a serial search (which was left over from when there were very
5037 few conditions). Also, it now recognizes conditions like "pam" even when
5038 the relevant support is not compiled in: a suitably worded error message is
5039 given if an attempt is made to use such a condition.
5041 11. Added ${time_interval:xxxxx}.
5043 12. A bug was causing one of the ddress fields not to be passed back correctly
5044 from remote delivery subprocesses. The field in question was not being
5045 subsequently used, so this caused to problems in practice.
5047 13. Added new log selectors queue_time and deliver_time.
5049 14. Might have fixed a bug in maildirsizefile handling that threw up
5050 "unexpected character" debug warnings, and recalculated the data
5051 unnecessarily. In any case, I expanded the warning message to give more
5054 15. Added the message "Restricted characters in address" to the statements in
5055 the default ACL that block characters like @ and % in local parts.
5057 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
5058 Three changes have been made:
5060 (a) There was a serious bug; a negative response to MAIL caused the whole
5061 recipient domain to be cached as invalid, thereby blocking all messages
5062 to all local parts at the same domain, from all senders. This bug has
5063 been fixed. The domain is no longer cached after a negative response to
5064 MAIL if the sender used is not empty.
5066 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
5069 (c) A new callout option, "use_sender" has been added for people who want
5070 the modified behaviour.
5076 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
5079 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
5080 indeed breaks things for older releases.
5082 3. Added additional logging to the case where there is a problem reading data
5083 from a filter that is running in a subprocess using a pipe, in order to
5084 try to track down a specific problem.
5086 4. Testing facility fudge: when running in the test harness and attempting
5087 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
5088 getting "No route to host". Convert this to a timeout.
5090 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
5093 6. Some OS don't have socklen_t but use size_t instead. This affects the
5094 fifth argument of getsockopt() amongst other things. This is now
5095 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
5096 can be set for individual OS. I have set it for SunOS5, OSF1, and
5097 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
5098 some earlier ones do not.
5100 7. Change 4.30/15 was not doing the test caselessly.
5102 8. The standard form for an IPv6 address literal was being rejected by address
5103 parsing in, for example, MAIL and RCPT commands. An example of this kind of
5104 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5105 this, as well as the form without the "IPv6" on the front (but only when
5106 address literals are enabled, of course).
5108 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
5110 10. Exim crashed if a message with an empty sender address specified by -f
5111 encountered a router with an errors_to setting. This could be provoked only
5112 by a command such as
5116 where an empty string was supplied; "<>" did not hit this bug.
5118 11. Installed PCRE release 4.5.
5120 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
5121 remained set. It is now erased.
5123 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
5124 times from message ids (which are base 36 rather than the normal 62).
5126 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
5127 were being counted as actual protocol errors, and logged if the log
5128 selector +smtp_protocol_error was set. One cannot be perfect in this test,
5129 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
5130 and DATA following a set of rejected RCPTs do not count as protocol errors.
5131 In other words, Exim assumes they were pipelined, though this may not
5132 actually be the case. Of course, in all cases the client gets an
5133 appropriate error code.
5135 15. If a lookup fails in an ACL condition, a message about the failure may
5136 be available; it is used if testing the ACL cannot continue, because most
5137 such messages specify what the cause of the deferral is. However, some
5138 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
5139 that caused an old message to be retained and used if a later statement
5140 caused a defer, replacing the real cause of the deferral.
5142 16. If an IP address had so many PTR records that the DNS lookup buffer
5143 was not large enough to hold them, Exim could crash while trying to process
5144 the truncated data. It now detects and logs this case.
5146 17. Further to 4.21/58, another change has been made: if (and only if) the
5147 first line of a message (the first header line) ends with CRLF, a bare LF
5148 in a subsequent header line has a space inserted after it, so as not to
5149 terminate the header.
5151 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
5152 unnecessarily, used atoi() instead of strtol(), and didn't check the
5153 termination when getting file sizes from file names by regex.
5155 19. Completely re-implemented the support for maildirsize files, in the light
5156 of a number of problems with the previous contributed implementation
5157 (4.30/29). In particular:
5159 . If the quota is zero, the maildirsize file is maintained, but no quota is
5162 . If the maildir directory does not exist, it is created before any attempt
5163 to write a maildirsize file.
5165 . The quota value in the file is just a cache; if the quota is changed in
5166 the transport, the new value overrides.
5168 . A regular expression is available for excluding directories from the
5171 20. The autoreply transport checks the characters in options that define the
5172 message's headers; it allows continued headers, but it was checking with
5173 isspace() after an embedded newline instead of explicitly looking for a
5176 21. If all the "regular" hosts to which an address was routed had passed their
5177 expiry times, and had not reached their retry times, the address was
5178 bounced, even if fallback hosts were defined. Now Exim should go on to try
5181 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
5182 equivalent code in the SMTP transport. Some hosts send humungous responses
5183 to HELO/EHLO, more than 1024 it seems.
5185 23. Refactoring: code in filter.c used (void *) for "any old type" but this
5186 gives compiler warnings in some environments. I've now done it "properly",
5189 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
5190 (because of problems with the built-in one) was declared to return uschar *
5191 instead of char *, causing compiler failure.
5193 25. Fixed a file descriptor leak when processing alias/forward files.
5195 26. Fixed a minor format string issue in dbfn.c.
5197 27. Typo in exim.c: ("dmbnz" for "dbmnz").
5199 28. If a filter file refered to $h_xxx or $message_headers, and the headers
5200 contained RFC 2047 "words", Exim's memory could, under certain conditions,
5203 29. When a sender address is verified, it is cached, to save repeating the test
5204 when there is more than one recipient in a message. However, when the
5205 verification involves a callout, it is possible for different callout
5206 options to be set for different recipients. It is too complicated to keep
5207 track of this in the cache, so now Exim always runs a verification when a
5208 callout is required, relying on the callout cache for the optimization.
5209 The overhead is duplication of the address routing, but this should not be
5212 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
5213 to be verified with callout=postmaster, and the main callout worked but the
5214 postmaster check failed, the verification correctly failed. However, if a
5215 subsequent RCPT command asked for sender verification *without* the
5216 postmaster check, incorrect caching caused this verification also to fail,
5219 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
5220 it was not caching the DNS options (qualify_single, search_parents) that
5221 were used when the lookup failed. A subsequent lookup with different
5222 options therefore always gave the same answer, though there were cases
5223 where it should not have. (Example: a "domains = !$mx_any" option on a
5224 dnslookup router: the "domains" option is always processed without any
5225 widening, but the router might have qualify_single set.) Now Exim uses the
5226 cached value only when the same options are set.
5228 32. Added John Jetmore's "exipick" utility to the distribution.
5230 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
5231 than a timeout (e.g. a certificate is required, and is not provided), an
5232 Exim server now closes the connection immediately. Previously it waited for
5233 the client to close - but if the client is SSL, it seems that they each
5234 wait for each other, leading to a delay before one of them times out.
5236 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
5237 maintained 0.8.x compatibility because I don't think many are using it, and
5238 it is clearly obsolete.
5240 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
5243 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
5244 client certificate was expired. A simple patch fixes this, though I don't
5245 understand the full logic of why the verify callback is called multiple
5248 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
5249 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
5250 which causes problems with some clients (such as the Certicom SSL Plus
5251 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
5252 disables the coutermeasure allowing Eudora to connect."
5254 38. Exim was not checking that a write() to a log file succeeded. This could
5255 lead to Bad Things if a log got too big, in particular if it hit a file
5256 size limit. Exim now panics and dies if it cannot write to a log file, just
5257 as it does if it cannot open a log file.
5259 39. Modified OS/Makefile-Linux so that it now contains
5261 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
5263 The two -D definitions ensure that Exim is compiled with large file
5264 support, which makes it possible to handle log files that are bigger than
5267 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
5268 instance) a domain was checked against a named list that involved a lookup,
5269 causing $domain_data to be set, then another domain was checked against the
5270 same list, then the first domain was re-checked, the value of $domain_data
5271 after the final check could be wrong. In particular, if the second check
5272 failed, it could be set empty. This bug probably also applied to
5275 41. The strip_trailing_dot option was not being applied to the address given
5276 with the -f command-line option.
5278 42. The code for reading a message's header from the spool was incrementing
5279 $received_count, but never initializing it. This meant that the value was
5280 incorrect (doubled) while delivering a message in the same process in which
5281 it was received. In the most common configuration of Exim, this never
5282 happens - a fresh exec is done - but it can happen when
5283 deliver_drop_privilege is set.
5285 43. When Exim logs an SMTP synchronization error - client data sent too soon -
5286 it now includes up to 150 characters of the unexpected data in the log
5289 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
5290 and building data strings. The size of both of these buffers was 10 000
5291 bytes - far larger than anybody would *ever* want, thought I. Needless to
5292 say, somebody hit the limit. I have increased the maximum line length to
5293 20 000 and the maximum data length of concatenated lines to 100 000. I have
5294 also fixed two bugs, because there was no checking on these buffers. Tsk,
5295 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
5296 buffer is too small.
5298 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
5299 lsearch lookups. Now it does.
5301 46. When parsing a route_list item in a manualroute router, a fixed-length
5302 buffer was used for the list of hosts. I made this 1024 bytes long,
5303 thinking that nobody would ever have a list of hosts that long. Wrong.
5304 Somebody had a whole pile of complicated expansion conditions, and the
5305 string was silently truncated, leading to an expansion error. It turns out
5306 that it is easier to change to an unlimited length (owing to other changes
5307 that have happened since this code was originally written) than to build
5308 structure for giving a limitation error. The length of the item that
5309 expands into the list of hosts is now unlimited.
5311 47. The lsearch lookup could not handle data where the length of text line was
5312 more than 4095 characters. Such lines were truncated, leading to shortened
5313 data being returned. It should now handle lines of any length.
5315 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
5316 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
5319 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
5320 replacing strings such as BIN_DIRECTORY in a source file: the replacement
5321 no longer happens in comment lines. A list of replacements is now placed
5322 at the head of all of the source files, except those whose only change is
5323 to replace PERL_COMMAND in the very first #! line.
5325 50. Replaced the slow insertion sort in queue.c, for sorting the list of
5326 messages on the queue, with a bottom-up merge sort, using code contributed
5327 by Michael Haardt. This should make operations like -bp somewhat faster on
5328 large queues. It won't affect queue runners, except when queue_run_in_order
5331 51. Installed eximstats 1.31 in the distribution.
5333 52. Added support for SRV lookups to the dnslookup router.
5335 53. If an ACL referred to $message_body or $message_body_end, the value was not
5336 reset for any messages that followed in the same SMTP session.
5338 54. The store-handling optimization for building very long strings was not
5339 differentiating between the different store pools. I don't think this
5340 actually made any difference in practice, but I've tidied it.
5342 55. While running the routers to verify a sender address, $sender_address
5343 was still set to the sender address. This is wrong, because when routing to
5344 send a bounce to the sender, it would be empty. Therefore, I have changed
5345 it so that, while verifying a sender address, $sender_address is set to <>.
5346 (There is no change to what happens when verifying a recipient address.)
5348 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
5349 target A or AAAA records (if not already returned) without resetting the
5350 qualify_single or search_parents options of the DNS resolver. These are
5351 inappropriate in this case because the targets of MX and SRV records must
5352 be FQDNs. A broken DNS record could cause trouble if it happened to have a
5353 target that, when qualified, matched something in the local domain. These
5354 two options are now turned off when doing these lookups.
5356 57. It seems that at least some releases of Reiserfs (which does not have the
5357 concept of a fixed number of inodes) returns zero and not -1 for the
5358 number of available inodes. This interacted badly with check_spool_inodes,
5359 which assumed that -1 was the "no such thing" setting. What I have done is
5360 to check that the total number of inodes is greater than zero before doing
5361 the test of how many are available.
5363 58. When a "warn" ACL statement has a log_message modifier, the message is
5364 remembered, and not repeated. This is to avoid a lot of repetition when a
5365 message has many recipients that cause the same warning to be written.
5366 Howewer, Exim was preserving the list of already written lines for an
5367 entire SMTP session, which doesn't seem right. The memory is now reset if a
5368 new message is started.
5370 59. The "rewrite" debugging flag was not showing the result of rewriting in the
5371 debugging output unless log_rewrite was also set.
5373 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
5374 of (int)(handle) when we know that handle contains (void *)(-1).
5376 61. The Exim daemon panic-logs an error return when it closes the incoming
5377 connection. However "connection reset by peer" seems to be common, and
5378 isn't really an error worthy of noting specially, so that particular error
5381 62. When Exim is trying to find all the local interfaces, it used to panic and
5382 die if the ioctl to get the interface flags failed. However, it seems that
5383 on at least one OS (Solaris 9) it is possible to have an interface that is
5384 included in the list of interfaces, but for which you get a failure error
5385 for this call. This happens when the interface is not "plumbed" into a
5386 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
5387 failure of the "get flags" call assumes that the interface is down.
5389 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
5390 makes life easier for people who are doing arithmetic on fields extracted
5391 from dates, where you often get leading zeros that should not be
5392 interpreted as octal.
5394 64. Added qualify_domain to the redirect router, to override the global
5397 65. If a pathologically long header line contained very many addresses (the
5398 report of this problem mentioned 10 000) and each of them was rewritten,
5399 Exim could use up a very large amount of memory. (It kept on making new
5400 copies of the header line as it rewrote, and never released the old ones.)
5401 At the expense of a bit more processing, the header rewriting function has
5402 been changed so that it no longer eats memory in this way.
5404 66. The generation of the Received: header has been moved from the time that a
5405 message starts to be received, to the time that it finishes. The timestamp
5406 in the Received: header should now be very close to that of the <= log
5407 line. There are two side-effects of this change:
5409 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
5410 logged header lines no longer include the local Received: line, because
5411 it has not yet been created. The same applies to a copy of the message
5412 that is returned to a non-SMTP sender when a message is rejected.
5414 (b) When a filter file is tested using -bf, no additional Received: header
5415 is added to the test message. After some thought, I decided that this
5418 This change does not affect the value of $received_for. It is still set
5419 after address rewriting, but before local_scan() is called.
5421 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
5423 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
5424 gave an unhelpful panic error message, and a defer error. I have managed to
5425 change this behaviour so that it now rejects any supplied certificate,
5426 which seems right, as the list of acceptable certificates is empty.
5428 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
5429 gave an unhelpful defer error. I have not managed to make this reject any
5430 supplied certificates, but the error message it gives is "no certificate
5431 supplied", which is not helpful.
5433 70. exigrep's output now also includes lines that are not associated with any
5434 message, but which match the given pattern. Implemented by a patch from
5435 Martin Sluka, which also tidied up the Perl a bit.
5437 71. Recipient callout verification, like sender verification, was using <> in
5438 the MAIL FROM command. This isn't really the right thing, since the actual
5439 sender may affect whether the remote host accepts the recipient or not. I
5440 have changed it to use the actual sender in the callout; this means that
5441 the cache record is now keyed on a recipient/sender pair, not just the
5442 recipient address. There doesn't seem to be a real danger of callout loops,
5443 since a callout by the remote host to check the sender would use <>.
5444 [SEE ABOVE: changed after hitting problems.]
5446 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
5447 temporary errors. However, in the case of such a code being given after
5448 the end of a data transmission (i.e. after ".") Exim was failing to write
5449 a retry record for the message. (Yes, there was some broken host that was
5450 actually sending 8xx at this point.)
5452 73. An unknown lookup type in a host list could cause Exim to panic-die when
5453 the list was checked. (An example that provoked this was putting <; in the
5454 middle of a list instead of at the start.) If this happened during a DATA
5455 ACL check, a -D file could be left lying around. This kind of configuration
5456 error no longer causes Exim to die; instead it causes a defer errror. The
5457 incident is still logged to the main and panic logs.
5459 74. Buglet left over from Exim 3 conversion. The message "too many messages
5460 in one connection" was written to the rejectlog but not the mainlog, except
5461 when address rewriting (yes!) was being logged.
5463 75. Added write_rejectlog option.
5465 76. When a system filter was run not as root (that is, when system_filter_user
5466 was set), the values of the $n variables were not being returned to the
5467 main process; thus, they were not subsequently available in the $sn
5470 77. Added +return_path_on_delivery log selector.
5472 78. A connection timeout was being treated differently from recipients deferred
5473 when testing hosts_max_try with a message that was older than the host's
5474 retry timeout. (The host should not be counted, thus allowing all hosts to
5475 be tried at least once before bouncing.) This may have been the cause of an
5476 occasionally reported bug whereby a message would remain on the queue
5477 longer than the retry timeout, but would be bounced if a delivery was
5478 forced. I say "may" because I never totally pinned down the problem;
5479 setting up timeout/retry tests is difficult. See also the next item.
5481 79. The ultimate address timeout was not being applied to errors that involved
5482 a combination of host plus message (for example, a timeout on a MAIL
5483 command). When an address resolved to a number of possible hosts, and they
5484 were not all tried for each delivery (e.g. because of hosts_max_try), a
5485 message could remain on the queue longer than the retry timeout.
5487 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
5490 81. Fixed an obscure SMTP outgoing bug which required at least the following
5491 conditions: (a) there was another message waiting for the same server;
5492 (b) the server returned 5xx to all RCPT commands in the first message so
5493 that the message was not completed; (c) the server dropped the connection
5494 or gave a negative response to the RSET that Exim sends to abort the
5495 transaction. The observed case was a dropped connection after DATA that had
5496 been sent in pipelining mode. That is, the server had advertised PIPELINING
5497 but was not implementing it correctly. The effect of the bug was incorrect
5498 behaviour, such as trying another host, and this could lead to a crash.
5504 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
5505 and daemon.c were passed as pointers to ints; they should have been
5506 pointers to socklen_t variables (which are typically unsigned ints).
5508 2. Some signed/unsigned type warnings in the os.c file for Linux have been
5511 3. Fixed a really odd bug that affected only the testing scheme; patching a
5512 certain fixed string in the binary changed the value of another string that
5513 happened to be identical to the end of the original first string.
5515 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
5516 name", it returns that address as the IP address. On some operating
5517 systems (e.g. Solaris), it also passes back the IP address string as the
5518 "host name". However, on others (e.g. Linux), it passes back an empty
5519 string. Exim wasn't checking for this, and was changing the host name to an
5520 empty string, assuming it had been canonicized.
5522 5. Although rare, it is permitted to have more than one PTR record for a given
5523 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
5524 all the names associated with an address, because they do in Solaris.
5525 However, it seems that they do not in Linux for data that comes from the
5526 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
5527 I found this out when I moved to a new Linux workstation and tried to run
5528 the Exim test suite.
5530 To get round this problem I have changed the code so that it now does its
5531 own call to the DNS to look up PTR records when searching for a host name.
5532 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
5533 addresses that are only in /etc/hosts are still found.
5535 This behaviour is, however, controlled by an option called host_lookup_
5536 order, which defaults to "bydns:byaddr". If people want to use the other
5537 order, or indeed, just use one or the other means of lookup, they can
5538 specify it in this variable.
5540 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
5541 some operating systems, this comes back from gethostbyaddr() as an empty
5542 string, and this is what Exim used to test for. However, it seems that in
5543 other systems, "." is yielded. Exim now tests for this case too.
5545 7. The values of check_spool_space and check_log_space are now held internally
5546 as a number of kilobytes instead of an absolute number of bytes. If a
5547 numbers is specified without 'K' or 'M', it is rounded up to the nearest
5548 kilobyte. This means that much larger values can be stored.
5550 8. Exim monitor: an attempt to get the action menu when not actually pointing
5551 at a message produces an empty menu entitled "No message selected". This
5552 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
5553 no entries in it ("Shell widget menu has zero width and/or height"). So I
5554 have added a single, blank menu entry in this case.
5556 9. Added ${quote_local_part.
5558 10. MIME decoding is now applied to the contents of Subject: header lines when
5561 11. Now that a reference to $sender_host_address automatically causes a reverse
5562 lookup to occur if necessary (4.13/18), there is no need to arrange for a
5563 host lookup before query-style lookups in lists that might use this
5564 variable. This has therefore been abolished, and the "net-" prefix is no
5565 longer necessary for query-style lookups.
5567 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
5568 have been a typo for LFLAGS, so it has been changed.
5570 13. The install script calls Exim with "-C /dev/null" in order to find the
5571 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
5572 to be output. Howeve, since Exim outputs its version number before the
5573 error, it didn't break the script. It just looked ugly. I fixed this by
5574 always allowing "-C /dev/null" if the caller is root.
5576 14. Ignore overlarge ACL variable number when reading spool file - insurance
5577 against a later release with more variables having written the file.
5579 15. The standard form for an IPv6 address literal was being rejected by EHLO.
5580 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5581 this, as well as the form without the "IPv6" on the front.
5583 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
5584 OS/Makefile-Darwin file.
5586 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
5587 only to LDAP libraries that do not have LDAP_OPT_DEREF.
5589 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
5590 variable. However, some OS use ints for inodes. I've added cast to long int
5591 to get rid of the compiler warning.
5593 19. I had forgotten to lock out "/../" in configuration file names when
5594 ALT_CONFIG_PREFIX was set.
5596 20. Routers used for verification do not need to specify transports. However,
5597 if such a router generated a host list, and callout was configured, Exim
5598 crashed, because it could not find a port number from the (non-existent)
5599 transport. It now assumes port 25 in this circumstance.
5601 21. Added the -t option to exigrep.
5603 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
5604 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
5605 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
5606 them is set, LOOKUP_LSEARCH is forced.
5608 23. "exim -bV" now outputs a list of lookups that are included in the binary.
5610 24. Added sender and host information to the "rejected by local_scan()" log
5611 line; previously there was no indication of these.
5613 25. Added .include_if_exists.
5615 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
5616 Linux. It turns out that not all file systems support this. Apparently some
5617 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
5618 do it.) To cope with this, the error EINVAL, which means that sync-ing is
5619 not supported on the file descriptor, is now ignored when Exim is trying to
5620 sync a directory. This applies only to Linux.
5622 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
5624 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
5625 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
5628 29. Added support for maildirsize files from supplied patch (modified a bit).
5630 30. The use of :fail: followed by an empty string could lead Exim to respond to
5631 sender verification failures with (e.g.):
5633 550 Verification failed for <xxx>
5634 550 Sender verify failed
5636 where the first response line was missing the '-' that indicates it is not
5637 the final line of the response.
5639 31. The loop for finding the name of the user that called Exim had a hardwired
5640 limit of 10; it now uses the value of finduser_retries, which is used for
5641 all other user lookups.
5643 32. Added $received_count variable, available in data and not_smtp ACLs, and at
5646 33. Exim was neglecting to zero errno before one call of strtol() when
5647 expanding a string and expecting an integer value. On some systems this
5648 resulted in spurious "integer overflow" errors. Also, it was casting the
5649 result into an int without checking.
5651 34. Testing for a connection timeout using "timeout_connect" in the retry rules
5652 did not work. The code looks as if it has *never* worked, though it appears
5653 to have been documented since at least releast 1.62. I have made it work.
5655 35. The "timeout_DNS" error in retry rules, also documented since at least
5656 1.62, also never worked. As it isn't clear exactly what this means, and
5657 clearly it isn't a major issue, I have abolished the feature by treating it
5658 as "timeout", and writing a warning to the main and panic logs.
5660 36. The display of retry rules for -brt wasn't always showing the error code
5663 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
5664 timeout_connect_A, timeout_connect_MX.
5666 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
5667 to the empty sender.
5669 39. The daemon was not analysing the content of -oX till after it had closed
5670 stderr and disconnected from the controlling terminal. This meant that any
5671 syntax errors were only noted on the panic log, and the return code from
5672 the command was 0. By re-arranging the code a little, I've made the
5673 decoding happen first, so such errors now appear on stderr, and the return
5674 code is 1. However, the actual setting up of the sockets still happens in
5675 the disconnected process, so errors there are still only recorded on the
5678 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
5679 connections (as happens on some IP stacks) was logged at start up time as
5680 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
5681 it from "IPv6 and IPv4", which means that two separate sockets are being
5684 41. The debug output for gethostbyname2() or getipnodebyname() failures now
5685 says whether AF_INET or AF_INET6 was passed as an argument.
5687 42. Exiwhat output was messed up when time zones were included in log
5690 43. Exiwhat now gives more information about the daemon's listening ports,
5691 and whether -tls-on-connect was used.
5693 44. The "port" option of the smtp transport is now expanded.
5695 45. A "message" modifier in a "warn" statement in a non-message ACL was being
5696 silently ignored. Now an error message is written to the main and panic
5699 46. There's a new ACL modifier called "logwrite" which writes to a log file
5700 as soon as it is encountered.
5702 47. Added $local_user_uid and $local_user_gid at routing time.
5704 48. Exim crashed when trying to verify a sender address that was being
5707 49. Exim was recognizing only a space character after ".include". It now also
5708 recognizes a tab character.
5710 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
5711 extracting the relevant information from the specification. The man page no
5712 longer contains scrambled data for the -d option, and I've added a section
5713 at the front about calling Exim under different names.
5715 51. Added "extra_headers" argument to the "mail" command in filter files.
5717 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
5720 53. Installed eximstats 1.29.
5722 54. Added transport_filter_timeout as a generic transport option.
5724 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
5725 Cc: header lines. This was required by RFC 822, but it not required by RFC
5728 56. Exim used to add From:, Date:, and Message-Id: header lines to any
5729 incoming messages that did not have them. Now it does so only if the
5730 message originates locally, that is, if there is no associated remote host
5731 address. When Resent- header lines are present, this applies to the Resent-
5732 lines rather than the non-Resent- lines.
5734 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
5735 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
5737 58. Messages for configuration errors now include the name of the main
5738 configuration file - useful now that there may be more than one file in a
5739 list (.included file names were always shown).
5741 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
5742 for those rare installations that do not start the daemon as root or run it
5743 setuid root. I've cut out the call to initgroups() if the daemon is not
5746 60. The Exim user and group can now be bound into the binary as text strings
5747 that are looked up at the start of Exim's processing.
5749 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
5751 62. Added $mailstore_basename variable.
5753 63. Installed patch to sieve.c from Michael Haardt.
5755 64. When Exim failed to open the panic log after failing to open the main log,
5756 the original message it was trying to log was written to stderr and debug
5757 output, but if they were not available (the usual case in production), it
5758 was lost. Now it is written to syslog before the two lines that record the
5759 failures to open the logs.
5761 65. Users' Exim filters run in subprocesses under the user's uid. It is
5762 possible for a "deliver" command or an alias in a "personal" command to
5763 provoke an address rewrite. If logging of address rewriting is configured,
5764 this fails because the process is not running as root or exim. There may be
5765 a better way of dealing with this, but for the moment (because 4.30 needs
5766 to be released), I have disabled address rewrite logging when running a
5767 filter in a non-root, non-exim process.
5773 1. The buildconfig auxiliary program wasn't quoting the value set for
5774 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
5775 not defined. This bug was masked in 4.22 by the effect that was fixed in
5778 2. Some messages that were rejected after a message id was allocated were
5779 shown as "incomplete" by exigrep. It no longer does this for messages that
5780 are rejected by local_scan() or the DATA or non-SMTP ACLs.
5782 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
5783 have allow_domain_literals set, the ID did not get logged in the <= line.
5784 Domain literals are now always recognized in Message-ID: header lines.
5786 4. The first argument for a ${extract expansion item is the key name or field
5787 number. Leading and trailing spaces in this item were not being ignored,
5788 causing some misleading effects.
5790 5. When deliver_drop_privilege was set, single queue runner processes started
5791 manually (i.e. by the command "exim -q") or by the daemon (which uses the
5792 same command in the process it spins off) were not dropping privilege.
5794 6. When the daemon running as "exim" started a queue runner, it always
5795 re-executed Exim in the spun-off process. This is a waste of effort when
5796 deliver_drop_privilege is set. The new process now just calls the
5797 queue-runner function directly.
5803 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
5806 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
5807 ignored. Though the use of -oP was forcing the writing of a pid file, it
5808 was always written to the default place.
5810 3. If the message "no IP address found for host xxxx" is generated during
5811 incoming verification, it is now followed by identification of the incoming
5812 connection (so you can more easily find what provoked it).
5814 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
5816 5. Added some features to "harden" Exim a bit more against certain attacks:
5818 (a) There is now a build-time option called FIXED_NEVER_USERS that can
5819 be put in Local/Makefile. This is like the never_users runtime option,
5820 but it cannot be overridden. The default setting is "root".
5822 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
5823 prefix string with which any file named in a -C command line option
5826 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
5827 is retained for -C and -D only if the caller of Exim is root. Without
5828 it, the exim user may also use -C and -D and retain privilege.
5830 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
5831 command line option is disabled.
5833 6. Macro names set by the -D option must start with an upper case letter, just
5834 like macro names defined in the configuration file.
5836 7. Added "dereference=" facility to LDAP.
5838 8. Two instances of the typo "uknown" in the source files are fixed.
5840 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
5841 the Configure-Makefile script screwed up while processing it.
5843 10. Incorporated PCRE 4.4.
5845 11. The SMTP synchronization check was not operating right at the start of an
5846 SMTP session. For example, it could not catch a HELO sent before the client
5847 waited for the greeting. There is now a check for outstanding input at the
5848 point when the greeting is written. Because of the duplex, asynchronous
5849 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
5850 way, but not yet received, when the check is performed.
5852 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
5853 on TCP/IP sockets, because this apparently causes some broken clients to
5856 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
5857 unchanged) from the Cygwin maintainer.
5859 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
5860 was supported instead of testing for mbx. Effectively a typo.
5862 15. The spa authenticator server code was not checking that the input it
5863 received was valid base64.
5865 16. The debug output line for the "set" modifier in ACLs was not showing the
5866 name of the variable that was being set.
5868 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
5870 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
5871 lookup on its own. Something else was needed to trigger the lookup. For
5872 example, a match in host_lookup or the need for a host name in a host list.
5873 Now, if $sender_host_name is referenced and the host name has not yet been
5874 looked up, a lookup is performed. If the lookup fails, the variable remains
5875 empty, and $host_lookup_failed is set to "1".
5877 19. Added "eqi" as a case-independent comparison operator.
5879 20. The saslauthd authentication condition could segfault if neither service
5880 nor realm was specified.
5882 21. If an overflowing value such as "2048M" was set for message_size_limit, the
5883 error message that was logged was misleading, and incoming SMTP
5884 connections were dropped. The message is now more accurate, and temporary
5885 errors are given to SMTP connections.
5887 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
5888 (except RSET) with a 421 error, until QUIT is received. However, it was
5889 failing to send a response to QUIT.
5891 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
5892 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
5893 ACL is now run after the helo_try_verify_hosts code.
5895 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
5896 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
5897 case-independent, so other case variants are also recognized). Apparently
5898 some systems use these upper case variants.
5900 25. If more than two messages were waiting for the same host, and a transport
5901 filter was specified for the transport, Exim sent two messages over the
5902 same TCP/IP connection, and then failed with "socket operation on non-
5903 socket" when it tried to send the third.
5905 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
5907 27. The extern definition of crypt16() in expand.c was not being excluded when
5908 the OS had its own crypt16() function.
5910 28. Added bounce_return_body as a new option, and bounce_return_size_limit
5911 as a preferred synonym for return_size_limit, both as an option and as an
5914 29. Added LIBS=-liconv to OS/Makefile-OSF1.
5916 30. Changed the default configuration ACL to relax the local part checking rule
5917 for addresses that are not in any local domains. For these addresses,
5918 slashes and pipe symbols are allowed within local parts, but the sequence
5919 /../ is explicitly forbidden.
5921 31. SPA server authentication was not clearing the challenge buffer before
5924 32. log_message in a "warn" ACL statement was writing to the reject log as
5925 well as to the main log, which contradicts the documentation and doesn't
5926 seem right (because no rejection is happening). So I have stopped it.
5928 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
5929 However, I am unable to do any testing of this.
5931 34. Fixed an infelicity in the appendfile transport. When checking directories
5932 for a mailbox, to see if any needed to be created, it was accidentally
5933 using path names with one or more superfluous leading slashes; tracing
5934 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
5936 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
5937 contents are added to the log line that is written for every discarded
5938 recipient. (Previously a log_message setting was ignored.)
5940 36. The ${quote: operator now quotes the string if it is empty.
5942 37. The install script runs exim in order to find its version number. If for
5943 some reason other than non-existence or emptiness, which it checks, it
5944 could not run './exim', it was installing it with an empty version number,
5945 i.e. as "exim-". This error state is now caught, and the installation is
5948 38. An argument was missing from the function that creates an error message
5949 when Exim fails to connect to the socket for saslauthd authentication.
5950 This could cause Exim to crash, or give a corrupted message.
5952 39. Added isip, isip4, and isip6 to ${if conditions.
5954 40. The ACL variables $acl_xx are now saved with the message, and can be
5955 accessed later in routers, transports, and filters.
5957 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
5958 strings in the file are not string-expanded.
5960 42. If a MAIL command specified a SIZE value that was too large to fit into an
5961 int variable, the check against message_size_limit failed. Such values are
5962 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
5963 day this will have to be increased, but I don't think I want to be around
5964 when emails are that large.
5971 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
5972 iconv() is not standard in FreeBSD.
5974 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
5975 IPv6 enabled. The observed symptom was a segmentation fault on return from
5976 the function os_common_find_running_interfaces() in src/os.c.
5978 3. In the check_special_case() function in daemon.c I had used "errno" as an
5979 argument name, which causes warnings on some systems. This was basically a
5980 typo, since it was named "eno" in the comments!
5982 4. The code that waits for the clock to tick (at a resolution of some fraction
5983 of a second) so as to ensure message-id uniqueness was always waiting for
5984 at least one whole tick, when it could have waited for less. [This is
5985 almost certainly not relevant at current processor speeds, where it is
5986 unlikely to ever wait at all. But we try to future-proof.]
5988 5. The function that sleeps for a time interval that includes fractions of a
5989 second contained a race. It did not block SIGALRM between setting the
5990 timer, and suspending (a couple of lines later). If the interval was short
5991 and the sigsuspend() was delayed until after it had expired, the suspension
5992 never ended. On busy systems this could lead to processes getting stuck for
5995 6. Some uncommon configurations may cause a lookup to happen in a queue runner
5996 process, before it forks any delivery processes. The open lookup caching
5997 mechanism meant that the open file or database connection was passed into
5998 the delivery process. The problem was that delivery processes always tidy
5999 up cached lookup data. This could cause a problem for the next delivery
6000 process started by the queue runner, because the external queue runner
6001 process does not know about the closure. So the next delivery process
6002 still has data in the lookup cache. In the case of a file lookup, there was
6003 no problem because closing a file descriptor in a subprocess doesn't affect
6004 the parent. However, if the lookup was caching a connection to a database,
6005 the connection was closed, and the second delivery process was likely to
6006 see errors such as "PGSQL: query failed: server closed the connection
6007 unexpectedly". The problem has been fixed by closing all cached lookups
6008 in a queue runner before running a delivery process.
6010 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
6011 seem to have the "const" qualifier which it has on other OS. I've
6014 8. Change 4.21/2 was too strict. It is only if there are two authenticators
6015 *of the same type* (client or server) with the same public name that an
6016 error should be diagnosed.
6018 9. When Exim looked up a host name for an IP address, but failed to find the
6019 original IP address when looking up the host name (a safety check), it
6020 output the message "<ip address> does not match any IP for NULL", which was
6021 confusing, to say the least. The bug was that the host name should have
6022 appeared instead of "NULL".
6024 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
6025 user that is built into the binary, and the -C or -D options is used, root
6026 privilege is dropped before the configuration file is read. In addition,
6027 logging is switched to stderr instead of the normal log files. If the
6028 configuration then re-defines the Exim user, the unprivileged environment
6029 is probably not what is expected, so Exim logs a panic warning message (but
6032 However, if deliver_drop_privilege is set, the unprivileged state may well
6033 be exactly what is intended, so the warning has been cut out in that case,
6034 and Exim is allowed to try to write to its normal log files.
6040 1. smtp_return_error_details was not giving details for temporary sender
6041 or receiver verification errors.
6043 2. Diagnose a configuration error if two authenticators have the same public
6046 3. Exim used not to create the message log file for a message until the first
6047 delivery attempt. This could be confusing when incoming messages were held
6048 for policy or load reasons. The message log file is now created at the time
6049 the message is received, and an initial "Received" line is written to it.
6051 4. The automatically generated man page for command line options had a minor
6052 bug that caused no ill effects; however, a more serious problem was that
6053 the procedure for building the man page automatically didn't always
6054 operate. Consequently, release 4.20 contains an out-of-date version. This
6055 shouldn't happen again.
6057 5. When building Exim with embedded Perl support, the script that builds the
6058 Makefile was calling 'perl' to find its compile-time parameters, ignoring
6059 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
6061 6. The freeze_tell option was not being used for messages that were frozen on
6062 arrival, either by an ACL or by local_scan().
6064 7. Added the smtp_incomplete_transaction log selector.
6066 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
6067 was accepting AUTH without a new EHLO.
6069 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
6070 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
6073 10. Logging of TCP/IP connections (when configured) now happens in the main
6074 daemon process instead of the child process, so that the TCP/IP connection
6075 count is more accurate (but it can never be perfect).
6077 11. The use of "drop" in a nested ACL was not being handled correctly in the
6078 outer ACL. Now, if condition failure induced by the nested "drop" causes
6079 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
6080 or "require"), the connection is dropped.
6082 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
6083 that yield "discard" can now be used with an "accept" or a "discard" verb,
6084 but an error is generated for any others (because I can't see a useful way
6085 to define what should happen).
6087 13. When an ACL is read dynamically from a file (or anywhere else), the lines
6088 are now processed in the same way as lines in the Exim configuration file.
6089 In particular, continuation lines are supported.
6091 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
6093 15. Added -ti meaning -t -i.
6095 16. Check for letters, digits, hyphens, and dots in the names of dnslist
6096 domains, and warn by logging if others are found.
6098 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
6099 returned from GIFCONF when Exim is trying to find the list of interfaces on
6100 a host. The code in os.c has been modified to copy each ifreq to an aligned
6101 structure in all cases.
6103 Also, in some cases, the returned ifreq's were being copied to a 'struct
6104 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
6105 means the last couple of bytes of an IPv6 address could be chopped if the
6106 ifreq contained only a normal sockaddr (14 bytes storage).
6108 18. Named domain lists were not supported in the hosts_treat_as_local option.
6109 An entry such as +xxxx was not recognized, and was treated as a literal
6112 19. Ensure that header lines added by a DATA ACL are included in the reject log
6113 if the ACL subsequently rejects the message.
6115 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
6116 MD5 (which is deprecated).
6118 21. When testing a filter file using -bf, Exim was writing a message when it
6119 took the sender from a "From " line in the message, but it was not doing so
6120 when it took $return_path from a Return-Path: header line. It now does.
6122 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
6123 with a valid header line field name (a series of printing characters
6124 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
6126 23. Changed "disc" in the source to "disk" to conform to the documentation and
6127 the book and for uniformity.
6129 24. Ignore Sendmail's -Ooption=value command line item.
6131 25. When execve() failed while trying to run a command in a pipe transport,
6132 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
6133 could be confused with a return value of 69 from the command itself. This
6134 has been changed to 127, the value the shell returns if it is asked to run
6135 a non-existent command. The wording for the related log line suggests a
6136 non-existent command as the problem.
6138 26. If received_header_text expands to an empty string, do not add a Received:
6139 header line to the message. (Well, it adds a token one on the spool, but
6140 marks it "old" so that it doesn't get used or transmitted.)
6142 27. Installed eximstats 1.28 (addition of -nt option).
6144 28. There was no check for failure on the call to getsockname() in the daemon
6145 code. This can fail if there is a shortage of resources on the system, with
6146 ENOMEM, for example. A temporary error is now given on failure.
6148 29. Contrary to the C standard, it seems that in some environments, the
6149 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
6150 program. Exim now does this explicitly; it affects the formatting of
6151 timestamps using strftime().
6153 30. If exiqsumm was given junk data, it threw up some uninitialized variable
6154 complaints. I've now initialized all the variables, to avoid this.
6156 32. Header lines added by a system filter were not being "seen" during
6157 transport-time rewrites.
6159 33. The info_callback() function passed to OpenSSL is set up with type void
6160 (*)(SSL *, int, int), as described somewhere. However, when calling the
6161 function (actually a macro) that sets it up, the type void(*)() is
6162 expected. I've put in a cast to prevent warnings from picky compilers.
6164 34. If a DNS black list lookup found a CNAME record, but there were no A
6165 records associated with the domain it pointed at, Exim crashed.
6167 35. If a DNS black list lookup returned more than one A record, Exim ignored
6168 all but the first. It now scans all returned addresses if a particular IP
6169 value is being sought. In this situation, the contents of the
6170 $dnslist_value variable are a list of all the addresses, separated by a
6173 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
6174 to accept a host name and all its aliases if the forward lookup for any of
6175 them yielded the IP address of the incoming connection. Now it accepts only
6176 those names whose forward lookup yields the correct IP address. Any other
6177 names are discarded. This closes a loophole whereby a rogue DNS
6178 administrator could create reverse DNS records to break through a
6179 wildcarded host restriction in an ACL.
6181 37. If a user filter or a system filter that ran in a subprocess used any of
6182 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
6183 the wrong values were passed to the pipe command ($thisaddress had the
6184 value of $0, $0 had the value of $1, etc). This bug was introduced by
6185 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
6187 38. Improved the line breaking for long SMTP error messages from ACLs.
6188 Previously, if there was no break point between 40 and 75 characters, Exim
6189 left the rest of the message alone. Two changes have been made: (a) I've
6190 reduced the minimum length to 35 characters; (b) if it can't find a break
6191 point between 35 and 75 characters, it looks ahead and uses the first one
6192 that it finds. This may give the occasional overlong line, but at least the
6193 remaining text gets split now.
6195 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
6196 file descriptors might be low, and that setting 1000 would always raise it.
6197 It turns out that in some environments, the limit is already over 1000 and
6198 that lowering it causes trouble. So now Exim takes care not to decrease it.
6200 40. When delivering a message, the value of $return_path is set to $sender_
6201 address at the start of routing (routers may change the value). By an
6202 oversight, this default was not being set up when an address was tested by
6203 -bt or -bv, which affected the outcome if any router or filter referred to
6206 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
6207 log only when "message" or "log_message" are set. However, if one of the
6208 conditions was an address verification, or a call to a nested ACL, the
6209 messages generated by the underlying test were being passed through. This
6210 no longer happens. The underlying message is available in $acl_verify_
6211 message for both "message" and "log_message" expansions, so it can be
6212 passed through if needed.
6214 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
6215 new expansion $bh_ to give the encoded byte string without charset
6216 translation. Translation happens only if iconv() is available; HAVE_ICONV
6217 indicates this at build time. HEADERS_CHARSET gives the charset to
6218 translate to; headers_charset can change it in the configuration, and
6219 "headers charset" can change it in an individual filter file.
6221 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
6222 that creates RFC 2047 encoded "words" labels them as that charset instead
6223 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
6224 expansion operator; (ii) when Exim creates a From: line for a local
6225 message; (iii) when a header line is rewritten to include a "phrase" part.
6227 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
6228 buggy, causing it to skip the first lines of messages whose message ID
6229 ended in 'D'. This would not have bitten before Exim release 4.14, because
6230 message IDs were unlikely to end in 'D' before then. The effect was to have
6231 incorrect size information for certain domains.
6233 45. #include "config.h" was missing at the start of the crypt16.c module. This
6234 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
6237 46. If there was a timeout during a "random" callout check, Exim treated it as
6238 a failure of the random address, and carried on sending RSET and the real
6239 address. If the delay was just some slowness somewhere, the response to the
6240 original RCPT would be taken as a response to RSET and so on, causing
6241 mayhem of various kinds.
6243 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
6244 when I implemented it. It didn't allow for the fact that some option values
6245 may legitimatetly be negative (e.g. size_addition), and it didn't even do
6246 the right test for positive values.
6248 48. Domain names in DNS records are case-independent. Exim always looks them up
6249 in lower case. Some resolvers return domain names in exactly the case they
6250 appear in the zone file, that is, they may contain uppercase letters. Not
6251 all resolvers do this - some return always lower case. Exim was treating a
6252 change of case by a resolver as a change of domain, similar to a widening
6253 of a domain abbreviation. This triggered its re-routing code and so it was
6254 trying to route what was effectively the same domain again. This normally
6255 caused routing to fail (because the router wouldn't handle the domain
6256 twice). Now Exim checks for this case specially, and just changes the
6257 casing of the domain that it ultimately uses when it transmits the message
6260 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
6263 50. If a filter generated a file delivery with a non-absolute name (possible if
6264 no home directory exists for the router), the forbid_file option was not
6267 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
6268 the existing equality matching.
6270 52. Exim was using ints instead of ino_t variables in some places where it was
6271 dealing with inode numbers.
6273 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
6274 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
6275 called TMPDIR, and if it finds it is different, it changes its value.
6277 54. The smtp_printf() function is now made available to local_scan() so
6278 additional output lines can be written before returning. There is also an
6279 smtp_fflush() function to enable the detection of a dropped connection.
6280 The variables smtp_input and smtp_batched_input are exported to
6283 55. Changed the default runtime configuration: the message "Unknown user"
6284 has been removed from the ACL, and instead placed on the localuser router,
6285 using the cannot_route_message feature. This means that any verification
6286 failures that generate their own messages won't get overridden. Similarly,
6287 the "Unrouteable address" message that was in the ACL for unverifiable
6288 relay addresses has also been removed.
6290 56. Added hosts_avoid_esmtp to the smtp transport.
6292 57. The exicyclog script was not checking for the esoteric option
6293 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
6294 will work only if exicyclog is run under the appropriate euid.
6296 58. Following a discussion on the list, the rules by which Exim recognises line
6297 endings on incoming messages have been changed. The -dropcr and drop_cr
6298 options are now no-ops, retained only for backwards compatibility. The
6299 following line terminators are recognized: LF CRLF CR. However, special
6300 processing applies to CR:
6302 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
6303 nor a local message in the state where . is a terminator.
6305 (ii) If a bare CR is encountered in a header line, an extra space is added
6306 after the line terminator so as not to end the header. The reasoning
6307 behind this is that bare CRs in header lines are most likely either
6308 to be mistakes, or people trying to play silly games.
6310 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
6311 was being incorrectly given as 18 bytes larger than it should have been.
6312 This is a VOB (very old bug).
6314 60. This may never have affected anything current, but just in case it has:
6315 When the local host is found other than at the start of a list of hosts,
6316 the local host, those with the same MX, and any that follow, are discarded.
6317 When the list in question was part of a longer list of hosts, the following
6318 hosts (not currently being processed) were also being discarded. This no
6319 longer happens. I'm not sure if this situation could ever has previously
6322 61. Added the "/MX" feature to lists of hosts in the manualroute and query
6325 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
6326 header. This is something that is recommended in a new Internet Draft, and
6327 is something that is documented as being done by Sendmail. There are two
6328 possible values. For messages generated by the autoreply transport, Exim
6331 Auto-Submitted: auto-replied
6333 whereas for all other generated messages (e.g. bounces) it adds
6335 Auto-Submitted: auto-generated
6337 63. The "personal" condition in filters now includes a test for the
6338 Auto-Submitted: header. If it contains the string "auto-" the message it
6339 not considered personal.
6341 64. Added rcpt_include_affixes as a generic transport option.
6343 65. Added queue_only_override (default true).
6345 66. Added the syslog_duplication option.
6347 67. If what should have been the first header line of a message consisted of
6348 a space followed by a colon, Exim was mis-interpreting it as a header line.
6349 It isn't of course - it is syntactically invalid and should therefore be
6350 treated as the start of the message body. The misbehaviour could have
6351 caused a number of strange effects, including loss of data in subsequent
6352 header lines, and spool format errors.
6354 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
6355 client host had authenticated. This control can now be exercised by an ACL
6356 for more flexibility.
6358 69. By default, callouts do not happen when testing with -bh. There is now a
6359 variant, -bhc, which does actually run the callout code, including
6360 consulting and updating the callout cache.
6362 70. Added support for saslauthd authentication, courtesy of Alexander
6365 71. If statvfs() failed on the spool or log directories while checking their
6366 size for availability, Exim confusingly gave the error "space shortage".
6367 Furthermore, in debugging mode it crashed with a floating point exception.
6368 These checks are done if check_{spool,log}_{space,inodes} are set, and when
6369 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
6370 serious problem, Exim now writes to the main and panic logs when this
6371 happens, with details of the failure. It then refuses to accept the
6372 incoming message, giving the message "spool directory problem" or "log
6373 directory problem" with a 421 code for SMTP messages.
6375 72. When Exim is about to re-exec itself, it ensures that the file descriptors
6376 0, 1, and 2 exist, because some OS complain for execs without them (see
6377 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
6378 descriptors. However, the code omitted to check that the open succeeded,
6379 causing mysterious errors if for some reason the permissions on /dev/null
6380 got screwed. Now Exim writes a message to the main and panic logs, and
6381 bombs out if it can't open /dev/null.
6383 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
6384 interact so that it is all more flexible. It is supposed to remain
6385 backwards compatible. Also added extra_local_interfaces.
6387 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
6388 to bomb out with an assertion failure - to the client this appears as a
6389 connection drop. This problem occurs in the part of the code that was taken
6390 from the Samba project. Fortunately, the assertion is in a very simple
6391 function, so I have fixed this by reproducing the function inline in the
6392 one place where it is called, and arranging for authentication to fail
6393 instead of killing the process with assert().
6395 75. The SPA client code was not working when the server requested OEM rather
6396 than Unicode encoding.
6398 76. Added code to make require_files with a specific uid setting more usable in
6399 the case where statting the file as root fails - usually a non-root-mounted
6400 NFS file system. When this happens and the failure is EACCES, Exim now
6401 forks a subprocess and does the per-uid checking as the relevant uid.
6403 77. Added process_log_path.
6405 78. If log_file_path was not explicitly set, a setting of check_log_space or
6406 check_log_inodes was ignored.
6408 79. If a space check for the spool or log partitions fails, the incident is now
6409 logged. Of course, in the latter case the data may get lost...
6411 80. Added the %p formatting code to string_format() so that it can be used to
6412 print addresses in debug_print(). Adjusted all the address printing in the
6413 debugging in store.c to use %p rather than %d.
6415 81. There was a concern that a line of code in smtp_in.c could overflow a
6416 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
6417 initially expressed, the concern was not well-founded, because trailing
6418 spaces are removed early. However, if the trailing spaces were followed by
6419 a NULL, they did not get removed, so the overflow was possible. Two fixes
6422 (a) I re-wrote the offending code in a cleaner fashion.
6423 (b) If an incoming SMTP command contains a NULL character, it is rejected
6426 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
6427 runs initgroups(), so that if the Exim user is in any additional groups,
6428 they will be used during message reception.
6434 The change log for 4.20 and earlier releases has been archived.
projects / exim.git / blob