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.
110 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
113 JH/02 The smtp transport option "multi_domain" is now expanded.
115 JH/03 The smtp transport now requests PRDR by default, if the server offers
118 JH/04 Certificate name checking on server certificates, when exim is a client,
119 is now done by default. The transport option tls_verify_cert_hostnames
120 can be used to disable this per-host. The build option
121 EXPERIMENTAL_CERTNAMES is withdrawn.
123 JH/05 The value of the tls_verify_certificates smtp transport and main options
124 default to the word "system" to access the system default CA bundle.
125 For GnuTLS, only version 3.0.20 or later.
127 JH/06 Verification of the server certificate for a TLS connection is now tried
128 (but not required) by default. The verification status is now logged by
129 default, for both outbound TLS and client-certificate supplying inbound
132 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
135 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
136 Status Notification (bounce) messages are now MIME format per RFC 3464.
137 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
138 under the control of the dsn_advertise_hosts option, and routers may
139 have a dsn_lasthop option.
141 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
142 default, modifiable by a malware= option. The list separator for
143 the options can now be changed in the usual way. Bug 68.
145 JH/10 The smtp_receive_timeout main option is now expanded before use.
147 JH/11 The incoming_interface log option now also enables logging of the
148 local interface on delivery outgoing connections.
150 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
151 if the interface and destination host and port all match.
153 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
156 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
157 Patch from Andrew Lewis.
159 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
160 now supports optional time-restrictions, weighting, and priority
161 modifiers per server. Patch originally by <rommer@active.by>.
163 JH/16 The spamd_address main option now supports a mixed list of local
164 and remote servers. Remote servers can be IPv6 addresses, and
165 specify a port-range.
167 JH/17 Bug 68: The spamd_address main option now supports an optional
168 timeout value per server.
170 JH/18 Bug 1581: Router and transport options headers_add/remove can
171 now have the list separator specified.
173 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
176 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
179 JH/21 Support for the A6 type of dns record is withdrawn.
181 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
182 rather than the verbs used.
184 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
185 from 255 to 1024 chars.
187 JH/24 Verification callouts now attempt to use TLS by default.
189 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
190 are generic router options now. The defaults didn't change.
192 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
193 Original patch from Alexander Shikoff, worked over by JH.
195 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
196 files if ZCAT_COMMAND is not executable.
198 JH/26 Bug 1539: Add timout/retry options on dnsdb lookups.
200 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
202 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
203 Normally benign, it bites when the pair was led to by a CNAME;
204 modern usage is to not canoicalize the domain to a CNAME target
205 (and we were inconsistent anyway for A-only vs AAAA+A).
207 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
209 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
210 when evaluating $sender_host_dnssec.
212 JH/31 Check the HELO verification lookup for DNSSEC, adding new
213 $sender_helo_dnssec variable.
215 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
217 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
219 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
221 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
222 documented as working, but never had. Support all but $spam_report.
224 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
225 added for tls authenticator.
230 TL/01 When running the test suite, the README says that variables such as
231 no_msglog_check are global and can be placed anywhere in a specific
232 test's script, however it was observed that placement needed to be near
233 the beginning for it to behave that way. Changed the runtest perl
234 script to read through the entire script once to detect and set these
235 variables, reset to the beginning of the script, and then run through
236 the script parsing/test process like normal.
238 TL/02 The BSD's have an arc4random API. One of the functions to induce
239 adding randomness was arc4random_stir(), but it has been removed in
240 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
241 function when detected.
243 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
244 cause callback expansion.
246 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
247 syntax errors in an expansion can be treated as a string instead of
248 logging or causing an error, due to the internal use of bool_lax
249 instead of bool when processing it.
251 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
252 server certificates when making smtp deliveries.
254 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
256 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
258 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
260 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
261 Merged patch from Sebastian Wiedenroth.
263 JH/05 Fix results-pipe from transport process. Several recipients, combined
264 with certificate use, exposed issues where response data items split
265 over buffer boundaries were not parsed properly. This eventually
266 resulted in duplicates being sent. This issue only became common enough
267 to notice due to the introduction of conection certificate information,
268 the item size being so much larger. Found and fixed by Wolfgang Breyha.
270 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
271 size buffer was used, resulting in syntax errors when an expansion
274 JH/07 Add support for directories of certificates when compiled with a GnuTLS
275 version 3.3.6 or later.
277 JH/08 Rename the TPDA expermimental facility to Event Actions. The #ifdef
278 is EXPERIMENTAL_EVENT, the main-configuration and transport options
279 both become "event_action", the variables become $event_name, $event_data
280 and $event_defer_errno. There is a new variable $verify_mode, usable in
281 routers, transports and related events. The tls:cert event is now also
282 raised for inbound connections, if the main configuration event_action
285 TL/06 In test suite, disable OCSP for old versions of openssl which contained
286 early OCSP support, but no stapling (appears to be less than 1.0.0).
288 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
289 server certificate names available under the smtp transport option
290 "tls_verify_cert_hostname" now do not permit multi-component wildcard
293 JH/10 Time-related extraction expansions from certificates now use the main
294 option "timezone" setting for output formatting, and are consistent
295 between OpenSSL and GnuTLS compilations. Bug 1541.
297 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
298 encoded parameter in the incoming message. Bug 1558.
300 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
301 include certificate info, eximon was claiming there were spoolfile
304 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
306 JH/14 Log delivery-related information more consistently, using the sequence
307 "H=<name> [<ip>]" wherever possible.
309 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
310 are problematic for Debian distribution, omit them from the release
313 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
315 JH/16 Fix string representation of time values on 64bit time_t anchitectures.
318 JH/17 Fix a null-indirection in certextract expansions when a nondefault
319 output list separator was used.
324 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
325 checkers that were complaining about end of non-void function with no
328 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
329 This was a regression intruduced in 4.83 by another bugfix.
331 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
333 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
334 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
340 TF/01 Correctly close the server side of TLS when forking for delivery.
342 When a message was received over SMTP with TLS, Exim failed to clear up
343 the incoming connection properly after forking off the child process to
344 deliver the message. In some situations the subsequent outgoing
345 delivery connection happened to have the same fd number as the incoming
346 connection previously had. Exim would try to use TLS and fail, logging
347 a "Bad file descriptor" error.
349 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
350 utilities have not been installed.
352 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
353 temporary space as the ACL may create new global variables.
355 TL/01 LDAP support uses per connection or global context settings, depending
356 upon the detected version of the libraries at build time.
358 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
359 to extract and use the src ip:port in logging and expansions as if it
360 were a direct connection from the outside internet. PPv2 support was
361 updated based on HAProxy spec change in May 2014.
363 JH/02 Add ${listextract {number}{list}{success}{fail}}.
365 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
366 Properly escape header and check for NULL return.
368 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
371 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
373 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
374 characters in header names, implemented as a verify condition.
375 Contributed by Michael Fischer v. Mollard.
377 TL/05 Rename SPF condition results err_perm and err_temp to standardized
378 results permerror and temperror. Previous values are deprecated but
379 still accepted. In a future release, err_perm and err_temp will be
380 completely removed, which will be a backward incompatibility if the
381 ACL tests for either of these two old results. Patch contributed by
382 user bes-internal on the mailing list.
384 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
386 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
387 selectors, in both main and reject logs.
389 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
392 JH/07 Add malware type "sock" for talking to simple daemon.
394 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
396 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
397 routers/transports under cutthrough routing.
399 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
400 numbers. Touch up "bool" conditional to keep the same definition.
402 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
404 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
406 JH/12 Expand items in router/transport headers_add or headers_remove lists
407 individually rather than the list as a whole. Bug 1452.
409 Required for reasonable handling of multiple headers_ options when
410 they may be empty; requires that headers_remove items with embedded
411 colons must have them doubled (or the list-separator changed).
413 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
414 view the policy declared in the DMARC record. Currently, $dmarc_status
415 is a combined value of both the record presence and the result of the
418 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
420 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
421 dnslookup router and the smtp transport (applying to the forward
424 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
425 of ldap servers used for a specific lookup. Patch provided by Heiko
428 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
429 New variable $lookup_dnssec_authenticated for observability.
431 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
432 Patch submitted by Lars Timman.
434 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
436 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
437 Requires trusted mode and valid format message id, aborts otherwise.
438 Patch contributed by Heiko Schlichting.
440 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
441 certextract with support for various fields. Bug 1358.
443 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
444 is requested by default, modifiable by smtp transport option
447 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
448 operate on certificate variables to give certificate fingerprints
449 Also new ${sha256:cert_variable}.
451 JH/23 The PRDR feature is moved from being Experimental into the mainline.
453 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
454 Christian Aistleitner.
456 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
458 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
459 file. Patch from Wolfgang Breyha.
461 JH/25 Expand the coverage of the delivery $host and $host_address to
462 client authenticators run in verify callout. Bug 1476.
464 JH/26 Port service names are now accepted for tls_on_connect_ports, to
465 align with daemon_smtp_ports. Bug 72.
467 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
468 support and error reporting did not work properly.
470 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
471 and is readable. Patch from Andrew Colin Kissa.
473 TL/14 Enhance documentation of ${run expansion and how it parses the
474 commandline after expansion, particularly in the case when an
475 unquoted variable expansion results in an empty value.
477 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
479 PP/02 Fix internal collision of T_APL on systems which support RFC3123
480 by renaming away from it. Addresses GH issue 15, reported by
483 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
485 TL/15 SECURITY: prevent double expansion in math comparison functions
486 (can expand unsanitized data). Not remotely exploitable.
493 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
495 PP/02 Make -n do something, by making it not do something.
496 When combined with -bP, the name of an option is not output.
498 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
501 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
502 $sender_host_name and config options to manage this, and basic check
505 PP/05 DSCP support for outbound connections and control modifier for inbound.
507 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
508 (Only plugin which currently uses this is kerberos4, which nobody should
509 be using, but we should make it available and other future plugins might
510 conceivably use it, even though it would break NAT; stuff *should* be
511 using channel bindings instead).
513 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
514 name; added for Sendmail compatibility; requires admin caller.
515 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
516 just ignore it); requires trusted caller.
517 Also parse but ignore: -Ac -Am -X<logfile>
520 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
522 TL/02 Add +smtp_confirmation as a default logging option.
524 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
525 Patch by Magnus Holmgren from 2007-02-20.
527 TL/04 Bugzilla 1281 - Spec typo.
528 Bugzilla 1283 - Spec typo.
529 Bugzilla 1290 - Spec grammar fixes.
531 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
533 TL/06 Add Experimental DMARC support using libopendmarc libraries.
535 TL/07 Fix an out of order global option causing a segfault. Reported to dev
536 mailing list by by Dmitry Isaikin.
538 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
540 JH/02 Support "G" suffix to numbers in ${if comparisons.
542 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
544 NM/01 Bugzilla 1197 - Spec typo
545 Bugzilla 1196 - Spec examples corrections
547 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
549 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
550 gnutls_enable_pkcs11, but renamed to more accurately indicate its
553 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
554 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
556 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
557 "acl {{name}{arg}...}", and optional args on acl condition
560 JH/05 Permit multiple router/transport headers_add/remove lines.
562 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
564 JH/07 Avoid using a waiting database for a single-message-only transport.
565 Performance patch from Paul Fisher. Bugzilla 1262.
567 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
570 JH/09 Add $headers_added variable, with content from use of ACL modifier
571 add_header (but not yet added to the message). Bugzilla 199.
573 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
574 Pulled from Bugzilla 817 by Wolfgang Breyha.
576 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
578 (nb: this is the same fix as in Exim 4.80.1)
580 JH/11 Add A= logging on delivery lines, and a client_set_id option on
583 JH/12 Add optional authenticated_sender logging to A= and a log_selector
586 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
588 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
589 advertise SMTP AUTH mechanism to us, instead of a generic
590 protocol violation error. Also, make Exim more robust to bad
591 data from the Dovecot auth socket.
593 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
595 When a queue runner is handling a message, Exim first routes the
596 recipient addresses, during which it prunes them based on the retry
597 hints database. After that it attempts to deliver the message to
598 any remaining recipients. It then updates the hints database using
601 So if a recipient address works intermittently, it can get repeatedly
602 deferred at routing time. The retry hints record remains fresh so the
603 address never reaches the final cutoff time.
605 This is a fairly common occurrence when a user is bumping up against
606 their storage quota. Exim had some logic in its local delivery code
607 to deal with this. However it did not apply to per-recipient defers
608 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
610 This change adds a proper retry rule check during routing so that the
611 final cutoff time is checked against the message's age. We only do
612 this check if there is an address retry record and there is not a
613 domain retry record; this implies that previous attempts to handle
614 the address had the retry_use_local_parts option turned on. We use
615 this as an approximation for the destination being like a local
616 delivery, as in LMTP.
618 I suspect this new check makes the old local delivery cutoff check
619 redundant, but I have not verified this so I left the code in place.
621 TF/02 Correct gecos expansion when From: is a prefix of the username.
623 Test 0254 submits a message to Exim with the header
627 When I ran the test suite under the user fanf2, Exim expanded
628 the header to contain my full name, whereas it should have added
629 a Resent-Sender: header. It erroneously treats any prefix of the
630 username as equal to the username.
632 This change corrects that bug.
634 GF/01 DCC debug and logging tidyup
635 Error conditions log to paniclog rather than rejectlog.
636 Debug lines prefixed by "DCC: " to remove any ambiguity.
638 TF/03 Avoid unnecessary rebuilds of lookup-related code.
640 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
641 Bug spotted by Jeremy Harris; was flawed since initial commit.
642 Would have resulted in OCSP responses post-SNI triggering an Exim
643 NULL dereference and crash.
645 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
647 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
648 Bug detection, analysis and fix by Samuel Thibault.
649 Bugzilla 1331, Debian bug #698092.
651 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
653 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
654 Server implementation by Todd Lyons, client by JH.
655 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
656 config variable "prdr_enable" controls whether the server
657 advertises the facility. If the client requests PRDR a new
658 acl_data_smtp_prdr ACL is called once for each recipient, after
659 the body content is received and before the acl_smtp_data ACL.
660 The client is controlled by bolth of: a hosts_try_prdr option
661 on the smtp transport, and the server advertisement.
662 Default client logging of deliveries and rejections involving
663 PRDR are flagged with the string "PRDR".
665 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
666 fclose(). Diagnosis by Todd Lyons.
668 PP/17 Update configure.default to handle IPv6 localhost better.
669 Patch by Alain Williams (plus minor tweaks).
672 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
673 This is now consistent with GnuTLS, and is now documented: the
674 previous undocumented portable approach to treating the option as
675 unset was to force an expansion failure. That still works, and
676 an empty string is now equivalent.
678 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
679 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
680 not performing validation itself.
682 PP/20 Added force_command boolean option to pipe transport.
683 Patch from Nick Koston, of cPanel Inc.
685 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
688 TF/04 Added udpsend ACL modifer and hexquote expansion operator
690 PP/21 Fix eximon continuous updating with timestamped log-files.
691 Broken in a format-string cleanup in 4.80, missed when I repaired the
692 other false fix of the same issue.
693 Report and fix from Heiko Schlichting.
696 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
697 Report from Prashanth Katuri.
699 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
700 It's SecureTransport, so affects any MacOS clients which use the
701 system-integrated TLS libraries, including email clients.
703 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
704 using a MIME ACL for non-SMTP local injection.
705 Report and assistance in diagnosis by Warren Baker.
707 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
709 JH/16 Fix comparisons for 64b. Bugzilla 1385.
711 TL/09 Add expansion variable $authenticated_fail_id to keep track of
712 last id that failed so it may be referenced in subsequent ACL's.
714 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
717 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
718 ldap library initialization, allowing self-signed CA's to be
719 used. Also properly sets require_cert option later in code by
720 using NULL (global ldap config) instead of ldap handle (per
721 session). Bug diagnosis and testing by alxgomz.
723 TL/12 Enhanced documentation in the ratelimit.pl script provided in
724 the src/util/ subdirectory.
726 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
727 renamed to Transport Post Delivery Action by Jeremy Harris, as
730 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
731 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
732 redis_servers = needs to be configured which will be used by the redis
733 lookup. Patch from Warren Baker, of The Packet Hub.
735 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
737 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
738 hostname or reverse DNS when processing a host list. Used suggestions
739 from multiple comments on this bug.
741 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
743 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
744 Missed a few lines, added it to make the runtest require no keyboard
747 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
748 contains upper case chars. Make router use caseful_local_part.
750 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
751 support when GnuTLS has been built with p11-kit.
757 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
759 This, or similar/improved, will also be change PP/11 of 4.82.
765 PP/01 Handle short writes when writing local log-files.
766 In practice, only affects FreeBSD (8 onwards).
767 Bugzilla 1053, with thanks to Dmitry Isaikin.
769 NM/01 Bugzilla 949 - Documentation tweak
771 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
774 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
776 PP/02 Implemented gsasl authenticator.
778 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
780 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
781 `pkg-config foo` for cflags/libs.
783 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
784 with rest of GSASL and with heimdal_gssapi.
786 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
787 `pkg-config foo` for cflags/libs for the TLS implementation.
789 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
790 properties get this fed in as external SSF. A number of robustness
791 and debugging improvements to the cyrus_sasl authenticator.
793 PP/08 cyrus_sasl server now expands the server_realm option.
795 PP/09 Bugzilla 1214 - Log authentication information in reject log.
796 Patch by Jeremy Harris.
798 PP/10 Added dbmjz lookup type.
800 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
802 PP/12 MAIL args handles TAB as well as SP, for better interop with
803 non-compliant senders.
804 Analysis and variant patch by Todd Lyons.
806 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
807 Bug report from Lars Müller <lars@samba.org> (via SUSE),
808 Patch from Dirk Mueller <dmueller@suse.com>
810 PP/13 tls_peerdn now print-escaped for spool files.
811 Observed some $tls_peerdn in wild which contained \n, which resulted
812 in spool file corruption.
814 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
815 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
816 or write after TLS renegotiation, which otherwise led to messages
819 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
820 as a tracking header (ie: a signed header comes before the signature).
821 Patch from Wolfgang Breyha.
823 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
824 comma-sep list; embedded commas doubled.
826 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
828 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
830 Report and patch from Dmitry Banschikov.
832 PP/16 Removed "dont_insert_empty_fragments" fron "openssl_options".
833 Removed SSL_clear() after SSL_new() which led to protocol negotiation
834 failures. We appear to now support TLS1.1+ with Exim.
836 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
837 lets Exim select keys and certificates based upon TLS SNI from client.
838 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
839 before an outbound SMTP session. New log_selector, +tls_sni.
841 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
842 NULL dereference. Report and patch from Alun Jones.
844 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
845 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
846 Not seeing resolver debug output on NetBSD, but suspect this is a
847 resolver implementation change.
849 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
850 Left warnings. Added "eximon gdb" invocation mode.
852 PP/21 Defaulting "accept_8bitmime" to true, not false.
854 PP/22 Added -bw for inetd wait mode support.
856 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
857 locate the relevant includes and libraries. Made this the default.
859 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
860 Bugzilla 1246, report and most of solution from Tomasz Kusy.
862 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
863 This may cause build issues on older platforms.
865 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
866 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
867 gnutls_require_mac & gnutls_require_protocols (no longer supported).
868 Added SNI support via GnuTLS too.
869 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
871 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
873 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
874 Applied second patch from Janne, implementing suggestion to default
875 multiple-strings-in-record handling to match SPF spec.
877 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
879 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
880 read-only, out of scope).
881 Patch from Wolfgang Breyha, report from Stuart Northfield.
883 PP/29 Fix three issues highlighted by clang analyser static analysis.
884 Only crash-plausible issue would require the Cambridge-specific
885 iplookup router and a misconfiguration.
886 Report from Marcin Mirosław.
888 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
890 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
891 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
892 As part of this, removing so much warning spew let me fix some minor
893 real issues in debug logging.
895 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
896 assignment on my part. Fixed.
898 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
899 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
900 Janne Snabb (who went above and beyond: thank you).
902 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
903 string otherwise requires a connection and a bunch more work and it's
904 relatively easy to get wrong. Should also expose TLS library linkage
907 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
908 64-bit ${eval} (JH/03).
910 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
911 GNU libc to support some of the 64-bit stuff, should not lead to
912 conflicts. Defined before os.h is pulled in, so if a given platform
913 needs to override this, it can.
915 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
916 protection layer was required, which is not implemented.
917 Bugzilla 1254, patch from Wolfgang Breyha.
919 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
920 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
921 tls_dhparam take prime identifiers. Also unbreak combination of
922 OpenSSL+DH_params+TLSSNI.
924 PP/39 Disable SSLv2 by default in OpenSSL support.
930 PP/01 Solaris build fix for Oracle's LDAP libraries.
931 Bugzilla 1109, patch from Stephen Usher.
933 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
935 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
938 TF/02 Fix a couple more cases where we did not log the error message
939 when unlink() failed. See also change 4.74-TF/03.
941 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
942 lock up or crash if it happened to be inside a call to libc when it
943 got a SIGUSR1 from exiwhat.
945 The SIGUSR1 handler appends the current process status to the process
946 log which is later printed by exiwhat. It used to use the general
947 purpose logging code to do this, but several functions it calls are
948 not safe for signals.
950 The new output code in the SIGUSR1 handler is specific to the process
951 log, and simple enough that it's easy to inspect for signal safety.
952 Removing some special cases also simplifies the general logging code.
953 Removing the spurious timestamps from the process log simplifies
956 TF/04 Improved ratelimit ACL condition.
958 The /noupdate option has been deprecated in favour of /readonly which
959 has clearer semantics. The /leaky, /strict, and /readonly update modes
960 are mutually exclusive. The update mode is no longer included in the
961 database key; it just determines when the database is updated. (This
962 means that when you upgrde Exim will forget old rate measurements.)
964 Exim now checks that the per_* options are used with an update mode that
965 makes sense for the current ACL. For example, when Exim is processing a
966 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
967 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
968 must specify per_mail/readonly. If you omit the update mode it defaults to
969 /leaky where that makes sense (as before) or /readonly where required.
971 The /noupdate option is now undocumented but still supported for
972 backwards compatibility. It is equivalent to /readonly except that in
973 ACLs where /readonly is required you may specify /leaky/noupdate or
974 /strict/noupdate which are treated the same as /readonly.
976 A useful new feature is the /count= option. This is a generalization
977 of the per_byte option, so that you can measure the throughput of other
978 aggregate values. For example, the per_byte option is now equivalent
979 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
981 The per_rcpt option has been generalized using the /count= mechanism
982 (though it's more complicated than the per_byte equivalence). When it is
983 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
984 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
985 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
986 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
987 Note that using per_rcpt with a non-readonly update mode in more than
988 one ACL will cause the recipients to be double-counted. (The per_mail
989 and per_byte options don't have this problem.)
991 The handling of very low rates has changed slightly. If the computed rate
992 is less than the event's count (usually one) then this event is the first
993 after a long gap. In this case the rate is set to the same as this event's
994 count, so that the first message of a spam run is counted properly.
996 The major new feature is a mechanism for counting the rate of unique
997 events. The new per_addr option counts the number of different
998 recipients that someone has sent messages to in the last time period. It
999 behaves like per_rcpt if all the recipient addresses are different, but
1000 duplicate recipient addresses do not increase the measured rate. Like
1001 the /count= option this is a general mechanism, so the per_addr option
1002 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
1003 example, measure the rate that a client uses different sender addresses
1004 with the options per_mail/unique=$sender_address. There are further
1005 details in the main documentation.
1007 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
1009 TF/06 Removed a few PCRE remnants.
1011 TF/07 Automatically extract Exim's version number from tags in the git
1012 repository when doing development or release builds.
1014 PP/02 Raise smtp_cmd_buffer_size to 16kB.
1015 Bugzilla 879. Patch from Paul Fisher.
1017 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
1018 Heavily based on revision 40f9a89a from Simon Arlott's tree.
1021 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
1023 PP/05 Variable $av_failed, true if the AV scanner deferred.
1024 Bugzilla 1078. Patch from John Horne.
1026 PP/06 Stop make process more reliably on build failure.
1027 Bugzilla 1087. Patch from Heiko Schlittermann.
1029 PP/07 Make maildir_use_size_file an _expandable_ boolean.
1030 Bugzilla 1089. Patch from Heiko Schlittermann.
1032 PP/08 Handle ${run} returning more data than OS pipe buffer size.
1033 Bugzilla 1131. Patch from Holger Weiß.
1035 PP/09 Handle IPv6 addresses with SPF.
1036 Bugzilla 860. Patch from Wolfgang Breyha.
1038 PP/10 GnuTLS: support TLS 1.2 & 1.1.
1040 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
1043 PP/11 match_* no longer expand right-hand-side by default.
1044 New compile-time build option, EXPAND_LISTMATCH_RHS.
1045 New expansion conditions, "inlist", "inlisti".
1047 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
1049 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
1051 PP/14 fix log_write() format string regression from TF/03.
1052 Bugzilla 1152. Patch from Dmitry Isaikin.
1058 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
1060 PP/02 Harmonised TLS library version reporting; only show if debugging.
1061 Layout now matches that introduced for other libraries in 4.74 PP/03.
1063 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
1065 PP/04 New "dns_use_edns0" global option.
1067 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
1070 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
1071 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
1073 TK/01 Updated PolarSSL code to 0.14.2.
1074 Bugzilla 1097. Patch from Andreas Metzler.
1076 PP/07 Catch divide-by-zero in ${eval:...}.
1077 Fixes bugzilla 1102.
1079 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
1082 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
1083 format-string attack -- SECURITY: remote arbitrary code execution.
1085 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
1086 time unintentionally subject to list matching rules, letting the header
1087 cause arbitrary Exim lookups (of items which can occur in lists, *not*
1088 arbitrary string expansion). This allowed for information disclosure.
1090 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
1091 INT_MIN/-1 -- value coerced to INT_MAX.
1097 NM/01 Workround for PCRE version dependency in version reporting
1100 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
1101 This fixes portability to compilers other than gcc, notably
1102 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
1104 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
1105 makefiles for portability to HP-UX and POSIX correctness.
1107 PP/01 Permit LOOKUP_foo enabling on the make command-line.
1108 Also via indented variable definition in the Makefile.
1109 (Debugging by Oliver Heesakkers).
1111 PP/02 Restore caching of spamd results with expanded spamd_address.
1112 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
1114 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
1115 Improves build reliability. Fix from: Frank Elsner
1117 NM/02 Fix wide character breakage in the rfc2047 coding
1118 Fixes bug 1064. Patch from Andrey N. Oktyabrski
1120 NM/03 Allow underscore in dnslist lookups
1121 Fixes bug 1026. Patch from Graeme Fowler
1123 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
1124 Code patches from Adam Ciarcinski of NetBSD.
1126 NM/04 Fixed exiqgrep to cope with mailq missing size issue
1129 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
1130 is logged, to avoid truncation. Patch from John Horne.
1132 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
1133 Patch from Jakob Hirsch.
1135 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
1136 SQL string expansion failure details.
1137 Patch from Andrey Oktyabrski.
1139 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
1140 Patch from Simon Arlott.
1142 PP/09 New lookups functionality failed to compile on old gcc which rejects
1143 extern declarations in function scope.
1144 Patch from Oliver Fleischmann
1146 PP/10 Use sig_atomic_t for flags set from signal handlers.
1147 Check getgroups() return and improve debugging.
1148 Fixed developed for diagnosis in bug 927 (which turned out to be
1151 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
1152 Patch from Mark Zealey.
1154 PP/12 Bugzilla 1056: Improved spamd server selection.
1155 Patch from Mark Zealey.
1157 PP/13 Bugzilla 1086: Deal with maildir quota file races.
1158 Based on patch from Heiko Schlittermann.
1160 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
1161 Patch from Uwe Doering, sign-off by Michael Haardt.
1163 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
1164 variable declaration deep within a block. Bug and patch from
1167 PP/15 lookups-Makefile IRIX compatibilty coercion.
1169 PP/16 Make DISABLE_DKIM build knob functional.
1171 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
1172 Patch by Simon Arlott
1174 TF/03 Fix valgrind.h portability to C89 compilers that do not support
1175 variable argument macros. Our copy now differs from upstream.
1181 TF/01 Failure to get a lock on a hints database can have serious
1182 consequences so log it to the panic log.
1184 TF/02 Log LMTP confirmation messages in the same way as SMTP,
1185 controlled using the smtp_confirmation log selector.
1187 TF/03 Include the error message when we fail to unlink a spool file.
1189 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
1190 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
1191 for maintaining out-of-tree patches for some time.
1193 PP/01 Bugzilla 139: Documentation and portability issues.
1194 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
1195 Handle per-OS dynamic-module compilation flags.
1197 PP/02 Let /dev/null have normal permissions.
1198 The 4.73 fixes were a little too stringent and complained about the
1199 permissions on /dev/null. Exempt it from some checks.
1200 Reported by Andreas M. Kirchwitz.
1202 PP/03 Report version information for many libraries, including
1203 Exim version information for dynamically loaded libraries. Created
1204 version.h, now support a version extension string for distributors
1205 who patch heavily. Dynamic module ABI change.
1207 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
1208 privilege escalation vulnerability whereby the Exim run-time user
1209 can cause root to append content of the attacker's choosing to
1212 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
1215 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
1216 If dropping privileges for untrusted macros, we disabled normal logging
1217 on the basis that it would fail; for the Exim run-time user, this is not
1218 the case, and it resulted in successful deliveries going unlogged.
1219 Fixed. Reported by Andreas Metzler.
1225 PP/01 Date: & Message-Id: revert to normally being appended to a message,
1226 only prepend for the Resent-* case. Fixes regression introduced in
1227 Exim 4.70 by NM/22 for Bugzilla 607.
1229 PP/02 Include check_rfc2047_length in configure.default because we're seeing
1230 increasing numbers of administrators be bitten by this.
1232 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
1234 PP/03 Bugzilla 994: added openssl_options main configuration option.
1236 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
1238 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
1240 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
1242 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
1243 without a peer certificate, leading to a segfault because of an
1244 assumption that peers always have certificates. Be a little more
1245 paranoid. Problem reported by Martin Tscholak.
1247 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
1248 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
1249 NB: ClamAV planning to remove STREAM in "middle of 2010".
1250 CL also introduces -bmalware, various -d+acl logging additions and
1251 more caution in buffer sizes.
1253 PP/09 Implemented reverse_ip expansion operator.
1255 PP/10 Bugzilla 937: provide a "debug" ACL control.
1257 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
1259 PP/12 Bugzilla 973: Implement --version.
1261 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
1263 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
1265 PP/15 Bugzilla 816: support multiple condition rules on Routers.
1267 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
1268 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
1269 ignore trailing whitespace.
1271 JJ/02 prevent non-panic DKIM error from being sent to paniclog
1273 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
1276 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
1277 Notification from Dr Andrew Aitchison.
1279 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
1280 ExtendedDetectionInfo response format.
1281 Notification from John Horne.
1283 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
1286 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
1287 XSL and documented dependency on system catalogs, with examples of how
1290 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
1293 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
1294 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
1295 configuration file which is writeable by the Exim user or group.
1297 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
1298 of configuration files to cover files specified with the -C option if
1299 they are going to be used with root privileges, not just the default
1302 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
1303 option (effectively making it always true).
1305 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
1306 files to be used while preserving root privileges.
1308 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
1309 that rogue child processes cannot use them.
1311 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
1312 run-time user, instead of root.
1314 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
1315 Exim run-time user without dropping privileges.
1317 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
1318 result string, instead of calling string_vformat() twice with the same
1321 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
1322 for other users. Others should always drop root privileges if they use
1323 -C on the command line, even for a whitelisted configure file.
1325 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
1327 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
1333 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
1334 $data_path, and $header_path variables; fixed documentation bugs and
1337 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
1338 exipick to access non-standard spools, including the "frozen" queue
1341 NM/01 Bugzilla 965: Support mysql stored procedures.
1342 Patch from Alain Williams
1344 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
1346 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
1347 Patch from Andreas Metzler
1349 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
1350 Patch from Kirill Miazine
1352 NM/05 Bugzilla 671: Added umask to procmail example.
1354 JJ/03 installed exipick 20100323.0, fixing doc bug
1356 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
1357 directory. Notification and patch from Dan Rosenberg.
1359 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
1361 TK/02 Improve log output when DKIM signing operation fails.
1363 MH/01 Treat the transport option dkim_domain as a colon separated
1364 list, not as a single string, and sign the message with each element,
1365 omitting multiple occurences of the same signer.
1367 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
1368 Bugzilla 985, 986. Patch by Simon Arlott
1370 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
1371 Patch by Simon Arlott
1373 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
1374 MBX locking. Notification from Dan Rosenberg.
1380 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
1382 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
1384 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
1386 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
1388 NM/04 Bugzilla 915: Flag broken perl installation during build.
1394 TK/01 Added patch by Johannes Berg that expands the main option
1395 "spamd_address" if it starts with a dollar sign.
1397 TK/02 Write list of recipients to X-Envelope-Sender header when building
1398 the mbox-format spool file for content scanning (suggested by Jakob
1401 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
1402 (http://www.dcc-servers.net/) support via dccifd. Activated by
1403 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
1405 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
1406 by Mark Daniel Reidel <mr@df.eu>.
1408 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
1409 When building exim an external PCRE library is now needed -
1410 PCRE is a system library on the majority of modern systems.
1411 See entry on PCRE_LIBS in EDITME file.
1413 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
1414 conversation. Added nologin parameter to request.
1415 Patch contributed by Kirill Miazine.
1417 TF/01 Do not log submission mode rewrites if they do not change the address.
1419 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
1421 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
1422 log files in place. Contributed by Roberto Lima.
1424 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
1426 TF/03 Bugzilla 615: When checking the local_parts router precondition
1427 after a local_part_suffix or local_part_prefix option, Exim now
1428 does not use the address's named list lookup cache, since this
1429 contains cached lookups for the whole local part.
1431 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
1432 Robert Millan. Documentation is in experimental-spec.txt.
1434 TF/04 Bugzilla 668: Fix parallel build (make -j).
1436 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
1438 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
1439 Patch provided by Jan Srzednicki.
1441 TF/05 Leading white space used to be stripped from $spam_report which
1442 wrecked the formatting. Now it is preserved.
1444 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
1445 that they are available at delivery time.
1447 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
1449 TF/08 TLS error reporting now respects the incoming_interface and
1450 incoming_port log selectors.
1452 TF/09 Produce a more useful error message if an SMTP transport's hosts
1453 setting expands to an empty string.
1455 NM/06 Bugzilla 744: EXPN did not work under TLS.
1456 Patch provided by Phil Pennock.
1458 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
1459 Patch provided by Richard Godbee.
1461 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
1462 acl_smtp_notquit, added index entry.
1464 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
1465 Patch provided by Eugene Bujak.
1467 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
1468 accept(). Patch provided by Maxim Dounin.
1470 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
1471 Patch provided by Phil Pennock.
1473 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
1475 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
1476 Patch provided by Brad "anomie" Jorsch.
1478 NM/14 Bugzilla 622: Added timeout setting to transport filter.
1479 Patch provided by Dean Brooks.
1481 TK/05 Add native DKIM support (does not depend on external libraries).
1483 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
1484 Patch provided by Graeme Fowler.
1486 NM/16 Bugzilla 851: Documentation example syntax fix.
1488 NM/17 Changed NOTICE file to remove references to embedded PCRE.
1490 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
1493 NM/19 Bugzilla 745: TLS version reporting.
1494 Patch provided by Phil Pennock.
1496 NM/20 Bugzilla 167: bool: condition support.
1497 Patch provided by Phil Pennock.
1499 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
1500 clients. Patch provided by Phil Pennock.
1502 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
1503 Patch provided by Brad "anomie" Jorsch.
1505 NM/23 Bugzilla 687: Fix misparses in eximstats.
1506 Patch provided by Heiko Schlittermann.
1508 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
1509 Patch provided by Heiko Schlittermann.
1511 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
1512 plus update to original patch.
1514 NM/26 Bugzilla 799: Documentation correction for ratelimit.
1516 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
1517 Patch provided by David Brownlee.
1519 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
1521 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
1523 NM/30 Bugzilla 888: TLS documentation bugfixes.
1525 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
1527 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
1528 Unlike the original bugzilla I have changed all shell scripts in src tree.
1530 NM/33 Bugzilla 898: Transport filter timeout fix.
1531 Patch by Todd Rinaldo.
1533 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
1534 Patch by Serge Demonchaux.
1536 NM/35 Bugzilla 39: Base64 decode bug fixes.
1537 Patch by Jakob Hirsch.
1539 NM/36 Bugzilla 909: Correct connect() call in dcc code.
1541 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
1543 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
1545 NM/39 Bugzilla 911: Fixed MakeLinks build script.
1551 TK/01 Add preliminary DKIM support. Currently requires a forked version of
1552 ALT-N's libdkim that I have put here:
1553 http://duncanthrax.net/exim-experimental/
1555 Note to Michael Haardt: I had to rename some vars in sieve.c. They
1556 were called 'true' and it seems that C99 defines that as a reserved
1557 keyword to be used with 'bool' variable types. That means you could
1558 not include C99-style headers which use bools without triggering
1559 build errors in sieve.c.
1561 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
1562 as mailq or other aliases. Changed the --help handling significantly
1563 to do whats expected. exim_usage() emits usage/help information.
1565 SC/01 Added the -bylocaldomain option to eximstats.
1567 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
1569 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
1571 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
1577 PH/01 Another patch from the Sieve maintainer.
1579 PH/02 When an IPv6 address is converted to a string for single-key lookup
1580 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
1581 dots are used instead of colons so that keys in lsearch files need not
1582 contain colons. This was done some time before quoting was made available
1583 in lsearch files. However, iplsearch files do require colons in IPv6 keys
1584 (notated using the quote facility) so as to distinguish them from IPv4
1585 keys. This meant that lookups for IP addresses in host lists did not work
1586 for iplsearch lookups.
1588 This has been fixed by arranging for IPv6 addresses to be expressed with
1589 colons if the lookup type is iplsearch. This is not incompatible, because
1590 previously such lookups could never work.
1592 The situation is now rather anomolous, since one *can* have colons in
1593 ordinary lsearch keys. However, making the change in all cases is
1594 incompatible and would probably break a number of configurations.
1596 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
1599 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
1600 conversion specification without a maximum field width, thereby enabling
1601 a rogue spamd server to cause a buffer overflow. While nobody in their
1602 right mind would setup Exim to query an untrusted spamd server, an
1603 attacker that gains access to a server running spamd could potentially
1604 exploit this vulnerability to run arbitrary code as the Exim user.
1606 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
1607 $primary_hostname instead of what libspf2 thinks the hosts name is.
1609 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
1610 a directory entry by the name of the lookup key. Previously, if a
1611 symlink pointed to a non-existing file or a file in a directory that
1612 Exim lacked permissions to read, a lookup for a key matching that
1613 symlink would fail. Now it is enough that a matching directory entry
1614 exists, symlink or not. (Bugzilla 503.)
1616 PH/03 The body_linecount and body_zerocount variables are now exported in the
1619 PH/04 Added the $dnslist_matched variable.
1621 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
1622 This means they are set thereafter only if the connection becomes
1625 PH/06 Added the client_condition to authenticators so that some can be skipped
1626 by clients under certain conditions.
1628 PH/07 The error message for a badly-placed control=no_multiline_responses left
1629 "_responses" off the end of the name.
1631 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
1633 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
1634 (without spaces) instead of just copying the configuration text.
1636 PH/10 Added the /noupdate option to the ratelimit ACL condition.
1638 PH/11 Added $max_received_linelength.
1640 PH/12 Added +ignore_defer and +include_defer to host lists.
1642 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
1643 way in which PCRE > 7.0 is built.
1645 PH/14 Implemented queue_only_load_latch.
1647 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
1648 MAIL command. The effect was to mangle the value on 64-bit systems.
1650 PH/16 Another patch from the Sieve maintainer.
1652 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
1654 PH/18 If a system quota error occurred while trying to create the file for
1655 a maildir delivery, the message "Mailbox is full" was not appended to the
1656 bounce if the delivery eventually timed out. Change 4.67/27 below applied
1657 only to a quota excession during the actual writing of the file.
1659 PH/19 It seems that peer DN values may contain newlines (and other non-printing
1660 characters?) which causes problems in log lines. The DN values are now
1661 passed through string_printing() before being added to log lines.
1663 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
1664 and InterBase are left for another time.)
1666 PH/21 Added message_body_newlines option.
1668 PH/22 Guard against possible overflow in moan_check_errorcopy().
1670 PH/23 POSIX allows open() to be a macro; guard against that.
1672 PH/24 If the recipient of an error message contained an @ in the local part
1673 (suitably quoted, of course), incorrect values were put in $domain and
1674 $local_part during the evaluation of errors_copy.
1680 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
1681 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
1684 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
1685 issue a MAIL command.
1687 PH/02 In an ACL statement such as
1689 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
1691 if a client was not listed at all, or was listed with a value other than
1692 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
1693 the condition was not true (as it should be), so access was not denied.
1694 The bug was that the ! inversion was incorrectly passed on to the second
1695 item. This has been fixed.
1697 PH/03 Added additional dnslists conditions == and =& which are different from
1698 = and & when the dns lookup returns more than one IP address.
1700 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
1701 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
1703 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
1704 FSYNC, which compiles an option called disable_fsync that allows for
1705 bypassing fsync(). The documentation is heavily laced with warnings.
1707 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
1709 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
1710 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
1711 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
1712 including adding "make clean"; (3) Added -fPIC when compiling the test
1713 dynamically loaded module, to get rid of a warning.
1715 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
1716 message fails, move_frozen_messages = true and ignore_bounce_errors_after
1717 = 0s. The bug is otherwise harmless.
1719 PH/07 There was a bug in the dovecot authenticator such that the value of
1720 $auth1 could be overwritten, and so not correctly preserved, after a
1721 successful authentication. This usually meant that the value preserved by
1722 the server_setid option was incorrect.
1724 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
1726 PH/09 Installed PCRE release 7.0.
1728 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
1729 run for batched SMTP input. It is now run at the start of every message
1730 in the batch. While fixing this I discovered that the process information
1731 (output by running exiwhat) was not always getting set for -bs and -bS
1732 input. This is fixed, and it now also says "batched" for BSMTP.
1734 PH/11 Added control=no_pipelining.
1736 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
1737 patch, slightly modified), and move the expansion of helo_data till after
1738 the connection is made in the smtp transport (so it can use these
1741 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
1743 PH/14 Added log_selector = +pid.
1745 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
1747 PH/16 Add ${if forany and ${if forall.
1749 PH/17 Added dsn_from option to vary the From: line in DSNs.
1751 PH/18 Flush SMTP output before performing a callout, unless control =
1752 no_callout_flush is set.
1754 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
1755 was true (the default) a successful delivery failed to delete the retry
1756 item, thus causing premature timeout of the address. The bug is now
1759 PH/20 Added hosts_avoid_pipelining to the smtp transport.
1761 PH/21 Long custom messages for fakedefer and fakereject are now split up
1762 into multiline reponses in the same way that messages for "deny" and
1763 other ACL rejections are.
1765 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
1766 with slight modification.
1768 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
1769 draft, changing the syntax and factoring some duplicate code".
1771 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
1772 for deliveries of the second and subsequent messages over the same SMTP
1775 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
1776 ${reduce, with only minor "tidies".
1778 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
1780 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
1781 expansion side effects.
1783 PH/27 When a message times out after an over-quota error from an Exim-imposed
1784 quota, the bounce message says "mailbox is full". This message was not
1785 being given when it was a system quota that was exceeded. It now should
1788 MH/03 Made $recipients available in local_scan(). local_scan() already has
1789 better access to the recipient list through recipients_list[], but
1790 $recipients can be useful in postmaster-provided expansion strings.
1792 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
1793 in the case of a MAIL command with additional options following the
1794 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
1795 were accidentally chopped off.
1797 PH/29 SMTP synchronization checks are implemented when a command is read -
1798 there is a check that no more input is waiting when there shouldn't be
1799 any. However, for some commands, a delay in an ACL can mean that it is
1800 some time before the response is written. In this time, more input might
1801 arrive, invalidly. So now there are extra checks after an ACL has run for
1802 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
1803 pipelining has not been advertised.
1805 PH/30 MH's patch to allow iscntrl() characters to be list separators.
1807 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
1808 returned in the SMTP response when smtp_return_error_details was false.
1809 This has been fixed.
1811 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
1812 instead of the C I/O that was originally supplied, because problems were
1813 reported on Solaris.
1815 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
1816 Exim which did not show up earlier: it was assuming that a call to
1817 SSL_CTX_set_info_callback() might give an error value. In fact, there is
1818 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
1819 was a macro that became an assignment, so it seemed to work. This has
1820 changed to a proper function call with a void return, hence the compile
1821 error. Exim's code has been fixed.
1823 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
1826 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
1828 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
1830 PH/37 If a message is not accepted after it has had an id assigned (e.g.
1831 because it turns out to be too big or there is a timeout) there is no
1832 "Completed" line in the log. When some messages of this type were
1833 selected by exigrep, they were listed as "not completed". Others were
1834 picked up by some special patterns. I have improved the selection
1835 criteria to be more general.
1837 PH/38 The host_find_failed option in the manualroute router can now be set
1838 to "ignore", to completely ignore a host whose IP address cannot be
1839 found. If all hosts are ignored, the behaviour is controlled by the new
1840 host_all_ignored option.
1842 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
1843 homing or because of multiple MX records with /mx) generated more than
1844 one IP address, and the following item turned out to be the local host,
1845 all the secondary addresses of the first item were incorrectly removed
1846 from the list, along with the local host and any following hosts (which
1847 is what is supposed to happen).
1849 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
1850 whoever called Exim into the -H file. In the case of the daemon it was
1851 behaving confusingly. When first started, it used values for whoever
1852 started the daemon, but after a SIGHUP it used the Exim user (because it
1853 calls itself on a restart). I have changed the code so that it now always
1856 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
1857 message are rejected with the same error (e.g. no authentication or bad
1858 sender address), and a DATA command is nevertheless sent (as can happen
1859 with PIPELINING or a stupid MUA), the error message that was given to the
1860 RCPT commands is included in the rejection of the DATA command. This is
1861 intended to be helpful for MUAs that show only the final error to their
1864 PH/42 Another patch from the Sieve maintainer.
1866 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
1867 Eximstats - Fixed some broken HTML links and added missing column headers
1869 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
1870 columns for Rejects, Temp Rejects, Ham, and Spam rows.
1872 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
1874 PH/43 Yet another patch from the Sieve maintainer.
1876 PH/44 I found a way to check for a TCP/IP connection going away before sending
1877 the response to the final '.' that terminates a message, but only in the
1878 case where the client has not sent further data following the '.'
1879 (unfortunately, this is allowed). However, in many cases there won't be
1880 any further data because there won't be any more messages to send. A call
1881 to select() can be used: if it shows that the input is "ready", there is
1882 either input waiting, or the socket has been closed. An attempt to read
1883 the next input character can distinguish the two cases. Previously, Exim
1884 would have sent an OK response which the client would never have see.
1885 This could lead to message repetition. This fix should cure that, at
1886 least in a lot of common cases.
1888 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
1889 advertised in response to EHLO.
1895 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
1896 fixed by 4.65/MH/01 (is this a record?) are fixed:
1898 (i) An empty string was always treated as zero by the numeric comparison
1899 operators. This behaviour has been restored.
1901 (ii) It is documented that the numeric comparison operators always treat
1902 their arguments as decimal numbers. This was broken in that numbers
1903 starting with 0 were being interpreted as octal.
1905 While fixing these problems I realized that there was another issue that
1906 hadn't been noticed. Values of message_size_limit (both the global option
1907 and the transport option) were treated as octal if they started with 0.
1908 The documentation was vague. These values are now always treated as
1909 decimal, and I will make that clear in the documentation.
1915 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
1916 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
1919 MH/01 Don't check that the operands of numeric comparison operators are
1920 integers when their expansion is in "skipping" mode (fixes bug
1921 introduced by 4.64-PH/07).
1923 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
1924 child addresses, Exim now panics and dies. Previously, because the count
1925 is held in a short int, deliveries were likely to be lost. As such a
1926 large number of recipients for a single message is ridiculous
1927 (performance will be very, very poor), I have chosen to impose a limit
1928 rather than extend the field.
1934 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
1935 leftover -K file (the existence of which was triggered by #402).
1936 While we were at it, introduced process PID as part of the -K
1937 filename. This should rule out race conditions when creating
1940 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
1941 processing considerably. Previous code took too long for large mails,
1942 triggering a timeout which in turn triggers #401.
1944 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
1945 in the DK code in transports.c. sendfile() is not really portable,
1946 hence the _LINUX specificness.
1948 TF/01 In the add_headers option to the mail command in an Exim filter,
1949 there was a bug that Exim would claim a syntax error in any
1950 header after the first one which had an odd number of characters
1953 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
1954 callout verification, Exim cached a "reject" for the entire domain. This
1955 is correct for most verifications, but it is not correct for a recipient
1956 verification with use_sender or use_postmaster set, because in that case
1957 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
1958 case of MAIL FROM:<> rejection from other early rejections (e.g.
1959 rejection of HELO). When verifying a recipient using a non-null MAIL
1960 address, the cache is ignored if it shows MAIL FROM:<> rejection.
1961 Whatever the result of the callout, the value of the domain cache is
1962 left unchanged (for any other kind of callout, getting as far as trying
1963 RCPT means that the domain itself is ok).
1965 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
1968 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
1969 manifest itself as EPIPE rather than ECONNECT. When tidying away a
1970 session, the daemon ignores ECONNECT errors and logs others; it now
1971 ignores EPIPE as well.
1973 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
1974 (quoted-printable decoding).
1976 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
1977 later the small subsequent patch to fix an introduced bug.
1979 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
1981 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
1983 PH/08 An error is now given if message_size_limit is specified negative.
1985 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
1986 to be given (somewhat) arbitrary names.
1988 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
1991 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
1992 miscellaneous code fixes
1994 PH/10 Added the log_reject_target ACL modifier to specify where to log
1997 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
1998 hostname. This is wrong, because it relates to the incoming message (and
1999 probably the interface on which it is arriving) and not to the outgoing
2000 callout (which could be using a different interface). This has been
2001 changed to use the value of the helo_data option from the smtp transport
2002 instead - this is what is used when a message is actually being sent. If
2003 there is no remote transport (possible with a router that sets up host
2004 addresses), $smtp_active_hostname is used.
2006 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
2007 tweaks were necessary in order to get it to work (see also 21 below):
2008 (a) The code assumed that strncpy() returns a negative number on buffer
2009 overflow, which isn't the case. Replaced with Exim's string_format()
2011 (b) There were several signed/unsigned issues. I just did the minimum
2012 hacking in of casts. There is scope for a larger refactoring.
2013 (c) The code used strcasecmp() which is not a standard C function.
2014 Replaced with Exim's strcmpic() function.
2015 (d) The code set only $1; it now sets $auth1 as well.
2016 (e) A simple test gave the error "authentication client didn't specify
2017 service in request". It would seem that Dovecot has changed its
2018 interface. Fortunately there's a specification; I followed it and
2019 changed what the client sends and it appears to be working now.
2021 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
2024 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
2025 address A is aliased to B and C, where B exists and C does not. Without
2026 -v the output is "A verified" because verification stops after a
2027 successful redirection if more than one address is generated. However,
2028 with -v the child addresses are also verified. Exim was outputting "A
2029 failed to verify" and then showing the successful verification for C,
2030 with its parentage. It now outputs "B failed to verify", showing B's
2031 parentage before showing the successful verification of C.
2033 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
2034 look up a TXT record in a specific list after matching in a combined
2037 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
2038 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
2039 they consult the DNS. I had assumed they would set it the way they
2040 wanted; and indeed my experiments on Linux seem to show that in some
2041 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
2042 To be on the safe side, however, I have now made the interface to
2043 host_find_byname() similar to host_find_bydns(), with an argument
2044 containing the DNS resolver options. The host_find_byname() function now
2045 sets these options at its start, just as host_find_bydns() does. The smtp
2046 transport options dns_qualify_single and dns_search_parents are passed to
2047 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
2048 of host_find_byname() use the default settings of RES_DEFNAMES
2049 (qualify_single) but not RES_DNSRCH (search_parents).
2051 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
2052 spool_read_header() do less string testing, by means of a preliminary
2053 switch on the second character of optional "-foo" lines. (This is
2054 overdue, caused by the large number of possibilities that now exist.
2055 Originally there were few.) While I was there, I also converted the
2056 str(n)cmp tests so they don't re-test the leading "-" and the first
2057 character, in the hope this might squeeze out yet more improvement.
2059 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
2060 flag allowing group syntax was set by the header_syntax check but not
2061 turned off, possible causing trouble later; (2) The flag was not being
2062 set at all for the header_verify test, causing "group"-style headers to
2063 be rejected. I have now set it in this case, and also caused header_
2064 verify to ignore an empty address taken from a group. While doing this, I
2065 came across some other cases where the code for allowing group syntax
2066 while scanning a header line wasn't quite right (mostly, not resetting
2067 the flag correctly in the right place). These bugs could have caused
2068 trouble for malformed header lines. I hope it is now all correct.
2070 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
2071 with the "reply" argument non-NULL. The code, however (which originally
2072 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
2073 but it didn't always do it. This confused somebody who was copying the
2074 code for some other use. I have removed all the tests.
2076 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
2077 feature that was used to support insecure browsers during the U.S. crypto
2078 embargo. It requires special client support, and Exim is probably the
2079 only MTA that supported it -- and would never use it because real RSA is
2080 always available. This code has been removed, because it had the bad
2081 effect of slowing Exim down by computing (never used) parameters for the
2082 RSA_EXPORT functionality.
2084 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
2085 authenticator to fail if there's a tab character in the incoming data
2086 (there should never be unless someone is messing about, as it's supposed
2087 to be base64-encoded). Also added, on Timo's advice, the "secured" option
2088 if the connection is using TLS or if the remote IP is the same as the
2089 local IP, and the "valid-client-cert option" if a client certificate has
2092 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
2093 authenticators. This can be used for authorization after authentication
2094 succeeds. (In the case of plaintext, it servers for both authentication
2097 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
2098 if any retry times were supplied.
2100 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
2101 connection, where there is no client IP address to check. In this
2102 situation, the verify now always succeeds.
2104 PH/25 Applied John Jetmore's -Mset patch.
2106 PH/26 Added -bem to be like -Mset, but loading a message from a file.
2108 PH/27 In a string expansion for a processed (not raw) header when multiple
2109 headers of the same name were present, leading whitespace was being
2110 removed from all of them, but trailing whitespace was being removed only
2111 from the last one. Now trailing whitespace is removed from each header
2112 before concatenation. Completely empty headers in a concatenation (as
2113 before) are ignored.
2115 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
2116 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
2118 PH/29 [Removed. This was a change that I later backed out, and forgot to
2119 correct the ChangeLog entry (that I had efficiently created) before
2120 committing the later change.]
2122 PH/30 Exim was sometimes attempting to deliver messages that had suffered
2123 address errors (4xx response to RCPT) over the same connection as other
2124 messages routed to the same hosts. Such deliveries are always "forced",
2125 so retry times are not inspected. This resulted in far too many retries
2126 for the affected addresses. The effect occurred only when there were more
2127 hosts than the hosts_max_try setting in the smtp transport when it had
2128 the 4xx errors. Those hosts that it had tried were not added to the list
2129 of hosts for which the message was waiting, so if all were tried, there
2130 was no problem. Two fixes have been applied:
2132 (i) If there are any address or message errors in an SMTP delivery, none
2133 of the hosts (tried or untried) are now added to the list of hosts
2134 for which the message is waiting, so the message should not be a
2135 candidate for sending over the same connection that was used for a
2136 successful delivery of some other message. This seems entirely
2137 reasonable: after all the message is NOT "waiting for some host".
2138 This is so "obvious" that I'm not sure why it wasn't done
2139 previously. Hope I haven't missed anything, but it can't do any
2140 harm, as the worst effect is to miss an optimization.
2142 (ii) If, despite (i), such a delivery is accidentally attempted, the
2143 routing retry time is respected, so at least it doesn't keep
2144 hammering the server.
2146 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
2147 in ${readsocket because some servers need this prod.
2149 PH/32 Added some extra debug output when updating a wait-xxx database.
2151 PH/33 The hint "could be header name not terminated by colon", which has been
2152 given for certain expansion errors for a long time, was not being given
2153 for the ${if def:h_colon_omitted{... case.
2155 PH/34 The spec says: "With one important exception, whenever a domain list is
2156 being scanned, $domain contains the subject domain." There was at least
2157 one case where this was not true.
2159 PH/35 The error "getsockname() failed: connection reset by peer" was being
2160 written to the panic log as well as the main log, but it isn't really
2161 panic-worthy as it just means the connection died rather early on. I have
2162 removed the panic log writing for the ECONNRESET error when getsockname()
2165 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
2166 runs only) independently of the message's sender address. This meant
2167 that, if the 4xx error was in fact related to the sender, a different
2168 message to the same recipient with a different sender could confuse
2169 things. In particualar, this can happen when sending to a greylisting
2170 server, but other circumstances could also provoke similar problems.
2171 I have changed the default so that the retry time for these errors is now
2172 based a combination of the sender and recipient addresses. This change
2173 can be overridden by setting address_retry_include_sender=false in the
2176 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
2177 remote server are returned as part of bounce messages. This was not
2178 happening for LMTP over a pipe (the lmtp transport), but now it is the
2179 same for both kinds of LMTP.
2181 PH/38 Despite being documented as not happening, Exim was rewriting addresses
2182 in header lines that were in fact CNAMEs. This is no longer the case.
2184 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
2185 and queue runs started by the daemon processed all messages. This has
2186 been fixed so that -R and -S can now usefully be given with -q<time>.
2188 PH/40 Import PCRE release 6.7 (fixes some bugs).
2190 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
2192 PH/42 Give an error if -q is specified more than once.
2194 PH/43 Renamed the variables $interface_address and $interface_port as
2195 $received_ip_address and $received_port, to make it clear that these
2196 values apply to message reception, and not to the outgoing interface when
2197 a message is delivered. (The old names remain recognized, of course.)
2199 PH/44 There was no timeout on the connect() call when using a Unix domain
2200 socket in the ${readsocket expansion. There now is.
2202 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
2203 be meaningful with "accept".
2205 SC/01 Eximstats V1.43
2206 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
2208 SC/02 Eximstats V1.44
2209 Use a glob alias rather than an array ref in the generated
2210 parser. This improves both readability and performance.
2212 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
2213 Collect SpamAssassin and rejection statistics.
2214 Don't display local sender or destination tables unless
2215 there is data to show.
2216 Added average volumes into the top table text output.
2218 SC/04 Eximstats V1.46
2219 Collect data on the number of addresses (recipients)
2220 as well as the number of messages.
2222 SC/05 Eximstats V1.47
2223 Added 'Message too big' to the list of mail rejection
2224 reasons (thanks to Marco Gaiarin).
2226 SC/06 Eximstats V1.48
2227 Mainlog lines which have GMT offsets and are too short to
2228 have a flag are now skipped.
2230 SC/07 Eximstats V1.49 (Alain Williams)
2231 Added the -emptyok flag.
2233 SC/08 Eximstats V1.50
2234 Fixes for obtaining the IP address from reject messages.
2236 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
2237 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
2238 whitesspace changes from 4.64-PH/27
2240 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
2243 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
2244 are found, allow negative numbers in numeric criteria)
2246 JJ/06 exipick.20061117.2, added new $message_body_missing variable
2248 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
2249 to match changes made in 4.64-PH/43
2251 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
2253 PH/47 Put in an explicit test for a DNS lookup of an address record where the
2254 "domain" is actually an IP address, and force a failure. This locks out
2255 those revolvers/nameservers that support "A-for-A" lookups, in
2256 contravention of the specifications.
2258 PH/48 When a host name was looked up from an IP address, and the subsequent
2259 forward lookup of the name timed out, the host name was left in
2260 $sender_host_name, contrary to the specification.
2262 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
2263 restricted to single-key lookups, Exim was not diagnosing an error if
2264 * or *@ was used with a query-style lookup.
2266 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
2268 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
2269 long ago, but noone interested enough thought of it. Let's just say that
2270 the "1.1" means that there are some new functions that weren't there at
2271 some point in the past.
2273 PH/51 Error processing for expansion failure of helo_data from an smtp
2274 transport during callout processing was broken.
2276 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
2277 tested/used via the -bh/-bhc/-bs options.
2279 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
2280 bug, fixed in subsequent PCRE releases).
2282 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
2283 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
2285 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
2291 SC/01 Use a glob alias rather than an array ref in eximstats generated
2292 parser. This improves both readability and performance.
2294 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
2295 Don't display local sender or destination tables in eximstats unless
2296 there is data to show.
2297 Added average volumes into the eximstats top table text output.
2299 SC/03 Collect data on the number of addresses (recipients) as well
2300 as the number of messages in eximstats.
2302 TF/01 Correct an error in the documentation for the redirect router. Exim
2303 does (usually) call initgroups() when daemonizing.
2305 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
2306 with consistent privilege compared to when running as a daemon.
2308 TF/03 Note in the spec that $authenticated_id is not set for local
2309 submissions from trusted users.
2311 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
2312 Thanks to Dean Brooks <dean@iglou.com> for the patch.
2314 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
2315 by adding some example configuration directives to the default
2316 configuration file. A little bit of work is required to uncomment the
2317 directives and define how usernames and passwords are checked, but
2318 there is now a framework to start from.
2320 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
2321 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
2322 without this. I don't know how relevant this is to other LDAP libraries.
2324 PH/02 Add the verb name to the "unknown ACL verb" error.
2326 PH/03 Magnus Holmgren's patch for filter_prepend_home.
2328 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
2330 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
2331 directory not expanded when it should be if an expanded home directory
2332 was set for the address (which is overridden by the transport).
2334 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
2337 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
2338 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
2339 because it is too late at that time, and has no effect.
2341 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
2342 security issue with \' (bugzilla #107). I could not use the
2343 PQescapeStringConn() function, because it needs a PGconn value as one of
2346 PH/08 When testing addresses using -bt, indicate those final addresses that
2347 are duplicates that would not cause an additional delivery. At least one
2348 person was confused, thinking that -bt output corresponded to deliveries.
2349 (Suppressing duplicates isn't a good idea as you lose the information
2350 about possibly different redirections that led to the duplicates.)
2352 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
2353 systems where poll() doesn't work, in particular OS X.
2355 PH/10 Added more information to debugging output for retry time not reached.
2357 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
2358 operations in malware.c.
2360 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
2363 PH/13 If write_rejectlog was set false when logging was sent to syslog with
2364 syslog_duplication set false, log lines that would normally be written
2365 both the the main log and to the reject log were not written to syslog at
2368 PH/14 In the default configuration, change the use of "message" in ACL warn
2369 statements to "add_header".
2371 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
2372 not followed by a command (e.g. "seen endif").
2374 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
2375 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
2378 PH/17 Added extra conditions to the default value of delay_warning_condition
2382 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
2383 { match{$h_precedence:}{(?i)bulk|list|junk} } \
2384 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
2387 The Auto-Submitted: and various List- headers are standardised, whereas I
2388 don't think Precedence: ever was.
2390 PH/18 Refactored debugging code in route_finduser() to show more information,
2391 in particular, the error code if getpwnam() issues one.
2393 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
2394 This is apparently needed in addition to the PH/07 change above to avoid
2395 any possible encoding problems.
2397 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
2398 but not after initializing Perl.
2400 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
2401 output them only if debugging. By default they are written stderr,
2402 apparently, which is not desirable.
2404 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
2407 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
2410 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
2412 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
2413 authenticated or an ident call has been made. Suppress the default
2414 values for $authenticated_id and $authenticated_sender (but permit -oMai
2415 and -oMas) when testing with -bh.
2417 PH/24 Re-jigged the order of the tests in the default configuration so that the
2418 tests for valid domains and recipients precede the DNS black list and CSA
2419 tests, on the grounds that those ones are more expensive.
2421 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
2422 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
2423 This bug exists in every version of Exim that I still have, right back to
2426 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
2427 However, an attempt to turn on case-sensitivity in a regex key by
2428 including (?-i) didn't work because the subject string was already
2429 lowercased, and the effects were non-intuitive. It turns out that a
2430 one-line patch can be used to allow (?-i) to work as expected.
2436 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
2437 other effects) broke the use of negated acl sub-conditions.
2439 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
2442 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
2443 "Deny" causes Exim to reject the incoming connection with a 554 error.
2444 Unfortunately, if there is a major crisis, such as a disk failure,
2445 tcp-wrappers gives "deny", whereas what one would like would be some
2446 kind of temporary error. A kludge has been added to help with this.
2447 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
2448 554 error is used if errno is still zero or contains ENOENT (which occurs
2449 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
2452 PH/03 Add -lutil to the default FreeBSD LIBS setting.
2454 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
2455 errors. Otherwise a message that provokes a temporary error (when other
2456 messages do not) can cause a whole host to time out.
2458 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
2459 addresses were routed directly to files or pipes from a redirect router.
2460 File deliveries just didn't batch; pipe deliveries might have suffered
2463 PH/06 A failure to get a lock for a hints database would erroneously always say
2464 "Failed to get write lock", even when it was really a read lock.
2466 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
2467 of 0600. This has been changed to use the value of the lockfile_mode
2468 option (which defaults to 0600).
2470 PH/08 Applied small patch from the Sieve maintainer.
2472 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
2473 folder from quota calculations, a direct delivery into this folder messed
2474 up the contents of the maildirsize file. This was because the regex was
2475 used only to exclude .Trash (or whatever) when the size of the mailbox
2476 was calculated. There was no check that a delivery was happening into an
2477 excluded directory. This bug has been fixed by ignoring all quota
2478 processing for deliveries into excluded directories.
2480 PH/10 Added the maildirfolder_create_regex option to appendfile.
2486 PH/01 The code for finding all the local interface addresses on a FreeBSD
2487 system running IPv6 was broken. This may well have applied to all BSD
2488 systems, as well as to others that have similar system calls. The broken
2489 code found IPv4 interfaces correctly, but gave incorrect values for the
2490 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
2491 that it would not match correctly against @[] and not recognize the IPv6
2494 PH/02 The ipliteral router was not recognizing addresses of the form user@
2495 [ipv6:....] because it didn't know about the "ipv6:" prefix.
2497 PH/03 Added disable_ipv6.
2499 PH/04 Changed $reply_address to use the raw form of the headers instead of the
2500 decoded form, because it is most often used to construct To: headers
2501 lines in autoreplies, and the decoded form may well be syntactically
2502 invalid. However, $reply_address has leading white space removed, and all
2503 newlines turned into spaces so that the autoreply transport does not
2506 PH/05 If group was specified without a user on a router, and no group or user
2507 was specified on a transport, the group from the router was ignored.
2509 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
2510 for visible compile-time settings that can be used to change these
2511 numbers, for those that want even more. Backwards compatibility with old
2512 spool files has been maintained. However, going back to a previous Exim
2513 release will lost any variables that are in spool files.
2515 PH/07 Two small changes when running in the test harness: increase delay when
2516 passing a TCP/IP connection to a new process, in case the original
2517 process has to generate a bounce, and remove special handling of
2518 127.0.0.2 (sic), which is no longer necessary.
2520 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
2521 be the same on different OS.
2523 PH/09 Moved a debug statement in filter processing to avoid a race problem when
2526 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
2527 whether --show-vars was specified or not
2529 JJ/02 exipick: Added support for new ACL variable spool format introduced
2532 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
2533 syntactically invalid From: or Reply-to: line, and a filter used this to
2534 generate an autoreply, and therefore failed to obtain an address for the
2535 autoreply, Exim could try to deliver to a non-existent relative file
2536 name, causing unrelated and misleading errors. What now happens is that
2537 it logs this as a hard delivery error, but does not attempt to create a
2540 PH/11 The exinext utility has a -C option for testing purposes, but although
2541 the given file was scanned by exinext itself; it wasn't being passed on
2542 when Exim was called.
2544 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
2545 an end-of-file indication when reading a command response.
2547 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
2548 compiled. In many other places in Exim, IPv6 addresses are always
2549 recognized, so I have changed this. It also means that IPv4 domain
2550 literals of the form [IPV4:n.n.n.n] are now always recognized.
2552 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
2553 used if the router is not running as root, for example, when verifying at
2554 ACL time, or when using -bh. The debugging output from this situation was
2555 non-existent - all you got was a failure to exec. I have made two
2558 (a) Failures to set uid/gid, the current directory, or a process leader
2559 in a subprocess such as that created by queryprogram now generate
2560 suitable debugging ouput when -d is set.
2562 (b) The queryprogram router detects when it is not running as root,
2563 outputs suitable debugging information if -d is set, and then runs
2564 the subprocess without attempting to change uid/gid.
2566 PH/15 Minor change to Makefile for building test_host (undocumented testing
2569 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
2570 additional section of a DNS packet that returns MX or SRV records.
2571 Instead, it always explicitly searches for A/AAAA records. This avoids
2572 major problems that occur when a DNS server includes only records of one
2573 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
2574 fixed another bug: if SRV records were looked up and the corresponding
2575 address records were *not* found in the additional section, the port
2576 values from the SRV records were lost.
2578 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
2579 using the correct key (the original address) when searching the retry
2580 rules in order to find which one to use for generating the retry hint.
2582 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
2583 adding the default one. Similarly, if it contains a Reply-To: header, the
2584 errors_reply_to option, if set, is not used.
2586 PH/19 When calculating a retry time, Exim used to measure the "time since
2587 failure" by looking at the "first failed" field in the retry record. Now
2588 it does not use this if it is later than than the arrival time of the
2589 message. Instead it uses the arrival time. This makes for better
2590 behaviour in cases where some deliveries succeed, thus re-setting the
2591 "first failed" field. An example is a quota failure for a huge message
2592 when small messages continue to be delivered. Without this change, the
2593 "time since failure" will always be short, possible causing more frequent
2594 delivery attempts for the huge message than are intended.
2595 [Note: This change was subsequently modified - see PH/04 for 4.62.]
2597 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
2598 $1, $2, $3) because the numerical variables can be reset during some
2599 expansion items (e.g. "match"), thereby losing the authentication data.
2601 PH/21 Make -bV show the size of off_t variables so that the test suite can
2602 decide whether to run tests for quotas > 2G.
2604 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
2605 mailbox_size, and mailbox_filecount in the appendfile transport. If a
2606 filecount value is greater than 2G or if a quota value is greater than 2G
2607 on a system where the size of off_t is not greater than 4, a panic error
2610 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
2611 never match. The debug and -bh output now contains an explicit error
2612 message indicating a malformed IPv4 address or mask.
2614 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
2615 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
2616 PH/23 above applies.
2618 PH/25 Do not write to syslog when running in the test harness. The only
2619 occasion when this arises is a failure to open the main or panic logs
2620 (for which there is an explicit test).
2622 PH/26 Added the /no_tell option to "control=freeze".
2624 PH/27 If a host name lookup failed very early in a connection, for example, if
2625 the IP address matched host_lookup and the reverse lookup yielded a name
2626 that did not have a forward lookup, an error message of the form "no IP
2627 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
2628 could be logged. Now it outputs the IP address instead of "NULL".
2630 PH/28 An enabling patch from MH: add new function child_open_exim2() which
2631 allows the sender and the authenticated sender to be set when
2632 submitting a message from within Exim. Since child_open_exim() is
2633 documented for local_scan(), the new function should be too.
2635 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
2636 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
2637 results in an empty string is now treated as unset.
2639 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
2641 PH/31 Added sender_verify_fail logging option.
2643 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
2644 needed by RFC 822 but not by RFC 2822 was commented out. I have now
2645 tidied the source and removed it altogether.
2647 PH/33 When a queue run was abandoned because the load average was too high, a
2648 log line was always written; now it is written only if the queue_run log
2649 selector is set. In addition, the log line for abandonment now contains
2650 information about the queue run such as the pid. This is always present
2651 in "start" and "stop" lines but was omitted from the "abandon" line.
2653 PH/34 Omit spaces between a header name and the colon in the error message that
2654 is given when verify = headers_syntax fails (if there are lots of them,
2655 the message gets confusing).
2657 PH/35 Change the default for dns_check_names_pattern to allow slashes within
2658 names, as there are now some PTR records that contain slashes. This check
2659 is only to protect against broken name servers that fall over on strange
2660 characters, so the fact that it applies to all lookups doesn't matter.
2662 PH/36 Now that the new test suite is complete, we can remove some of the
2663 special code in Exim that was needed for the old test suite. For example,
2664 sorting DNS records because real resolvers return them in an arbitrary
2665 order. The new test suite's fake resolver always returns records in the
2668 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
2669 bounces) except when queue_only is set, to avoid logging races between
2670 the different processes.
2672 PH/38 Panic-die if .include specifies a non-absolute path.
2674 PH/39 A tweak to the "H" retry rule from its user.
2676 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
2677 a label. They prevented compilation on older perls.
2679 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
2680 a warning to be raised on newish perls.
2682 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
2683 on queue. Changes to match documented behaviour of showing count of
2684 messages matching specified criteria.
2686 PH/40 Changed the default ident timeout from 30s to 5s.
2688 PH/41 Added support for the use of login_cap features, on those BSD systems
2689 that have them, for controlling the resources used by pipe deliveries.
2691 PH/42 The content-scanning code uses fopen() to create files in which to put
2692 message data. Previously it was not paying any attention to the mode of
2693 the files. Exim runs with umask(0) because the rest of the code creates
2694 files with open(), and sets the required mode explicitly. Thus, these
2695 files were ending up world-writeable. This was not a big issue, because,
2696 being within the spool directory, they were not world-accessible. I have
2697 created a function called modefopen, which takes an additional mode
2698 argument. It sets umask(777), creates the file, chmods it to the required
2699 mode, then resets the umask. All the relevant calls to fopen() in the
2700 content scanning code have been changed to use this function.
2702 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
2703 to 24 hours. This avoids potential overflow problems when processing G
2704 and H retry rules. I suspect nobody ever tinkers with this value.
2706 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
2708 PH/45 When the plaintext authenticator is running as a client, the server's
2709 challenges are checked to ensure they are valid base64 strings. By
2710 default, the authentication attempt is cancelled if an invalid string is
2711 received. Setting client_ignore_invalid_base64 true ignores these errors.
2712 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
2713 they are received. Thus, the responses can be made to depend on the
2714 challenges. If an invalid string is ignored, an empty string is placed in
2717 PH/46 Messages that are created by the autoreply transport now contains a
2718 References: header, in accordance with RFCs 2822 and 3834.
2720 PH/47 Added authenticated_sender_force to the smtp transport.
2722 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
2724 PH/49 Installed latest patch from the Sieve maintainer.
2726 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
2727 was also set, the appendfile transport was unnecessarily scanning a
2728 directory of message files (e.g. for maildir delivery) to find the count
2729 of files (along with the size), even though it did not need this
2730 information. It now does the scan only if it needs to find either the
2731 size of the count of files.
2733 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
2735 PH/52 Two bugs concerned with error handling when the smtp transport is
2738 (i) Exim was not creating retry information for temporary errors given
2739 for individual recipients after the DATA command when the smtp transport
2740 was used in LMTP mode. This meant that they could be retried too
2741 frequently, and not timed out correctly.
2743 (ii) Exim was setting the flag that allows error details to be returned
2744 for LMTP errors on RCPT commands, but not for LMTP errors for individual
2745 recipients that were returned after the DATA command.
2747 PH/53 This is related to PH/52, but is more general: for any failing address,
2748 when detailed error information was permitted to be returned to the
2749 sender, but the error was temporary, then after the final timeout, only
2750 "retry timeout exceeded" was returned. Now it returns the full error as
2751 well as "retry timeout exceeded".
2753 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
2754 do this, and (what is worse) MTAs that accept it.
2756 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
2757 will now be deprecated.
2759 PH/56 New os.c-cygwin from the Cygwin maintainer.
2761 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
2762 formats (previously only available in exim formats via -bpr, -bpru,
2763 and -bpra. Now also available in native and exiqgrep formats)
2765 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
2766 with very large, slow to parse queues
2768 JJ/08 exipick: added ! as generic prefix to negate any criteria format
2770 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
2772 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
2773 responses to authentication challenges, though it was showing the
2774 challenges; (ii) I've removed the CR characters from the debug output for
2777 PH/58 Allow for the insertion of a newline as well as a space when a string
2778 is turned into more than one encoded-word during RFC 2047 encoding. The
2779 Sieve code now uses this.
2781 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
2782 data_4xx, lost_connection, tls_required.
2784 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
2785 message was being sent as an SMTP response.
2787 PH/61 Add -l and -k options to exicyclog.
2789 PH/62 When verifying, if an address was redirected to one new address, so that
2790 verification continued, and the new address failed or deferred after
2791 having set something in $address_data, the value of $address_data was not
2792 passed back to the ACL. This was different to the case when no
2793 redirection occurred. The value is now passed back in both cases.
2795 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
2796 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
2797 use login_cap.h, so on its own it isn't the distinguishing feature. The
2798 new name refers directly to the setclassresources() function.
2800 PH/65 Added configuration files for NetBSD3.
2802 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
2804 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
2805 is preferred over IPv4.
2807 PH/68 The bounce_return_message and bounce_return_body options were not being
2808 honoured for bounces generated during the reception of non-SMTP messages.
2809 In particular, this applied to messages rejected by the ACL. This bug has
2810 been fixed. However, if bounce_return_message is true and bounce_return_
2811 body is false, the headers that are returned for a non-SMTP message
2812 include only those that have been read before the error was detected.
2813 (In the case of an ACL rejection, they have all been read.)
2815 PH/69 The HTML version of the specification is now built in a directory called
2816 spec_html instead of spec.html, because the latter looks like a path with
2817 a MIME-type, and this confuses some software.
2819 PH/70 Catch two compiler warnings in sieve.c.
2821 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
2822 function verify_get_ident() calls ip_connect() to connect a socket, but
2823 if the "connect()" function timed out, ip_connect() used to close the
2824 socket. However, verify_get_ident() also closes the socket later, and in
2825 between Exim writes to the log, which may get opened at this point. When
2826 the socket was closed in ip_connect(), the log could get the same file
2827 descriptor number as the socket. This naturally causes chaos. The fix is
2828 not to close the socket in ip_connect(); the socket should be closed by
2829 the function that creates it. There was only one place in the code where
2830 this was missing, in the iplookup router, which I don't think anybody now
2831 uses, but I've fixed it anyway.
2833 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
2834 well as to direct DNS lookups. Otherwise the handling of names in host
2835 lists is inconsistent and therefore confusing.
2841 PH/01 Two changes to the default runtime configuration:
2843 (1) Move the checks for relay_from_hosts and authenticated clients from
2844 after to before the (commented out) DNS black list checks.
2846 (2) Add control=submission to the relay_from_hosts and authenticated
2847 clients checks, on the grounds that messages accepted by these
2848 statements are most likely to be submissions.
2850 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
2852 (1) Generate an error if the third argument for the ${prvs expansion is
2855 (2) Treat a missing third argument of ${prvscheck as if it were an empty
2858 (3) Reset the variables that are obtained from the first argument of
2859 ${prvscheck and used in the second argument before leaving the code,
2860 because their memory is reclaimed, so using them afterwards may do
2863 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
2864 one (it's much easier than Tom thought :-).
2866 (5) Because of (4), we can now allow for the use of $prvscheck_result
2867 inside the third argument.
2869 PH/03 For some reason, the default setting of PATH when running a command from
2870 a pipe transport was just "/usr/bin". I have changed it to
2873 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
2874 anything to be listed in the output from -bV.
2876 PH/05 When a filter generated an autoreply, the entire To: header line was
2877 quoted in the delivery log line, like this:
2879 => >A.N.Other <ano@some.domain> <original@ddress> ...
2881 This has been changed so that it extracts the operative address. There
2882 may be more than one such address. If so, they are comma-separated, like
2885 => >ano@some.domain,ona@other.domain <original@ddress> ...
2887 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
2888 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
2889 not being looked up in the rDNS to get a host name, Exim was showing the
2890 IP address twice in Received: lines, even though the IP addresses were
2891 identical. For example:
2893 Received: from [1.2.3.4] (helo=[1.2.3.4])
2895 However, if the real host name was known, it was omitting the HELO data
2896 if it matched the actual IP address. This has been tidied up so that it
2897 doesn't show the same IP address twice.
2899 PH/07 When both +timestamp and +memory debugging was on, the value given by
2900 $tod_xxx expansions could be wrong, because the tod_stamp() function was
2901 called by the debug printing, thereby overwriting the timestamp buffer.
2902 Debugging no longer uses the tod_stamp() function when +timestamp is set.
2904 PH/08 When the original message was included in an autoreply transport, it
2905 always said "this is a copy of the message, including all the headers",
2906 even if body_only or headers_only was set. It now gives an appropriate
2909 PH/09 Applied a patch from the Sieve maintainer which:
2911 o fixes some comments
2912 o adds the (disabled) notify extension core
2913 o adds some debug output for the result of if/elsif tests
2914 o points to the current vacation draft in the documentation
2915 and documents the missing references header update
2919 o fixes a bug in processing the envelope test (when testing
2920 multiple envelope elements, the last element determinted the
2923 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
2924 Electronic Mail") by including:
2926 Auto-submitted: auto-generated
2928 in the messages that it generates (bounce messages and others, such as
2929 warnings). In the case of bounce messages for non-SMTP mesages, there was
2930 also a typo: it was using "Auto_submitted" (underscore instead of
2931 hyphen). Since every message generated by Exim is necessarily in response
2932 to another message, thes have all been changed to:
2934 Auto-Submitted: auto-replied
2936 in accordance with these statements in the RFC:
2938 The auto-replied keyword:
2940 - SHOULD be used on messages sent in direct response to another
2941 message by an automatic process,
2943 - MUST NOT be used on manually-generated messages,
2945 - MAY be used on Delivery Status Notifications (DSNs) and Message
2946 Disposition Notifications (MDNs),
2948 - MUST NOT be used on messages generated by automatic or periodic
2949 processes, except for messages which are automatic responses to
2952 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
2953 to the default Received: header definition.
2955 PH/12 Added log selector acl_warn_skipped (default on).
2957 PH/13 After a successful wildlsearch lookup, discard the values of numeric
2958 variables because (a) they are in the wrong storage pool and (b) even if
2959 they were copied, it wouldn't work properly because of the caching.
2961 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
2962 checking when decoding. Apparently there are clients that generate
2963 overlong encoded strings. Why am I not surprised?
2965 PH/15 If the first argument of "${if match_address" was not empty, but did not
2966 contain an "@" character, Exim crashed. Now it writes a panic log message
2967 and treats the condition as false.
2969 PH/16 In autoreply, treat an empty string for "once" the same as unset.
2971 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
2972 extension "envelope-auth". The code is finished and in agreement with
2973 other implementations, but there is no documentation so far and in fact,
2974 nobody wrote the draft yet. This extension is currently #undef'ed, thus
2975 not changing the active code.
2977 Print executed "if" and "elsif" statements when debugging is used. This
2978 helps a great deal to understand what a filter does.
2980 Document more things not specified clearly in RFC3028. I had all this
2981 sorted out, when out of a sudden new issues came to my mind. Oops."
2983 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
2986 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
2987 canonical form (as documented). However, after a host name lookup from
2988 the IP address, check_host() was doing a simple string comparison with
2989 addresses acquired from the DNS when checking that the found name did
2990 have the original IP as one of its addresses. Since any found IPv6
2991 addresses are likely to be in abbreviated form, the comparison could
2992 fail. Luckily, there already exists a function for doing the comparison
2993 by converting both addresses to binary, so now that is used instead of
2994 the text comparison.
2996 PH/20 There was another similar case to PH/19, when a complete host name was
2997 given in a host list; looking up its IP address could give an abbreviated
2998 form, whereas the current host's name might or might not be abbreviated.
2999 The same fix has been applied.
3005 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
3006 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
3009 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
3010 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
3012 PH/03 Typo: missing ".o" in src/pcre/Makefile.
3014 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
3015 header line, restrict the check to what is listed in RFCs 2369 and 2929.
3016 Also, for "Auto-Submitted", treat anything other than "no" as
3017 non-personal, in accordance with RFC 3834. (Previously it treated
3018 anything starting "auto-" as non-personal.)
3020 TF/01 The control=submission/name=... option had a problem with syntax
3021 errors if the name included a slash character. The /name= option
3022 now slurps the rest of the string, so it can include any characters
3023 but it must come last in the list of options (after /sender_retain
3026 PH/05 Some modifications to the interface to the fake nameserver for the new
3034 TK/01 Added the "success_on_redirect" address verification option. See
3035 NewStuff for rationale and an example.
3037 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
3039 PH/02 Patch to exigrep to allow it to work on syslog lines.
3041 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
3042 fread() to skip over the body file's header line, because in Cygwin the
3043 header line is locked and is inaccessible.
3045 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
3046 co-exist for some time) to make it clear that it is the Exim ID that is
3047 referenced, not the Message-ID: header line.
3049 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
3050 string_format() function, because snprintf() does not exist on all
3053 PH/06 The use of forbid_filter_existstest now also locks out the use of the
3054 ${stat: expansion item.
3056 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
3057 protocol synchronization error", to keep the pedants happy.
3059 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
3060 well as for IRIX systems, when gcc is being used. See the host.c source
3063 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
3065 PH/10 Named domain lists were not working if used in a queue_smtp_domains
3068 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
3069 transport and to the smtp transport in LMTP mode.
3071 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
3073 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
3074 run a filter in a subprocess. This could lead to confusion in subsequent
3075 lookups in the parent process. There should also be a search_tidyup() at
3076 the end of the subprocess.
3078 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
3079 only if the host matched helo_try_verify_hosts, which caused the
3080 verification to occur when the EHLO/HELO command was issued. The ACL just
3081 tested the remembered result. Now, if a previous verification attempt has
3082 not happened, "verify = helo" does it there and then.
3084 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
3086 TK/03 Fix log output including CR from clamd.
3088 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
3089 exist provoked a memory error which could cause a segfault.
3091 PH/15 Installed PCRE 6.2
3093 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
3095 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
3096 of the problem. Specifically, suggested +O2 rather than +O1 for the
3099 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
3101 PH/20 If a delivery was routed to a non-standard port by means of an SRV
3102 record, the port was not correctly logged when the outgoing_port log
3103 selector was set (it logged the transort's default port).
3105 PH/21 Added support for host-specific ports to manualroute, queryprogram,
3106 fallback_hosts, and "hosts" in the smtp transport.
3108 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
3109 host errors such as "Connection refused".
3111 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
3112 authentication with radiusclient 0.4.9:
3114 - Error returned from rc_read_config was caught wrongly
3115 - Username/password not passed on to radius server due to wrong length.
3117 The presumption is that some radiusclient API changes for 4.51/PH/17
3118 were not taken care of correctly. The code is still untested by me (my
3119 Linux distribution still has 0.3.2 of radiusclient), but it was
3120 contributed by a Radius user.
3122 PH/24 When doing a callout, the value of $domain wasn't set correctly when
3123 expanding the "port" option of the smtp transport.
3125 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
3126 while reading a MIME header. Thanks to Tom Hughes for a patch.
3128 PH/24 Include config.h inside local_scan.h so that configuration settings are
3131 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
3132 that in an ACL for RCPT (for example), you can examine exactly what was
3135 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
3136 commands, but it was not correctly comparing the address with the actual
3137 client host address. Thus, it would show the EHLO address in Received:
3138 header lines when this was not necessary.
3140 PH/27 Added the % operator to ${eval:}.
3142 PH/28 Exim tries to create and chdir to its spool directory when it starts;
3143 it should be ignoring failures (because with -C, for example, it has lost
3144 privilege). It wasn't ignoring creation failures other than "already
3147 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
3150 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
3151 because an input line was too long, either on its own, or by virtue of
3152 too many continuations, the temporary file was not being removed, and the
3153 return code was incorrect.
3155 PH/31 Missing "BOOL" in function definition in filtertest.c.
3157 PH/32 Applied Sieve patches from the maintainer.
3159 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
3161 PH/33 Added "verify = not_blind".
3163 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
3164 Local/Makefile (with some defaults set). These are used in built scripts
3165 such as exicyclog, but they have never been used in the exim_install
3166 script (though there are many overriding facilities there). I have
3167 arranged that the exim_install script now takes note of these two
3170 PH/35 Installed configuration files for Dragonfly.
3172 PH/36 When a locally submitted message by a trusted user did not contain a
3173 From: header, and the sender address was obtained from -f or from an SMTP
3174 MAIL command, and the trusted user did not use -F to supply a sender
3175 name, $originator_name was incorrectly used when constructing a From:
3176 header. Furthermore, $originator_name was used for submission mode
3177 messages from external hosts without From: headers in a similar way,
3178 which is clearly wrong.
3180 PH/37 Added control=suppress_local_fixups.
3182 PH/38 When log_selector = +received_sender was set, and the addition of the
3183 sender made the log line's construction buffer exactly full, or one byte
3184 less than full, an overflow happened when the terminating "\n" was
3187 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
3188 when the result of a list match is failure because a DNS lookup failed.
3190 PH/40 RM_COMMAND is now used in the building process.
3192 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
3193 the "build-* directories that it finds.
3195 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
3196 address in a domain literal was a prefix of an interface address.
3198 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
3199 when verifying a sender address, unless rewrite_headers is false.
3201 PH/44 Wrote a long comment about why errors_to addresses are verified as
3202 recipients, not senders.
3204 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
3205 the ratelimit ACL was added.
3207 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
3209 PH/46 Added extra information about PostgreSQL errors to the error string.
3211 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
3212 avoiding the need to install special zones in a real server. This is
3213 backwards compatible; if it can't find the fake resolver, it drops back.
3214 Thus, both old and new test suites can be run.
3216 TF/02 Added util/ratelimit.pl
3218 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
3219 clock is set back in time.
3221 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
3222 Candler <B.Candler@pobox.com>.
3224 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
3225 OK for addresses that are the result of redirections.
3227 PH/48 A number of further additions for the benefit of the new test suite,
3228 including a fake gethostbyname() that interfaces to the fake DNS resolver
3231 TF/06 The fix for widen_domains has also been applied to qualify_single and
3232 search_parents which are the other dnslookup options that can cause
3235 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
3238 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
3240 TF/07 Exim produced the error message "an SRV record indicated no SMTP
3241 service" if it encountered an MX record with an empty target hostname.
3242 The message is now "an MX or SRV record indicated no SMTP service".
3244 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
3245 if the DNS of the sending site is misconfigured. This is quite a
3246 common situation. This change restores the behaviour of treating a
3247 helo verification defer as a failure.
3249 PH/51 If self=fail was set on a router, the bounce message did not include the
3250 actual error message.
3256 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
3258 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
3259 command itself ended in error, the underlying message about the transport
3260 filter timeout was being overwritten with the pipe command error. Now the
3261 underlying error message should be appended to the second error message.
3263 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
3265 PH/02 Reduce the amount of output that "make" produces by default. Full output
3266 can still be requested.
3268 PH/03 The warning log line about a condition test deferring for a "warn" verb
3269 was being output only once per connection, rather than after each
3270 occurrence (because it was using the same function as for successful
3271 "warn" verbs). This seems wrong, so I have changed it.
3273 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
3274 it should not have, which might have caused a crash in the right
3275 circumstances, but probably never did.
3277 PH/04 Installed a modified version of Tony Finch's patch to make submission
3278 mode fix the return path as well as the Sender: header line, and to
3279 add a /name= option so that you can make the user's friendly name appear
3282 TF/03 Added the control = fakedefer ACL modifier.
3284 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
3285 Mark Lowes for thorough testing.
3287 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
3289 TK/03 Merged latest SRS patch from Miles Wilton.
3291 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
3292 with the definition in sysexits.h (which is #included earlier).
3293 Fortunately, Exim does not actually use EX_OK. The code used to try to
3294 preserve the sysexits.h value, by assumimg that macro definitions were
3295 scanned for macro replacements. I have been disabused of this notion,
3296 so now the code just undefines EX_OK before #including unistd.h.
3298 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
3299 in the smtp transport. When a block could not be written in a single
3300 write() function, the timeout was being re-applied to each part-write.
3301 This seems wrong - if the receiver was accepting one byte at a time it
3302 would take for ever. The timeout is now adjusted when this happens. It
3303 doesn't have to be particularly precise.
3305 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
3306 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
3308 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
3309 without a domain if the check to <postmaster@domain> fails.
3311 SC/01 Eximstats: added -xls and the ability to specify output files
3312 (patch written by Frank Heydlauf).
3314 SC/02 Eximstats: use FileHandles for outputing results.
3316 SC/03 Eximstats: allow any combination of xls, txt, and html output.
3318 SC/04 Eximstats: fixed display of large numbers with -nvr option
3320 SC/05 Eximstats: fixed merging of reports with empty tables.
3322 SC/06 Eximstats: added the -include_original_destination flag
3324 SC/07 Eximstats: removed tabs and trailing whitespace.
3326 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
3328 TK/06 MBOX spool code: Add real "From " MBOX separator line
3329 so the .eml file is really in mbox format (even though
3330 most programs do not really care). Patch from Alex Miller.
3332 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
3333 The latter is generated from $received_to and is only set if the
3334 message has one envelope recipient. SA can use these headers,
3335 obviously out-of-the-box. Patch from Alex Miller.
3337 PH/08 The ${def test on a variable was returning false if the variable's
3338 value was "0", contrary to what the specification has always said!
3339 The result should be true unless the variable is empty.
3341 PH/09 The syntax error of a character other than { following "${if
3342 def:variable_name" (after optional whitespace) was not being diagnosed.
3343 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
3344 accidental colon was present, for example, could give incorrect results.
3346 PH/10 Tidied the code in a number of places where the st_size field of a stat()
3347 result is used (not including appendfile, where other changes are about
3350 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
3351 This involved changing a lot of size variables from int to off_t. It
3352 should work with maildirs and everything.
3354 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
3355 spamd dying while we are connected to it.
3357 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
3360 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
3361 function for BDB 4.3.
3363 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
3365 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
3366 the sources for PCRE have been re-arranged and more files are now
3369 PH/15 The code I had for printing potentially long long variables in PH/11
3370 above was not the best (it lost precision). The length of off_t variables
3371 is now inspected at build time, and an appropriate printing format (%ld
3372 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
3373 to be "long long int" or "long int". This is needed for the internal
3374 formatting function string_vformat().
3376 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
3377 the configuration file to be ":syslog", then the script "guesses" where
3378 the logs files are, rather than using the compiled in default. In our
3379 case the guess is not the same as the compiled default, so the script
3380 suddenly stopped working when I started to use syslog. The patch checks
3381 to see if log_file_path is "". If so, it attempts to read it from exim
3382 with no configuration file to get the compiled in version, before it
3383 falls back to the previous guessing code."
3385 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
3386 implementing BATV in an Exim configuration. See NewStuff for the gory
3389 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
3390 Makefile that are specific to HP-UX.
3392 PH/18 If the "use_postmaster" option was set for a recipient callout together
3393 with the "random" option, the postmaster address was used as the MAIL
3394 FROM address for the random test, but not for the subsequent recipient
3395 test. It is now used for both.
3397 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
3398 patch removes a few documentation additions to RFC 3028, because the
3399 latest draft now contains them. It adds the new en;ascii-case comparator
3400 and a new error check for 8bit text in MIME parts. Comparator and
3401 require names are now matched exactly. I enabled the subaddress
3402 extension, but it is not well tested yet (read: it works for me)."
3404 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
3405 rework some of the code of TK/09 above to avoid the hardwired use of
3406 "%lld" and "long long". Replaced the call to snprintf() with a call to
3409 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
3410 records point to non-existent hosts", "retry timeout exceeded", and
3411 "retry time not reached for any host after a long failure period".
3413 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
3414 experimental DomainKeys support:
3416 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
3417 (2) On an error such as an illegally used "control", the wrong name for
3418 the control was given.
3420 These problems did NOT occur unless DomainKeys support was compiled.
3422 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
3424 PH/24 Added ${if match_ip condition.
3426 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
3427 databases so that it will be absolutely obvious if a crash occurs in the
3428 DB library. This is a regular occurrence (often caused by mis-matched
3431 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
3432 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
3433 picked up on a user's system that detects such things. There doesn't seem
3434 to be a gcc warning option for this - only an attribute that has to be
3435 put on the function's prototype. It seems that in Fedora Core 4 they have
3436 set this on a number of new functions. No doubt there will be more in due
3439 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
3440 specify a transport. However, if an address that was verified by such a
3441 router was the subject of a callout, Exim crashed because it tried to
3442 read the rcpt_include_affixes from the non-existent transport. Now it
3443 just assumes that the setting of that option is false. This bug was
3444 introduced by 4.51/PH/31.
3446 PH/28 Changed -d+all to exclude +memory, because that information is very
3447 rarely of interest, but it makes the output a lot bigger. People tend to
3448 do -d+all out of habit.
3450 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
3451 code in os-type was giving problems when libc.so lives in lib64, like on
3454 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
3455 aren't the modern standard, and it seems that some systems' include files
3456 don't always have them. Exim was already checking for some of the newer
3457 ones like T_AAAA, and defining it itself. I've added checks for all the
3458 record types that Exim uses.
3460 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
3461 not automatically generating a new one, as it is supposed to. This
3462 prevented TLS from working. If the file did exist, but contained invalid
3463 data, a new version was generated, as expected. It was only the case of a
3464 non-existent file that was broken.
3466 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
3467 with a change in libdomainkeys > 0.64.
3469 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
3470 from DNS. If the selector record carries the flag, it now has
3471 precedence over the domain-wide flag.
3473 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
3475 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
3476 the use of a port name that isn't defined in /etc/services) occurred, the
3477 message was deferred as in a normal delivery, and thus remained on the
3478 spool, instead of being failed because of the mua_wrapper setting. This
3479 is now fixed, and I tidied up some of the mua_wrapper messages at the
3482 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
3483 the messages in a hash of arrays rather than using individual hashes.
3484 This is a bit cleaner and results in dramatic memory savings, albeit
3485 at a slight CPU cost.
3487 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
3488 as requested by Marc Sherman.
3490 SC/10 Eximstats: added histograms for user specified patterns as requested
3493 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
3495 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
3496 fopen() in the content-scanning modules that did not already have it.
3502 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
3503 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
3505 TK/02 Fix ACL "control" statement not being available in MIME ACL.
3507 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
3509 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
3510 to test Sieve filters that use "vacation".
3512 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
3513 that changes the way the GnuTLS parameters are stored in the cache file.
3514 The new format can be generated externally. For backward compatibility,
3515 if the data in the cache doesn't make sense, Exim assumes it has read an
3516 old-format file, and it generates new data and writes a new file. This
3517 means that you can't go back to an older release without removing the
3520 PH/03 A redirect router that has both "unseen" and "one_time" set does not
3521 work if there are any delivery delays because "one_time" forces the
3522 parent to be marked "delivered", so its unseen clone is never tried
3523 again. For this reason, Exim now forbids the simultaneous setting of
3526 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
3527 redirected to themselves ("homonym" addresses). Read the long ChangeLog
3528 entry if you want to know the details. The fix, however, neglected to
3529 consider the case when local delivery batching is involved. The test for
3530 "previously delivered" was not happening when checking to see if an
3531 address could be batched with a previous (undelivered) one; under
3532 certain circumstances this could lead to multiple deliveries to the same
3535 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
3536 in its include files, and this causes problems building Exim.
3538 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
3539 header_syntax) but Exim was just ignoring anything given after a slash.
3540 In particular, this caused confusion with an attempt to use "verify =
3541 reverse_host_lookup/defer_ok". An error is now given when options are
3542 supplied for verify items that do not have them. (Maybe reverse_host_
3543 lookup should have a defer_ok option, but that's a different point.)
3545 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
3546 defined by RFC 821) to 2048, because there were problems with some AUTH
3547 commands, and RFC 1869 says the size should be increased for extended
3548 SMTP commands that take arguments.
3550 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
3553 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
3554 "unknown" error; now it says that the functionality isn't in the binary.
3556 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
3557 an address' error message when a string expansion fails (syntax or
3558 whatever). Otherwise the password may appear in the log. Following change
3559 PH/42 below, there is no longer a chance of it appearing in a bounce
3562 PH/11 Installed exipick version 20050225.0 from John Jetmore.
3564 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
3565 of its addresses was ever tried. (Bugzilla bug #2.)
3567 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
3568 the result incorrectly in the debug output. (It correctly added a newline
3569 to what was transported.)
3571 TF/01 Added $received_time.
3573 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
3574 commented out examples of how to interface to a virus scanner and to
3575 SpamAssassin. Also added commented examples of av_scanner and
3576 spamd_address settings.
3578 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
3579 and controls are allowed in which ACLs. There were a couple of minor
3580 errors. Some of the entries in the conditions table (which is a table of
3581 where they are NOT allowed) were getting very unwieldy; rewrote them as a
3582 negation of where the condition IS allowed.
3584 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
3586 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
3587 header file does not have a version number, so I've had to invent a new
3588 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
3589 API. The code is untested by me (my Linux distribution still has 0.3.2 of
3590 radiusclient), but it was contributed by a Radius user.
3592 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
3593 files or directories, for OpenSSL.
3595 PH/19 When an Exim process that is running as root has to create an Exim log
3596 file, it does so in a subprocess that runs as exim:exim so as to get the
3597 ownership right at creation (otherwise, other Exim processes might see
3598 the file with the wrong ownership). There was no test for failure of this
3599 fork() call, which would lead to the process getting stuck as it waited
3600 for a non-existent subprocess. Forks do occasionally fail when resources
3601 run out. I reviewed all the other calls to fork(); they all seem to check
3604 PH/20 When checking for unexpected SMTP input at connect time (before writing
3605 the banner), Exim was not dealing correctly with a non-positive return
3606 from the read() function. If the client had disconnected by this time,
3607 the result was a log entry for a synchronization error with an empty
3608 string after "input=" when read() returned zero. If read() returned -1
3609 (an event I could not check), uninitialized data bytes were printed.
3610 There were reports of junk text (parts of files, etc) appearing after
3613 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
3615 PH/22 Added support for macro redefinition, and (re)definition in between
3616 driver and ACL definitions.
3618 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
3619 forgetting to use the resulting value; it was using the unexpanded value.
3621 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
3622 hadn't been configured. The fix is from Juergen Kreileder, who
3623 understands it better than I do:
3625 "Here's what I see happening with three configured cyrus_sasl
3626 authenticators configured (plain, login, cram-md5):
3628 On startup auth_cyrus_sasl_init() gets called for each of these.
3629 This means three calls to sasl_listmech() without a specified mech_list.
3630 => SASL tests which mechs of all available mechs actually work
3631 => three warnings about OTP not working
3632 => the returned list contains: plain, login, cram-md5, digest-md5, ...
3634 With the patch, sasl_listmech() also gets called three times. But now
3635 SASL's mech_list option is set to the server_mech specified in the the
3636 authenticator. Or in other words, the answer from sasl_listmech()
3637 gets limited to just the mech you're testing for (which is different
3639 => the return list contains just 'plain' or 'login', 'cram-md5' or
3640 nothing depending on the value of ob->server_mech.
3642 I've just tested the patch: Authentication still works fine,
3643 unavailable mechs specified in the exim configuration are still
3644 caught, and the auth.log warnings about OTP are gone."
3646 PH/25 When debugging is enabled, the contents of the command line are added
3647 to the debugging output, even when log_selector=+arguments is not
3650 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
3651 answer is "GNU", and only if the return is "GNU/something" is the answer
3654 PH/27 $acl_verify_message is now set immediately after the failure of a
3655 verification in an ACL, and so is available in subsequent modifiers. In
3656 particular, the message can be preserved by coding like this:
3658 warn !verify = sender
3659 set acl_m0 = $acl_verify_message
3661 Previously, $acl_verify_message was set only while expanding "message"
3662 and "log_message" when a very denied access.
3664 PH/28 Modified OS/os.c-Linux with
3666 -#ifndef OS_LOAD_AVERAGE
3667 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
3669 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
3670 nomenclature these days.)
3672 PH/29 Installed patch from the Sieve maintainer that adds the options
3673 sieve_useraddress and sieve_subaddress to the redirect router.
3675 PH/30 In these circumstances:
3676 . Two addresses routed to the same list of hosts;
3677 . First host does not offer TLS;
3678 . First host accepts first address;
3679 . First host gives temporary error to second address;
3680 . Second host offers TLS and a TLS session is established;
3681 . Second host accepts second address.
3682 Exim incorrectly logged both deliveries with the TLS parameters (cipher
3683 and peerdn, if requested) that were in fact used only for the second
3686 PH/31 When doing a callout as part of verifying an address, Exim was not paying
3687 attention to any local part prefix or suffix that was matched by the
3688 router that accepted the address. It now behaves in the same way as it
3689 does for delivery: the affixes are removed from the local part unless
3690 rcpt_include_affixes is set on the transport.
3692 PH/32 Add the sender address, as F=<...>, to the log line when logging a
3693 timeout during the DATA phase of an incoming message.
3695 PH/33 Sieve envelope tests were broken for match types other than :is. I have
3696 applied a patch sanctioned by the Sieve maintainer.
3698 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
3699 the uid or gid is negative. A case of a negative gid caused this to be
3700 noticed. The fix allows for either to be negative.
3702 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
3703 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
3706 PH/36 The change PH/12 above was broken. Fixed it.
3708 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
3709 the grounds that routing the same address twice would always produce the
3710 same answer. This might have been true once, but it is certainly no
3711 longer true now. Routing a child address may depend on the previous
3712 routing that produced that child. Some complicated redirection strategies
3713 went wrong when messages had multiple recipients, and made Exim's
3714 behaviour dependent on the order in which the addresses were given.
3716 I have moved the duplicate checking until after the routing is complete.
3717 Exim scans the addresses that are assigned to local and remote
3718 transports, and removes any duplicates. This means that more work will be
3719 done, as duplicates will always all be routed, but duplicates are
3720 presumably rare, so I don't expect this is of any significance.
3722 For deliveries to pipes, files, and autoreplies, the duplicate checking
3723 still happens during the routing process, since they are not going to be
3726 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
3727 It corrects a timeout issue with spamd. This is Ian's comment: "The
3728 background is that sometimes spamd either never reads data from a
3729 connection it has accepted, or it never writes response data. The exiscan
3730 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
3731 blindly assumes that writes won't block so it may never time out."
3733 PH/39 Allow G after quota size as well as K and M.
3735 PH/40 The value set for $authenticated_id in an authenticator may not contain
3736 binary zeroes or newlines because the value is written to log lines and
3737 to spool files. There was no check on this. Now the value is run through
3738 the string_printing() function so that such characters are converted to
3739 printable escape sequences.
3741 PH/41 $message_linecount is a new variable that contains the total number of
3742 lines in the message. Compare $body_linecount, which is the count for the
3745 PH/42 Exim no longer gives details of delivery errors for specific addresses in
3746 bounce and delay warning messages, except in certain special cases, which
3749 (a) An SMTP error message from a remote host;
3750 (b) A message specified in a :fail: redirection;
3751 (c) A message specified in a "fail" command in a system filter;
3752 (d) A message specified in a FAIL return from the queryprogram router;
3753 (e) A message specified by the cannot_route_message router option.
3755 In these cases only, Exim does include the error details in bounce and
3756 warning messages. There are also a few cases where bland messages such
3757 as "unrouteable address" or "local delivery error" are given.
3759 PH/43 $value is now also set for the "else" part of a ${run expansion.
3761 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
3762 being worked on, but at least Exim now implements the latest version to
3765 PH/45 In a pipe transport, although a timeout while waiting for the pipe
3766 process to complete was treated as a delivery failure, a timeout while
3767 writing the message to the pipe was logged, but erroneously treated as a
3768 successful delivery. Such timeouts include transport filter timeouts. For
3769 consistency with the overall process timeout, these timeouts are now
3770 treated as errors, giving rise to delivery failures by default. However,
3771 there is now a new Boolean option for the pipe transport called
3772 timeout_defer, which, if set TRUE, converts the failures into defers for
3773 both kinds of timeout. A transport filter timeout is now identified in
3776 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
3777 systems where "make" and "gmake" are different, calling "gmake" at top
3778 level broke things. I've arranged for the value of $(MAKE) to be passed
3779 from the Makefile to this script so that it can call the same version of
3783 A note about Exim versions 4.44 and 4.50
3784 ----------------------------------------
3786 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
3787 changes of various kinds. As a consequence, a big documentation update was
3788 needed. This delayed the release for rather longer than seemed good, especially
3789 in the light of a couple of (minor) security issues. Therefore, the changes
3790 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
3791 release. So 4.44 and 4.50 are in effect two different branches that both start
3794 I have left the 4.50 change log unchanged; it contains all the changes since
3795 4.43. The change log for 4.44 is below; many of its items are identical to
3796 those for 4.50. This seems to be the most sensible way to preserve the
3797 historical information.
3803 1. Minor wording change to the doc/README.SIEVE file.
3805 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
3806 computation of the current number of files was incorrect.
3808 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
3809 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
3812 4. Give more explanation in the error message when the command for a transport
3813 filter fails to execute.
3815 5. There are several places where Exim runs a non-Exim command in a
3816 subprocess. The SIGUSR1 signal should be disabled for these processes. This
3817 was being done only for the command run by the queryprogram router. It is
3818 now done for all such subprocesses. The other cases are: ${run, transport
3819 filters, and the commands run by the lmtp and pipe transports.
3821 6. Added CONFIGURE_GROUP build-time option.
3823 7. Some older OS have a limit of 256 on the maximum number of file
3824 descriptors. Exim was using setrlimit() to set 1000 as a large value
3825 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
3826 systems. I've change it so that if it can't get 1000, it tries for 256.
3828 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
3829 was an oversight, and furthermore, ever since the addition of extra
3830 controls (e.g. 4.43/32), the checks on when to allow different forms of
3831 "control" were broken. There should now be diagnostics for all cases when a
3832 control that does not make sense is encountered.
3834 9. Added the /retain_sender option to "control=submission".
3836 10. $recipients is now available in the predata ACL (oversight).
3838 11. Tidy the search cache before the fork to do a delivery from a message
3839 received from the command line. Otherwise the child will trigger a lookup
3840 failure and thereby defer the delivery if it tries to use (for example) a
3841 cached ldap connection that the parent has called unbind on.
3843 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
3844 of $address_data from the recipient verification was clobbered by the
3845 sender verification.
3847 13. The value of address_data from a sender verification is now available in
3848 $sender_address_data in subsequent conditions in the ACL statement.
3850 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
3852 15. Added a new option "connect=<time>" to callout options, to set a different
3855 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
3856 was its contents. (It was OK if the option was not defined at all.)
3858 17. A "Completed" log line is now written for messages that are removed from
3859 the spool by the -Mrm option.
3861 18. New variables $sender_verify_failure and $recipient_verify_failure contain
3862 information about exactly what failed.
3864 19. Added -dd to debug only the daemon process.
3866 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
3867 handles timeouts, both on the server side and network timeouts. Renamed the
3868 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
3870 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
3871 It is now set to "smtps".
3873 22. $host_address is now set to the target address during the checking of
3874 ignore_target_hosts.
3876 23. When checking ignore_target_hosts for an ipliteral router, no host name was
3877 being passed; this would have caused $sender_host_name to have been used if
3878 matching the list had actually called for a host name (not very likely,
3879 since this list is usually IP addresses). A host name is now passed as
3882 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
3883 code that specifies a non-restarting handler (typically sigaction() in
3884 modern systems) in an attempt to fix a rare and obscure crash bug.
3886 25. Narrowed the window for a race in the daemon that could cause it to ignore
3887 SIGCHLD signals. This is not a major problem, because they are used only to
3888 wake it up if nothing else does.
3890 26. A malformed maildirsize file could cause Exim to calculate negative values
3891 for the mailbox size or file count. Odd effects could occur as a result.
3892 The maildirsize information is now recalculated if the size or filecount
3895 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
3896 support for a long time. Removed HAVE_SYS_VFS_H.
3898 28. Installed the latest version of exipick from John Jetmore.
3900 29. In an address list, if the pattern was not a regular expression, an empty
3901 subject address (from a bounce message) matched only if the pattern was an
3902 empty string. Non-empty patterns were not even tested. This was the wrong
3903 because it is perfectly reasonable to use an empty address as part of a
3904 database query. An empty address is now tested by patterns that are
3905 lookups. However, all the other forms of pattern expect the subject to
3906 contain a local part and a domain, and therefore, for them, an empty
3907 address still always fails if the pattern is not itself empty.
3909 30. Exim went into a mad DNS loop when attempting to do a callout where the
3910 host was specified on an smtp transport, and looking it up yielded more
3911 than one IP address.
3913 31. Re-factored the code for checking spool and log partition space into a
3914 function that finds that data and another that does the check. The former
3915 is then used to implement four new variables: $spool_space, $log_space,
3916 $spool_inodes, and $log_inodes.
3918 32. The RFC2047 encoding function was originally intended for short strings
3919 such as real names; it was not keeping to the 75-character limit for
3920 encoded words that the RFC imposes. It now respects the limit, and
3921 generates multiple encoded words if necessary. To be on the safe side, I
3922 have increased the buffer size for the ${rfc2047: expansion operator from
3925 33. It is now permitted to omit both strings after an "if" condition; if the
3926 condition is true, the result is "true". As before, when the second string
3927 is omitted, a false condition yields an empty string. This makes it less
3928 cumbersome to write custom ACL and router conditions.
3930 34. Failure to deliver a bounce message always caused it to be frozen, even if
3931 there was an errors_to setting on the router. The errors_to setting is now
3934 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
3935 canonical form (fully expanded) before being placed in
3936 $sender_host_address.
3938 36. The table in the code that translates DNS record types into text (T_A to
3939 "A" for instance) was missing entries for NS and CNAME. It is just possible
3940 that this could have caused confusion if both these types were looked up
3941 for the same domain, because the text type is used as part of Exim's
3942 per-process caching. But the chance of anyone hitting this buglet seems
3945 37. The dnsdb lookup has been extended in a number of ways.
3947 (1) There is a new type, "zns", which walks up the domain tree until it
3948 finds some nameserver records. It should be used with care.
3950 (2) There is a new type, "mxh", which is like "mx" except that it returns
3951 just the host names, not the priorities.
3953 (3) It is now possible to give a list of domains (or IP addresses) to be
3954 looked up. The behaviour when one of the lookups defers can be
3955 controlled by a keyword.
3957 (4) It is now possible to specify the separator character for use when
3958 multiple records are returned.
3960 38. The dnslists ACL condition has been extended: it is now possible to supply
3961 a list of IP addresses and/or domains to be looked up in a particular DNS
3964 39. Added log_selector=+queue_time_overall.
3966 40. When running the queue in the test harness, wait just a tad after forking a
3967 delivery process, to get repeatability of debugging output.
3969 41. Include certificate and key file names in error message when GnuTLS fails
3970 to set them up, because the GnuTLS error message doesn't include the name
3971 of the failing file when there is a problem reading it.
3973 42. Allow both -bf and -bF in the same test run.
3975 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
3977 44. The "Exiscan patch" is now merged into the mainline Exim source.
3979 45. Sometimes the final signoff response after QUIT could fail to get
3980 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
3981 < 0 before doing a fflush(). This bug looks as though it goes back to the
3982 introduction of TLS in release 3.20, but "sometimes" must have been rare
3983 because the tests only now provoked it.
3985 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
3986 (this can affect the format of dates).
3988 47. exim_tidydb, when checking for the continued existence of a message for
3989 which it has found a message-specific retry record, was not finding
3990 messages that were in split spool directories. Consequently, it was
3991 deleting retry records that should have stayed in existence.
3993 48. Steve fixed some bugs in eximstats.
3995 49. The SPA authentication driver was not abandoning authentication and moving
3996 on to the next authenticator when an expansion was forced to fail,
3997 contradicting the general specification for all authenticators. Instead it
3998 was generating a temporary error. It now behaves as specified.
4000 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
4001 (the order specifies the preference for clients). The order is now AES256,
4002 AES128, 3DES, ARCFOUR128.
4004 51. Small patch to Sieve code - explicitly set From: when generating an
4007 52. Exim crashed if a remote delivery caused a very long error message to be
4008 recorded - for instance if somebody sent an entire SpamAssassin report back
4009 as a large number of 550 error lines. This bug was coincidentally fixed by
4010 increasing the size of one of Exim's internal buffers (big_buffer) that
4011 happened as part of the Exiscan merge. However, to be on the safe side, I
4012 have made the code more robust (and fixed the comments that describe what
4015 53. Now that there can be additional text after "Completed" in log lines (if
4016 the queue_time_overall log selector is set), a one-byte patch to exigrep
4017 was needed to allow it to recognize "Completed" as not the last thing in
4020 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
4021 patch that reportedly fixes this has been added. I am not expert enough to
4022 create a test for it. This is what the patch creator wrote:
4024 "I found a little strange behaviour of ldap code when working with
4025 Windows 2003 AD Domain, where users was placed in more than one
4026 Organization Units. When I tried to give exim partial DN, the exit code
4027 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
4028 But simultaneously result of request was absolutely normal ldap result,
4029 so I produce this patch..."
4031 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
4032 REFERENCE, so I have modified the code to exclude the patch when that macro
4035 55. Some experimental protocols are using DNS PTR records for new purposes. The
4036 keys for these records are domain names, not reversed IP addresses. The
4037 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4038 leaves it alone. Component reversal etc. now happens only for IP addresses.
4041 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
4043 57. Double the size of the debug message buffer (to 2048) so that more of very
4044 long debug lines gets shown.
4046 58. The exicyclog utility now does better if the number of log files to keep
4047 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
4049 59. Two changes related to the smtp_active_hostname option:
4051 (1) $smtp_active_hostname is now available as a variable.
4052 (2) The default for smtp_banner uses $smtp_active_hostname instead
4053 of $primary_hostname.
4055 60. The host_aton() function is supposed to be passed a string that is known
4056 to be a valid IP address. However, in the case of IPv6 addresses, it was
4057 not checking this. This is a hostage to fortune. Exim now panics and dies
4058 if the condition is not met. A case was found where this could be provoked
4059 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4060 components; fortuitously, this particular loophole had already been fixed
4061 by change 4.50/55 above.
4063 If there are any other similar loopholes, the new check in host_aton()
4064 itself should stop them being exploited. The report I received stated that
4065 data on the command line could provoke the exploit when Exim was running as
4066 exim, but did not say which command line option was involved. All I could
4067 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4068 running as the user.
4071 61. There was a buffer overflow vulnerability in the SPA authentication code
4072 (which came originally from the Samba project). I have added a test to the
4073 spa_base64_to_bits() function which I hope fixes it.
4076 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
4077 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
4079 63. The daemon start-up calls getloadavg() while still root for those OS that
4080 need the first call to be done as root, but it missed one case: when
4081 deliver_queue_load_max is set with deliver_drop_privilege. This is
4082 necessary for the benefit of the queue runner, because there is no re-exec
4083 when deliver_drop_privilege is set.
4085 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
4086 This has been fixed.
4088 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4089 was in use, was not putting the data itself into the right store pool;
4090 consequently, it could be overwritten for a subsequent message in the same
4091 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4094 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
4096 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
4097 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
4098 the function were treating the return as a boolean value, which happened to
4099 work because 0=false and not-0=true, but is not correct code.
4101 68. The host_aton() function was not handling scoped IPv6 addresses (those
4102 with, for example, "%eth0" on the end) correctly.
4104 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
4105 negated items (that is, ~something) in unsigned ints. Some compilers
4106 apparently mutter when there is no cast.
4108 70. If an address verification called from an ACL failed, and did not produce a
4109 user-specific message (i.e. there was only a "system" message), nothing was
4110 put in $acl_verify_message. In this situation, it now puts the system
4113 71. Change 4.23/11 added synchronization checking at the start of an SMTP
4114 session; change 4.31/43 added the unwanted input to the log line - except
4115 that it did not do this in the start of session case. It now does.
4117 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
4118 This is wrong and can cause the other end to generate a synchronization
4119 error if it is another Exim or anything else that does the synchronization
4120 check. A QUIT command is no longer sent after a timeout.
4122 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
4123 during host lookups.
4125 74. The defer_ok option of callout verification was not working if it was used
4126 when verifying addresses in header lines, that is, for this case:
4128 verify = header_sender/callout=defer_ok
4130 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
4131 those file descriptors could be used for SMTP connections. If anything
4132 wrote to stderr (the example that came up was "warn" in embedded Perl), it
4133 could be sent to the SMTP client, causing chaos. The daemon now opens
4134 stdin, stdout, and stderr to /dev/null when it puts itself into the
4137 76. Arrange for output from Perl's "warn" command to be written to Exim's main
4138 log by default. The user can override this with suitable Perl magic.
4140 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
4141 the log message when discard triggers, was not working for the DATA ACL or
4142 for the non-SMTP ACL.
4144 78. Error message wording change in sieve.c.
4146 79. If smtp_accept_max_per_host was set, the number of connections could be
4147 restricted to fewer than expected, because the daemon was trying to set up
4148 a new connection before checking whether the processes handling previous
4149 connections had finished. The check for completed processes is now done
4150 earlier. On busy systems, this bug wouldn't be noticed because something
4151 else would have woken the daemon, and it would have reaped the completed
4154 80. If a message was submitted locally by a user whose login name contained one
4155 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
4156 It caused a spool format error. I have fixed the spool reading code. A
4157 related problem was that the "from" clause in the Received: line became
4158 illegal because of the space(s). It is now covered by ${quote_local_part.
4160 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
4163 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
4165 83. Patch from Sieve maintainer to fix unterminated string problem in
4166 "vacation" handling.
4168 84. Some minor changes to the Linux configuration files to help with other
4169 OS variants using glibc.
4171 85. One more patch for Sieve to update vacation handling to latest spec.
4174 ----------------------------------------------------
4175 See the note above about the 4.44 and 4.50 releases.
4176 ----------------------------------------------------
4182 1. Change 4.43/35 introduced a bug that caused file counts to be
4183 incorrectly computed when quota_filecount was set in an appendfile
4186 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4187 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4190 3. Give more explanation in the error message when the command for a transport
4191 filter fails to execute.
4193 4. There are several places where Exim runs a non-Exim command in a
4194 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4195 was being done only for the command run by the queryprogram router. It is
4196 now done for all such subprocesses. The other cases are: ${run, transport
4197 filters, and the commands run by the lmtp and pipe transports.
4199 5. Some older OS have a limit of 256 on the maximum number of file
4200 descriptors. Exim was using setrlimit() to set 1000 as a large value
4201 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4202 systems. I've change it so that if it can't get 1000, it tries for 256.
4204 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4205 was an oversight, and furthermore, ever since the addition of extra
4206 controls (e.g. 4.43/32), the checks on when to allow different forms of
4207 "control" were broken. There should now be diagnostics for all cases when a
4208 control that does not make sense is encountered.
4210 7. $recipients is now available in the predata ACL (oversight).
4212 8. Tidy the search cache before the fork to do a delivery from a message
4213 received from the command line. Otherwise the child will trigger a lookup
4214 failure and thereby defer the delivery if it tries to use (for example) a
4215 cached ldap connection that the parent has called unbind on.
4217 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4218 of $address_data from the recipient verification was clobbered by the
4219 sender verification.
4221 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4222 was its contents. (It was OK if the option was not defined at all.)
4224 11. A "Completed" log line is now written for messages that are removed from
4225 the spool by the -Mrm option.
4227 12. $host_address is now set to the target address during the checking of
4228 ignore_target_hosts.
4230 13. When checking ignore_target_hosts for an ipliteral router, no host name was
4231 being passed; this would have caused $sender_host_name to have been used if
4232 matching the list had actually called for a host name (not very likely,
4233 since this list is usually IP addresses). A host name is now passed as
4236 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4237 code that specifies a non-restarting handler (typically sigaction() in
4238 modern systems) in an attempt to fix a rare and obscure crash bug.
4240 15. Narrowed the window for a race in the daemon that could cause it to ignore
4241 SIGCHLD signals. This is not a major problem, because they are used only to
4242 wake it up if nothing else does.
4244 16. A malformed maildirsize file could cause Exim to calculate negative values
4245 for the mailbox size or file count. Odd effects could occur as a result.
4246 The maildirsize information is now recalculated if the size or filecount
4249 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4250 support for a long time. Removed HAVE_SYS_VFS_H.
4252 18. Updated exipick to current release from John Jetmore.
4254 19. Allow an empty sender to be matched against a lookup in an address list.
4255 Previously the only cases considered were a regular expression, or an
4258 20. Exim went into a mad DNS lookup loop when doing a callout where the
4259 host was specified on the transport, if the DNS lookup yielded more than
4262 21. The RFC2047 encoding function was originally intended for short strings
4263 such as real names; it was not keeping to the 75-character limit for
4264 encoded words that the RFC imposes. It now respects the limit, and
4265 generates multiple encoded words if necessary. To be on the safe side, I
4266 have increased the buffer size for the ${rfc2047: expansion operator from
4269 22. Failure to deliver a bounce message always caused it to be frozen, even if
4270 there was an errors_to setting on the router. The errors_to setting is now
4273 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4274 canonical form (fully expanded) before being placed in
4275 $sender_host_address.
4277 24. Updated eximstats to version 1.33
4279 25. Include certificate and key file names in error message when GnuTLS fails
4280 to set them up, because the GnuTLS error message doesn't include the name
4281 of the failing file when there is a problem reading it.
4283 26. Expand error message when OpenSSL has problems setting up cert/key files.
4286 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
4287 (this can affect the format of dates).
4289 28. exim_tidydb, when checking for the continued existence of a message for
4290 which it has found a message-specific retry record, was not finding
4291 messages that were in split spool directories. Consequently, it was
4292 deleting retry records that should have stayed in existence.
4294 29. eximstats updated to version 1.35
4295 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
4296 1.35 - bugfix such that pie charts by volume are generated correctly
4298 30. The SPA authentication driver was not abandoning authentication and moving
4299 on to the next authenticator when an expansion was forced to fail,
4300 contradicting the general specification for all authenticators. Instead it
4301 was generating a temporary error. It now behaves as specified.
4303 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
4304 (the order specifies the preference for clients). The order is now AES256,
4305 AES128, 3DES, ARCFOUR128.
4307 31. Small patch to Sieve code - explicitly set From: when generating an
4310 32. Exim crashed if a remote delivery caused a very long error message to be
4311 recorded - for instance if somebody sent an entire SpamAssassin report back
4312 as a large number of 550 error lines. This bug was coincidentally fixed by
4313 increasing the size of one of Exim's internal buffers (big_buffer) that
4314 happened as part of the Exiscan merge. However, to be on the safe side, I
4315 have made the code more robust (and fixed the comments that describe what
4318 33. Some experimental protocols are using DNS PTR records for new purposes. The
4319 keys for these records are domain names, not reversed IP addresses. The
4320 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4321 leaves it alone. Component reversal etc. now happens only for IP addresses.
4324 34. The host_aton() function is supposed to be passed a string that is known
4325 to be a valid IP address. However, in the case of IPv6 addresses, it was
4326 not checking this. This is a hostage to fortune. Exim now panics and dies
4327 if the condition is not met. A case was found where this could be provoked
4328 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4329 components; fortuitously, this particular loophole had already been fixed
4330 by change 4.50/55 or 4.44/33 above.
4332 If there are any other similar loopholes, the new check in host_aton()
4333 itself should stop them being exploited. The report I received stated that
4334 data on the command line could provoke the exploit when Exim was running as
4335 exim, but did not say which command line option was involved. All I could
4336 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4337 running as the user.
4340 35. There was a buffer overflow vulnerability in the SPA authentication code
4341 (which came originally from the Samba project). I have added a test to the
4342 spa_base64_to_bits() function which I hope fixes it.
4345 36. The daemon start-up calls getloadavg() while still root for those OS that
4346 need the first call to be done as root, but it missed one case: when
4347 deliver_queue_load_max is set with deliver_drop_privilege. This is
4348 necessary for the benefit of the queue runner, because there is no re-exec
4349 when deliver_drop_privilege is set.
4351 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4352 was in use, was not putting the data itself into the right store pool;
4353 consequently, it could be overwritten for a subsequent message in the same
4354 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4357 38. Sometimes the final signoff response after QUIT could fail to get
4358 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4359 < 0 before doing a fflush(). This bug looks as though it goes back to the
4360 introduction of TLS in release 3.20, but "sometimes" must have been rare
4361 because the tests only now provoked it.
4367 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
4368 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
4369 Now it returns an integer. A number of calls were still expecting a T/F
4370 return. Fortuitously, in all cases, the tests worked in OK situations,
4371 which is the norm. However, things would have gone wrong on any write
4372 failures on the smtp file descriptor. This function is used when sending
4373 messages over SMTP and also when doing verify callouts.
4375 2. When Exim is called to do synchronous delivery of a locally submitted
4376 message (the -odf or -odi options), it no longer closes stderr before doing
4379 3. Implemented the mua_wrapper option.
4381 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
4383 5. Implemented the functions header_remove(), header_testname(),
4384 header_add_at_position(), and receive_remove_recipient(), and exported them
4387 6. If an ACL "warn" statement specified the addition of headers, Exim already
4388 inserted X-ACL-Warn: at the start if there was no header name. However, it
4389 was not making this test for the second and subsequent header lines if
4390 there were newlines in the string. This meant that an invalid header could
4391 be inserted if Exim was badly configured.
4393 7. Allow an ACL "warn" statement to add header lines at the start or after all
4394 the Received: headers, as well as at the end.
4396 8. Added the rcpt_4xx retry error code.
4398 9. Added postmaster_mailfrom=xxx to callout verification option.
4400 10. Added mailfrom=xxxx to the callout verification option, for verify=
4403 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
4404 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
4406 12. Inserted some casts to stop certain compilers warning when using pointer
4407 differences as field lengths or precisions in printf-type calls (mostly
4408 affecting debugging statements).
4410 13. Added optional readline() support for -be (dynamically loaded).
4412 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
4413 same clock tick as a message's arrival, so that its received time was the
4414 same as the "first fail" time on the retry record, and that message
4415 remained on the queue past the ultimate address timeout, every queue runner
4416 would try a delivery (because it was past the ultimate address timeout) but
4417 after another failure, the ultimate address timeout, which should have then
4418 bounced the address, did not kick in. This was a "< instead of <=" error;
4419 in most cases the first failure would have been in the next clock tick
4420 after the received time, and all would be well.
4422 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
4423 being recognized when the domain list was tested by the match_domain
4424 condition in an expansion string.
4426 16. Added the ${str2b64: operator.
4428 17. Exim was always calling setrlimit() to set a large limit for the number of
4429 processes, without checking whether the existing limit was already
4430 adequate. (It did check for the limit on file descriptors.) Furthermore,
4431 errors from getrlimit() and setrlimit() were being ignored. Now they are
4432 logged to the main and panic logs, but Exim does carry on, to try to do its
4433 job under whatever limits there are.
4435 18. Imported PCRE 5.0.
4437 19. Trivial typo in log message " temporarily refused connection" (the leading
4440 20. If the log selector return_path_on_delivery was set and an address was
4441 redirected to /dev/null, the delivery process crashed because it assumed
4442 that a return path would always be set for a "successful" delivery. In this
4443 case, the whole delivery is bypassed as an optimization, and therefore no
4446 21. Internal re-arrangement: the function for sending a challenge and reading
4447 a response while authentication was assuming a zero-terminated challenge
4448 string. It's now changed to take a pointer and a length, to allow for
4449 binary data in such strings.
4451 22. Added the cyrus_sasl authenticator (code supplied by MBM).
4453 23. Exim was not respecting finduser_retries when seeking the login of the
4454 uid under which it was called; it was always trying 10 times. (The default
4455 setting of finduser_retries is zero.) Also, it was sleeping after the final
4456 failure, which is pointless.
4458 24. Implemented tls_on_connect_ports.
4460 25. Implemented acl_smtp_predata.
4462 26. If the domain in control=submission is set empty, Exim assumes that the
4463 authenticated id is a complete email address when it generates From: or
4464 Sender: header lines.
4466 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
4467 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
4468 chown and chgrp in /bin and hostname in /usr/bin.
4470 28. Exim was keeping the "process log" file open after each use, just as it
4471 does for the main log. This opens the possibility of it remaining open for
4472 long periods when the USR1 signal hits a daemon. Occasional processlog
4473 errors were reported, that could have been caused by this. Anyway, it seems
4474 much more sensible not to leave this file open at all, so that is what now
4477 29. The long-running daemon process does not normally write to the log once it
4478 has entered its main loop, and it closes the log before doing so. This is
4479 so that log files can straightforwardly be renamed and moved. However,
4480 there are a couple of unusual error situations where the daemon does write
4481 log entries, and I had neglected to close the log afterwards.
4483 30. The text of an SMTP error response that was received during a remote
4484 delivery was being truncated at 512 bytes. This is too short for some of
4485 the long messages that one sometimes sees. I've increased the limit to
4488 31. It is now possible to make retry rules that apply only when a message has a
4489 specific sender, in particular, an empty sender.
4491 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
4492 it possible to be selective about when SMTP synchronization is enforced.
4494 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
4496 32. Implemented hosts_connection_nolog.
4498 33. Added an ACL for QUIT.
4500 34. Setting "delay_warning=" to disable warnings was not working; it gave a
4503 35. Added mailbox_size and mailbox_filecount to appendfile.
4505 36. Added control = no_multiline_responses to ACLs.
4507 37. There was a bug in the logic of the code that waits for the clock to tick
4508 in the case where the clock went backwards by a substantial amount such
4509 that the microsecond fraction of "now" was more than the microsecond
4510 fraction of "then" (but the whole seconds number was less).
4512 38. Added support for the libradius Radius client library this is found on
4513 FreeBSD (previously only the radiusclient library was supported).
4519 1. When certain lookups returned multiple values in the form name=value, the
4520 quoting of the values was not always being done properly. Specifically:
4521 (a) If the value started with a double quote, but contained no whitespace,
4523 (b) If the value contained whitespace other than a space character (i.e.
4524 tabs or newlines or carriage returns) it was not quoted.
4525 This fix has been applied to the mysql and pgsql lookups by writing a
4526 separate quoting function and calling it from the lookup code. The fix
4527 should probably also be applied to nisplus, ibase and oracle lookups, but
4528 since I cannot test any of those, I have not disturbed their existing code.
4530 2. A hit in the callout cache for a specific address caused a log line with no
4531 reason for rejecting RCPT. Now it says "Previous (cached) callout
4532 verification failure".
4534 3. There was an off-by-one bug in the queryprogram router. An over-long
4535 return line was truncated at 256 instead of 255 characters, thereby
4536 overflowing its buffer with the terminating zero. As well as fixing this, I
4537 have increased the buffer size to 1024 (and made a note to document this).
4539 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
4540 when Exim is waiting for an SMTP response from a remote server, Exim
4541 restarts its select() call on the socket, thereby resetting its timeout.
4542 This is not a problem when such interrupts are rare. Somebody set up a cron
4543 job to run exiwhat every 2 minutes, which is less than the normal select()
4544 timeout (5 or 10 minutes). This meant that the select() timeout never
4545 kicked in because it was always reset. I have fixed this by comparing the
4546 time when an interrupt arrives with the time at the start of the first call
4547 to select(). If more time than the timeout has elapsed, the interrupt is
4548 treated as a timeout.
4550 5. Some internal re-factoring in preparation for the addition of Sieve
4551 extensions (by MH). In particular, the "personal" test is moved to a
4552 separate function, and given an option for scanning Cc: and Bcc: (which is
4553 not set for Exim filters).
4555 6. When Exim created an email address using the login of the caller as the
4556 local part (e.g. when creating a From: or Sender: header line), it was not
4557 quoting the local part when it contained special characters such as @.
4559 7. Installed new OpenBSD configuration files.
4561 8. Reworded some messages for syntax errors in "and" and "or" conditions to
4562 try to make them clearer.
4564 9. Callout options, other than the timeout value, were being ignored when
4565 verifying sender addresses in header lines. For example, when using
4567 verify = header_sender/callout=no_cache
4569 the cache was (incorrectly) being used.
4571 10. Added a missing instance of ${EXE} to the exim_install script; this affects
4572 only the Cygwin environment.
4574 11. When return_path_on_delivery was set as a log selector, if different remote
4575 addresses in the same message used different return paths and parallel
4576 remote delivery occurred, the wrong values would sometimes be logged.
4577 (Whenever a remote delivery process finished, the return path value from
4578 the most recently started remote delivery process was logged.)
4580 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
4581 lines when AUTH and/or TLS are in use. This is the "received protocol"
4582 field. Exim used to use "asmtp" for authenticated SMTP, without any
4583 indication (in the protocol name) for TLS use. Now it follows the RFC and
4584 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
4585 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
4586 names appear in log lines as well as in Received: header lines.
4588 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
4589 extensions, and comparison tests, and to fix some bugs.
4591 14. Changes to the "personal" filter test:
4593 (1) The test was buggy in that it was just doing the equivalent of
4594 "contains" tests on header lines. For example, if a user's address was
4595 anne@some.where, the "personal" test would incorrectly be true for
4597 To: susanne@some.where
4599 This test is now done by extracting each address from the header in turn,
4600 and checking the entire address. Other tests that are part of "personal"
4601 are now done using regular expressions (for example, to check local parts
4602 of addresses in From: header lines).
4604 (2) The list of non-personal local parts in From: addresses has been
4605 extended to include "listserv", "majordomo", "*-request", and "owner-*",
4606 taken from the Sieve specification recommendations.
4608 (3) If the message contains any header line starting with "List-" it is
4609 treated as non-personal.
4611 (4) The test for "circular" in the Subject: header line has been removed
4612 because it now seems ill-conceived.
4614 15. Minor typos in src/EDITME comments corrected.
4616 16. Installed latest exipick from John Jetmore.
4618 17. If headers_add on a router specified a text string that was too long for
4619 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
4620 of string_sprintf() is now avoided.
4622 18. $message_body_size was not set (it was always zero) when running the DATA
4623 ACL and the local_scan() function.
4625 19. For the "mail" command in an Exim filter, no default was being set for
4626 the once_repeat time, causing a random time value to be used if "once" was
4627 specified. (If the value happened to be <= 0, no repeat happened.) The
4628 default is now 0s, meaning "never repeat". The "vacation" command was OK
4629 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
4630 (I found it when inspecting the code).
4632 20. There is now an overall timeout for performing a callout verification. It
4633 defaults to 4 times the callout timeout, which applies to individual SMTP
4634 commands during the callout. The overall timeout applies when there is more
4635 than one host that can be tried. The timeout is checked before trying the
4636 next host. This prevents very long delays if there are a large number of
4637 hosts and all are timing out (e.g. when the network connections are timing
4638 out). The value of the overall timeout can be changed by specifying an
4639 additional sub-option for "callout", called "maxwait". For example:
4641 verify = sender/callout=5s,maxwait=20s
4643 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
4644 to the end before writing, but this should make it even safer).
4646 22. Exim was forgetting that it had advertised PIPELINING for the second and
4647 subsequent messages on an SMTP connection. It was also not resetting its
4648 memory on STARTTLS and an internal HELO.
4650 23. When Exim logs an SMTP synchronization error within a session, it now
4651 records whether PIPELINING has been advertised or not.
4653 24. Added 3 instances of "(long int)" casts to time_t variables that were being
4654 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
4655 rather than long int.
4657 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
4659 26. Added the never_mail option to autoreply.
4665 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
4666 crash if the getsockname() call failed; this can happen if a connection is
4667 closed very soon after it is established. The problem was simply in the
4668 order in which certain operations were done, causing Exim to try to write
4669 to the SMTP stream before it had set up the file descriptor. The bug has
4670 been fixed by making things happen in the correct order.
4676 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
4677 before the connection was closed, thus losing the rejection response.
4679 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
4680 some early Solaris releases, but causes trouble in current releases where
4681 socklen_t is defined.
4683 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
4686 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
4689 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
4690 that releasing the top of it at the end releases what was used for sub-
4691 expansions (unless the block got too big). However, discard this block if
4692 the first thing is a variable or header, so that we can use its block when
4693 it is dynamic (useful for very large $message_headers, for example).
4695 6. Lookups now cache *every* query, not just the most recent. A new, separate
4696 store pool is used for this. It can be recovered when all lookup caches are
4697 flushed. Lookups now release memory at the end of their result strings.
4698 This has involved some general refactoring of the lookup sources.
4700 7. Some code has been added to the store_xxx() functions to reduce the amount
4701 of flapping under certain conditions.
4703 8. log_incoming_interface used to affect only the <= reception log lines. Now
4704 it causes the local interface and port to be added to several more SMTP log
4705 lines, for example "SMTP connection from", and rejection lines.
4707 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
4709 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
4711 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
4712 could be overwritten at the end of the current message (or the start of a
4713 new message if it was set in a HELO ACL). The value is now preserved for
4714 the duration of the SMTP connection.
4716 12. If a transport had a headers_rewrite setting, and a matching header line
4717 contained an unqualified address, that address was qualified, even if it
4718 did not match any rewriting rules. The underlying bug was that the values
4719 of the flags that permit the existence of unqualified sender and recipient
4720 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
4721 non-local messages, and by -bnq for local messages) were not being
4722 preserved with the message after it was received.
4724 13. When Exim was logging an SMTP synchronization error, it could sometimes log
4725 "next input=" as part of the text comprising the host identity instead of
4726 the correct text. The code was using the same buffer for two different
4727 strings. However, depending on which order the printing function evaluated
4728 its arguments, the bug did not always show up. Under Linux, for example, my
4729 test suite worked just fine.
4731 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
4732 doesn't work with some older versions of Perl. It has been changed to "my",
4733 which in any case is probably the better facility to use.
4735 15. A really picky compiler found some instances of statements for creating
4736 error messages that either had too many or two few arguments for the format
4739 16. The size of the buffer for calls to the DNS resolver has been increased
4740 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
4741 for addresses that have a lot of PTR records. This alleviates a problem; it
4742 does not fully solve it.
4744 17. A dnsdb lookup for PTR records that receives more data than will fit in the
4745 buffer now truncates the list and logs the incident, which is the same
4746 action as happens when Exim is looking up a host name and its aliases.
4747 Previously in this situation something unpredictable would happen;
4748 sometimes it was "internal error: store_reset failed".
4750 18. If a server dropped the connection unexpectedly when an Exim client was
4751 using GnuTLS and trying to read a response, the client delivery process
4752 crashed while trying to generate an error log message.
4754 19. If a "warn" verb in an ACL added multiple headers to a message in a single
4755 string, for example:
4757 warn message = H1: something\nH2: something
4759 the text was added as a single header line from Exim's point of view
4760 though it ended up OK in the delivered message. However, searching for the
4761 second and subsequent header lines using $h_h2: did not work. This has been
4762 fixed. Similarly, if a system filter added multiple headers in this way,
4763 the routers could not see them.
4765 20. Expanded the error message when iplsearch is called with an invalid key to
4766 suggest using net-iplsearch in a host list.
4768 21. When running tests using -bh, any delays imposed by "delay" modifiers in
4769 ACLs are no longer actually imposed (and a message to that effect is
4772 22. If a "gecos" field in a passwd entry contained escaped characters, in
4773 particular, if it contained a \" sequence, Exim got it wrong when building
4774 a From: or a Sender: header from that name. A second bug also caused
4775 incorrect handling when an unquoted " was present following a character
4776 that needed quoting.
4778 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
4779 was not being matched caselessly.
4781 24. Arranged for all hyphens in the exim.8 source to be escaped with
4784 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
4785 properly. Recipient callout cache records were still being keyed to include
4786 the sender, even when use_sender was set false. This led to far more
4787 callouts that were necessary. The sender is no longer included in the key
4788 when use_sender is false.
4790 26. Added "control = submission" modifier to ACLs.
4792 27. Added the ${base62d: operator to decode base 62 numbers.
4794 28. dnsdb lookups can now access SRV records.
4796 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
4797 the configuration file.
4799 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
4800 (-v) mode. This makes the output for a verbose queue run more intelligible.
4802 31. Added a use_postmaster feature to recipient callouts.
4804 32. Added the $body_zerocount variable, containing the number of binary zero
4805 bytes in the message body.
4807 33. The time of last modification of the "new" subdirectory is now used as the
4808 "mailbox time last read" when there is a quota error for a maildir
4811 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
4813 35. Added +ignore_unknown as a special item in host lists.
4815 36. Code for decoding IPv6 addresses in host lists is now included, even if
4816 IPv6 support is not being compiled. This fixes a bug in which an IPv6
4817 address was recognized as an IP address, but was then not correctly decoded
4818 into binary, causing unexpected and incorrect effects when compared with
4825 1. Very minor rewording of debugging text in manualroute to say "list of
4826 hosts" instead of "hostlist".
4828 2. If verify=header_syntax was set, and a header line with an unqualified
4829 address (no domain) and a large number of spaces between the end of the
4830 name and the colon was received, the reception process suffered a buffer
4831 overflow, and (when I tested it) crashed. This was caused by some obsolete
4832 code that should have been removed. The fix is to remove it!
4834 3. When running in the test harness, delay a bit after writing a bounce
4835 message to get a bit more predictability in the log output.
4837 4. Added a call to search_tidyup() just before forking a reception process. In
4838 theory, someone could use a lookup in the expansion of smtp_accept_max_
4839 per_host which, without the tidyup, could leave open a database connection.
4841 5. Added the variables $recipient_data and $sender_data which get set from a
4842 lookup success in an ACL "recipients" or "senders" condition, or a router
4843 "senders" option, similar to $domain_data and $local_part_data.
4845 6. Moved the writing of debug_print from before to after the "senders" test
4848 7. Change 4.31/66 (moving the time when the Received: is generated) caused
4849 problems for message scanning, either using a data ACL, or using
4850 local_scan() because the Received: header was not generated till after they
4851 were called (in order to set the time as the time of reception completion).
4852 I have revised the way this works. The header is now generated after the
4853 body is received, but before the ACL or local_scan() are called. After they
4854 are run, the timestamp in the header is updated.
4860 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
4861 before starting a queue runner without re-exec. This happened only when
4862 deliver_drop_privilege was set or when the Exim user was set to root. The
4863 effect of the bug was that timeouts during subsequent deliveries caused
4864 crashes instead of being properly handled. The handler is now left at its
4865 default (and expected) setting.
4867 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
4868 message, again when deliver_drop_privilege is set or Exim is run as root.
4869 The bug described in (1) was not present in this case, but the tidying up
4870 of the other signals was missing. I have made the two cases consistent.
4872 3. The ignore_target_hosts setting on a manualroute router was being ignored
4873 for hosts that were looked up using the /MX notation.
4875 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
4878 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
4879 operated on the sender address. After changing the $sender_address to <>
4880 for the sender address verify, Exim was re-instated it as the original
4881 (before rewriting) address, but remembering that it had rewritten it, so it
4882 wasn't rewriting it again. This bug also had the effect of breaking the
4883 sender address verification caching when the sender address was rewritten.
4885 6. The ignore_target_hosts option was being ignored by the ipliteral router.
4886 This has been changed so that if the ip literal address matches
4887 ignore_target_hosts, the router declines.
4889 7. Added expansion conditions match_domain, match_address, and match_local_
4890 part (NOT match_host).
4892 8. The placeholder for the Received: header didn't have a length field set.
4894 9. Added code to Exim itself and to exim_lock to test for a specific race
4895 condition that could lead to file corruption when using MBX delivery. The
4896 issue is with the lockfile that is created in /tmp. If this file is removed
4897 after a process has opened it but before that process has acquired a lock,
4898 there is the potential for a second process to recreate the file and also
4899 acquire a lock. This could lead to two Exim processes writing to the file
4900 at the same time. The added code performs the same test as UW imapd; it
4901 checks after acquiring the lock that its file descriptor still refers to
4902 the same named file.
4904 10. The buffer for building added header lines was of fixed size, 8192 bytes.
4905 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
4908 11. Added the smtp_active_hostname option. If used, this will typically be made
4909 to depend on the incoming interface address. Because $interface_address is
4910 not set up until the daemon has forked a reception process, error responses
4911 that can happen earlier (such as "too many connections") no longer contain
4914 12. If an expansion in a condition on a "warn" statement fails because a lookup
4915 defers, the "warn" statement is abandoned, and the next ACL statement is
4916 processed. Previously this caused the whole ACL to be aborted.
4918 13. Added the iplsearch lookup type.
4920 14. Added ident_timeout as a log selector.
4922 15. Added tls_certificate_verified as a log selector.
4924 16. Added a global option tls_require_ciphers (compare the smtp transport
4925 option of the same name). This controls incoming TLS connections.
4927 17. I finally figured out how to make tls_require_ciphers do a similar thing
4928 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
4929 before starting the TLS session.
4931 18. Tabs are now shown as \t in -bP output.
4933 19. If the log selector return_path_on_delivery was set, Exim crashed when
4934 bouncing a message because it had too many Received: header lines.
4936 20. If two routers both had headers_remove settings, and the first one included
4937 a superfluous trailing colon, the final name in the first list and the
4938 first name in the second list were incorrectly joined into one item (with a
4939 colon in the middle).
4945 1. Added -C and -D options to the exinext utility, mainly to make it easier
4946 to include in the automated testing, but these could be helpful when
4947 multiple configurations are in use.
4949 2. The exinext utility was not formatting the output nicely when there was
4950 an alternate port involved in the retry record key, nor when there was a
4951 message id as well (for retries that were specific to a specific message
4952 and a specific host). It was also confused by IPv6 addresses, because of
4953 the additional colons they contain. I have fixed the IPv4 problem, and
4954 patched it up to do a reasonable job for IPv6.
4956 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
4957 delivery, the log line now contains "pipelined" if PIPELINING was used.
4959 4. An SMTP transport process used to panic and die if the bind() call to set
4960 an explicit outgoing interface failed. This has been changed; it is now
4961 treated in the same way as a connect() failure.
4963 5. A reference to $sender_host_name in the part of a conditional expansion
4964 that was being skipped was still causing a DNS lookup. This no longer
4967 6. The def: expansion condition was not recognizing references to header lines
4968 that used bh_ and bheader_.
4970 7. Added the _cache feature to named lists.
4972 8. The code for checking quota_filecount in the appendfile transport was
4973 allowing one more file than it should have been.
4975 9. For compatibility with Sendmail, the command line option
4983 and sets the incoming protocol and host name (for trusted callers). The
4984 host name and its colon can be omitted when only the protocol is to be set.
4985 Note the Exim already has two private options, -pd and -ps, that refer to
4986 embedded Perl. It is therefore impossible to set a protocol value of "d" or
4987 "s", but I don't think that's a major issue.
4989 10. A number of refactoring changes to the code, none of which should affect
4992 (a) The number of logging options was getting close to filling up the
4993 32-bit word that was used as a bit map. I have split them into two classes:
4994 those that are passed in the argument to log_write(), and those that are
4995 only ever tested independently outside of that function. These are now in
4996 separate 32-bit words, so there is plenty of room for expansion again.
4997 There is no change in the user interface or the logging behaviour.
4999 (b) When building, for example, log lines, the code previously used a
5000 macro that called string_cat() twice, in order to add two strings. This is
5001 not really sufficiently general. Furthermore, there was one instance where
5002 it was actually wrong because one of the argument was used twice, and in
5003 one call a function was used. (As it happened, calling the function twice
5004 did not affect the overall behaviour.) The macro has been replaced by a
5005 function that can join an arbitrary number of extra strings onto a growing
5008 (c) The code for expansion conditions now uses a table and a binary chop
5009 instead of a serial search (which was left over from when there were very
5010 few conditions). Also, it now recognizes conditions like "pam" even when
5011 the relevant support is not compiled in: a suitably worded error message is
5012 given if an attempt is made to use such a condition.
5014 11. Added ${time_interval:xxxxx}.
5016 12. A bug was causing one of the ddress fields not to be passed back correctly
5017 from remote delivery subprocesses. The field in question was not being
5018 subsequently used, so this caused to problems in practice.
5020 13. Added new log selectors queue_time and deliver_time.
5022 14. Might have fixed a bug in maildirsizefile handling that threw up
5023 "unexpected character" debug warnings, and recalculated the data
5024 unnecessarily. In any case, I expanded the warning message to give more
5027 15. Added the message "Restricted characters in address" to the statements in
5028 the default ACL that block characters like @ and % in local parts.
5030 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
5031 Three changes have been made:
5033 (a) There was a serious bug; a negative response to MAIL caused the whole
5034 recipient domain to be cached as invalid, thereby blocking all messages
5035 to all local parts at the same domain, from all senders. This bug has
5036 been fixed. The domain is no longer cached after a negative response to
5037 MAIL if the sender used is not empty.
5039 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
5042 (c) A new callout option, "use_sender" has been added for people who want
5043 the modified behaviour.
5049 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
5052 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
5053 indeed breaks things for older releases.
5055 3. Added additional logging to the case where there is a problem reading data
5056 from a filter that is running in a subprocess using a pipe, in order to
5057 try to track down a specific problem.
5059 4. Testing facility fudge: when running in the test harness and attempting
5060 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
5061 getting "No route to host". Convert this to a timeout.
5063 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
5066 6. Some OS don't have socklen_t but use size_t instead. This affects the
5067 fifth argument of getsockopt() amongst other things. This is now
5068 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
5069 can be set for individual OS. I have set it for SunOS5, OSF1, and
5070 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
5071 some earlier ones do not.
5073 7. Change 4.30/15 was not doing the test caselessly.
5075 8. The standard form for an IPv6 address literal was being rejected by address
5076 parsing in, for example, MAIL and RCPT commands. An example of this kind of
5077 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5078 this, as well as the form without the "IPv6" on the front (but only when
5079 address literals are enabled, of course).
5081 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
5083 10. Exim crashed if a message with an empty sender address specified by -f
5084 encountered a router with an errors_to setting. This could be provoked only
5085 by a command such as
5089 where an empty string was supplied; "<>" did not hit this bug.
5091 11. Installed PCRE release 4.5.
5093 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
5094 remained set. It is now erased.
5096 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
5097 times from message ids (which are base 36 rather than the normal 62).
5099 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
5100 were being counted as actual protocol errors, and logged if the log
5101 selector +smtp_protocol_error was set. One cannot be perfect in this test,
5102 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
5103 and DATA following a set of rejected RCPTs do not count as protocol errors.
5104 In other words, Exim assumes they were pipelined, though this may not
5105 actually be the case. Of course, in all cases the client gets an
5106 appropriate error code.
5108 15. If a lookup fails in an ACL condition, a message about the failure may
5109 be available; it is used if testing the ACL cannot continue, because most
5110 such messages specify what the cause of the deferral is. However, some
5111 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
5112 that caused an old message to be retained and used if a later statement
5113 caused a defer, replacing the real cause of the deferral.
5115 16. If an IP address had so many PTR records that the DNS lookup buffer
5116 was not large enough to hold them, Exim could crash while trying to process
5117 the truncated data. It now detects and logs this case.
5119 17. Further to 4.21/58, another change has been made: if (and only if) the
5120 first line of a message (the first header line) ends with CRLF, a bare LF
5121 in a subsequent header line has a space inserted after it, so as not to
5122 terminate the header.
5124 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
5125 unnecessarily, used atoi() instead of strtol(), and didn't check the
5126 termination when getting file sizes from file names by regex.
5128 19. Completely re-implemented the support for maildirsize files, in the light
5129 of a number of problems with the previous contributed implementation
5130 (4.30/29). In particular:
5132 . If the quota is zero, the maildirsize file is maintained, but no quota is
5135 . If the maildir directory does not exist, it is created before any attempt
5136 to write a maildirsize file.
5138 . The quota value in the file is just a cache; if the quota is changed in
5139 the transport, the new value overrides.
5141 . A regular expression is available for excluding directories from the
5144 20. The autoreply transport checks the characters in options that define the
5145 message's headers; it allows continued headers, but it was checking with
5146 isspace() after an embedded newline instead of explicitly looking for a
5149 21. If all the "regular" hosts to which an address was routed had passed their
5150 expiry times, and had not reached their retry times, the address was
5151 bounced, even if fallback hosts were defined. Now Exim should go on to try
5154 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
5155 equivalent code in the SMTP transport. Some hosts send humungous responses
5156 to HELO/EHLO, more than 1024 it seems.
5158 23. Refactoring: code in filter.c used (void *) for "any old type" but this
5159 gives compiler warnings in some environments. I've now done it "properly",
5162 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
5163 (because of problems with the built-in one) was declared to return uschar *
5164 instead of char *, causing compiler failure.
5166 25. Fixed a file descriptor leak when processing alias/forward files.
5168 26. Fixed a minor format string issue in dbfn.c.
5170 27. Typo in exim.c: ("dmbnz" for "dbmnz").
5172 28. If a filter file refered to $h_xxx or $message_headers, and the headers
5173 contained RFC 2047 "words", Exim's memory could, under certain conditions,
5176 29. When a sender address is verified, it is cached, to save repeating the test
5177 when there is more than one recipient in a message. However, when the
5178 verification involves a callout, it is possible for different callout
5179 options to be set for different recipients. It is too complicated to keep
5180 track of this in the cache, so now Exim always runs a verification when a
5181 callout is required, relying on the callout cache for the optimization.
5182 The overhead is duplication of the address routing, but this should not be
5185 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
5186 to be verified with callout=postmaster, and the main callout worked but the
5187 postmaster check failed, the verification correctly failed. However, if a
5188 subsequent RCPT command asked for sender verification *without* the
5189 postmaster check, incorrect caching caused this verification also to fail,
5192 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
5193 it was not caching the DNS options (qualify_single, search_parents) that
5194 were used when the lookup failed. A subsequent lookup with different
5195 options therefore always gave the same answer, though there were cases
5196 where it should not have. (Example: a "domains = !$mx_any" option on a
5197 dnslookup router: the "domains" option is always processed without any
5198 widening, but the router might have qualify_single set.) Now Exim uses the
5199 cached value only when the same options are set.
5201 32. Added John Jetmore's "exipick" utility to the distribution.
5203 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
5204 than a timeout (e.g. a certificate is required, and is not provided), an
5205 Exim server now closes the connection immediately. Previously it waited for
5206 the client to close - but if the client is SSL, it seems that they each
5207 wait for each other, leading to a delay before one of them times out.
5209 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
5210 maintained 0.8.x compatibility because I don't think many are using it, and
5211 it is clearly obsolete.
5213 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
5216 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
5217 client certificate was expired. A simple patch fixes this, though I don't
5218 understand the full logic of why the verify callback is called multiple
5221 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
5222 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
5223 which causes problems with some clients (such as the Certicom SSL Plus
5224 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
5225 disables the coutermeasure allowing Eudora to connect."
5227 38. Exim was not checking that a write() to a log file succeeded. This could
5228 lead to Bad Things if a log got too big, in particular if it hit a file
5229 size limit. Exim now panics and dies if it cannot write to a log file, just
5230 as it does if it cannot open a log file.
5232 39. Modified OS/Makefile-Linux so that it now contains
5234 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
5236 The two -D definitions ensure that Exim is compiled with large file
5237 support, which makes it possible to handle log files that are bigger than
5240 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
5241 instance) a domain was checked against a named list that involved a lookup,
5242 causing $domain_data to be set, then another domain was checked against the
5243 same list, then the first domain was re-checked, the value of $domain_data
5244 after the final check could be wrong. In particular, if the second check
5245 failed, it could be set empty. This bug probably also applied to
5248 41. The strip_trailing_dot option was not being applied to the address given
5249 with the -f command-line option.
5251 42. The code for reading a message's header from the spool was incrementing
5252 $received_count, but never initializing it. This meant that the value was
5253 incorrect (doubled) while delivering a message in the same process in which
5254 it was received. In the most common configuration of Exim, this never
5255 happens - a fresh exec is done - but it can happen when
5256 deliver_drop_privilege is set.
5258 43. When Exim logs an SMTP synchronization error - client data sent too soon -
5259 it now includes up to 150 characters of the unexpected data in the log
5262 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
5263 and building data strings. The size of both of these buffers was 10 000
5264 bytes - far larger than anybody would *ever* want, thought I. Needless to
5265 say, somebody hit the limit. I have increased the maximum line length to
5266 20 000 and the maximum data length of concatenated lines to 100 000. I have
5267 also fixed two bugs, because there was no checking on these buffers. Tsk,
5268 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
5269 buffer is too small.
5271 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
5272 lsearch lookups. Now it does.
5274 46. When parsing a route_list item in a manualroute router, a fixed-length
5275 buffer was used for the list of hosts. I made this 1024 bytes long,
5276 thinking that nobody would ever have a list of hosts that long. Wrong.
5277 Somebody had a whole pile of complicated expansion conditions, and the
5278 string was silently truncated, leading to an expansion error. It turns out
5279 that it is easier to change to an unlimited length (owing to other changes
5280 that have happened since this code was originally written) than to build
5281 structure for giving a limitation error. The length of the item that
5282 expands into the list of hosts is now unlimited.
5284 47. The lsearch lookup could not handle data where the length of text line was
5285 more than 4095 characters. Such lines were truncated, leading to shortened
5286 data being returned. It should now handle lines of any length.
5288 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
5289 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
5292 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
5293 replacing strings such as BIN_DIRECTORY in a source file: the replacement
5294 no longer happens in comment lines. A list of replacements is now placed
5295 at the head of all of the source files, except those whose only change is
5296 to replace PERL_COMMAND in the very first #! line.
5298 50. Replaced the slow insertion sort in queue.c, for sorting the list of
5299 messages on the queue, with a bottom-up merge sort, using code contributed
5300 by Michael Haardt. This should make operations like -bp somewhat faster on
5301 large queues. It won't affect queue runners, except when queue_run_in_order
5304 51. Installed eximstats 1.31 in the distribution.
5306 52. Added support for SRV lookups to the dnslookup router.
5308 53. If an ACL referred to $message_body or $message_body_end, the value was not
5309 reset for any messages that followed in the same SMTP session.
5311 54. The store-handling optimization for building very long strings was not
5312 differentiating between the different store pools. I don't think this
5313 actually made any difference in practice, but I've tidied it.
5315 55. While running the routers to verify a sender address, $sender_address
5316 was still set to the sender address. This is wrong, because when routing to
5317 send a bounce to the sender, it would be empty. Therefore, I have changed
5318 it so that, while verifying a sender address, $sender_address is set to <>.
5319 (There is no change to what happens when verifying a recipient address.)
5321 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
5322 target A or AAAA records (if not already returned) without resetting the
5323 qualify_single or search_parents options of the DNS resolver. These are
5324 inappropriate in this case because the targets of MX and SRV records must
5325 be FQDNs. A broken DNS record could cause trouble if it happened to have a
5326 target that, when qualified, matched something in the local domain. These
5327 two options are now turned off when doing these lookups.
5329 57. It seems that at least some releases of Reiserfs (which does not have the
5330 concept of a fixed number of inodes) returns zero and not -1 for the
5331 number of available inodes. This interacted badly with check_spool_inodes,
5332 which assumed that -1 was the "no such thing" setting. What I have done is
5333 to check that the total number of inodes is greater than zero before doing
5334 the test of how many are available.
5336 58. When a "warn" ACL statement has a log_message modifier, the message is
5337 remembered, and not repeated. This is to avoid a lot of repetition when a
5338 message has many recipients that cause the same warning to be written.
5339 Howewer, Exim was preserving the list of already written lines for an
5340 entire SMTP session, which doesn't seem right. The memory is now reset if a
5341 new message is started.
5343 59. The "rewrite" debugging flag was not showing the result of rewriting in the
5344 debugging output unless log_rewrite was also set.
5346 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
5347 of (int)(handle) when we know that handle contains (void *)(-1).
5349 61. The Exim daemon panic-logs an error return when it closes the incoming
5350 connection. However "connection reset by peer" seems to be common, and
5351 isn't really an error worthy of noting specially, so that particular error
5354 62. When Exim is trying to find all the local interfaces, it used to panic and
5355 die if the ioctl to get the interface flags failed. However, it seems that
5356 on at least one OS (Solaris 9) it is possible to have an interface that is
5357 included in the list of interfaces, but for which you get a failure error
5358 for this call. This happens when the interface is not "plumbed" into a
5359 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
5360 failure of the "get flags" call assumes that the interface is down.
5362 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
5363 makes life easier for people who are doing arithmetic on fields extracted
5364 from dates, where you often get leading zeros that should not be
5365 interpreted as octal.
5367 64. Added qualify_domain to the redirect router, to override the global
5370 65. If a pathologically long header line contained very many addresses (the
5371 report of this problem mentioned 10 000) and each of them was rewritten,
5372 Exim could use up a very large amount of memory. (It kept on making new
5373 copies of the header line as it rewrote, and never released the old ones.)
5374 At the expense of a bit more processing, the header rewriting function has
5375 been changed so that it no longer eats memory in this way.
5377 66. The generation of the Received: header has been moved from the time that a
5378 message starts to be received, to the time that it finishes. The timestamp
5379 in the Received: header should now be very close to that of the <= log
5380 line. There are two side-effects of this change:
5382 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
5383 logged header lines no longer include the local Received: line, because
5384 it has not yet been created. The same applies to a copy of the message
5385 that is returned to a non-SMTP sender when a message is rejected.
5387 (b) When a filter file is tested using -bf, no additional Received: header
5388 is added to the test message. After some thought, I decided that this
5391 This change does not affect the value of $received_for. It is still set
5392 after address rewriting, but before local_scan() is called.
5394 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
5396 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
5397 gave an unhelpful panic error message, and a defer error. I have managed to
5398 change this behaviour so that it now rejects any supplied certificate,
5399 which seems right, as the list of acceptable certificates is empty.
5401 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
5402 gave an unhelpful defer error. I have not managed to make this reject any
5403 supplied certificates, but the error message it gives is "no certificate
5404 supplied", which is not helpful.
5406 70. exigrep's output now also includes lines that are not associated with any
5407 message, but which match the given pattern. Implemented by a patch from
5408 Martin Sluka, which also tidied up the Perl a bit.
5410 71. Recipient callout verification, like sender verification, was using <> in
5411 the MAIL FROM command. This isn't really the right thing, since the actual
5412 sender may affect whether the remote host accepts the recipient or not. I
5413 have changed it to use the actual sender in the callout; this means that
5414 the cache record is now keyed on a recipient/sender pair, not just the
5415 recipient address. There doesn't seem to be a real danger of callout loops,
5416 since a callout by the remote host to check the sender would use <>.
5417 [SEE ABOVE: changed after hitting problems.]
5419 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
5420 temporary errors. However, in the case of such a code being given after
5421 the end of a data transmission (i.e. after ".") Exim was failing to write
5422 a retry record for the message. (Yes, there was some broken host that was
5423 actually sending 8xx at this point.)
5425 73. An unknown lookup type in a host list could cause Exim to panic-die when
5426 the list was checked. (An example that provoked this was putting <; in the
5427 middle of a list instead of at the start.) If this happened during a DATA
5428 ACL check, a -D file could be left lying around. This kind of configuration
5429 error no longer causes Exim to die; instead it causes a defer errror. The
5430 incident is still logged to the main and panic logs.
5432 74. Buglet left over from Exim 3 conversion. The message "too many messages
5433 in one connection" was written to the rejectlog but not the mainlog, except
5434 when address rewriting (yes!) was being logged.
5436 75. Added write_rejectlog option.
5438 76. When a system filter was run not as root (that is, when system_filter_user
5439 was set), the values of the $n variables were not being returned to the
5440 main process; thus, they were not subsequently available in the $sn
5443 77. Added +return_path_on_delivery log selector.
5445 78. A connection timeout was being treated differently from recipients deferred
5446 when testing hosts_max_try with a message that was older than the host's
5447 retry timeout. (The host should not be counted, thus allowing all hosts to
5448 be tried at least once before bouncing.) This may have been the cause of an
5449 occasionally reported bug whereby a message would remain on the queue
5450 longer than the retry timeout, but would be bounced if a delivery was
5451 forced. I say "may" because I never totally pinned down the problem;
5452 setting up timeout/retry tests is difficult. See also the next item.
5454 79. The ultimate address timeout was not being applied to errors that involved
5455 a combination of host plus message (for example, a timeout on a MAIL
5456 command). When an address resolved to a number of possible hosts, and they
5457 were not all tried for each delivery (e.g. because of hosts_max_try), a
5458 message could remain on the queue longer than the retry timeout.
5460 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
5463 81. Fixed an obscure SMTP outgoing bug which required at least the following
5464 conditions: (a) there was another message waiting for the same server;
5465 (b) the server returned 5xx to all RCPT commands in the first message so
5466 that the message was not completed; (c) the server dropped the connection
5467 or gave a negative response to the RSET that Exim sends to abort the
5468 transaction. The observed case was a dropped connection after DATA that had
5469 been sent in pipelining mode. That is, the server had advertised PIPELINING
5470 but was not implementing it correctly. The effect of the bug was incorrect
5471 behaviour, such as trying another host, and this could lead to a crash.
5477 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
5478 and daemon.c were passed as pointers to ints; they should have been
5479 pointers to socklen_t variables (which are typically unsigned ints).
5481 2. Some signed/unsigned type warnings in the os.c file for Linux have been
5484 3. Fixed a really odd bug that affected only the testing scheme; patching a
5485 certain fixed string in the binary changed the value of another string that
5486 happened to be identical to the end of the original first string.
5488 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
5489 name", it returns that address as the IP address. On some operating
5490 systems (e.g. Solaris), it also passes back the IP address string as the
5491 "host name". However, on others (e.g. Linux), it passes back an empty
5492 string. Exim wasn't checking for this, and was changing the host name to an
5493 empty string, assuming it had been canonicized.
5495 5. Although rare, it is permitted to have more than one PTR record for a given
5496 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
5497 all the names associated with an address, because they do in Solaris.
5498 However, it seems that they do not in Linux for data that comes from the
5499 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
5500 I found this out when I moved to a new Linux workstation and tried to run
5501 the Exim test suite.
5503 To get round this problem I have changed the code so that it now does its
5504 own call to the DNS to look up PTR records when searching for a host name.
5505 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
5506 addresses that are only in /etc/hosts are still found.
5508 This behaviour is, however, controlled by an option called host_lookup_
5509 order, which defaults to "bydns:byaddr". If people want to use the other
5510 order, or indeed, just use one or the other means of lookup, they can
5511 specify it in this variable.
5513 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
5514 some operating systems, this comes back from gethostbyaddr() as an empty
5515 string, and this is what Exim used to test for. However, it seems that in
5516 other systems, "." is yielded. Exim now tests for this case too.
5518 7. The values of check_spool_space and check_log_space are now held internally
5519 as a number of kilobytes instead of an absolute number of bytes. If a
5520 numbers is specified without 'K' or 'M', it is rounded up to the nearest
5521 kilobyte. This means that much larger values can be stored.
5523 8. Exim monitor: an attempt to get the action menu when not actually pointing
5524 at a message produces an empty menu entitled "No message selected". This
5525 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
5526 no entries in it ("Shell widget menu has zero width and/or height"). So I
5527 have added a single, blank menu entry in this case.
5529 9. Added ${quote_local_part.
5531 10. MIME decoding is now applied to the contents of Subject: header lines when
5534 11. Now that a reference to $sender_host_address automatically causes a reverse
5535 lookup to occur if necessary (4.13/18), there is no need to arrange for a
5536 host lookup before query-style lookups in lists that might use this
5537 variable. This has therefore been abolished, and the "net-" prefix is no
5538 longer necessary for query-style lookups.
5540 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
5541 have been a typo for LFLAGS, so it has been changed.
5543 13. The install script calls Exim with "-C /dev/null" in order to find the
5544 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
5545 to be output. Howeve, since Exim outputs its version number before the
5546 error, it didn't break the script. It just looked ugly. I fixed this by
5547 always allowing "-C /dev/null" if the caller is root.
5549 14. Ignore overlarge ACL variable number when reading spool file - insurance
5550 against a later release with more variables having written the file.
5552 15. The standard form for an IPv6 address literal was being rejected by EHLO.
5553 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5554 this, as well as the form without the "IPv6" on the front.
5556 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
5557 OS/Makefile-Darwin file.
5559 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
5560 only to LDAP libraries that do not have LDAP_OPT_DEREF.
5562 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
5563 variable. However, some OS use ints for inodes. I've added cast to long int
5564 to get rid of the compiler warning.
5566 19. I had forgotten to lock out "/../" in configuration file names when
5567 ALT_CONFIG_PREFIX was set.
5569 20. Routers used for verification do not need to specify transports. However,
5570 if such a router generated a host list, and callout was configured, Exim
5571 crashed, because it could not find a port number from the (non-existent)
5572 transport. It now assumes port 25 in this circumstance.
5574 21. Added the -t option to exigrep.
5576 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
5577 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
5578 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
5579 them is set, LOOKUP_LSEARCH is forced.
5581 23. "exim -bV" now outputs a list of lookups that are included in the binary.
5583 24. Added sender and host information to the "rejected by local_scan()" log
5584 line; previously there was no indication of these.
5586 25. Added .include_if_exists.
5588 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
5589 Linux. It turns out that not all file systems support this. Apparently some
5590 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
5591 do it.) To cope with this, the error EINVAL, which means that sync-ing is
5592 not supported on the file descriptor, is now ignored when Exim is trying to
5593 sync a directory. This applies only to Linux.
5595 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
5597 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
5598 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
5601 29. Added support for maildirsize files from supplied patch (modified a bit).
5603 30. The use of :fail: followed by an empty string could lead Exim to respond to
5604 sender verification failures with (e.g.):
5606 550 Verification failed for <xxx>
5607 550 Sender verify failed
5609 where the first response line was missing the '-' that indicates it is not
5610 the final line of the response.
5612 31. The loop for finding the name of the user that called Exim had a hardwired
5613 limit of 10; it now uses the value of finduser_retries, which is used for
5614 all other user lookups.
5616 32. Added $received_count variable, available in data and not_smtp ACLs, and at
5619 33. Exim was neglecting to zero errno before one call of strtol() when
5620 expanding a string and expecting an integer value. On some systems this
5621 resulted in spurious "integer overflow" errors. Also, it was casting the
5622 result into an int without checking.
5624 34. Testing for a connection timeout using "timeout_connect" in the retry rules
5625 did not work. The code looks as if it has *never* worked, though it appears
5626 to have been documented since at least releast 1.62. I have made it work.
5628 35. The "timeout_DNS" error in retry rules, also documented since at least
5629 1.62, also never worked. As it isn't clear exactly what this means, and
5630 clearly it isn't a major issue, I have abolished the feature by treating it
5631 as "timeout", and writing a warning to the main and panic logs.
5633 36. The display of retry rules for -brt wasn't always showing the error code
5636 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
5637 timeout_connect_A, timeout_connect_MX.
5639 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
5640 to the empty sender.
5642 39. The daemon was not analysing the content of -oX till after it had closed
5643 stderr and disconnected from the controlling terminal. This meant that any
5644 syntax errors were only noted on the panic log, and the return code from
5645 the command was 0. By re-arranging the code a little, I've made the
5646 decoding happen first, so such errors now appear on stderr, and the return
5647 code is 1. However, the actual setting up of the sockets still happens in
5648 the disconnected process, so errors there are still only recorded on the
5651 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
5652 connections (as happens on some IP stacks) was logged at start up time as
5653 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
5654 it from "IPv6 and IPv4", which means that two separate sockets are being
5657 41. The debug output for gethostbyname2() or getipnodebyname() failures now
5658 says whether AF_INET or AF_INET6 was passed as an argument.
5660 42. Exiwhat output was messed up when time zones were included in log
5663 43. Exiwhat now gives more information about the daemon's listening ports,
5664 and whether -tls-on-connect was used.
5666 44. The "port" option of the smtp transport is now expanded.
5668 45. A "message" modifier in a "warn" statement in a non-message ACL was being
5669 silently ignored. Now an error message is written to the main and panic
5672 46. There's a new ACL modifier called "logwrite" which writes to a log file
5673 as soon as it is encountered.
5675 47. Added $local_user_uid and $local_user_gid at routing time.
5677 48. Exim crashed when trying to verify a sender address that was being
5680 49. Exim was recognizing only a space character after ".include". It now also
5681 recognizes a tab character.
5683 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
5684 extracting the relevant information from the specification. The man page no
5685 longer contains scrambled data for the -d option, and I've added a section
5686 at the front about calling Exim under different names.
5688 51. Added "extra_headers" argument to the "mail" command in filter files.
5690 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
5693 53. Installed eximstats 1.29.
5695 54. Added transport_filter_timeout as a generic transport option.
5697 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
5698 Cc: header lines. This was required by RFC 822, but it not required by RFC
5701 56. Exim used to add From:, Date:, and Message-Id: header lines to any
5702 incoming messages that did not have them. Now it does so only if the
5703 message originates locally, that is, if there is no associated remote host
5704 address. When Resent- header lines are present, this applies to the Resent-
5705 lines rather than the non-Resent- lines.
5707 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
5708 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
5710 58. Messages for configuration errors now include the name of the main
5711 configuration file - useful now that there may be more than one file in a
5712 list (.included file names were always shown).
5714 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
5715 for those rare installations that do not start the daemon as root or run it
5716 setuid root. I've cut out the call to initgroups() if the daemon is not
5719 60. The Exim user and group can now be bound into the binary as text strings
5720 that are looked up at the start of Exim's processing.
5722 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
5724 62. Added $mailstore_basename variable.
5726 63. Installed patch to sieve.c from Michael Haardt.
5728 64. When Exim failed to open the panic log after failing to open the main log,
5729 the original message it was trying to log was written to stderr and debug
5730 output, but if they were not available (the usual case in production), it
5731 was lost. Now it is written to syslog before the two lines that record the
5732 failures to open the logs.
5734 65. Users' Exim filters run in subprocesses under the user's uid. It is
5735 possible for a "deliver" command or an alias in a "personal" command to
5736 provoke an address rewrite. If logging of address rewriting is configured,
5737 this fails because the process is not running as root or exim. There may be
5738 a better way of dealing with this, but for the moment (because 4.30 needs
5739 to be released), I have disabled address rewrite logging when running a
5740 filter in a non-root, non-exim process.
5746 1. The buildconfig auxiliary program wasn't quoting the value set for
5747 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
5748 not defined. This bug was masked in 4.22 by the effect that was fixed in
5751 2. Some messages that were rejected after a message id was allocated were
5752 shown as "incomplete" by exigrep. It no longer does this for messages that
5753 are rejected by local_scan() or the DATA or non-SMTP ACLs.
5755 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
5756 have allow_domain_literals set, the ID did not get logged in the <= line.
5757 Domain literals are now always recognized in Message-ID: header lines.
5759 4. The first argument for a ${extract expansion item is the key name or field
5760 number. Leading and trailing spaces in this item were not being ignored,
5761 causing some misleading effects.
5763 5. When deliver_drop_privilege was set, single queue runner processes started
5764 manually (i.e. by the command "exim -q") or by the daemon (which uses the
5765 same command in the process it spins off) were not dropping privilege.
5767 6. When the daemon running as "exim" started a queue runner, it always
5768 re-executed Exim in the spun-off process. This is a waste of effort when
5769 deliver_drop_privilege is set. The new process now just calls the
5770 queue-runner function directly.
5776 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
5779 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
5780 ignored. Though the use of -oP was forcing the writing of a pid file, it
5781 was always written to the default place.
5783 3. If the message "no IP address found for host xxxx" is generated during
5784 incoming verification, it is now followed by identification of the incoming
5785 connection (so you can more easily find what provoked it).
5787 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
5789 5. Added some features to "harden" Exim a bit more against certain attacks:
5791 (a) There is now a build-time option called FIXED_NEVER_USERS that can
5792 be put in Local/Makefile. This is like the never_users runtime option,
5793 but it cannot be overridden. The default setting is "root".
5795 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
5796 prefix string with which any file named in a -C command line option
5799 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
5800 is retained for -C and -D only if the caller of Exim is root. Without
5801 it, the exim user may also use -C and -D and retain privilege.
5803 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
5804 command line option is disabled.
5806 6. Macro names set by the -D option must start with an upper case letter, just
5807 like macro names defined in the configuration file.
5809 7. Added "dereference=" facility to LDAP.
5811 8. Two instances of the typo "uknown" in the source files are fixed.
5813 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
5814 the Configure-Makefile script screwed up while processing it.
5816 10. Incorporated PCRE 4.4.
5818 11. The SMTP synchronization check was not operating right at the start of an
5819 SMTP session. For example, it could not catch a HELO sent before the client
5820 waited for the greeting. There is now a check for outstanding input at the
5821 point when the greeting is written. Because of the duplex, asynchronous
5822 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
5823 way, but not yet received, when the check is performed.
5825 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
5826 on TCP/IP sockets, because this apparently causes some broken clients to
5829 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
5830 unchanged) from the Cygwin maintainer.
5832 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
5833 was supported instead of testing for mbx. Effectively a typo.
5835 15. The spa authenticator server code was not checking that the input it
5836 received was valid base64.
5838 16. The debug output line for the "set" modifier in ACLs was not showing the
5839 name of the variable that was being set.
5841 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
5843 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
5844 lookup on its own. Something else was needed to trigger the lookup. For
5845 example, a match in host_lookup or the need for a host name in a host list.
5846 Now, if $sender_host_name is referenced and the host name has not yet been
5847 looked up, a lookup is performed. If the lookup fails, the variable remains
5848 empty, and $host_lookup_failed is set to "1".
5850 19. Added "eqi" as a case-independent comparison operator.
5852 20. The saslauthd authentication condition could segfault if neither service
5853 nor realm was specified.
5855 21. If an overflowing value such as "2048M" was set for message_size_limit, the
5856 error message that was logged was misleading, and incoming SMTP
5857 connections were dropped. The message is now more accurate, and temporary
5858 errors are given to SMTP connections.
5860 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
5861 (except RSET) with a 421 error, until QUIT is received. However, it was
5862 failing to send a response to QUIT.
5864 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
5865 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
5866 ACL is now run after the helo_try_verify_hosts code.
5868 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
5869 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
5870 case-independent, so other case variants are also recognized). Apparently
5871 some systems use these upper case variants.
5873 25. If more than two messages were waiting for the same host, and a transport
5874 filter was specified for the transport, Exim sent two messages over the
5875 same TCP/IP connection, and then failed with "socket operation on non-
5876 socket" when it tried to send the third.
5878 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
5880 27. The extern definition of crypt16() in expand.c was not being excluded when
5881 the OS had its own crypt16() function.
5883 28. Added bounce_return_body as a new option, and bounce_return_size_limit
5884 as a preferred synonym for return_size_limit, both as an option and as an
5887 29. Added LIBS=-liconv to OS/Makefile-OSF1.
5889 30. Changed the default configuration ACL to relax the local part checking rule
5890 for addresses that are not in any local domains. For these addresses,
5891 slashes and pipe symbols are allowed within local parts, but the sequence
5892 /../ is explicitly forbidden.
5894 31. SPA server authentication was not clearing the challenge buffer before
5897 32. log_message in a "warn" ACL statement was writing to the reject log as
5898 well as to the main log, which contradicts the documentation and doesn't
5899 seem right (because no rejection is happening). So I have stopped it.
5901 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
5902 However, I am unable to do any testing of this.
5904 34. Fixed an infelicity in the appendfile transport. When checking directories
5905 for a mailbox, to see if any needed to be created, it was accidentally
5906 using path names with one or more superfluous leading slashes; tracing
5907 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
5909 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
5910 contents are added to the log line that is written for every discarded
5911 recipient. (Previously a log_message setting was ignored.)
5913 36. The ${quote: operator now quotes the string if it is empty.
5915 37. The install script runs exim in order to find its version number. If for
5916 some reason other than non-existence or emptiness, which it checks, it
5917 could not run './exim', it was installing it with an empty version number,
5918 i.e. as "exim-". This error state is now caught, and the installation is
5921 38. An argument was missing from the function that creates an error message
5922 when Exim fails to connect to the socket for saslauthd authentication.
5923 This could cause Exim to crash, or give a corrupted message.
5925 39. Added isip, isip4, and isip6 to ${if conditions.
5927 40. The ACL variables $acl_xx are now saved with the message, and can be
5928 accessed later in routers, transports, and filters.
5930 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
5931 strings in the file are not string-expanded.
5933 42. If a MAIL command specified a SIZE value that was too large to fit into an
5934 int variable, the check against message_size_limit failed. Such values are
5935 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
5936 day this will have to be increased, but I don't think I want to be around
5937 when emails are that large.
5944 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
5945 iconv() is not standard in FreeBSD.
5947 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
5948 IPv6 enabled. The observed symptom was a segmentation fault on return from
5949 the function os_common_find_running_interfaces() in src/os.c.
5951 3. In the check_special_case() function in daemon.c I had used "errno" as an
5952 argument name, which causes warnings on some systems. This was basically a
5953 typo, since it was named "eno" in the comments!
5955 4. The code that waits for the clock to tick (at a resolution of some fraction
5956 of a second) so as to ensure message-id uniqueness was always waiting for
5957 at least one whole tick, when it could have waited for less. [This is
5958 almost certainly not relevant at current processor speeds, where it is
5959 unlikely to ever wait at all. But we try to future-proof.]
5961 5. The function that sleeps for a time interval that includes fractions of a
5962 second contained a race. It did not block SIGALRM between setting the
5963 timer, and suspending (a couple of lines later). If the interval was short
5964 and the sigsuspend() was delayed until after it had expired, the suspension
5965 never ended. On busy systems this could lead to processes getting stuck for
5968 6. Some uncommon configurations may cause a lookup to happen in a queue runner
5969 process, before it forks any delivery processes. The open lookup caching
5970 mechanism meant that the open file or database connection was passed into
5971 the delivery process. The problem was that delivery processes always tidy
5972 up cached lookup data. This could cause a problem for the next delivery
5973 process started by the queue runner, because the external queue runner
5974 process does not know about the closure. So the next delivery process
5975 still has data in the lookup cache. In the case of a file lookup, there was
5976 no problem because closing a file descriptor in a subprocess doesn't affect
5977 the parent. However, if the lookup was caching a connection to a database,
5978 the connection was closed, and the second delivery process was likely to
5979 see errors such as "PGSQL: query failed: server closed the connection
5980 unexpectedly". The problem has been fixed by closing all cached lookups
5981 in a queue runner before running a delivery process.
5983 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
5984 seem to have the "const" qualifier which it has on other OS. I've
5987 8. Change 4.21/2 was too strict. It is only if there are two authenticators
5988 *of the same type* (client or server) with the same public name that an
5989 error should be diagnosed.
5991 9. When Exim looked up a host name for an IP address, but failed to find the
5992 original IP address when looking up the host name (a safety check), it
5993 output the message "<ip address> does not match any IP for NULL", which was
5994 confusing, to say the least. The bug was that the host name should have
5995 appeared instead of "NULL".
5997 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
5998 user that is built into the binary, and the -C or -D options is used, root
5999 privilege is dropped before the configuration file is read. In addition,
6000 logging is switched to stderr instead of the normal log files. If the
6001 configuration then re-defines the Exim user, the unprivileged environment
6002 is probably not what is expected, so Exim logs a panic warning message (but
6005 However, if deliver_drop_privilege is set, the unprivileged state may well
6006 be exactly what is intended, so the warning has been cut out in that case,
6007 and Exim is allowed to try to write to its normal log files.
6013 1. smtp_return_error_details was not giving details for temporary sender
6014 or receiver verification errors.
6016 2. Diagnose a configuration error if two authenticators have the same public
6019 3. Exim used not to create the message log file for a message until the first
6020 delivery attempt. This could be confusing when incoming messages were held
6021 for policy or load reasons. The message log file is now created at the time
6022 the message is received, and an initial "Received" line is written to it.
6024 4. The automatically generated man page for command line options had a minor
6025 bug that caused no ill effects; however, a more serious problem was that
6026 the procedure for building the man page automatically didn't always
6027 operate. Consequently, release 4.20 contains an out-of-date version. This
6028 shouldn't happen again.
6030 5. When building Exim with embedded Perl support, the script that builds the
6031 Makefile was calling 'perl' to find its compile-time parameters, ignoring
6032 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
6034 6. The freeze_tell option was not being used for messages that were frozen on
6035 arrival, either by an ACL or by local_scan().
6037 7. Added the smtp_incomplete_transaction log selector.
6039 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
6040 was accepting AUTH without a new EHLO.
6042 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
6043 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
6046 10. Logging of TCP/IP connections (when configured) now happens in the main
6047 daemon process instead of the child process, so that the TCP/IP connection
6048 count is more accurate (but it can never be perfect).
6050 11. The use of "drop" in a nested ACL was not being handled correctly in the
6051 outer ACL. Now, if condition failure induced by the nested "drop" causes
6052 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
6053 or "require"), the connection is dropped.
6055 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
6056 that yield "discard" can now be used with an "accept" or a "discard" verb,
6057 but an error is generated for any others (because I can't see a useful way
6058 to define what should happen).
6060 13. When an ACL is read dynamically from a file (or anywhere else), the lines
6061 are now processed in the same way as lines in the Exim configuration file.
6062 In particular, continuation lines are supported.
6064 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
6066 15. Added -ti meaning -t -i.
6068 16. Check for letters, digits, hyphens, and dots in the names of dnslist
6069 domains, and warn by logging if others are found.
6071 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
6072 returned from GIFCONF when Exim is trying to find the list of interfaces on
6073 a host. The code in os.c has been modified to copy each ifreq to an aligned
6074 structure in all cases.
6076 Also, in some cases, the returned ifreq's were being copied to a 'struct
6077 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
6078 means the last couple of bytes of an IPv6 address could be chopped if the
6079 ifreq contained only a normal sockaddr (14 bytes storage).
6081 18. Named domain lists were not supported in the hosts_treat_as_local option.
6082 An entry such as +xxxx was not recognized, and was treated as a literal
6085 19. Ensure that header lines added by a DATA ACL are included in the reject log
6086 if the ACL subsequently rejects the message.
6088 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
6089 MD5 (which is deprecated).
6091 21. When testing a filter file using -bf, Exim was writing a message when it
6092 took the sender from a "From " line in the message, but it was not doing so
6093 when it took $return_path from a Return-Path: header line. It now does.
6095 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
6096 with a valid header line field name (a series of printing characters
6097 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
6099 23. Changed "disc" in the source to "disk" to conform to the documentation and
6100 the book and for uniformity.
6102 24. Ignore Sendmail's -Ooption=value command line item.
6104 25. When execve() failed while trying to run a command in a pipe transport,
6105 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
6106 could be confused with a return value of 69 from the command itself. This
6107 has been changed to 127, the value the shell returns if it is asked to run
6108 a non-existent command. The wording for the related log line suggests a
6109 non-existent command as the problem.
6111 26. If received_header_text expands to an empty string, do not add a Received:
6112 header line to the message. (Well, it adds a token one on the spool, but
6113 marks it "old" so that it doesn't get used or transmitted.)
6115 27. Installed eximstats 1.28 (addition of -nt option).
6117 28. There was no check for failure on the call to getsockname() in the daemon
6118 code. This can fail if there is a shortage of resources on the system, with
6119 ENOMEM, for example. A temporary error is now given on failure.
6121 29. Contrary to the C standard, it seems that in some environments, the
6122 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
6123 program. Exim now does this explicitly; it affects the formatting of
6124 timestamps using strftime().
6126 30. If exiqsumm was given junk data, it threw up some uninitialized variable
6127 complaints. I've now initialized all the variables, to avoid this.
6129 32. Header lines added by a system filter were not being "seen" during
6130 transport-time rewrites.
6132 33. The info_callback() function passed to OpenSSL is set up with type void
6133 (*)(SSL *, int, int), as described somewhere. However, when calling the
6134 function (actually a macro) that sets it up, the type void(*)() is
6135 expected. I've put in a cast to prevent warnings from picky compilers.
6137 34. If a DNS black list lookup found a CNAME record, but there were no A
6138 records associated with the domain it pointed at, Exim crashed.
6140 35. If a DNS black list lookup returned more than one A record, Exim ignored
6141 all but the first. It now scans all returned addresses if a particular IP
6142 value is being sought. In this situation, the contents of the
6143 $dnslist_value variable are a list of all the addresses, separated by a
6146 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
6147 to accept a host name and all its aliases if the forward lookup for any of
6148 them yielded the IP address of the incoming connection. Now it accepts only
6149 those names whose forward lookup yields the correct IP address. Any other
6150 names are discarded. This closes a loophole whereby a rogue DNS
6151 administrator could create reverse DNS records to break through a
6152 wildcarded host restriction in an ACL.
6154 37. If a user filter or a system filter that ran in a subprocess used any of
6155 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
6156 the wrong values were passed to the pipe command ($thisaddress had the
6157 value of $0, $0 had the value of $1, etc). This bug was introduced by
6158 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
6160 38. Improved the line breaking for long SMTP error messages from ACLs.
6161 Previously, if there was no break point between 40 and 75 characters, Exim
6162 left the rest of the message alone. Two changes have been made: (a) I've
6163 reduced the minimum length to 35 characters; (b) if it can't find a break
6164 point between 35 and 75 characters, it looks ahead and uses the first one
6165 that it finds. This may give the occasional overlong line, but at least the
6166 remaining text gets split now.
6168 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
6169 file descriptors might be low, and that setting 1000 would always raise it.
6170 It turns out that in some environments, the limit is already over 1000 and
6171 that lowering it causes trouble. So now Exim takes care not to decrease it.
6173 40. When delivering a message, the value of $return_path is set to $sender_
6174 address at the start of routing (routers may change the value). By an
6175 oversight, this default was not being set up when an address was tested by
6176 -bt or -bv, which affected the outcome if any router or filter referred to
6179 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
6180 log only when "message" or "log_message" are set. However, if one of the
6181 conditions was an address verification, or a call to a nested ACL, the
6182 messages generated by the underlying test were being passed through. This
6183 no longer happens. The underlying message is available in $acl_verify_
6184 message for both "message" and "log_message" expansions, so it can be
6185 passed through if needed.
6187 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
6188 new expansion $bh_ to give the encoded byte string without charset
6189 translation. Translation happens only if iconv() is available; HAVE_ICONV
6190 indicates this at build time. HEADERS_CHARSET gives the charset to
6191 translate to; headers_charset can change it in the configuration, and
6192 "headers charset" can change it in an individual filter file.
6194 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
6195 that creates RFC 2047 encoded "words" labels them as that charset instead
6196 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
6197 expansion operator; (ii) when Exim creates a From: line for a local
6198 message; (iii) when a header line is rewritten to include a "phrase" part.
6200 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
6201 buggy, causing it to skip the first lines of messages whose message ID
6202 ended in 'D'. This would not have bitten before Exim release 4.14, because
6203 message IDs were unlikely to end in 'D' before then. The effect was to have
6204 incorrect size information for certain domains.
6206 45. #include "config.h" was missing at the start of the crypt16.c module. This
6207 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
6210 46. If there was a timeout during a "random" callout check, Exim treated it as
6211 a failure of the random address, and carried on sending RSET and the real
6212 address. If the delay was just some slowness somewhere, the response to the
6213 original RCPT would be taken as a response to RSET and so on, causing
6214 mayhem of various kinds.
6216 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
6217 when I implemented it. It didn't allow for the fact that some option values
6218 may legitimatetly be negative (e.g. size_addition), and it didn't even do
6219 the right test for positive values.
6221 48. Domain names in DNS records are case-independent. Exim always looks them up
6222 in lower case. Some resolvers return domain names in exactly the case they
6223 appear in the zone file, that is, they may contain uppercase letters. Not
6224 all resolvers do this - some return always lower case. Exim was treating a
6225 change of case by a resolver as a change of domain, similar to a widening
6226 of a domain abbreviation. This triggered its re-routing code and so it was
6227 trying to route what was effectively the same domain again. This normally
6228 caused routing to fail (because the router wouldn't handle the domain
6229 twice). Now Exim checks for this case specially, and just changes the
6230 casing of the domain that it ultimately uses when it transmits the message
6233 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
6236 50. If a filter generated a file delivery with a non-absolute name (possible if
6237 no home directory exists for the router), the forbid_file option was not
6240 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
6241 the existing equality matching.
6243 52. Exim was using ints instead of ino_t variables in some places where it was
6244 dealing with inode numbers.
6246 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
6247 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
6248 called TMPDIR, and if it finds it is different, it changes its value.
6250 54. The smtp_printf() function is now made available to local_scan() so
6251 additional output lines can be written before returning. There is also an
6252 smtp_fflush() function to enable the detection of a dropped connection.
6253 The variables smtp_input and smtp_batched_input are exported to
6256 55. Changed the default runtime configuration: the message "Unknown user"
6257 has been removed from the ACL, and instead placed on the localuser router,
6258 using the cannot_route_message feature. This means that any verification
6259 failures that generate their own messages won't get overridden. Similarly,
6260 the "Unrouteable address" message that was in the ACL for unverifiable
6261 relay addresses has also been removed.
6263 56. Added hosts_avoid_esmtp to the smtp transport.
6265 57. The exicyclog script was not checking for the esoteric option
6266 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
6267 will work only if exicyclog is run under the appropriate euid.
6269 58. Following a discussion on the list, the rules by which Exim recognises line
6270 endings on incoming messages have been changed. The -dropcr and drop_cr
6271 options are now no-ops, retained only for backwards compatibility. The
6272 following line terminators are recognized: LF CRLF CR. However, special
6273 processing applies to CR:
6275 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
6276 nor a local message in the state where . is a terminator.
6278 (ii) If a bare CR is encountered in a header line, an extra space is added
6279 after the line terminator so as not to end the header. The reasoning
6280 behind this is that bare CRs in header lines are most likely either
6281 to be mistakes, or people trying to play silly games.
6283 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
6284 was being incorrectly given as 18 bytes larger than it should have been.
6285 This is a VOB (very old bug).
6287 60. This may never have affected anything current, but just in case it has:
6288 When the local host is found other than at the start of a list of hosts,
6289 the local host, those with the same MX, and any that follow, are discarded.
6290 When the list in question was part of a longer list of hosts, the following
6291 hosts (not currently being processed) were also being discarded. This no
6292 longer happens. I'm not sure if this situation could ever has previously
6295 61. Added the "/MX" feature to lists of hosts in the manualroute and query
6298 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
6299 header. This is something that is recommended in a new Internet Draft, and
6300 is something that is documented as being done by Sendmail. There are two
6301 possible values. For messages generated by the autoreply transport, Exim
6304 Auto-Submitted: auto-replied
6306 whereas for all other generated messages (e.g. bounces) it adds
6308 Auto-Submitted: auto-generated
6310 63. The "personal" condition in filters now includes a test for the
6311 Auto-Submitted: header. If it contains the string "auto-" the message it
6312 not considered personal.
6314 64. Added rcpt_include_affixes as a generic transport option.
6316 65. Added queue_only_override (default true).
6318 66. Added the syslog_duplication option.
6320 67. If what should have been the first header line of a message consisted of
6321 a space followed by a colon, Exim was mis-interpreting it as a header line.
6322 It isn't of course - it is syntactically invalid and should therefore be
6323 treated as the start of the message body. The misbehaviour could have
6324 caused a number of strange effects, including loss of data in subsequent
6325 header lines, and spool format errors.
6327 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
6328 client host had authenticated. This control can now be exercised by an ACL
6329 for more flexibility.
6331 69. By default, callouts do not happen when testing with -bh. There is now a
6332 variant, -bhc, which does actually run the callout code, including
6333 consulting and updating the callout cache.
6335 70. Added support for saslauthd authentication, courtesy of Alexander
6338 71. If statvfs() failed on the spool or log directories while checking their
6339 size for availability, Exim confusingly gave the error "space shortage".
6340 Furthermore, in debugging mode it crashed with a floating point exception.
6341 These checks are done if check_{spool,log}_{space,inodes} are set, and when
6342 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
6343 serious problem, Exim now writes to the main and panic logs when this
6344 happens, with details of the failure. It then refuses to accept the
6345 incoming message, giving the message "spool directory problem" or "log
6346 directory problem" with a 421 code for SMTP messages.
6348 72. When Exim is about to re-exec itself, it ensures that the file descriptors
6349 0, 1, and 2 exist, because some OS complain for execs without them (see
6350 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
6351 descriptors. However, the code omitted to check that the open succeeded,
6352 causing mysterious errors if for some reason the permissions on /dev/null
6353 got screwed. Now Exim writes a message to the main and panic logs, and
6354 bombs out if it can't open /dev/null.
6356 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
6357 interact so that it is all more flexible. It is supposed to remain
6358 backwards compatible. Also added extra_local_interfaces.
6360 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
6361 to bomb out with an assertion failure - to the client this appears as a
6362 connection drop. This problem occurs in the part of the code that was taken
6363 from the Samba project. Fortunately, the assertion is in a very simple
6364 function, so I have fixed this by reproducing the function inline in the
6365 one place where it is called, and arranging for authentication to fail
6366 instead of killing the process with assert().
6368 75. The SPA client code was not working when the server requested OEM rather
6369 than Unicode encoding.
6371 76. Added code to make require_files with a specific uid setting more usable in
6372 the case where statting the file as root fails - usually a non-root-mounted
6373 NFS file system. When this happens and the failure is EACCES, Exim now
6374 forks a subprocess and does the per-uid checking as the relevant uid.
6376 77. Added process_log_path.
6378 78. If log_file_path was not explicitly set, a setting of check_log_space or
6379 check_log_inodes was ignored.
6381 79. If a space check for the spool or log partitions fails, the incident is now
6382 logged. Of course, in the latter case the data may get lost...
6384 80. Added the %p formatting code to string_format() so that it can be used to
6385 print addresses in debug_print(). Adjusted all the address printing in the
6386 debugging in store.c to use %p rather than %d.
6388 81. There was a concern that a line of code in smtp_in.c could overflow a
6389 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
6390 initially expressed, the concern was not well-founded, because trailing
6391 spaces are removed early. However, if the trailing spaces were followed by
6392 a NULL, they did not get removed, so the overflow was possible. Two fixes
6395 (a) I re-wrote the offending code in a cleaner fashion.
6396 (b) If an incoming SMTP command contains a NULL character, it is rejected
6399 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
6400 runs initgroups(), so that if the Exim user is in any additional groups,
6401 they will be used during message reception.
6407 The change log for 4.20 and earlier releases has been archived.