1 Change log file for Exim from version 4.21
2 -------------------------------------------
7 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
8 and 3.4.4 - once the server is enabled to respond to an OCSP request
9 it does even when not requested, resulting in a stapling non-aware
10 client dropping the TLS connection.
12 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
13 support variable-length bit vectors. No functional change.
15 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
16 The I= interface field on outgoing lines is now after the H= remote
17 host field, same as incoming lines. There is a separate
18 outgoing_interface log selector which allows you to disable the
21 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
22 If not running log_selector +smtp_connection the mainlog would be held
23 open indefinitely after a "too many connections" event, including to a
24 deleted file after a log rotate. Leave the per net connection logging
25 leaving it open for efficiency as that will be quickly detected by the
26 check on the next write.
28 HS/01 Bug 1671: Fix post transport crash.
29 Processing the wait-<transport> messages could crash the delivery
30 process if the message IDs didn't exist for some reason. When
31 using 'split_spool_directory=yes' the construction of the spool
32 file name failed already, exposing the same netto behaviour.
34 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
35 mime_regex ACL conditions.
37 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
38 to DSN fail messages (bounces): remote IP, remote greeting, remote response
39 to HELO, local diagnostic string.
41 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
42 log line to debug. Even when configured with a tls authenticator many
43 client connections are expected to not authenticate in this way, so
44 an authenticate fail is not an error.
46 HS/02 Add the Exim version string to the process info. This way exiwhat
47 gives some more detail about the running daemon.
49 JH/06 Bug 1395: time-limit cacheing of DNS lookups, to the TTL value. This may
50 matter for fast-change records such as DNSBLs.
52 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
53 retry record, even if constant. There may be multiple transports with
54 different interface settings and the retry behaviour needs to be kept
57 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
59 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
61 JH/10 Bug 840: fix log_defer_output option of pipe transport
63 JH/11 Bug 830: use same host for all RCPTS of a message, even under
64 hosts_randomize. This matters a lot when combined with mua_wrapper.
66 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
67 ${quote_pgsql:<string>} operator.
69 JH/13 Bug 1708: avoid misaligned access in cached lookup.
71 JH/14 Change header file name for freeradius-client. Relevant if compiling
72 with Radius support; from the Gentoo tree and checked under Fedora.
74 JH/15 Bug 1712: Introduce $prdr_requested flag variable
76 JH/16 Bug 1714: Permit an empty string as expansion result for transport
77 option transport_filter, meaning no filtering.
79 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
81 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
82 defaults to "*" (all hosts). The variable is now available when not built
83 with TLS, default unset, mainly to enable keeping the testuite sane.
84 If a server certificate is not supplied (via tls_certificate) an error is
85 logged, and clients will find TLS connections fail on startup. Presumably
86 they will retry in-clear.
87 Packagers of Exim are strongly encouraged to create a server certificate
90 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
91 with the $config_file variable.
93 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
94 in transport context, after the attempt, and per-recipient. The latter type
95 is per host attempted. The event data is the error message, and the errno
96 information encodes the lookup type (A vs. MX) used for the (first) host,
97 and the trailing two digits of the smtp 4xx reponse.
99 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
100 to write to mainlog (or rejectlog, paniclog) in the window between file
101 creation and permissions/ownership being changed. Particularly affects
102 installations where exicyclog is run as root, rather than exim user;
103 result is that the running daemon panics and dies.
105 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
107 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
108 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
109 "pri" and "weight". Note that the previous implicit priority given by the
110 list order is no longer honoured.
112 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalisation
115 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
116 by defining SUPPORT_SOCKS.
118 JH/26 Move PROXY support from Experimental to mainline, enabled for a build
119 by defining SUPPORT_PROXY. Note that the proxy_required_hosts option
120 is renamed to hosts_proxy, and the proxy_{host,target}_{address,port}.
121 variables are renamed to proxy_{local,external}_{address,port}.
123 JH/27 Move Internationalisation support from Experimental to mainline, enabled
124 for a build by defining SUPPORT_I18N
129 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
132 JH/02 The smtp transport option "multi_domain" is now expanded.
134 JH/03 The smtp transport now requests PRDR by default, if the server offers
137 JH/04 Certificate name checking on server certificates, when exim is a client,
138 is now done by default. The transport option tls_verify_cert_hostnames
139 can be used to disable this per-host. The build option
140 EXPERIMENTAL_CERTNAMES is withdrawn.
142 JH/05 The value of the tls_verify_certificates smtp transport and main options
143 default to the word "system" to access the system default CA bundle.
144 For GnuTLS, only version 3.0.20 or later.
146 JH/06 Verification of the server certificate for a TLS connection is now tried
147 (but not required) by default. The verification status is now logged by
148 default, for both outbound TLS and client-certificate supplying inbound
151 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
154 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
155 Status Notification (bounce) messages are now MIME format per RFC 3464.
156 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
157 under the control of the dsn_advertise_hosts option, and routers may
158 have a dsn_lasthop option.
160 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
161 default, modifiable by a malware= option. The list separator for
162 the options can now be changed in the usual way. Bug 68.
164 JH/10 The smtp_receive_timeout main option is now expanded before use.
166 JH/11 The incoming_interface log option now also enables logging of the
167 local interface on delivery outgoing connections.
169 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
170 if the interface and destination host and port all match.
172 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
175 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
176 Patch from Andrew Lewis.
178 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
179 now supports optional time-restrictions, weighting, and priority
180 modifiers per server. Patch originally by <rommer@active.by>.
182 JH/16 The spamd_address main option now supports a mixed list of local
183 and remote servers. Remote servers can be IPv6 addresses, and
184 specify a port-range.
186 JH/17 Bug 68: The spamd_address main option now supports an optional
187 timeout value per server.
189 JH/18 Bug 1581: Router and transport options headers_add/remove can
190 now have the list separator specified.
192 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
195 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
198 JH/21 Support for the A6 type of dns record is withdrawn.
200 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
201 rather than the verbs used.
203 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
204 from 255 to 1024 chars.
206 JH/24 Verification callouts now attempt to use TLS by default.
208 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
209 are generic router options now. The defaults didn't change.
211 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
212 Original patch from Alexander Shikoff, worked over by JH.
214 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
215 files if ZCAT_COMMAND is not executable.
217 JH/26 Bug 1539: Add timout/retry options on dnsdb lookups.
219 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
221 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
222 Normally benign, it bites when the pair was led to by a CNAME;
223 modern usage is to not canoicalize the domain to a CNAME target
224 (and we were inconsistent anyway for A-only vs AAAA+A).
226 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
228 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
229 when evaluating $sender_host_dnssec.
231 JH/31 Check the HELO verification lookup for DNSSEC, adding new
232 $sender_helo_dnssec variable.
234 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
236 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
238 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
240 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
241 documented as working, but never had. Support all but $spam_report.
243 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
244 added for tls authenticator.
249 TL/01 When running the test suite, the README says that variables such as
250 no_msglog_check are global and can be placed anywhere in a specific
251 test's script, however it was observed that placement needed to be near
252 the beginning for it to behave that way. Changed the runtest perl
253 script to read through the entire script once to detect and set these
254 variables, reset to the beginning of the script, and then run through
255 the script parsing/test process like normal.
257 TL/02 The BSD's have an arc4random API. One of the functions to induce
258 adding randomness was arc4random_stir(), but it has been removed in
259 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
260 function when detected.
262 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
263 cause callback expansion.
265 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
266 syntax errors in an expansion can be treated as a string instead of
267 logging or causing an error, due to the internal use of bool_lax
268 instead of bool when processing it.
270 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
271 server certificates when making smtp deliveries.
273 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
275 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
277 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
279 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
280 Merged patch from Sebastian Wiedenroth.
282 JH/05 Fix results-pipe from transport process. Several recipients, combined
283 with certificate use, exposed issues where response data items split
284 over buffer boundaries were not parsed properly. This eventually
285 resulted in duplicates being sent. This issue only became common enough
286 to notice due to the introduction of conection certificate information,
287 the item size being so much larger. Found and fixed by Wolfgang Breyha.
289 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
290 size buffer was used, resulting in syntax errors when an expansion
293 JH/07 Add support for directories of certificates when compiled with a GnuTLS
294 version 3.3.6 or later.
296 JH/08 Rename the TPDA expermimental facility to Event Actions. The #ifdef
297 is EXPERIMENTAL_EVENT, the main-configuration and transport options
298 both become "event_action", the variables become $event_name, $event_data
299 and $event_defer_errno. There is a new variable $verify_mode, usable in
300 routers, transports and related events. The tls:cert event is now also
301 raised for inbound connections, if the main configuration event_action
304 TL/06 In test suite, disable OCSP for old versions of openssl which contained
305 early OCSP support, but no stapling (appears to be less than 1.0.0).
307 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
308 server certificate names available under the smtp transport option
309 "tls_verify_cert_hostname" now do not permit multi-component wildcard
312 JH/10 Time-related extraction expansions from certificates now use the main
313 option "timezone" setting for output formatting, and are consistent
314 between OpenSSL and GnuTLS compilations. Bug 1541.
316 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
317 encoded parameter in the incoming message. Bug 1558.
319 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
320 include certificate info, eximon was claiming there were spoolfile
323 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
325 JH/14 Log delivery-related information more consistently, using the sequence
326 "H=<name> [<ip>]" wherever possible.
328 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
329 are problematic for Debian distribution, omit them from the release
332 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
334 JH/16 Fix string representation of time values on 64bit time_t anchitectures.
337 JH/17 Fix a null-indirection in certextract expansions when a nondefault
338 output list separator was used.
343 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
344 checkers that were complaining about end of non-void function with no
347 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
348 This was a regression intruduced in 4.83 by another bugfix.
350 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
352 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
353 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
359 TF/01 Correctly close the server side of TLS when forking for delivery.
361 When a message was received over SMTP with TLS, Exim failed to clear up
362 the incoming connection properly after forking off the child process to
363 deliver the message. In some situations the subsequent outgoing
364 delivery connection happened to have the same fd number as the incoming
365 connection previously had. Exim would try to use TLS and fail, logging
366 a "Bad file descriptor" error.
368 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
369 utilities have not been installed.
371 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
372 temporary space as the ACL may create new global variables.
374 TL/01 LDAP support uses per connection or global context settings, depending
375 upon the detected version of the libraries at build time.
377 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
378 to extract and use the src ip:port in logging and expansions as if it
379 were a direct connection from the outside internet. PPv2 support was
380 updated based on HAProxy spec change in May 2014.
382 JH/02 Add ${listextract {number}{list}{success}{fail}}.
384 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
385 Properly escape header and check for NULL return.
387 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
390 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
392 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
393 characters in header names, implemented as a verify condition.
394 Contributed by Michael Fischer v. Mollard.
396 TL/05 Rename SPF condition results err_perm and err_temp to standardized
397 results permerror and temperror. Previous values are deprecated but
398 still accepted. In a future release, err_perm and err_temp will be
399 completely removed, which will be a backward incompatibility if the
400 ACL tests for either of these two old results. Patch contributed by
401 user bes-internal on the mailing list.
403 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
405 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
406 selectors, in both main and reject logs.
408 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
411 JH/07 Add malware type "sock" for talking to simple daemon.
413 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
415 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
416 routers/transports under cutthrough routing.
418 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
419 numbers. Touch up "bool" conditional to keep the same definition.
421 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
423 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
425 JH/12 Expand items in router/transport headers_add or headers_remove lists
426 individually rather than the list as a whole. Bug 1452.
428 Required for reasonable handling of multiple headers_ options when
429 they may be empty; requires that headers_remove items with embedded
430 colons must have them doubled (or the list-separator changed).
432 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
433 view the policy declared in the DMARC record. Currently, $dmarc_status
434 is a combined value of both the record presence and the result of the
437 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
439 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
440 dnslookup router and the smtp transport (applying to the forward
443 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
444 of ldap servers used for a specific lookup. Patch provided by Heiko
447 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
448 New variable $lookup_dnssec_authenticated for observability.
450 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
451 Patch submitted by Lars Timman.
453 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
455 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
456 Requires trusted mode and valid format message id, aborts otherwise.
457 Patch contributed by Heiko Schlichting.
459 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
460 certextract with support for various fields. Bug 1358.
462 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
463 is requested by default, modifiable by smtp transport option
466 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
467 operate on certificate variables to give certificate fingerprints
468 Also new ${sha256:cert_variable}.
470 JH/23 The PRDR feature is moved from being Experimental into the mainline.
472 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
473 Christian Aistleitner.
475 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
477 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
478 file. Patch from Wolfgang Breyha.
480 JH/25 Expand the coverage of the delivery $host and $host_address to
481 client authenticators run in verify callout. Bug 1476.
483 JH/26 Port service names are now accepted for tls_on_connect_ports, to
484 align with daemon_smtp_ports. Bug 72.
486 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
487 support and error reporting did not work properly.
489 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
490 and is readable. Patch from Andrew Colin Kissa.
492 TL/14 Enhance documentation of ${run expansion and how it parses the
493 commandline after expansion, particularly in the case when an
494 unquoted variable expansion results in an empty value.
496 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
498 PP/02 Fix internal collision of T_APL on systems which support RFC3123
499 by renaming away from it. Addresses GH issue 15, reported by
502 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
504 TL/15 SECURITY: prevent double expansion in math comparison functions
505 (can expand unsanitized data). Not remotely exploitable.
512 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
514 PP/02 Make -n do something, by making it not do something.
515 When combined with -bP, the name of an option is not output.
517 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
520 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
521 $sender_host_name and config options to manage this, and basic check
524 PP/05 DSCP support for outbound connections and control modifier for inbound.
526 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
527 (Only plugin which currently uses this is kerberos4, which nobody should
528 be using, but we should make it available and other future plugins might
529 conceivably use it, even though it would break NAT; stuff *should* be
530 using channel bindings instead).
532 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
533 name; added for Sendmail compatibility; requires admin caller.
534 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
535 just ignore it); requires trusted caller.
536 Also parse but ignore: -Ac -Am -X<logfile>
539 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
541 TL/02 Add +smtp_confirmation as a default logging option.
543 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
544 Patch by Magnus Holmgren from 2007-02-20.
546 TL/04 Bugzilla 1281 - Spec typo.
547 Bugzilla 1283 - Spec typo.
548 Bugzilla 1290 - Spec grammar fixes.
550 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
552 TL/06 Add Experimental DMARC support using libopendmarc libraries.
554 TL/07 Fix an out of order global option causing a segfault. Reported to dev
555 mailing list by by Dmitry Isaikin.
557 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
559 JH/02 Support "G" suffix to numbers in ${if comparisons.
561 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
563 NM/01 Bugzilla 1197 - Spec typo
564 Bugzilla 1196 - Spec examples corrections
566 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
568 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
569 gnutls_enable_pkcs11, but renamed to more accurately indicate its
572 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
573 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
575 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
576 "acl {{name}{arg}...}", and optional args on acl condition
579 JH/05 Permit multiple router/transport headers_add/remove lines.
581 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
583 JH/07 Avoid using a waiting database for a single-message-only transport.
584 Performance patch from Paul Fisher. Bugzilla 1262.
586 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
589 JH/09 Add $headers_added variable, with content from use of ACL modifier
590 add_header (but not yet added to the message). Bugzilla 199.
592 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
593 Pulled from Bugzilla 817 by Wolfgang Breyha.
595 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
597 (nb: this is the same fix as in Exim 4.80.1)
599 JH/11 Add A= logging on delivery lines, and a client_set_id option on
602 JH/12 Add optional authenticated_sender logging to A= and a log_selector
605 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
607 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
608 advertise SMTP AUTH mechanism to us, instead of a generic
609 protocol violation error. Also, make Exim more robust to bad
610 data from the Dovecot auth socket.
612 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
614 When a queue runner is handling a message, Exim first routes the
615 recipient addresses, during which it prunes them based on the retry
616 hints database. After that it attempts to deliver the message to
617 any remaining recipients. It then updates the hints database using
620 So if a recipient address works intermittently, it can get repeatedly
621 deferred at routing time. The retry hints record remains fresh so the
622 address never reaches the final cutoff time.
624 This is a fairly common occurrence when a user is bumping up against
625 their storage quota. Exim had some logic in its local delivery code
626 to deal with this. However it did not apply to per-recipient defers
627 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
629 This change adds a proper retry rule check during routing so that the
630 final cutoff time is checked against the message's age. We only do
631 this check if there is an address retry record and there is not a
632 domain retry record; this implies that previous attempts to handle
633 the address had the retry_use_local_parts option turned on. We use
634 this as an approximation for the destination being like a local
635 delivery, as in LMTP.
637 I suspect this new check makes the old local delivery cutoff check
638 redundant, but I have not verified this so I left the code in place.
640 TF/02 Correct gecos expansion when From: is a prefix of the username.
642 Test 0254 submits a message to Exim with the header
646 When I ran the test suite under the user fanf2, Exim expanded
647 the header to contain my full name, whereas it should have added
648 a Resent-Sender: header. It erroneously treats any prefix of the
649 username as equal to the username.
651 This change corrects that bug.
653 GF/01 DCC debug and logging tidyup
654 Error conditions log to paniclog rather than rejectlog.
655 Debug lines prefixed by "DCC: " to remove any ambiguity.
657 TF/03 Avoid unnecessary rebuilds of lookup-related code.
659 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
660 Bug spotted by Jeremy Harris; was flawed since initial commit.
661 Would have resulted in OCSP responses post-SNI triggering an Exim
662 NULL dereference and crash.
664 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
666 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
667 Bug detection, analysis and fix by Samuel Thibault.
668 Bugzilla 1331, Debian bug #698092.
670 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
672 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
673 Server implementation by Todd Lyons, client by JH.
674 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
675 config variable "prdr_enable" controls whether the server
676 advertises the facility. If the client requests PRDR a new
677 acl_data_smtp_prdr ACL is called once for each recipient, after
678 the body content is received and before the acl_smtp_data ACL.
679 The client is controlled by bolth of: a hosts_try_prdr option
680 on the smtp transport, and the server advertisement.
681 Default client logging of deliveries and rejections involving
682 PRDR are flagged with the string "PRDR".
684 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
685 fclose(). Diagnosis by Todd Lyons.
687 PP/17 Update configure.default to handle IPv6 localhost better.
688 Patch by Alain Williams (plus minor tweaks).
691 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
692 This is now consistent with GnuTLS, and is now documented: the
693 previous undocumented portable approach to treating the option as
694 unset was to force an expansion failure. That still works, and
695 an empty string is now equivalent.
697 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
698 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
699 not performing validation itself.
701 PP/20 Added force_command boolean option to pipe transport.
702 Patch from Nick Koston, of cPanel Inc.
704 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
707 TF/04 Added udpsend ACL modifer and hexquote expansion operator
709 PP/21 Fix eximon continuous updating with timestamped log-files.
710 Broken in a format-string cleanup in 4.80, missed when I repaired the
711 other false fix of the same issue.
712 Report and fix from Heiko Schlichting.
715 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
716 Report from Prashanth Katuri.
718 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
719 It's SecureTransport, so affects any MacOS clients which use the
720 system-integrated TLS libraries, including email clients.
722 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
723 using a MIME ACL for non-SMTP local injection.
724 Report and assistance in diagnosis by Warren Baker.
726 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
728 JH/16 Fix comparisons for 64b. Bugzilla 1385.
730 TL/09 Add expansion variable $authenticated_fail_id to keep track of
731 last id that failed so it may be referenced in subsequent ACL's.
733 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
736 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
737 ldap library initialization, allowing self-signed CA's to be
738 used. Also properly sets require_cert option later in code by
739 using NULL (global ldap config) instead of ldap handle (per
740 session). Bug diagnosis and testing by alxgomz.
742 TL/12 Enhanced documentation in the ratelimit.pl script provided in
743 the src/util/ subdirectory.
745 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
746 renamed to Transport Post Delivery Action by Jeremy Harris, as
749 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
750 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
751 redis_servers = needs to be configured which will be used by the redis
752 lookup. Patch from Warren Baker, of The Packet Hub.
754 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
756 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
757 hostname or reverse DNS when processing a host list. Used suggestions
758 from multiple comments on this bug.
760 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
762 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
763 Missed a few lines, added it to make the runtest require no keyboard
766 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
767 contains upper case chars. Make router use caseful_local_part.
769 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
770 support when GnuTLS has been built with p11-kit.
776 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
778 This, or similar/improved, will also be change PP/11 of 4.82.
784 PP/01 Handle short writes when writing local log-files.
785 In practice, only affects FreeBSD (8 onwards).
786 Bugzilla 1053, with thanks to Dmitry Isaikin.
788 NM/01 Bugzilla 949 - Documentation tweak
790 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
793 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
795 PP/02 Implemented gsasl authenticator.
797 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
799 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
800 `pkg-config foo` for cflags/libs.
802 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
803 with rest of GSASL and with heimdal_gssapi.
805 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
806 `pkg-config foo` for cflags/libs for the TLS implementation.
808 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
809 properties get this fed in as external SSF. A number of robustness
810 and debugging improvements to the cyrus_sasl authenticator.
812 PP/08 cyrus_sasl server now expands the server_realm option.
814 PP/09 Bugzilla 1214 - Log authentication information in reject log.
815 Patch by Jeremy Harris.
817 PP/10 Added dbmjz lookup type.
819 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
821 PP/12 MAIL args handles TAB as well as SP, for better interop with
822 non-compliant senders.
823 Analysis and variant patch by Todd Lyons.
825 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
826 Bug report from Lars Müller <lars@samba.org> (via SUSE),
827 Patch from Dirk Mueller <dmueller@suse.com>
829 PP/13 tls_peerdn now print-escaped for spool files.
830 Observed some $tls_peerdn in wild which contained \n, which resulted
831 in spool file corruption.
833 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
834 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
835 or write after TLS renegotiation, which otherwise led to messages
838 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
839 as a tracking header (ie: a signed header comes before the signature).
840 Patch from Wolfgang Breyha.
842 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
843 comma-sep list; embedded commas doubled.
845 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
847 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
849 Report and patch from Dmitry Banschikov.
851 PP/16 Removed "dont_insert_empty_fragments" fron "openssl_options".
852 Removed SSL_clear() after SSL_new() which led to protocol negotiation
853 failures. We appear to now support TLS1.1+ with Exim.
855 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
856 lets Exim select keys and certificates based upon TLS SNI from client.
857 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
858 before an outbound SMTP session. New log_selector, +tls_sni.
860 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
861 NULL dereference. Report and patch from Alun Jones.
863 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
864 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
865 Not seeing resolver debug output on NetBSD, but suspect this is a
866 resolver implementation change.
868 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
869 Left warnings. Added "eximon gdb" invocation mode.
871 PP/21 Defaulting "accept_8bitmime" to true, not false.
873 PP/22 Added -bw for inetd wait mode support.
875 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
876 locate the relevant includes and libraries. Made this the default.
878 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
879 Bugzilla 1246, report and most of solution from Tomasz Kusy.
881 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
882 This may cause build issues on older platforms.
884 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
885 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
886 gnutls_require_mac & gnutls_require_protocols (no longer supported).
887 Added SNI support via GnuTLS too.
888 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
890 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
892 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
893 Applied second patch from Janne, implementing suggestion to default
894 multiple-strings-in-record handling to match SPF spec.
896 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
898 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
899 read-only, out of scope).
900 Patch from Wolfgang Breyha, report from Stuart Northfield.
902 PP/29 Fix three issues highlighted by clang analyser static analysis.
903 Only crash-plausible issue would require the Cambridge-specific
904 iplookup router and a misconfiguration.
905 Report from Marcin Mirosław.
907 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
909 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
910 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
911 As part of this, removing so much warning spew let me fix some minor
912 real issues in debug logging.
914 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
915 assignment on my part. Fixed.
917 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
918 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
919 Janne Snabb (who went above and beyond: thank you).
921 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
922 string otherwise requires a connection and a bunch more work and it's
923 relatively easy to get wrong. Should also expose TLS library linkage
926 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
927 64-bit ${eval} (JH/03).
929 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
930 GNU libc to support some of the 64-bit stuff, should not lead to
931 conflicts. Defined before os.h is pulled in, so if a given platform
932 needs to override this, it can.
934 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
935 protection layer was required, which is not implemented.
936 Bugzilla 1254, patch from Wolfgang Breyha.
938 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
939 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
940 tls_dhparam take prime identifiers. Also unbreak combination of
941 OpenSSL+DH_params+TLSSNI.
943 PP/39 Disable SSLv2 by default in OpenSSL support.
949 PP/01 Solaris build fix for Oracle's LDAP libraries.
950 Bugzilla 1109, patch from Stephen Usher.
952 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
954 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
957 TF/02 Fix a couple more cases where we did not log the error message
958 when unlink() failed. See also change 4.74-TF/03.
960 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
961 lock up or crash if it happened to be inside a call to libc when it
962 got a SIGUSR1 from exiwhat.
964 The SIGUSR1 handler appends the current process status to the process
965 log which is later printed by exiwhat. It used to use the general
966 purpose logging code to do this, but several functions it calls are
967 not safe for signals.
969 The new output code in the SIGUSR1 handler is specific to the process
970 log, and simple enough that it's easy to inspect for signal safety.
971 Removing some special cases also simplifies the general logging code.
972 Removing the spurious timestamps from the process log simplifies
975 TF/04 Improved ratelimit ACL condition.
977 The /noupdate option has been deprecated in favour of /readonly which
978 has clearer semantics. The /leaky, /strict, and /readonly update modes
979 are mutually exclusive. The update mode is no longer included in the
980 database key; it just determines when the database is updated. (This
981 means that when you upgrde Exim will forget old rate measurements.)
983 Exim now checks that the per_* options are used with an update mode that
984 makes sense for the current ACL. For example, when Exim is processing a
985 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
986 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
987 must specify per_mail/readonly. If you omit the update mode it defaults to
988 /leaky where that makes sense (as before) or /readonly where required.
990 The /noupdate option is now undocumented but still supported for
991 backwards compatibility. It is equivalent to /readonly except that in
992 ACLs where /readonly is required you may specify /leaky/noupdate or
993 /strict/noupdate which are treated the same as /readonly.
995 A useful new feature is the /count= option. This is a generalization
996 of the per_byte option, so that you can measure the throughput of other
997 aggregate values. For example, the per_byte option is now equivalent
998 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
1000 The per_rcpt option has been generalized using the /count= mechanism
1001 (though it's more complicated than the per_byte equivalence). When it is
1002 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
1003 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
1004 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
1005 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
1006 Note that using per_rcpt with a non-readonly update mode in more than
1007 one ACL will cause the recipients to be double-counted. (The per_mail
1008 and per_byte options don't have this problem.)
1010 The handling of very low rates has changed slightly. If the computed rate
1011 is less than the event's count (usually one) then this event is the first
1012 after a long gap. In this case the rate is set to the same as this event's
1013 count, so that the first message of a spam run is counted properly.
1015 The major new feature is a mechanism for counting the rate of unique
1016 events. The new per_addr option counts the number of different
1017 recipients that someone has sent messages to in the last time period. It
1018 behaves like per_rcpt if all the recipient addresses are different, but
1019 duplicate recipient addresses do not increase the measured rate. Like
1020 the /count= option this is a general mechanism, so the per_addr option
1021 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
1022 example, measure the rate that a client uses different sender addresses
1023 with the options per_mail/unique=$sender_address. There are further
1024 details in the main documentation.
1026 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
1028 TF/06 Removed a few PCRE remnants.
1030 TF/07 Automatically extract Exim's version number from tags in the git
1031 repository when doing development or release builds.
1033 PP/02 Raise smtp_cmd_buffer_size to 16kB.
1034 Bugzilla 879. Patch from Paul Fisher.
1036 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
1037 Heavily based on revision 40f9a89a from Simon Arlott's tree.
1040 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
1042 PP/05 Variable $av_failed, true if the AV scanner deferred.
1043 Bugzilla 1078. Patch from John Horne.
1045 PP/06 Stop make process more reliably on build failure.
1046 Bugzilla 1087. Patch from Heiko Schlittermann.
1048 PP/07 Make maildir_use_size_file an _expandable_ boolean.
1049 Bugzilla 1089. Patch from Heiko Schlittermann.
1051 PP/08 Handle ${run} returning more data than OS pipe buffer size.
1052 Bugzilla 1131. Patch from Holger Weiß.
1054 PP/09 Handle IPv6 addresses with SPF.
1055 Bugzilla 860. Patch from Wolfgang Breyha.
1057 PP/10 GnuTLS: support TLS 1.2 & 1.1.
1059 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
1062 PP/11 match_* no longer expand right-hand-side by default.
1063 New compile-time build option, EXPAND_LISTMATCH_RHS.
1064 New expansion conditions, "inlist", "inlisti".
1066 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
1068 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
1070 PP/14 fix log_write() format string regression from TF/03.
1071 Bugzilla 1152. Patch from Dmitry Isaikin.
1077 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
1079 PP/02 Harmonised TLS library version reporting; only show if debugging.
1080 Layout now matches that introduced for other libraries in 4.74 PP/03.
1082 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
1084 PP/04 New "dns_use_edns0" global option.
1086 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
1089 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
1090 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
1092 TK/01 Updated PolarSSL code to 0.14.2.
1093 Bugzilla 1097. Patch from Andreas Metzler.
1095 PP/07 Catch divide-by-zero in ${eval:...}.
1096 Fixes bugzilla 1102.
1098 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
1101 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
1102 format-string attack -- SECURITY: remote arbitrary code execution.
1104 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
1105 time unintentionally subject to list matching rules, letting the header
1106 cause arbitrary Exim lookups (of items which can occur in lists, *not*
1107 arbitrary string expansion). This allowed for information disclosure.
1109 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
1110 INT_MIN/-1 -- value coerced to INT_MAX.
1116 NM/01 Workround for PCRE version dependency in version reporting
1119 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
1120 This fixes portability to compilers other than gcc, notably
1121 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
1123 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
1124 makefiles for portability to HP-UX and POSIX correctness.
1126 PP/01 Permit LOOKUP_foo enabling on the make command-line.
1127 Also via indented variable definition in the Makefile.
1128 (Debugging by Oliver Heesakkers).
1130 PP/02 Restore caching of spamd results with expanded spamd_address.
1131 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
1133 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
1134 Improves build reliability. Fix from: Frank Elsner
1136 NM/02 Fix wide character breakage in the rfc2047 coding
1137 Fixes bug 1064. Patch from Andrey N. Oktyabrski
1139 NM/03 Allow underscore in dnslist lookups
1140 Fixes bug 1026. Patch from Graeme Fowler
1142 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
1143 Code patches from Adam Ciarcinski of NetBSD.
1145 NM/04 Fixed exiqgrep to cope with mailq missing size issue
1148 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
1149 is logged, to avoid truncation. Patch from John Horne.
1151 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
1152 Patch from Jakob Hirsch.
1154 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
1155 SQL string expansion failure details.
1156 Patch from Andrey Oktyabrski.
1158 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
1159 Patch from Simon Arlott.
1161 PP/09 New lookups functionality failed to compile on old gcc which rejects
1162 extern declarations in function scope.
1163 Patch from Oliver Fleischmann
1165 PP/10 Use sig_atomic_t for flags set from signal handlers.
1166 Check getgroups() return and improve debugging.
1167 Fixed developed for diagnosis in bug 927 (which turned out to be
1170 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
1171 Patch from Mark Zealey.
1173 PP/12 Bugzilla 1056: Improved spamd server selection.
1174 Patch from Mark Zealey.
1176 PP/13 Bugzilla 1086: Deal with maildir quota file races.
1177 Based on patch from Heiko Schlittermann.
1179 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
1180 Patch from Uwe Doering, sign-off by Michael Haardt.
1182 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
1183 variable declaration deep within a block. Bug and patch from
1186 PP/15 lookups-Makefile IRIX compatibilty coercion.
1188 PP/16 Make DISABLE_DKIM build knob functional.
1190 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
1191 Patch by Simon Arlott
1193 TF/03 Fix valgrind.h portability to C89 compilers that do not support
1194 variable argument macros. Our copy now differs from upstream.
1200 TF/01 Failure to get a lock on a hints database can have serious
1201 consequences so log it to the panic log.
1203 TF/02 Log LMTP confirmation messages in the same way as SMTP,
1204 controlled using the smtp_confirmation log selector.
1206 TF/03 Include the error message when we fail to unlink a spool file.
1208 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
1209 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
1210 for maintaining out-of-tree patches for some time.
1212 PP/01 Bugzilla 139: Documentation and portability issues.
1213 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
1214 Handle per-OS dynamic-module compilation flags.
1216 PP/02 Let /dev/null have normal permissions.
1217 The 4.73 fixes were a little too stringent and complained about the
1218 permissions on /dev/null. Exempt it from some checks.
1219 Reported by Andreas M. Kirchwitz.
1221 PP/03 Report version information for many libraries, including
1222 Exim version information for dynamically loaded libraries. Created
1223 version.h, now support a version extension string for distributors
1224 who patch heavily. Dynamic module ABI change.
1226 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
1227 privilege escalation vulnerability whereby the Exim run-time user
1228 can cause root to append content of the attacker's choosing to
1231 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
1234 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
1235 If dropping privileges for untrusted macros, we disabled normal logging
1236 on the basis that it would fail; for the Exim run-time user, this is not
1237 the case, and it resulted in successful deliveries going unlogged.
1238 Fixed. Reported by Andreas Metzler.
1244 PP/01 Date: & Message-Id: revert to normally being appended to a message,
1245 only prepend for the Resent-* case. Fixes regression introduced in
1246 Exim 4.70 by NM/22 for Bugzilla 607.
1248 PP/02 Include check_rfc2047_length in configure.default because we're seeing
1249 increasing numbers of administrators be bitten by this.
1251 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
1253 PP/03 Bugzilla 994: added openssl_options main configuration option.
1255 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
1257 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
1259 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
1261 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
1262 without a peer certificate, leading to a segfault because of an
1263 assumption that peers always have certificates. Be a little more
1264 paranoid. Problem reported by Martin Tscholak.
1266 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
1267 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
1268 NB: ClamAV planning to remove STREAM in "middle of 2010".
1269 CL also introduces -bmalware, various -d+acl logging additions and
1270 more caution in buffer sizes.
1272 PP/09 Implemented reverse_ip expansion operator.
1274 PP/10 Bugzilla 937: provide a "debug" ACL control.
1276 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
1278 PP/12 Bugzilla 973: Implement --version.
1280 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
1282 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
1284 PP/15 Bugzilla 816: support multiple condition rules on Routers.
1286 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
1287 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
1288 ignore trailing whitespace.
1290 JJ/02 prevent non-panic DKIM error from being sent to paniclog
1292 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
1295 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
1296 Notification from Dr Andrew Aitchison.
1298 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
1299 ExtendedDetectionInfo response format.
1300 Notification from John Horne.
1302 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
1305 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
1306 XSL and documented dependency on system catalogs, with examples of how
1309 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
1312 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
1313 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
1314 configuration file which is writeable by the Exim user or group.
1316 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
1317 of configuration files to cover files specified with the -C option if
1318 they are going to be used with root privileges, not just the default
1321 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
1322 option (effectively making it always true).
1324 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
1325 files to be used while preserving root privileges.
1327 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
1328 that rogue child processes cannot use them.
1330 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
1331 run-time user, instead of root.
1333 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
1334 Exim run-time user without dropping privileges.
1336 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
1337 result string, instead of calling string_vformat() twice with the same
1340 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
1341 for other users. Others should always drop root privileges if they use
1342 -C on the command line, even for a whitelisted configure file.
1344 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
1346 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
1352 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
1353 $data_path, and $header_path variables; fixed documentation bugs and
1356 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
1357 exipick to access non-standard spools, including the "frozen" queue
1360 NM/01 Bugzilla 965: Support mysql stored procedures.
1361 Patch from Alain Williams
1363 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
1365 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
1366 Patch from Andreas Metzler
1368 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
1369 Patch from Kirill Miazine
1371 NM/05 Bugzilla 671: Added umask to procmail example.
1373 JJ/03 installed exipick 20100323.0, fixing doc bug
1375 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
1376 directory. Notification and patch from Dan Rosenberg.
1378 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
1380 TK/02 Improve log output when DKIM signing operation fails.
1382 MH/01 Treat the transport option dkim_domain as a colon separated
1383 list, not as a single string, and sign the message with each element,
1384 omitting multiple occurences of the same signer.
1386 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
1387 Bugzilla 985, 986. Patch by Simon Arlott
1389 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
1390 Patch by Simon Arlott
1392 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
1393 MBX locking. Notification from Dan Rosenberg.
1399 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
1401 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
1403 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
1405 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
1407 NM/04 Bugzilla 915: Flag broken perl installation during build.
1413 TK/01 Added patch by Johannes Berg that expands the main option
1414 "spamd_address" if it starts with a dollar sign.
1416 TK/02 Write list of recipients to X-Envelope-Sender header when building
1417 the mbox-format spool file for content scanning (suggested by Jakob
1420 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
1421 (http://www.dcc-servers.net/) support via dccifd. Activated by
1422 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
1424 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
1425 by Mark Daniel Reidel <mr@df.eu>.
1427 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
1428 When building exim an external PCRE library is now needed -
1429 PCRE is a system library on the majority of modern systems.
1430 See entry on PCRE_LIBS in EDITME file.
1432 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
1433 conversation. Added nologin parameter to request.
1434 Patch contributed by Kirill Miazine.
1436 TF/01 Do not log submission mode rewrites if they do not change the address.
1438 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
1440 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
1441 log files in place. Contributed by Roberto Lima.
1443 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
1445 TF/03 Bugzilla 615: When checking the local_parts router precondition
1446 after a local_part_suffix or local_part_prefix option, Exim now
1447 does not use the address's named list lookup cache, since this
1448 contains cached lookups for the whole local part.
1450 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
1451 Robert Millan. Documentation is in experimental-spec.txt.
1453 TF/04 Bugzilla 668: Fix parallel build (make -j).
1455 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
1457 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
1458 Patch provided by Jan Srzednicki.
1460 TF/05 Leading white space used to be stripped from $spam_report which
1461 wrecked the formatting. Now it is preserved.
1463 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
1464 that they are available at delivery time.
1466 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
1468 TF/08 TLS error reporting now respects the incoming_interface and
1469 incoming_port log selectors.
1471 TF/09 Produce a more useful error message if an SMTP transport's hosts
1472 setting expands to an empty string.
1474 NM/06 Bugzilla 744: EXPN did not work under TLS.
1475 Patch provided by Phil Pennock.
1477 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
1478 Patch provided by Richard Godbee.
1480 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
1481 acl_smtp_notquit, added index entry.
1483 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
1484 Patch provided by Eugene Bujak.
1486 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
1487 accept(). Patch provided by Maxim Dounin.
1489 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
1490 Patch provided by Phil Pennock.
1492 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
1494 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
1495 Patch provided by Brad "anomie" Jorsch.
1497 NM/14 Bugzilla 622: Added timeout setting to transport filter.
1498 Patch provided by Dean Brooks.
1500 TK/05 Add native DKIM support (does not depend on external libraries).
1502 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
1503 Patch provided by Graeme Fowler.
1505 NM/16 Bugzilla 851: Documentation example syntax fix.
1507 NM/17 Changed NOTICE file to remove references to embedded PCRE.
1509 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
1512 NM/19 Bugzilla 745: TLS version reporting.
1513 Patch provided by Phil Pennock.
1515 NM/20 Bugzilla 167: bool: condition support.
1516 Patch provided by Phil Pennock.
1518 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
1519 clients. Patch provided by Phil Pennock.
1521 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
1522 Patch provided by Brad "anomie" Jorsch.
1524 NM/23 Bugzilla 687: Fix misparses in eximstats.
1525 Patch provided by Heiko Schlittermann.
1527 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
1528 Patch provided by Heiko Schlittermann.
1530 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
1531 plus update to original patch.
1533 NM/26 Bugzilla 799: Documentation correction for ratelimit.
1535 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
1536 Patch provided by David Brownlee.
1538 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
1540 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
1542 NM/30 Bugzilla 888: TLS documentation bugfixes.
1544 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
1546 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
1547 Unlike the original bugzilla I have changed all shell scripts in src tree.
1549 NM/33 Bugzilla 898: Transport filter timeout fix.
1550 Patch by Todd Rinaldo.
1552 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
1553 Patch by Serge Demonchaux.
1555 NM/35 Bugzilla 39: Base64 decode bug fixes.
1556 Patch by Jakob Hirsch.
1558 NM/36 Bugzilla 909: Correct connect() call in dcc code.
1560 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
1562 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
1564 NM/39 Bugzilla 911: Fixed MakeLinks build script.
1570 TK/01 Add preliminary DKIM support. Currently requires a forked version of
1571 ALT-N's libdkim that I have put here:
1572 http://duncanthrax.net/exim-experimental/
1574 Note to Michael Haardt: I had to rename some vars in sieve.c. They
1575 were called 'true' and it seems that C99 defines that as a reserved
1576 keyword to be used with 'bool' variable types. That means you could
1577 not include C99-style headers which use bools without triggering
1578 build errors in sieve.c.
1580 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
1581 as mailq or other aliases. Changed the --help handling significantly
1582 to do whats expected. exim_usage() emits usage/help information.
1584 SC/01 Added the -bylocaldomain option to eximstats.
1586 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
1588 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
1590 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
1596 PH/01 Another patch from the Sieve maintainer.
1598 PH/02 When an IPv6 address is converted to a string for single-key lookup
1599 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
1600 dots are used instead of colons so that keys in lsearch files need not
1601 contain colons. This was done some time before quoting was made available
1602 in lsearch files. However, iplsearch files do require colons in IPv6 keys
1603 (notated using the quote facility) so as to distinguish them from IPv4
1604 keys. This meant that lookups for IP addresses in host lists did not work
1605 for iplsearch lookups.
1607 This has been fixed by arranging for IPv6 addresses to be expressed with
1608 colons if the lookup type is iplsearch. This is not incompatible, because
1609 previously such lookups could never work.
1611 The situation is now rather anomolous, since one *can* have colons in
1612 ordinary lsearch keys. However, making the change in all cases is
1613 incompatible and would probably break a number of configurations.
1615 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
1618 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
1619 conversion specification without a maximum field width, thereby enabling
1620 a rogue spamd server to cause a buffer overflow. While nobody in their
1621 right mind would setup Exim to query an untrusted spamd server, an
1622 attacker that gains access to a server running spamd could potentially
1623 exploit this vulnerability to run arbitrary code as the Exim user.
1625 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
1626 $primary_hostname instead of what libspf2 thinks the hosts name is.
1628 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
1629 a directory entry by the name of the lookup key. Previously, if a
1630 symlink pointed to a non-existing file or a file in a directory that
1631 Exim lacked permissions to read, a lookup for a key matching that
1632 symlink would fail. Now it is enough that a matching directory entry
1633 exists, symlink or not. (Bugzilla 503.)
1635 PH/03 The body_linecount and body_zerocount variables are now exported in the
1638 PH/04 Added the $dnslist_matched variable.
1640 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
1641 This means they are set thereafter only if the connection becomes
1644 PH/06 Added the client_condition to authenticators so that some can be skipped
1645 by clients under certain conditions.
1647 PH/07 The error message for a badly-placed control=no_multiline_responses left
1648 "_responses" off the end of the name.
1650 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
1652 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
1653 (without spaces) instead of just copying the configuration text.
1655 PH/10 Added the /noupdate option to the ratelimit ACL condition.
1657 PH/11 Added $max_received_linelength.
1659 PH/12 Added +ignore_defer and +include_defer to host lists.
1661 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
1662 way in which PCRE > 7.0 is built.
1664 PH/14 Implemented queue_only_load_latch.
1666 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
1667 MAIL command. The effect was to mangle the value on 64-bit systems.
1669 PH/16 Another patch from the Sieve maintainer.
1671 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
1673 PH/18 If a system quota error occurred while trying to create the file for
1674 a maildir delivery, the message "Mailbox is full" was not appended to the
1675 bounce if the delivery eventually timed out. Change 4.67/27 below applied
1676 only to a quota excession during the actual writing of the file.
1678 PH/19 It seems that peer DN values may contain newlines (and other non-printing
1679 characters?) which causes problems in log lines. The DN values are now
1680 passed through string_printing() before being added to log lines.
1682 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
1683 and InterBase are left for another time.)
1685 PH/21 Added message_body_newlines option.
1687 PH/22 Guard against possible overflow in moan_check_errorcopy().
1689 PH/23 POSIX allows open() to be a macro; guard against that.
1691 PH/24 If the recipient of an error message contained an @ in the local part
1692 (suitably quoted, of course), incorrect values were put in $domain and
1693 $local_part during the evaluation of errors_copy.
1699 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
1700 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
1703 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
1704 issue a MAIL command.
1706 PH/02 In an ACL statement such as
1708 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
1710 if a client was not listed at all, or was listed with a value other than
1711 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
1712 the condition was not true (as it should be), so access was not denied.
1713 The bug was that the ! inversion was incorrectly passed on to the second
1714 item. This has been fixed.
1716 PH/03 Added additional dnslists conditions == and =& which are different from
1717 = and & when the dns lookup returns more than one IP address.
1719 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
1720 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
1722 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
1723 FSYNC, which compiles an option called disable_fsync that allows for
1724 bypassing fsync(). The documentation is heavily laced with warnings.
1726 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
1728 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
1729 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
1730 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
1731 including adding "make clean"; (3) Added -fPIC when compiling the test
1732 dynamically loaded module, to get rid of a warning.
1734 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
1735 message fails, move_frozen_messages = true and ignore_bounce_errors_after
1736 = 0s. The bug is otherwise harmless.
1738 PH/07 There was a bug in the dovecot authenticator such that the value of
1739 $auth1 could be overwritten, and so not correctly preserved, after a
1740 successful authentication. This usually meant that the value preserved by
1741 the server_setid option was incorrect.
1743 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
1745 PH/09 Installed PCRE release 7.0.
1747 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
1748 run for batched SMTP input. It is now run at the start of every message
1749 in the batch. While fixing this I discovered that the process information
1750 (output by running exiwhat) was not always getting set for -bs and -bS
1751 input. This is fixed, and it now also says "batched" for BSMTP.
1753 PH/11 Added control=no_pipelining.
1755 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
1756 patch, slightly modified), and move the expansion of helo_data till after
1757 the connection is made in the smtp transport (so it can use these
1760 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
1762 PH/14 Added log_selector = +pid.
1764 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
1766 PH/16 Add ${if forany and ${if forall.
1768 PH/17 Added dsn_from option to vary the From: line in DSNs.
1770 PH/18 Flush SMTP output before performing a callout, unless control =
1771 no_callout_flush is set.
1773 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
1774 was true (the default) a successful delivery failed to delete the retry
1775 item, thus causing premature timeout of the address. The bug is now
1778 PH/20 Added hosts_avoid_pipelining to the smtp transport.
1780 PH/21 Long custom messages for fakedefer and fakereject are now split up
1781 into multiline reponses in the same way that messages for "deny" and
1782 other ACL rejections are.
1784 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
1785 with slight modification.
1787 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
1788 draft, changing the syntax and factoring some duplicate code".
1790 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
1791 for deliveries of the second and subsequent messages over the same SMTP
1794 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
1795 ${reduce, with only minor "tidies".
1797 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
1799 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
1800 expansion side effects.
1802 PH/27 When a message times out after an over-quota error from an Exim-imposed
1803 quota, the bounce message says "mailbox is full". This message was not
1804 being given when it was a system quota that was exceeded. It now should
1807 MH/03 Made $recipients available in local_scan(). local_scan() already has
1808 better access to the recipient list through recipients_list[], but
1809 $recipients can be useful in postmaster-provided expansion strings.
1811 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
1812 in the case of a MAIL command with additional options following the
1813 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
1814 were accidentally chopped off.
1816 PH/29 SMTP synchronization checks are implemented when a command is read -
1817 there is a check that no more input is waiting when there shouldn't be
1818 any. However, for some commands, a delay in an ACL can mean that it is
1819 some time before the response is written. In this time, more input might
1820 arrive, invalidly. So now there are extra checks after an ACL has run for
1821 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
1822 pipelining has not been advertised.
1824 PH/30 MH's patch to allow iscntrl() characters to be list separators.
1826 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
1827 returned in the SMTP response when smtp_return_error_details was false.
1828 This has been fixed.
1830 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
1831 instead of the C I/O that was originally supplied, because problems were
1832 reported on Solaris.
1834 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
1835 Exim which did not show up earlier: it was assuming that a call to
1836 SSL_CTX_set_info_callback() might give an error value. In fact, there is
1837 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
1838 was a macro that became an assignment, so it seemed to work. This has
1839 changed to a proper function call with a void return, hence the compile
1840 error. Exim's code has been fixed.
1842 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
1845 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
1847 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
1849 PH/37 If a message is not accepted after it has had an id assigned (e.g.
1850 because it turns out to be too big or there is a timeout) there is no
1851 "Completed" line in the log. When some messages of this type were
1852 selected by exigrep, they were listed as "not completed". Others were
1853 picked up by some special patterns. I have improved the selection
1854 criteria to be more general.
1856 PH/38 The host_find_failed option in the manualroute router can now be set
1857 to "ignore", to completely ignore a host whose IP address cannot be
1858 found. If all hosts are ignored, the behaviour is controlled by the new
1859 host_all_ignored option.
1861 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
1862 homing or because of multiple MX records with /mx) generated more than
1863 one IP address, and the following item turned out to be the local host,
1864 all the secondary addresses of the first item were incorrectly removed
1865 from the list, along with the local host and any following hosts (which
1866 is what is supposed to happen).
1868 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
1869 whoever called Exim into the -H file. In the case of the daemon it was
1870 behaving confusingly. When first started, it used values for whoever
1871 started the daemon, but after a SIGHUP it used the Exim user (because it
1872 calls itself on a restart). I have changed the code so that it now always
1875 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
1876 message are rejected with the same error (e.g. no authentication or bad
1877 sender address), and a DATA command is nevertheless sent (as can happen
1878 with PIPELINING or a stupid MUA), the error message that was given to the
1879 RCPT commands is included in the rejection of the DATA command. This is
1880 intended to be helpful for MUAs that show only the final error to their
1883 PH/42 Another patch from the Sieve maintainer.
1885 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
1886 Eximstats - Fixed some broken HTML links and added missing column headers
1888 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
1889 columns for Rejects, Temp Rejects, Ham, and Spam rows.
1891 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
1893 PH/43 Yet another patch from the Sieve maintainer.
1895 PH/44 I found a way to check for a TCP/IP connection going away before sending
1896 the response to the final '.' that terminates a message, but only in the
1897 case where the client has not sent further data following the '.'
1898 (unfortunately, this is allowed). However, in many cases there won't be
1899 any further data because there won't be any more messages to send. A call
1900 to select() can be used: if it shows that the input is "ready", there is
1901 either input waiting, or the socket has been closed. An attempt to read
1902 the next input character can distinguish the two cases. Previously, Exim
1903 would have sent an OK response which the client would never have see.
1904 This could lead to message repetition. This fix should cure that, at
1905 least in a lot of common cases.
1907 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
1908 advertised in response to EHLO.
1914 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
1915 fixed by 4.65/MH/01 (is this a record?) are fixed:
1917 (i) An empty string was always treated as zero by the numeric comparison
1918 operators. This behaviour has been restored.
1920 (ii) It is documented that the numeric comparison operators always treat
1921 their arguments as decimal numbers. This was broken in that numbers
1922 starting with 0 were being interpreted as octal.
1924 While fixing these problems I realized that there was another issue that
1925 hadn't been noticed. Values of message_size_limit (both the global option
1926 and the transport option) were treated as octal if they started with 0.
1927 The documentation was vague. These values are now always treated as
1928 decimal, and I will make that clear in the documentation.
1934 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
1935 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
1938 MH/01 Don't check that the operands of numeric comparison operators are
1939 integers when their expansion is in "skipping" mode (fixes bug
1940 introduced by 4.64-PH/07).
1942 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
1943 child addresses, Exim now panics and dies. Previously, because the count
1944 is held in a short int, deliveries were likely to be lost. As such a
1945 large number of recipients for a single message is ridiculous
1946 (performance will be very, very poor), I have chosen to impose a limit
1947 rather than extend the field.
1953 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
1954 leftover -K file (the existence of which was triggered by #402).
1955 While we were at it, introduced process PID as part of the -K
1956 filename. This should rule out race conditions when creating
1959 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
1960 processing considerably. Previous code took too long for large mails,
1961 triggering a timeout which in turn triggers #401.
1963 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
1964 in the DK code in transports.c. sendfile() is not really portable,
1965 hence the _LINUX specificness.
1967 TF/01 In the add_headers option to the mail command in an Exim filter,
1968 there was a bug that Exim would claim a syntax error in any
1969 header after the first one which had an odd number of characters
1972 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
1973 callout verification, Exim cached a "reject" for the entire domain. This
1974 is correct for most verifications, but it is not correct for a recipient
1975 verification with use_sender or use_postmaster set, because in that case
1976 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
1977 case of MAIL FROM:<> rejection from other early rejections (e.g.
1978 rejection of HELO). When verifying a recipient using a non-null MAIL
1979 address, the cache is ignored if it shows MAIL FROM:<> rejection.
1980 Whatever the result of the callout, the value of the domain cache is
1981 left unchanged (for any other kind of callout, getting as far as trying
1982 RCPT means that the domain itself is ok).
1984 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
1987 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
1988 manifest itself as EPIPE rather than ECONNECT. When tidying away a
1989 session, the daemon ignores ECONNECT errors and logs others; it now
1990 ignores EPIPE as well.
1992 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
1993 (quoted-printable decoding).
1995 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
1996 later the small subsequent patch to fix an introduced bug.
1998 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
2000 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
2002 PH/08 An error is now given if message_size_limit is specified negative.
2004 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
2005 to be given (somewhat) arbitrary names.
2007 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
2010 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
2011 miscellaneous code fixes
2013 PH/10 Added the log_reject_target ACL modifier to specify where to log
2016 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
2017 hostname. This is wrong, because it relates to the incoming message (and
2018 probably the interface on which it is arriving) and not to the outgoing
2019 callout (which could be using a different interface). This has been
2020 changed to use the value of the helo_data option from the smtp transport
2021 instead - this is what is used when a message is actually being sent. If
2022 there is no remote transport (possible with a router that sets up host
2023 addresses), $smtp_active_hostname is used.
2025 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
2026 tweaks were necessary in order to get it to work (see also 21 below):
2027 (a) The code assumed that strncpy() returns a negative number on buffer
2028 overflow, which isn't the case. Replaced with Exim's string_format()
2030 (b) There were several signed/unsigned issues. I just did the minimum
2031 hacking in of casts. There is scope for a larger refactoring.
2032 (c) The code used strcasecmp() which is not a standard C function.
2033 Replaced with Exim's strcmpic() function.
2034 (d) The code set only $1; it now sets $auth1 as well.
2035 (e) A simple test gave the error "authentication client didn't specify
2036 service in request". It would seem that Dovecot has changed its
2037 interface. Fortunately there's a specification; I followed it and
2038 changed what the client sends and it appears to be working now.
2040 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
2043 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
2044 address A is aliased to B and C, where B exists and C does not. Without
2045 -v the output is "A verified" because verification stops after a
2046 successful redirection if more than one address is generated. However,
2047 with -v the child addresses are also verified. Exim was outputting "A
2048 failed to verify" and then showing the successful verification for C,
2049 with its parentage. It now outputs "B failed to verify", showing B's
2050 parentage before showing the successful verification of C.
2052 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
2053 look up a TXT record in a specific list after matching in a combined
2056 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
2057 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
2058 they consult the DNS. I had assumed they would set it the way they
2059 wanted; and indeed my experiments on Linux seem to show that in some
2060 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
2061 To be on the safe side, however, I have now made the interface to
2062 host_find_byname() similar to host_find_bydns(), with an argument
2063 containing the DNS resolver options. The host_find_byname() function now
2064 sets these options at its start, just as host_find_bydns() does. The smtp
2065 transport options dns_qualify_single and dns_search_parents are passed to
2066 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
2067 of host_find_byname() use the default settings of RES_DEFNAMES
2068 (qualify_single) but not RES_DNSRCH (search_parents).
2070 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
2071 spool_read_header() do less string testing, by means of a preliminary
2072 switch on the second character of optional "-foo" lines. (This is
2073 overdue, caused by the large number of possibilities that now exist.
2074 Originally there were few.) While I was there, I also converted the
2075 str(n)cmp tests so they don't re-test the leading "-" and the first
2076 character, in the hope this might squeeze out yet more improvement.
2078 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
2079 flag allowing group syntax was set by the header_syntax check but not
2080 turned off, possible causing trouble later; (2) The flag was not being
2081 set at all for the header_verify test, causing "group"-style headers to
2082 be rejected. I have now set it in this case, and also caused header_
2083 verify to ignore an empty address taken from a group. While doing this, I
2084 came across some other cases where the code for allowing group syntax
2085 while scanning a header line wasn't quite right (mostly, not resetting
2086 the flag correctly in the right place). These bugs could have caused
2087 trouble for malformed header lines. I hope it is now all correct.
2089 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
2090 with the "reply" argument non-NULL. The code, however (which originally
2091 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
2092 but it didn't always do it. This confused somebody who was copying the
2093 code for some other use. I have removed all the tests.
2095 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
2096 feature that was used to support insecure browsers during the U.S. crypto
2097 embargo. It requires special client support, and Exim is probably the
2098 only MTA that supported it -- and would never use it because real RSA is
2099 always available. This code has been removed, because it had the bad
2100 effect of slowing Exim down by computing (never used) parameters for the
2101 RSA_EXPORT functionality.
2103 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
2104 authenticator to fail if there's a tab character in the incoming data
2105 (there should never be unless someone is messing about, as it's supposed
2106 to be base64-encoded). Also added, on Timo's advice, the "secured" option
2107 if the connection is using TLS or if the remote IP is the same as the
2108 local IP, and the "valid-client-cert option" if a client certificate has
2111 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
2112 authenticators. This can be used for authorization after authentication
2113 succeeds. (In the case of plaintext, it servers for both authentication
2116 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
2117 if any retry times were supplied.
2119 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
2120 connection, where there is no client IP address to check. In this
2121 situation, the verify now always succeeds.
2123 PH/25 Applied John Jetmore's -Mset patch.
2125 PH/26 Added -bem to be like -Mset, but loading a message from a file.
2127 PH/27 In a string expansion for a processed (not raw) header when multiple
2128 headers of the same name were present, leading whitespace was being
2129 removed from all of them, but trailing whitespace was being removed only
2130 from the last one. Now trailing whitespace is removed from each header
2131 before concatenation. Completely empty headers in a concatenation (as
2132 before) are ignored.
2134 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
2135 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
2137 PH/29 [Removed. This was a change that I later backed out, and forgot to
2138 correct the ChangeLog entry (that I had efficiently created) before
2139 committing the later change.]
2141 PH/30 Exim was sometimes attempting to deliver messages that had suffered
2142 address errors (4xx response to RCPT) over the same connection as other
2143 messages routed to the same hosts. Such deliveries are always "forced",
2144 so retry times are not inspected. This resulted in far too many retries
2145 for the affected addresses. The effect occurred only when there were more
2146 hosts than the hosts_max_try setting in the smtp transport when it had
2147 the 4xx errors. Those hosts that it had tried were not added to the list
2148 of hosts for which the message was waiting, so if all were tried, there
2149 was no problem. Two fixes have been applied:
2151 (i) If there are any address or message errors in an SMTP delivery, none
2152 of the hosts (tried or untried) are now added to the list of hosts
2153 for which the message is waiting, so the message should not be a
2154 candidate for sending over the same connection that was used for a
2155 successful delivery of some other message. This seems entirely
2156 reasonable: after all the message is NOT "waiting for some host".
2157 This is so "obvious" that I'm not sure why it wasn't done
2158 previously. Hope I haven't missed anything, but it can't do any
2159 harm, as the worst effect is to miss an optimization.
2161 (ii) If, despite (i), such a delivery is accidentally attempted, the
2162 routing retry time is respected, so at least it doesn't keep
2163 hammering the server.
2165 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
2166 in ${readsocket because some servers need this prod.
2168 PH/32 Added some extra debug output when updating a wait-xxx database.
2170 PH/33 The hint "could be header name not terminated by colon", which has been
2171 given for certain expansion errors for a long time, was not being given
2172 for the ${if def:h_colon_omitted{... case.
2174 PH/34 The spec says: "With one important exception, whenever a domain list is
2175 being scanned, $domain contains the subject domain." There was at least
2176 one case where this was not true.
2178 PH/35 The error "getsockname() failed: connection reset by peer" was being
2179 written to the panic log as well as the main log, but it isn't really
2180 panic-worthy as it just means the connection died rather early on. I have
2181 removed the panic log writing for the ECONNRESET error when getsockname()
2184 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
2185 runs only) independently of the message's sender address. This meant
2186 that, if the 4xx error was in fact related to the sender, a different
2187 message to the same recipient with a different sender could confuse
2188 things. In particualar, this can happen when sending to a greylisting
2189 server, but other circumstances could also provoke similar problems.
2190 I have changed the default so that the retry time for these errors is now
2191 based a combination of the sender and recipient addresses. This change
2192 can be overridden by setting address_retry_include_sender=false in the
2195 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
2196 remote server are returned as part of bounce messages. This was not
2197 happening for LMTP over a pipe (the lmtp transport), but now it is the
2198 same for both kinds of LMTP.
2200 PH/38 Despite being documented as not happening, Exim was rewriting addresses
2201 in header lines that were in fact CNAMEs. This is no longer the case.
2203 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
2204 and queue runs started by the daemon processed all messages. This has
2205 been fixed so that -R and -S can now usefully be given with -q<time>.
2207 PH/40 Import PCRE release 6.7 (fixes some bugs).
2209 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
2211 PH/42 Give an error if -q is specified more than once.
2213 PH/43 Renamed the variables $interface_address and $interface_port as
2214 $received_ip_address and $received_port, to make it clear that these
2215 values apply to message reception, and not to the outgoing interface when
2216 a message is delivered. (The old names remain recognized, of course.)
2218 PH/44 There was no timeout on the connect() call when using a Unix domain
2219 socket in the ${readsocket expansion. There now is.
2221 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
2222 be meaningful with "accept".
2224 SC/01 Eximstats V1.43
2225 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
2227 SC/02 Eximstats V1.44
2228 Use a glob alias rather than an array ref in the generated
2229 parser. This improves both readability and performance.
2231 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
2232 Collect SpamAssassin and rejection statistics.
2233 Don't display local sender or destination tables unless
2234 there is data to show.
2235 Added average volumes into the top table text output.
2237 SC/04 Eximstats V1.46
2238 Collect data on the number of addresses (recipients)
2239 as well as the number of messages.
2241 SC/05 Eximstats V1.47
2242 Added 'Message too big' to the list of mail rejection
2243 reasons (thanks to Marco Gaiarin).
2245 SC/06 Eximstats V1.48
2246 Mainlog lines which have GMT offsets and are too short to
2247 have a flag are now skipped.
2249 SC/07 Eximstats V1.49 (Alain Williams)
2250 Added the -emptyok flag.
2252 SC/08 Eximstats V1.50
2253 Fixes for obtaining the IP address from reject messages.
2255 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
2256 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
2257 whitesspace changes from 4.64-PH/27
2259 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
2262 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
2263 are found, allow negative numbers in numeric criteria)
2265 JJ/06 exipick.20061117.2, added new $message_body_missing variable
2267 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
2268 to match changes made in 4.64-PH/43
2270 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
2272 PH/47 Put in an explicit test for a DNS lookup of an address record where the
2273 "domain" is actually an IP address, and force a failure. This locks out
2274 those revolvers/nameservers that support "A-for-A" lookups, in
2275 contravention of the specifications.
2277 PH/48 When a host name was looked up from an IP address, and the subsequent
2278 forward lookup of the name timed out, the host name was left in
2279 $sender_host_name, contrary to the specification.
2281 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
2282 restricted to single-key lookups, Exim was not diagnosing an error if
2283 * or *@ was used with a query-style lookup.
2285 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
2287 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
2288 long ago, but noone interested enough thought of it. Let's just say that
2289 the "1.1" means that there are some new functions that weren't there at
2290 some point in the past.
2292 PH/51 Error processing for expansion failure of helo_data from an smtp
2293 transport during callout processing was broken.
2295 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
2296 tested/used via the -bh/-bhc/-bs options.
2298 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
2299 bug, fixed in subsequent PCRE releases).
2301 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
2302 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
2304 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
2310 SC/01 Use a glob alias rather than an array ref in eximstats generated
2311 parser. This improves both readability and performance.
2313 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
2314 Don't display local sender or destination tables in eximstats unless
2315 there is data to show.
2316 Added average volumes into the eximstats top table text output.
2318 SC/03 Collect data on the number of addresses (recipients) as well
2319 as the number of messages in eximstats.
2321 TF/01 Correct an error in the documentation for the redirect router. Exim
2322 does (usually) call initgroups() when daemonizing.
2324 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
2325 with consistent privilege compared to when running as a daemon.
2327 TF/03 Note in the spec that $authenticated_id is not set for local
2328 submissions from trusted users.
2330 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
2331 Thanks to Dean Brooks <dean@iglou.com> for the patch.
2333 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
2334 by adding some example configuration directives to the default
2335 configuration file. A little bit of work is required to uncomment the
2336 directives and define how usernames and passwords are checked, but
2337 there is now a framework to start from.
2339 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
2340 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
2341 without this. I don't know how relevant this is to other LDAP libraries.
2343 PH/02 Add the verb name to the "unknown ACL verb" error.
2345 PH/03 Magnus Holmgren's patch for filter_prepend_home.
2347 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
2349 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
2350 directory not expanded when it should be if an expanded home directory
2351 was set for the address (which is overridden by the transport).
2353 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
2356 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
2357 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
2358 because it is too late at that time, and has no effect.
2360 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
2361 security issue with \' (bugzilla #107). I could not use the
2362 PQescapeStringConn() function, because it needs a PGconn value as one of
2365 PH/08 When testing addresses using -bt, indicate those final addresses that
2366 are duplicates that would not cause an additional delivery. At least one
2367 person was confused, thinking that -bt output corresponded to deliveries.
2368 (Suppressing duplicates isn't a good idea as you lose the information
2369 about possibly different redirections that led to the duplicates.)
2371 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
2372 systems where poll() doesn't work, in particular OS X.
2374 PH/10 Added more information to debugging output for retry time not reached.
2376 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
2377 operations in malware.c.
2379 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
2382 PH/13 If write_rejectlog was set false when logging was sent to syslog with
2383 syslog_duplication set false, log lines that would normally be written
2384 both the the main log and to the reject log were not written to syslog at
2387 PH/14 In the default configuration, change the use of "message" in ACL warn
2388 statements to "add_header".
2390 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
2391 not followed by a command (e.g. "seen endif").
2393 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
2394 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
2397 PH/17 Added extra conditions to the default value of delay_warning_condition
2401 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
2402 { match{$h_precedence:}{(?i)bulk|list|junk} } \
2403 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
2406 The Auto-Submitted: and various List- headers are standardised, whereas I
2407 don't think Precedence: ever was.
2409 PH/18 Refactored debugging code in route_finduser() to show more information,
2410 in particular, the error code if getpwnam() issues one.
2412 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
2413 This is apparently needed in addition to the PH/07 change above to avoid
2414 any possible encoding problems.
2416 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
2417 but not after initializing Perl.
2419 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
2420 output them only if debugging. By default they are written stderr,
2421 apparently, which is not desirable.
2423 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
2426 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
2429 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
2431 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
2432 authenticated or an ident call has been made. Suppress the default
2433 values for $authenticated_id and $authenticated_sender (but permit -oMai
2434 and -oMas) when testing with -bh.
2436 PH/24 Re-jigged the order of the tests in the default configuration so that the
2437 tests for valid domains and recipients precede the DNS black list and CSA
2438 tests, on the grounds that those ones are more expensive.
2440 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
2441 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
2442 This bug exists in every version of Exim that I still have, right back to
2445 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
2446 However, an attempt to turn on case-sensitivity in a regex key by
2447 including (?-i) didn't work because the subject string was already
2448 lowercased, and the effects were non-intuitive. It turns out that a
2449 one-line patch can be used to allow (?-i) to work as expected.
2455 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
2456 other effects) broke the use of negated acl sub-conditions.
2458 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
2461 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
2462 "Deny" causes Exim to reject the incoming connection with a 554 error.
2463 Unfortunately, if there is a major crisis, such as a disk failure,
2464 tcp-wrappers gives "deny", whereas what one would like would be some
2465 kind of temporary error. A kludge has been added to help with this.
2466 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
2467 554 error is used if errno is still zero or contains ENOENT (which occurs
2468 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
2471 PH/03 Add -lutil to the default FreeBSD LIBS setting.
2473 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
2474 errors. Otherwise a message that provokes a temporary error (when other
2475 messages do not) can cause a whole host to time out.
2477 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
2478 addresses were routed directly to files or pipes from a redirect router.
2479 File deliveries just didn't batch; pipe deliveries might have suffered
2482 PH/06 A failure to get a lock for a hints database would erroneously always say
2483 "Failed to get write lock", even when it was really a read lock.
2485 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
2486 of 0600. This has been changed to use the value of the lockfile_mode
2487 option (which defaults to 0600).
2489 PH/08 Applied small patch from the Sieve maintainer.
2491 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
2492 folder from quota calculations, a direct delivery into this folder messed
2493 up the contents of the maildirsize file. This was because the regex was
2494 used only to exclude .Trash (or whatever) when the size of the mailbox
2495 was calculated. There was no check that a delivery was happening into an
2496 excluded directory. This bug has been fixed by ignoring all quota
2497 processing for deliveries into excluded directories.
2499 PH/10 Added the maildirfolder_create_regex option to appendfile.
2505 PH/01 The code for finding all the local interface addresses on a FreeBSD
2506 system running IPv6 was broken. This may well have applied to all BSD
2507 systems, as well as to others that have similar system calls. The broken
2508 code found IPv4 interfaces correctly, but gave incorrect values for the
2509 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
2510 that it would not match correctly against @[] and not recognize the IPv6
2513 PH/02 The ipliteral router was not recognizing addresses of the form user@
2514 [ipv6:....] because it didn't know about the "ipv6:" prefix.
2516 PH/03 Added disable_ipv6.
2518 PH/04 Changed $reply_address to use the raw form of the headers instead of the
2519 decoded form, because it is most often used to construct To: headers
2520 lines in autoreplies, and the decoded form may well be syntactically
2521 invalid. However, $reply_address has leading white space removed, and all
2522 newlines turned into spaces so that the autoreply transport does not
2525 PH/05 If group was specified without a user on a router, and no group or user
2526 was specified on a transport, the group from the router was ignored.
2528 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
2529 for visible compile-time settings that can be used to change these
2530 numbers, for those that want even more. Backwards compatibility with old
2531 spool files has been maintained. However, going back to a previous Exim
2532 release will lost any variables that are in spool files.
2534 PH/07 Two small changes when running in the test harness: increase delay when
2535 passing a TCP/IP connection to a new process, in case the original
2536 process has to generate a bounce, and remove special handling of
2537 127.0.0.2 (sic), which is no longer necessary.
2539 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
2540 be the same on different OS.
2542 PH/09 Moved a debug statement in filter processing to avoid a race problem when
2545 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
2546 whether --show-vars was specified or not
2548 JJ/02 exipick: Added support for new ACL variable spool format introduced
2551 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
2552 syntactically invalid From: or Reply-to: line, and a filter used this to
2553 generate an autoreply, and therefore failed to obtain an address for the
2554 autoreply, Exim could try to deliver to a non-existent relative file
2555 name, causing unrelated and misleading errors. What now happens is that
2556 it logs this as a hard delivery error, but does not attempt to create a
2559 PH/11 The exinext utility has a -C option for testing purposes, but although
2560 the given file was scanned by exinext itself; it wasn't being passed on
2561 when Exim was called.
2563 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
2564 an end-of-file indication when reading a command response.
2566 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
2567 compiled. In many other places in Exim, IPv6 addresses are always
2568 recognized, so I have changed this. It also means that IPv4 domain
2569 literals of the form [IPV4:n.n.n.n] are now always recognized.
2571 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
2572 used if the router is not running as root, for example, when verifying at
2573 ACL time, or when using -bh. The debugging output from this situation was
2574 non-existent - all you got was a failure to exec. I have made two
2577 (a) Failures to set uid/gid, the current directory, or a process leader
2578 in a subprocess such as that created by queryprogram now generate
2579 suitable debugging ouput when -d is set.
2581 (b) The queryprogram router detects when it is not running as root,
2582 outputs suitable debugging information if -d is set, and then runs
2583 the subprocess without attempting to change uid/gid.
2585 PH/15 Minor change to Makefile for building test_host (undocumented testing
2588 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
2589 additional section of a DNS packet that returns MX or SRV records.
2590 Instead, it always explicitly searches for A/AAAA records. This avoids
2591 major problems that occur when a DNS server includes only records of one
2592 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
2593 fixed another bug: if SRV records were looked up and the corresponding
2594 address records were *not* found in the additional section, the port
2595 values from the SRV records were lost.
2597 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
2598 using the correct key (the original address) when searching the retry
2599 rules in order to find which one to use for generating the retry hint.
2601 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
2602 adding the default one. Similarly, if it contains a Reply-To: header, the
2603 errors_reply_to option, if set, is not used.
2605 PH/19 When calculating a retry time, Exim used to measure the "time since
2606 failure" by looking at the "first failed" field in the retry record. Now
2607 it does not use this if it is later than than the arrival time of the
2608 message. Instead it uses the arrival time. This makes for better
2609 behaviour in cases where some deliveries succeed, thus re-setting the
2610 "first failed" field. An example is a quota failure for a huge message
2611 when small messages continue to be delivered. Without this change, the
2612 "time since failure" will always be short, possible causing more frequent
2613 delivery attempts for the huge message than are intended.
2614 [Note: This change was subsequently modified - see PH/04 for 4.62.]
2616 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
2617 $1, $2, $3) because the numerical variables can be reset during some
2618 expansion items (e.g. "match"), thereby losing the authentication data.
2620 PH/21 Make -bV show the size of off_t variables so that the test suite can
2621 decide whether to run tests for quotas > 2G.
2623 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
2624 mailbox_size, and mailbox_filecount in the appendfile transport. If a
2625 filecount value is greater than 2G or if a quota value is greater than 2G
2626 on a system where the size of off_t is not greater than 4, a panic error
2629 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
2630 never match. The debug and -bh output now contains an explicit error
2631 message indicating a malformed IPv4 address or mask.
2633 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
2634 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
2635 PH/23 above applies.
2637 PH/25 Do not write to syslog when running in the test harness. The only
2638 occasion when this arises is a failure to open the main or panic logs
2639 (for which there is an explicit test).
2641 PH/26 Added the /no_tell option to "control=freeze".
2643 PH/27 If a host name lookup failed very early in a connection, for example, if
2644 the IP address matched host_lookup and the reverse lookup yielded a name
2645 that did not have a forward lookup, an error message of the form "no IP
2646 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
2647 could be logged. Now it outputs the IP address instead of "NULL".
2649 PH/28 An enabling patch from MH: add new function child_open_exim2() which
2650 allows the sender and the authenticated sender to be set when
2651 submitting a message from within Exim. Since child_open_exim() is
2652 documented for local_scan(), the new function should be too.
2654 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
2655 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
2656 results in an empty string is now treated as unset.
2658 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
2660 PH/31 Added sender_verify_fail logging option.
2662 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
2663 needed by RFC 822 but not by RFC 2822 was commented out. I have now
2664 tidied the source and removed it altogether.
2666 PH/33 When a queue run was abandoned because the load average was too high, a
2667 log line was always written; now it is written only if the queue_run log
2668 selector is set. In addition, the log line for abandonment now contains
2669 information about the queue run such as the pid. This is always present
2670 in "start" and "stop" lines but was omitted from the "abandon" line.
2672 PH/34 Omit spaces between a header name and the colon in the error message that
2673 is given when verify = headers_syntax fails (if there are lots of them,
2674 the message gets confusing).
2676 PH/35 Change the default for dns_check_names_pattern to allow slashes within
2677 names, as there are now some PTR records that contain slashes. This check
2678 is only to protect against broken name servers that fall over on strange
2679 characters, so the fact that it applies to all lookups doesn't matter.
2681 PH/36 Now that the new test suite is complete, we can remove some of the
2682 special code in Exim that was needed for the old test suite. For example,
2683 sorting DNS records because real resolvers return them in an arbitrary
2684 order. The new test suite's fake resolver always returns records in the
2687 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
2688 bounces) except when queue_only is set, to avoid logging races between
2689 the different processes.
2691 PH/38 Panic-die if .include specifies a non-absolute path.
2693 PH/39 A tweak to the "H" retry rule from its user.
2695 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
2696 a label. They prevented compilation on older perls.
2698 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
2699 a warning to be raised on newish perls.
2701 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
2702 on queue. Changes to match documented behaviour of showing count of
2703 messages matching specified criteria.
2705 PH/40 Changed the default ident timeout from 30s to 5s.
2707 PH/41 Added support for the use of login_cap features, on those BSD systems
2708 that have them, for controlling the resources used by pipe deliveries.
2710 PH/42 The content-scanning code uses fopen() to create files in which to put
2711 message data. Previously it was not paying any attention to the mode of
2712 the files. Exim runs with umask(0) because the rest of the code creates
2713 files with open(), and sets the required mode explicitly. Thus, these
2714 files were ending up world-writeable. This was not a big issue, because,
2715 being within the spool directory, they were not world-accessible. I have
2716 created a function called modefopen, which takes an additional mode
2717 argument. It sets umask(777), creates the file, chmods it to the required
2718 mode, then resets the umask. All the relevant calls to fopen() in the
2719 content scanning code have been changed to use this function.
2721 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
2722 to 24 hours. This avoids potential overflow problems when processing G
2723 and H retry rules. I suspect nobody ever tinkers with this value.
2725 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
2727 PH/45 When the plaintext authenticator is running as a client, the server's
2728 challenges are checked to ensure they are valid base64 strings. By
2729 default, the authentication attempt is cancelled if an invalid string is
2730 received. Setting client_ignore_invalid_base64 true ignores these errors.
2731 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
2732 they are received. Thus, the responses can be made to depend on the
2733 challenges. If an invalid string is ignored, an empty string is placed in
2736 PH/46 Messages that are created by the autoreply transport now contains a
2737 References: header, in accordance with RFCs 2822 and 3834.
2739 PH/47 Added authenticated_sender_force to the smtp transport.
2741 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
2743 PH/49 Installed latest patch from the Sieve maintainer.
2745 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
2746 was also set, the appendfile transport was unnecessarily scanning a
2747 directory of message files (e.g. for maildir delivery) to find the count
2748 of files (along with the size), even though it did not need this
2749 information. It now does the scan only if it needs to find either the
2750 size of the count of files.
2752 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
2754 PH/52 Two bugs concerned with error handling when the smtp transport is
2757 (i) Exim was not creating retry information for temporary errors given
2758 for individual recipients after the DATA command when the smtp transport
2759 was used in LMTP mode. This meant that they could be retried too
2760 frequently, and not timed out correctly.
2762 (ii) Exim was setting the flag that allows error details to be returned
2763 for LMTP errors on RCPT commands, but not for LMTP errors for individual
2764 recipients that were returned after the DATA command.
2766 PH/53 This is related to PH/52, but is more general: for any failing address,
2767 when detailed error information was permitted to be returned to the
2768 sender, but the error was temporary, then after the final timeout, only
2769 "retry timeout exceeded" was returned. Now it returns the full error as
2770 well as "retry timeout exceeded".
2772 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
2773 do this, and (what is worse) MTAs that accept it.
2775 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
2776 will now be deprecated.
2778 PH/56 New os.c-cygwin from the Cygwin maintainer.
2780 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
2781 formats (previously only available in exim formats via -bpr, -bpru,
2782 and -bpra. Now also available in native and exiqgrep formats)
2784 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
2785 with very large, slow to parse queues
2787 JJ/08 exipick: added ! as generic prefix to negate any criteria format
2789 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
2791 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
2792 responses to authentication challenges, though it was showing the
2793 challenges; (ii) I've removed the CR characters from the debug output for
2796 PH/58 Allow for the insertion of a newline as well as a space when a string
2797 is turned into more than one encoded-word during RFC 2047 encoding. The
2798 Sieve code now uses this.
2800 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
2801 data_4xx, lost_connection, tls_required.
2803 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
2804 message was being sent as an SMTP response.
2806 PH/61 Add -l and -k options to exicyclog.
2808 PH/62 When verifying, if an address was redirected to one new address, so that
2809 verification continued, and the new address failed or deferred after
2810 having set something in $address_data, the value of $address_data was not
2811 passed back to the ACL. This was different to the case when no
2812 redirection occurred. The value is now passed back in both cases.
2814 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
2815 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
2816 use login_cap.h, so on its own it isn't the distinguishing feature. The
2817 new name refers directly to the setclassresources() function.
2819 PH/65 Added configuration files for NetBSD3.
2821 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
2823 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
2824 is preferred over IPv4.
2826 PH/68 The bounce_return_message and bounce_return_body options were not being
2827 honoured for bounces generated during the reception of non-SMTP messages.
2828 In particular, this applied to messages rejected by the ACL. This bug has
2829 been fixed. However, if bounce_return_message is true and bounce_return_
2830 body is false, the headers that are returned for a non-SMTP message
2831 include only those that have been read before the error was detected.
2832 (In the case of an ACL rejection, they have all been read.)
2834 PH/69 The HTML version of the specification is now built in a directory called
2835 spec_html instead of spec.html, because the latter looks like a path with
2836 a MIME-type, and this confuses some software.
2838 PH/70 Catch two compiler warnings in sieve.c.
2840 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
2841 function verify_get_ident() calls ip_connect() to connect a socket, but
2842 if the "connect()" function timed out, ip_connect() used to close the
2843 socket. However, verify_get_ident() also closes the socket later, and in
2844 between Exim writes to the log, which may get opened at this point. When
2845 the socket was closed in ip_connect(), the log could get the same file
2846 descriptor number as the socket. This naturally causes chaos. The fix is
2847 not to close the socket in ip_connect(); the socket should be closed by
2848 the function that creates it. There was only one place in the code where
2849 this was missing, in the iplookup router, which I don't think anybody now
2850 uses, but I've fixed it anyway.
2852 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
2853 well as to direct DNS lookups. Otherwise the handling of names in host
2854 lists is inconsistent and therefore confusing.
2860 PH/01 Two changes to the default runtime configuration:
2862 (1) Move the checks for relay_from_hosts and authenticated clients from
2863 after to before the (commented out) DNS black list checks.
2865 (2) Add control=submission to the relay_from_hosts and authenticated
2866 clients checks, on the grounds that messages accepted by these
2867 statements are most likely to be submissions.
2869 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
2871 (1) Generate an error if the third argument for the ${prvs expansion is
2874 (2) Treat a missing third argument of ${prvscheck as if it were an empty
2877 (3) Reset the variables that are obtained from the first argument of
2878 ${prvscheck and used in the second argument before leaving the code,
2879 because their memory is reclaimed, so using them afterwards may do
2882 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
2883 one (it's much easier than Tom thought :-).
2885 (5) Because of (4), we can now allow for the use of $prvscheck_result
2886 inside the third argument.
2888 PH/03 For some reason, the default setting of PATH when running a command from
2889 a pipe transport was just "/usr/bin". I have changed it to
2892 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
2893 anything to be listed in the output from -bV.
2895 PH/05 When a filter generated an autoreply, the entire To: header line was
2896 quoted in the delivery log line, like this:
2898 => >A.N.Other <ano@some.domain> <original@ddress> ...
2900 This has been changed so that it extracts the operative address. There
2901 may be more than one such address. If so, they are comma-separated, like
2904 => >ano@some.domain,ona@other.domain <original@ddress> ...
2906 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
2907 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
2908 not being looked up in the rDNS to get a host name, Exim was showing the
2909 IP address twice in Received: lines, even though the IP addresses were
2910 identical. For example:
2912 Received: from [1.2.3.4] (helo=[1.2.3.4])
2914 However, if the real host name was known, it was omitting the HELO data
2915 if it matched the actual IP address. This has been tidied up so that it
2916 doesn't show the same IP address twice.
2918 PH/07 When both +timestamp and +memory debugging was on, the value given by
2919 $tod_xxx expansions could be wrong, because the tod_stamp() function was
2920 called by the debug printing, thereby overwriting the timestamp buffer.
2921 Debugging no longer uses the tod_stamp() function when +timestamp is set.
2923 PH/08 When the original message was included in an autoreply transport, it
2924 always said "this is a copy of the message, including all the headers",
2925 even if body_only or headers_only was set. It now gives an appropriate
2928 PH/09 Applied a patch from the Sieve maintainer which:
2930 o fixes some comments
2931 o adds the (disabled) notify extension core
2932 o adds some debug output for the result of if/elsif tests
2933 o points to the current vacation draft in the documentation
2934 and documents the missing references header update
2938 o fixes a bug in processing the envelope test (when testing
2939 multiple envelope elements, the last element determinted the
2942 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
2943 Electronic Mail") by including:
2945 Auto-submitted: auto-generated
2947 in the messages that it generates (bounce messages and others, such as
2948 warnings). In the case of bounce messages for non-SMTP mesages, there was
2949 also a typo: it was using "Auto_submitted" (underscore instead of
2950 hyphen). Since every message generated by Exim is necessarily in response
2951 to another message, thes have all been changed to:
2953 Auto-Submitted: auto-replied
2955 in accordance with these statements in the RFC:
2957 The auto-replied keyword:
2959 - SHOULD be used on messages sent in direct response to another
2960 message by an automatic process,
2962 - MUST NOT be used on manually-generated messages,
2964 - MAY be used on Delivery Status Notifications (DSNs) and Message
2965 Disposition Notifications (MDNs),
2967 - MUST NOT be used on messages generated by automatic or periodic
2968 processes, except for messages which are automatic responses to
2971 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
2972 to the default Received: header definition.
2974 PH/12 Added log selector acl_warn_skipped (default on).
2976 PH/13 After a successful wildlsearch lookup, discard the values of numeric
2977 variables because (a) they are in the wrong storage pool and (b) even if
2978 they were copied, it wouldn't work properly because of the caching.
2980 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
2981 checking when decoding. Apparently there are clients that generate
2982 overlong encoded strings. Why am I not surprised?
2984 PH/15 If the first argument of "${if match_address" was not empty, but did not
2985 contain an "@" character, Exim crashed. Now it writes a panic log message
2986 and treats the condition as false.
2988 PH/16 In autoreply, treat an empty string for "once" the same as unset.
2990 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
2991 extension "envelope-auth". The code is finished and in agreement with
2992 other implementations, but there is no documentation so far and in fact,
2993 nobody wrote the draft yet. This extension is currently #undef'ed, thus
2994 not changing the active code.
2996 Print executed "if" and "elsif" statements when debugging is used. This
2997 helps a great deal to understand what a filter does.
2999 Document more things not specified clearly in RFC3028. I had all this
3000 sorted out, when out of a sudden new issues came to my mind. Oops."
3002 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
3005 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
3006 canonical form (as documented). However, after a host name lookup from
3007 the IP address, check_host() was doing a simple string comparison with
3008 addresses acquired from the DNS when checking that the found name did
3009 have the original IP as one of its addresses. Since any found IPv6
3010 addresses are likely to be in abbreviated form, the comparison could
3011 fail. Luckily, there already exists a function for doing the comparison
3012 by converting both addresses to binary, so now that is used instead of
3013 the text comparison.
3015 PH/20 There was another similar case to PH/19, when a complete host name was
3016 given in a host list; looking up its IP address could give an abbreviated
3017 form, whereas the current host's name might or might not be abbreviated.
3018 The same fix has been applied.
3024 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
3025 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
3028 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
3029 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
3031 PH/03 Typo: missing ".o" in src/pcre/Makefile.
3033 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
3034 header line, restrict the check to what is listed in RFCs 2369 and 2929.
3035 Also, for "Auto-Submitted", treat anything other than "no" as
3036 non-personal, in accordance with RFC 3834. (Previously it treated
3037 anything starting "auto-" as non-personal.)
3039 TF/01 The control=submission/name=... option had a problem with syntax
3040 errors if the name included a slash character. The /name= option
3041 now slurps the rest of the string, so it can include any characters
3042 but it must come last in the list of options (after /sender_retain
3045 PH/05 Some modifications to the interface to the fake nameserver for the new
3053 TK/01 Added the "success_on_redirect" address verification option. See
3054 NewStuff for rationale and an example.
3056 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
3058 PH/02 Patch to exigrep to allow it to work on syslog lines.
3060 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
3061 fread() to skip over the body file's header line, because in Cygwin the
3062 header line is locked and is inaccessible.
3064 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
3065 co-exist for some time) to make it clear that it is the Exim ID that is
3066 referenced, not the Message-ID: header line.
3068 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
3069 string_format() function, because snprintf() does not exist on all
3072 PH/06 The use of forbid_filter_existstest now also locks out the use of the
3073 ${stat: expansion item.
3075 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
3076 protocol synchronization error", to keep the pedants happy.
3078 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
3079 well as for IRIX systems, when gcc is being used. See the host.c source
3082 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
3084 PH/10 Named domain lists were not working if used in a queue_smtp_domains
3087 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
3088 transport and to the smtp transport in LMTP mode.
3090 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
3092 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
3093 run a filter in a subprocess. This could lead to confusion in subsequent
3094 lookups in the parent process. There should also be a search_tidyup() at
3095 the end of the subprocess.
3097 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
3098 only if the host matched helo_try_verify_hosts, which caused the
3099 verification to occur when the EHLO/HELO command was issued. The ACL just
3100 tested the remembered result. Now, if a previous verification attempt has
3101 not happened, "verify = helo" does it there and then.
3103 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
3105 TK/03 Fix log output including CR from clamd.
3107 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
3108 exist provoked a memory error which could cause a segfault.
3110 PH/15 Installed PCRE 6.2
3112 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
3114 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
3115 of the problem. Specifically, suggested +O2 rather than +O1 for the
3118 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
3120 PH/20 If a delivery was routed to a non-standard port by means of an SRV
3121 record, the port was not correctly logged when the outgoing_port log
3122 selector was set (it logged the transort's default port).
3124 PH/21 Added support for host-specific ports to manualroute, queryprogram,
3125 fallback_hosts, and "hosts" in the smtp transport.
3127 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
3128 host errors such as "Connection refused".
3130 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
3131 authentication with radiusclient 0.4.9:
3133 - Error returned from rc_read_config was caught wrongly
3134 - Username/password not passed on to radius server due to wrong length.
3136 The presumption is that some radiusclient API changes for 4.51/PH/17
3137 were not taken care of correctly. The code is still untested by me (my
3138 Linux distribution still has 0.3.2 of radiusclient), but it was
3139 contributed by a Radius user.
3141 PH/24 When doing a callout, the value of $domain wasn't set correctly when
3142 expanding the "port" option of the smtp transport.
3144 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
3145 while reading a MIME header. Thanks to Tom Hughes for a patch.
3147 PH/24 Include config.h inside local_scan.h so that configuration settings are
3150 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
3151 that in an ACL for RCPT (for example), you can examine exactly what was
3154 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
3155 commands, but it was not correctly comparing the address with the actual
3156 client host address. Thus, it would show the EHLO address in Received:
3157 header lines when this was not necessary.
3159 PH/27 Added the % operator to ${eval:}.
3161 PH/28 Exim tries to create and chdir to its spool directory when it starts;
3162 it should be ignoring failures (because with -C, for example, it has lost
3163 privilege). It wasn't ignoring creation failures other than "already
3166 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
3169 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
3170 because an input line was too long, either on its own, or by virtue of
3171 too many continuations, the temporary file was not being removed, and the
3172 return code was incorrect.
3174 PH/31 Missing "BOOL" in function definition in filtertest.c.
3176 PH/32 Applied Sieve patches from the maintainer.
3178 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
3180 PH/33 Added "verify = not_blind".
3182 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
3183 Local/Makefile (with some defaults set). These are used in built scripts
3184 such as exicyclog, but they have never been used in the exim_install
3185 script (though there are many overriding facilities there). I have
3186 arranged that the exim_install script now takes note of these two
3189 PH/35 Installed configuration files for Dragonfly.
3191 PH/36 When a locally submitted message by a trusted user did not contain a
3192 From: header, and the sender address was obtained from -f or from an SMTP
3193 MAIL command, and the trusted user did not use -F to supply a sender
3194 name, $originator_name was incorrectly used when constructing a From:
3195 header. Furthermore, $originator_name was used for submission mode
3196 messages from external hosts without From: headers in a similar way,
3197 which is clearly wrong.
3199 PH/37 Added control=suppress_local_fixups.
3201 PH/38 When log_selector = +received_sender was set, and the addition of the
3202 sender made the log line's construction buffer exactly full, or one byte
3203 less than full, an overflow happened when the terminating "\n" was
3206 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
3207 when the result of a list match is failure because a DNS lookup failed.
3209 PH/40 RM_COMMAND is now used in the building process.
3211 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
3212 the "build-* directories that it finds.
3214 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
3215 address in a domain literal was a prefix of an interface address.
3217 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
3218 when verifying a sender address, unless rewrite_headers is false.
3220 PH/44 Wrote a long comment about why errors_to addresses are verified as
3221 recipients, not senders.
3223 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
3224 the ratelimit ACL was added.
3226 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
3228 PH/46 Added extra information about PostgreSQL errors to the error string.
3230 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
3231 avoiding the need to install special zones in a real server. This is
3232 backwards compatible; if it can't find the fake resolver, it drops back.
3233 Thus, both old and new test suites can be run.
3235 TF/02 Added util/ratelimit.pl
3237 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
3238 clock is set back in time.
3240 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
3241 Candler <B.Candler@pobox.com>.
3243 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
3244 OK for addresses that are the result of redirections.
3246 PH/48 A number of further additions for the benefit of the new test suite,
3247 including a fake gethostbyname() that interfaces to the fake DNS resolver
3250 TF/06 The fix for widen_domains has also been applied to qualify_single and
3251 search_parents which are the other dnslookup options that can cause
3254 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
3257 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
3259 TF/07 Exim produced the error message "an SRV record indicated no SMTP
3260 service" if it encountered an MX record with an empty target hostname.
3261 The message is now "an MX or SRV record indicated no SMTP service".
3263 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
3264 if the DNS of the sending site is misconfigured. This is quite a
3265 common situation. This change restores the behaviour of treating a
3266 helo verification defer as a failure.
3268 PH/51 If self=fail was set on a router, the bounce message did not include the
3269 actual error message.
3275 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
3277 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
3278 command itself ended in error, the underlying message about the transport
3279 filter timeout was being overwritten with the pipe command error. Now the
3280 underlying error message should be appended to the second error message.
3282 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
3284 PH/02 Reduce the amount of output that "make" produces by default. Full output
3285 can still be requested.
3287 PH/03 The warning log line about a condition test deferring for a "warn" verb
3288 was being output only once per connection, rather than after each
3289 occurrence (because it was using the same function as for successful
3290 "warn" verbs). This seems wrong, so I have changed it.
3292 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
3293 it should not have, which might have caused a crash in the right
3294 circumstances, but probably never did.
3296 PH/04 Installed a modified version of Tony Finch's patch to make submission
3297 mode fix the return path as well as the Sender: header line, and to
3298 add a /name= option so that you can make the user's friendly name appear
3301 TF/03 Added the control = fakedefer ACL modifier.
3303 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
3304 Mark Lowes for thorough testing.
3306 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
3308 TK/03 Merged latest SRS patch from Miles Wilton.
3310 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
3311 with the definition in sysexits.h (which is #included earlier).
3312 Fortunately, Exim does not actually use EX_OK. The code used to try to
3313 preserve the sysexits.h value, by assumimg that macro definitions were
3314 scanned for macro replacements. I have been disabused of this notion,
3315 so now the code just undefines EX_OK before #including unistd.h.
3317 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
3318 in the smtp transport. When a block could not be written in a single
3319 write() function, the timeout was being re-applied to each part-write.
3320 This seems wrong - if the receiver was accepting one byte at a time it
3321 would take for ever. The timeout is now adjusted when this happens. It
3322 doesn't have to be particularly precise.
3324 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
3325 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
3327 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
3328 without a domain if the check to <postmaster@domain> fails.
3330 SC/01 Eximstats: added -xls and the ability to specify output files
3331 (patch written by Frank Heydlauf).
3333 SC/02 Eximstats: use FileHandles for outputing results.
3335 SC/03 Eximstats: allow any combination of xls, txt, and html output.
3337 SC/04 Eximstats: fixed display of large numbers with -nvr option
3339 SC/05 Eximstats: fixed merging of reports with empty tables.
3341 SC/06 Eximstats: added the -include_original_destination flag
3343 SC/07 Eximstats: removed tabs and trailing whitespace.
3345 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
3347 TK/06 MBOX spool code: Add real "From " MBOX separator line
3348 so the .eml file is really in mbox format (even though
3349 most programs do not really care). Patch from Alex Miller.
3351 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
3352 The latter is generated from $received_to and is only set if the
3353 message has one envelope recipient. SA can use these headers,
3354 obviously out-of-the-box. Patch from Alex Miller.
3356 PH/08 The ${def test on a variable was returning false if the variable's
3357 value was "0", contrary to what the specification has always said!
3358 The result should be true unless the variable is empty.
3360 PH/09 The syntax error of a character other than { following "${if
3361 def:variable_name" (after optional whitespace) was not being diagnosed.
3362 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
3363 accidental colon was present, for example, could give incorrect results.
3365 PH/10 Tidied the code in a number of places where the st_size field of a stat()
3366 result is used (not including appendfile, where other changes are about
3369 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
3370 This involved changing a lot of size variables from int to off_t. It
3371 should work with maildirs and everything.
3373 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
3374 spamd dying while we are connected to it.
3376 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
3379 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
3380 function for BDB 4.3.
3382 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
3384 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
3385 the sources for PCRE have been re-arranged and more files are now
3388 PH/15 The code I had for printing potentially long long variables in PH/11
3389 above was not the best (it lost precision). The length of off_t variables
3390 is now inspected at build time, and an appropriate printing format (%ld
3391 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
3392 to be "long long int" or "long int". This is needed for the internal
3393 formatting function string_vformat().
3395 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
3396 the configuration file to be ":syslog", then the script "guesses" where
3397 the logs files are, rather than using the compiled in default. In our
3398 case the guess is not the same as the compiled default, so the script
3399 suddenly stopped working when I started to use syslog. The patch checks
3400 to see if log_file_path is "". If so, it attempts to read it from exim
3401 with no configuration file to get the compiled in version, before it
3402 falls back to the previous guessing code."
3404 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
3405 implementing BATV in an Exim configuration. See NewStuff for the gory
3408 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
3409 Makefile that are specific to HP-UX.
3411 PH/18 If the "use_postmaster" option was set for a recipient callout together
3412 with the "random" option, the postmaster address was used as the MAIL
3413 FROM address for the random test, but not for the subsequent recipient
3414 test. It is now used for both.
3416 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
3417 patch removes a few documentation additions to RFC 3028, because the
3418 latest draft now contains them. It adds the new en;ascii-case comparator
3419 and a new error check for 8bit text in MIME parts. Comparator and
3420 require names are now matched exactly. I enabled the subaddress
3421 extension, but it is not well tested yet (read: it works for me)."
3423 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
3424 rework some of the code of TK/09 above to avoid the hardwired use of
3425 "%lld" and "long long". Replaced the call to snprintf() with a call to
3428 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
3429 records point to non-existent hosts", "retry timeout exceeded", and
3430 "retry time not reached for any host after a long failure period".
3432 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
3433 experimental DomainKeys support:
3435 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
3436 (2) On an error such as an illegally used "control", the wrong name for
3437 the control was given.
3439 These problems did NOT occur unless DomainKeys support was compiled.
3441 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
3443 PH/24 Added ${if match_ip condition.
3445 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
3446 databases so that it will be absolutely obvious if a crash occurs in the
3447 DB library. This is a regular occurrence (often caused by mis-matched
3450 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
3451 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
3452 picked up on a user's system that detects such things. There doesn't seem
3453 to be a gcc warning option for this - only an attribute that has to be
3454 put on the function's prototype. It seems that in Fedora Core 4 they have
3455 set this on a number of new functions. No doubt there will be more in due
3458 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
3459 specify a transport. However, if an address that was verified by such a
3460 router was the subject of a callout, Exim crashed because it tried to
3461 read the rcpt_include_affixes from the non-existent transport. Now it
3462 just assumes that the setting of that option is false. This bug was
3463 introduced by 4.51/PH/31.
3465 PH/28 Changed -d+all to exclude +memory, because that information is very
3466 rarely of interest, but it makes the output a lot bigger. People tend to
3467 do -d+all out of habit.
3469 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
3470 code in os-type was giving problems when libc.so lives in lib64, like on
3473 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
3474 aren't the modern standard, and it seems that some systems' include files
3475 don't always have them. Exim was already checking for some of the newer
3476 ones like T_AAAA, and defining it itself. I've added checks for all the
3477 record types that Exim uses.
3479 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
3480 not automatically generating a new one, as it is supposed to. This
3481 prevented TLS from working. If the file did exist, but contained invalid
3482 data, a new version was generated, as expected. It was only the case of a
3483 non-existent file that was broken.
3485 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
3486 with a change in libdomainkeys > 0.64.
3488 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
3489 from DNS. If the selector record carries the flag, it now has
3490 precedence over the domain-wide flag.
3492 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
3494 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
3495 the use of a port name that isn't defined in /etc/services) occurred, the
3496 message was deferred as in a normal delivery, and thus remained on the
3497 spool, instead of being failed because of the mua_wrapper setting. This
3498 is now fixed, and I tidied up some of the mua_wrapper messages at the
3501 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
3502 the messages in a hash of arrays rather than using individual hashes.
3503 This is a bit cleaner and results in dramatic memory savings, albeit
3504 at a slight CPU cost.
3506 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
3507 as requested by Marc Sherman.
3509 SC/10 Eximstats: added histograms for user specified patterns as requested
3512 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
3514 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
3515 fopen() in the content-scanning modules that did not already have it.
3521 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
3522 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
3524 TK/02 Fix ACL "control" statement not being available in MIME ACL.
3526 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
3528 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
3529 to test Sieve filters that use "vacation".
3531 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
3532 that changes the way the GnuTLS parameters are stored in the cache file.
3533 The new format can be generated externally. For backward compatibility,
3534 if the data in the cache doesn't make sense, Exim assumes it has read an
3535 old-format file, and it generates new data and writes a new file. This
3536 means that you can't go back to an older release without removing the
3539 PH/03 A redirect router that has both "unseen" and "one_time" set does not
3540 work if there are any delivery delays because "one_time" forces the
3541 parent to be marked "delivered", so its unseen clone is never tried
3542 again. For this reason, Exim now forbids the simultaneous setting of
3545 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
3546 redirected to themselves ("homonym" addresses). Read the long ChangeLog
3547 entry if you want to know the details. The fix, however, neglected to
3548 consider the case when local delivery batching is involved. The test for
3549 "previously delivered" was not happening when checking to see if an
3550 address could be batched with a previous (undelivered) one; under
3551 certain circumstances this could lead to multiple deliveries to the same
3554 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
3555 in its include files, and this causes problems building Exim.
3557 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
3558 header_syntax) but Exim was just ignoring anything given after a slash.
3559 In particular, this caused confusion with an attempt to use "verify =
3560 reverse_host_lookup/defer_ok". An error is now given when options are
3561 supplied for verify items that do not have them. (Maybe reverse_host_
3562 lookup should have a defer_ok option, but that's a different point.)
3564 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
3565 defined by RFC 821) to 2048, because there were problems with some AUTH
3566 commands, and RFC 1869 says the size should be increased for extended
3567 SMTP commands that take arguments.
3569 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
3572 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
3573 "unknown" error; now it says that the functionality isn't in the binary.
3575 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
3576 an address' error message when a string expansion fails (syntax or
3577 whatever). Otherwise the password may appear in the log. Following change
3578 PH/42 below, there is no longer a chance of it appearing in a bounce
3581 PH/11 Installed exipick version 20050225.0 from John Jetmore.
3583 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
3584 of its addresses was ever tried. (Bugzilla bug #2.)
3586 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
3587 the result incorrectly in the debug output. (It correctly added a newline
3588 to what was transported.)
3590 TF/01 Added $received_time.
3592 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
3593 commented out examples of how to interface to a virus scanner and to
3594 SpamAssassin. Also added commented examples of av_scanner and
3595 spamd_address settings.
3597 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
3598 and controls are allowed in which ACLs. There were a couple of minor
3599 errors. Some of the entries in the conditions table (which is a table of
3600 where they are NOT allowed) were getting very unwieldy; rewrote them as a
3601 negation of where the condition IS allowed.
3603 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
3605 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
3606 header file does not have a version number, so I've had to invent a new
3607 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
3608 API. The code is untested by me (my Linux distribution still has 0.3.2 of
3609 radiusclient), but it was contributed by a Radius user.
3611 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
3612 files or directories, for OpenSSL.
3614 PH/19 When an Exim process that is running as root has to create an Exim log
3615 file, it does so in a subprocess that runs as exim:exim so as to get the
3616 ownership right at creation (otherwise, other Exim processes might see
3617 the file with the wrong ownership). There was no test for failure of this
3618 fork() call, which would lead to the process getting stuck as it waited
3619 for a non-existent subprocess. Forks do occasionally fail when resources
3620 run out. I reviewed all the other calls to fork(); they all seem to check
3623 PH/20 When checking for unexpected SMTP input at connect time (before writing
3624 the banner), Exim was not dealing correctly with a non-positive return
3625 from the read() function. If the client had disconnected by this time,
3626 the result was a log entry for a synchronization error with an empty
3627 string after "input=" when read() returned zero. If read() returned -1
3628 (an event I could not check), uninitialized data bytes were printed.
3629 There were reports of junk text (parts of files, etc) appearing after
3632 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
3634 PH/22 Added support for macro redefinition, and (re)definition in between
3635 driver and ACL definitions.
3637 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
3638 forgetting to use the resulting value; it was using the unexpanded value.
3640 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
3641 hadn't been configured. The fix is from Juergen Kreileder, who
3642 understands it better than I do:
3644 "Here's what I see happening with three configured cyrus_sasl
3645 authenticators configured (plain, login, cram-md5):
3647 On startup auth_cyrus_sasl_init() gets called for each of these.
3648 This means three calls to sasl_listmech() without a specified mech_list.
3649 => SASL tests which mechs of all available mechs actually work
3650 => three warnings about OTP not working
3651 => the returned list contains: plain, login, cram-md5, digest-md5, ...
3653 With the patch, sasl_listmech() also gets called three times. But now
3654 SASL's mech_list option is set to the server_mech specified in the the
3655 authenticator. Or in other words, the answer from sasl_listmech()
3656 gets limited to just the mech you're testing for (which is different
3658 => the return list contains just 'plain' or 'login', 'cram-md5' or
3659 nothing depending on the value of ob->server_mech.
3661 I've just tested the patch: Authentication still works fine,
3662 unavailable mechs specified in the exim configuration are still
3663 caught, and the auth.log warnings about OTP are gone."
3665 PH/25 When debugging is enabled, the contents of the command line are added
3666 to the debugging output, even when log_selector=+arguments is not
3669 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
3670 answer is "GNU", and only if the return is "GNU/something" is the answer
3673 PH/27 $acl_verify_message is now set immediately after the failure of a
3674 verification in an ACL, and so is available in subsequent modifiers. In
3675 particular, the message can be preserved by coding like this:
3677 warn !verify = sender
3678 set acl_m0 = $acl_verify_message
3680 Previously, $acl_verify_message was set only while expanding "message"
3681 and "log_message" when a very denied access.
3683 PH/28 Modified OS/os.c-Linux with
3685 -#ifndef OS_LOAD_AVERAGE
3686 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
3688 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
3689 nomenclature these days.)
3691 PH/29 Installed patch from the Sieve maintainer that adds the options
3692 sieve_useraddress and sieve_subaddress to the redirect router.
3694 PH/30 In these circumstances:
3695 . Two addresses routed to the same list of hosts;
3696 . First host does not offer TLS;
3697 . First host accepts first address;
3698 . First host gives temporary error to second address;
3699 . Second host offers TLS and a TLS session is established;
3700 . Second host accepts second address.
3701 Exim incorrectly logged both deliveries with the TLS parameters (cipher
3702 and peerdn, if requested) that were in fact used only for the second
3705 PH/31 When doing a callout as part of verifying an address, Exim was not paying
3706 attention to any local part prefix or suffix that was matched by the
3707 router that accepted the address. It now behaves in the same way as it
3708 does for delivery: the affixes are removed from the local part unless
3709 rcpt_include_affixes is set on the transport.
3711 PH/32 Add the sender address, as F=<...>, to the log line when logging a
3712 timeout during the DATA phase of an incoming message.
3714 PH/33 Sieve envelope tests were broken for match types other than :is. I have
3715 applied a patch sanctioned by the Sieve maintainer.
3717 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
3718 the uid or gid is negative. A case of a negative gid caused this to be
3719 noticed. The fix allows for either to be negative.
3721 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
3722 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
3725 PH/36 The change PH/12 above was broken. Fixed it.
3727 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
3728 the grounds that routing the same address twice would always produce the
3729 same answer. This might have been true once, but it is certainly no
3730 longer true now. Routing a child address may depend on the previous
3731 routing that produced that child. Some complicated redirection strategies
3732 went wrong when messages had multiple recipients, and made Exim's
3733 behaviour dependent on the order in which the addresses were given.
3735 I have moved the duplicate checking until after the routing is complete.
3736 Exim scans the addresses that are assigned to local and remote
3737 transports, and removes any duplicates. This means that more work will be
3738 done, as duplicates will always all be routed, but duplicates are
3739 presumably rare, so I don't expect this is of any significance.
3741 For deliveries to pipes, files, and autoreplies, the duplicate checking
3742 still happens during the routing process, since they are not going to be
3745 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
3746 It corrects a timeout issue with spamd. This is Ian's comment: "The
3747 background is that sometimes spamd either never reads data from a
3748 connection it has accepted, or it never writes response data. The exiscan
3749 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
3750 blindly assumes that writes won't block so it may never time out."
3752 PH/39 Allow G after quota size as well as K and M.
3754 PH/40 The value set for $authenticated_id in an authenticator may not contain
3755 binary zeroes or newlines because the value is written to log lines and
3756 to spool files. There was no check on this. Now the value is run through
3757 the string_printing() function so that such characters are converted to
3758 printable escape sequences.
3760 PH/41 $message_linecount is a new variable that contains the total number of
3761 lines in the message. Compare $body_linecount, which is the count for the
3764 PH/42 Exim no longer gives details of delivery errors for specific addresses in
3765 bounce and delay warning messages, except in certain special cases, which
3768 (a) An SMTP error message from a remote host;
3769 (b) A message specified in a :fail: redirection;
3770 (c) A message specified in a "fail" command in a system filter;
3771 (d) A message specified in a FAIL return from the queryprogram router;
3772 (e) A message specified by the cannot_route_message router option.
3774 In these cases only, Exim does include the error details in bounce and
3775 warning messages. There are also a few cases where bland messages such
3776 as "unrouteable address" or "local delivery error" are given.
3778 PH/43 $value is now also set for the "else" part of a ${run expansion.
3780 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
3781 being worked on, but at least Exim now implements the latest version to
3784 PH/45 In a pipe transport, although a timeout while waiting for the pipe
3785 process to complete was treated as a delivery failure, a timeout while
3786 writing the message to the pipe was logged, but erroneously treated as a
3787 successful delivery. Such timeouts include transport filter timeouts. For
3788 consistency with the overall process timeout, these timeouts are now
3789 treated as errors, giving rise to delivery failures by default. However,
3790 there is now a new Boolean option for the pipe transport called
3791 timeout_defer, which, if set TRUE, converts the failures into defers for
3792 both kinds of timeout. A transport filter timeout is now identified in
3795 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
3796 systems where "make" and "gmake" are different, calling "gmake" at top
3797 level broke things. I've arranged for the value of $(MAKE) to be passed
3798 from the Makefile to this script so that it can call the same version of
3802 A note about Exim versions 4.44 and 4.50
3803 ----------------------------------------
3805 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
3806 changes of various kinds. As a consequence, a big documentation update was
3807 needed. This delayed the release for rather longer than seemed good, especially
3808 in the light of a couple of (minor) security issues. Therefore, the changes
3809 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
3810 release. So 4.44 and 4.50 are in effect two different branches that both start
3813 I have left the 4.50 change log unchanged; it contains all the changes since
3814 4.43. The change log for 4.44 is below; many of its items are identical to
3815 those for 4.50. This seems to be the most sensible way to preserve the
3816 historical information.
3822 1. Minor wording change to the doc/README.SIEVE file.
3824 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
3825 computation of the current number of files was incorrect.
3827 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
3828 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
3831 4. Give more explanation in the error message when the command for a transport
3832 filter fails to execute.
3834 5. There are several places where Exim runs a non-Exim command in a
3835 subprocess. The SIGUSR1 signal should be disabled for these processes. This
3836 was being done only for the command run by the queryprogram router. It is
3837 now done for all such subprocesses. The other cases are: ${run, transport
3838 filters, and the commands run by the lmtp and pipe transports.
3840 6. Added CONFIGURE_GROUP build-time option.
3842 7. Some older OS have a limit of 256 on the maximum number of file
3843 descriptors. Exim was using setrlimit() to set 1000 as a large value
3844 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
3845 systems. I've change it so that if it can't get 1000, it tries for 256.
3847 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
3848 was an oversight, and furthermore, ever since the addition of extra
3849 controls (e.g. 4.43/32), the checks on when to allow different forms of
3850 "control" were broken. There should now be diagnostics for all cases when a
3851 control that does not make sense is encountered.
3853 9. Added the /retain_sender option to "control=submission".
3855 10. $recipients is now available in the predata ACL (oversight).
3857 11. Tidy the search cache before the fork to do a delivery from a message
3858 received from the command line. Otherwise the child will trigger a lookup
3859 failure and thereby defer the delivery if it tries to use (for example) a
3860 cached ldap connection that the parent has called unbind on.
3862 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
3863 of $address_data from the recipient verification was clobbered by the
3864 sender verification.
3866 13. The value of address_data from a sender verification is now available in
3867 $sender_address_data in subsequent conditions in the ACL statement.
3869 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
3871 15. Added a new option "connect=<time>" to callout options, to set a different
3874 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
3875 was its contents. (It was OK if the option was not defined at all.)
3877 17. A "Completed" log line is now written for messages that are removed from
3878 the spool by the -Mrm option.
3880 18. New variables $sender_verify_failure and $recipient_verify_failure contain
3881 information about exactly what failed.
3883 19. Added -dd to debug only the daemon process.
3885 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
3886 handles timeouts, both on the server side and network timeouts. Renamed the
3887 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
3889 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
3890 It is now set to "smtps".
3892 22. $host_address is now set to the target address during the checking of
3893 ignore_target_hosts.
3895 23. When checking ignore_target_hosts for an ipliteral router, no host name was
3896 being passed; this would have caused $sender_host_name to have been used if
3897 matching the list had actually called for a host name (not very likely,
3898 since this list is usually IP addresses). A host name is now passed as
3901 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
3902 code that specifies a non-restarting handler (typically sigaction() in
3903 modern systems) in an attempt to fix a rare and obscure crash bug.
3905 25. Narrowed the window for a race in the daemon that could cause it to ignore
3906 SIGCHLD signals. This is not a major problem, because they are used only to
3907 wake it up if nothing else does.
3909 26. A malformed maildirsize file could cause Exim to calculate negative values
3910 for the mailbox size or file count. Odd effects could occur as a result.
3911 The maildirsize information is now recalculated if the size or filecount
3914 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
3915 support for a long time. Removed HAVE_SYS_VFS_H.
3917 28. Installed the latest version of exipick from John Jetmore.
3919 29. In an address list, if the pattern was not a regular expression, an empty
3920 subject address (from a bounce message) matched only if the pattern was an
3921 empty string. Non-empty patterns were not even tested. This was the wrong
3922 because it is perfectly reasonable to use an empty address as part of a
3923 database query. An empty address is now tested by patterns that are
3924 lookups. However, all the other forms of pattern expect the subject to
3925 contain a local part and a domain, and therefore, for them, an empty
3926 address still always fails if the pattern is not itself empty.
3928 30. Exim went into a mad DNS loop when attempting to do a callout where the
3929 host was specified on an smtp transport, and looking it up yielded more
3930 than one IP address.
3932 31. Re-factored the code for checking spool and log partition space into a
3933 function that finds that data and another that does the check. The former
3934 is then used to implement four new variables: $spool_space, $log_space,
3935 $spool_inodes, and $log_inodes.
3937 32. The RFC2047 encoding function was originally intended for short strings
3938 such as real names; it was not keeping to the 75-character limit for
3939 encoded words that the RFC imposes. It now respects the limit, and
3940 generates multiple encoded words if necessary. To be on the safe side, I
3941 have increased the buffer size for the ${rfc2047: expansion operator from
3944 33. It is now permitted to omit both strings after an "if" condition; if the
3945 condition is true, the result is "true". As before, when the second string
3946 is omitted, a false condition yields an empty string. This makes it less
3947 cumbersome to write custom ACL and router conditions.
3949 34. Failure to deliver a bounce message always caused it to be frozen, even if
3950 there was an errors_to setting on the router. The errors_to setting is now
3953 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
3954 canonical form (fully expanded) before being placed in
3955 $sender_host_address.
3957 36. The table in the code that translates DNS record types into text (T_A to
3958 "A" for instance) was missing entries for NS and CNAME. It is just possible
3959 that this could have caused confusion if both these types were looked up
3960 for the same domain, because the text type is used as part of Exim's
3961 per-process caching. But the chance of anyone hitting this buglet seems
3964 37. The dnsdb lookup has been extended in a number of ways.
3966 (1) There is a new type, "zns", which walks up the domain tree until it
3967 finds some nameserver records. It should be used with care.
3969 (2) There is a new type, "mxh", which is like "mx" except that it returns
3970 just the host names, not the priorities.
3972 (3) It is now possible to give a list of domains (or IP addresses) to be
3973 looked up. The behaviour when one of the lookups defers can be
3974 controlled by a keyword.
3976 (4) It is now possible to specify the separator character for use when
3977 multiple records are returned.
3979 38. The dnslists ACL condition has been extended: it is now possible to supply
3980 a list of IP addresses and/or domains to be looked up in a particular DNS
3983 39. Added log_selector=+queue_time_overall.
3985 40. When running the queue in the test harness, wait just a tad after forking a
3986 delivery process, to get repeatability of debugging output.
3988 41. Include certificate and key file names in error message when GnuTLS fails
3989 to set them up, because the GnuTLS error message doesn't include the name
3990 of the failing file when there is a problem reading it.
3992 42. Allow both -bf and -bF in the same test run.
3994 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
3996 44. The "Exiscan patch" is now merged into the mainline Exim source.
3998 45. Sometimes the final signoff response after QUIT could fail to get
3999 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4000 < 0 before doing a fflush(). This bug looks as though it goes back to the
4001 introduction of TLS in release 3.20, but "sometimes" must have been rare
4002 because the tests only now provoked it.
4004 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
4005 (this can affect the format of dates).
4007 47. exim_tidydb, when checking for the continued existence of a message for
4008 which it has found a message-specific retry record, was not finding
4009 messages that were in split spool directories. Consequently, it was
4010 deleting retry records that should have stayed in existence.
4012 48. Steve fixed some bugs in eximstats.
4014 49. The SPA authentication driver was not abandoning authentication and moving
4015 on to the next authenticator when an expansion was forced to fail,
4016 contradicting the general specification for all authenticators. Instead it
4017 was generating a temporary error. It now behaves as specified.
4019 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
4020 (the order specifies the preference for clients). The order is now AES256,
4021 AES128, 3DES, ARCFOUR128.
4023 51. Small patch to Sieve code - explicitly set From: when generating an
4026 52. Exim crashed if a remote delivery caused a very long error message to be
4027 recorded - for instance if somebody sent an entire SpamAssassin report back
4028 as a large number of 550 error lines. This bug was coincidentally fixed by
4029 increasing the size of one of Exim's internal buffers (big_buffer) that
4030 happened as part of the Exiscan merge. However, to be on the safe side, I
4031 have made the code more robust (and fixed the comments that describe what
4034 53. Now that there can be additional text after "Completed" in log lines (if
4035 the queue_time_overall log selector is set), a one-byte patch to exigrep
4036 was needed to allow it to recognize "Completed" as not the last thing in
4039 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
4040 patch that reportedly fixes this has been added. I am not expert enough to
4041 create a test for it. This is what the patch creator wrote:
4043 "I found a little strange behaviour of ldap code when working with
4044 Windows 2003 AD Domain, where users was placed in more than one
4045 Organization Units. When I tried to give exim partial DN, the exit code
4046 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
4047 But simultaneously result of request was absolutely normal ldap result,
4048 so I produce this patch..."
4050 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
4051 REFERENCE, so I have modified the code to exclude the patch when that macro
4054 55. Some experimental protocols are using DNS PTR records for new purposes. The
4055 keys for these records are domain names, not reversed IP addresses. The
4056 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4057 leaves it alone. Component reversal etc. now happens only for IP addresses.
4060 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
4062 57. Double the size of the debug message buffer (to 2048) so that more of very
4063 long debug lines gets shown.
4065 58. The exicyclog utility now does better if the number of log files to keep
4066 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
4068 59. Two changes related to the smtp_active_hostname option:
4070 (1) $smtp_active_hostname is now available as a variable.
4071 (2) The default for smtp_banner uses $smtp_active_hostname instead
4072 of $primary_hostname.
4074 60. The host_aton() function is supposed to be passed a string that is known
4075 to be a valid IP address. However, in the case of IPv6 addresses, it was
4076 not checking this. This is a hostage to fortune. Exim now panics and dies
4077 if the condition is not met. A case was found where this could be provoked
4078 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4079 components; fortuitously, this particular loophole had already been fixed
4080 by change 4.50/55 above.
4082 If there are any other similar loopholes, the new check in host_aton()
4083 itself should stop them being exploited. The report I received stated that
4084 data on the command line could provoke the exploit when Exim was running as
4085 exim, but did not say which command line option was involved. All I could
4086 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4087 running as the user.
4090 61. There was a buffer overflow vulnerability in the SPA authentication code
4091 (which came originally from the Samba project). I have added a test to the
4092 spa_base64_to_bits() function which I hope fixes it.
4095 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
4096 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
4098 63. The daemon start-up calls getloadavg() while still root for those OS that
4099 need the first call to be done as root, but it missed one case: when
4100 deliver_queue_load_max is set with deliver_drop_privilege. This is
4101 necessary for the benefit of the queue runner, because there is no re-exec
4102 when deliver_drop_privilege is set.
4104 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
4105 This has been fixed.
4107 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4108 was in use, was not putting the data itself into the right store pool;
4109 consequently, it could be overwritten for a subsequent message in the same
4110 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4113 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
4115 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
4116 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
4117 the function were treating the return as a boolean value, which happened to
4118 work because 0=false and not-0=true, but is not correct code.
4120 68. The host_aton() function was not handling scoped IPv6 addresses (those
4121 with, for example, "%eth0" on the end) correctly.
4123 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
4124 negated items (that is, ~something) in unsigned ints. Some compilers
4125 apparently mutter when there is no cast.
4127 70. If an address verification called from an ACL failed, and did not produce a
4128 user-specific message (i.e. there was only a "system" message), nothing was
4129 put in $acl_verify_message. In this situation, it now puts the system
4132 71. Change 4.23/11 added synchronization checking at the start of an SMTP
4133 session; change 4.31/43 added the unwanted input to the log line - except
4134 that it did not do this in the start of session case. It now does.
4136 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
4137 This is wrong and can cause the other end to generate a synchronization
4138 error if it is another Exim or anything else that does the synchronization
4139 check. A QUIT command is no longer sent after a timeout.
4141 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
4142 during host lookups.
4144 74. The defer_ok option of callout verification was not working if it was used
4145 when verifying addresses in header lines, that is, for this case:
4147 verify = header_sender/callout=defer_ok
4149 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
4150 those file descriptors could be used for SMTP connections. If anything
4151 wrote to stderr (the example that came up was "warn" in embedded Perl), it
4152 could be sent to the SMTP client, causing chaos. The daemon now opens
4153 stdin, stdout, and stderr to /dev/null when it puts itself into the
4156 76. Arrange for output from Perl's "warn" command to be written to Exim's main
4157 log by default. The user can override this with suitable Perl magic.
4159 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
4160 the log message when discard triggers, was not working for the DATA ACL or
4161 for the non-SMTP ACL.
4163 78. Error message wording change in sieve.c.
4165 79. If smtp_accept_max_per_host was set, the number of connections could be
4166 restricted to fewer than expected, because the daemon was trying to set up
4167 a new connection before checking whether the processes handling previous
4168 connections had finished. The check for completed processes is now done
4169 earlier. On busy systems, this bug wouldn't be noticed because something
4170 else would have woken the daemon, and it would have reaped the completed
4173 80. If a message was submitted locally by a user whose login name contained one
4174 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
4175 It caused a spool format error. I have fixed the spool reading code. A
4176 related problem was that the "from" clause in the Received: line became
4177 illegal because of the space(s). It is now covered by ${quote_local_part.
4179 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
4182 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
4184 83. Patch from Sieve maintainer to fix unterminated string problem in
4185 "vacation" handling.
4187 84. Some minor changes to the Linux configuration files to help with other
4188 OS variants using glibc.
4190 85. One more patch for Sieve to update vacation handling to latest spec.
4193 ----------------------------------------------------
4194 See the note above about the 4.44 and 4.50 releases.
4195 ----------------------------------------------------
4201 1. Change 4.43/35 introduced a bug that caused file counts to be
4202 incorrectly computed when quota_filecount was set in an appendfile
4205 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4206 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4209 3. Give more explanation in the error message when the command for a transport
4210 filter fails to execute.
4212 4. There are several places where Exim runs a non-Exim command in a
4213 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4214 was being done only for the command run by the queryprogram router. It is
4215 now done for all such subprocesses. The other cases are: ${run, transport
4216 filters, and the commands run by the lmtp and pipe transports.
4218 5. Some older OS have a limit of 256 on the maximum number of file
4219 descriptors. Exim was using setrlimit() to set 1000 as a large value
4220 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4221 systems. I've change it so that if it can't get 1000, it tries for 256.
4223 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4224 was an oversight, and furthermore, ever since the addition of extra
4225 controls (e.g. 4.43/32), the checks on when to allow different forms of
4226 "control" were broken. There should now be diagnostics for all cases when a
4227 control that does not make sense is encountered.
4229 7. $recipients is now available in the predata ACL (oversight).
4231 8. Tidy the search cache before the fork to do a delivery from a message
4232 received from the command line. Otherwise the child will trigger a lookup
4233 failure and thereby defer the delivery if it tries to use (for example) a
4234 cached ldap connection that the parent has called unbind on.
4236 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4237 of $address_data from the recipient verification was clobbered by the
4238 sender verification.
4240 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4241 was its contents. (It was OK if the option was not defined at all.)
4243 11. A "Completed" log line is now written for messages that are removed from
4244 the spool by the -Mrm option.
4246 12. $host_address is now set to the target address during the checking of
4247 ignore_target_hosts.
4249 13. When checking ignore_target_hosts for an ipliteral router, no host name was
4250 being passed; this would have caused $sender_host_name to have been used if
4251 matching the list had actually called for a host name (not very likely,
4252 since this list is usually IP addresses). A host name is now passed as
4255 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4256 code that specifies a non-restarting handler (typically sigaction() in
4257 modern systems) in an attempt to fix a rare and obscure crash bug.
4259 15. Narrowed the window for a race in the daemon that could cause it to ignore
4260 SIGCHLD signals. This is not a major problem, because they are used only to
4261 wake it up if nothing else does.
4263 16. A malformed maildirsize file could cause Exim to calculate negative values
4264 for the mailbox size or file count. Odd effects could occur as a result.
4265 The maildirsize information is now recalculated if the size or filecount
4268 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4269 support for a long time. Removed HAVE_SYS_VFS_H.
4271 18. Updated exipick to current release from John Jetmore.
4273 19. Allow an empty sender to be matched against a lookup in an address list.
4274 Previously the only cases considered were a regular expression, or an
4277 20. Exim went into a mad DNS lookup loop when doing a callout where the
4278 host was specified on the transport, if the DNS lookup yielded more than
4281 21. The RFC2047 encoding function was originally intended for short strings
4282 such as real names; it was not keeping to the 75-character limit for
4283 encoded words that the RFC imposes. It now respects the limit, and
4284 generates multiple encoded words if necessary. To be on the safe side, I
4285 have increased the buffer size for the ${rfc2047: expansion operator from
4288 22. Failure to deliver a bounce message always caused it to be frozen, even if
4289 there was an errors_to setting on the router. The errors_to setting is now
4292 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4293 canonical form (fully expanded) before being placed in
4294 $sender_host_address.
4296 24. Updated eximstats to version 1.33
4298 25. Include certificate and key file names in error message when GnuTLS fails
4299 to set them up, because the GnuTLS error message doesn't include the name
4300 of the failing file when there is a problem reading it.
4302 26. Expand error message when OpenSSL has problems setting up cert/key files.
4305 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
4306 (this can affect the format of dates).
4308 28. exim_tidydb, when checking for the continued existence of a message for
4309 which it has found a message-specific retry record, was not finding
4310 messages that were in split spool directories. Consequently, it was
4311 deleting retry records that should have stayed in existence.
4313 29. eximstats updated to version 1.35
4314 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
4315 1.35 - bugfix such that pie charts by volume are generated correctly
4317 30. The SPA authentication driver was not abandoning authentication and moving
4318 on to the next authenticator when an expansion was forced to fail,
4319 contradicting the general specification for all authenticators. Instead it
4320 was generating a temporary error. It now behaves as specified.
4322 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
4323 (the order specifies the preference for clients). The order is now AES256,
4324 AES128, 3DES, ARCFOUR128.
4326 31. Small patch to Sieve code - explicitly set From: when generating an
4329 32. Exim crashed if a remote delivery caused a very long error message to be
4330 recorded - for instance if somebody sent an entire SpamAssassin report back
4331 as a large number of 550 error lines. This bug was coincidentally fixed by
4332 increasing the size of one of Exim's internal buffers (big_buffer) that
4333 happened as part of the Exiscan merge. However, to be on the safe side, I
4334 have made the code more robust (and fixed the comments that describe what
4337 33. Some experimental protocols are using DNS PTR records for new purposes. The
4338 keys for these records are domain names, not reversed IP addresses. The
4339 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4340 leaves it alone. Component reversal etc. now happens only for IP addresses.
4343 34. The host_aton() function is supposed to be passed a string that is known
4344 to be a valid IP address. However, in the case of IPv6 addresses, it was
4345 not checking this. This is a hostage to fortune. Exim now panics and dies
4346 if the condition is not met. A case was found where this could be provoked
4347 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4348 components; fortuitously, this particular loophole had already been fixed
4349 by change 4.50/55 or 4.44/33 above.
4351 If there are any other similar loopholes, the new check in host_aton()
4352 itself should stop them being exploited. The report I received stated that
4353 data on the command line could provoke the exploit when Exim was running as
4354 exim, but did not say which command line option was involved. All I could
4355 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4356 running as the user.
4359 35. There was a buffer overflow vulnerability in the SPA authentication code
4360 (which came originally from the Samba project). I have added a test to the
4361 spa_base64_to_bits() function which I hope fixes it.
4364 36. The daemon start-up calls getloadavg() while still root for those OS that
4365 need the first call to be done as root, but it missed one case: when
4366 deliver_queue_load_max is set with deliver_drop_privilege. This is
4367 necessary for the benefit of the queue runner, because there is no re-exec
4368 when deliver_drop_privilege is set.
4370 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4371 was in use, was not putting the data itself into the right store pool;
4372 consequently, it could be overwritten for a subsequent message in the same
4373 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4376 38. Sometimes the final signoff response after QUIT could fail to get
4377 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4378 < 0 before doing a fflush(). This bug looks as though it goes back to the
4379 introduction of TLS in release 3.20, but "sometimes" must have been rare
4380 because the tests only now provoked it.
4386 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
4387 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
4388 Now it returns an integer. A number of calls were still expecting a T/F
4389 return. Fortuitously, in all cases, the tests worked in OK situations,
4390 which is the norm. However, things would have gone wrong on any write
4391 failures on the smtp file descriptor. This function is used when sending
4392 messages over SMTP and also when doing verify callouts.
4394 2. When Exim is called to do synchronous delivery of a locally submitted
4395 message (the -odf or -odi options), it no longer closes stderr before doing
4398 3. Implemented the mua_wrapper option.
4400 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
4402 5. Implemented the functions header_remove(), header_testname(),
4403 header_add_at_position(), and receive_remove_recipient(), and exported them
4406 6. If an ACL "warn" statement specified the addition of headers, Exim already
4407 inserted X-ACL-Warn: at the start if there was no header name. However, it
4408 was not making this test for the second and subsequent header lines if
4409 there were newlines in the string. This meant that an invalid header could
4410 be inserted if Exim was badly configured.
4412 7. Allow an ACL "warn" statement to add header lines at the start or after all
4413 the Received: headers, as well as at the end.
4415 8. Added the rcpt_4xx retry error code.
4417 9. Added postmaster_mailfrom=xxx to callout verification option.
4419 10. Added mailfrom=xxxx to the callout verification option, for verify=
4422 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
4423 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
4425 12. Inserted some casts to stop certain compilers warning when using pointer
4426 differences as field lengths or precisions in printf-type calls (mostly
4427 affecting debugging statements).
4429 13. Added optional readline() support for -be (dynamically loaded).
4431 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
4432 same clock tick as a message's arrival, so that its received time was the
4433 same as the "first fail" time on the retry record, and that message
4434 remained on the queue past the ultimate address timeout, every queue runner
4435 would try a delivery (because it was past the ultimate address timeout) but
4436 after another failure, the ultimate address timeout, which should have then
4437 bounced the address, did not kick in. This was a "< instead of <=" error;
4438 in most cases the first failure would have been in the next clock tick
4439 after the received time, and all would be well.
4441 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
4442 being recognized when the domain list was tested by the match_domain
4443 condition in an expansion string.
4445 16. Added the ${str2b64: operator.
4447 17. Exim was always calling setrlimit() to set a large limit for the number of
4448 processes, without checking whether the existing limit was already
4449 adequate. (It did check for the limit on file descriptors.) Furthermore,
4450 errors from getrlimit() and setrlimit() were being ignored. Now they are
4451 logged to the main and panic logs, but Exim does carry on, to try to do its
4452 job under whatever limits there are.
4454 18. Imported PCRE 5.0.
4456 19. Trivial typo in log message " temporarily refused connection" (the leading
4459 20. If the log selector return_path_on_delivery was set and an address was
4460 redirected to /dev/null, the delivery process crashed because it assumed
4461 that a return path would always be set for a "successful" delivery. In this
4462 case, the whole delivery is bypassed as an optimization, and therefore no
4465 21. Internal re-arrangement: the function for sending a challenge and reading
4466 a response while authentication was assuming a zero-terminated challenge
4467 string. It's now changed to take a pointer and a length, to allow for
4468 binary data in such strings.
4470 22. Added the cyrus_sasl authenticator (code supplied by MBM).
4472 23. Exim was not respecting finduser_retries when seeking the login of the
4473 uid under which it was called; it was always trying 10 times. (The default
4474 setting of finduser_retries is zero.) Also, it was sleeping after the final
4475 failure, which is pointless.
4477 24. Implemented tls_on_connect_ports.
4479 25. Implemented acl_smtp_predata.
4481 26. If the domain in control=submission is set empty, Exim assumes that the
4482 authenticated id is a complete email address when it generates From: or
4483 Sender: header lines.
4485 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
4486 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
4487 chown and chgrp in /bin and hostname in /usr/bin.
4489 28. Exim was keeping the "process log" file open after each use, just as it
4490 does for the main log. This opens the possibility of it remaining open for
4491 long periods when the USR1 signal hits a daemon. Occasional processlog
4492 errors were reported, that could have been caused by this. Anyway, it seems
4493 much more sensible not to leave this file open at all, so that is what now
4496 29. The long-running daemon process does not normally write to the log once it
4497 has entered its main loop, and it closes the log before doing so. This is
4498 so that log files can straightforwardly be renamed and moved. However,
4499 there are a couple of unusual error situations where the daemon does write
4500 log entries, and I had neglected to close the log afterwards.
4502 30. The text of an SMTP error response that was received during a remote
4503 delivery was being truncated at 512 bytes. This is too short for some of
4504 the long messages that one sometimes sees. I've increased the limit to
4507 31. It is now possible to make retry rules that apply only when a message has a
4508 specific sender, in particular, an empty sender.
4510 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
4511 it possible to be selective about when SMTP synchronization is enforced.
4513 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
4515 32. Implemented hosts_connection_nolog.
4517 33. Added an ACL for QUIT.
4519 34. Setting "delay_warning=" to disable warnings was not working; it gave a
4522 35. Added mailbox_size and mailbox_filecount to appendfile.
4524 36. Added control = no_multiline_responses to ACLs.
4526 37. There was a bug in the logic of the code that waits for the clock to tick
4527 in the case where the clock went backwards by a substantial amount such
4528 that the microsecond fraction of "now" was more than the microsecond
4529 fraction of "then" (but the whole seconds number was less).
4531 38. Added support for the libradius Radius client library this is found on
4532 FreeBSD (previously only the radiusclient library was supported).
4538 1. When certain lookups returned multiple values in the form name=value, the
4539 quoting of the values was not always being done properly. Specifically:
4540 (a) If the value started with a double quote, but contained no whitespace,
4542 (b) If the value contained whitespace other than a space character (i.e.
4543 tabs or newlines or carriage returns) it was not quoted.
4544 This fix has been applied to the mysql and pgsql lookups by writing a
4545 separate quoting function and calling it from the lookup code. The fix
4546 should probably also be applied to nisplus, ibase and oracle lookups, but
4547 since I cannot test any of those, I have not disturbed their existing code.
4549 2. A hit in the callout cache for a specific address caused a log line with no
4550 reason for rejecting RCPT. Now it says "Previous (cached) callout
4551 verification failure".
4553 3. There was an off-by-one bug in the queryprogram router. An over-long
4554 return line was truncated at 256 instead of 255 characters, thereby
4555 overflowing its buffer with the terminating zero. As well as fixing this, I
4556 have increased the buffer size to 1024 (and made a note to document this).
4558 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
4559 when Exim is waiting for an SMTP response from a remote server, Exim
4560 restarts its select() call on the socket, thereby resetting its timeout.
4561 This is not a problem when such interrupts are rare. Somebody set up a cron
4562 job to run exiwhat every 2 minutes, which is less than the normal select()
4563 timeout (5 or 10 minutes). This meant that the select() timeout never
4564 kicked in because it was always reset. I have fixed this by comparing the
4565 time when an interrupt arrives with the time at the start of the first call
4566 to select(). If more time than the timeout has elapsed, the interrupt is
4567 treated as a timeout.
4569 5. Some internal re-factoring in preparation for the addition of Sieve
4570 extensions (by MH). In particular, the "personal" test is moved to a
4571 separate function, and given an option for scanning Cc: and Bcc: (which is
4572 not set for Exim filters).
4574 6. When Exim created an email address using the login of the caller as the
4575 local part (e.g. when creating a From: or Sender: header line), it was not
4576 quoting the local part when it contained special characters such as @.
4578 7. Installed new OpenBSD configuration files.
4580 8. Reworded some messages for syntax errors in "and" and "or" conditions to
4581 try to make them clearer.
4583 9. Callout options, other than the timeout value, were being ignored when
4584 verifying sender addresses in header lines. For example, when using
4586 verify = header_sender/callout=no_cache
4588 the cache was (incorrectly) being used.
4590 10. Added a missing instance of ${EXE} to the exim_install script; this affects
4591 only the Cygwin environment.
4593 11. When return_path_on_delivery was set as a log selector, if different remote
4594 addresses in the same message used different return paths and parallel
4595 remote delivery occurred, the wrong values would sometimes be logged.
4596 (Whenever a remote delivery process finished, the return path value from
4597 the most recently started remote delivery process was logged.)
4599 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
4600 lines when AUTH and/or TLS are in use. This is the "received protocol"
4601 field. Exim used to use "asmtp" for authenticated SMTP, without any
4602 indication (in the protocol name) for TLS use. Now it follows the RFC and
4603 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
4604 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
4605 names appear in log lines as well as in Received: header lines.
4607 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
4608 extensions, and comparison tests, and to fix some bugs.
4610 14. Changes to the "personal" filter test:
4612 (1) The test was buggy in that it was just doing the equivalent of
4613 "contains" tests on header lines. For example, if a user's address was
4614 anne@some.where, the "personal" test would incorrectly be true for
4616 To: susanne@some.where
4618 This test is now done by extracting each address from the header in turn,
4619 and checking the entire address. Other tests that are part of "personal"
4620 are now done using regular expressions (for example, to check local parts
4621 of addresses in From: header lines).
4623 (2) The list of non-personal local parts in From: addresses has been
4624 extended to include "listserv", "majordomo", "*-request", and "owner-*",
4625 taken from the Sieve specification recommendations.
4627 (3) If the message contains any header line starting with "List-" it is
4628 treated as non-personal.
4630 (4) The test for "circular" in the Subject: header line has been removed
4631 because it now seems ill-conceived.
4633 15. Minor typos in src/EDITME comments corrected.
4635 16. Installed latest exipick from John Jetmore.
4637 17. If headers_add on a router specified a text string that was too long for
4638 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
4639 of string_sprintf() is now avoided.
4641 18. $message_body_size was not set (it was always zero) when running the DATA
4642 ACL and the local_scan() function.
4644 19. For the "mail" command in an Exim filter, no default was being set for
4645 the once_repeat time, causing a random time value to be used if "once" was
4646 specified. (If the value happened to be <= 0, no repeat happened.) The
4647 default is now 0s, meaning "never repeat". The "vacation" command was OK
4648 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
4649 (I found it when inspecting the code).
4651 20. There is now an overall timeout for performing a callout verification. It
4652 defaults to 4 times the callout timeout, which applies to individual SMTP
4653 commands during the callout. The overall timeout applies when there is more
4654 than one host that can be tried. The timeout is checked before trying the
4655 next host. This prevents very long delays if there are a large number of
4656 hosts and all are timing out (e.g. when the network connections are timing
4657 out). The value of the overall timeout can be changed by specifying an
4658 additional sub-option for "callout", called "maxwait". For example:
4660 verify = sender/callout=5s,maxwait=20s
4662 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
4663 to the end before writing, but this should make it even safer).
4665 22. Exim was forgetting that it had advertised PIPELINING for the second and
4666 subsequent messages on an SMTP connection. It was also not resetting its
4667 memory on STARTTLS and an internal HELO.
4669 23. When Exim logs an SMTP synchronization error within a session, it now
4670 records whether PIPELINING has been advertised or not.
4672 24. Added 3 instances of "(long int)" casts to time_t variables that were being
4673 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
4674 rather than long int.
4676 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
4678 26. Added the never_mail option to autoreply.
4684 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
4685 crash if the getsockname() call failed; this can happen if a connection is
4686 closed very soon after it is established. The problem was simply in the
4687 order in which certain operations were done, causing Exim to try to write
4688 to the SMTP stream before it had set up the file descriptor. The bug has
4689 been fixed by making things happen in the correct order.
4695 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
4696 before the connection was closed, thus losing the rejection response.
4698 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
4699 some early Solaris releases, but causes trouble in current releases where
4700 socklen_t is defined.
4702 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
4705 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
4708 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
4709 that releasing the top of it at the end releases what was used for sub-
4710 expansions (unless the block got too big). However, discard this block if
4711 the first thing is a variable or header, so that we can use its block when
4712 it is dynamic (useful for very large $message_headers, for example).
4714 6. Lookups now cache *every* query, not just the most recent. A new, separate
4715 store pool is used for this. It can be recovered when all lookup caches are
4716 flushed. Lookups now release memory at the end of their result strings.
4717 This has involved some general refactoring of the lookup sources.
4719 7. Some code has been added to the store_xxx() functions to reduce the amount
4720 of flapping under certain conditions.
4722 8. log_incoming_interface used to affect only the <= reception log lines. Now
4723 it causes the local interface and port to be added to several more SMTP log
4724 lines, for example "SMTP connection from", and rejection lines.
4726 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
4728 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
4730 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
4731 could be overwritten at the end of the current message (or the start of a
4732 new message if it was set in a HELO ACL). The value is now preserved for
4733 the duration of the SMTP connection.
4735 12. If a transport had a headers_rewrite setting, and a matching header line
4736 contained an unqualified address, that address was qualified, even if it
4737 did not match any rewriting rules. The underlying bug was that the values
4738 of the flags that permit the existence of unqualified sender and recipient
4739 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
4740 non-local messages, and by -bnq for local messages) were not being
4741 preserved with the message after it was received.
4743 13. When Exim was logging an SMTP synchronization error, it could sometimes log
4744 "next input=" as part of the text comprising the host identity instead of
4745 the correct text. The code was using the same buffer for two different
4746 strings. However, depending on which order the printing function evaluated
4747 its arguments, the bug did not always show up. Under Linux, for example, my
4748 test suite worked just fine.
4750 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
4751 doesn't work with some older versions of Perl. It has been changed to "my",
4752 which in any case is probably the better facility to use.
4754 15. A really picky compiler found some instances of statements for creating
4755 error messages that either had too many or two few arguments for the format
4758 16. The size of the buffer for calls to the DNS resolver has been increased
4759 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
4760 for addresses that have a lot of PTR records. This alleviates a problem; it
4761 does not fully solve it.
4763 17. A dnsdb lookup for PTR records that receives more data than will fit in the
4764 buffer now truncates the list and logs the incident, which is the same
4765 action as happens when Exim is looking up a host name and its aliases.
4766 Previously in this situation something unpredictable would happen;
4767 sometimes it was "internal error: store_reset failed".
4769 18. If a server dropped the connection unexpectedly when an Exim client was
4770 using GnuTLS and trying to read a response, the client delivery process
4771 crashed while trying to generate an error log message.
4773 19. If a "warn" verb in an ACL added multiple headers to a message in a single
4774 string, for example:
4776 warn message = H1: something\nH2: something
4778 the text was added as a single header line from Exim's point of view
4779 though it ended up OK in the delivered message. However, searching for the
4780 second and subsequent header lines using $h_h2: did not work. This has been
4781 fixed. Similarly, if a system filter added multiple headers in this way,
4782 the routers could not see them.
4784 20. Expanded the error message when iplsearch is called with an invalid key to
4785 suggest using net-iplsearch in a host list.
4787 21. When running tests using -bh, any delays imposed by "delay" modifiers in
4788 ACLs are no longer actually imposed (and a message to that effect is
4791 22. If a "gecos" field in a passwd entry contained escaped characters, in
4792 particular, if it contained a \" sequence, Exim got it wrong when building
4793 a From: or a Sender: header from that name. A second bug also caused
4794 incorrect handling when an unquoted " was present following a character
4795 that needed quoting.
4797 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
4798 was not being matched caselessly.
4800 24. Arranged for all hyphens in the exim.8 source to be escaped with
4803 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
4804 properly. Recipient callout cache records were still being keyed to include
4805 the sender, even when use_sender was set false. This led to far more
4806 callouts that were necessary. The sender is no longer included in the key
4807 when use_sender is false.
4809 26. Added "control = submission" modifier to ACLs.
4811 27. Added the ${base62d: operator to decode base 62 numbers.
4813 28. dnsdb lookups can now access SRV records.
4815 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
4816 the configuration file.
4818 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
4819 (-v) mode. This makes the output for a verbose queue run more intelligible.
4821 31. Added a use_postmaster feature to recipient callouts.
4823 32. Added the $body_zerocount variable, containing the number of binary zero
4824 bytes in the message body.
4826 33. The time of last modification of the "new" subdirectory is now used as the
4827 "mailbox time last read" when there is a quota error for a maildir
4830 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
4832 35. Added +ignore_unknown as a special item in host lists.
4834 36. Code for decoding IPv6 addresses in host lists is now included, even if
4835 IPv6 support is not being compiled. This fixes a bug in which an IPv6
4836 address was recognized as an IP address, but was then not correctly decoded
4837 into binary, causing unexpected and incorrect effects when compared with
4844 1. Very minor rewording of debugging text in manualroute to say "list of
4845 hosts" instead of "hostlist".
4847 2. If verify=header_syntax was set, and a header line with an unqualified
4848 address (no domain) and a large number of spaces between the end of the
4849 name and the colon was received, the reception process suffered a buffer
4850 overflow, and (when I tested it) crashed. This was caused by some obsolete
4851 code that should have been removed. The fix is to remove it!
4853 3. When running in the test harness, delay a bit after writing a bounce
4854 message to get a bit more predictability in the log output.
4856 4. Added a call to search_tidyup() just before forking a reception process. In
4857 theory, someone could use a lookup in the expansion of smtp_accept_max_
4858 per_host which, without the tidyup, could leave open a database connection.
4860 5. Added the variables $recipient_data and $sender_data which get set from a
4861 lookup success in an ACL "recipients" or "senders" condition, or a router
4862 "senders" option, similar to $domain_data and $local_part_data.
4864 6. Moved the writing of debug_print from before to after the "senders" test
4867 7. Change 4.31/66 (moving the time when the Received: is generated) caused
4868 problems for message scanning, either using a data ACL, or using
4869 local_scan() because the Received: header was not generated till after they
4870 were called (in order to set the time as the time of reception completion).
4871 I have revised the way this works. The header is now generated after the
4872 body is received, but before the ACL or local_scan() are called. After they
4873 are run, the timestamp in the header is updated.
4879 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
4880 before starting a queue runner without re-exec. This happened only when
4881 deliver_drop_privilege was set or when the Exim user was set to root. The
4882 effect of the bug was that timeouts during subsequent deliveries caused
4883 crashes instead of being properly handled. The handler is now left at its
4884 default (and expected) setting.
4886 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
4887 message, again when deliver_drop_privilege is set or Exim is run as root.
4888 The bug described in (1) was not present in this case, but the tidying up
4889 of the other signals was missing. I have made the two cases consistent.
4891 3. The ignore_target_hosts setting on a manualroute router was being ignored
4892 for hosts that were looked up using the /MX notation.
4894 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
4897 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
4898 operated on the sender address. After changing the $sender_address to <>
4899 for the sender address verify, Exim was re-instated it as the original
4900 (before rewriting) address, but remembering that it had rewritten it, so it
4901 wasn't rewriting it again. This bug also had the effect of breaking the
4902 sender address verification caching when the sender address was rewritten.
4904 6. The ignore_target_hosts option was being ignored by the ipliteral router.
4905 This has been changed so that if the ip literal address matches
4906 ignore_target_hosts, the router declines.
4908 7. Added expansion conditions match_domain, match_address, and match_local_
4909 part (NOT match_host).
4911 8. The placeholder for the Received: header didn't have a length field set.
4913 9. Added code to Exim itself and to exim_lock to test for a specific race
4914 condition that could lead to file corruption when using MBX delivery. The
4915 issue is with the lockfile that is created in /tmp. If this file is removed
4916 after a process has opened it but before that process has acquired a lock,
4917 there is the potential for a second process to recreate the file and also
4918 acquire a lock. This could lead to two Exim processes writing to the file
4919 at the same time. The added code performs the same test as UW imapd; it
4920 checks after acquiring the lock that its file descriptor still refers to
4921 the same named file.
4923 10. The buffer for building added header lines was of fixed size, 8192 bytes.
4924 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
4927 11. Added the smtp_active_hostname option. If used, this will typically be made
4928 to depend on the incoming interface address. Because $interface_address is
4929 not set up until the daemon has forked a reception process, error responses
4930 that can happen earlier (such as "too many connections") no longer contain
4933 12. If an expansion in a condition on a "warn" statement fails because a lookup
4934 defers, the "warn" statement is abandoned, and the next ACL statement is
4935 processed. Previously this caused the whole ACL to be aborted.
4937 13. Added the iplsearch lookup type.
4939 14. Added ident_timeout as a log selector.
4941 15. Added tls_certificate_verified as a log selector.
4943 16. Added a global option tls_require_ciphers (compare the smtp transport
4944 option of the same name). This controls incoming TLS connections.
4946 17. I finally figured out how to make tls_require_ciphers do a similar thing
4947 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
4948 before starting the TLS session.
4950 18. Tabs are now shown as \t in -bP output.
4952 19. If the log selector return_path_on_delivery was set, Exim crashed when
4953 bouncing a message because it had too many Received: header lines.
4955 20. If two routers both had headers_remove settings, and the first one included
4956 a superfluous trailing colon, the final name in the first list and the
4957 first name in the second list were incorrectly joined into one item (with a
4958 colon in the middle).
4964 1. Added -C and -D options to the exinext utility, mainly to make it easier
4965 to include in the automated testing, but these could be helpful when
4966 multiple configurations are in use.
4968 2. The exinext utility was not formatting the output nicely when there was
4969 an alternate port involved in the retry record key, nor when there was a
4970 message id as well (for retries that were specific to a specific message
4971 and a specific host). It was also confused by IPv6 addresses, because of
4972 the additional colons they contain. I have fixed the IPv4 problem, and
4973 patched it up to do a reasonable job for IPv6.
4975 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
4976 delivery, the log line now contains "pipelined" if PIPELINING was used.
4978 4. An SMTP transport process used to panic and die if the bind() call to set
4979 an explicit outgoing interface failed. This has been changed; it is now
4980 treated in the same way as a connect() failure.
4982 5. A reference to $sender_host_name in the part of a conditional expansion
4983 that was being skipped was still causing a DNS lookup. This no longer
4986 6. The def: expansion condition was not recognizing references to header lines
4987 that used bh_ and bheader_.
4989 7. Added the _cache feature to named lists.
4991 8. The code for checking quota_filecount in the appendfile transport was
4992 allowing one more file than it should have been.
4994 9. For compatibility with Sendmail, the command line option
5002 and sets the incoming protocol and host name (for trusted callers). The
5003 host name and its colon can be omitted when only the protocol is to be set.
5004 Note the Exim already has two private options, -pd and -ps, that refer to
5005 embedded Perl. It is therefore impossible to set a protocol value of "d" or
5006 "s", but I don't think that's a major issue.
5008 10. A number of refactoring changes to the code, none of which should affect
5011 (a) The number of logging options was getting close to filling up the
5012 32-bit word that was used as a bit map. I have split them into two classes:
5013 those that are passed in the argument to log_write(), and those that are
5014 only ever tested independently outside of that function. These are now in
5015 separate 32-bit words, so there is plenty of room for expansion again.
5016 There is no change in the user interface or the logging behaviour.
5018 (b) When building, for example, log lines, the code previously used a
5019 macro that called string_cat() twice, in order to add two strings. This is
5020 not really sufficiently general. Furthermore, there was one instance where
5021 it was actually wrong because one of the argument was used twice, and in
5022 one call a function was used. (As it happened, calling the function twice
5023 did not affect the overall behaviour.) The macro has been replaced by a
5024 function that can join an arbitrary number of extra strings onto a growing
5027 (c) The code for expansion conditions now uses a table and a binary chop
5028 instead of a serial search (which was left over from when there were very
5029 few conditions). Also, it now recognizes conditions like "pam" even when
5030 the relevant support is not compiled in: a suitably worded error message is
5031 given if an attempt is made to use such a condition.
5033 11. Added ${time_interval:xxxxx}.
5035 12. A bug was causing one of the ddress fields not to be passed back correctly
5036 from remote delivery subprocesses. The field in question was not being
5037 subsequently used, so this caused to problems in practice.
5039 13. Added new log selectors queue_time and deliver_time.
5041 14. Might have fixed a bug in maildirsizefile handling that threw up
5042 "unexpected character" debug warnings, and recalculated the data
5043 unnecessarily. In any case, I expanded the warning message to give more
5046 15. Added the message "Restricted characters in address" to the statements in
5047 the default ACL that block characters like @ and % in local parts.
5049 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
5050 Three changes have been made:
5052 (a) There was a serious bug; a negative response to MAIL caused the whole
5053 recipient domain to be cached as invalid, thereby blocking all messages
5054 to all local parts at the same domain, from all senders. This bug has
5055 been fixed. The domain is no longer cached after a negative response to
5056 MAIL if the sender used is not empty.
5058 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
5061 (c) A new callout option, "use_sender" has been added for people who want
5062 the modified behaviour.
5068 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
5071 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
5072 indeed breaks things for older releases.
5074 3. Added additional logging to the case where there is a problem reading data
5075 from a filter that is running in a subprocess using a pipe, in order to
5076 try to track down a specific problem.
5078 4. Testing facility fudge: when running in the test harness and attempting
5079 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
5080 getting "No route to host". Convert this to a timeout.
5082 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
5085 6. Some OS don't have socklen_t but use size_t instead. This affects the
5086 fifth argument of getsockopt() amongst other things. This is now
5087 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
5088 can be set for individual OS. I have set it for SunOS5, OSF1, and
5089 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
5090 some earlier ones do not.
5092 7. Change 4.30/15 was not doing the test caselessly.
5094 8. The standard form for an IPv6 address literal was being rejected by address
5095 parsing in, for example, MAIL and RCPT commands. An example of this kind of
5096 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5097 this, as well as the form without the "IPv6" on the front (but only when
5098 address literals are enabled, of course).
5100 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
5102 10. Exim crashed if a message with an empty sender address specified by -f
5103 encountered a router with an errors_to setting. This could be provoked only
5104 by a command such as
5108 where an empty string was supplied; "<>" did not hit this bug.
5110 11. Installed PCRE release 4.5.
5112 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
5113 remained set. It is now erased.
5115 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
5116 times from message ids (which are base 36 rather than the normal 62).
5118 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
5119 were being counted as actual protocol errors, and logged if the log
5120 selector +smtp_protocol_error was set. One cannot be perfect in this test,
5121 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
5122 and DATA following a set of rejected RCPTs do not count as protocol errors.
5123 In other words, Exim assumes they were pipelined, though this may not
5124 actually be the case. Of course, in all cases the client gets an
5125 appropriate error code.
5127 15. If a lookup fails in an ACL condition, a message about the failure may
5128 be available; it is used if testing the ACL cannot continue, because most
5129 such messages specify what the cause of the deferral is. However, some
5130 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
5131 that caused an old message to be retained and used if a later statement
5132 caused a defer, replacing the real cause of the deferral.
5134 16. If an IP address had so many PTR records that the DNS lookup buffer
5135 was not large enough to hold them, Exim could crash while trying to process
5136 the truncated data. It now detects and logs this case.
5138 17. Further to 4.21/58, another change has been made: if (and only if) the
5139 first line of a message (the first header line) ends with CRLF, a bare LF
5140 in a subsequent header line has a space inserted after it, so as not to
5141 terminate the header.
5143 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
5144 unnecessarily, used atoi() instead of strtol(), and didn't check the
5145 termination when getting file sizes from file names by regex.
5147 19. Completely re-implemented the support for maildirsize files, in the light
5148 of a number of problems with the previous contributed implementation
5149 (4.30/29). In particular:
5151 . If the quota is zero, the maildirsize file is maintained, but no quota is
5154 . If the maildir directory does not exist, it is created before any attempt
5155 to write a maildirsize file.
5157 . The quota value in the file is just a cache; if the quota is changed in
5158 the transport, the new value overrides.
5160 . A regular expression is available for excluding directories from the
5163 20. The autoreply transport checks the characters in options that define the
5164 message's headers; it allows continued headers, but it was checking with
5165 isspace() after an embedded newline instead of explicitly looking for a
5168 21. If all the "regular" hosts to which an address was routed had passed their
5169 expiry times, and had not reached their retry times, the address was
5170 bounced, even if fallback hosts were defined. Now Exim should go on to try
5173 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
5174 equivalent code in the SMTP transport. Some hosts send humungous responses
5175 to HELO/EHLO, more than 1024 it seems.
5177 23. Refactoring: code in filter.c used (void *) for "any old type" but this
5178 gives compiler warnings in some environments. I've now done it "properly",
5181 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
5182 (because of problems with the built-in one) was declared to return uschar *
5183 instead of char *, causing compiler failure.
5185 25. Fixed a file descriptor leak when processing alias/forward files.
5187 26. Fixed a minor format string issue in dbfn.c.
5189 27. Typo in exim.c: ("dmbnz" for "dbmnz").
5191 28. If a filter file refered to $h_xxx or $message_headers, and the headers
5192 contained RFC 2047 "words", Exim's memory could, under certain conditions,
5195 29. When a sender address is verified, it is cached, to save repeating the test
5196 when there is more than one recipient in a message. However, when the
5197 verification involves a callout, it is possible for different callout
5198 options to be set for different recipients. It is too complicated to keep
5199 track of this in the cache, so now Exim always runs a verification when a
5200 callout is required, relying on the callout cache for the optimization.
5201 The overhead is duplication of the address routing, but this should not be
5204 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
5205 to be verified with callout=postmaster, and the main callout worked but the
5206 postmaster check failed, the verification correctly failed. However, if a
5207 subsequent RCPT command asked for sender verification *without* the
5208 postmaster check, incorrect caching caused this verification also to fail,
5211 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
5212 it was not caching the DNS options (qualify_single, search_parents) that
5213 were used when the lookup failed. A subsequent lookup with different
5214 options therefore always gave the same answer, though there were cases
5215 where it should not have. (Example: a "domains = !$mx_any" option on a
5216 dnslookup router: the "domains" option is always processed without any
5217 widening, but the router might have qualify_single set.) Now Exim uses the
5218 cached value only when the same options are set.
5220 32. Added John Jetmore's "exipick" utility to the distribution.
5222 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
5223 than a timeout (e.g. a certificate is required, and is not provided), an
5224 Exim server now closes the connection immediately. Previously it waited for
5225 the client to close - but if the client is SSL, it seems that they each
5226 wait for each other, leading to a delay before one of them times out.
5228 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
5229 maintained 0.8.x compatibility because I don't think many are using it, and
5230 it is clearly obsolete.
5232 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
5235 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
5236 client certificate was expired. A simple patch fixes this, though I don't
5237 understand the full logic of why the verify callback is called multiple
5240 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
5241 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
5242 which causes problems with some clients (such as the Certicom SSL Plus
5243 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
5244 disables the coutermeasure allowing Eudora to connect."
5246 38. Exim was not checking that a write() to a log file succeeded. This could
5247 lead to Bad Things if a log got too big, in particular if it hit a file
5248 size limit. Exim now panics and dies if it cannot write to a log file, just
5249 as it does if it cannot open a log file.
5251 39. Modified OS/Makefile-Linux so that it now contains
5253 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
5255 The two -D definitions ensure that Exim is compiled with large file
5256 support, which makes it possible to handle log files that are bigger than
5259 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
5260 instance) a domain was checked against a named list that involved a lookup,
5261 causing $domain_data to be set, then another domain was checked against the
5262 same list, then the first domain was re-checked, the value of $domain_data
5263 after the final check could be wrong. In particular, if the second check
5264 failed, it could be set empty. This bug probably also applied to
5267 41. The strip_trailing_dot option was not being applied to the address given
5268 with the -f command-line option.
5270 42. The code for reading a message's header from the spool was incrementing
5271 $received_count, but never initializing it. This meant that the value was
5272 incorrect (doubled) while delivering a message in the same process in which
5273 it was received. In the most common configuration of Exim, this never
5274 happens - a fresh exec is done - but it can happen when
5275 deliver_drop_privilege is set.
5277 43. When Exim logs an SMTP synchronization error - client data sent too soon -
5278 it now includes up to 150 characters of the unexpected data in the log
5281 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
5282 and building data strings. The size of both of these buffers was 10 000
5283 bytes - far larger than anybody would *ever* want, thought I. Needless to
5284 say, somebody hit the limit. I have increased the maximum line length to
5285 20 000 and the maximum data length of concatenated lines to 100 000. I have
5286 also fixed two bugs, because there was no checking on these buffers. Tsk,
5287 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
5288 buffer is too small.
5290 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
5291 lsearch lookups. Now it does.
5293 46. When parsing a route_list item in a manualroute router, a fixed-length
5294 buffer was used for the list of hosts. I made this 1024 bytes long,
5295 thinking that nobody would ever have a list of hosts that long. Wrong.
5296 Somebody had a whole pile of complicated expansion conditions, and the
5297 string was silently truncated, leading to an expansion error. It turns out
5298 that it is easier to change to an unlimited length (owing to other changes
5299 that have happened since this code was originally written) than to build
5300 structure for giving a limitation error. The length of the item that
5301 expands into the list of hosts is now unlimited.
5303 47. The lsearch lookup could not handle data where the length of text line was
5304 more than 4095 characters. Such lines were truncated, leading to shortened
5305 data being returned. It should now handle lines of any length.
5307 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
5308 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
5311 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
5312 replacing strings such as BIN_DIRECTORY in a source file: the replacement
5313 no longer happens in comment lines. A list of replacements is now placed
5314 at the head of all of the source files, except those whose only change is
5315 to replace PERL_COMMAND in the very first #! line.
5317 50. Replaced the slow insertion sort in queue.c, for sorting the list of
5318 messages on the queue, with a bottom-up merge sort, using code contributed
5319 by Michael Haardt. This should make operations like -bp somewhat faster on
5320 large queues. It won't affect queue runners, except when queue_run_in_order
5323 51. Installed eximstats 1.31 in the distribution.
5325 52. Added support for SRV lookups to the dnslookup router.
5327 53. If an ACL referred to $message_body or $message_body_end, the value was not
5328 reset for any messages that followed in the same SMTP session.
5330 54. The store-handling optimization for building very long strings was not
5331 differentiating between the different store pools. I don't think this
5332 actually made any difference in practice, but I've tidied it.
5334 55. While running the routers to verify a sender address, $sender_address
5335 was still set to the sender address. This is wrong, because when routing to
5336 send a bounce to the sender, it would be empty. Therefore, I have changed
5337 it so that, while verifying a sender address, $sender_address is set to <>.
5338 (There is no change to what happens when verifying a recipient address.)
5340 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
5341 target A or AAAA records (if not already returned) without resetting the
5342 qualify_single or search_parents options of the DNS resolver. These are
5343 inappropriate in this case because the targets of MX and SRV records must
5344 be FQDNs. A broken DNS record could cause trouble if it happened to have a
5345 target that, when qualified, matched something in the local domain. These
5346 two options are now turned off when doing these lookups.
5348 57. It seems that at least some releases of Reiserfs (which does not have the
5349 concept of a fixed number of inodes) returns zero and not -1 for the
5350 number of available inodes. This interacted badly with check_spool_inodes,
5351 which assumed that -1 was the "no such thing" setting. What I have done is
5352 to check that the total number of inodes is greater than zero before doing
5353 the test of how many are available.
5355 58. When a "warn" ACL statement has a log_message modifier, the message is
5356 remembered, and not repeated. This is to avoid a lot of repetition when a
5357 message has many recipients that cause the same warning to be written.
5358 Howewer, Exim was preserving the list of already written lines for an
5359 entire SMTP session, which doesn't seem right. The memory is now reset if a
5360 new message is started.
5362 59. The "rewrite" debugging flag was not showing the result of rewriting in the
5363 debugging output unless log_rewrite was also set.
5365 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
5366 of (int)(handle) when we know that handle contains (void *)(-1).
5368 61. The Exim daemon panic-logs an error return when it closes the incoming
5369 connection. However "connection reset by peer" seems to be common, and
5370 isn't really an error worthy of noting specially, so that particular error
5373 62. When Exim is trying to find all the local interfaces, it used to panic and
5374 die if the ioctl to get the interface flags failed. However, it seems that
5375 on at least one OS (Solaris 9) it is possible to have an interface that is
5376 included in the list of interfaces, but for which you get a failure error
5377 for this call. This happens when the interface is not "plumbed" into a
5378 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
5379 failure of the "get flags" call assumes that the interface is down.
5381 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
5382 makes life easier for people who are doing arithmetic on fields extracted
5383 from dates, where you often get leading zeros that should not be
5384 interpreted as octal.
5386 64. Added qualify_domain to the redirect router, to override the global
5389 65. If a pathologically long header line contained very many addresses (the
5390 report of this problem mentioned 10 000) and each of them was rewritten,
5391 Exim could use up a very large amount of memory. (It kept on making new
5392 copies of the header line as it rewrote, and never released the old ones.)
5393 At the expense of a bit more processing, the header rewriting function has
5394 been changed so that it no longer eats memory in this way.
5396 66. The generation of the Received: header has been moved from the time that a
5397 message starts to be received, to the time that it finishes. The timestamp
5398 in the Received: header should now be very close to that of the <= log
5399 line. There are two side-effects of this change:
5401 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
5402 logged header lines no longer include the local Received: line, because
5403 it has not yet been created. The same applies to a copy of the message
5404 that is returned to a non-SMTP sender when a message is rejected.
5406 (b) When a filter file is tested using -bf, no additional Received: header
5407 is added to the test message. After some thought, I decided that this
5410 This change does not affect the value of $received_for. It is still set
5411 after address rewriting, but before local_scan() is called.
5413 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
5415 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
5416 gave an unhelpful panic error message, and a defer error. I have managed to
5417 change this behaviour so that it now rejects any supplied certificate,
5418 which seems right, as the list of acceptable certificates is empty.
5420 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
5421 gave an unhelpful defer error. I have not managed to make this reject any
5422 supplied certificates, but the error message it gives is "no certificate
5423 supplied", which is not helpful.
5425 70. exigrep's output now also includes lines that are not associated with any
5426 message, but which match the given pattern. Implemented by a patch from
5427 Martin Sluka, which also tidied up the Perl a bit.
5429 71. Recipient callout verification, like sender verification, was using <> in
5430 the MAIL FROM command. This isn't really the right thing, since the actual
5431 sender may affect whether the remote host accepts the recipient or not. I
5432 have changed it to use the actual sender in the callout; this means that
5433 the cache record is now keyed on a recipient/sender pair, not just the
5434 recipient address. There doesn't seem to be a real danger of callout loops,
5435 since a callout by the remote host to check the sender would use <>.
5436 [SEE ABOVE: changed after hitting problems.]
5438 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
5439 temporary errors. However, in the case of such a code being given after
5440 the end of a data transmission (i.e. after ".") Exim was failing to write
5441 a retry record for the message. (Yes, there was some broken host that was
5442 actually sending 8xx at this point.)
5444 73. An unknown lookup type in a host list could cause Exim to panic-die when
5445 the list was checked. (An example that provoked this was putting <; in the
5446 middle of a list instead of at the start.) If this happened during a DATA
5447 ACL check, a -D file could be left lying around. This kind of configuration
5448 error no longer causes Exim to die; instead it causes a defer errror. The
5449 incident is still logged to the main and panic logs.
5451 74. Buglet left over from Exim 3 conversion. The message "too many messages
5452 in one connection" was written to the rejectlog but not the mainlog, except
5453 when address rewriting (yes!) was being logged.
5455 75. Added write_rejectlog option.
5457 76. When a system filter was run not as root (that is, when system_filter_user
5458 was set), the values of the $n variables were not being returned to the
5459 main process; thus, they were not subsequently available in the $sn
5462 77. Added +return_path_on_delivery log selector.
5464 78. A connection timeout was being treated differently from recipients deferred
5465 when testing hosts_max_try with a message that was older than the host's
5466 retry timeout. (The host should not be counted, thus allowing all hosts to
5467 be tried at least once before bouncing.) This may have been the cause of an
5468 occasionally reported bug whereby a message would remain on the queue
5469 longer than the retry timeout, but would be bounced if a delivery was
5470 forced. I say "may" because I never totally pinned down the problem;
5471 setting up timeout/retry tests is difficult. See also the next item.
5473 79. The ultimate address timeout was not being applied to errors that involved
5474 a combination of host plus message (for example, a timeout on a MAIL
5475 command). When an address resolved to a number of possible hosts, and they
5476 were not all tried for each delivery (e.g. because of hosts_max_try), a
5477 message could remain on the queue longer than the retry timeout.
5479 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
5482 81. Fixed an obscure SMTP outgoing bug which required at least the following
5483 conditions: (a) there was another message waiting for the same server;
5484 (b) the server returned 5xx to all RCPT commands in the first message so
5485 that the message was not completed; (c) the server dropped the connection
5486 or gave a negative response to the RSET that Exim sends to abort the
5487 transaction. The observed case was a dropped connection after DATA that had
5488 been sent in pipelining mode. That is, the server had advertised PIPELINING
5489 but was not implementing it correctly. The effect of the bug was incorrect
5490 behaviour, such as trying another host, and this could lead to a crash.
5496 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
5497 and daemon.c were passed as pointers to ints; they should have been
5498 pointers to socklen_t variables (which are typically unsigned ints).
5500 2. Some signed/unsigned type warnings in the os.c file for Linux have been
5503 3. Fixed a really odd bug that affected only the testing scheme; patching a
5504 certain fixed string in the binary changed the value of another string that
5505 happened to be identical to the end of the original first string.
5507 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
5508 name", it returns that address as the IP address. On some operating
5509 systems (e.g. Solaris), it also passes back the IP address string as the
5510 "host name". However, on others (e.g. Linux), it passes back an empty
5511 string. Exim wasn't checking for this, and was changing the host name to an
5512 empty string, assuming it had been canonicized.
5514 5. Although rare, it is permitted to have more than one PTR record for a given
5515 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
5516 all the names associated with an address, because they do in Solaris.
5517 However, it seems that they do not in Linux for data that comes from the
5518 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
5519 I found this out when I moved to a new Linux workstation and tried to run
5520 the Exim test suite.
5522 To get round this problem I have changed the code so that it now does its
5523 own call to the DNS to look up PTR records when searching for a host name.
5524 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
5525 addresses that are only in /etc/hosts are still found.
5527 This behaviour is, however, controlled by an option called host_lookup_
5528 order, which defaults to "bydns:byaddr". If people want to use the other
5529 order, or indeed, just use one or the other means of lookup, they can
5530 specify it in this variable.
5532 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
5533 some operating systems, this comes back from gethostbyaddr() as an empty
5534 string, and this is what Exim used to test for. However, it seems that in
5535 other systems, "." is yielded. Exim now tests for this case too.
5537 7. The values of check_spool_space and check_log_space are now held internally
5538 as a number of kilobytes instead of an absolute number of bytes. If a
5539 numbers is specified without 'K' or 'M', it is rounded up to the nearest
5540 kilobyte. This means that much larger values can be stored.
5542 8. Exim monitor: an attempt to get the action menu when not actually pointing
5543 at a message produces an empty menu entitled "No message selected". This
5544 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
5545 no entries in it ("Shell widget menu has zero width and/or height"). So I
5546 have added a single, blank menu entry in this case.
5548 9. Added ${quote_local_part.
5550 10. MIME decoding is now applied to the contents of Subject: header lines when
5553 11. Now that a reference to $sender_host_address automatically causes a reverse
5554 lookup to occur if necessary (4.13/18), there is no need to arrange for a
5555 host lookup before query-style lookups in lists that might use this
5556 variable. This has therefore been abolished, and the "net-" prefix is no
5557 longer necessary for query-style lookups.
5559 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
5560 have been a typo for LFLAGS, so it has been changed.
5562 13. The install script calls Exim with "-C /dev/null" in order to find the
5563 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
5564 to be output. Howeve, since Exim outputs its version number before the
5565 error, it didn't break the script. It just looked ugly. I fixed this by
5566 always allowing "-C /dev/null" if the caller is root.
5568 14. Ignore overlarge ACL variable number when reading spool file - insurance
5569 against a later release with more variables having written the file.
5571 15. The standard form for an IPv6 address literal was being rejected by EHLO.
5572 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5573 this, as well as the form without the "IPv6" on the front.
5575 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
5576 OS/Makefile-Darwin file.
5578 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
5579 only to LDAP libraries that do not have LDAP_OPT_DEREF.
5581 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
5582 variable. However, some OS use ints for inodes. I've added cast to long int
5583 to get rid of the compiler warning.
5585 19. I had forgotten to lock out "/../" in configuration file names when
5586 ALT_CONFIG_PREFIX was set.
5588 20. Routers used for verification do not need to specify transports. However,
5589 if such a router generated a host list, and callout was configured, Exim
5590 crashed, because it could not find a port number from the (non-existent)
5591 transport. It now assumes port 25 in this circumstance.
5593 21. Added the -t option to exigrep.
5595 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
5596 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
5597 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
5598 them is set, LOOKUP_LSEARCH is forced.
5600 23. "exim -bV" now outputs a list of lookups that are included in the binary.
5602 24. Added sender and host information to the "rejected by local_scan()" log
5603 line; previously there was no indication of these.
5605 25. Added .include_if_exists.
5607 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
5608 Linux. It turns out that not all file systems support this. Apparently some
5609 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
5610 do it.) To cope with this, the error EINVAL, which means that sync-ing is
5611 not supported on the file descriptor, is now ignored when Exim is trying to
5612 sync a directory. This applies only to Linux.
5614 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
5616 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
5617 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
5620 29. Added support for maildirsize files from supplied patch (modified a bit).
5622 30. The use of :fail: followed by an empty string could lead Exim to respond to
5623 sender verification failures with (e.g.):
5625 550 Verification failed for <xxx>
5626 550 Sender verify failed
5628 where the first response line was missing the '-' that indicates it is not
5629 the final line of the response.
5631 31. The loop for finding the name of the user that called Exim had a hardwired
5632 limit of 10; it now uses the value of finduser_retries, which is used for
5633 all other user lookups.
5635 32. Added $received_count variable, available in data and not_smtp ACLs, and at
5638 33. Exim was neglecting to zero errno before one call of strtol() when
5639 expanding a string and expecting an integer value. On some systems this
5640 resulted in spurious "integer overflow" errors. Also, it was casting the
5641 result into an int without checking.
5643 34. Testing for a connection timeout using "timeout_connect" in the retry rules
5644 did not work. The code looks as if it has *never* worked, though it appears
5645 to have been documented since at least releast 1.62. I have made it work.
5647 35. The "timeout_DNS" error in retry rules, also documented since at least
5648 1.62, also never worked. As it isn't clear exactly what this means, and
5649 clearly it isn't a major issue, I have abolished the feature by treating it
5650 as "timeout", and writing a warning to the main and panic logs.
5652 36. The display of retry rules for -brt wasn't always showing the error code
5655 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
5656 timeout_connect_A, timeout_connect_MX.
5658 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
5659 to the empty sender.
5661 39. The daemon was not analysing the content of -oX till after it had closed
5662 stderr and disconnected from the controlling terminal. This meant that any
5663 syntax errors were only noted on the panic log, and the return code from
5664 the command was 0. By re-arranging the code a little, I've made the
5665 decoding happen first, so such errors now appear on stderr, and the return
5666 code is 1. However, the actual setting up of the sockets still happens in
5667 the disconnected process, so errors there are still only recorded on the
5670 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
5671 connections (as happens on some IP stacks) was logged at start up time as
5672 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
5673 it from "IPv6 and IPv4", which means that two separate sockets are being
5676 41. The debug output for gethostbyname2() or getipnodebyname() failures now
5677 says whether AF_INET or AF_INET6 was passed as an argument.
5679 42. Exiwhat output was messed up when time zones were included in log
5682 43. Exiwhat now gives more information about the daemon's listening ports,
5683 and whether -tls-on-connect was used.
5685 44. The "port" option of the smtp transport is now expanded.
5687 45. A "message" modifier in a "warn" statement in a non-message ACL was being
5688 silently ignored. Now an error message is written to the main and panic
5691 46. There's a new ACL modifier called "logwrite" which writes to a log file
5692 as soon as it is encountered.
5694 47. Added $local_user_uid and $local_user_gid at routing time.
5696 48. Exim crashed when trying to verify a sender address that was being
5699 49. Exim was recognizing only a space character after ".include". It now also
5700 recognizes a tab character.
5702 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
5703 extracting the relevant information from the specification. The man page no
5704 longer contains scrambled data for the -d option, and I've added a section
5705 at the front about calling Exim under different names.
5707 51. Added "extra_headers" argument to the "mail" command in filter files.
5709 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
5712 53. Installed eximstats 1.29.
5714 54. Added transport_filter_timeout as a generic transport option.
5716 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
5717 Cc: header lines. This was required by RFC 822, but it not required by RFC
5720 56. Exim used to add From:, Date:, and Message-Id: header lines to any
5721 incoming messages that did not have them. Now it does so only if the
5722 message originates locally, that is, if there is no associated remote host
5723 address. When Resent- header lines are present, this applies to the Resent-
5724 lines rather than the non-Resent- lines.
5726 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
5727 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
5729 58. Messages for configuration errors now include the name of the main
5730 configuration file - useful now that there may be more than one file in a
5731 list (.included file names were always shown).
5733 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
5734 for those rare installations that do not start the daemon as root or run it
5735 setuid root. I've cut out the call to initgroups() if the daemon is not
5738 60. The Exim user and group can now be bound into the binary as text strings
5739 that are looked up at the start of Exim's processing.
5741 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
5743 62. Added $mailstore_basename variable.
5745 63. Installed patch to sieve.c from Michael Haardt.
5747 64. When Exim failed to open the panic log after failing to open the main log,
5748 the original message it was trying to log was written to stderr and debug
5749 output, but if they were not available (the usual case in production), it
5750 was lost. Now it is written to syslog before the two lines that record the
5751 failures to open the logs.
5753 65. Users' Exim filters run in subprocesses under the user's uid. It is
5754 possible for a "deliver" command or an alias in a "personal" command to
5755 provoke an address rewrite. If logging of address rewriting is configured,
5756 this fails because the process is not running as root or exim. There may be
5757 a better way of dealing with this, but for the moment (because 4.30 needs
5758 to be released), I have disabled address rewrite logging when running a
5759 filter in a non-root, non-exim process.
5765 1. The buildconfig auxiliary program wasn't quoting the value set for
5766 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
5767 not defined. This bug was masked in 4.22 by the effect that was fixed in
5770 2. Some messages that were rejected after a message id was allocated were
5771 shown as "incomplete" by exigrep. It no longer does this for messages that
5772 are rejected by local_scan() or the DATA or non-SMTP ACLs.
5774 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
5775 have allow_domain_literals set, the ID did not get logged in the <= line.
5776 Domain literals are now always recognized in Message-ID: header lines.
5778 4. The first argument for a ${extract expansion item is the key name or field
5779 number. Leading and trailing spaces in this item were not being ignored,
5780 causing some misleading effects.
5782 5. When deliver_drop_privilege was set, single queue runner processes started
5783 manually (i.e. by the command "exim -q") or by the daemon (which uses the
5784 same command in the process it spins off) were not dropping privilege.
5786 6. When the daemon running as "exim" started a queue runner, it always
5787 re-executed Exim in the spun-off process. This is a waste of effort when
5788 deliver_drop_privilege is set. The new process now just calls the
5789 queue-runner function directly.
5795 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
5798 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
5799 ignored. Though the use of -oP was forcing the writing of a pid file, it
5800 was always written to the default place.
5802 3. If the message "no IP address found for host xxxx" is generated during
5803 incoming verification, it is now followed by identification of the incoming
5804 connection (so you can more easily find what provoked it).
5806 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
5808 5. Added some features to "harden" Exim a bit more against certain attacks:
5810 (a) There is now a build-time option called FIXED_NEVER_USERS that can
5811 be put in Local/Makefile. This is like the never_users runtime option,
5812 but it cannot be overridden. The default setting is "root".
5814 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
5815 prefix string with which any file named in a -C command line option
5818 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
5819 is retained for -C and -D only if the caller of Exim is root. Without
5820 it, the exim user may also use -C and -D and retain privilege.
5822 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
5823 command line option is disabled.
5825 6. Macro names set by the -D option must start with an upper case letter, just
5826 like macro names defined in the configuration file.
5828 7. Added "dereference=" facility to LDAP.
5830 8. Two instances of the typo "uknown" in the source files are fixed.
5832 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
5833 the Configure-Makefile script screwed up while processing it.
5835 10. Incorporated PCRE 4.4.
5837 11. The SMTP synchronization check was not operating right at the start of an
5838 SMTP session. For example, it could not catch a HELO sent before the client
5839 waited for the greeting. There is now a check for outstanding input at the
5840 point when the greeting is written. Because of the duplex, asynchronous
5841 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
5842 way, but not yet received, when the check is performed.
5844 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
5845 on TCP/IP sockets, because this apparently causes some broken clients to
5848 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
5849 unchanged) from the Cygwin maintainer.
5851 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
5852 was supported instead of testing for mbx. Effectively a typo.
5854 15. The spa authenticator server code was not checking that the input it
5855 received was valid base64.
5857 16. The debug output line for the "set" modifier in ACLs was not showing the
5858 name of the variable that was being set.
5860 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
5862 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
5863 lookup on its own. Something else was needed to trigger the lookup. For
5864 example, a match in host_lookup or the need for a host name in a host list.
5865 Now, if $sender_host_name is referenced and the host name has not yet been
5866 looked up, a lookup is performed. If the lookup fails, the variable remains
5867 empty, and $host_lookup_failed is set to "1".
5869 19. Added "eqi" as a case-independent comparison operator.
5871 20. The saslauthd authentication condition could segfault if neither service
5872 nor realm was specified.
5874 21. If an overflowing value such as "2048M" was set for message_size_limit, the
5875 error message that was logged was misleading, and incoming SMTP
5876 connections were dropped. The message is now more accurate, and temporary
5877 errors are given to SMTP connections.
5879 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
5880 (except RSET) with a 421 error, until QUIT is received. However, it was
5881 failing to send a response to QUIT.
5883 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
5884 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
5885 ACL is now run after the helo_try_verify_hosts code.
5887 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
5888 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
5889 case-independent, so other case variants are also recognized). Apparently
5890 some systems use these upper case variants.
5892 25. If more than two messages were waiting for the same host, and a transport
5893 filter was specified for the transport, Exim sent two messages over the
5894 same TCP/IP connection, and then failed with "socket operation on non-
5895 socket" when it tried to send the third.
5897 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
5899 27. The extern definition of crypt16() in expand.c was not being excluded when
5900 the OS had its own crypt16() function.
5902 28. Added bounce_return_body as a new option, and bounce_return_size_limit
5903 as a preferred synonym for return_size_limit, both as an option and as an
5906 29. Added LIBS=-liconv to OS/Makefile-OSF1.
5908 30. Changed the default configuration ACL to relax the local part checking rule
5909 for addresses that are not in any local domains. For these addresses,
5910 slashes and pipe symbols are allowed within local parts, but the sequence
5911 /../ is explicitly forbidden.
5913 31. SPA server authentication was not clearing the challenge buffer before
5916 32. log_message in a "warn" ACL statement was writing to the reject log as
5917 well as to the main log, which contradicts the documentation and doesn't
5918 seem right (because no rejection is happening). So I have stopped it.
5920 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
5921 However, I am unable to do any testing of this.
5923 34. Fixed an infelicity in the appendfile transport. When checking directories
5924 for a mailbox, to see if any needed to be created, it was accidentally
5925 using path names with one or more superfluous leading slashes; tracing
5926 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
5928 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
5929 contents are added to the log line that is written for every discarded
5930 recipient. (Previously a log_message setting was ignored.)
5932 36. The ${quote: operator now quotes the string if it is empty.
5934 37. The install script runs exim in order to find its version number. If for
5935 some reason other than non-existence or emptiness, which it checks, it
5936 could not run './exim', it was installing it with an empty version number,
5937 i.e. as "exim-". This error state is now caught, and the installation is
5940 38. An argument was missing from the function that creates an error message
5941 when Exim fails to connect to the socket for saslauthd authentication.
5942 This could cause Exim to crash, or give a corrupted message.
5944 39. Added isip, isip4, and isip6 to ${if conditions.
5946 40. The ACL variables $acl_xx are now saved with the message, and can be
5947 accessed later in routers, transports, and filters.
5949 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
5950 strings in the file are not string-expanded.
5952 42. If a MAIL command specified a SIZE value that was too large to fit into an
5953 int variable, the check against message_size_limit failed. Such values are
5954 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
5955 day this will have to be increased, but I don't think I want to be around
5956 when emails are that large.
5963 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
5964 iconv() is not standard in FreeBSD.
5966 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
5967 IPv6 enabled. The observed symptom was a segmentation fault on return from
5968 the function os_common_find_running_interfaces() in src/os.c.
5970 3. In the check_special_case() function in daemon.c I had used "errno" as an
5971 argument name, which causes warnings on some systems. This was basically a
5972 typo, since it was named "eno" in the comments!
5974 4. The code that waits for the clock to tick (at a resolution of some fraction
5975 of a second) so as to ensure message-id uniqueness was always waiting for
5976 at least one whole tick, when it could have waited for less. [This is
5977 almost certainly not relevant at current processor speeds, where it is
5978 unlikely to ever wait at all. But we try to future-proof.]
5980 5. The function that sleeps for a time interval that includes fractions of a
5981 second contained a race. It did not block SIGALRM between setting the
5982 timer, and suspending (a couple of lines later). If the interval was short
5983 and the sigsuspend() was delayed until after it had expired, the suspension
5984 never ended. On busy systems this could lead to processes getting stuck for
5987 6. Some uncommon configurations may cause a lookup to happen in a queue runner
5988 process, before it forks any delivery processes. The open lookup caching
5989 mechanism meant that the open file or database connection was passed into
5990 the delivery process. The problem was that delivery processes always tidy
5991 up cached lookup data. This could cause a problem for the next delivery
5992 process started by the queue runner, because the external queue runner
5993 process does not know about the closure. So the next delivery process
5994 still has data in the lookup cache. In the case of a file lookup, there was
5995 no problem because closing a file descriptor in a subprocess doesn't affect
5996 the parent. However, if the lookup was caching a connection to a database,
5997 the connection was closed, and the second delivery process was likely to
5998 see errors such as "PGSQL: query failed: server closed the connection
5999 unexpectedly". The problem has been fixed by closing all cached lookups
6000 in a queue runner before running a delivery process.
6002 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
6003 seem to have the "const" qualifier which it has on other OS. I've
6006 8. Change 4.21/2 was too strict. It is only if there are two authenticators
6007 *of the same type* (client or server) with the same public name that an
6008 error should be diagnosed.
6010 9. When Exim looked up a host name for an IP address, but failed to find the
6011 original IP address when looking up the host name (a safety check), it
6012 output the message "<ip address> does not match any IP for NULL", which was
6013 confusing, to say the least. The bug was that the host name should have
6014 appeared instead of "NULL".
6016 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
6017 user that is built into the binary, and the -C or -D options is used, root
6018 privilege is dropped before the configuration file is read. In addition,
6019 logging is switched to stderr instead of the normal log files. If the
6020 configuration then re-defines the Exim user, the unprivileged environment
6021 is probably not what is expected, so Exim logs a panic warning message (but
6024 However, if deliver_drop_privilege is set, the unprivileged state may well
6025 be exactly what is intended, so the warning has been cut out in that case,
6026 and Exim is allowed to try to write to its normal log files.
6032 1. smtp_return_error_details was not giving details for temporary sender
6033 or receiver verification errors.
6035 2. Diagnose a configuration error if two authenticators have the same public
6038 3. Exim used not to create the message log file for a message until the first
6039 delivery attempt. This could be confusing when incoming messages were held
6040 for policy or load reasons. The message log file is now created at the time
6041 the message is received, and an initial "Received" line is written to it.
6043 4. The automatically generated man page for command line options had a minor
6044 bug that caused no ill effects; however, a more serious problem was that
6045 the procedure for building the man page automatically didn't always
6046 operate. Consequently, release 4.20 contains an out-of-date version. This
6047 shouldn't happen again.
6049 5. When building Exim with embedded Perl support, the script that builds the
6050 Makefile was calling 'perl' to find its compile-time parameters, ignoring
6051 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
6053 6. The freeze_tell option was not being used for messages that were frozen on
6054 arrival, either by an ACL or by local_scan().
6056 7. Added the smtp_incomplete_transaction log selector.
6058 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
6059 was accepting AUTH without a new EHLO.
6061 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
6062 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
6065 10. Logging of TCP/IP connections (when configured) now happens in the main
6066 daemon process instead of the child process, so that the TCP/IP connection
6067 count is more accurate (but it can never be perfect).
6069 11. The use of "drop" in a nested ACL was not being handled correctly in the
6070 outer ACL. Now, if condition failure induced by the nested "drop" causes
6071 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
6072 or "require"), the connection is dropped.
6074 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
6075 that yield "discard" can now be used with an "accept" or a "discard" verb,
6076 but an error is generated for any others (because I can't see a useful way
6077 to define what should happen).
6079 13. When an ACL is read dynamically from a file (or anywhere else), the lines
6080 are now processed in the same way as lines in the Exim configuration file.
6081 In particular, continuation lines are supported.
6083 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
6085 15. Added -ti meaning -t -i.
6087 16. Check for letters, digits, hyphens, and dots in the names of dnslist
6088 domains, and warn by logging if others are found.
6090 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
6091 returned from GIFCONF when Exim is trying to find the list of interfaces on
6092 a host. The code in os.c has been modified to copy each ifreq to an aligned
6093 structure in all cases.
6095 Also, in some cases, the returned ifreq's were being copied to a 'struct
6096 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
6097 means the last couple of bytes of an IPv6 address could be chopped if the
6098 ifreq contained only a normal sockaddr (14 bytes storage).
6100 18. Named domain lists were not supported in the hosts_treat_as_local option.
6101 An entry such as +xxxx was not recognized, and was treated as a literal
6104 19. Ensure that header lines added by a DATA ACL are included in the reject log
6105 if the ACL subsequently rejects the message.
6107 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
6108 MD5 (which is deprecated).
6110 21. When testing a filter file using -bf, Exim was writing a message when it
6111 took the sender from a "From " line in the message, but it was not doing so
6112 when it took $return_path from a Return-Path: header line. It now does.
6114 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
6115 with a valid header line field name (a series of printing characters
6116 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
6118 23. Changed "disc" in the source to "disk" to conform to the documentation and
6119 the book and for uniformity.
6121 24. Ignore Sendmail's -Ooption=value command line item.
6123 25. When execve() failed while trying to run a command in a pipe transport,
6124 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
6125 could be confused with a return value of 69 from the command itself. This
6126 has been changed to 127, the value the shell returns if it is asked to run
6127 a non-existent command. The wording for the related log line suggests a
6128 non-existent command as the problem.
6130 26. If received_header_text expands to an empty string, do not add a Received:
6131 header line to the message. (Well, it adds a token one on the spool, but
6132 marks it "old" so that it doesn't get used or transmitted.)
6134 27. Installed eximstats 1.28 (addition of -nt option).
6136 28. There was no check for failure on the call to getsockname() in the daemon
6137 code. This can fail if there is a shortage of resources on the system, with
6138 ENOMEM, for example. A temporary error is now given on failure.
6140 29. Contrary to the C standard, it seems that in some environments, the
6141 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
6142 program. Exim now does this explicitly; it affects the formatting of
6143 timestamps using strftime().
6145 30. If exiqsumm was given junk data, it threw up some uninitialized variable
6146 complaints. I've now initialized all the variables, to avoid this.
6148 32. Header lines added by a system filter were not being "seen" during
6149 transport-time rewrites.
6151 33. The info_callback() function passed to OpenSSL is set up with type void
6152 (*)(SSL *, int, int), as described somewhere. However, when calling the
6153 function (actually a macro) that sets it up, the type void(*)() is
6154 expected. I've put in a cast to prevent warnings from picky compilers.
6156 34. If a DNS black list lookup found a CNAME record, but there were no A
6157 records associated with the domain it pointed at, Exim crashed.
6159 35. If a DNS black list lookup returned more than one A record, Exim ignored
6160 all but the first. It now scans all returned addresses if a particular IP
6161 value is being sought. In this situation, the contents of the
6162 $dnslist_value variable are a list of all the addresses, separated by a
6165 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
6166 to accept a host name and all its aliases if the forward lookup for any of
6167 them yielded the IP address of the incoming connection. Now it accepts only
6168 those names whose forward lookup yields the correct IP address. Any other
6169 names are discarded. This closes a loophole whereby a rogue DNS
6170 administrator could create reverse DNS records to break through a
6171 wildcarded host restriction in an ACL.
6173 37. If a user filter or a system filter that ran in a subprocess used any of
6174 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
6175 the wrong values were passed to the pipe command ($thisaddress had the
6176 value of $0, $0 had the value of $1, etc). This bug was introduced by
6177 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
6179 38. Improved the line breaking for long SMTP error messages from ACLs.
6180 Previously, if there was no break point between 40 and 75 characters, Exim
6181 left the rest of the message alone. Two changes have been made: (a) I've
6182 reduced the minimum length to 35 characters; (b) if it can't find a break
6183 point between 35 and 75 characters, it looks ahead and uses the first one
6184 that it finds. This may give the occasional overlong line, but at least the
6185 remaining text gets split now.
6187 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
6188 file descriptors might be low, and that setting 1000 would always raise it.
6189 It turns out that in some environments, the limit is already over 1000 and
6190 that lowering it causes trouble. So now Exim takes care not to decrease it.
6192 40. When delivering a message, the value of $return_path is set to $sender_
6193 address at the start of routing (routers may change the value). By an
6194 oversight, this default was not being set up when an address was tested by
6195 -bt or -bv, which affected the outcome if any router or filter referred to
6198 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
6199 log only when "message" or "log_message" are set. However, if one of the
6200 conditions was an address verification, or a call to a nested ACL, the
6201 messages generated by the underlying test were being passed through. This
6202 no longer happens. The underlying message is available in $acl_verify_
6203 message for both "message" and "log_message" expansions, so it can be
6204 passed through if needed.
6206 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
6207 new expansion $bh_ to give the encoded byte string without charset
6208 translation. Translation happens only if iconv() is available; HAVE_ICONV
6209 indicates this at build time. HEADERS_CHARSET gives the charset to
6210 translate to; headers_charset can change it in the configuration, and
6211 "headers charset" can change it in an individual filter file.
6213 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
6214 that creates RFC 2047 encoded "words" labels them as that charset instead
6215 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
6216 expansion operator; (ii) when Exim creates a From: line for a local
6217 message; (iii) when a header line is rewritten to include a "phrase" part.
6219 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
6220 buggy, causing it to skip the first lines of messages whose message ID
6221 ended in 'D'. This would not have bitten before Exim release 4.14, because
6222 message IDs were unlikely to end in 'D' before then. The effect was to have
6223 incorrect size information for certain domains.
6225 45. #include "config.h" was missing at the start of the crypt16.c module. This
6226 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
6229 46. If there was a timeout during a "random" callout check, Exim treated it as
6230 a failure of the random address, and carried on sending RSET and the real
6231 address. If the delay was just some slowness somewhere, the response to the
6232 original RCPT would be taken as a response to RSET and so on, causing
6233 mayhem of various kinds.
6235 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
6236 when I implemented it. It didn't allow for the fact that some option values
6237 may legitimatetly be negative (e.g. size_addition), and it didn't even do
6238 the right test for positive values.
6240 48. Domain names in DNS records are case-independent. Exim always looks them up
6241 in lower case. Some resolvers return domain names in exactly the case they
6242 appear in the zone file, that is, they may contain uppercase letters. Not
6243 all resolvers do this - some return always lower case. Exim was treating a
6244 change of case by a resolver as a change of domain, similar to a widening
6245 of a domain abbreviation. This triggered its re-routing code and so it was
6246 trying to route what was effectively the same domain again. This normally
6247 caused routing to fail (because the router wouldn't handle the domain
6248 twice). Now Exim checks for this case specially, and just changes the
6249 casing of the domain that it ultimately uses when it transmits the message
6252 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
6255 50. If a filter generated a file delivery with a non-absolute name (possible if
6256 no home directory exists for the router), the forbid_file option was not
6259 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
6260 the existing equality matching.
6262 52. Exim was using ints instead of ino_t variables in some places where it was
6263 dealing with inode numbers.
6265 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
6266 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
6267 called TMPDIR, and if it finds it is different, it changes its value.
6269 54. The smtp_printf() function is now made available to local_scan() so
6270 additional output lines can be written before returning. There is also an
6271 smtp_fflush() function to enable the detection of a dropped connection.
6272 The variables smtp_input and smtp_batched_input are exported to
6275 55. Changed the default runtime configuration: the message "Unknown user"
6276 has been removed from the ACL, and instead placed on the localuser router,
6277 using the cannot_route_message feature. This means that any verification
6278 failures that generate their own messages won't get overridden. Similarly,
6279 the "Unrouteable address" message that was in the ACL for unverifiable
6280 relay addresses has also been removed.
6282 56. Added hosts_avoid_esmtp to the smtp transport.
6284 57. The exicyclog script was not checking for the esoteric option
6285 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
6286 will work only if exicyclog is run under the appropriate euid.
6288 58. Following a discussion on the list, the rules by which Exim recognises line
6289 endings on incoming messages have been changed. The -dropcr and drop_cr
6290 options are now no-ops, retained only for backwards compatibility. The
6291 following line terminators are recognized: LF CRLF CR. However, special
6292 processing applies to CR:
6294 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
6295 nor a local message in the state where . is a terminator.
6297 (ii) If a bare CR is encountered in a header line, an extra space is added
6298 after the line terminator so as not to end the header. The reasoning
6299 behind this is that bare CRs in header lines are most likely either
6300 to be mistakes, or people trying to play silly games.
6302 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
6303 was being incorrectly given as 18 bytes larger than it should have been.
6304 This is a VOB (very old bug).
6306 60. This may never have affected anything current, but just in case it has:
6307 When the local host is found other than at the start of a list of hosts,
6308 the local host, those with the same MX, and any that follow, are discarded.
6309 When the list in question was part of a longer list of hosts, the following
6310 hosts (not currently being processed) were also being discarded. This no
6311 longer happens. I'm not sure if this situation could ever has previously
6314 61. Added the "/MX" feature to lists of hosts in the manualroute and query
6317 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
6318 header. This is something that is recommended in a new Internet Draft, and
6319 is something that is documented as being done by Sendmail. There are two
6320 possible values. For messages generated by the autoreply transport, Exim
6323 Auto-Submitted: auto-replied
6325 whereas for all other generated messages (e.g. bounces) it adds
6327 Auto-Submitted: auto-generated
6329 63. The "personal" condition in filters now includes a test for the
6330 Auto-Submitted: header. If it contains the string "auto-" the message it
6331 not considered personal.
6333 64. Added rcpt_include_affixes as a generic transport option.
6335 65. Added queue_only_override (default true).
6337 66. Added the syslog_duplication option.
6339 67. If what should have been the first header line of a message consisted of
6340 a space followed by a colon, Exim was mis-interpreting it as a header line.
6341 It isn't of course - it is syntactically invalid and should therefore be
6342 treated as the start of the message body. The misbehaviour could have
6343 caused a number of strange effects, including loss of data in subsequent
6344 header lines, and spool format errors.
6346 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
6347 client host had authenticated. This control can now be exercised by an ACL
6348 for more flexibility.
6350 69. By default, callouts do not happen when testing with -bh. There is now a
6351 variant, -bhc, which does actually run the callout code, including
6352 consulting and updating the callout cache.
6354 70. Added support for saslauthd authentication, courtesy of Alexander
6357 71. If statvfs() failed on the spool or log directories while checking their
6358 size for availability, Exim confusingly gave the error "space shortage".
6359 Furthermore, in debugging mode it crashed with a floating point exception.
6360 These checks are done if check_{spool,log}_{space,inodes} are set, and when
6361 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
6362 serious problem, Exim now writes to the main and panic logs when this
6363 happens, with details of the failure. It then refuses to accept the
6364 incoming message, giving the message "spool directory problem" or "log
6365 directory problem" with a 421 code for SMTP messages.
6367 72. When Exim is about to re-exec itself, it ensures that the file descriptors
6368 0, 1, and 2 exist, because some OS complain for execs without them (see
6369 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
6370 descriptors. However, the code omitted to check that the open succeeded,
6371 causing mysterious errors if for some reason the permissions on /dev/null
6372 got screwed. Now Exim writes a message to the main and panic logs, and
6373 bombs out if it can't open /dev/null.
6375 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
6376 interact so that it is all more flexible. It is supposed to remain
6377 backwards compatible. Also added extra_local_interfaces.
6379 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
6380 to bomb out with an assertion failure - to the client this appears as a
6381 connection drop. This problem occurs in the part of the code that was taken
6382 from the Samba project. Fortunately, the assertion is in a very simple
6383 function, so I have fixed this by reproducing the function inline in the
6384 one place where it is called, and arranging for authentication to fail
6385 instead of killing the process with assert().
6387 75. The SPA client code was not working when the server requested OEM rather
6388 than Unicode encoding.
6390 76. Added code to make require_files with a specific uid setting more usable in
6391 the case where statting the file as root fails - usually a non-root-mounted
6392 NFS file system. When this happens and the failure is EACCES, Exim now
6393 forks a subprocess and does the per-uid checking as the relevant uid.
6395 77. Added process_log_path.
6397 78. If log_file_path was not explicitly set, a setting of check_log_space or
6398 check_log_inodes was ignored.
6400 79. If a space check for the spool or log partitions fails, the incident is now
6401 logged. Of course, in the latter case the data may get lost...
6403 80. Added the %p formatting code to string_format() so that it can be used to
6404 print addresses in debug_print(). Adjusted all the address printing in the
6405 debugging in store.c to use %p rather than %d.
6407 81. There was a concern that a line of code in smtp_in.c could overflow a
6408 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
6409 initially expressed, the concern was not well-founded, because trailing
6410 spaces are removed early. However, if the trailing spaces were followed by
6411 a NULL, they did not get removed, so the overflow was possible. Two fixes
6414 (a) I re-wrote the offending code in a cleaner fashion.
6415 (b) If an incoming SMTP command contains a NULL character, it is rejected
6418 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
6419 runs initgroups(), so that if the Exim user is in any additional groups,
6420 they will be used during message reception.
6426 The change log for 4.20 and earlier releases has been archived.
projects / exim.git / blob