1 Change log file for Exim from version 4.21
2 -------------------------------------------
7 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
8 and 3.4.4 - once the server is enabled to respond to an OCSP request
9 it does even when not requested, resulting in a stapling non-aware
10 client dropping the TLS connection.
12 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
13 support variable-length bit vectors. No functional change.
15 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
16 The I= interface field on outgoing lines is now after the H= remote
17 host field, same as incoming lines. There is a separate
18 outgoing_interface log selector which allows you to disable the
21 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
22 If not running log_selector +smtp_connection the mainlog would be held
23 open indefinitely after a "too many connections" event, including to a
24 deleted file after a log rotate. Leave the per net connection logging
25 leaving it open for efficiency as that will be quickly detected by the
26 check on the next write.
28 HS/01 Bug 1671: Fix post transport crash.
29 Processing the wait-<transport> messages could crash the delivery
30 process if the message IDs didn't exist for some reason. When
31 using 'split_spool_directory=yes' the construction of the spool
32 file name failed already, exposing the same netto behaviour.
34 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
35 mime_regex ACL conditions.
37 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
38 to DSN fail messages (bounces): remote IP, remote greeting, remote response
39 to HELO, local diagnostic string.
41 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
42 log line to debug. Even when configured with a tls authenticator many
43 client connections are expected to not authenticate in this way, so
44 an authenticate fail is not an error.
46 HS/02 Add the Exim version string to the process info. This way exiwhat
47 gives some more detail about the running daemon.
49 JH/06 Bug 1395: time-limit cacheing of DNS lookups, to the TTL value. This may
50 matter for fast-change records such as DNSBLs.
52 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
53 retry record, even if constant. There may be multiple transports with
54 different interface settings and the retry behaviour needs to be kept
57 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
59 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
61 JH/10 Bug 840: fix log_defer_output option of pipe transport
63 JH/11 Bug 830: use same host for all RCPTS of a message, even under
64 hosts_randomize. This matters a lot when combined with mua_wrapper.
66 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
67 ${quote_pgsql:<string>} operator.
69 JH/13 Bug 1708: avoid misaligned access in cached lookup.
71 JH/14 Change header file name for freeradius-client. Relevant if compiling
72 with Radius support; from the Gentoo tree and checked under Fedora.
74 JH/15 Bug 1712: Introduce $prdr_requested flag variable
76 JH/16 Bug 1714: Permit an empty string as expansion result for transport
77 option transport_filter, meaning no filtering.
79 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
81 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
82 defaults to "*" (all hosts). The variable is now available when not built
83 with TLS, default unset, mainly to enable keeping the testuite sane.
84 If a server certificate is not supplied (via tls_certificate) an error is
85 logged, and clients will find TLS connections fail on startup. Presumably
86 they will retry in-clear.
87 Packagers of Exim are strongly encouraged to create a server certificate
90 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
91 with the $config_file variable.
93 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
94 in transport context, after the attempt, and per-recipient. The latter type
95 is per host attempted. The event data is the error message, and the errno
96 information encodes the lookup type (A vs. MX) used for the (first) host,
97 and the trailing two digits of the smtp 4xx reponse.
99 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
100 to write to mainlog (or rejectlog, paniclog) in the window between file
101 creation and permissions/ownership being changed. Particularly affects
102 installations where exicyclog is run as root, rather than exim user;
103 result is that the running daemon panics and dies.
105 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
107 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
108 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
109 "pri" and "weight". Note that the previous implicit priority given by the
110 list order is no longer honoured.
112 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalisation
115 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
116 by defining SUPPORT_SOCKS.
118 JH/26 Move PROXY support from Experimental to mainline, enabled for a build
119 by defining SUPPORT_PROXY. Note that the proxy_required_hosts option
120 is renamed to hosts_proxy, and the proxy_{host,target}_{address,port}.
121 variables are renamed to proxy_{local,external}_{address,port}.
123 JH/27 Move Internationalisation support from Experimental to mainline, enabled
124 for a build by defining SUPPORT_I18N
126 JH/28 Bug 1745: Fix redis lookups to handle (quoted) spaces embedded in parts
127 of the query string, and make ${quote_redis:} do that quoting.
132 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
135 JH/02 The smtp transport option "multi_domain" is now expanded.
137 JH/03 The smtp transport now requests PRDR by default, if the server offers
140 JH/04 Certificate name checking on server certificates, when exim is a client,
141 is now done by default. The transport option tls_verify_cert_hostnames
142 can be used to disable this per-host. The build option
143 EXPERIMENTAL_CERTNAMES is withdrawn.
145 JH/05 The value of the tls_verify_certificates smtp transport and main options
146 default to the word "system" to access the system default CA bundle.
147 For GnuTLS, only version 3.0.20 or later.
149 JH/06 Verification of the server certificate for a TLS connection is now tried
150 (but not required) by default. The verification status is now logged by
151 default, for both outbound TLS and client-certificate supplying inbound
154 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
157 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
158 Status Notification (bounce) messages are now MIME format per RFC 3464.
159 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
160 under the control of the dsn_advertise_hosts option, and routers may
161 have a dsn_lasthop option.
163 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
164 default, modifiable by a malware= option. The list separator for
165 the options can now be changed in the usual way. Bug 68.
167 JH/10 The smtp_receive_timeout main option is now expanded before use.
169 JH/11 The incoming_interface log option now also enables logging of the
170 local interface on delivery outgoing connections.
172 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
173 if the interface and destination host and port all match.
175 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
178 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
179 Patch from Andrew Lewis.
181 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
182 now supports optional time-restrictions, weighting, and priority
183 modifiers per server. Patch originally by <rommer@active.by>.
185 JH/16 The spamd_address main option now supports a mixed list of local
186 and remote servers. Remote servers can be IPv6 addresses, and
187 specify a port-range.
189 JH/17 Bug 68: The spamd_address main option now supports an optional
190 timeout value per server.
192 JH/18 Bug 1581: Router and transport options headers_add/remove can
193 now have the list separator specified.
195 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
198 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
201 JH/21 Support for the A6 type of dns record is withdrawn.
203 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
204 rather than the verbs used.
206 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
207 from 255 to 1024 chars.
209 JH/24 Verification callouts now attempt to use TLS by default.
211 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
212 are generic router options now. The defaults didn't change.
214 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
215 Original patch from Alexander Shikoff, worked over by JH.
217 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
218 files if ZCAT_COMMAND is not executable.
220 JH/26 Bug 1539: Add timout/retry options on dnsdb lookups.
222 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
224 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
225 Normally benign, it bites when the pair was led to by a CNAME;
226 modern usage is to not canoicalize the domain to a CNAME target
227 (and we were inconsistent anyway for A-only vs AAAA+A).
229 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
231 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
232 when evaluating $sender_host_dnssec.
234 JH/31 Check the HELO verification lookup for DNSSEC, adding new
235 $sender_helo_dnssec variable.
237 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
239 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
241 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
243 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
244 documented as working, but never had. Support all but $spam_report.
246 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
247 added for tls authenticator.
252 TL/01 When running the test suite, the README says that variables such as
253 no_msglog_check are global and can be placed anywhere in a specific
254 test's script, however it was observed that placement needed to be near
255 the beginning for it to behave that way. Changed the runtest perl
256 script to read through the entire script once to detect and set these
257 variables, reset to the beginning of the script, and then run through
258 the script parsing/test process like normal.
260 TL/02 The BSD's have an arc4random API. One of the functions to induce
261 adding randomness was arc4random_stir(), but it has been removed in
262 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
263 function when detected.
265 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
266 cause callback expansion.
268 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
269 syntax errors in an expansion can be treated as a string instead of
270 logging or causing an error, due to the internal use of bool_lax
271 instead of bool when processing it.
273 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
274 server certificates when making smtp deliveries.
276 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
278 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
280 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
282 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
283 Merged patch from Sebastian Wiedenroth.
285 JH/05 Fix results-pipe from transport process. Several recipients, combined
286 with certificate use, exposed issues where response data items split
287 over buffer boundaries were not parsed properly. This eventually
288 resulted in duplicates being sent. This issue only became common enough
289 to notice due to the introduction of conection certificate information,
290 the item size being so much larger. Found and fixed by Wolfgang Breyha.
292 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
293 size buffer was used, resulting in syntax errors when an expansion
296 JH/07 Add support for directories of certificates when compiled with a GnuTLS
297 version 3.3.6 or later.
299 JH/08 Rename the TPDA expermimental facility to Event Actions. The #ifdef
300 is EXPERIMENTAL_EVENT, the main-configuration and transport options
301 both become "event_action", the variables become $event_name, $event_data
302 and $event_defer_errno. There is a new variable $verify_mode, usable in
303 routers, transports and related events. The tls:cert event is now also
304 raised for inbound connections, if the main configuration event_action
307 TL/06 In test suite, disable OCSP for old versions of openssl which contained
308 early OCSP support, but no stapling (appears to be less than 1.0.0).
310 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
311 server certificate names available under the smtp transport option
312 "tls_verify_cert_hostname" now do not permit multi-component wildcard
315 JH/10 Time-related extraction expansions from certificates now use the main
316 option "timezone" setting for output formatting, and are consistent
317 between OpenSSL and GnuTLS compilations. Bug 1541.
319 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
320 encoded parameter in the incoming message. Bug 1558.
322 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
323 include certificate info, eximon was claiming there were spoolfile
326 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
328 JH/14 Log delivery-related information more consistently, using the sequence
329 "H=<name> [<ip>]" wherever possible.
331 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
332 are problematic for Debian distribution, omit them from the release
335 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
337 JH/16 Fix string representation of time values on 64bit time_t anchitectures.
340 JH/17 Fix a null-indirection in certextract expansions when a nondefault
341 output list separator was used.
346 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
347 checkers that were complaining about end of non-void function with no
350 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
351 This was a regression intruduced in 4.83 by another bugfix.
353 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
355 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
356 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
362 TF/01 Correctly close the server side of TLS when forking for delivery.
364 When a message was received over SMTP with TLS, Exim failed to clear up
365 the incoming connection properly after forking off the child process to
366 deliver the message. In some situations the subsequent outgoing
367 delivery connection happened to have the same fd number as the incoming
368 connection previously had. Exim would try to use TLS and fail, logging
369 a "Bad file descriptor" error.
371 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
372 utilities have not been installed.
374 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
375 temporary space as the ACL may create new global variables.
377 TL/01 LDAP support uses per connection or global context settings, depending
378 upon the detected version of the libraries at build time.
380 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
381 to extract and use the src ip:port in logging and expansions as if it
382 were a direct connection from the outside internet. PPv2 support was
383 updated based on HAProxy spec change in May 2014.
385 JH/02 Add ${listextract {number}{list}{success}{fail}}.
387 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
388 Properly escape header and check for NULL return.
390 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
393 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
395 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
396 characters in header names, implemented as a verify condition.
397 Contributed by Michael Fischer v. Mollard.
399 TL/05 Rename SPF condition results err_perm and err_temp to standardized
400 results permerror and temperror. Previous values are deprecated but
401 still accepted. In a future release, err_perm and err_temp will be
402 completely removed, which will be a backward incompatibility if the
403 ACL tests for either of these two old results. Patch contributed by
404 user bes-internal on the mailing list.
406 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
408 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
409 selectors, in both main and reject logs.
411 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
414 JH/07 Add malware type "sock" for talking to simple daemon.
416 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
418 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
419 routers/transports under cutthrough routing.
421 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
422 numbers. Touch up "bool" conditional to keep the same definition.
424 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
426 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
428 JH/12 Expand items in router/transport headers_add or headers_remove lists
429 individually rather than the list as a whole. Bug 1452.
431 Required for reasonable handling of multiple headers_ options when
432 they may be empty; requires that headers_remove items with embedded
433 colons must have them doubled (or the list-separator changed).
435 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
436 view the policy declared in the DMARC record. Currently, $dmarc_status
437 is a combined value of both the record presence and the result of the
440 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
442 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
443 dnslookup router and the smtp transport (applying to the forward
446 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
447 of ldap servers used for a specific lookup. Patch provided by Heiko
450 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
451 New variable $lookup_dnssec_authenticated for observability.
453 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
454 Patch submitted by Lars Timman.
456 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
458 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
459 Requires trusted mode and valid format message id, aborts otherwise.
460 Patch contributed by Heiko Schlichting.
462 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
463 certextract with support for various fields. Bug 1358.
465 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
466 is requested by default, modifiable by smtp transport option
469 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
470 operate on certificate variables to give certificate fingerprints
471 Also new ${sha256:cert_variable}.
473 JH/23 The PRDR feature is moved from being Experimental into the mainline.
475 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
476 Christian Aistleitner.
478 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
480 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
481 file. Patch from Wolfgang Breyha.
483 JH/25 Expand the coverage of the delivery $host and $host_address to
484 client authenticators run in verify callout. Bug 1476.
486 JH/26 Port service names are now accepted for tls_on_connect_ports, to
487 align with daemon_smtp_ports. Bug 72.
489 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
490 support and error reporting did not work properly.
492 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
493 and is readable. Patch from Andrew Colin Kissa.
495 TL/14 Enhance documentation of ${run expansion and how it parses the
496 commandline after expansion, particularly in the case when an
497 unquoted variable expansion results in an empty value.
499 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
501 PP/02 Fix internal collision of T_APL on systems which support RFC3123
502 by renaming away from it. Addresses GH issue 15, reported by
505 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
507 TL/15 SECURITY: prevent double expansion in math comparison functions
508 (can expand unsanitized data). Not remotely exploitable.
515 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
517 PP/02 Make -n do something, by making it not do something.
518 When combined with -bP, the name of an option is not output.
520 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
523 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
524 $sender_host_name and config options to manage this, and basic check
527 PP/05 DSCP support for outbound connections and control modifier for inbound.
529 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
530 (Only plugin which currently uses this is kerberos4, which nobody should
531 be using, but we should make it available and other future plugins might
532 conceivably use it, even though it would break NAT; stuff *should* be
533 using channel bindings instead).
535 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
536 name; added for Sendmail compatibility; requires admin caller.
537 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
538 just ignore it); requires trusted caller.
539 Also parse but ignore: -Ac -Am -X<logfile>
542 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
544 TL/02 Add +smtp_confirmation as a default logging option.
546 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
547 Patch by Magnus Holmgren from 2007-02-20.
549 TL/04 Bugzilla 1281 - Spec typo.
550 Bugzilla 1283 - Spec typo.
551 Bugzilla 1290 - Spec grammar fixes.
553 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
555 TL/06 Add Experimental DMARC support using libopendmarc libraries.
557 TL/07 Fix an out of order global option causing a segfault. Reported to dev
558 mailing list by by Dmitry Isaikin.
560 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
562 JH/02 Support "G" suffix to numbers in ${if comparisons.
564 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
566 NM/01 Bugzilla 1197 - Spec typo
567 Bugzilla 1196 - Spec examples corrections
569 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
571 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
572 gnutls_enable_pkcs11, but renamed to more accurately indicate its
575 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
576 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
578 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
579 "acl {{name}{arg}...}", and optional args on acl condition
582 JH/05 Permit multiple router/transport headers_add/remove lines.
584 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
586 JH/07 Avoid using a waiting database for a single-message-only transport.
587 Performance patch from Paul Fisher. Bugzilla 1262.
589 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
592 JH/09 Add $headers_added variable, with content from use of ACL modifier
593 add_header (but not yet added to the message). Bugzilla 199.
595 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
596 Pulled from Bugzilla 817 by Wolfgang Breyha.
598 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
600 (nb: this is the same fix as in Exim 4.80.1)
602 JH/11 Add A= logging on delivery lines, and a client_set_id option on
605 JH/12 Add optional authenticated_sender logging to A= and a log_selector
608 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
610 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
611 advertise SMTP AUTH mechanism to us, instead of a generic
612 protocol violation error. Also, make Exim more robust to bad
613 data from the Dovecot auth socket.
615 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
617 When a queue runner is handling a message, Exim first routes the
618 recipient addresses, during which it prunes them based on the retry
619 hints database. After that it attempts to deliver the message to
620 any remaining recipients. It then updates the hints database using
623 So if a recipient address works intermittently, it can get repeatedly
624 deferred at routing time. The retry hints record remains fresh so the
625 address never reaches the final cutoff time.
627 This is a fairly common occurrence when a user is bumping up against
628 their storage quota. Exim had some logic in its local delivery code
629 to deal with this. However it did not apply to per-recipient defers
630 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
632 This change adds a proper retry rule check during routing so that the
633 final cutoff time is checked against the message's age. We only do
634 this check if there is an address retry record and there is not a
635 domain retry record; this implies that previous attempts to handle
636 the address had the retry_use_local_parts option turned on. We use
637 this as an approximation for the destination being like a local
638 delivery, as in LMTP.
640 I suspect this new check makes the old local delivery cutoff check
641 redundant, but I have not verified this so I left the code in place.
643 TF/02 Correct gecos expansion when From: is a prefix of the username.
645 Test 0254 submits a message to Exim with the header
649 When I ran the test suite under the user fanf2, Exim expanded
650 the header to contain my full name, whereas it should have added
651 a Resent-Sender: header. It erroneously treats any prefix of the
652 username as equal to the username.
654 This change corrects that bug.
656 GF/01 DCC debug and logging tidyup
657 Error conditions log to paniclog rather than rejectlog.
658 Debug lines prefixed by "DCC: " to remove any ambiguity.
660 TF/03 Avoid unnecessary rebuilds of lookup-related code.
662 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
663 Bug spotted by Jeremy Harris; was flawed since initial commit.
664 Would have resulted in OCSP responses post-SNI triggering an Exim
665 NULL dereference and crash.
667 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
669 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
670 Bug detection, analysis and fix by Samuel Thibault.
671 Bugzilla 1331, Debian bug #698092.
673 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
675 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
676 Server implementation by Todd Lyons, client by JH.
677 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
678 config variable "prdr_enable" controls whether the server
679 advertises the facility. If the client requests PRDR a new
680 acl_data_smtp_prdr ACL is called once for each recipient, after
681 the body content is received and before the acl_smtp_data ACL.
682 The client is controlled by bolth of: a hosts_try_prdr option
683 on the smtp transport, and the server advertisement.
684 Default client logging of deliveries and rejections involving
685 PRDR are flagged with the string "PRDR".
687 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
688 fclose(). Diagnosis by Todd Lyons.
690 PP/17 Update configure.default to handle IPv6 localhost better.
691 Patch by Alain Williams (plus minor tweaks).
694 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
695 This is now consistent with GnuTLS, and is now documented: the
696 previous undocumented portable approach to treating the option as
697 unset was to force an expansion failure. That still works, and
698 an empty string is now equivalent.
700 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
701 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
702 not performing validation itself.
704 PP/20 Added force_command boolean option to pipe transport.
705 Patch from Nick Koston, of cPanel Inc.
707 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
710 TF/04 Added udpsend ACL modifer and hexquote expansion operator
712 PP/21 Fix eximon continuous updating with timestamped log-files.
713 Broken in a format-string cleanup in 4.80, missed when I repaired the
714 other false fix of the same issue.
715 Report and fix from Heiko Schlichting.
718 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
719 Report from Prashanth Katuri.
721 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
722 It's SecureTransport, so affects any MacOS clients which use the
723 system-integrated TLS libraries, including email clients.
725 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
726 using a MIME ACL for non-SMTP local injection.
727 Report and assistance in diagnosis by Warren Baker.
729 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
731 JH/16 Fix comparisons for 64b. Bugzilla 1385.
733 TL/09 Add expansion variable $authenticated_fail_id to keep track of
734 last id that failed so it may be referenced in subsequent ACL's.
736 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
739 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
740 ldap library initialization, allowing self-signed CA's to be
741 used. Also properly sets require_cert option later in code by
742 using NULL (global ldap config) instead of ldap handle (per
743 session). Bug diagnosis and testing by alxgomz.
745 TL/12 Enhanced documentation in the ratelimit.pl script provided in
746 the src/util/ subdirectory.
748 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
749 renamed to Transport Post Delivery Action by Jeremy Harris, as
752 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
753 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
754 redis_servers = needs to be configured which will be used by the redis
755 lookup. Patch from Warren Baker, of The Packet Hub.
757 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
759 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
760 hostname or reverse DNS when processing a host list. Used suggestions
761 from multiple comments on this bug.
763 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
765 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
766 Missed a few lines, added it to make the runtest require no keyboard
769 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
770 contains upper case chars. Make router use caseful_local_part.
772 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
773 support when GnuTLS has been built with p11-kit.
779 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
781 This, or similar/improved, will also be change PP/11 of 4.82.
787 PP/01 Handle short writes when writing local log-files.
788 In practice, only affects FreeBSD (8 onwards).
789 Bugzilla 1053, with thanks to Dmitry Isaikin.
791 NM/01 Bugzilla 949 - Documentation tweak
793 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
796 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
798 PP/02 Implemented gsasl authenticator.
800 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
802 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
803 `pkg-config foo` for cflags/libs.
805 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
806 with rest of GSASL and with heimdal_gssapi.
808 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
809 `pkg-config foo` for cflags/libs for the TLS implementation.
811 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
812 properties get this fed in as external SSF. A number of robustness
813 and debugging improvements to the cyrus_sasl authenticator.
815 PP/08 cyrus_sasl server now expands the server_realm option.
817 PP/09 Bugzilla 1214 - Log authentication information in reject log.
818 Patch by Jeremy Harris.
820 PP/10 Added dbmjz lookup type.
822 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
824 PP/12 MAIL args handles TAB as well as SP, for better interop with
825 non-compliant senders.
826 Analysis and variant patch by Todd Lyons.
828 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
829 Bug report from Lars Müller <lars@samba.org> (via SUSE),
830 Patch from Dirk Mueller <dmueller@suse.com>
832 PP/13 tls_peerdn now print-escaped for spool files.
833 Observed some $tls_peerdn in wild which contained \n, which resulted
834 in spool file corruption.
836 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
837 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
838 or write after TLS renegotiation, which otherwise led to messages
841 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
842 as a tracking header (ie: a signed header comes before the signature).
843 Patch from Wolfgang Breyha.
845 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
846 comma-sep list; embedded commas doubled.
848 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
850 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
852 Report and patch from Dmitry Banschikov.
854 PP/16 Removed "dont_insert_empty_fragments" fron "openssl_options".
855 Removed SSL_clear() after SSL_new() which led to protocol negotiation
856 failures. We appear to now support TLS1.1+ with Exim.
858 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
859 lets Exim select keys and certificates based upon TLS SNI from client.
860 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
861 before an outbound SMTP session. New log_selector, +tls_sni.
863 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
864 NULL dereference. Report and patch from Alun Jones.
866 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
867 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
868 Not seeing resolver debug output on NetBSD, but suspect this is a
869 resolver implementation change.
871 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
872 Left warnings. Added "eximon gdb" invocation mode.
874 PP/21 Defaulting "accept_8bitmime" to true, not false.
876 PP/22 Added -bw for inetd wait mode support.
878 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
879 locate the relevant includes and libraries. Made this the default.
881 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
882 Bugzilla 1246, report and most of solution from Tomasz Kusy.
884 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
885 This may cause build issues on older platforms.
887 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
888 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
889 gnutls_require_mac & gnutls_require_protocols (no longer supported).
890 Added SNI support via GnuTLS too.
891 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
893 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
895 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
896 Applied second patch from Janne, implementing suggestion to default
897 multiple-strings-in-record handling to match SPF spec.
899 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
901 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
902 read-only, out of scope).
903 Patch from Wolfgang Breyha, report from Stuart Northfield.
905 PP/29 Fix three issues highlighted by clang analyser static analysis.
906 Only crash-plausible issue would require the Cambridge-specific
907 iplookup router and a misconfiguration.
908 Report from Marcin Mirosław.
910 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
912 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
913 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
914 As part of this, removing so much warning spew let me fix some minor
915 real issues in debug logging.
917 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
918 assignment on my part. Fixed.
920 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
921 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
922 Janne Snabb (who went above and beyond: thank you).
924 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
925 string otherwise requires a connection and a bunch more work and it's
926 relatively easy to get wrong. Should also expose TLS library linkage
929 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
930 64-bit ${eval} (JH/03).
932 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
933 GNU libc to support some of the 64-bit stuff, should not lead to
934 conflicts. Defined before os.h is pulled in, so if a given platform
935 needs to override this, it can.
937 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
938 protection layer was required, which is not implemented.
939 Bugzilla 1254, patch from Wolfgang Breyha.
941 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
942 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
943 tls_dhparam take prime identifiers. Also unbreak combination of
944 OpenSSL+DH_params+TLSSNI.
946 PP/39 Disable SSLv2 by default in OpenSSL support.
952 PP/01 Solaris build fix for Oracle's LDAP libraries.
953 Bugzilla 1109, patch from Stephen Usher.
955 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
957 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
960 TF/02 Fix a couple more cases where we did not log the error message
961 when unlink() failed. See also change 4.74-TF/03.
963 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
964 lock up or crash if it happened to be inside a call to libc when it
965 got a SIGUSR1 from exiwhat.
967 The SIGUSR1 handler appends the current process status to the process
968 log which is later printed by exiwhat. It used to use the general
969 purpose logging code to do this, but several functions it calls are
970 not safe for signals.
972 The new output code in the SIGUSR1 handler is specific to the process
973 log, and simple enough that it's easy to inspect for signal safety.
974 Removing some special cases also simplifies the general logging code.
975 Removing the spurious timestamps from the process log simplifies
978 TF/04 Improved ratelimit ACL condition.
980 The /noupdate option has been deprecated in favour of /readonly which
981 has clearer semantics. The /leaky, /strict, and /readonly update modes
982 are mutually exclusive. The update mode is no longer included in the
983 database key; it just determines when the database is updated. (This
984 means that when you upgrde Exim will forget old rate measurements.)
986 Exim now checks that the per_* options are used with an update mode that
987 makes sense for the current ACL. For example, when Exim is processing a
988 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
989 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
990 must specify per_mail/readonly. If you omit the update mode it defaults to
991 /leaky where that makes sense (as before) or /readonly where required.
993 The /noupdate option is now undocumented but still supported for
994 backwards compatibility. It is equivalent to /readonly except that in
995 ACLs where /readonly is required you may specify /leaky/noupdate or
996 /strict/noupdate which are treated the same as /readonly.
998 A useful new feature is the /count= option. This is a generalization
999 of the per_byte option, so that you can measure the throughput of other
1000 aggregate values. For example, the per_byte option is now equivalent
1001 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
1003 The per_rcpt option has been generalized using the /count= mechanism
1004 (though it's more complicated than the per_byte equivalence). When it is
1005 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
1006 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
1007 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
1008 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
1009 Note that using per_rcpt with a non-readonly update mode in more than
1010 one ACL will cause the recipients to be double-counted. (The per_mail
1011 and per_byte options don't have this problem.)
1013 The handling of very low rates has changed slightly. If the computed rate
1014 is less than the event's count (usually one) then this event is the first
1015 after a long gap. In this case the rate is set to the same as this event's
1016 count, so that the first message of a spam run is counted properly.
1018 The major new feature is a mechanism for counting the rate of unique
1019 events. The new per_addr option counts the number of different
1020 recipients that someone has sent messages to in the last time period. It
1021 behaves like per_rcpt if all the recipient addresses are different, but
1022 duplicate recipient addresses do not increase the measured rate. Like
1023 the /count= option this is a general mechanism, so the per_addr option
1024 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
1025 example, measure the rate that a client uses different sender addresses
1026 with the options per_mail/unique=$sender_address. There are further
1027 details in the main documentation.
1029 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
1031 TF/06 Removed a few PCRE remnants.
1033 TF/07 Automatically extract Exim's version number from tags in the git
1034 repository when doing development or release builds.
1036 PP/02 Raise smtp_cmd_buffer_size to 16kB.
1037 Bugzilla 879. Patch from Paul Fisher.
1039 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
1040 Heavily based on revision 40f9a89a from Simon Arlott's tree.
1043 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
1045 PP/05 Variable $av_failed, true if the AV scanner deferred.
1046 Bugzilla 1078. Patch from John Horne.
1048 PP/06 Stop make process more reliably on build failure.
1049 Bugzilla 1087. Patch from Heiko Schlittermann.
1051 PP/07 Make maildir_use_size_file an _expandable_ boolean.
1052 Bugzilla 1089. Patch from Heiko Schlittermann.
1054 PP/08 Handle ${run} returning more data than OS pipe buffer size.
1055 Bugzilla 1131. Patch from Holger Weiß.
1057 PP/09 Handle IPv6 addresses with SPF.
1058 Bugzilla 860. Patch from Wolfgang Breyha.
1060 PP/10 GnuTLS: support TLS 1.2 & 1.1.
1062 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
1065 PP/11 match_* no longer expand right-hand-side by default.
1066 New compile-time build option, EXPAND_LISTMATCH_RHS.
1067 New expansion conditions, "inlist", "inlisti".
1069 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
1071 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
1073 PP/14 fix log_write() format string regression from TF/03.
1074 Bugzilla 1152. Patch from Dmitry Isaikin.
1080 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
1082 PP/02 Harmonised TLS library version reporting; only show if debugging.
1083 Layout now matches that introduced for other libraries in 4.74 PP/03.
1085 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
1087 PP/04 New "dns_use_edns0" global option.
1089 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
1092 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
1093 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
1095 TK/01 Updated PolarSSL code to 0.14.2.
1096 Bugzilla 1097. Patch from Andreas Metzler.
1098 PP/07 Catch divide-by-zero in ${eval:...}.
1099 Fixes bugzilla 1102.
1101 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
1104 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
1105 format-string attack -- SECURITY: remote arbitrary code execution.
1107 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
1108 time unintentionally subject to list matching rules, letting the header
1109 cause arbitrary Exim lookups (of items which can occur in lists, *not*
1110 arbitrary string expansion). This allowed for information disclosure.
1112 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
1113 INT_MIN/-1 -- value coerced to INT_MAX.
1119 NM/01 Workround for PCRE version dependency in version reporting
1122 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
1123 This fixes portability to compilers other than gcc, notably
1124 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
1126 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
1127 makefiles for portability to HP-UX and POSIX correctness.
1129 PP/01 Permit LOOKUP_foo enabling on the make command-line.
1130 Also via indented variable definition in the Makefile.
1131 (Debugging by Oliver Heesakkers).
1133 PP/02 Restore caching of spamd results with expanded spamd_address.
1134 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
1136 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
1137 Improves build reliability. Fix from: Frank Elsner
1139 NM/02 Fix wide character breakage in the rfc2047 coding
1140 Fixes bug 1064. Patch from Andrey N. Oktyabrski
1142 NM/03 Allow underscore in dnslist lookups
1143 Fixes bug 1026. Patch from Graeme Fowler
1145 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
1146 Code patches from Adam Ciarcinski of NetBSD.
1148 NM/04 Fixed exiqgrep to cope with mailq missing size issue
1151 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
1152 is logged, to avoid truncation. Patch from John Horne.
1154 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
1155 Patch from Jakob Hirsch.
1157 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
1158 SQL string expansion failure details.
1159 Patch from Andrey Oktyabrski.
1161 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
1162 Patch from Simon Arlott.
1164 PP/09 New lookups functionality failed to compile on old gcc which rejects
1165 extern declarations in function scope.
1166 Patch from Oliver Fleischmann
1168 PP/10 Use sig_atomic_t for flags set from signal handlers.
1169 Check getgroups() return and improve debugging.
1170 Fixed developed for diagnosis in bug 927 (which turned out to be
1173 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
1174 Patch from Mark Zealey.
1176 PP/12 Bugzilla 1056: Improved spamd server selection.
1177 Patch from Mark Zealey.
1179 PP/13 Bugzilla 1086: Deal with maildir quota file races.
1180 Based on patch from Heiko Schlittermann.
1182 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
1183 Patch from Uwe Doering, sign-off by Michael Haardt.
1185 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
1186 variable declaration deep within a block. Bug and patch from
1189 PP/15 lookups-Makefile IRIX compatibilty coercion.
1191 PP/16 Make DISABLE_DKIM build knob functional.
1193 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
1194 Patch by Simon Arlott
1196 TF/03 Fix valgrind.h portability to C89 compilers that do not support
1197 variable argument macros. Our copy now differs from upstream.
1203 TF/01 Failure to get a lock on a hints database can have serious
1204 consequences so log it to the panic log.
1206 TF/02 Log LMTP confirmation messages in the same way as SMTP,
1207 controlled using the smtp_confirmation log selector.
1209 TF/03 Include the error message when we fail to unlink a spool file.
1211 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
1212 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
1213 for maintaining out-of-tree patches for some time.
1215 PP/01 Bugzilla 139: Documentation and portability issues.
1216 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
1217 Handle per-OS dynamic-module compilation flags.
1219 PP/02 Let /dev/null have normal permissions.
1220 The 4.73 fixes were a little too stringent and complained about the
1221 permissions on /dev/null. Exempt it from some checks.
1222 Reported by Andreas M. Kirchwitz.
1224 PP/03 Report version information for many libraries, including
1225 Exim version information for dynamically loaded libraries. Created
1226 version.h, now support a version extension string for distributors
1227 who patch heavily. Dynamic module ABI change.
1229 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
1230 privilege escalation vulnerability whereby the Exim run-time user
1231 can cause root to append content of the attacker's choosing to
1234 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
1237 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
1238 If dropping privileges for untrusted macros, we disabled normal logging
1239 on the basis that it would fail; for the Exim run-time user, this is not
1240 the case, and it resulted in successful deliveries going unlogged.
1241 Fixed. Reported by Andreas Metzler.
1247 PP/01 Date: & Message-Id: revert to normally being appended to a message,
1248 only prepend for the Resent-* case. Fixes regression introduced in
1249 Exim 4.70 by NM/22 for Bugzilla 607.
1251 PP/02 Include check_rfc2047_length in configure.default because we're seeing
1252 increasing numbers of administrators be bitten by this.
1254 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
1256 PP/03 Bugzilla 994: added openssl_options main configuration option.
1258 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
1260 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
1262 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
1264 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
1265 without a peer certificate, leading to a segfault because of an
1266 assumption that peers always have certificates. Be a little more
1267 paranoid. Problem reported by Martin Tscholak.
1269 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
1270 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
1271 NB: ClamAV planning to remove STREAM in "middle of 2010".
1272 CL also introduces -bmalware, various -d+acl logging additions and
1273 more caution in buffer sizes.
1275 PP/09 Implemented reverse_ip expansion operator.
1277 PP/10 Bugzilla 937: provide a "debug" ACL control.
1279 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
1281 PP/12 Bugzilla 973: Implement --version.
1283 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
1285 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
1287 PP/15 Bugzilla 816: support multiple condition rules on Routers.
1289 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
1290 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
1291 ignore trailing whitespace.
1293 JJ/02 prevent non-panic DKIM error from being sent to paniclog
1295 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
1298 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
1299 Notification from Dr Andrew Aitchison.
1301 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
1302 ExtendedDetectionInfo response format.
1303 Notification from John Horne.
1305 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
1308 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
1309 XSL and documented dependency on system catalogs, with examples of how
1312 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
1315 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
1316 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
1317 configuration file which is writeable by the Exim user or group.
1319 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
1320 of configuration files to cover files specified with the -C option if
1321 they are going to be used with root privileges, not just the default
1324 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
1325 option (effectively making it always true).
1327 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
1328 files to be used while preserving root privileges.
1330 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
1331 that rogue child processes cannot use them.
1333 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
1334 run-time user, instead of root.
1336 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
1337 Exim run-time user without dropping privileges.
1339 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
1340 result string, instead of calling string_vformat() twice with the same
1343 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
1344 for other users. Others should always drop root privileges if they use
1345 -C on the command line, even for a whitelisted configure file.
1347 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
1349 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
1355 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
1356 $data_path, and $header_path variables; fixed documentation bugs and
1359 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
1360 exipick to access non-standard spools, including the "frozen" queue
1363 NM/01 Bugzilla 965: Support mysql stored procedures.
1364 Patch from Alain Williams
1366 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
1368 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
1369 Patch from Andreas Metzler
1371 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
1372 Patch from Kirill Miazine
1374 NM/05 Bugzilla 671: Added umask to procmail example.
1376 JJ/03 installed exipick 20100323.0, fixing doc bug
1378 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
1379 directory. Notification and patch from Dan Rosenberg.
1381 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
1383 TK/02 Improve log output when DKIM signing operation fails.
1385 MH/01 Treat the transport option dkim_domain as a colon separated
1386 list, not as a single string, and sign the message with each element,
1387 omitting multiple occurences of the same signer.
1389 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
1390 Bugzilla 985, 986. Patch by Simon Arlott
1392 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
1393 Patch by Simon Arlott
1395 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
1396 MBX locking. Notification from Dan Rosenberg.
1402 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
1404 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
1406 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
1408 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
1410 NM/04 Bugzilla 915: Flag broken perl installation during build.
1416 TK/01 Added patch by Johannes Berg that expands the main option
1417 "spamd_address" if it starts with a dollar sign.
1419 TK/02 Write list of recipients to X-Envelope-Sender header when building
1420 the mbox-format spool file for content scanning (suggested by Jakob
1423 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
1424 (http://www.dcc-servers.net/) support via dccifd. Activated by
1425 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
1427 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
1428 by Mark Daniel Reidel <mr@df.eu>.
1430 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
1431 When building exim an external PCRE library is now needed -
1432 PCRE is a system library on the majority of modern systems.
1433 See entry on PCRE_LIBS in EDITME file.
1435 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
1436 conversation. Added nologin parameter to request.
1437 Patch contributed by Kirill Miazine.
1439 TF/01 Do not log submission mode rewrites if they do not change the address.
1441 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
1443 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
1444 log files in place. Contributed by Roberto Lima.
1446 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
1448 TF/03 Bugzilla 615: When checking the local_parts router precondition
1449 after a local_part_suffix or local_part_prefix option, Exim now
1450 does not use the address's named list lookup cache, since this
1451 contains cached lookups for the whole local part.
1453 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
1454 Robert Millan. Documentation is in experimental-spec.txt.
1456 TF/04 Bugzilla 668: Fix parallel build (make -j).
1458 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
1460 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
1461 Patch provided by Jan Srzednicki.
1463 TF/05 Leading white space used to be stripped from $spam_report which
1464 wrecked the formatting. Now it is preserved.
1466 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
1467 that they are available at delivery time.
1469 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
1471 TF/08 TLS error reporting now respects the incoming_interface and
1472 incoming_port log selectors.
1474 TF/09 Produce a more useful error message if an SMTP transport's hosts
1475 setting expands to an empty string.
1477 NM/06 Bugzilla 744: EXPN did not work under TLS.
1478 Patch provided by Phil Pennock.
1480 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
1481 Patch provided by Richard Godbee.
1483 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
1484 acl_smtp_notquit, added index entry.
1486 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
1487 Patch provided by Eugene Bujak.
1489 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
1490 accept(). Patch provided by Maxim Dounin.
1492 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
1493 Patch provided by Phil Pennock.
1495 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
1497 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
1498 Patch provided by Brad "anomie" Jorsch.
1500 NM/14 Bugzilla 622: Added timeout setting to transport filter.
1501 Patch provided by Dean Brooks.
1503 TK/05 Add native DKIM support (does not depend on external libraries).
1505 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
1506 Patch provided by Graeme Fowler.
1508 NM/16 Bugzilla 851: Documentation example syntax fix.
1510 NM/17 Changed NOTICE file to remove references to embedded PCRE.
1512 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
1515 NM/19 Bugzilla 745: TLS version reporting.
1516 Patch provided by Phil Pennock.
1518 NM/20 Bugzilla 167: bool: condition support.
1519 Patch provided by Phil Pennock.
1521 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
1522 clients. Patch provided by Phil Pennock.
1524 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
1525 Patch provided by Brad "anomie" Jorsch.
1527 NM/23 Bugzilla 687: Fix misparses in eximstats.
1528 Patch provided by Heiko Schlittermann.
1530 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
1531 Patch provided by Heiko Schlittermann.
1533 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
1534 plus update to original patch.
1536 NM/26 Bugzilla 799: Documentation correction for ratelimit.
1538 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
1539 Patch provided by David Brownlee.
1541 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
1543 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
1545 NM/30 Bugzilla 888: TLS documentation bugfixes.
1547 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
1549 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
1550 Unlike the original bugzilla I have changed all shell scripts in src tree.
1552 NM/33 Bugzilla 898: Transport filter timeout fix.
1553 Patch by Todd Rinaldo.
1555 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
1556 Patch by Serge Demonchaux.
1558 NM/35 Bugzilla 39: Base64 decode bug fixes.
1559 Patch by Jakob Hirsch.
1561 NM/36 Bugzilla 909: Correct connect() call in dcc code.
1563 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
1565 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
1567 NM/39 Bugzilla 911: Fixed MakeLinks build script.
1573 TK/01 Add preliminary DKIM support. Currently requires a forked version of
1574 ALT-N's libdkim that I have put here:
1575 http://duncanthrax.net/exim-experimental/
1577 Note to Michael Haardt: I had to rename some vars in sieve.c. They
1578 were called 'true' and it seems that C99 defines that as a reserved
1579 keyword to be used with 'bool' variable types. That means you could
1580 not include C99-style headers which use bools without triggering
1581 build errors in sieve.c.
1583 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
1584 as mailq or other aliases. Changed the --help handling significantly
1585 to do whats expected. exim_usage() emits usage/help information.
1587 SC/01 Added the -bylocaldomain option to eximstats.
1589 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
1591 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
1593 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
1599 PH/01 Another patch from the Sieve maintainer.
1601 PH/02 When an IPv6 address is converted to a string for single-key lookup
1602 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
1603 dots are used instead of colons so that keys in lsearch files need not
1604 contain colons. This was done some time before quoting was made available
1605 in lsearch files. However, iplsearch files do require colons in IPv6 keys
1606 (notated using the quote facility) so as to distinguish them from IPv4
1607 keys. This meant that lookups for IP addresses in host lists did not work
1608 for iplsearch lookups.
1610 This has been fixed by arranging for IPv6 addresses to be expressed with
1611 colons if the lookup type is iplsearch. This is not incompatible, because
1612 previously such lookups could never work.
1614 The situation is now rather anomolous, since one *can* have colons in
1615 ordinary lsearch keys. However, making the change in all cases is
1616 incompatible and would probably break a number of configurations.
1618 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
1621 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
1622 conversion specification without a maximum field width, thereby enabling
1623 a rogue spamd server to cause a buffer overflow. While nobody in their
1624 right mind would setup Exim to query an untrusted spamd server, an
1625 attacker that gains access to a server running spamd could potentially
1626 exploit this vulnerability to run arbitrary code as the Exim user.
1628 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
1629 $primary_hostname instead of what libspf2 thinks the hosts name is.
1631 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
1632 a directory entry by the name of the lookup key. Previously, if a
1633 symlink pointed to a non-existing file or a file in a directory that
1634 Exim lacked permissions to read, a lookup for a key matching that
1635 symlink would fail. Now it is enough that a matching directory entry
1636 exists, symlink or not. (Bugzilla 503.)
1638 PH/03 The body_linecount and body_zerocount variables are now exported in the
1641 PH/04 Added the $dnslist_matched variable.
1643 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
1644 This means they are set thereafter only if the connection becomes
1647 PH/06 Added the client_condition to authenticators so that some can be skipped
1648 by clients under certain conditions.
1650 PH/07 The error message for a badly-placed control=no_multiline_responses left
1651 "_responses" off the end of the name.
1653 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
1655 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
1656 (without spaces) instead of just copying the configuration text.
1658 PH/10 Added the /noupdate option to the ratelimit ACL condition.
1660 PH/11 Added $max_received_linelength.
1662 PH/12 Added +ignore_defer and +include_defer to host lists.
1664 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
1665 way in which PCRE > 7.0 is built.
1667 PH/14 Implemented queue_only_load_latch.
1669 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
1670 MAIL command. The effect was to mangle the value on 64-bit systems.
1672 PH/16 Another patch from the Sieve maintainer.
1674 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
1676 PH/18 If a system quota error occurred while trying to create the file for
1677 a maildir delivery, the message "Mailbox is full" was not appended to the
1678 bounce if the delivery eventually timed out. Change 4.67/27 below applied
1679 only to a quota excession during the actual writing of the file.
1681 PH/19 It seems that peer DN values may contain newlines (and other non-printing
1682 characters?) which causes problems in log lines. The DN values are now
1683 passed through string_printing() before being added to log lines.
1685 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
1686 and InterBase are left for another time.)
1688 PH/21 Added message_body_newlines option.
1690 PH/22 Guard against possible overflow in moan_check_errorcopy().
1692 PH/23 POSIX allows open() to be a macro; guard against that.
1694 PH/24 If the recipient of an error message contained an @ in the local part
1695 (suitably quoted, of course), incorrect values were put in $domain and
1696 $local_part during the evaluation of errors_copy.
1702 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
1703 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
1706 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
1707 issue a MAIL command.
1709 PH/02 In an ACL statement such as
1711 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
1713 if a client was not listed at all, or was listed with a value other than
1714 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
1715 the condition was not true (as it should be), so access was not denied.
1716 The bug was that the ! inversion was incorrectly passed on to the second
1717 item. This has been fixed.
1719 PH/03 Added additional dnslists conditions == and =& which are different from
1720 = and & when the dns lookup returns more than one IP address.
1722 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
1723 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
1725 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
1726 FSYNC, which compiles an option called disable_fsync that allows for
1727 bypassing fsync(). The documentation is heavily laced with warnings.
1729 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
1731 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
1732 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
1733 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
1734 including adding "make clean"; (3) Added -fPIC when compiling the test
1735 dynamically loaded module, to get rid of a warning.
1737 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
1738 message fails, move_frozen_messages = true and ignore_bounce_errors_after
1739 = 0s. The bug is otherwise harmless.
1741 PH/07 There was a bug in the dovecot authenticator such that the value of
1742 $auth1 could be overwritten, and so not correctly preserved, after a
1743 successful authentication. This usually meant that the value preserved by
1744 the server_setid option was incorrect.
1746 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
1748 PH/09 Installed PCRE release 7.0.
1750 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
1751 run for batched SMTP input. It is now run at the start of every message
1752 in the batch. While fixing this I discovered that the process information
1753 (output by running exiwhat) was not always getting set for -bs and -bS
1754 input. This is fixed, and it now also says "batched" for BSMTP.
1756 PH/11 Added control=no_pipelining.
1758 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
1759 patch, slightly modified), and move the expansion of helo_data till after
1760 the connection is made in the smtp transport (so it can use these
1763 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
1765 PH/14 Added log_selector = +pid.
1767 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
1769 PH/16 Add ${if forany and ${if forall.
1771 PH/17 Added dsn_from option to vary the From: line in DSNs.
1773 PH/18 Flush SMTP output before performing a callout, unless control =
1774 no_callout_flush is set.
1776 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
1777 was true (the default) a successful delivery failed to delete the retry
1778 item, thus causing premature timeout of the address. The bug is now
1781 PH/20 Added hosts_avoid_pipelining to the smtp transport.
1783 PH/21 Long custom messages for fakedefer and fakereject are now split up
1784 into multiline reponses in the same way that messages for "deny" and
1785 other ACL rejections are.
1787 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
1788 with slight modification.
1790 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
1791 draft, changing the syntax and factoring some duplicate code".
1793 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
1794 for deliveries of the second and subsequent messages over the same SMTP
1797 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
1798 ${reduce, with only minor "tidies".
1800 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
1802 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
1803 expansion side effects.
1805 PH/27 When a message times out after an over-quota error from an Exim-imposed
1806 quota, the bounce message says "mailbox is full". This message was not
1807 being given when it was a system quota that was exceeded. It now should
1810 MH/03 Made $recipients available in local_scan(). local_scan() already has
1811 better access to the recipient list through recipients_list[], but
1812 $recipients can be useful in postmaster-provided expansion strings.
1814 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
1815 in the case of a MAIL command with additional options following the
1816 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
1817 were accidentally chopped off.
1819 PH/29 SMTP synchronization checks are implemented when a command is read -
1820 there is a check that no more input is waiting when there shouldn't be
1821 any. However, for some commands, a delay in an ACL can mean that it is
1822 some time before the response is written. In this time, more input might
1823 arrive, invalidly. So now there are extra checks after an ACL has run for
1824 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
1825 pipelining has not been advertised.
1827 PH/30 MH's patch to allow iscntrl() characters to be list separators.
1829 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
1830 returned in the SMTP response when smtp_return_error_details was false.
1831 This has been fixed.
1833 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
1834 instead of the C I/O that was originally supplied, because problems were
1835 reported on Solaris.
1837 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
1838 Exim which did not show up earlier: it was assuming that a call to
1839 SSL_CTX_set_info_callback() might give an error value. In fact, there is
1840 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
1841 was a macro that became an assignment, so it seemed to work. This has
1842 changed to a proper function call with a void return, hence the compile
1843 error. Exim's code has been fixed.
1845 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
1848 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
1850 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
1852 PH/37 If a message is not accepted after it has had an id assigned (e.g.
1853 because it turns out to be too big or there is a timeout) there is no
1854 "Completed" line in the log. When some messages of this type were
1855 selected by exigrep, they were listed as "not completed". Others were
1856 picked up by some special patterns. I have improved the selection
1857 criteria to be more general.
1859 PH/38 The host_find_failed option in the manualroute router can now be set
1860 to "ignore", to completely ignore a host whose IP address cannot be
1861 found. If all hosts are ignored, the behaviour is controlled by the new
1862 host_all_ignored option.
1864 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
1865 homing or because of multiple MX records with /mx) generated more than
1866 one IP address, and the following item turned out to be the local host,
1867 all the secondary addresses of the first item were incorrectly removed
1868 from the list, along with the local host and any following hosts (which
1869 is what is supposed to happen).
1871 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
1872 whoever called Exim into the -H file. In the case of the daemon it was
1873 behaving confusingly. When first started, it used values for whoever
1874 started the daemon, but after a SIGHUP it used the Exim user (because it
1875 calls itself on a restart). I have changed the code so that it now always
1878 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
1879 message are rejected with the same error (e.g. no authentication or bad
1880 sender address), and a DATA command is nevertheless sent (as can happen
1881 with PIPELINING or a stupid MUA), the error message that was given to the
1882 RCPT commands is included in the rejection of the DATA command. This is
1883 intended to be helpful for MUAs that show only the final error to their
1886 PH/42 Another patch from the Sieve maintainer.
1888 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
1889 Eximstats - Fixed some broken HTML links and added missing column headers
1891 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
1892 columns for Rejects, Temp Rejects, Ham, and Spam rows.
1894 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
1896 PH/43 Yet another patch from the Sieve maintainer.
1898 PH/44 I found a way to check for a TCP/IP connection going away before sending
1899 the response to the final '.' that terminates a message, but only in the
1900 case where the client has not sent further data following the '.'
1901 (unfortunately, this is allowed). However, in many cases there won't be
1902 any further data because there won't be any more messages to send. A call
1903 to select() can be used: if it shows that the input is "ready", there is
1904 either input waiting, or the socket has been closed. An attempt to read
1905 the next input character can distinguish the two cases. Previously, Exim
1906 would have sent an OK response which the client would never have see.
1907 This could lead to message repetition. This fix should cure that, at
1908 least in a lot of common cases.
1910 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
1911 advertised in response to EHLO.
1917 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
1918 fixed by 4.65/MH/01 (is this a record?) are fixed:
1920 (i) An empty string was always treated as zero by the numeric comparison
1921 operators. This behaviour has been restored.
1923 (ii) It is documented that the numeric comparison operators always treat
1924 their arguments as decimal numbers. This was broken in that numbers
1925 starting with 0 were being interpreted as octal.
1927 While fixing these problems I realized that there was another issue that
1928 hadn't been noticed. Values of message_size_limit (both the global option
1929 and the transport option) were treated as octal if they started with 0.
1930 The documentation was vague. These values are now always treated as
1931 decimal, and I will make that clear in the documentation.
1937 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
1938 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
1941 MH/01 Don't check that the operands of numeric comparison operators are
1942 integers when their expansion is in "skipping" mode (fixes bug
1943 introduced by 4.64-PH/07).
1945 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
1946 child addresses, Exim now panics and dies. Previously, because the count
1947 is held in a short int, deliveries were likely to be lost. As such a
1948 large number of recipients for a single message is ridiculous
1949 (performance will be very, very poor), I have chosen to impose a limit
1950 rather than extend the field.
1956 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
1957 leftover -K file (the existence of which was triggered by #402).
1958 While we were at it, introduced process PID as part of the -K
1959 filename. This should rule out race conditions when creating
1962 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
1963 processing considerably. Previous code took too long for large mails,
1964 triggering a timeout which in turn triggers #401.
1966 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
1967 in the DK code in transports.c. sendfile() is not really portable,
1968 hence the _LINUX specificness.
1970 TF/01 In the add_headers option to the mail command in an Exim filter,
1971 there was a bug that Exim would claim a syntax error in any
1972 header after the first one which had an odd number of characters
1975 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
1976 callout verification, Exim cached a "reject" for the entire domain. This
1977 is correct for most verifications, but it is not correct for a recipient
1978 verification with use_sender or use_postmaster set, because in that case
1979 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
1980 case of MAIL FROM:<> rejection from other early rejections (e.g.
1981 rejection of HELO). When verifying a recipient using a non-null MAIL
1982 address, the cache is ignored if it shows MAIL FROM:<> rejection.
1983 Whatever the result of the callout, the value of the domain cache is
1984 left unchanged (for any other kind of callout, getting as far as trying
1985 RCPT means that the domain itself is ok).
1987 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
1990 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
1991 manifest itself as EPIPE rather than ECONNECT. When tidying away a
1992 session, the daemon ignores ECONNECT errors and logs others; it now
1993 ignores EPIPE as well.
1995 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
1996 (quoted-printable decoding).
1998 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
1999 later the small subsequent patch to fix an introduced bug.
2001 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
2003 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
2005 PH/08 An error is now given if message_size_limit is specified negative.
2007 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
2008 to be given (somewhat) arbitrary names.
2010 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
2013 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
2014 miscellaneous code fixes
2016 PH/10 Added the log_reject_target ACL modifier to specify where to log
2019 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
2020 hostname. This is wrong, because it relates to the incoming message (and
2021 probably the interface on which it is arriving) and not to the outgoing
2022 callout (which could be using a different interface). This has been
2023 changed to use the value of the helo_data option from the smtp transport
2024 instead - this is what is used when a message is actually being sent. If
2025 there is no remote transport (possible with a router that sets up host
2026 addresses), $smtp_active_hostname is used.
2028 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
2029 tweaks were necessary in order to get it to work (see also 21 below):
2030 (a) The code assumed that strncpy() returns a negative number on buffer
2031 overflow, which isn't the case. Replaced with Exim's string_format()
2033 (b) There were several signed/unsigned issues. I just did the minimum
2034 hacking in of casts. There is scope for a larger refactoring.
2035 (c) The code used strcasecmp() which is not a standard C function.
2036 Replaced with Exim's strcmpic() function.
2037 (d) The code set only $1; it now sets $auth1 as well.
2038 (e) A simple test gave the error "authentication client didn't specify
2039 service in request". It would seem that Dovecot has changed its
2040 interface. Fortunately there's a specification; I followed it and
2041 changed what the client sends and it appears to be working now.
2043 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
2046 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
2047 address A is aliased to B and C, where B exists and C does not. Without
2048 -v the output is "A verified" because verification stops after a
2049 successful redirection if more than one address is generated. However,
2050 with -v the child addresses are also verified. Exim was outputting "A
2051 failed to verify" and then showing the successful verification for C,
2052 with its parentage. It now outputs "B failed to verify", showing B's
2053 parentage before showing the successful verification of C.
2055 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
2056 look up a TXT record in a specific list after matching in a combined
2059 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
2060 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
2061 they consult the DNS. I had assumed they would set it the way they
2062 wanted; and indeed my experiments on Linux seem to show that in some
2063 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
2064 To be on the safe side, however, I have now made the interface to
2065 host_find_byname() similar to host_find_bydns(), with an argument
2066 containing the DNS resolver options. The host_find_byname() function now
2067 sets these options at its start, just as host_find_bydns() does. The smtp
2068 transport options dns_qualify_single and dns_search_parents are passed to
2069 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
2070 of host_find_byname() use the default settings of RES_DEFNAMES
2071 (qualify_single) but not RES_DNSRCH (search_parents).
2073 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
2074 spool_read_header() do less string testing, by means of a preliminary
2075 switch on the second character of optional "-foo" lines. (This is
2076 overdue, caused by the large number of possibilities that now exist.
2077 Originally there were few.) While I was there, I also converted the
2078 str(n)cmp tests so they don't re-test the leading "-" and the first
2079 character, in the hope this might squeeze out yet more improvement.
2081 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
2082 flag allowing group syntax was set by the header_syntax check but not
2083 turned off, possible causing trouble later; (2) The flag was not being
2084 set at all for the header_verify test, causing "group"-style headers to
2085 be rejected. I have now set it in this case, and also caused header_
2086 verify to ignore an empty address taken from a group. While doing this, I
2087 came across some other cases where the code for allowing group syntax
2088 while scanning a header line wasn't quite right (mostly, not resetting
2089 the flag correctly in the right place). These bugs could have caused
2090 trouble for malformed header lines. I hope it is now all correct.
2092 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
2093 with the "reply" argument non-NULL. The code, however (which originally
2094 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
2095 but it didn't always do it. This confused somebody who was copying the
2096 code for some other use. I have removed all the tests.
2098 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
2099 feature that was used to support insecure browsers during the U.S. crypto
2100 embargo. It requires special client support, and Exim is probably the
2101 only MTA that supported it -- and would never use it because real RSA is
2102 always available. This code has been removed, because it had the bad
2103 effect of slowing Exim down by computing (never used) parameters for the
2104 RSA_EXPORT functionality.
2106 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
2107 authenticator to fail if there's a tab character in the incoming data
2108 (there should never be unless someone is messing about, as it's supposed
2109 to be base64-encoded). Also added, on Timo's advice, the "secured" option
2110 if the connection is using TLS or if the remote IP is the same as the
2111 local IP, and the "valid-client-cert option" if a client certificate has
2114 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
2115 authenticators. This can be used for authorization after authentication
2116 succeeds. (In the case of plaintext, it servers for both authentication
2119 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
2120 if any retry times were supplied.
2122 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
2123 connection, where there is no client IP address to check. In this
2124 situation, the verify now always succeeds.
2126 PH/25 Applied John Jetmore's -Mset patch.
2128 PH/26 Added -bem to be like -Mset, but loading a message from a file.
2130 PH/27 In a string expansion for a processed (not raw) header when multiple
2131 headers of the same name were present, leading whitespace was being
2132 removed from all of them, but trailing whitespace was being removed only
2133 from the last one. Now trailing whitespace is removed from each header
2134 before concatenation. Completely empty headers in a concatenation (as
2135 before) are ignored.
2137 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
2138 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
2140 PH/29 [Removed. This was a change that I later backed out, and forgot to
2141 correct the ChangeLog entry (that I had efficiently created) before
2142 committing the later change.]
2144 PH/30 Exim was sometimes attempting to deliver messages that had suffered
2145 address errors (4xx response to RCPT) over the same connection as other
2146 messages routed to the same hosts. Such deliveries are always "forced",
2147 so retry times are not inspected. This resulted in far too many retries
2148 for the affected addresses. The effect occurred only when there were more
2149 hosts than the hosts_max_try setting in the smtp transport when it had
2150 the 4xx errors. Those hosts that it had tried were not added to the list
2151 of hosts for which the message was waiting, so if all were tried, there
2152 was no problem. Two fixes have been applied:
2154 (i) If there are any address or message errors in an SMTP delivery, none
2155 of the hosts (tried or untried) are now added to the list of hosts
2156 for which the message is waiting, so the message should not be a
2157 candidate for sending over the same connection that was used for a
2158 successful delivery of some other message. This seems entirely
2159 reasonable: after all the message is NOT "waiting for some host".
2160 This is so "obvious" that I'm not sure why it wasn't done
2161 previously. Hope I haven't missed anything, but it can't do any
2162 harm, as the worst effect is to miss an optimization.
2164 (ii) If, despite (i), such a delivery is accidentally attempted, the
2165 routing retry time is respected, so at least it doesn't keep
2166 hammering the server.
2168 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
2169 in ${readsocket because some servers need this prod.
2171 PH/32 Added some extra debug output when updating a wait-xxx database.
2173 PH/33 The hint "could be header name not terminated by colon", which has been
2174 given for certain expansion errors for a long time, was not being given
2175 for the ${if def:h_colon_omitted{... case.
2177 PH/34 The spec says: "With one important exception, whenever a domain list is
2178 being scanned, $domain contains the subject domain." There was at least
2179 one case where this was not true.
2181 PH/35 The error "getsockname() failed: connection reset by peer" was being
2182 written to the panic log as well as the main log, but it isn't really
2183 panic-worthy as it just means the connection died rather early on. I have
2184 removed the panic log writing for the ECONNRESET error when getsockname()
2187 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
2188 runs only) independently of the message's sender address. This meant
2189 that, if the 4xx error was in fact related to the sender, a different
2190 message to the same recipient with a different sender could confuse
2191 things. In particualar, this can happen when sending to a greylisting
2192 server, but other circumstances could also provoke similar problems.
2193 I have changed the default so that the retry time for these errors is now
2194 based a combination of the sender and recipient addresses. This change
2195 can be overridden by setting address_retry_include_sender=false in the
2198 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
2199 remote server are returned as part of bounce messages. This was not
2200 happening for LMTP over a pipe (the lmtp transport), but now it is the
2201 same for both kinds of LMTP.
2203 PH/38 Despite being documented as not happening, Exim was rewriting addresses
2204 in header lines that were in fact CNAMEs. This is no longer the case.
2206 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
2207 and queue runs started by the daemon processed all messages. This has
2208 been fixed so that -R and -S can now usefully be given with -q<time>.
2210 PH/40 Import PCRE release 6.7 (fixes some bugs).
2212 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
2214 PH/42 Give an error if -q is specified more than once.
2216 PH/43 Renamed the variables $interface_address and $interface_port as
2217 $received_ip_address and $received_port, to make it clear that these
2218 values apply to message reception, and not to the outgoing interface when
2219 a message is delivered. (The old names remain recognized, of course.)
2221 PH/44 There was no timeout on the connect() call when using a Unix domain
2222 socket in the ${readsocket expansion. There now is.
2224 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
2225 be meaningful with "accept".
2227 SC/01 Eximstats V1.43
2228 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
2230 SC/02 Eximstats V1.44
2231 Use a glob alias rather than an array ref in the generated
2232 parser. This improves both readability and performance.
2234 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
2235 Collect SpamAssassin and rejection statistics.
2236 Don't display local sender or destination tables unless
2237 there is data to show.
2238 Added average volumes into the top table text output.
2240 SC/04 Eximstats V1.46
2241 Collect data on the number of addresses (recipients)
2242 as well as the number of messages.
2244 SC/05 Eximstats V1.47
2245 Added 'Message too big' to the list of mail rejection
2246 reasons (thanks to Marco Gaiarin).
2248 SC/06 Eximstats V1.48
2249 Mainlog lines which have GMT offsets and are too short to
2250 have a flag are now skipped.
2252 SC/07 Eximstats V1.49 (Alain Williams)
2253 Added the -emptyok flag.
2255 SC/08 Eximstats V1.50
2256 Fixes for obtaining the IP address from reject messages.
2258 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
2259 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
2260 whitesspace changes from 4.64-PH/27
2262 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
2265 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
2266 are found, allow negative numbers in numeric criteria)
2268 JJ/06 exipick.20061117.2, added new $message_body_missing variable
2270 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
2271 to match changes made in 4.64-PH/43
2273 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
2275 PH/47 Put in an explicit test for a DNS lookup of an address record where the
2276 "domain" is actually an IP address, and force a failure. This locks out
2277 those revolvers/nameservers that support "A-for-A" lookups, in
2278 contravention of the specifications.
2280 PH/48 When a host name was looked up from an IP address, and the subsequent
2281 forward lookup of the name timed out, the host name was left in
2282 $sender_host_name, contrary to the specification.
2284 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
2285 restricted to single-key lookups, Exim was not diagnosing an error if
2286 * or *@ was used with a query-style lookup.
2288 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
2290 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
2291 long ago, but noone interested enough thought of it. Let's just say that
2292 the "1.1" means that there are some new functions that weren't there at
2293 some point in the past.
2295 PH/51 Error processing for expansion failure of helo_data from an smtp
2296 transport during callout processing was broken.
2298 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
2299 tested/used via the -bh/-bhc/-bs options.
2301 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
2302 bug, fixed in subsequent PCRE releases).
2304 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
2305 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
2307 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
2313 SC/01 Use a glob alias rather than an array ref in eximstats generated
2314 parser. This improves both readability and performance.
2316 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
2317 Don't display local sender or destination tables in eximstats unless
2318 there is data to show.
2319 Added average volumes into the eximstats top table text output.
2321 SC/03 Collect data on the number of addresses (recipients) as well
2322 as the number of messages in eximstats.
2324 TF/01 Correct an error in the documentation for the redirect router. Exim
2325 does (usually) call initgroups() when daemonizing.
2327 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
2328 with consistent privilege compared to when running as a daemon.
2330 TF/03 Note in the spec that $authenticated_id is not set for local
2331 submissions from trusted users.
2333 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
2334 Thanks to Dean Brooks <dean@iglou.com> for the patch.
2336 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
2337 by adding some example configuration directives to the default
2338 configuration file. A little bit of work is required to uncomment the
2339 directives and define how usernames and passwords are checked, but
2340 there is now a framework to start from.
2342 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
2343 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
2344 without this. I don't know how relevant this is to other LDAP libraries.
2346 PH/02 Add the verb name to the "unknown ACL verb" error.
2348 PH/03 Magnus Holmgren's patch for filter_prepend_home.
2350 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
2352 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
2353 directory not expanded when it should be if an expanded home directory
2354 was set for the address (which is overridden by the transport).
2356 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
2359 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
2360 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
2361 because it is too late at that time, and has no effect.
2363 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
2364 security issue with \' (bugzilla #107). I could not use the
2365 PQescapeStringConn() function, because it needs a PGconn value as one of
2368 PH/08 When testing addresses using -bt, indicate those final addresses that
2369 are duplicates that would not cause an additional delivery. At least one
2370 person was confused, thinking that -bt output corresponded to deliveries.
2371 (Suppressing duplicates isn't a good idea as you lose the information
2372 about possibly different redirections that led to the duplicates.)
2374 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
2375 systems where poll() doesn't work, in particular OS X.
2377 PH/10 Added more information to debugging output for retry time not reached.
2379 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
2380 operations in malware.c.
2382 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
2385 PH/13 If write_rejectlog was set false when logging was sent to syslog with
2386 syslog_duplication set false, log lines that would normally be written
2387 both the the main log and to the reject log were not written to syslog at
2390 PH/14 In the default configuration, change the use of "message" in ACL warn
2391 statements to "add_header".
2393 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
2394 not followed by a command (e.g. "seen endif").
2396 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
2397 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
2400 PH/17 Added extra conditions to the default value of delay_warning_condition
2404 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
2405 { match{$h_precedence:}{(?i)bulk|list|junk} } \
2406 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
2409 The Auto-Submitted: and various List- headers are standardised, whereas I
2410 don't think Precedence: ever was.
2412 PH/18 Refactored debugging code in route_finduser() to show more information,
2413 in particular, the error code if getpwnam() issues one.
2415 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
2416 This is apparently needed in addition to the PH/07 change above to avoid
2417 any possible encoding problems.
2419 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
2420 but not after initializing Perl.
2422 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
2423 output them only if debugging. By default they are written stderr,
2424 apparently, which is not desirable.
2426 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
2429 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
2432 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
2434 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
2435 authenticated or an ident call has been made. Suppress the default
2436 values for $authenticated_id and $authenticated_sender (but permit -oMai
2437 and -oMas) when testing with -bh.
2439 PH/24 Re-jigged the order of the tests in the default configuration so that the
2440 tests for valid domains and recipients precede the DNS black list and CSA
2441 tests, on the grounds that those ones are more expensive.
2443 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
2444 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
2445 This bug exists in every version of Exim that I still have, right back to
2448 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
2449 However, an attempt to turn on case-sensitivity in a regex key by
2450 including (?-i) didn't work because the subject string was already
2451 lowercased, and the effects were non-intuitive. It turns out that a
2452 one-line patch can be used to allow (?-i) to work as expected.
2458 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
2459 other effects) broke the use of negated acl sub-conditions.
2461 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
2464 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
2465 "Deny" causes Exim to reject the incoming connection with a 554 error.
2466 Unfortunately, if there is a major crisis, such as a disk failure,
2467 tcp-wrappers gives "deny", whereas what one would like would be some
2468 kind of temporary error. A kludge has been added to help with this.
2469 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
2470 554 error is used if errno is still zero or contains ENOENT (which occurs
2471 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
2474 PH/03 Add -lutil to the default FreeBSD LIBS setting.
2476 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
2477 errors. Otherwise a message that provokes a temporary error (when other
2478 messages do not) can cause a whole host to time out.
2480 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
2481 addresses were routed directly to files or pipes from a redirect router.
2482 File deliveries just didn't batch; pipe deliveries might have suffered
2485 PH/06 A failure to get a lock for a hints database would erroneously always say
2486 "Failed to get write lock", even when it was really a read lock.
2488 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
2489 of 0600. This has been changed to use the value of the lockfile_mode
2490 option (which defaults to 0600).
2492 PH/08 Applied small patch from the Sieve maintainer.
2494 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
2495 folder from quota calculations, a direct delivery into this folder messed
2496 up the contents of the maildirsize file. This was because the regex was
2497 used only to exclude .Trash (or whatever) when the size of the mailbox
2498 was calculated. There was no check that a delivery was happening into an
2499 excluded directory. This bug has been fixed by ignoring all quota
2500 processing for deliveries into excluded directories.
2502 PH/10 Added the maildirfolder_create_regex option to appendfile.
2508 PH/01 The code for finding all the local interface addresses on a FreeBSD
2509 system running IPv6 was broken. This may well have applied to all BSD
2510 systems, as well as to others that have similar system calls. The broken
2511 code found IPv4 interfaces correctly, but gave incorrect values for the
2512 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
2513 that it would not match correctly against @[] and not recognize the IPv6
2516 PH/02 The ipliteral router was not recognizing addresses of the form user@
2517 [ipv6:....] because it didn't know about the "ipv6:" prefix.
2519 PH/03 Added disable_ipv6.
2521 PH/04 Changed $reply_address to use the raw form of the headers instead of the
2522 decoded form, because it is most often used to construct To: headers
2523 lines in autoreplies, and the decoded form may well be syntactically
2524 invalid. However, $reply_address has leading white space removed, and all
2525 newlines turned into spaces so that the autoreply transport does not
2528 PH/05 If group was specified without a user on a router, and no group or user
2529 was specified on a transport, the group from the router was ignored.
2531 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
2532 for visible compile-time settings that can be used to change these
2533 numbers, for those that want even more. Backwards compatibility with old
2534 spool files has been maintained. However, going back to a previous Exim
2535 release will lost any variables that are in spool files.
2537 PH/07 Two small changes when running in the test harness: increase delay when
2538 passing a TCP/IP connection to a new process, in case the original
2539 process has to generate a bounce, and remove special handling of
2540 127.0.0.2 (sic), which is no longer necessary.
2542 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
2543 be the same on different OS.
2545 PH/09 Moved a debug statement in filter processing to avoid a race problem when
2548 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
2549 whether --show-vars was specified or not
2551 JJ/02 exipick: Added support for new ACL variable spool format introduced
2554 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
2555 syntactically invalid From: or Reply-to: line, and a filter used this to
2556 generate an autoreply, and therefore failed to obtain an address for the
2557 autoreply, Exim could try to deliver to a non-existent relative file
2558 name, causing unrelated and misleading errors. What now happens is that
2559 it logs this as a hard delivery error, but does not attempt to create a
2562 PH/11 The exinext utility has a -C option for testing purposes, but although
2563 the given file was scanned by exinext itself; it wasn't being passed on
2564 when Exim was called.
2566 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
2567 an end-of-file indication when reading a command response.
2569 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
2570 compiled. In many other places in Exim, IPv6 addresses are always
2571 recognized, so I have changed this. It also means that IPv4 domain
2572 literals of the form [IPV4:n.n.n.n] are now always recognized.
2574 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
2575 used if the router is not running as root, for example, when verifying at
2576 ACL time, or when using -bh. The debugging output from this situation was
2577 non-existent - all you got was a failure to exec. I have made two
2580 (a) Failures to set uid/gid, the current directory, or a process leader
2581 in a subprocess such as that created by queryprogram now generate
2582 suitable debugging ouput when -d is set.
2584 (b) The queryprogram router detects when it is not running as root,
2585 outputs suitable debugging information if -d is set, and then runs
2586 the subprocess without attempting to change uid/gid.
2588 PH/15 Minor change to Makefile for building test_host (undocumented testing
2591 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
2592 additional section of a DNS packet that returns MX or SRV records.
2593 Instead, it always explicitly searches for A/AAAA records. This avoids
2594 major problems that occur when a DNS server includes only records of one
2595 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
2596 fixed another bug: if SRV records were looked up and the corresponding
2597 address records were *not* found in the additional section, the port
2598 values from the SRV records were lost.
2600 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
2601 using the correct key (the original address) when searching the retry
2602 rules in order to find which one to use for generating the retry hint.
2604 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
2605 adding the default one. Similarly, if it contains a Reply-To: header, the
2606 errors_reply_to option, if set, is not used.
2608 PH/19 When calculating a retry time, Exim used to measure the "time since
2609 failure" by looking at the "first failed" field in the retry record. Now
2610 it does not use this if it is later than than the arrival time of the
2611 message. Instead it uses the arrival time. This makes for better
2612 behaviour in cases where some deliveries succeed, thus re-setting the
2613 "first failed" field. An example is a quota failure for a huge message
2614 when small messages continue to be delivered. Without this change, the
2615 "time since failure" will always be short, possible causing more frequent
2616 delivery attempts for the huge message than are intended.
2617 [Note: This change was subsequently modified - see PH/04 for 4.62.]
2619 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
2620 $1, $2, $3) because the numerical variables can be reset during some
2621 expansion items (e.g. "match"), thereby losing the authentication data.
2623 PH/21 Make -bV show the size of off_t variables so that the test suite can
2624 decide whether to run tests for quotas > 2G.
2626 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
2627 mailbox_size, and mailbox_filecount in the appendfile transport. If a
2628 filecount value is greater than 2G or if a quota value is greater than 2G
2629 on a system where the size of off_t is not greater than 4, a panic error
2632 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
2633 never match. The debug and -bh output now contains an explicit error
2634 message indicating a malformed IPv4 address or mask.
2636 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
2637 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
2638 PH/23 above applies.
2640 PH/25 Do not write to syslog when running in the test harness. The only
2641 occasion when this arises is a failure to open the main or panic logs
2642 (for which there is an explicit test).
2644 PH/26 Added the /no_tell option to "control=freeze".
2646 PH/27 If a host name lookup failed very early in a connection, for example, if
2647 the IP address matched host_lookup and the reverse lookup yielded a name
2648 that did not have a forward lookup, an error message of the form "no IP
2649 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
2650 could be logged. Now it outputs the IP address instead of "NULL".
2652 PH/28 An enabling patch from MH: add new function child_open_exim2() which
2653 allows the sender and the authenticated sender to be set when
2654 submitting a message from within Exim. Since child_open_exim() is
2655 documented for local_scan(), the new function should be too.
2657 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
2658 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
2659 results in an empty string is now treated as unset.
2661 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
2663 PH/31 Added sender_verify_fail logging option.
2665 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
2666 needed by RFC 822 but not by RFC 2822 was commented out. I have now
2667 tidied the source and removed it altogether.
2669 PH/33 When a queue run was abandoned because the load average was too high, a
2670 log line was always written; now it is written only if the queue_run log
2671 selector is set. In addition, the log line for abandonment now contains
2672 information about the queue run such as the pid. This is always present
2673 in "start" and "stop" lines but was omitted from the "abandon" line.
2675 PH/34 Omit spaces between a header name and the colon in the error message that
2676 is given when verify = headers_syntax fails (if there are lots of them,
2677 the message gets confusing).
2679 PH/35 Change the default for dns_check_names_pattern to allow slashes within
2680 names, as there are now some PTR records that contain slashes. This check
2681 is only to protect against broken name servers that fall over on strange
2682 characters, so the fact that it applies to all lookups doesn't matter.
2684 PH/36 Now that the new test suite is complete, we can remove some of the
2685 special code in Exim that was needed for the old test suite. For example,
2686 sorting DNS records because real resolvers return them in an arbitrary
2687 order. The new test suite's fake resolver always returns records in the
2690 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
2691 bounces) except when queue_only is set, to avoid logging races between
2692 the different processes.
2694 PH/38 Panic-die if .include specifies a non-absolute path.
2696 PH/39 A tweak to the "H" retry rule from its user.
2698 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
2699 a label. They prevented compilation on older perls.
2701 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
2702 a warning to be raised on newish perls.
2704 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
2705 on queue. Changes to match documented behaviour of showing count of
2706 messages matching specified criteria.
2708 PH/40 Changed the default ident timeout from 30s to 5s.
2710 PH/41 Added support for the use of login_cap features, on those BSD systems
2711 that have them, for controlling the resources used by pipe deliveries.
2713 PH/42 The content-scanning code uses fopen() to create files in which to put
2714 message data. Previously it was not paying any attention to the mode of
2715 the files. Exim runs with umask(0) because the rest of the code creates
2716 files with open(), and sets the required mode explicitly. Thus, these
2717 files were ending up world-writeable. This was not a big issue, because,
2718 being within the spool directory, they were not world-accessible. I have
2719 created a function called modefopen, which takes an additional mode
2720 argument. It sets umask(777), creates the file, chmods it to the required
2721 mode, then resets the umask. All the relevant calls to fopen() in the
2722 content scanning code have been changed to use this function.
2724 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
2725 to 24 hours. This avoids potential overflow problems when processing G
2726 and H retry rules. I suspect nobody ever tinkers with this value.
2728 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
2730 PH/45 When the plaintext authenticator is running as a client, the server's
2731 challenges are checked to ensure they are valid base64 strings. By
2732 default, the authentication attempt is cancelled if an invalid string is
2733 received. Setting client_ignore_invalid_base64 true ignores these errors.
2734 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
2735 they are received. Thus, the responses can be made to depend on the
2736 challenges. If an invalid string is ignored, an empty string is placed in
2739 PH/46 Messages that are created by the autoreply transport now contains a
2740 References: header, in accordance with RFCs 2822 and 3834.
2742 PH/47 Added authenticated_sender_force to the smtp transport.
2744 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
2746 PH/49 Installed latest patch from the Sieve maintainer.
2748 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
2749 was also set, the appendfile transport was unnecessarily scanning a
2750 directory of message files (e.g. for maildir delivery) to find the count
2751 of files (along with the size), even though it did not need this
2752 information. It now does the scan only if it needs to find either the
2753 size of the count of files.
2755 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
2757 PH/52 Two bugs concerned with error handling when the smtp transport is
2760 (i) Exim was not creating retry information for temporary errors given
2761 for individual recipients after the DATA command when the smtp transport
2762 was used in LMTP mode. This meant that they could be retried too
2763 frequently, and not timed out correctly.
2765 (ii) Exim was setting the flag that allows error details to be returned
2766 for LMTP errors on RCPT commands, but not for LMTP errors for individual
2767 recipients that were returned after the DATA command.
2769 PH/53 This is related to PH/52, but is more general: for any failing address,
2770 when detailed error information was permitted to be returned to the
2771 sender, but the error was temporary, then after the final timeout, only
2772 "retry timeout exceeded" was returned. Now it returns the full error as
2773 well as "retry timeout exceeded".
2775 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
2776 do this, and (what is worse) MTAs that accept it.
2778 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
2779 will now be deprecated.
2781 PH/56 New os.c-cygwin from the Cygwin maintainer.
2783 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
2784 formats (previously only available in exim formats via -bpr, -bpru,
2785 and -bpra. Now also available in native and exiqgrep formats)
2787 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
2788 with very large, slow to parse queues
2790 JJ/08 exipick: added ! as generic prefix to negate any criteria format
2792 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
2794 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
2795 responses to authentication challenges, though it was showing the
2796 challenges; (ii) I've removed the CR characters from the debug output for
2799 PH/58 Allow for the insertion of a newline as well as a space when a string
2800 is turned into more than one encoded-word during RFC 2047 encoding. The
2801 Sieve code now uses this.
2803 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
2804 data_4xx, lost_connection, tls_required.
2806 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
2807 message was being sent as an SMTP response.
2809 PH/61 Add -l and -k options to exicyclog.
2811 PH/62 When verifying, if an address was redirected to one new address, so that
2812 verification continued, and the new address failed or deferred after
2813 having set something in $address_data, the value of $address_data was not
2814 passed back to the ACL. This was different to the case when no
2815 redirection occurred. The value is now passed back in both cases.
2817 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
2818 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
2819 use login_cap.h, so on its own it isn't the distinguishing feature. The
2820 new name refers directly to the setclassresources() function.
2822 PH/65 Added configuration files for NetBSD3.
2824 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
2826 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
2827 is preferred over IPv4.
2829 PH/68 The bounce_return_message and bounce_return_body options were not being
2830 honoured for bounces generated during the reception of non-SMTP messages.
2831 In particular, this applied to messages rejected by the ACL. This bug has
2832 been fixed. However, if bounce_return_message is true and bounce_return_
2833 body is false, the headers that are returned for a non-SMTP message
2834 include only those that have been read before the error was detected.
2835 (In the case of an ACL rejection, they have all been read.)
2837 PH/69 The HTML version of the specification is now built in a directory called
2838 spec_html instead of spec.html, because the latter looks like a path with
2839 a MIME-type, and this confuses some software.
2841 PH/70 Catch two compiler warnings in sieve.c.
2843 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
2844 function verify_get_ident() calls ip_connect() to connect a socket, but
2845 if the "connect()" function timed out, ip_connect() used to close the
2846 socket. However, verify_get_ident() also closes the socket later, and in
2847 between Exim writes to the log, which may get opened at this point. When
2848 the socket was closed in ip_connect(), the log could get the same file
2849 descriptor number as the socket. This naturally causes chaos. The fix is
2850 not to close the socket in ip_connect(); the socket should be closed by
2851 the function that creates it. There was only one place in the code where
2852 this was missing, in the iplookup router, which I don't think anybody now
2853 uses, but I've fixed it anyway.
2855 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
2856 well as to direct DNS lookups. Otherwise the handling of names in host
2857 lists is inconsistent and therefore confusing.
2863 PH/01 Two changes to the default runtime configuration:
2865 (1) Move the checks for relay_from_hosts and authenticated clients from
2866 after to before the (commented out) DNS black list checks.
2868 (2) Add control=submission to the relay_from_hosts and authenticated
2869 clients checks, on the grounds that messages accepted by these
2870 statements are most likely to be submissions.
2872 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
2874 (1) Generate an error if the third argument for the ${prvs expansion is
2877 (2) Treat a missing third argument of ${prvscheck as if it were an empty
2880 (3) Reset the variables that are obtained from the first argument of
2881 ${prvscheck and used in the second argument before leaving the code,
2882 because their memory is reclaimed, so using them afterwards may do
2885 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
2886 one (it's much easier than Tom thought :-).
2888 (5) Because of (4), we can now allow for the use of $prvscheck_result
2889 inside the third argument.
2891 PH/03 For some reason, the default setting of PATH when running a command from
2892 a pipe transport was just "/usr/bin". I have changed it to
2895 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
2896 anything to be listed in the output from -bV.
2898 PH/05 When a filter generated an autoreply, the entire To: header line was
2899 quoted in the delivery log line, like this:
2901 => >A.N.Other <ano@some.domain> <original@ddress> ...
2903 This has been changed so that it extracts the operative address. There
2904 may be more than one such address. If so, they are comma-separated, like
2907 => >ano@some.domain,ona@other.domain <original@ddress> ...
2909 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
2910 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
2911 not being looked up in the rDNS to get a host name, Exim was showing the
2912 IP address twice in Received: lines, even though the IP addresses were
2913 identical. For example:
2915 Received: from [1.2.3.4] (helo=[1.2.3.4])
2917 However, if the real host name was known, it was omitting the HELO data
2918 if it matched the actual IP address. This has been tidied up so that it
2919 doesn't show the same IP address twice.
2921 PH/07 When both +timestamp and +memory debugging was on, the value given by
2922 $tod_xxx expansions could be wrong, because the tod_stamp() function was
2923 called by the debug printing, thereby overwriting the timestamp buffer.
2924 Debugging no longer uses the tod_stamp() function when +timestamp is set.
2926 PH/08 When the original message was included in an autoreply transport, it
2927 always said "this is a copy of the message, including all the headers",
2928 even if body_only or headers_only was set. It now gives an appropriate
2931 PH/09 Applied a patch from the Sieve maintainer which:
2933 o fixes some comments
2934 o adds the (disabled) notify extension core
2935 o adds some debug output for the result of if/elsif tests
2936 o points to the current vacation draft in the documentation
2937 and documents the missing references header update
2941 o fixes a bug in processing the envelope test (when testing
2942 multiple envelope elements, the last element determinted the
2945 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
2946 Electronic Mail") by including:
2948 Auto-submitted: auto-generated
2950 in the messages that it generates (bounce messages and others, such as
2951 warnings). In the case of bounce messages for non-SMTP mesages, there was
2952 also a typo: it was using "Auto_submitted" (underscore instead of
2953 hyphen). Since every message generated by Exim is necessarily in response
2954 to another message, thes have all been changed to:
2956 Auto-Submitted: auto-replied
2958 in accordance with these statements in the RFC:
2960 The auto-replied keyword:
2962 - SHOULD be used on messages sent in direct response to another
2963 message by an automatic process,
2965 - MUST NOT be used on manually-generated messages,
2967 - MAY be used on Delivery Status Notifications (DSNs) and Message
2968 Disposition Notifications (MDNs),
2970 - MUST NOT be used on messages generated by automatic or periodic
2971 processes, except for messages which are automatic responses to
2974 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
2975 to the default Received: header definition.
2977 PH/12 Added log selector acl_warn_skipped (default on).
2979 PH/13 After a successful wildlsearch lookup, discard the values of numeric
2980 variables because (a) they are in the wrong storage pool and (b) even if
2981 they were copied, it wouldn't work properly because of the caching.
2983 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
2984 checking when decoding. Apparently there are clients that generate
2985 overlong encoded strings. Why am I not surprised?
2987 PH/15 If the first argument of "${if match_address" was not empty, but did not
2988 contain an "@" character, Exim crashed. Now it writes a panic log message
2989 and treats the condition as false.
2991 PH/16 In autoreply, treat an empty string for "once" the same as unset.
2993 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
2994 extension "envelope-auth". The code is finished and in agreement with
2995 other implementations, but there is no documentation so far and in fact,
2996 nobody wrote the draft yet. This extension is currently #undef'ed, thus
2997 not changing the active code.
2999 Print executed "if" and "elsif" statements when debugging is used. This
3000 helps a great deal to understand what a filter does.
3002 Document more things not specified clearly in RFC3028. I had all this
3003 sorted out, when out of a sudden new issues came to my mind. Oops."
3005 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
3008 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
3009 canonical form (as documented). However, after a host name lookup from
3010 the IP address, check_host() was doing a simple string comparison with
3011 addresses acquired from the DNS when checking that the found name did
3012 have the original IP as one of its addresses. Since any found IPv6
3013 addresses are likely to be in abbreviated form, the comparison could
3014 fail. Luckily, there already exists a function for doing the comparison
3015 by converting both addresses to binary, so now that is used instead of
3016 the text comparison.
3018 PH/20 There was another similar case to PH/19, when a complete host name was
3019 given in a host list; looking up its IP address could give an abbreviated
3020 form, whereas the current host's name might or might not be abbreviated.
3021 The same fix has been applied.
3027 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
3028 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
3031 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
3032 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
3034 PH/03 Typo: missing ".o" in src/pcre/Makefile.
3036 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
3037 header line, restrict the check to what is listed in RFCs 2369 and 2929.
3038 Also, for "Auto-Submitted", treat anything other than "no" as
3039 non-personal, in accordance with RFC 3834. (Previously it treated
3040 anything starting "auto-" as non-personal.)
3042 TF/01 The control=submission/name=... option had a problem with syntax
3043 errors if the name included a slash character. The /name= option
3044 now slurps the rest of the string, so it can include any characters
3045 but it must come last in the list of options (after /sender_retain
3048 PH/05 Some modifications to the interface to the fake nameserver for the new
3056 TK/01 Added the "success_on_redirect" address verification option. See
3057 NewStuff for rationale and an example.
3059 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
3061 PH/02 Patch to exigrep to allow it to work on syslog lines.
3063 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
3064 fread() to skip over the body file's header line, because in Cygwin the
3065 header line is locked and is inaccessible.
3067 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
3068 co-exist for some time) to make it clear that it is the Exim ID that is
3069 referenced, not the Message-ID: header line.
3071 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
3072 string_format() function, because snprintf() does not exist on all
3075 PH/06 The use of forbid_filter_existstest now also locks out the use of the
3076 ${stat: expansion item.
3078 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
3079 protocol synchronization error", to keep the pedants happy.
3081 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
3082 well as for IRIX systems, when gcc is being used. See the host.c source
3085 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
3087 PH/10 Named domain lists were not working if used in a queue_smtp_domains
3090 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
3091 transport and to the smtp transport in LMTP mode.
3093 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
3095 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
3096 run a filter in a subprocess. This could lead to confusion in subsequent
3097 lookups in the parent process. There should also be a search_tidyup() at
3098 the end of the subprocess.
3100 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
3101 only if the host matched helo_try_verify_hosts, which caused the
3102 verification to occur when the EHLO/HELO command was issued. The ACL just
3103 tested the remembered result. Now, if a previous verification attempt has
3104 not happened, "verify = helo" does it there and then.
3106 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
3108 TK/03 Fix log output including CR from clamd.
3110 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
3111 exist provoked a memory error which could cause a segfault.
3113 PH/15 Installed PCRE 6.2
3115 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
3117 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
3118 of the problem. Specifically, suggested +O2 rather than +O1 for the
3121 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
3123 PH/20 If a delivery was routed to a non-standard port by means of an SRV
3124 record, the port was not correctly logged when the outgoing_port log
3125 selector was set (it logged the transort's default port).
3127 PH/21 Added support for host-specific ports to manualroute, queryprogram,
3128 fallback_hosts, and "hosts" in the smtp transport.
3130 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
3131 host errors such as "Connection refused".
3133 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
3134 authentication with radiusclient 0.4.9:
3136 - Error returned from rc_read_config was caught wrongly
3137 - Username/password not passed on to radius server due to wrong length.
3139 The presumption is that some radiusclient API changes for 4.51/PH/17
3140 were not taken care of correctly. The code is still untested by me (my
3141 Linux distribution still has 0.3.2 of radiusclient), but it was
3142 contributed by a Radius user.
3144 PH/24 When doing a callout, the value of $domain wasn't set correctly when
3145 expanding the "port" option of the smtp transport.
3147 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
3148 while reading a MIME header. Thanks to Tom Hughes for a patch.
3150 PH/24 Include config.h inside local_scan.h so that configuration settings are
3153 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
3154 that in an ACL for RCPT (for example), you can examine exactly what was
3157 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
3158 commands, but it was not correctly comparing the address with the actual
3159 client host address. Thus, it would show the EHLO address in Received:
3160 header lines when this was not necessary.
3162 PH/27 Added the % operator to ${eval:}.
3164 PH/28 Exim tries to create and chdir to its spool directory when it starts;
3165 it should be ignoring failures (because with -C, for example, it has lost
3166 privilege). It wasn't ignoring creation failures other than "already
3169 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
3172 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
3173 because an input line was too long, either on its own, or by virtue of
3174 too many continuations, the temporary file was not being removed, and the
3175 return code was incorrect.
3177 PH/31 Missing "BOOL" in function definition in filtertest.c.
3179 PH/32 Applied Sieve patches from the maintainer.
3181 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
3183 PH/33 Added "verify = not_blind".
3185 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
3186 Local/Makefile (with some defaults set). These are used in built scripts
3187 such as exicyclog, but they have never been used in the exim_install
3188 script (though there are many overriding facilities there). I have
3189 arranged that the exim_install script now takes note of these two
3192 PH/35 Installed configuration files for Dragonfly.
3194 PH/36 When a locally submitted message by a trusted user did not contain a
3195 From: header, and the sender address was obtained from -f or from an SMTP
3196 MAIL command, and the trusted user did not use -F to supply a sender
3197 name, $originator_name was incorrectly used when constructing a From:
3198 header. Furthermore, $originator_name was used for submission mode
3199 messages from external hosts without From: headers in a similar way,
3200 which is clearly wrong.
3202 PH/37 Added control=suppress_local_fixups.
3204 PH/38 When log_selector = +received_sender was set, and the addition of the
3205 sender made the log line's construction buffer exactly full, or one byte
3206 less than full, an overflow happened when the terminating "\n" was
3209 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
3210 when the result of a list match is failure because a DNS lookup failed.
3212 PH/40 RM_COMMAND is now used in the building process.
3214 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
3215 the "build-* directories that it finds.
3217 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
3218 address in a domain literal was a prefix of an interface address.
3220 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
3221 when verifying a sender address, unless rewrite_headers is false.
3223 PH/44 Wrote a long comment about why errors_to addresses are verified as
3224 recipients, not senders.
3226 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
3227 the ratelimit ACL was added.
3229 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
3231 PH/46 Added extra information about PostgreSQL errors to the error string.
3233 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
3234 avoiding the need to install special zones in a real server. This is
3235 backwards compatible; if it can't find the fake resolver, it drops back.
3236 Thus, both old and new test suites can be run.
3238 TF/02 Added util/ratelimit.pl
3240 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
3241 clock is set back in time.
3243 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
3244 Candler <B.Candler@pobox.com>.
3246 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
3247 OK for addresses that are the result of redirections.
3249 PH/48 A number of further additions for the benefit of the new test suite,
3250 including a fake gethostbyname() that interfaces to the fake DNS resolver
3253 TF/06 The fix for widen_domains has also been applied to qualify_single and
3254 search_parents which are the other dnslookup options that can cause
3257 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
3260 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
3262 TF/07 Exim produced the error message "an SRV record indicated no SMTP
3263 service" if it encountered an MX record with an empty target hostname.
3264 The message is now "an MX or SRV record indicated no SMTP service".
3266 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
3267 if the DNS of the sending site is misconfigured. This is quite a
3268 common situation. This change restores the behaviour of treating a
3269 helo verification defer as a failure.
3271 PH/51 If self=fail was set on a router, the bounce message did not include the
3272 actual error message.
3278 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
3280 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
3281 command itself ended in error, the underlying message about the transport
3282 filter timeout was being overwritten with the pipe command error. Now the
3283 underlying error message should be appended to the second error message.
3285 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
3287 PH/02 Reduce the amount of output that "make" produces by default. Full output
3288 can still be requested.
3290 PH/03 The warning log line about a condition test deferring for a "warn" verb
3291 was being output only once per connection, rather than after each
3292 occurrence (because it was using the same function as for successful
3293 "warn" verbs). This seems wrong, so I have changed it.
3295 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
3296 it should not have, which might have caused a crash in the right
3297 circumstances, but probably never did.
3299 PH/04 Installed a modified version of Tony Finch's patch to make submission
3300 mode fix the return path as well as the Sender: header line, and to
3301 add a /name= option so that you can make the user's friendly name appear
3304 TF/03 Added the control = fakedefer ACL modifier.
3306 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
3307 Mark Lowes for thorough testing.
3309 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
3311 TK/03 Merged latest SRS patch from Miles Wilton.
3313 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
3314 with the definition in sysexits.h (which is #included earlier).
3315 Fortunately, Exim does not actually use EX_OK. The code used to try to
3316 preserve the sysexits.h value, by assumimg that macro definitions were
3317 scanned for macro replacements. I have been disabused of this notion,
3318 so now the code just undefines EX_OK before #including unistd.h.
3320 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
3321 in the smtp transport. When a block could not be written in a single
3322 write() function, the timeout was being re-applied to each part-write.
3323 This seems wrong - if the receiver was accepting one byte at a time it
3324 would take for ever. The timeout is now adjusted when this happens. It
3325 doesn't have to be particularly precise.
3327 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
3328 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
3330 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
3331 without a domain if the check to <postmaster@domain> fails.
3333 SC/01 Eximstats: added -xls and the ability to specify output files
3334 (patch written by Frank Heydlauf).
3336 SC/02 Eximstats: use FileHandles for outputing results.
3338 SC/03 Eximstats: allow any combination of xls, txt, and html output.
3340 SC/04 Eximstats: fixed display of large numbers with -nvr option
3342 SC/05 Eximstats: fixed merging of reports with empty tables.
3344 SC/06 Eximstats: added the -include_original_destination flag
3346 SC/07 Eximstats: removed tabs and trailing whitespace.
3348 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
3350 TK/06 MBOX spool code: Add real "From " MBOX separator line
3351 so the .eml file is really in mbox format (even though
3352 most programs do not really care). Patch from Alex Miller.
3354 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
3355 The latter is generated from $received_to and is only set if the
3356 message has one envelope recipient. SA can use these headers,
3357 obviously out-of-the-box. Patch from Alex Miller.
3359 PH/08 The ${def test on a variable was returning false if the variable's
3360 value was "0", contrary to what the specification has always said!
3361 The result should be true unless the variable is empty.
3363 PH/09 The syntax error of a character other than { following "${if
3364 def:variable_name" (after optional whitespace) was not being diagnosed.
3365 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
3366 accidental colon was present, for example, could give incorrect results.
3368 PH/10 Tidied the code in a number of places where the st_size field of a stat()
3369 result is used (not including appendfile, where other changes are about
3372 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
3373 This involved changing a lot of size variables from int to off_t. It
3374 should work with maildirs and everything.
3376 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
3377 spamd dying while we are connected to it.
3379 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
3382 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
3383 function for BDB 4.3.
3385 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
3387 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
3388 the sources for PCRE have been re-arranged and more files are now
3391 PH/15 The code I had for printing potentially long long variables in PH/11
3392 above was not the best (it lost precision). The length of off_t variables
3393 is now inspected at build time, and an appropriate printing format (%ld
3394 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
3395 to be "long long int" or "long int". This is needed for the internal
3396 formatting function string_vformat().
3398 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
3399 the configuration file to be ":syslog", then the script "guesses" where
3400 the logs files are, rather than using the compiled in default. In our
3401 case the guess is not the same as the compiled default, so the script
3402 suddenly stopped working when I started to use syslog. The patch checks
3403 to see if log_file_path is "". If so, it attempts to read it from exim
3404 with no configuration file to get the compiled in version, before it
3405 falls back to the previous guessing code."
3407 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
3408 implementing BATV in an Exim configuration. See NewStuff for the gory
3411 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
3412 Makefile that are specific to HP-UX.
3414 PH/18 If the "use_postmaster" option was set for a recipient callout together
3415 with the "random" option, the postmaster address was used as the MAIL
3416 FROM address for the random test, but not for the subsequent recipient
3417 test. It is now used for both.
3419 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
3420 patch removes a few documentation additions to RFC 3028, because the
3421 latest draft now contains them. It adds the new en;ascii-case comparator
3422 and a new error check for 8bit text in MIME parts. Comparator and
3423 require names are now matched exactly. I enabled the subaddress
3424 extension, but it is not well tested yet (read: it works for me)."
3426 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
3427 rework some of the code of TK/09 above to avoid the hardwired use of
3428 "%lld" and "long long". Replaced the call to snprintf() with a call to
3431 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
3432 records point to non-existent hosts", "retry timeout exceeded", and
3433 "retry time not reached for any host after a long failure period".
3435 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
3436 experimental DomainKeys support:
3438 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
3439 (2) On an error such as an illegally used "control", the wrong name for
3440 the control was given.
3442 These problems did NOT occur unless DomainKeys support was compiled.
3444 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
3446 PH/24 Added ${if match_ip condition.
3448 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
3449 databases so that it will be absolutely obvious if a crash occurs in the
3450 DB library. This is a regular occurrence (often caused by mis-matched
3453 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
3454 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
3455 picked up on a user's system that detects such things. There doesn't seem
3456 to be a gcc warning option for this - only an attribute that has to be
3457 put on the function's prototype. It seems that in Fedora Core 4 they have
3458 set this on a number of new functions. No doubt there will be more in due
3461 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
3462 specify a transport. However, if an address that was verified by such a
3463 router was the subject of a callout, Exim crashed because it tried to
3464 read the rcpt_include_affixes from the non-existent transport. Now it
3465 just assumes that the setting of that option is false. This bug was
3466 introduced by 4.51/PH/31.
3468 PH/28 Changed -d+all to exclude +memory, because that information is very
3469 rarely of interest, but it makes the output a lot bigger. People tend to
3470 do -d+all out of habit.
3472 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
3473 code in os-type was giving problems when libc.so lives in lib64, like on
3476 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
3477 aren't the modern standard, and it seems that some systems' include files
3478 don't always have them. Exim was already checking for some of the newer
3479 ones like T_AAAA, and defining it itself. I've added checks for all the
3480 record types that Exim uses.
3482 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
3483 not automatically generating a new one, as it is supposed to. This
3484 prevented TLS from working. If the file did exist, but contained invalid
3485 data, a new version was generated, as expected. It was only the case of a
3486 non-existent file that was broken.
3488 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
3489 with a change in libdomainkeys > 0.64.
3491 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
3492 from DNS. If the selector record carries the flag, it now has
3493 precedence over the domain-wide flag.
3495 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
3497 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
3498 the use of a port name that isn't defined in /etc/services) occurred, the
3499 message was deferred as in a normal delivery, and thus remained on the
3500 spool, instead of being failed because of the mua_wrapper setting. This
3501 is now fixed, and I tidied up some of the mua_wrapper messages at the
3504 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
3505 the messages in a hash of arrays rather than using individual hashes.
3506 This is a bit cleaner and results in dramatic memory savings, albeit
3507 at a slight CPU cost.
3509 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
3510 as requested by Marc Sherman.
3512 SC/10 Eximstats: added histograms for user specified patterns as requested
3515 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
3517 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
3518 fopen() in the content-scanning modules that did not already have it.
3524 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
3525 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
3527 TK/02 Fix ACL "control" statement not being available in MIME ACL.
3529 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
3531 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
3532 to test Sieve filters that use "vacation".
3534 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
3535 that changes the way the GnuTLS parameters are stored in the cache file.
3536 The new format can be generated externally. For backward compatibility,
3537 if the data in the cache doesn't make sense, Exim assumes it has read an
3538 old-format file, and it generates new data and writes a new file. This
3539 means that you can't go back to an older release without removing the
3542 PH/03 A redirect router that has both "unseen" and "one_time" set does not
3543 work if there are any delivery delays because "one_time" forces the
3544 parent to be marked "delivered", so its unseen clone is never tried
3545 again. For this reason, Exim now forbids the simultaneous setting of
3548 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
3549 redirected to themselves ("homonym" addresses). Read the long ChangeLog
3550 entry if you want to know the details. The fix, however, neglected to
3551 consider the case when local delivery batching is involved. The test for
3552 "previously delivered" was not happening when checking to see if an
3553 address could be batched with a previous (undelivered) one; under
3554 certain circumstances this could lead to multiple deliveries to the same
3557 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
3558 in its include files, and this causes problems building Exim.
3560 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
3561 header_syntax) but Exim was just ignoring anything given after a slash.
3562 In particular, this caused confusion with an attempt to use "verify =
3563 reverse_host_lookup/defer_ok". An error is now given when options are
3564 supplied for verify items that do not have them. (Maybe reverse_host_
3565 lookup should have a defer_ok option, but that's a different point.)
3567 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
3568 defined by RFC 821) to 2048, because there were problems with some AUTH
3569 commands, and RFC 1869 says the size should be increased for extended
3570 SMTP commands that take arguments.
3572 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
3575 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
3576 "unknown" error; now it says that the functionality isn't in the binary.
3578 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
3579 an address' error message when a string expansion fails (syntax or
3580 whatever). Otherwise the password may appear in the log. Following change
3581 PH/42 below, there is no longer a chance of it appearing in a bounce
3584 PH/11 Installed exipick version 20050225.0 from John Jetmore.
3586 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
3587 of its addresses was ever tried. (Bugzilla bug #2.)
3589 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
3590 the result incorrectly in the debug output. (It correctly added a newline
3591 to what was transported.)
3593 TF/01 Added $received_time.
3595 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
3596 commented out examples of how to interface to a virus scanner and to
3597 SpamAssassin. Also added commented examples of av_scanner and
3598 spamd_address settings.
3600 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
3601 and controls are allowed in which ACLs. There were a couple of minor
3602 errors. Some of the entries in the conditions table (which is a table of
3603 where they are NOT allowed) were getting very unwieldy; rewrote them as a
3604 negation of where the condition IS allowed.
3606 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
3608 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
3609 header file does not have a version number, so I've had to invent a new
3610 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
3611 API. The code is untested by me (my Linux distribution still has 0.3.2 of
3612 radiusclient), but it was contributed by a Radius user.
3614 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
3615 files or directories, for OpenSSL.
3617 PH/19 When an Exim process that is running as root has to create an Exim log
3618 file, it does so in a subprocess that runs as exim:exim so as to get the
3619 ownership right at creation (otherwise, other Exim processes might see
3620 the file with the wrong ownership). There was no test for failure of this
3621 fork() call, which would lead to the process getting stuck as it waited
3622 for a non-existent subprocess. Forks do occasionally fail when resources
3623 run out. I reviewed all the other calls to fork(); they all seem to check
3626 PH/20 When checking for unexpected SMTP input at connect time (before writing
3627 the banner), Exim was not dealing correctly with a non-positive return
3628 from the read() function. If the client had disconnected by this time,
3629 the result was a log entry for a synchronization error with an empty
3630 string after "input=" when read() returned zero. If read() returned -1
3631 (an event I could not check), uninitialized data bytes were printed.
3632 There were reports of junk text (parts of files, etc) appearing after
3635 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
3637 PH/22 Added support for macro redefinition, and (re)definition in between
3638 driver and ACL definitions.
3640 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
3641 forgetting to use the resulting value; it was using the unexpanded value.
3643 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
3644 hadn't been configured. The fix is from Juergen Kreileder, who
3645 understands it better than I do:
3647 "Here's what I see happening with three configured cyrus_sasl
3648 authenticators configured (plain, login, cram-md5):
3650 On startup auth_cyrus_sasl_init() gets called for each of these.
3651 This means three calls to sasl_listmech() without a specified mech_list.
3652 => SASL tests which mechs of all available mechs actually work
3653 => three warnings about OTP not working
3654 => the returned list contains: plain, login, cram-md5, digest-md5, ...
3656 With the patch, sasl_listmech() also gets called three times. But now
3657 SASL's mech_list option is set to the server_mech specified in the the
3658 authenticator. Or in other words, the answer from sasl_listmech()
3659 gets limited to just the mech you're testing for (which is different
3661 => the return list contains just 'plain' or 'login', 'cram-md5' or
3662 nothing depending on the value of ob->server_mech.
3664 I've just tested the patch: Authentication still works fine,
3665 unavailable mechs specified in the exim configuration are still
3666 caught, and the auth.log warnings about OTP are gone."
3668 PH/25 When debugging is enabled, the contents of the command line are added
3669 to the debugging output, even when log_selector=+arguments is not
3672 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
3673 answer is "GNU", and only if the return is "GNU/something" is the answer
3676 PH/27 $acl_verify_message is now set immediately after the failure of a
3677 verification in an ACL, and so is available in subsequent modifiers. In
3678 particular, the message can be preserved by coding like this:
3680 warn !verify = sender
3681 set acl_m0 = $acl_verify_message
3683 Previously, $acl_verify_message was set only while expanding "message"
3684 and "log_message" when a very denied access.
3686 PH/28 Modified OS/os.c-Linux with
3688 -#ifndef OS_LOAD_AVERAGE
3689 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
3691 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
3692 nomenclature these days.)
3694 PH/29 Installed patch from the Sieve maintainer that adds the options
3695 sieve_useraddress and sieve_subaddress to the redirect router.
3697 PH/30 In these circumstances:
3698 . Two addresses routed to the same list of hosts;
3699 . First host does not offer TLS;
3700 . First host accepts first address;
3701 . First host gives temporary error to second address;
3702 . Second host offers TLS and a TLS session is established;
3703 . Second host accepts second address.
3704 Exim incorrectly logged both deliveries with the TLS parameters (cipher
3705 and peerdn, if requested) that were in fact used only for the second
3708 PH/31 When doing a callout as part of verifying an address, Exim was not paying
3709 attention to any local part prefix or suffix that was matched by the
3710 router that accepted the address. It now behaves in the same way as it
3711 does for delivery: the affixes are removed from the local part unless
3712 rcpt_include_affixes is set on the transport.
3714 PH/32 Add the sender address, as F=<...>, to the log line when logging a
3715 timeout during the DATA phase of an incoming message.
3717 PH/33 Sieve envelope tests were broken for match types other than :is. I have
3718 applied a patch sanctioned by the Sieve maintainer.
3720 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
3721 the uid or gid is negative. A case of a negative gid caused this to be
3722 noticed. The fix allows for either to be negative.
3724 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
3725 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
3728 PH/36 The change PH/12 above was broken. Fixed it.
3730 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
3731 the grounds that routing the same address twice would always produce the
3732 same answer. This might have been true once, but it is certainly no
3733 longer true now. Routing a child address may depend on the previous
3734 routing that produced that child. Some complicated redirection strategies
3735 went wrong when messages had multiple recipients, and made Exim's
3736 behaviour dependent on the order in which the addresses were given.
3738 I have moved the duplicate checking until after the routing is complete.
3739 Exim scans the addresses that are assigned to local and remote
3740 transports, and removes any duplicates. This means that more work will be
3741 done, as duplicates will always all be routed, but duplicates are
3742 presumably rare, so I don't expect this is of any significance.
3744 For deliveries to pipes, files, and autoreplies, the duplicate checking
3745 still happens during the routing process, since they are not going to be
3748 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
3749 It corrects a timeout issue with spamd. This is Ian's comment: "The
3750 background is that sometimes spamd either never reads data from a
3751 connection it has accepted, or it never writes response data. The exiscan
3752 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
3753 blindly assumes that writes won't block so it may never time out."
3755 PH/39 Allow G after quota size as well as K and M.
3757 PH/40 The value set for $authenticated_id in an authenticator may not contain
3758 binary zeroes or newlines because the value is written to log lines and
3759 to spool files. There was no check on this. Now the value is run through
3760 the string_printing() function so that such characters are converted to
3761 printable escape sequences.
3763 PH/41 $message_linecount is a new variable that contains the total number of
3764 lines in the message. Compare $body_linecount, which is the count for the
3767 PH/42 Exim no longer gives details of delivery errors for specific addresses in
3768 bounce and delay warning messages, except in certain special cases, which
3771 (a) An SMTP error message from a remote host;
3772 (b) A message specified in a :fail: redirection;
3773 (c) A message specified in a "fail" command in a system filter;
3774 (d) A message specified in a FAIL return from the queryprogram router;
3775 (e) A message specified by the cannot_route_message router option.
3777 In these cases only, Exim does include the error details in bounce and
3778 warning messages. There are also a few cases where bland messages such
3779 as "unrouteable address" or "local delivery error" are given.
3781 PH/43 $value is now also set for the "else" part of a ${run expansion.
3783 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
3784 being worked on, but at least Exim now implements the latest version to
3787 PH/45 In a pipe transport, although a timeout while waiting for the pipe
3788 process to complete was treated as a delivery failure, a timeout while
3789 writing the message to the pipe was logged, but erroneously treated as a
3790 successful delivery. Such timeouts include transport filter timeouts. For
3791 consistency with the overall process timeout, these timeouts are now
3792 treated as errors, giving rise to delivery failures by default. However,
3793 there is now a new Boolean option for the pipe transport called
3794 timeout_defer, which, if set TRUE, converts the failures into defers for
3795 both kinds of timeout. A transport filter timeout is now identified in
3798 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
3799 systems where "make" and "gmake" are different, calling "gmake" at top
3800 level broke things. I've arranged for the value of $(MAKE) to be passed
3801 from the Makefile to this script so that it can call the same version of
3805 A note about Exim versions 4.44 and 4.50
3806 ----------------------------------------
3808 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
3809 changes of various kinds. As a consequence, a big documentation update was
3810 needed. This delayed the release for rather longer than seemed good, especially
3811 in the light of a couple of (minor) security issues. Therefore, the changes
3812 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
3813 release. So 4.44 and 4.50 are in effect two different branches that both start
3816 I have left the 4.50 change log unchanged; it contains all the changes since
3817 4.43. The change log for 4.44 is below; many of its items are identical to
3818 those for 4.50. This seems to be the most sensible way to preserve the
3819 historical information.
3825 1. Minor wording change to the doc/README.SIEVE file.
3827 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
3828 computation of the current number of files was incorrect.
3830 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
3831 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
3834 4. Give more explanation in the error message when the command for a transport
3835 filter fails to execute.
3837 5. There are several places where Exim runs a non-Exim command in a
3838 subprocess. The SIGUSR1 signal should be disabled for these processes. This
3839 was being done only for the command run by the queryprogram router. It is
3840 now done for all such subprocesses. The other cases are: ${run, transport
3841 filters, and the commands run by the lmtp and pipe transports.
3843 6. Added CONFIGURE_GROUP build-time option.
3845 7. Some older OS have a limit of 256 on the maximum number of file
3846 descriptors. Exim was using setrlimit() to set 1000 as a large value
3847 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
3848 systems. I've change it so that if it can't get 1000, it tries for 256.
3850 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
3851 was an oversight, and furthermore, ever since the addition of extra
3852 controls (e.g. 4.43/32), the checks on when to allow different forms of
3853 "control" were broken. There should now be diagnostics for all cases when a
3854 control that does not make sense is encountered.
3856 9. Added the /retain_sender option to "control=submission".
3858 10. $recipients is now available in the predata ACL (oversight).
3860 11. Tidy the search cache before the fork to do a delivery from a message
3861 received from the command line. Otherwise the child will trigger a lookup
3862 failure and thereby defer the delivery if it tries to use (for example) a
3863 cached ldap connection that the parent has called unbind on.
3865 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
3866 of $address_data from the recipient verification was clobbered by the
3867 sender verification.
3869 13. The value of address_data from a sender verification is now available in
3870 $sender_address_data in subsequent conditions in the ACL statement.
3872 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
3874 15. Added a new option "connect=<time>" to callout options, to set a different
3877 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
3878 was its contents. (It was OK if the option was not defined at all.)
3880 17. A "Completed" log line is now written for messages that are removed from
3881 the spool by the -Mrm option.
3883 18. New variables $sender_verify_failure and $recipient_verify_failure contain
3884 information about exactly what failed.
3886 19. Added -dd to debug only the daemon process.
3888 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
3889 handles timeouts, both on the server side and network timeouts. Renamed the
3890 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
3892 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
3893 It is now set to "smtps".
3895 22. $host_address is now set to the target address during the checking of
3896 ignore_target_hosts.
3898 23. When checking ignore_target_hosts for an ipliteral router, no host name was
3899 being passed; this would have caused $sender_host_name to have been used if
3900 matching the list had actually called for a host name (not very likely,
3901 since this list is usually IP addresses). A host name is now passed as
3904 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
3905 code that specifies a non-restarting handler (typically sigaction() in
3906 modern systems) in an attempt to fix a rare and obscure crash bug.
3908 25. Narrowed the window for a race in the daemon that could cause it to ignore
3909 SIGCHLD signals. This is not a major problem, because they are used only to
3910 wake it up if nothing else does.
3912 26. A malformed maildirsize file could cause Exim to calculate negative values
3913 for the mailbox size or file count. Odd effects could occur as a result.
3914 The maildirsize information is now recalculated if the size or filecount
3917 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
3918 support for a long time. Removed HAVE_SYS_VFS_H.
3920 28. Installed the latest version of exipick from John Jetmore.
3922 29. In an address list, if the pattern was not a regular expression, an empty
3923 subject address (from a bounce message) matched only if the pattern was an
3924 empty string. Non-empty patterns were not even tested. This was the wrong
3925 because it is perfectly reasonable to use an empty address as part of a
3926 database query. An empty address is now tested by patterns that are
3927 lookups. However, all the other forms of pattern expect the subject to
3928 contain a local part and a domain, and therefore, for them, an empty
3929 address still always fails if the pattern is not itself empty.
3931 30. Exim went into a mad DNS loop when attempting to do a callout where the
3932 host was specified on an smtp transport, and looking it up yielded more
3933 than one IP address.
3935 31. Re-factored the code for checking spool and log partition space into a
3936 function that finds that data and another that does the check. The former
3937 is then used to implement four new variables: $spool_space, $log_space,
3938 $spool_inodes, and $log_inodes.
3940 32. The RFC2047 encoding function was originally intended for short strings
3941 such as real names; it was not keeping to the 75-character limit for
3942 encoded words that the RFC imposes. It now respects the limit, and
3943 generates multiple encoded words if necessary. To be on the safe side, I
3944 have increased the buffer size for the ${rfc2047: expansion operator from
3947 33. It is now permitted to omit both strings after an "if" condition; if the
3948 condition is true, the result is "true". As before, when the second string
3949 is omitted, a false condition yields an empty string. This makes it less
3950 cumbersome to write custom ACL and router conditions.
3952 34. Failure to deliver a bounce message always caused it to be frozen, even if
3953 there was an errors_to setting on the router. The errors_to setting is now
3956 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
3957 canonical form (fully expanded) before being placed in
3958 $sender_host_address.
3960 36. The table in the code that translates DNS record types into text (T_A to
3961 "A" for instance) was missing entries for NS and CNAME. It is just possible
3962 that this could have caused confusion if both these types were looked up
3963 for the same domain, because the text type is used as part of Exim's
3964 per-process caching. But the chance of anyone hitting this buglet seems
3967 37. The dnsdb lookup has been extended in a number of ways.
3969 (1) There is a new type, "zns", which walks up the domain tree until it
3970 finds some nameserver records. It should be used with care.
3972 (2) There is a new type, "mxh", which is like "mx" except that it returns
3973 just the host names, not the priorities.
3975 (3) It is now possible to give a list of domains (or IP addresses) to be
3976 looked up. The behaviour when one of the lookups defers can be
3977 controlled by a keyword.
3979 (4) It is now possible to specify the separator character for use when
3980 multiple records are returned.
3982 38. The dnslists ACL condition has been extended: it is now possible to supply
3983 a list of IP addresses and/or domains to be looked up in a particular DNS
3986 39. Added log_selector=+queue_time_overall.
3988 40. When running the queue in the test harness, wait just a tad after forking a
3989 delivery process, to get repeatability of debugging output.
3991 41. Include certificate and key file names in error message when GnuTLS fails
3992 to set them up, because the GnuTLS error message doesn't include the name
3993 of the failing file when there is a problem reading it.
3995 42. Allow both -bf and -bF in the same test run.
3997 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
3999 44. The "Exiscan patch" is now merged into the mainline Exim source.
4001 45. Sometimes the final signoff response after QUIT could fail to get
4002 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4003 < 0 before doing a fflush(). This bug looks as though it goes back to the
4004 introduction of TLS in release 3.20, but "sometimes" must have been rare
4005 because the tests only now provoked it.
4007 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
4008 (this can affect the format of dates).
4010 47. exim_tidydb, when checking for the continued existence of a message for
4011 which it has found a message-specific retry record, was not finding
4012 messages that were in split spool directories. Consequently, it was
4013 deleting retry records that should have stayed in existence.
4015 48. Steve fixed some bugs in eximstats.
4017 49. The SPA authentication driver was not abandoning authentication and moving
4018 on to the next authenticator when an expansion was forced to fail,
4019 contradicting the general specification for all authenticators. Instead it
4020 was generating a temporary error. It now behaves as specified.
4022 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
4023 (the order specifies the preference for clients). The order is now AES256,
4024 AES128, 3DES, ARCFOUR128.
4026 51. Small patch to Sieve code - explicitly set From: when generating an
4029 52. Exim crashed if a remote delivery caused a very long error message to be
4030 recorded - for instance if somebody sent an entire SpamAssassin report back
4031 as a large number of 550 error lines. This bug was coincidentally fixed by
4032 increasing the size of one of Exim's internal buffers (big_buffer) that
4033 happened as part of the Exiscan merge. However, to be on the safe side, I
4034 have made the code more robust (and fixed the comments that describe what
4037 53. Now that there can be additional text after "Completed" in log lines (if
4038 the queue_time_overall log selector is set), a one-byte patch to exigrep
4039 was needed to allow it to recognize "Completed" as not the last thing in
4042 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
4043 patch that reportedly fixes this has been added. I am not expert enough to
4044 create a test for it. This is what the patch creator wrote:
4046 "I found a little strange behaviour of ldap code when working with
4047 Windows 2003 AD Domain, where users was placed in more than one
4048 Organization Units. When I tried to give exim partial DN, the exit code
4049 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
4050 But simultaneously result of request was absolutely normal ldap result,
4051 so I produce this patch..."
4053 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
4054 REFERENCE, so I have modified the code to exclude the patch when that macro
4057 55. Some experimental protocols are using DNS PTR records for new purposes. The
4058 keys for these records are domain names, not reversed IP addresses. The
4059 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4060 leaves it alone. Component reversal etc. now happens only for IP addresses.
4063 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
4065 57. Double the size of the debug message buffer (to 2048) so that more of very
4066 long debug lines gets shown.
4068 58. The exicyclog utility now does better if the number of log files to keep
4069 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
4071 59. Two changes related to the smtp_active_hostname option:
4073 (1) $smtp_active_hostname is now available as a variable.
4074 (2) The default for smtp_banner uses $smtp_active_hostname instead
4075 of $primary_hostname.
4077 60. The host_aton() function is supposed to be passed a string that is known
4078 to be a valid IP address. However, in the case of IPv6 addresses, it was
4079 not checking this. This is a hostage to fortune. Exim now panics and dies
4080 if the condition is not met. A case was found where this could be provoked
4081 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4082 components; fortuitously, this particular loophole had already been fixed
4083 by change 4.50/55 above.
4085 If there are any other similar loopholes, the new check in host_aton()
4086 itself should stop them being exploited. The report I received stated that
4087 data on the command line could provoke the exploit when Exim was running as
4088 exim, but did not say which command line option was involved. All I could
4089 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4090 running as the user.
4093 61. There was a buffer overflow vulnerability in the SPA authentication code
4094 (which came originally from the Samba project). I have added a test to the
4095 spa_base64_to_bits() function which I hope fixes it.
4098 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
4099 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
4101 63. The daemon start-up calls getloadavg() while still root for those OS that
4102 need the first call to be done as root, but it missed one case: when
4103 deliver_queue_load_max is set with deliver_drop_privilege. This is
4104 necessary for the benefit of the queue runner, because there is no re-exec
4105 when deliver_drop_privilege is set.
4107 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
4108 This has been fixed.
4110 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4111 was in use, was not putting the data itself into the right store pool;
4112 consequently, it could be overwritten for a subsequent message in the same
4113 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4116 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
4118 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
4119 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
4120 the function were treating the return as a boolean value, which happened to
4121 work because 0=false and not-0=true, but is not correct code.
4123 68. The host_aton() function was not handling scoped IPv6 addresses (those
4124 with, for example, "%eth0" on the end) correctly.
4126 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
4127 negated items (that is, ~something) in unsigned ints. Some compilers
4128 apparently mutter when there is no cast.
4130 70. If an address verification called from an ACL failed, and did not produce a
4131 user-specific message (i.e. there was only a "system" message), nothing was
4132 put in $acl_verify_message. In this situation, it now puts the system
4135 71. Change 4.23/11 added synchronization checking at the start of an SMTP
4136 session; change 4.31/43 added the unwanted input to the log line - except
4137 that it did not do this in the start of session case. It now does.
4139 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
4140 This is wrong and can cause the other end to generate a synchronization
4141 error if it is another Exim or anything else that does the synchronization
4142 check. A QUIT command is no longer sent after a timeout.
4144 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
4145 during host lookups.
4147 74. The defer_ok option of callout verification was not working if it was used
4148 when verifying addresses in header lines, that is, for this case:
4150 verify = header_sender/callout=defer_ok
4152 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
4153 those file descriptors could be used for SMTP connections. If anything
4154 wrote to stderr (the example that came up was "warn" in embedded Perl), it
4155 could be sent to the SMTP client, causing chaos. The daemon now opens
4156 stdin, stdout, and stderr to /dev/null when it puts itself into the
4159 76. Arrange for output from Perl's "warn" command to be written to Exim's main
4160 log by default. The user can override this with suitable Perl magic.
4162 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
4163 the log message when discard triggers, was not working for the DATA ACL or
4164 for the non-SMTP ACL.
4166 78. Error message wording change in sieve.c.
4168 79. If smtp_accept_max_per_host was set, the number of connections could be
4169 restricted to fewer than expected, because the daemon was trying to set up
4170 a new connection before checking whether the processes handling previous
4171 connections had finished. The check for completed processes is now done
4172 earlier. On busy systems, this bug wouldn't be noticed because something
4173 else would have woken the daemon, and it would have reaped the completed
4176 80. If a message was submitted locally by a user whose login name contained one
4177 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
4178 It caused a spool format error. I have fixed the spool reading code. A
4179 related problem was that the "from" clause in the Received: line became
4180 illegal because of the space(s). It is now covered by ${quote_local_part.
4182 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
4185 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
4187 83. Patch from Sieve maintainer to fix unterminated string problem in
4188 "vacation" handling.
4190 84. Some minor changes to the Linux configuration files to help with other
4191 OS variants using glibc.
4193 85. One more patch for Sieve to update vacation handling to latest spec.
4196 ----------------------------------------------------
4197 See the note above about the 4.44 and 4.50 releases.
4198 ----------------------------------------------------
4204 1. Change 4.43/35 introduced a bug that caused file counts to be
4205 incorrectly computed when quota_filecount was set in an appendfile
4208 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4209 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4212 3. Give more explanation in the error message when the command for a transport
4213 filter fails to execute.
4215 4. There are several places where Exim runs a non-Exim command in a
4216 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4217 was being done only for the command run by the queryprogram router. It is
4218 now done for all such subprocesses. The other cases are: ${run, transport
4219 filters, and the commands run by the lmtp and pipe transports.
4221 5. Some older OS have a limit of 256 on the maximum number of file
4222 descriptors. Exim was using setrlimit() to set 1000 as a large value
4223 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4224 systems. I've change it so that if it can't get 1000, it tries for 256.
4226 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4227 was an oversight, and furthermore, ever since the addition of extra
4228 controls (e.g. 4.43/32), the checks on when to allow different forms of
4229 "control" were broken. There should now be diagnostics for all cases when a
4230 control that does not make sense is encountered.
4232 7. $recipients is now available in the predata ACL (oversight).
4234 8. Tidy the search cache before the fork to do a delivery from a message
4235 received from the command line. Otherwise the child will trigger a lookup
4236 failure and thereby defer the delivery if it tries to use (for example) a
4237 cached ldap connection that the parent has called unbind on.
4239 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4240 of $address_data from the recipient verification was clobbered by the
4241 sender verification.
4243 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4244 was its contents. (It was OK if the option was not defined at all.)
4246 11. A "Completed" log line is now written for messages that are removed from
4247 the spool by the -Mrm option.
4249 12. $host_address is now set to the target address during the checking of
4250 ignore_target_hosts.
4252 13. When checking ignore_target_hosts for an ipliteral router, no host name was
4253 being passed; this would have caused $sender_host_name to have been used if
4254 matching the list had actually called for a host name (not very likely,
4255 since this list is usually IP addresses). A host name is now passed as
4258 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4259 code that specifies a non-restarting handler (typically sigaction() in
4260 modern systems) in an attempt to fix a rare and obscure crash bug.
4262 15. Narrowed the window for a race in the daemon that could cause it to ignore
4263 SIGCHLD signals. This is not a major problem, because they are used only to
4264 wake it up if nothing else does.
4266 16. A malformed maildirsize file could cause Exim to calculate negative values
4267 for the mailbox size or file count. Odd effects could occur as a result.
4268 The maildirsize information is now recalculated if the size or filecount
4271 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4272 support for a long time. Removed HAVE_SYS_VFS_H.
4274 18. Updated exipick to current release from John Jetmore.
4276 19. Allow an empty sender to be matched against a lookup in an address list.
4277 Previously the only cases considered were a regular expression, or an
4280 20. Exim went into a mad DNS lookup loop when doing a callout where the
4281 host was specified on the transport, if the DNS lookup yielded more than
4284 21. The RFC2047 encoding function was originally intended for short strings
4285 such as real names; it was not keeping to the 75-character limit for
4286 encoded words that the RFC imposes. It now respects the limit, and
4287 generates multiple encoded words if necessary. To be on the safe side, I
4288 have increased the buffer size for the ${rfc2047: expansion operator from
4291 22. Failure to deliver a bounce message always caused it to be frozen, even if
4292 there was an errors_to setting on the router. The errors_to setting is now
4295 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4296 canonical form (fully expanded) before being placed in
4297 $sender_host_address.
4299 24. Updated eximstats to version 1.33
4301 25. Include certificate and key file names in error message when GnuTLS fails
4302 to set them up, because the GnuTLS error message doesn't include the name
4303 of the failing file when there is a problem reading it.
4305 26. Expand error message when OpenSSL has problems setting up cert/key files.
4308 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
4309 (this can affect the format of dates).
4311 28. exim_tidydb, when checking for the continued existence of a message for
4312 which it has found a message-specific retry record, was not finding
4313 messages that were in split spool directories. Consequently, it was
4314 deleting retry records that should have stayed in existence.
4316 29. eximstats updated to version 1.35
4317 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
4318 1.35 - bugfix such that pie charts by volume are generated correctly
4320 30. The SPA authentication driver was not abandoning authentication and moving
4321 on to the next authenticator when an expansion was forced to fail,
4322 contradicting the general specification for all authenticators. Instead it
4323 was generating a temporary error. It now behaves as specified.
4325 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
4326 (the order specifies the preference for clients). The order is now AES256,
4327 AES128, 3DES, ARCFOUR128.
4329 31. Small patch to Sieve code - explicitly set From: when generating an
4332 32. Exim crashed if a remote delivery caused a very long error message to be
4333 recorded - for instance if somebody sent an entire SpamAssassin report back
4334 as a large number of 550 error lines. This bug was coincidentally fixed by
4335 increasing the size of one of Exim's internal buffers (big_buffer) that
4336 happened as part of the Exiscan merge. However, to be on the safe side, I
4337 have made the code more robust (and fixed the comments that describe what
4340 33. Some experimental protocols are using DNS PTR records for new purposes. The
4341 keys for these records are domain names, not reversed IP addresses. The
4342 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4343 leaves it alone. Component reversal etc. now happens only for IP addresses.
4346 34. The host_aton() function is supposed to be passed a string that is known
4347 to be a valid IP address. However, in the case of IPv6 addresses, it was
4348 not checking this. This is a hostage to fortune. Exim now panics and dies
4349 if the condition is not met. A case was found where this could be provoked
4350 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4351 components; fortuitously, this particular loophole had already been fixed
4352 by change 4.50/55 or 4.44/33 above.
4354 If there are any other similar loopholes, the new check in host_aton()
4355 itself should stop them being exploited. The report I received stated that
4356 data on the command line could provoke the exploit when Exim was running as
4357 exim, but did not say which command line option was involved. All I could
4358 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4359 running as the user.
4362 35. There was a buffer overflow vulnerability in the SPA authentication code
4363 (which came originally from the Samba project). I have added a test to the
4364 spa_base64_to_bits() function which I hope fixes it.
4367 36. The daemon start-up calls getloadavg() while still root for those OS that
4368 need the first call to be done as root, but it missed one case: when
4369 deliver_queue_load_max is set with deliver_drop_privilege. This is
4370 necessary for the benefit of the queue runner, because there is no re-exec
4371 when deliver_drop_privilege is set.
4373 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4374 was in use, was not putting the data itself into the right store pool;
4375 consequently, it could be overwritten for a subsequent message in the same
4376 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4379 38. Sometimes the final signoff response after QUIT could fail to get
4380 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4381 < 0 before doing a fflush(). This bug looks as though it goes back to the
4382 introduction of TLS in release 3.20, but "sometimes" must have been rare
4383 because the tests only now provoked it.
4389 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
4390 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
4391 Now it returns an integer. A number of calls were still expecting a T/F
4392 return. Fortuitously, in all cases, the tests worked in OK situations,
4393 which is the norm. However, things would have gone wrong on any write
4394 failures on the smtp file descriptor. This function is used when sending
4395 messages over SMTP and also when doing verify callouts.
4397 2. When Exim is called to do synchronous delivery of a locally submitted
4398 message (the -odf or -odi options), it no longer closes stderr before doing
4401 3. Implemented the mua_wrapper option.
4403 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
4405 5. Implemented the functions header_remove(), header_testname(),
4406 header_add_at_position(), and receive_remove_recipient(), and exported them
4409 6. If an ACL "warn" statement specified the addition of headers, Exim already
4410 inserted X-ACL-Warn: at the start if there was no header name. However, it
4411 was not making this test for the second and subsequent header lines if
4412 there were newlines in the string. This meant that an invalid header could
4413 be inserted if Exim was badly configured.
4415 7. Allow an ACL "warn" statement to add header lines at the start or after all
4416 the Received: headers, as well as at the end.
4418 8. Added the rcpt_4xx retry error code.
4420 9. Added postmaster_mailfrom=xxx to callout verification option.
4422 10. Added mailfrom=xxxx to the callout verification option, for verify=
4425 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
4426 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
4428 12. Inserted some casts to stop certain compilers warning when using pointer
4429 differences as field lengths or precisions in printf-type calls (mostly
4430 affecting debugging statements).
4432 13. Added optional readline() support for -be (dynamically loaded).
4434 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
4435 same clock tick as a message's arrival, so that its received time was the
4436 same as the "first fail" time on the retry record, and that message
4437 remained on the queue past the ultimate address timeout, every queue runner
4438 would try a delivery (because it was past the ultimate address timeout) but
4439 after another failure, the ultimate address timeout, which should have then
4440 bounced the address, did not kick in. This was a "< instead of <=" error;
4441 in most cases the first failure would have been in the next clock tick
4442 after the received time, and all would be well.
4444 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
4445 being recognized when the domain list was tested by the match_domain
4446 condition in an expansion string.
4448 16. Added the ${str2b64: operator.
4450 17. Exim was always calling setrlimit() to set a large limit for the number of
4451 processes, without checking whether the existing limit was already
4452 adequate. (It did check for the limit on file descriptors.) Furthermore,
4453 errors from getrlimit() and setrlimit() were being ignored. Now they are
4454 logged to the main and panic logs, but Exim does carry on, to try to do its
4455 job under whatever limits there are.
4457 18. Imported PCRE 5.0.
4459 19. Trivial typo in log message " temporarily refused connection" (the leading
4462 20. If the log selector return_path_on_delivery was set and an address was
4463 redirected to /dev/null, the delivery process crashed because it assumed
4464 that a return path would always be set for a "successful" delivery. In this
4465 case, the whole delivery is bypassed as an optimization, and therefore no
4468 21. Internal re-arrangement: the function for sending a challenge and reading
4469 a response while authentication was assuming a zero-terminated challenge
4470 string. It's now changed to take a pointer and a length, to allow for
4471 binary data in such strings.
4473 22. Added the cyrus_sasl authenticator (code supplied by MBM).
4475 23. Exim was not respecting finduser_retries when seeking the login of the
4476 uid under which it was called; it was always trying 10 times. (The default
4477 setting of finduser_retries is zero.) Also, it was sleeping after the final
4478 failure, which is pointless.
4480 24. Implemented tls_on_connect_ports.
4482 25. Implemented acl_smtp_predata.
4484 26. If the domain in control=submission is set empty, Exim assumes that the
4485 authenticated id is a complete email address when it generates From: or
4486 Sender: header lines.
4488 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
4489 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
4490 chown and chgrp in /bin and hostname in /usr/bin.
4492 28. Exim was keeping the "process log" file open after each use, just as it
4493 does for the main log. This opens the possibility of it remaining open for
4494 long periods when the USR1 signal hits a daemon. Occasional processlog
4495 errors were reported, that could have been caused by this. Anyway, it seems
4496 much more sensible not to leave this file open at all, so that is what now
4499 29. The long-running daemon process does not normally write to the log once it
4500 has entered its main loop, and it closes the log before doing so. This is
4501 so that log files can straightforwardly be renamed and moved. However,
4502 there are a couple of unusual error situations where the daemon does write
4503 log entries, and I had neglected to close the log afterwards.
4505 30. The text of an SMTP error response that was received during a remote
4506 delivery was being truncated at 512 bytes. This is too short for some of
4507 the long messages that one sometimes sees. I've increased the limit to
4510 31. It is now possible to make retry rules that apply only when a message has a
4511 specific sender, in particular, an empty sender.
4513 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
4514 it possible to be selective about when SMTP synchronization is enforced.
4516 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
4518 32. Implemented hosts_connection_nolog.
4520 33. Added an ACL for QUIT.
4522 34. Setting "delay_warning=" to disable warnings was not working; it gave a
4525 35. Added mailbox_size and mailbox_filecount to appendfile.
4527 36. Added control = no_multiline_responses to ACLs.
4529 37. There was a bug in the logic of the code that waits for the clock to tick
4530 in the case where the clock went backwards by a substantial amount such
4531 that the microsecond fraction of "now" was more than the microsecond
4532 fraction of "then" (but the whole seconds number was less).
4534 38. Added support for the libradius Radius client library this is found on
4535 FreeBSD (previously only the radiusclient library was supported).
4541 1. When certain lookups returned multiple values in the form name=value, the
4542 quoting of the values was not always being done properly. Specifically:
4543 (a) If the value started with a double quote, but contained no whitespace,
4545 (b) If the value contained whitespace other than a space character (i.e.
4546 tabs or newlines or carriage returns) it was not quoted.
4547 This fix has been applied to the mysql and pgsql lookups by writing a
4548 separate quoting function and calling it from the lookup code. The fix
4549 should probably also be applied to nisplus, ibase and oracle lookups, but
4550 since I cannot test any of those, I have not disturbed their existing code.
4552 2. A hit in the callout cache for a specific address caused a log line with no
4553 reason for rejecting RCPT. Now it says "Previous (cached) callout
4554 verification failure".
4556 3. There was an off-by-one bug in the queryprogram router. An over-long
4557 return line was truncated at 256 instead of 255 characters, thereby
4558 overflowing its buffer with the terminating zero. As well as fixing this, I
4559 have increased the buffer size to 1024 (and made a note to document this).
4561 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
4562 when Exim is waiting for an SMTP response from a remote server, Exim
4563 restarts its select() call on the socket, thereby resetting its timeout.
4564 This is not a problem when such interrupts are rare. Somebody set up a cron
4565 job to run exiwhat every 2 minutes, which is less than the normal select()
4566 timeout (5 or 10 minutes). This meant that the select() timeout never
4567 kicked in because it was always reset. I have fixed this by comparing the
4568 time when an interrupt arrives with the time at the start of the first call
4569 to select(). If more time than the timeout has elapsed, the interrupt is
4570 treated as a timeout.
4572 5. Some internal re-factoring in preparation for the addition of Sieve
4573 extensions (by MH). In particular, the "personal" test is moved to a
4574 separate function, and given an option for scanning Cc: and Bcc: (which is
4575 not set for Exim filters).
4577 6. When Exim created an email address using the login of the caller as the
4578 local part (e.g. when creating a From: or Sender: header line), it was not
4579 quoting the local part when it contained special characters such as @.
4581 7. Installed new OpenBSD configuration files.
4583 8. Reworded some messages for syntax errors in "and" and "or" conditions to
4584 try to make them clearer.
4586 9. Callout options, other than the timeout value, were being ignored when
4587 verifying sender addresses in header lines. For example, when using
4589 verify = header_sender/callout=no_cache
4591 the cache was (incorrectly) being used.
4593 10. Added a missing instance of ${EXE} to the exim_install script; this affects
4594 only the Cygwin environment.
4596 11. When return_path_on_delivery was set as a log selector, if different remote
4597 addresses in the same message used different return paths and parallel
4598 remote delivery occurred, the wrong values would sometimes be logged.
4599 (Whenever a remote delivery process finished, the return path value from
4600 the most recently started remote delivery process was logged.)
4602 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
4603 lines when AUTH and/or TLS are in use. This is the "received protocol"
4604 field. Exim used to use "asmtp" for authenticated SMTP, without any
4605 indication (in the protocol name) for TLS use. Now it follows the RFC and
4606 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
4607 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
4608 names appear in log lines as well as in Received: header lines.
4610 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
4611 extensions, and comparison tests, and to fix some bugs.
4613 14. Changes to the "personal" filter test:
4615 (1) The test was buggy in that it was just doing the equivalent of
4616 "contains" tests on header lines. For example, if a user's address was
4617 anne@some.where, the "personal" test would incorrectly be true for
4619 To: susanne@some.where
4621 This test is now done by extracting each address from the header in turn,
4622 and checking the entire address. Other tests that are part of "personal"
4623 are now done using regular expressions (for example, to check local parts
4624 of addresses in From: header lines).
4626 (2) The list of non-personal local parts in From: addresses has been
4627 extended to include "listserv", "majordomo", "*-request", and "owner-*",
4628 taken from the Sieve specification recommendations.
4630 (3) If the message contains any header line starting with "List-" it is
4631 treated as non-personal.
4633 (4) The test for "circular" in the Subject: header line has been removed
4634 because it now seems ill-conceived.
4636 15. Minor typos in src/EDITME comments corrected.
4638 16. Installed latest exipick from John Jetmore.
4640 17. If headers_add on a router specified a text string that was too long for
4641 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
4642 of string_sprintf() is now avoided.
4644 18. $message_body_size was not set (it was always zero) when running the DATA
4645 ACL and the local_scan() function.
4647 19. For the "mail" command in an Exim filter, no default was being set for
4648 the once_repeat time, causing a random time value to be used if "once" was
4649 specified. (If the value happened to be <= 0, no repeat happened.) The
4650 default is now 0s, meaning "never repeat". The "vacation" command was OK
4651 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
4652 (I found it when inspecting the code).
4654 20. There is now an overall timeout for performing a callout verification. It
4655 defaults to 4 times the callout timeout, which applies to individual SMTP
4656 commands during the callout. The overall timeout applies when there is more
4657 than one host that can be tried. The timeout is checked before trying the
4658 next host. This prevents very long delays if there are a large number of
4659 hosts and all are timing out (e.g. when the network connections are timing
4660 out). The value of the overall timeout can be changed by specifying an
4661 additional sub-option for "callout", called "maxwait". For example:
4663 verify = sender/callout=5s,maxwait=20s
4665 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
4666 to the end before writing, but this should make it even safer).
4668 22. Exim was forgetting that it had advertised PIPELINING for the second and
4669 subsequent messages on an SMTP connection. It was also not resetting its
4670 memory on STARTTLS and an internal HELO.
4672 23. When Exim logs an SMTP synchronization error within a session, it now
4673 records whether PIPELINING has been advertised or not.
4675 24. Added 3 instances of "(long int)" casts to time_t variables that were being
4676 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
4677 rather than long int.
4679 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
4681 26. Added the never_mail option to autoreply.
4687 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
4688 crash if the getsockname() call failed; this can happen if a connection is
4689 closed very soon after it is established. The problem was simply in the
4690 order in which certain operations were done, causing Exim to try to write
4691 to the SMTP stream before it had set up the file descriptor. The bug has
4692 been fixed by making things happen in the correct order.
4698 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
4699 before the connection was closed, thus losing the rejection response.
4701 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
4702 some early Solaris releases, but causes trouble in current releases where
4703 socklen_t is defined.
4705 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
4708 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
4711 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
4712 that releasing the top of it at the end releases what was used for sub-
4713 expansions (unless the block got too big). However, discard this block if
4714 the first thing is a variable or header, so that we can use its block when
4715 it is dynamic (useful for very large $message_headers, for example).
4717 6. Lookups now cache *every* query, not just the most recent. A new, separate
4718 store pool is used for this. It can be recovered when all lookup caches are
4719 flushed. Lookups now release memory at the end of their result strings.
4720 This has involved some general refactoring of the lookup sources.
4722 7. Some code has been added to the store_xxx() functions to reduce the amount
4723 of flapping under certain conditions.
4725 8. log_incoming_interface used to affect only the <= reception log lines. Now
4726 it causes the local interface and port to be added to several more SMTP log
4727 lines, for example "SMTP connection from", and rejection lines.
4729 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
4731 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
4733 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
4734 could be overwritten at the end of the current message (or the start of a
4735 new message if it was set in a HELO ACL). The value is now preserved for
4736 the duration of the SMTP connection.
4738 12. If a transport had a headers_rewrite setting, and a matching header line
4739 contained an unqualified address, that address was qualified, even if it
4740 did not match any rewriting rules. The underlying bug was that the values
4741 of the flags that permit the existence of unqualified sender and recipient
4742 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
4743 non-local messages, and by -bnq for local messages) were not being
4744 preserved with the message after it was received.
4746 13. When Exim was logging an SMTP synchronization error, it could sometimes log
4747 "next input=" as part of the text comprising the host identity instead of
4748 the correct text. The code was using the same buffer for two different
4749 strings. However, depending on which order the printing function evaluated
4750 its arguments, the bug did not always show up. Under Linux, for example, my
4751 test suite worked just fine.
4753 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
4754 doesn't work with some older versions of Perl. It has been changed to "my",
4755 which in any case is probably the better facility to use.
4757 15. A really picky compiler found some instances of statements for creating
4758 error messages that either had too many or two few arguments for the format
4761 16. The size of the buffer for calls to the DNS resolver has been increased
4762 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
4763 for addresses that have a lot of PTR records. This alleviates a problem; it
4764 does not fully solve it.
4766 17. A dnsdb lookup for PTR records that receives more data than will fit in the
4767 buffer now truncates the list and logs the incident, which is the same
4768 action as happens when Exim is looking up a host name and its aliases.
4769 Previously in this situation something unpredictable would happen;
4770 sometimes it was "internal error: store_reset failed".
4772 18. If a server dropped the connection unexpectedly when an Exim client was
4773 using GnuTLS and trying to read a response, the client delivery process
4774 crashed while trying to generate an error log message.
4776 19. If a "warn" verb in an ACL added multiple headers to a message in a single
4777 string, for example:
4779 warn message = H1: something\nH2: something
4781 the text was added as a single header line from Exim's point of view
4782 though it ended up OK in the delivered message. However, searching for the
4783 second and subsequent header lines using $h_h2: did not work. This has been
4784 fixed. Similarly, if a system filter added multiple headers in this way,
4785 the routers could not see them.
4787 20. Expanded the error message when iplsearch is called with an invalid key to
4788 suggest using net-iplsearch in a host list.
4790 21. When running tests using -bh, any delays imposed by "delay" modifiers in
4791 ACLs are no longer actually imposed (and a message to that effect is
4794 22. If a "gecos" field in a passwd entry contained escaped characters, in
4795 particular, if it contained a \" sequence, Exim got it wrong when building
4796 a From: or a Sender: header from that name. A second bug also caused
4797 incorrect handling when an unquoted " was present following a character
4798 that needed quoting.
4800 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
4801 was not being matched caselessly.
4803 24. Arranged for all hyphens in the exim.8 source to be escaped with
4806 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
4807 properly. Recipient callout cache records were still being keyed to include
4808 the sender, even when use_sender was set false. This led to far more
4809 callouts that were necessary. The sender is no longer included in the key
4810 when use_sender is false.
4812 26. Added "control = submission" modifier to ACLs.
4814 27. Added the ${base62d: operator to decode base 62 numbers.
4816 28. dnsdb lookups can now access SRV records.
4818 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
4819 the configuration file.
4821 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
4822 (-v) mode. This makes the output for a verbose queue run more intelligible.
4824 31. Added a use_postmaster feature to recipient callouts.
4826 32. Added the $body_zerocount variable, containing the number of binary zero
4827 bytes in the message body.
4829 33. The time of last modification of the "new" subdirectory is now used as the
4830 "mailbox time last read" when there is a quota error for a maildir
4833 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
4835 35. Added +ignore_unknown as a special item in host lists.
4837 36. Code for decoding IPv6 addresses in host lists is now included, even if
4838 IPv6 support is not being compiled. This fixes a bug in which an IPv6
4839 address was recognized as an IP address, but was then not correctly decoded
4840 into binary, causing unexpected and incorrect effects when compared with
4847 1. Very minor rewording of debugging text in manualroute to say "list of
4848 hosts" instead of "hostlist".
4850 2. If verify=header_syntax was set, and a header line with an unqualified
4851 address (no domain) and a large number of spaces between the end of the
4852 name and the colon was received, the reception process suffered a buffer
4853 overflow, and (when I tested it) crashed. This was caused by some obsolete
4854 code that should have been removed. The fix is to remove it!
4856 3. When running in the test harness, delay a bit after writing a bounce
4857 message to get a bit more predictability in the log output.
4859 4. Added a call to search_tidyup() just before forking a reception process. In
4860 theory, someone could use a lookup in the expansion of smtp_accept_max_
4861 per_host which, without the tidyup, could leave open a database connection.
4863 5. Added the variables $recipient_data and $sender_data which get set from a
4864 lookup success in an ACL "recipients" or "senders" condition, or a router
4865 "senders" option, similar to $domain_data and $local_part_data.
4867 6. Moved the writing of debug_print from before to after the "senders" test
4870 7. Change 4.31/66 (moving the time when the Received: is generated) caused
4871 problems for message scanning, either using a data ACL, or using
4872 local_scan() because the Received: header was not generated till after they
4873 were called (in order to set the time as the time of reception completion).
4874 I have revised the way this works. The header is now generated after the
4875 body is received, but before the ACL or local_scan() are called. After they
4876 are run, the timestamp in the header is updated.
4882 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
4883 before starting a queue runner without re-exec. This happened only when
4884 deliver_drop_privilege was set or when the Exim user was set to root. The
4885 effect of the bug was that timeouts during subsequent deliveries caused
4886 crashes instead of being properly handled. The handler is now left at its
4887 default (and expected) setting.
4889 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
4890 message, again when deliver_drop_privilege is set or Exim is run as root.
4891 The bug described in (1) was not present in this case, but the tidying up
4892 of the other signals was missing. I have made the two cases consistent.
4894 3. The ignore_target_hosts setting on a manualroute router was being ignored
4895 for hosts that were looked up using the /MX notation.
4897 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
4900 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
4901 operated on the sender address. After changing the $sender_address to <>
4902 for the sender address verify, Exim was re-instated it as the original
4903 (before rewriting) address, but remembering that it had rewritten it, so it
4904 wasn't rewriting it again. This bug also had the effect of breaking the
4905 sender address verification caching when the sender address was rewritten.
4907 6. The ignore_target_hosts option was being ignored by the ipliteral router.
4908 This has been changed so that if the ip literal address matches
4909 ignore_target_hosts, the router declines.
4911 7. Added expansion conditions match_domain, match_address, and match_local_
4912 part (NOT match_host).
4914 8. The placeholder for the Received: header didn't have a length field set.
4916 9. Added code to Exim itself and to exim_lock to test for a specific race
4917 condition that could lead to file corruption when using MBX delivery. The
4918 issue is with the lockfile that is created in /tmp. If this file is removed
4919 after a process has opened it but before that process has acquired a lock,
4920 there is the potential for a second process to recreate the file and also
4921 acquire a lock. This could lead to two Exim processes writing to the file
4922 at the same time. The added code performs the same test as UW imapd; it
4923 checks after acquiring the lock that its file descriptor still refers to
4924 the same named file.
4926 10. The buffer for building added header lines was of fixed size, 8192 bytes.
4927 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
4930 11. Added the smtp_active_hostname option. If used, this will typically be made
4931 to depend on the incoming interface address. Because $interface_address is
4932 not set up until the daemon has forked a reception process, error responses
4933 that can happen earlier (such as "too many connections") no longer contain
4936 12. If an expansion in a condition on a "warn" statement fails because a lookup
4937 defers, the "warn" statement is abandoned, and the next ACL statement is
4938 processed. Previously this caused the whole ACL to be aborted.
4940 13. Added the iplsearch lookup type.
4942 14. Added ident_timeout as a log selector.
4944 15. Added tls_certificate_verified as a log selector.
4946 16. Added a global option tls_require_ciphers (compare the smtp transport
4947 option of the same name). This controls incoming TLS connections.
4949 17. I finally figured out how to make tls_require_ciphers do a similar thing
4950 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
4951 before starting the TLS session.
4953 18. Tabs are now shown as \t in -bP output.
4955 19. If the log selector return_path_on_delivery was set, Exim crashed when
4956 bouncing a message because it had too many Received: header lines.
4958 20. If two routers both had headers_remove settings, and the first one included
4959 a superfluous trailing colon, the final name in the first list and the
4960 first name in the second list were incorrectly joined into one item (with a
4961 colon in the middle).
4967 1. Added -C and -D options to the exinext utility, mainly to make it easier
4968 to include in the automated testing, but these could be helpful when
4969 multiple configurations are in use.
4971 2. The exinext utility was not formatting the output nicely when there was
4972 an alternate port involved in the retry record key, nor when there was a
4973 message id as well (for retries that were specific to a specific message
4974 and a specific host). It was also confused by IPv6 addresses, because of
4975 the additional colons they contain. I have fixed the IPv4 problem, and
4976 patched it up to do a reasonable job for IPv6.
4978 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
4979 delivery, the log line now contains "pipelined" if PIPELINING was used.
4981 4. An SMTP transport process used to panic and die if the bind() call to set
4982 an explicit outgoing interface failed. This has been changed; it is now
4983 treated in the same way as a connect() failure.
4985 5. A reference to $sender_host_name in the part of a conditional expansion
4986 that was being skipped was still causing a DNS lookup. This no longer
4989 6. The def: expansion condition was not recognizing references to header lines
4990 that used bh_ and bheader_.
4992 7. Added the _cache feature to named lists.
4994 8. The code for checking quota_filecount in the appendfile transport was
4995 allowing one more file than it should have been.
4997 9. For compatibility with Sendmail, the command line option
5005 and sets the incoming protocol and host name (for trusted callers). The
5006 host name and its colon can be omitted when only the protocol is to be set.
5007 Note the Exim already has two private options, -pd and -ps, that refer to
5008 embedded Perl. It is therefore impossible to set a protocol value of "d" or
5009 "s", but I don't think that's a major issue.
5011 10. A number of refactoring changes to the code, none of which should affect
5014 (a) The number of logging options was getting close to filling up the
5015 32-bit word that was used as a bit map. I have split them into two classes:
5016 those that are passed in the argument to log_write(), and those that are
5017 only ever tested independently outside of that function. These are now in
5018 separate 32-bit words, so there is plenty of room for expansion again.
5019 There is no change in the user interface or the logging behaviour.
5021 (b) When building, for example, log lines, the code previously used a
5022 macro that called string_cat() twice, in order to add two strings. This is
5023 not really sufficiently general. Furthermore, there was one instance where
5024 it was actually wrong because one of the argument was used twice, and in
5025 one call a function was used. (As it happened, calling the function twice
5026 did not affect the overall behaviour.) The macro has been replaced by a
5027 function that can join an arbitrary number of extra strings onto a growing
5030 (c) The code for expansion conditions now uses a table and a binary chop
5031 instead of a serial search (which was left over from when there were very
5032 few conditions). Also, it now recognizes conditions like "pam" even when
5033 the relevant support is not compiled in: a suitably worded error message is
5034 given if an attempt is made to use such a condition.
5036 11. Added ${time_interval:xxxxx}.
5038 12. A bug was causing one of the ddress fields not to be passed back correctly
5039 from remote delivery subprocesses. The field in question was not being
5040 subsequently used, so this caused to problems in practice.
5042 13. Added new log selectors queue_time and deliver_time.
5044 14. Might have fixed a bug in maildirsizefile handling that threw up
5045 "unexpected character" debug warnings, and recalculated the data
5046 unnecessarily. In any case, I expanded the warning message to give more
5049 15. Added the message "Restricted characters in address" to the statements in
5050 the default ACL that block characters like @ and % in local parts.
5052 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
5053 Three changes have been made:
5055 (a) There was a serious bug; a negative response to MAIL caused the whole
5056 recipient domain to be cached as invalid, thereby blocking all messages
5057 to all local parts at the same domain, from all senders. This bug has
5058 been fixed. The domain is no longer cached after a negative response to
5059 MAIL if the sender used is not empty.
5061 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
5064 (c) A new callout option, "use_sender" has been added for people who want
5065 the modified behaviour.
5071 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
5074 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
5075 indeed breaks things for older releases.
5077 3. Added additional logging to the case where there is a problem reading data
5078 from a filter that is running in a subprocess using a pipe, in order to
5079 try to track down a specific problem.
5081 4. Testing facility fudge: when running in the test harness and attempting
5082 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
5083 getting "No route to host". Convert this to a timeout.
5085 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
5088 6. Some OS don't have socklen_t but use size_t instead. This affects the
5089 fifth argument of getsockopt() amongst other things. This is now
5090 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
5091 can be set for individual OS. I have set it for SunOS5, OSF1, and
5092 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
5093 some earlier ones do not.
5095 7. Change 4.30/15 was not doing the test caselessly.
5097 8. The standard form for an IPv6 address literal was being rejected by address
5098 parsing in, for example, MAIL and RCPT commands. An example of this kind of
5099 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5100 this, as well as the form without the "IPv6" on the front (but only when
5101 address literals are enabled, of course).
5103 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
5105 10. Exim crashed if a message with an empty sender address specified by -f
5106 encountered a router with an errors_to setting. This could be provoked only
5107 by a command such as
5111 where an empty string was supplied; "<>" did not hit this bug.
5113 11. Installed PCRE release 4.5.
5115 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
5116 remained set. It is now erased.
5118 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
5119 times from message ids (which are base 36 rather than the normal 62).
5121 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
5122 were being counted as actual protocol errors, and logged if the log
5123 selector +smtp_protocol_error was set. One cannot be perfect in this test,
5124 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
5125 and DATA following a set of rejected RCPTs do not count as protocol errors.
5126 In other words, Exim assumes they were pipelined, though this may not
5127 actually be the case. Of course, in all cases the client gets an
5128 appropriate error code.
5130 15. If a lookup fails in an ACL condition, a message about the failure may
5131 be available; it is used if testing the ACL cannot continue, because most
5132 such messages specify what the cause of the deferral is. However, some
5133 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
5134 that caused an old message to be retained and used if a later statement
5135 caused a defer, replacing the real cause of the deferral.
5137 16. If an IP address had so many PTR records that the DNS lookup buffer
5138 was not large enough to hold them, Exim could crash while trying to process
5139 the truncated data. It now detects and logs this case.
5141 17. Further to 4.21/58, another change has been made: if (and only if) the
5142 first line of a message (the first header line) ends with CRLF, a bare LF
5143 in a subsequent header line has a space inserted after it, so as not to
5144 terminate the header.
5146 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
5147 unnecessarily, used atoi() instead of strtol(), and didn't check the
5148 termination when getting file sizes from file names by regex.
5150 19. Completely re-implemented the support for maildirsize files, in the light
5151 of a number of problems with the previous contributed implementation
5152 (4.30/29). In particular:
5154 . If the quota is zero, the maildirsize file is maintained, but no quota is
5157 . If the maildir directory does not exist, it is created before any attempt
5158 to write a maildirsize file.
5160 . The quota value in the file is just a cache; if the quota is changed in
5161 the transport, the new value overrides.
5163 . A regular expression is available for excluding directories from the
5166 20. The autoreply transport checks the characters in options that define the
5167 message's headers; it allows continued headers, but it was checking with
5168 isspace() after an embedded newline instead of explicitly looking for a
5171 21. If all the "regular" hosts to which an address was routed had passed their
5172 expiry times, and had not reached their retry times, the address was
5173 bounced, even if fallback hosts were defined. Now Exim should go on to try
5176 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
5177 equivalent code in the SMTP transport. Some hosts send humungous responses
5178 to HELO/EHLO, more than 1024 it seems.
5180 23. Refactoring: code in filter.c used (void *) for "any old type" but this
5181 gives compiler warnings in some environments. I've now done it "properly",
5184 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
5185 (because of problems with the built-in one) was declared to return uschar *
5186 instead of char *, causing compiler failure.
5188 25. Fixed a file descriptor leak when processing alias/forward files.
5190 26. Fixed a minor format string issue in dbfn.c.
5192 27. Typo in exim.c: ("dmbnz" for "dbmnz").
5194 28. If a filter file refered to $h_xxx or $message_headers, and the headers
5195 contained RFC 2047 "words", Exim's memory could, under certain conditions,
5198 29. When a sender address is verified, it is cached, to save repeating the test
5199 when there is more than one recipient in a message. However, when the
5200 verification involves a callout, it is possible for different callout
5201 options to be set for different recipients. It is too complicated to keep
5202 track of this in the cache, so now Exim always runs a verification when a
5203 callout is required, relying on the callout cache for the optimization.
5204 The overhead is duplication of the address routing, but this should not be
5207 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
5208 to be verified with callout=postmaster, and the main callout worked but the
5209 postmaster check failed, the verification correctly failed. However, if a
5210 subsequent RCPT command asked for sender verification *without* the
5211 postmaster check, incorrect caching caused this verification also to fail,
5214 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
5215 it was not caching the DNS options (qualify_single, search_parents) that
5216 were used when the lookup failed. A subsequent lookup with different
5217 options therefore always gave the same answer, though there were cases
5218 where it should not have. (Example: a "domains = !$mx_any" option on a
5219 dnslookup router: the "domains" option is always processed without any
5220 widening, but the router might have qualify_single set.) Now Exim uses the
5221 cached value only when the same options are set.
5223 32. Added John Jetmore's "exipick" utility to the distribution.
5225 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
5226 than a timeout (e.g. a certificate is required, and is not provided), an
5227 Exim server now closes the connection immediately. Previously it waited for
5228 the client to close - but if the client is SSL, it seems that they each
5229 wait for each other, leading to a delay before one of them times out.
5231 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
5232 maintained 0.8.x compatibility because I don't think many are using it, and
5233 it is clearly obsolete.
5235 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
5238 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
5239 client certificate was expired. A simple patch fixes this, though I don't
5240 understand the full logic of why the verify callback is called multiple
5243 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
5244 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
5245 which causes problems with some clients (such as the Certicom SSL Plus
5246 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
5247 disables the coutermeasure allowing Eudora to connect."
5249 38. Exim was not checking that a write() to a log file succeeded. This could
5250 lead to Bad Things if a log got too big, in particular if it hit a file
5251 size limit. Exim now panics and dies if it cannot write to a log file, just
5252 as it does if it cannot open a log file.
5254 39. Modified OS/Makefile-Linux so that it now contains
5256 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
5258 The two -D definitions ensure that Exim is compiled with large file
5259 support, which makes it possible to handle log files that are bigger than
5262 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
5263 instance) a domain was checked against a named list that involved a lookup,
5264 causing $domain_data to be set, then another domain was checked against the
5265 same list, then the first domain was re-checked, the value of $domain_data
5266 after the final check could be wrong. In particular, if the second check
5267 failed, it could be set empty. This bug probably also applied to
5270 41. The strip_trailing_dot option was not being applied to the address given
5271 with the -f command-line option.
5273 42. The code for reading a message's header from the spool was incrementing
5274 $received_count, but never initializing it. This meant that the value was
5275 incorrect (doubled) while delivering a message in the same process in which
5276 it was received. In the most common configuration of Exim, this never
5277 happens - a fresh exec is done - but it can happen when
5278 deliver_drop_privilege is set.
5280 43. When Exim logs an SMTP synchronization error - client data sent too soon -
5281 it now includes up to 150 characters of the unexpected data in the log
5284 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
5285 and building data strings. The size of both of these buffers was 10 000
5286 bytes - far larger than anybody would *ever* want, thought I. Needless to
5287 say, somebody hit the limit. I have increased the maximum line length to
5288 20 000 and the maximum data length of concatenated lines to 100 000. I have
5289 also fixed two bugs, because there was no checking on these buffers. Tsk,
5290 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
5291 buffer is too small.
5293 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
5294 lsearch lookups. Now it does.
5296 46. When parsing a route_list item in a manualroute router, a fixed-length
5297 buffer was used for the list of hosts. I made this 1024 bytes long,
5298 thinking that nobody would ever have a list of hosts that long. Wrong.
5299 Somebody had a whole pile of complicated expansion conditions, and the
5300 string was silently truncated, leading to an expansion error. It turns out
5301 that it is easier to change to an unlimited length (owing to other changes
5302 that have happened since this code was originally written) than to build
5303 structure for giving a limitation error. The length of the item that
5304 expands into the list of hosts is now unlimited.
5306 47. The lsearch lookup could not handle data where the length of text line was
5307 more than 4095 characters. Such lines were truncated, leading to shortened
5308 data being returned. It should now handle lines of any length.
5310 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
5311 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
5314 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
5315 replacing strings such as BIN_DIRECTORY in a source file: the replacement
5316 no longer happens in comment lines. A list of replacements is now placed
5317 at the head of all of the source files, except those whose only change is
5318 to replace PERL_COMMAND in the very first #! line.
5320 50. Replaced the slow insertion sort in queue.c, for sorting the list of
5321 messages on the queue, with a bottom-up merge sort, using code contributed
5322 by Michael Haardt. This should make operations like -bp somewhat faster on
5323 large queues. It won't affect queue runners, except when queue_run_in_order
5326 51. Installed eximstats 1.31 in the distribution.
5328 52. Added support for SRV lookups to the dnslookup router.
5330 53. If an ACL referred to $message_body or $message_body_end, the value was not
5331 reset for any messages that followed in the same SMTP session.
5333 54. The store-handling optimization for building very long strings was not
5334 differentiating between the different store pools. I don't think this
5335 actually made any difference in practice, but I've tidied it.
5337 55. While running the routers to verify a sender address, $sender_address
5338 was still set to the sender address. This is wrong, because when routing to
5339 send a bounce to the sender, it would be empty. Therefore, I have changed
5340 it so that, while verifying a sender address, $sender_address is set to <>.
5341 (There is no change to what happens when verifying a recipient address.)
5343 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
5344 target A or AAAA records (if not already returned) without resetting the
5345 qualify_single or search_parents options of the DNS resolver. These are
5346 inappropriate in this case because the targets of MX and SRV records must
5347 be FQDNs. A broken DNS record could cause trouble if it happened to have a
5348 target that, when qualified, matched something in the local domain. These
5349 two options are now turned off when doing these lookups.
5351 57. It seems that at least some releases of Reiserfs (which does not have the
5352 concept of a fixed number of inodes) returns zero and not -1 for the
5353 number of available inodes. This interacted badly with check_spool_inodes,
5354 which assumed that -1 was the "no such thing" setting. What I have done is
5355 to check that the total number of inodes is greater than zero before doing
5356 the test of how many are available.
5358 58. When a "warn" ACL statement has a log_message modifier, the message is
5359 remembered, and not repeated. This is to avoid a lot of repetition when a
5360 message has many recipients that cause the same warning to be written.
5361 Howewer, Exim was preserving the list of already written lines for an
5362 entire SMTP session, which doesn't seem right. The memory is now reset if a
5363 new message is started.
5365 59. The "rewrite" debugging flag was not showing the result of rewriting in the
5366 debugging output unless log_rewrite was also set.
5368 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
5369 of (int)(handle) when we know that handle contains (void *)(-1).
5371 61. The Exim daemon panic-logs an error return when it closes the incoming
5372 connection. However "connection reset by peer" seems to be common, and
5373 isn't really an error worthy of noting specially, so that particular error
5376 62. When Exim is trying to find all the local interfaces, it used to panic and
5377 die if the ioctl to get the interface flags failed. However, it seems that
5378 on at least one OS (Solaris 9) it is possible to have an interface that is
5379 included in the list of interfaces, but for which you get a failure error
5380 for this call. This happens when the interface is not "plumbed" into a
5381 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
5382 failure of the "get flags" call assumes that the interface is down.
5384 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
5385 makes life easier for people who are doing arithmetic on fields extracted
5386 from dates, where you often get leading zeros that should not be
5387 interpreted as octal.
5389 64. Added qualify_domain to the redirect router, to override the global
5392 65. If a pathologically long header line contained very many addresses (the
5393 report of this problem mentioned 10 000) and each of them was rewritten,
5394 Exim could use up a very large amount of memory. (It kept on making new
5395 copies of the header line as it rewrote, and never released the old ones.)
5396 At the expense of a bit more processing, the header rewriting function has
5397 been changed so that it no longer eats memory in this way.
5399 66. The generation of the Received: header has been moved from the time that a
5400 message starts to be received, to the time that it finishes. The timestamp
5401 in the Received: header should now be very close to that of the <= log
5402 line. There are two side-effects of this change:
5404 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
5405 logged header lines no longer include the local Received: line, because
5406 it has not yet been created. The same applies to a copy of the message
5407 that is returned to a non-SMTP sender when a message is rejected.
5409 (b) When a filter file is tested using -bf, no additional Received: header
5410 is added to the test message. After some thought, I decided that this
5413 This change does not affect the value of $received_for. It is still set
5414 after address rewriting, but before local_scan() is called.
5416 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
5418 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
5419 gave an unhelpful panic error message, and a defer error. I have managed to
5420 change this behaviour so that it now rejects any supplied certificate,
5421 which seems right, as the list of acceptable certificates is empty.
5423 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
5424 gave an unhelpful defer error. I have not managed to make this reject any
5425 supplied certificates, but the error message it gives is "no certificate
5426 supplied", which is not helpful.
5428 70. exigrep's output now also includes lines that are not associated with any
5429 message, but which match the given pattern. Implemented by a patch from
5430 Martin Sluka, which also tidied up the Perl a bit.
5432 71. Recipient callout verification, like sender verification, was using <> in
5433 the MAIL FROM command. This isn't really the right thing, since the actual
5434 sender may affect whether the remote host accepts the recipient or not. I
5435 have changed it to use the actual sender in the callout; this means that
5436 the cache record is now keyed on a recipient/sender pair, not just the
5437 recipient address. There doesn't seem to be a real danger of callout loops,
5438 since a callout by the remote host to check the sender would use <>.
5439 [SEE ABOVE: changed after hitting problems.]
5441 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
5442 temporary errors. However, in the case of such a code being given after
5443 the end of a data transmission (i.e. after ".") Exim was failing to write
5444 a retry record for the message. (Yes, there was some broken host that was
5445 actually sending 8xx at this point.)
5447 73. An unknown lookup type in a host list could cause Exim to panic-die when
5448 the list was checked. (An example that provoked this was putting <; in the
5449 middle of a list instead of at the start.) If this happened during a DATA
5450 ACL check, a -D file could be left lying around. This kind of configuration
5451 error no longer causes Exim to die; instead it causes a defer errror. The
5452 incident is still logged to the main and panic logs.
5454 74. Buglet left over from Exim 3 conversion. The message "too many messages
5455 in one connection" was written to the rejectlog but not the mainlog, except
5456 when address rewriting (yes!) was being logged.
5458 75. Added write_rejectlog option.
5460 76. When a system filter was run not as root (that is, when system_filter_user
5461 was set), the values of the $n variables were not being returned to the
5462 main process; thus, they were not subsequently available in the $sn
5465 77. Added +return_path_on_delivery log selector.
5467 78. A connection timeout was being treated differently from recipients deferred
5468 when testing hosts_max_try with a message that was older than the host's
5469 retry timeout. (The host should not be counted, thus allowing all hosts to
5470 be tried at least once before bouncing.) This may have been the cause of an
5471 occasionally reported bug whereby a message would remain on the queue
5472 longer than the retry timeout, but would be bounced if a delivery was
5473 forced. I say "may" because I never totally pinned down the problem;
5474 setting up timeout/retry tests is difficult. See also the next item.
5476 79. The ultimate address timeout was not being applied to errors that involved
5477 a combination of host plus message (for example, a timeout on a MAIL
5478 command). When an address resolved to a number of possible hosts, and they
5479 were not all tried for each delivery (e.g. because of hosts_max_try), a
5480 message could remain on the queue longer than the retry timeout.
5482 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
5485 81. Fixed an obscure SMTP outgoing bug which required at least the following
5486 conditions: (a) there was another message waiting for the same server;
5487 (b) the server returned 5xx to all RCPT commands in the first message so
5488 that the message was not completed; (c) the server dropped the connection
5489 or gave a negative response to the RSET that Exim sends to abort the
5490 transaction. The observed case was a dropped connection after DATA that had
5491 been sent in pipelining mode. That is, the server had advertised PIPELINING
5492 but was not implementing it correctly. The effect of the bug was incorrect
5493 behaviour, such as trying another host, and this could lead to a crash.
5499 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
5500 and daemon.c were passed as pointers to ints; they should have been
5501 pointers to socklen_t variables (which are typically unsigned ints).
5503 2. Some signed/unsigned type warnings in the os.c file for Linux have been
5506 3. Fixed a really odd bug that affected only the testing scheme; patching a
5507 certain fixed string in the binary changed the value of another string that
5508 happened to be identical to the end of the original first string.
5510 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
5511 name", it returns that address as the IP address. On some operating
5512 systems (e.g. Solaris), it also passes back the IP address string as the
5513 "host name". However, on others (e.g. Linux), it passes back an empty
5514 string. Exim wasn't checking for this, and was changing the host name to an
5515 empty string, assuming it had been canonicized.
5517 5. Although rare, it is permitted to have more than one PTR record for a given
5518 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
5519 all the names associated with an address, because they do in Solaris.
5520 However, it seems that they do not in Linux for data that comes from the
5521 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
5522 I found this out when I moved to a new Linux workstation and tried to run
5523 the Exim test suite.
5525 To get round this problem I have changed the code so that it now does its
5526 own call to the DNS to look up PTR records when searching for a host name.
5527 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
5528 addresses that are only in /etc/hosts are still found.
5530 This behaviour is, however, controlled by an option called host_lookup_
5531 order, which defaults to "bydns:byaddr". If people want to use the other
5532 order, or indeed, just use one or the other means of lookup, they can
5533 specify it in this variable.
5535 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
5536 some operating systems, this comes back from gethostbyaddr() as an empty
5537 string, and this is what Exim used to test for. However, it seems that in
5538 other systems, "." is yielded. Exim now tests for this case too.
5540 7. The values of check_spool_space and check_log_space are now held internally
5541 as a number of kilobytes instead of an absolute number of bytes. If a
5542 numbers is specified without 'K' or 'M', it is rounded up to the nearest
5543 kilobyte. This means that much larger values can be stored.
5545 8. Exim monitor: an attempt to get the action menu when not actually pointing
5546 at a message produces an empty menu entitled "No message selected". This
5547 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
5548 no entries in it ("Shell widget menu has zero width and/or height"). So I
5549 have added a single, blank menu entry in this case.
5551 9. Added ${quote_local_part.
5553 10. MIME decoding is now applied to the contents of Subject: header lines when
5556 11. Now that a reference to $sender_host_address automatically causes a reverse
5557 lookup to occur if necessary (4.13/18), there is no need to arrange for a
5558 host lookup before query-style lookups in lists that might use this
5559 variable. This has therefore been abolished, and the "net-" prefix is no
5560 longer necessary for query-style lookups.
5562 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
5563 have been a typo for LFLAGS, so it has been changed.
5565 13. The install script calls Exim with "-C /dev/null" in order to find the
5566 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
5567 to be output. Howeve, since Exim outputs its version number before the
5568 error, it didn't break the script. It just looked ugly. I fixed this by
5569 always allowing "-C /dev/null" if the caller is root.
5571 14. Ignore overlarge ACL variable number when reading spool file - insurance
5572 against a later release with more variables having written the file.
5574 15. The standard form for an IPv6 address literal was being rejected by EHLO.
5575 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5576 this, as well as the form without the "IPv6" on the front.
5578 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
5579 OS/Makefile-Darwin file.
5581 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
5582 only to LDAP libraries that do not have LDAP_OPT_DEREF.
5584 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
5585 variable. However, some OS use ints for inodes. I've added cast to long int
5586 to get rid of the compiler warning.
5588 19. I had forgotten to lock out "/../" in configuration file names when
5589 ALT_CONFIG_PREFIX was set.
5591 20. Routers used for verification do not need to specify transports. However,
5592 if such a router generated a host list, and callout was configured, Exim
5593 crashed, because it could not find a port number from the (non-existent)
5594 transport. It now assumes port 25 in this circumstance.
5596 21. Added the -t option to exigrep.
5598 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
5599 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
5600 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
5601 them is set, LOOKUP_LSEARCH is forced.
5603 23. "exim -bV" now outputs a list of lookups that are included in the binary.
5605 24. Added sender and host information to the "rejected by local_scan()" log
5606 line; previously there was no indication of these.
5608 25. Added .include_if_exists.
5610 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
5611 Linux. It turns out that not all file systems support this. Apparently some
5612 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
5613 do it.) To cope with this, the error EINVAL, which means that sync-ing is
5614 not supported on the file descriptor, is now ignored when Exim is trying to
5615 sync a directory. This applies only to Linux.
5617 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
5619 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
5620 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
5623 29. Added support for maildirsize files from supplied patch (modified a bit).
5625 30. The use of :fail: followed by an empty string could lead Exim to respond to
5626 sender verification failures with (e.g.):
5628 550 Verification failed for <xxx>
5629 550 Sender verify failed
5631 where the first response line was missing the '-' that indicates it is not
5632 the final line of the response.
5634 31. The loop for finding the name of the user that called Exim had a hardwired
5635 limit of 10; it now uses the value of finduser_retries, which is used for
5636 all other user lookups.
5638 32. Added $received_count variable, available in data and not_smtp ACLs, and at
5641 33. Exim was neglecting to zero errno before one call of strtol() when
5642 expanding a string and expecting an integer value. On some systems this
5643 resulted in spurious "integer overflow" errors. Also, it was casting the
5644 result into an int without checking.
5646 34. Testing for a connection timeout using "timeout_connect" in the retry rules
5647 did not work. The code looks as if it has *never* worked, though it appears
5648 to have been documented since at least releast 1.62. I have made it work.
5650 35. The "timeout_DNS" error in retry rules, also documented since at least
5651 1.62, also never worked. As it isn't clear exactly what this means, and
5652 clearly it isn't a major issue, I have abolished the feature by treating it
5653 as "timeout", and writing a warning to the main and panic logs.
5655 36. The display of retry rules for -brt wasn't always showing the error code
5658 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
5659 timeout_connect_A, timeout_connect_MX.
5661 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
5662 to the empty sender.
5664 39. The daemon was not analysing the content of -oX till after it had closed
5665 stderr and disconnected from the controlling terminal. This meant that any
5666 syntax errors were only noted on the panic log, and the return code from
5667 the command was 0. By re-arranging the code a little, I've made the
5668 decoding happen first, so such errors now appear on stderr, and the return
5669 code is 1. However, the actual setting up of the sockets still happens in
5670 the disconnected process, so errors there are still only recorded on the
5673 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
5674 connections (as happens on some IP stacks) was logged at start up time as
5675 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
5676 it from "IPv6 and IPv4", which means that two separate sockets are being
5679 41. The debug output for gethostbyname2() or getipnodebyname() failures now
5680 says whether AF_INET or AF_INET6 was passed as an argument.
5682 42. Exiwhat output was messed up when time zones were included in log
5685 43. Exiwhat now gives more information about the daemon's listening ports,
5686 and whether -tls-on-connect was used.
5688 44. The "port" option of the smtp transport is now expanded.
5690 45. A "message" modifier in a "warn" statement in a non-message ACL was being
5691 silently ignored. Now an error message is written to the main and panic
5694 46. There's a new ACL modifier called "logwrite" which writes to a log file
5695 as soon as it is encountered.
5697 47. Added $local_user_uid and $local_user_gid at routing time.
5699 48. Exim crashed when trying to verify a sender address that was being
5702 49. Exim was recognizing only a space character after ".include". It now also
5703 recognizes a tab character.
5705 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
5706 extracting the relevant information from the specification. The man page no
5707 longer contains scrambled data for the -d option, and I've added a section
5708 at the front about calling Exim under different names.
5710 51. Added "extra_headers" argument to the "mail" command in filter files.
5712 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
5715 53. Installed eximstats 1.29.
5717 54. Added transport_filter_timeout as a generic transport option.
5719 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
5720 Cc: header lines. This was required by RFC 822, but it not required by RFC
5723 56. Exim used to add From:, Date:, and Message-Id: header lines to any
5724 incoming messages that did not have them. Now it does so only if the
5725 message originates locally, that is, if there is no associated remote host
5726 address. When Resent- header lines are present, this applies to the Resent-
5727 lines rather than the non-Resent- lines.
5729 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
5730 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
5732 58. Messages for configuration errors now include the name of the main
5733 configuration file - useful now that there may be more than one file in a
5734 list (.included file names were always shown).
5736 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
5737 for those rare installations that do not start the daemon as root or run it
5738 setuid root. I've cut out the call to initgroups() if the daemon is not
5741 60. The Exim user and group can now be bound into the binary as text strings
5742 that are looked up at the start of Exim's processing.
5744 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
5746 62. Added $mailstore_basename variable.
5748 63. Installed patch to sieve.c from Michael Haardt.
5750 64. When Exim failed to open the panic log after failing to open the main log,
5751 the original message it was trying to log was written to stderr and debug
5752 output, but if they were not available (the usual case in production), it
5753 was lost. Now it is written to syslog before the two lines that record the
5754 failures to open the logs.
5756 65. Users' Exim filters run in subprocesses under the user's uid. It is
5757 possible for a "deliver" command or an alias in a "personal" command to
5758 provoke an address rewrite. If logging of address rewriting is configured,
5759 this fails because the process is not running as root or exim. There may be
5760 a better way of dealing with this, but for the moment (because 4.30 needs
5761 to be released), I have disabled address rewrite logging when running a
5762 filter in a non-root, non-exim process.
5768 1. The buildconfig auxiliary program wasn't quoting the value set for
5769 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
5770 not defined. This bug was masked in 4.22 by the effect that was fixed in
5773 2. Some messages that were rejected after a message id was allocated were
5774 shown as "incomplete" by exigrep. It no longer does this for messages that
5775 are rejected by local_scan() or the DATA or non-SMTP ACLs.
5777 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
5778 have allow_domain_literals set, the ID did not get logged in the <= line.
5779 Domain literals are now always recognized in Message-ID: header lines.
5781 4. The first argument for a ${extract expansion item is the key name or field
5782 number. Leading and trailing spaces in this item were not being ignored,
5783 causing some misleading effects.
5785 5. When deliver_drop_privilege was set, single queue runner processes started
5786 manually (i.e. by the command "exim -q") or by the daemon (which uses the
5787 same command in the process it spins off) were not dropping privilege.
5789 6. When the daemon running as "exim" started a queue runner, it always
5790 re-executed Exim in the spun-off process. This is a waste of effort when
5791 deliver_drop_privilege is set. The new process now just calls the
5792 queue-runner function directly.
5798 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
5801 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
5802 ignored. Though the use of -oP was forcing the writing of a pid file, it
5803 was always written to the default place.
5805 3. If the message "no IP address found for host xxxx" is generated during
5806 incoming verification, it is now followed by identification of the incoming
5807 connection (so you can more easily find what provoked it).
5809 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
5811 5. Added some features to "harden" Exim a bit more against certain attacks:
5813 (a) There is now a build-time option called FIXED_NEVER_USERS that can
5814 be put in Local/Makefile. This is like the never_users runtime option,
5815 but it cannot be overridden. The default setting is "root".
5817 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
5818 prefix string with which any file named in a -C command line option
5821 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
5822 is retained for -C and -D only if the caller of Exim is root. Without
5823 it, the exim user may also use -C and -D and retain privilege.
5825 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
5826 command line option is disabled.
5828 6. Macro names set by the -D option must start with an upper case letter, just
5829 like macro names defined in the configuration file.
5831 7. Added "dereference=" facility to LDAP.
5833 8. Two instances of the typo "uknown" in the source files are fixed.
5835 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
5836 the Configure-Makefile script screwed up while processing it.
5838 10. Incorporated PCRE 4.4.
5840 11. The SMTP synchronization check was not operating right at the start of an
5841 SMTP session. For example, it could not catch a HELO sent before the client
5842 waited for the greeting. There is now a check for outstanding input at the
5843 point when the greeting is written. Because of the duplex, asynchronous
5844 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
5845 way, but not yet received, when the check is performed.
5847 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
5848 on TCP/IP sockets, because this apparently causes some broken clients to
5851 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
5852 unchanged) from the Cygwin maintainer.
5854 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
5855 was supported instead of testing for mbx. Effectively a typo.
5857 15. The spa authenticator server code was not checking that the input it
5858 received was valid base64.
5860 16. The debug output line for the "set" modifier in ACLs was not showing the
5861 name of the variable that was being set.
5863 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
5865 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
5866 lookup on its own. Something else was needed to trigger the lookup. For
5867 example, a match in host_lookup or the need for a host name in a host list.
5868 Now, if $sender_host_name is referenced and the host name has not yet been
5869 looked up, a lookup is performed. If the lookup fails, the variable remains
5870 empty, and $host_lookup_failed is set to "1".
5872 19. Added "eqi" as a case-independent comparison operator.
5874 20. The saslauthd authentication condition could segfault if neither service
5875 nor realm was specified.
5877 21. If an overflowing value such as "2048M" was set for message_size_limit, the
5878 error message that was logged was misleading, and incoming SMTP
5879 connections were dropped. The message is now more accurate, and temporary
5880 errors are given to SMTP connections.
5882 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
5883 (except RSET) with a 421 error, until QUIT is received. However, it was
5884 failing to send a response to QUIT.
5886 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
5887 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
5888 ACL is now run after the helo_try_verify_hosts code.
5890 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
5891 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
5892 case-independent, so other case variants are also recognized). Apparently
5893 some systems use these upper case variants.
5895 25. If more than two messages were waiting for the same host, and a transport
5896 filter was specified for the transport, Exim sent two messages over the
5897 same TCP/IP connection, and then failed with "socket operation on non-
5898 socket" when it tried to send the third.
5900 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
5902 27. The extern definition of crypt16() in expand.c was not being excluded when
5903 the OS had its own crypt16() function.
5905 28. Added bounce_return_body as a new option, and bounce_return_size_limit
5906 as a preferred synonym for return_size_limit, both as an option and as an
5909 29. Added LIBS=-liconv to OS/Makefile-OSF1.
5911 30. Changed the default configuration ACL to relax the local part checking rule
5912 for addresses that are not in any local domains. For these addresses,
5913 slashes and pipe symbols are allowed within local parts, but the sequence
5914 /../ is explicitly forbidden.
5916 31. SPA server authentication was not clearing the challenge buffer before
5919 32. log_message in a "warn" ACL statement was writing to the reject log as
5920 well as to the main log, which contradicts the documentation and doesn't
5921 seem right (because no rejection is happening). So I have stopped it.
5923 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
5924 However, I am unable to do any testing of this.
5926 34. Fixed an infelicity in the appendfile transport. When checking directories
5927 for a mailbox, to see if any needed to be created, it was accidentally
5928 using path names with one or more superfluous leading slashes; tracing
5929 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
5931 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
5932 contents are added to the log line that is written for every discarded
5933 recipient. (Previously a log_message setting was ignored.)
5935 36. The ${quote: operator now quotes the string if it is empty.
5937 37. The install script runs exim in order to find its version number. If for
5938 some reason other than non-existence or emptiness, which it checks, it
5939 could not run './exim', it was installing it with an empty version number,
5940 i.e. as "exim-". This error state is now caught, and the installation is
5943 38. An argument was missing from the function that creates an error message
5944 when Exim fails to connect to the socket for saslauthd authentication.
5945 This could cause Exim to crash, or give a corrupted message.
5947 39. Added isip, isip4, and isip6 to ${if conditions.
5949 40. The ACL variables $acl_xx are now saved with the message, and can be
5950 accessed later in routers, transports, and filters.
5952 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
5953 strings in the file are not string-expanded.
5955 42. If a MAIL command specified a SIZE value that was too large to fit into an
5956 int variable, the check against message_size_limit failed. Such values are
5957 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
5958 day this will have to be increased, but I don't think I want to be around
5959 when emails are that large.
5966 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
5967 iconv() is not standard in FreeBSD.
5969 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
5970 IPv6 enabled. The observed symptom was a segmentation fault on return from
5971 the function os_common_find_running_interfaces() in src/os.c.
5973 3. In the check_special_case() function in daemon.c I had used "errno" as an
5974 argument name, which causes warnings on some systems. This was basically a
5975 typo, since it was named "eno" in the comments!
5977 4. The code that waits for the clock to tick (at a resolution of some fraction
5978 of a second) so as to ensure message-id uniqueness was always waiting for
5979 at least one whole tick, when it could have waited for less. [This is
5980 almost certainly not relevant at current processor speeds, where it is
5981 unlikely to ever wait at all. But we try to future-proof.]
5983 5. The function that sleeps for a time interval that includes fractions of a
5984 second contained a race. It did not block SIGALRM between setting the
5985 timer, and suspending (a couple of lines later). If the interval was short
5986 and the sigsuspend() was delayed until after it had expired, the suspension
5987 never ended. On busy systems this could lead to processes getting stuck for
5990 6. Some uncommon configurations may cause a lookup to happen in a queue runner
5991 process, before it forks any delivery processes. The open lookup caching
5992 mechanism meant that the open file or database connection was passed into
5993 the delivery process. The problem was that delivery processes always tidy
5994 up cached lookup data. This could cause a problem for the next delivery
5995 process started by the queue runner, because the external queue runner
5996 process does not know about the closure. So the next delivery process
5997 still has data in the lookup cache. In the case of a file lookup, there was
5998 no problem because closing a file descriptor in a subprocess doesn't affect
5999 the parent. However, if the lookup was caching a connection to a database,
6000 the connection was closed, and the second delivery process was likely to
6001 see errors such as "PGSQL: query failed: server closed the connection
6002 unexpectedly". The problem has been fixed by closing all cached lookups
6003 in a queue runner before running a delivery process.
6005 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
6006 seem to have the "const" qualifier which it has on other OS. I've
6009 8. Change 4.21/2 was too strict. It is only if there are two authenticators
6010 *of the same type* (client or server) with the same public name that an
6011 error should be diagnosed.
6013 9. When Exim looked up a host name for an IP address, but failed to find the
6014 original IP address when looking up the host name (a safety check), it
6015 output the message "<ip address> does not match any IP for NULL", which was
6016 confusing, to say the least. The bug was that the host name should have
6017 appeared instead of "NULL".
6019 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
6020 user that is built into the binary, and the -C or -D options is used, root
6021 privilege is dropped before the configuration file is read. In addition,
6022 logging is switched to stderr instead of the normal log files. If the
6023 configuration then re-defines the Exim user, the unprivileged environment
6024 is probably not what is expected, so Exim logs a panic warning message (but
6027 However, if deliver_drop_privilege is set, the unprivileged state may well
6028 be exactly what is intended, so the warning has been cut out in that case,
6029 and Exim is allowed to try to write to its normal log files.
6035 1. smtp_return_error_details was not giving details for temporary sender
6036 or receiver verification errors.
6038 2. Diagnose a configuration error if two authenticators have the same public
6041 3. Exim used not to create the message log file for a message until the first
6042 delivery attempt. This could be confusing when incoming messages were held
6043 for policy or load reasons. The message log file is now created at the time
6044 the message is received, and an initial "Received" line is written to it.
6046 4. The automatically generated man page for command line options had a minor
6047 bug that caused no ill effects; however, a more serious problem was that
6048 the procedure for building the man page automatically didn't always
6049 operate. Consequently, release 4.20 contains an out-of-date version. This
6050 shouldn't happen again.
6052 5. When building Exim with embedded Perl support, the script that builds the
6053 Makefile was calling 'perl' to find its compile-time parameters, ignoring
6054 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
6056 6. The freeze_tell option was not being used for messages that were frozen on
6057 arrival, either by an ACL or by local_scan().
6059 7. Added the smtp_incomplete_transaction log selector.
6061 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
6062 was accepting AUTH without a new EHLO.
6064 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
6065 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
6068 10. Logging of TCP/IP connections (when configured) now happens in the main
6069 daemon process instead of the child process, so that the TCP/IP connection
6070 count is more accurate (but it can never be perfect).
6072 11. The use of "drop" in a nested ACL was not being handled correctly in the
6073 outer ACL. Now, if condition failure induced by the nested "drop" causes
6074 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
6075 or "require"), the connection is dropped.
6077 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
6078 that yield "discard" can now be used with an "accept" or a "discard" verb,
6079 but an error is generated for any others (because I can't see a useful way
6080 to define what should happen).
6082 13. When an ACL is read dynamically from a file (or anywhere else), the lines
6083 are now processed in the same way as lines in the Exim configuration file.
6084 In particular, continuation lines are supported.
6086 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
6088 15. Added -ti meaning -t -i.
6090 16. Check for letters, digits, hyphens, and dots in the names of dnslist
6091 domains, and warn by logging if others are found.
6093 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
6094 returned from GIFCONF when Exim is trying to find the list of interfaces on
6095 a host. The code in os.c has been modified to copy each ifreq to an aligned
6096 structure in all cases.
6098 Also, in some cases, the returned ifreq's were being copied to a 'struct
6099 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
6100 means the last couple of bytes of an IPv6 address could be chopped if the
6101 ifreq contained only a normal sockaddr (14 bytes storage).
6103 18. Named domain lists were not supported in the hosts_treat_as_local option.
6104 An entry such as +xxxx was not recognized, and was treated as a literal
6107 19. Ensure that header lines added by a DATA ACL are included in the reject log
6108 if the ACL subsequently rejects the message.
6110 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
6111 MD5 (which is deprecated).
6113 21. When testing a filter file using -bf, Exim was writing a message when it
6114 took the sender from a "From " line in the message, but it was not doing so
6115 when it took $return_path from a Return-Path: header line. It now does.
6117 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
6118 with a valid header line field name (a series of printing characters
6119 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
6121 23. Changed "disc" in the source to "disk" to conform to the documentation and
6122 the book and for uniformity.
6124 24. Ignore Sendmail's -Ooption=value command line item.
6126 25. When execve() failed while trying to run a command in a pipe transport,
6127 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
6128 could be confused with a return value of 69 from the command itself. This
6129 has been changed to 127, the value the shell returns if it is asked to run
6130 a non-existent command. The wording for the related log line suggests a
6131 non-existent command as the problem.
6133 26. If received_header_text expands to an empty string, do not add a Received:
6134 header line to the message. (Well, it adds a token one on the spool, but
6135 marks it "old" so that it doesn't get used or transmitted.)
6137 27. Installed eximstats 1.28 (addition of -nt option).
6139 28. There was no check for failure on the call to getsockname() in the daemon
6140 code. This can fail if there is a shortage of resources on the system, with
6141 ENOMEM, for example. A temporary error is now given on failure.
6143 29. Contrary to the C standard, it seems that in some environments, the
6144 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
6145 program. Exim now does this explicitly; it affects the formatting of
6146 timestamps using strftime().
6148 30. If exiqsumm was given junk data, it threw up some uninitialized variable
6149 complaints. I've now initialized all the variables, to avoid this.
6151 32. Header lines added by a system filter were not being "seen" during
6152 transport-time rewrites.
6154 33. The info_callback() function passed to OpenSSL is set up with type void
6155 (*)(SSL *, int, int), as described somewhere. However, when calling the
6156 function (actually a macro) that sets it up, the type void(*)() is
6157 expected. I've put in a cast to prevent warnings from picky compilers.
6159 34. If a DNS black list lookup found a CNAME record, but there were no A
6160 records associated with the domain it pointed at, Exim crashed.
6162 35. If a DNS black list lookup returned more than one A record, Exim ignored
6163 all but the first. It now scans all returned addresses if a particular IP
6164 value is being sought. In this situation, the contents of the
6165 $dnslist_value variable are a list of all the addresses, separated by a
6168 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
6169 to accept a host name and all its aliases if the forward lookup for any of
6170 them yielded the IP address of the incoming connection. Now it accepts only
6171 those names whose forward lookup yields the correct IP address. Any other
6172 names are discarded. This closes a loophole whereby a rogue DNS
6173 administrator could create reverse DNS records to break through a
6174 wildcarded host restriction in an ACL.
6176 37. If a user filter or a system filter that ran in a subprocess used any of
6177 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
6178 the wrong values were passed to the pipe command ($thisaddress had the
6179 value of $0, $0 had the value of $1, etc). This bug was introduced by
6180 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
6182 38. Improved the line breaking for long SMTP error messages from ACLs.
6183 Previously, if there was no break point between 40 and 75 characters, Exim
6184 left the rest of the message alone. Two changes have been made: (a) I've
6185 reduced the minimum length to 35 characters; (b) if it can't find a break
6186 point between 35 and 75 characters, it looks ahead and uses the first one
6187 that it finds. This may give the occasional overlong line, but at least the
6188 remaining text gets split now.
6190 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
6191 file descriptors might be low, and that setting 1000 would always raise it.
6192 It turns out that in some environments, the limit is already over 1000 and
6193 that lowering it causes trouble. So now Exim takes care not to decrease it.
6195 40. When delivering a message, the value of $return_path is set to $sender_
6196 address at the start of routing (routers may change the value). By an
6197 oversight, this default was not being set up when an address was tested by
6198 -bt or -bv, which affected the outcome if any router or filter referred to
6201 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
6202 log only when "message" or "log_message" are set. However, if one of the
6203 conditions was an address verification, or a call to a nested ACL, the
6204 messages generated by the underlying test were being passed through. This
6205 no longer happens. The underlying message is available in $acl_verify_
6206 message for both "message" and "log_message" expansions, so it can be
6207 passed through if needed.
6209 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
6210 new expansion $bh_ to give the encoded byte string without charset
6211 translation. Translation happens only if iconv() is available; HAVE_ICONV
6212 indicates this at build time. HEADERS_CHARSET gives the charset to
6213 translate to; headers_charset can change it in the configuration, and
6214 "headers charset" can change it in an individual filter file.
6216 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
6217 that creates RFC 2047 encoded "words" labels them as that charset instead
6218 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
6219 expansion operator; (ii) when Exim creates a From: line for a local
6220 message; (iii) when a header line is rewritten to include a "phrase" part.
6222 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
6223 buggy, causing it to skip the first lines of messages whose message ID
6224 ended in 'D'. This would not have bitten before Exim release 4.14, because
6225 message IDs were unlikely to end in 'D' before then. The effect was to have
6226 incorrect size information for certain domains.
6228 45. #include "config.h" was missing at the start of the crypt16.c module. This
6229 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
6232 46. If there was a timeout during a "random" callout check, Exim treated it as
6233 a failure of the random address, and carried on sending RSET and the real
6234 address. If the delay was just some slowness somewhere, the response to the
6235 original RCPT would be taken as a response to RSET and so on, causing
6236 mayhem of various kinds.
6238 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
6239 when I implemented it. It didn't allow for the fact that some option values
6240 may legitimatetly be negative (e.g. size_addition), and it didn't even do
6241 the right test for positive values.
6243 48. Domain names in DNS records are case-independent. Exim always looks them up
6244 in lower case. Some resolvers return domain names in exactly the case they
6245 appear in the zone file, that is, they may contain uppercase letters. Not
6246 all resolvers do this - some return always lower case. Exim was treating a
6247 change of case by a resolver as a change of domain, similar to a widening
6248 of a domain abbreviation. This triggered its re-routing code and so it was
6249 trying to route what was effectively the same domain again. This normally
6250 caused routing to fail (because the router wouldn't handle the domain
6251 twice). Now Exim checks for this case specially, and just changes the
6252 casing of the domain that it ultimately uses when it transmits the message
6255 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
6258 50. If a filter generated a file delivery with a non-absolute name (possible if
6259 no home directory exists for the router), the forbid_file option was not
6262 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
6263 the existing equality matching.
6265 52. Exim was using ints instead of ino_t variables in some places where it was
6266 dealing with inode numbers.
6268 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
6269 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
6270 called TMPDIR, and if it finds it is different, it changes its value.
6272 54. The smtp_printf() function is now made available to local_scan() so
6273 additional output lines can be written before returning. There is also an
6274 smtp_fflush() function to enable the detection of a dropped connection.
6275 The variables smtp_input and smtp_batched_input are exported to
6278 55. Changed the default runtime configuration: the message "Unknown user"
6279 has been removed from the ACL, and instead placed on the localuser router,
6280 using the cannot_route_message feature. This means that any verification
6281 failures that generate their own messages won't get overridden. Similarly,
6282 the "Unrouteable address" message that was in the ACL for unverifiable
6283 relay addresses has also been removed.
6285 56. Added hosts_avoid_esmtp to the smtp transport.
6287 57. The exicyclog script was not checking for the esoteric option
6288 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
6289 will work only if exicyclog is run under the appropriate euid.
6291 58. Following a discussion on the list, the rules by which Exim recognises line
6292 endings on incoming messages have been changed. The -dropcr and drop_cr
6293 options are now no-ops, retained only for backwards compatibility. The
6294 following line terminators are recognized: LF CRLF CR. However, special
6295 processing applies to CR:
6297 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
6298 nor a local message in the state where . is a terminator.
6300 (ii) If a bare CR is encountered in a header line, an extra space is added
6301 after the line terminator so as not to end the header. The reasoning
6302 behind this is that bare CRs in header lines are most likely either
6303 to be mistakes, or people trying to play silly games.
6305 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
6306 was being incorrectly given as 18 bytes larger than it should have been.
6307 This is a VOB (very old bug).
6309 60. This may never have affected anything current, but just in case it has:
6310 When the local host is found other than at the start of a list of hosts,
6311 the local host, those with the same MX, and any that follow, are discarded.
6312 When the list in question was part of a longer list of hosts, the following
6313 hosts (not currently being processed) were also being discarded. This no
6314 longer happens. I'm not sure if this situation could ever has previously
6317 61. Added the "/MX" feature to lists of hosts in the manualroute and query
6320 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
6321 header. This is something that is recommended in a new Internet Draft, and
6322 is something that is documented as being done by Sendmail. There are two
6323 possible values. For messages generated by the autoreply transport, Exim
6326 Auto-Submitted: auto-replied
6328 whereas for all other generated messages (e.g. bounces) it adds
6330 Auto-Submitted: auto-generated
6332 63. The "personal" condition in filters now includes a test for the
6333 Auto-Submitted: header. If it contains the string "auto-" the message it
6334 not considered personal.
6336 64. Added rcpt_include_affixes as a generic transport option.
6338 65. Added queue_only_override (default true).
6340 66. Added the syslog_duplication option.
6342 67. If what should have been the first header line of a message consisted of
6343 a space followed by a colon, Exim was mis-interpreting it as a header line.
6344 It isn't of course - it is syntactically invalid and should therefore be
6345 treated as the start of the message body. The misbehaviour could have
6346 caused a number of strange effects, including loss of data in subsequent
6347 header lines, and spool format errors.
6349 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
6350 client host had authenticated. This control can now be exercised by an ACL
6351 for more flexibility.
6353 69. By default, callouts do not happen when testing with -bh. There is now a
6354 variant, -bhc, which does actually run the callout code, including
6355 consulting and updating the callout cache.
6357 70. Added support for saslauthd authentication, courtesy of Alexander
6360 71. If statvfs() failed on the spool or log directories while checking their
6361 size for availability, Exim confusingly gave the error "space shortage".
6362 Furthermore, in debugging mode it crashed with a floating point exception.
6363 These checks are done if check_{spool,log}_{space,inodes} are set, and when
6364 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
6365 serious problem, Exim now writes to the main and panic logs when this
6366 happens, with details of the failure. It then refuses to accept the
6367 incoming message, giving the message "spool directory problem" or "log
6368 directory problem" with a 421 code for SMTP messages.
6370 72. When Exim is about to re-exec itself, it ensures that the file descriptors
6371 0, 1, and 2 exist, because some OS complain for execs without them (see
6372 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
6373 descriptors. However, the code omitted to check that the open succeeded,
6374 causing mysterious errors if for some reason the permissions on /dev/null
6375 got screwed. Now Exim writes a message to the main and panic logs, and
6376 bombs out if it can't open /dev/null.
6378 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
6379 interact so that it is all more flexible. It is supposed to remain
6380 backwards compatible. Also added extra_local_interfaces.
6382 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
6383 to bomb out with an assertion failure - to the client this appears as a
6384 connection drop. This problem occurs in the part of the code that was taken
6385 from the Samba project. Fortunately, the assertion is in a very simple
6386 function, so I have fixed this by reproducing the function inline in the
6387 one place where it is called, and arranging for authentication to fail
6388 instead of killing the process with assert().
6390 75. The SPA client code was not working when the server requested OEM rather
6391 than Unicode encoding.
6393 76. Added code to make require_files with a specific uid setting more usable in
6394 the case where statting the file as root fails - usually a non-root-mounted
6395 NFS file system. When this happens and the failure is EACCES, Exim now
6396 forks a subprocess and does the per-uid checking as the relevant uid.
6398 77. Added process_log_path.
6400 78. If log_file_path was not explicitly set, a setting of check_log_space or
6401 check_log_inodes was ignored.
6403 79. If a space check for the spool or log partitions fails, the incident is now
6404 logged. Of course, in the latter case the data may get lost...
6406 80. Added the %p formatting code to string_format() so that it can be used to
6407 print addresses in debug_print(). Adjusted all the address printing in the
6408 debugging in store.c to use %p rather than %d.
6410 81. There was a concern that a line of code in smtp_in.c could overflow a
6411 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
6412 initially expressed, the concern was not well-founded, because trailing
6413 spaces are removed early. However, if the trailing spaces were followed by
6414 a NULL, they did not get removed, so the overflow was possible. Two fixes
6417 (a) I re-wrote the offending code in a cleaner fashion.
6418 (b) If an incoming SMTP command contains a NULL character, it is rejected
6421 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
6422 runs initgroups(), so that if the Exim user is in any additional groups,
6423 they will be used during message reception.
6429 The change log for 4.20 and earlier releases has been archived.
projects / exim.git / blob