1 Change log file for Exim from version 4.21
2 -------------------------------------------
6 JH/01 Use SIZE on MAIL FROM in a cutthrough connection, if the destination
7 supports it and a size is available (ie. the sending peer gave us one).
9 JH/02 The obsolete acl condition "demime" is removed (finally, after ten
10 years of being deprecated). The replacements are the ACLs
11 acl_smtp_mime and acl_not_smtp_mime.
13 JH/03 Upgrade security requirements imposed for hosts_try_dane: previously
14 a downgraded non-dane trust-anchor for the TLS connection (CA-style)
15 or even an in-clear connection were permitted. Now, if the host lookup
16 was dnssec and dane was requested then the host is only used if the
17 TLSA lookup succeeds and is dnssec. Further hosts (eg. lower priority
18 MXs) will be tried (for hosts_try_dane though not for hosts_require_dane)
19 if one fails this test.
20 This means that a poorly-configured remote DNS will make it incommunicado;
21 but it protects against a DNS-interception attack on it.
23 JH/04 Bug 1810: make continued-use of an open smtp transport connection
24 non-noisy when a race steals the message being considered.
26 JH/05 If main configuration option tls_certificate is unset, generate a
27 selfsigned certificate for inbound TLS connections.
29 JH/06 Bug 165: hide more cases of password exposure - this time in expansions
30 in rewrites and routers.
32 JH/07 Retire gnutls_require_mac et.al. These were nonfunctional since 4.80
33 and logged a warning sing 4.83; now they are a configuration file error.
35 JH/08 Bug 1836: Fix crash in VRFY handling when handed an unqualified name
36 (lacking @domain). Apply the same qualification processing as RCPT.
38 JH/09 Bug 1804: Avoid writing msglog files when in -bh or -bhc mode.
43 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
44 and 3.4.4 - once the server is enabled to respond to an OCSP request
45 it does even when not requested, resulting in a stapling non-aware
46 client dropping the TLS connection.
48 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
49 support variable-length bit vectors. No functional change.
51 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
52 The I= interface field on outgoing lines is now after the H= remote
53 host field, same as incoming lines. There is a separate
54 outgoing_interface log selector which allows you to disable the
57 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
58 If not running log_selector +smtp_connection the mainlog would be held
59 open indefinitely after a "too many connections" event, including to a
60 deleted file after a log rotate. Leave the per net connection logging
61 leaving it open for efficiency as that will be quickly detected by the
62 check on the next write.
64 HS/01 Bug 1671: Fix post transport crash.
65 Processing the wait-<transport> messages could crash the delivery
66 process if the message IDs didn't exist for some reason. When
67 using 'split_spool_directory=yes' the construction of the spool
68 file name failed already, exposing the same netto behaviour.
70 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
71 mime_regex ACL conditions.
73 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
74 to DSN fail messages (bounces): remote IP, remote greeting, remote response
75 to HELO, local diagnostic string.
77 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
78 log line to debug. Even when configured with a tls authenticator many
79 client connections are expected to not authenticate in this way, so
80 an authenticate fail is not an error.
82 HS/02 Add the Exim version string to the process info. This way exiwhat
83 gives some more detail about the running daemon.
85 JH/06 Bug 1395: time-limit cacheing of DNS lookups, to the TTL value. This may
86 matter for fast-change records such as DNSBLs.
88 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
89 retry record, even if constant. There may be multiple transports with
90 different interface settings and the retry behaviour needs to be kept
93 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
95 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
97 JH/10 Bug 840: fix log_defer_output option of pipe transport
99 JH/11 Bug 830: use same host for all RCPTS of a message, even under
100 hosts_randomize. This matters a lot when combined with mua_wrapper.
102 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
103 ${quote_pgsql:<string>} operator.
105 JH/13 Bug 1708: avoid misaligned access in cached lookup.
107 JH/14 Change header file name for freeradius-client. Relevant if compiling
108 with Radius support; from the Gentoo tree and checked under Fedora.
110 JH/15 Bug 1712: Introduce $prdr_requested flag variable
112 JH/16 Bug 1714: Permit an empty string as expansion result for transport
113 option transport_filter, meaning no filtering.
115 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
117 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
118 defaults to "*" (all hosts). The variable is now available when not built
119 with TLS, default unset, mainly to enable keeping the testuite sane.
120 If a server certificate is not supplied (via tls_certificate) an error is
121 logged, and clients will find TLS connections fail on startup. Presumably
122 they will retry in-clear.
123 Packagers of Exim are strongly encouraged to create a server certificate
124 at installation time.
126 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
127 with the $config_file variable.
129 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
130 in transport context, after the attempt, and per-recipient. The latter type
131 is per host attempted. The event data is the error message, and the errno
132 information encodes the lookup type (A vs. MX) used for the (first) host,
133 and the trailing two digits of the smtp 4xx reponse.
135 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
136 to write to mainlog (or rejectlog, paniclog) in the window between file
137 creation and permissions/ownership being changed. Particularly affects
138 installations where exicyclog is run as root, rather than exim user;
139 result is that the running daemon panics and dies.
141 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
143 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
144 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
145 "pri" and "weight". Note that the previous implicit priority given by the
146 list order is no longer honoured.
148 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalisation
151 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
152 by defining SUPPORT_SOCKS.
154 JH/26 Move PROXY support from Experimental to mainline, enabled for a build
155 by defining SUPPORT_PROXY. Note that the proxy_required_hosts option
156 is renamed to hosts_proxy, and the proxy_{host,target}_{address,port}.
157 variables are renamed to proxy_{local,external}_{address,port}.
159 JH/27 Move Internationalisation support from Experimental to mainline, enabled
160 for a build by defining SUPPORT_I18N
162 JH/28 Bug 1745: Fix redis lookups to handle (quoted) spaces embedded in parts
163 of the query string, and make ${quote_redis:} do that quoting.
165 JH/29 Move Events support from Experimental to mainline, enabled by default
166 and removable for a build by defining DISABLE_EVENT.
168 JH/30 Updated DANE implementation code to current from Viktor Dukhovni.
170 JH/31 Fix bug with hosts_connection_nolog and named-lists which were wrongly
171 cached by the daemon.
173 JH/32 Move Redis support from Experimental to mainline, enabled for a build
174 by defining LOOKUP_REDIS. The libhiredis library is required.
176 JH/33 Bug 1748: Permit ACL dnslists= condition in non-smtp ACLs if explicit
177 keys are given for lookup.
179 JH/34 Bug 1192: replace the embedded copy of PolarSSL RSA routines in the DKIM
180 support, by using OpenSSL or GnuTLS library ones. This means DKIM is
181 only supported when built with TLS support. The PolarSSL SHA routines
182 are still used when the TLS library is too old for convenient support.
184 JH/35 Require SINGLE_DH_USE by default in OpenSSL (main config option
185 openssl_options), for security. OpenSSL forces this from version 1.1.0
186 server-side so match that on older versions.
188 JH/36 Bug 1778: longstanding bug in memory use by the ${run } expansion: A fresh
189 allocation for $value could be released as the expansion processing
190 concluded, but leaving the global pointer active for it.
192 JH/37 Bug 1769: Permit a VRFY ACL to override the default 252 response,
193 and to use the domains and local_parts ACL conditions.
195 JH/38 Fix cutthrough bug with body lines having a single dot. The dot was
196 incorrectly not doubled on cutthrough transmission, hence seen as a
197 body-termination at the receiving system - resulting in truncated mails.
198 Commonly the sender saw a TCP-level error, and retransmitted the message
199 via the normal store-and-forward channel. This could result in duplicates
200 received - but deduplicating mailstores were liable to retain only the
201 initial truncated version.
203 JH/39 Bug 1781: Fix use of DKIM private-keys having trailing '=' in the base-64.
205 JH/40 Fix crash in queryprogram router when compiled with EXPERIMENTAL_SRS.
207 JH/41 Bug 1792: Fix selection of headers to sign for DKIM: bottom-up. While
208 we're in there, support oversigning also; bug 1309.
210 JH/42 Bug 1796: Fix error logged on a malware scanner connection failure.
212 HS/04 Add support for keep_environment and add_environment options.
214 JH/43 Tidy coding issues detected by gcc --fsanitize=undefined. Some remain;
215 either intentional arithmetic overflow during PRNG, or testing config-
218 JH/44 Bug 1800: The combination of a -bhc commandline option and cutthrough
219 delivery resulted in actual delivery. Cancel cutthrough before DATA
222 JH/45 Fix cutthrough, when connection not opened by verify and target hard-
223 rejects a recipient: pass the reject to the originator.
225 JH/46 Multiple issues raised by Coverity. Some were obvious or plausible bugs.
226 Many were false-positives and ignorable, but it's worth fixing the
229 JH/47 Fix build on HP-UX and older Solaris, which need (un)setenv now also
230 for the new environment-manipulation done at startup. Move the routines
231 from being local to tls.c to being global via the os.c file.
233 JH/48 Bug 1807: Fix ${extract } for the numeric/3-string case. While preparsing
234 an extract embedded as result-arg for a map, the first arg for extract
235 is unavailable so we cannot tell if this is a numbered or keyed
236 extraction. Accept either.
242 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
245 JH/02 The smtp transport option "multi_domain" is now expanded.
247 JH/03 The smtp transport now requests PRDR by default, if the server offers
250 JH/04 Certificate name checking on server certificates, when exim is a client,
251 is now done by default. The transport option tls_verify_cert_hostnames
252 can be used to disable this per-host. The build option
253 EXPERIMENTAL_CERTNAMES is withdrawn.
255 JH/05 The value of the tls_verify_certificates smtp transport and main options
256 default to the word "system" to access the system default CA bundle.
257 For GnuTLS, only version 3.0.20 or later.
259 JH/06 Verification of the server certificate for a TLS connection is now tried
260 (but not required) by default. The verification status is now logged by
261 default, for both outbound TLS and client-certificate supplying inbound
264 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
267 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
268 Status Notification (bounce) messages are now MIME format per RFC 3464.
269 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
270 under the control of the dsn_advertise_hosts option, and routers may
271 have a dsn_lasthop option.
273 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
274 default, modifiable by a malware= option. The list separator for
275 the options can now be changed in the usual way. Bug 68.
277 JH/10 The smtp_receive_timeout main option is now expanded before use.
279 JH/11 The incoming_interface log option now also enables logging of the
280 local interface on delivery outgoing connections.
282 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
283 if the interface and destination host and port all match.
285 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
288 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
289 Patch from Andrew Lewis.
291 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
292 now supports optional time-restrictions, weighting, and priority
293 modifiers per server. Patch originally by <rommer@active.by>.
295 JH/16 The spamd_address main option now supports a mixed list of local
296 and remote servers. Remote servers can be IPv6 addresses, and
297 specify a port-range.
299 JH/17 Bug 68: The spamd_address main option now supports an optional
300 timeout value per server.
302 JH/18 Bug 1581: Router and transport options headers_add/remove can
303 now have the list separator specified.
305 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
308 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
311 JH/21 Support for the A6 type of dns record is withdrawn.
313 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
314 rather than the verbs used.
316 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
317 from 255 to 1024 chars.
319 JH/24 Verification callouts now attempt to use TLS by default.
321 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
322 are generic router options now. The defaults didn't change.
324 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
325 Original patch from Alexander Shikoff, worked over by JH.
327 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
328 files if ZCAT_COMMAND is not executable.
330 JH/26 Bug 1539: Add timout/retry options on dnsdb lookups.
332 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
334 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
335 Normally benign, it bites when the pair was led to by a CNAME;
336 modern usage is to not canoicalize the domain to a CNAME target
337 (and we were inconsistent anyway for A-only vs AAAA+A).
339 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
341 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
342 when evaluating $sender_host_dnssec.
344 JH/31 Check the HELO verification lookup for DNSSEC, adding new
345 $sender_helo_dnssec variable.
347 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
349 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
351 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
353 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
354 documented as working, but never had. Support all but $spam_report.
356 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
357 added for tls authenticator.
359 HS/03 Add perl_taintmode main config option
364 TL/01 When running the test suite, the README says that variables such as
365 no_msglog_check are global and can be placed anywhere in a specific
366 test's script, however it was observed that placement needed to be near
367 the beginning for it to behave that way. Changed the runtest perl
368 script to read through the entire script once to detect and set these
369 variables, reset to the beginning of the script, and then run through
370 the script parsing/test process like normal.
372 TL/02 The BSD's have an arc4random API. One of the functions to induce
373 adding randomness was arc4random_stir(), but it has been removed in
374 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
375 function when detected.
377 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
378 cause callback expansion.
380 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
381 syntax errors in an expansion can be treated as a string instead of
382 logging or causing an error, due to the internal use of bool_lax
383 instead of bool when processing it.
385 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
386 server certificates when making smtp deliveries.
388 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
390 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
392 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
394 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
395 Merged patch from Sebastian Wiedenroth.
397 JH/05 Fix results-pipe from transport process. Several recipients, combined
398 with certificate use, exposed issues where response data items split
399 over buffer boundaries were not parsed properly. This eventually
400 resulted in duplicates being sent. This issue only became common enough
401 to notice due to the introduction of conection certificate information,
402 the item size being so much larger. Found and fixed by Wolfgang Breyha.
404 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
405 size buffer was used, resulting in syntax errors when an expansion
408 JH/07 Add support for directories of certificates when compiled with a GnuTLS
409 version 3.3.6 or later.
411 JH/08 Rename the TPDA expermimental facility to Event Actions. The #ifdef
412 is EXPERIMENTAL_EVENT, the main-configuration and transport options
413 both become "event_action", the variables become $event_name, $event_data
414 and $event_defer_errno. There is a new variable $verify_mode, usable in
415 routers, transports and related events. The tls:cert event is now also
416 raised for inbound connections, if the main configuration event_action
419 TL/06 In test suite, disable OCSP for old versions of openssl which contained
420 early OCSP support, but no stapling (appears to be less than 1.0.0).
422 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
423 server certificate names available under the smtp transport option
424 "tls_verify_cert_hostname" now do not permit multi-component wildcard
427 JH/10 Time-related extraction expansions from certificates now use the main
428 option "timezone" setting for output formatting, and are consistent
429 between OpenSSL and GnuTLS compilations. Bug 1541.
431 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
432 encoded parameter in the incoming message. Bug 1558.
434 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
435 include certificate info, eximon was claiming there were spoolfile
438 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
440 JH/14 Log delivery-related information more consistently, using the sequence
441 "H=<name> [<ip>]" wherever possible.
443 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
444 are problematic for Debian distribution, omit them from the release
447 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
449 JH/16 Fix string representation of time values on 64bit time_t anchitectures.
452 JH/17 Fix a null-indirection in certextract expansions when a nondefault
453 output list separator was used.
458 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
459 checkers that were complaining about end of non-void function with no
462 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
463 This was a regression intruduced in 4.83 by another bugfix.
465 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
467 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
468 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
474 TF/01 Correctly close the server side of TLS when forking for delivery.
476 When a message was received over SMTP with TLS, Exim failed to clear up
477 the incoming connection properly after forking off the child process to
478 deliver the message. In some situations the subsequent outgoing
479 delivery connection happened to have the same fd number as the incoming
480 connection previously had. Exim would try to use TLS and fail, logging
481 a "Bad file descriptor" error.
483 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
484 utilities have not been installed.
486 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
487 temporary space as the ACL may create new global variables.
489 TL/01 LDAP support uses per connection or global context settings, depending
490 upon the detected version of the libraries at build time.
492 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
493 to extract and use the src ip:port in logging and expansions as if it
494 were a direct connection from the outside internet. PPv2 support was
495 updated based on HAProxy spec change in May 2014.
497 JH/02 Add ${listextract {number}{list}{success}{fail}}.
499 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
500 Properly escape header and check for NULL return.
502 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
505 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
507 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
508 characters in header names, implemented as a verify condition.
509 Contributed by Michael Fischer v. Mollard.
511 TL/05 Rename SPF condition results err_perm and err_temp to standardized
512 results permerror and temperror. Previous values are deprecated but
513 still accepted. In a future release, err_perm and err_temp will be
514 completely removed, which will be a backward incompatibility if the
515 ACL tests for either of these two old results. Patch contributed by
516 user bes-internal on the mailing list.
518 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
520 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
521 selectors, in both main and reject logs.
523 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
526 JH/07 Add malware type "sock" for talking to simple daemon.
528 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
530 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
531 routers/transports under cutthrough routing.
533 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
534 numbers. Touch up "bool" conditional to keep the same definition.
536 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
538 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
540 JH/12 Expand items in router/transport headers_add or headers_remove lists
541 individually rather than the list as a whole. Bug 1452.
543 Required for reasonable handling of multiple headers_ options when
544 they may be empty; requires that headers_remove items with embedded
545 colons must have them doubled (or the list-separator changed).
547 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
548 view the policy declared in the DMARC record. Currently, $dmarc_status
549 is a combined value of both the record presence and the result of the
552 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
554 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
555 dnslookup router and the smtp transport (applying to the forward
558 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
559 of ldap servers used for a specific lookup. Patch provided by Heiko
562 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
563 New variable $lookup_dnssec_authenticated for observability.
565 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
566 Patch submitted by Lars Timman.
568 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
570 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
571 Requires trusted mode and valid format message id, aborts otherwise.
572 Patch contributed by Heiko Schlichting.
574 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
575 certextract with support for various fields. Bug 1358.
577 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
578 is requested by default, modifiable by smtp transport option
581 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
582 operate on certificate variables to give certificate fingerprints
583 Also new ${sha256:cert_variable}.
585 JH/23 The PRDR feature is moved from being Experimental into the mainline.
587 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
588 Christian Aistleitner.
590 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
592 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
593 file. Patch from Wolfgang Breyha.
595 JH/25 Expand the coverage of the delivery $host and $host_address to
596 client authenticators run in verify callout. Bug 1476.
598 JH/26 Port service names are now accepted for tls_on_connect_ports, to
599 align with daemon_smtp_ports. Bug 72.
601 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
602 support and error reporting did not work properly.
604 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
605 and is readable. Patch from Andrew Colin Kissa.
607 TL/14 Enhance documentation of ${run expansion and how it parses the
608 commandline after expansion, particularly in the case when an
609 unquoted variable expansion results in an empty value.
611 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
613 PP/02 Fix internal collision of T_APL on systems which support RFC3123
614 by renaming away from it. Addresses GH issue 15, reported by
617 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
619 TL/15 SECURITY: prevent double expansion in math comparison functions
620 (can expand unsanitized data). Not remotely exploitable.
627 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
629 PP/02 Make -n do something, by making it not do something.
630 When combined with -bP, the name of an option is not output.
632 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
635 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
636 $sender_host_name and config options to manage this, and basic check
639 PP/05 DSCP support for outbound connections and control modifier for inbound.
641 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
642 (Only plugin which currently uses this is kerberos4, which nobody should
643 be using, but we should make it available and other future plugins might
644 conceivably use it, even though it would break NAT; stuff *should* be
645 using channel bindings instead).
647 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
648 name; added for Sendmail compatibility; requires admin caller.
649 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
650 just ignore it); requires trusted caller.
651 Also parse but ignore: -Ac -Am -X<logfile>
654 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
656 TL/02 Add +smtp_confirmation as a default logging option.
658 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
659 Patch by Magnus Holmgren from 2007-02-20.
661 TL/04 Bugzilla 1281 - Spec typo.
662 Bugzilla 1283 - Spec typo.
663 Bugzilla 1290 - Spec grammar fixes.
665 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
667 TL/06 Add Experimental DMARC support using libopendmarc libraries.
669 TL/07 Fix an out of order global option causing a segfault. Reported to dev
670 mailing list by by Dmitry Isaikin.
672 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
674 JH/02 Support "G" suffix to numbers in ${if comparisons.
676 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
678 NM/01 Bugzilla 1197 - Spec typo
679 Bugzilla 1196 - Spec examples corrections
681 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
683 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
684 gnutls_enable_pkcs11, but renamed to more accurately indicate its
687 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
688 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
690 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
691 "acl {{name}{arg}...}", and optional args on acl condition
694 JH/05 Permit multiple router/transport headers_add/remove lines.
696 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
698 JH/07 Avoid using a waiting database for a single-message-only transport.
699 Performance patch from Paul Fisher. Bugzilla 1262.
701 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
704 JH/09 Add $headers_added variable, with content from use of ACL modifier
705 add_header (but not yet added to the message). Bugzilla 199.
707 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
708 Pulled from Bugzilla 817 by Wolfgang Breyha.
710 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
712 (nb: this is the same fix as in Exim 4.80.1)
714 JH/11 Add A= logging on delivery lines, and a client_set_id option on
717 JH/12 Add optional authenticated_sender logging to A= and a log_selector
720 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
722 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
723 advertise SMTP AUTH mechanism to us, instead of a generic
724 protocol violation error. Also, make Exim more robust to bad
725 data from the Dovecot auth socket.
727 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
729 When a queue runner is handling a message, Exim first routes the
730 recipient addresses, during which it prunes them based on the retry
731 hints database. After that it attempts to deliver the message to
732 any remaining recipients. It then updates the hints database using
735 So if a recipient address works intermittently, it can get repeatedly
736 deferred at routing time. The retry hints record remains fresh so the
737 address never reaches the final cutoff time.
739 This is a fairly common occurrence when a user is bumping up against
740 their storage quota. Exim had some logic in its local delivery code
741 to deal with this. However it did not apply to per-recipient defers
742 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
744 This change adds a proper retry rule check during routing so that the
745 final cutoff time is checked against the message's age. We only do
746 this check if there is an address retry record and there is not a
747 domain retry record; this implies that previous attempts to handle
748 the address had the retry_use_local_parts option turned on. We use
749 this as an approximation for the destination being like a local
750 delivery, as in LMTP.
752 I suspect this new check makes the old local delivery cutoff check
753 redundant, but I have not verified this so I left the code in place.
755 TF/02 Correct gecos expansion when From: is a prefix of the username.
757 Test 0254 submits a message to Exim with the header
761 When I ran the test suite under the user fanf2, Exim expanded
762 the header to contain my full name, whereas it should have added
763 a Resent-Sender: header. It erroneously treats any prefix of the
764 username as equal to the username.
766 This change corrects that bug.
768 GF/01 DCC debug and logging tidyup
769 Error conditions log to paniclog rather than rejectlog.
770 Debug lines prefixed by "DCC: " to remove any ambiguity.
772 TF/03 Avoid unnecessary rebuilds of lookup-related code.
774 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
775 Bug spotted by Jeremy Harris; was flawed since initial commit.
776 Would have resulted in OCSP responses post-SNI triggering an Exim
777 NULL dereference and crash.
779 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
781 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
782 Bug detection, analysis and fix by Samuel Thibault.
783 Bugzilla 1331, Debian bug #698092.
785 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
787 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
788 Server implementation by Todd Lyons, client by JH.
789 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
790 config variable "prdr_enable" controls whether the server
791 advertises the facility. If the client requests PRDR a new
792 acl_data_smtp_prdr ACL is called once for each recipient, after
793 the body content is received and before the acl_smtp_data ACL.
794 The client is controlled by bolth of: a hosts_try_prdr option
795 on the smtp transport, and the server advertisement.
796 Default client logging of deliveries and rejections involving
797 PRDR are flagged with the string "PRDR".
799 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
800 fclose(). Diagnosis by Todd Lyons.
802 PP/17 Update configure.default to handle IPv6 localhost better.
803 Patch by Alain Williams (plus minor tweaks).
806 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
807 This is now consistent with GnuTLS, and is now documented: the
808 previous undocumented portable approach to treating the option as
809 unset was to force an expansion failure. That still works, and
810 an empty string is now equivalent.
812 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
813 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
814 not performing validation itself.
816 PP/20 Added force_command boolean option to pipe transport.
817 Patch from Nick Koston, of cPanel Inc.
819 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
822 TF/04 Added udpsend ACL modifer and hexquote expansion operator
824 PP/21 Fix eximon continuous updating with timestamped log-files.
825 Broken in a format-string cleanup in 4.80, missed when I repaired the
826 other false fix of the same issue.
827 Report and fix from Heiko Schlichting.
830 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
831 Report from Prashanth Katuri.
833 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
834 It's SecureTransport, so affects any MacOS clients which use the
835 system-integrated TLS libraries, including email clients.
837 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
838 using a MIME ACL for non-SMTP local injection.
839 Report and assistance in diagnosis by Warren Baker.
841 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
843 JH/16 Fix comparisons for 64b. Bugzilla 1385.
845 TL/09 Add expansion variable $authenticated_fail_id to keep track of
846 last id that failed so it may be referenced in subsequent ACL's.
848 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
851 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
852 ldap library initialization, allowing self-signed CA's to be
853 used. Also properly sets require_cert option later in code by
854 using NULL (global ldap config) instead of ldap handle (per
855 session). Bug diagnosis and testing by alxgomz.
857 TL/12 Enhanced documentation in the ratelimit.pl script provided in
858 the src/util/ subdirectory.
860 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
861 renamed to Transport Post Delivery Action by Jeremy Harris, as
864 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
865 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
866 redis_servers = needs to be configured which will be used by the redis
867 lookup. Patch from Warren Baker, of The Packet Hub.
869 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
871 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
872 hostname or reverse DNS when processing a host list. Used suggestions
873 from multiple comments on this bug.
875 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
877 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
878 Missed a few lines, added it to make the runtest require no keyboard
881 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
882 contains upper case chars. Make router use caseful_local_part.
884 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
885 support when GnuTLS has been built with p11-kit.
891 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
893 This, or similar/improved, will also be change PP/11 of 4.82.
899 PP/01 Handle short writes when writing local log-files.
900 In practice, only affects FreeBSD (8 onwards).
901 Bugzilla 1053, with thanks to Dmitry Isaikin.
903 NM/01 Bugzilla 949 - Documentation tweak
905 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
908 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
910 PP/02 Implemented gsasl authenticator.
912 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
914 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
915 `pkg-config foo` for cflags/libs.
917 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
918 with rest of GSASL and with heimdal_gssapi.
920 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
921 `pkg-config foo` for cflags/libs for the TLS implementation.
923 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
924 properties get this fed in as external SSF. A number of robustness
925 and debugging improvements to the cyrus_sasl authenticator.
927 PP/08 cyrus_sasl server now expands the server_realm option.
929 PP/09 Bugzilla 1214 - Log authentication information in reject log.
930 Patch by Jeremy Harris.
932 PP/10 Added dbmjz lookup type.
934 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
936 PP/12 MAIL args handles TAB as well as SP, for better interop with
937 non-compliant senders.
938 Analysis and variant patch by Todd Lyons.
940 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
941 Bug report from Lars Müller <lars@samba.org> (via SUSE),
942 Patch from Dirk Mueller <dmueller@suse.com>
944 PP/13 tls_peerdn now print-escaped for spool files.
945 Observed some $tls_peerdn in wild which contained \n, which resulted
946 in spool file corruption.
948 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
949 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
950 or write after TLS renegotiation, which otherwise led to messages
953 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
954 as a tracking header (ie: a signed header comes before the signature).
955 Patch from Wolfgang Breyha.
957 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
958 comma-sep list; embedded commas doubled.
960 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
962 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
964 Report and patch from Dmitry Banschikov.
966 PP/16 Removed "dont_insert_empty_fragments" fron "openssl_options".
967 Removed SSL_clear() after SSL_new() which led to protocol negotiation
968 failures. We appear to now support TLS1.1+ with Exim.
970 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
971 lets Exim select keys and certificates based upon TLS SNI from client.
972 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
973 before an outbound SMTP session. New log_selector, +tls_sni.
975 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
976 NULL dereference. Report and patch from Alun Jones.
978 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
979 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
980 Not seeing resolver debug output on NetBSD, but suspect this is a
981 resolver implementation change.
983 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
984 Left warnings. Added "eximon gdb" invocation mode.
986 PP/21 Defaulting "accept_8bitmime" to true, not false.
988 PP/22 Added -bw for inetd wait mode support.
990 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
991 locate the relevant includes and libraries. Made this the default.
993 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
994 Bugzilla 1246, report and most of solution from Tomasz Kusy.
996 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
997 This may cause build issues on older platforms.
999 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
1000 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
1001 gnutls_require_mac & gnutls_require_protocols (no longer supported).
1002 Added SNI support via GnuTLS too.
1003 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
1005 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
1007 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
1008 Applied second patch from Janne, implementing suggestion to default
1009 multiple-strings-in-record handling to match SPF spec.
1011 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
1013 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
1014 read-only, out of scope).
1015 Patch from Wolfgang Breyha, report from Stuart Northfield.
1017 PP/29 Fix three issues highlighted by clang analyser static analysis.
1018 Only crash-plausible issue would require the Cambridge-specific
1019 iplookup router and a misconfiguration.
1020 Report from Marcin Mirosław.
1022 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
1024 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
1025 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
1026 As part of this, removing so much warning spew let me fix some minor
1027 real issues in debug logging.
1029 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
1030 assignment on my part. Fixed.
1032 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
1033 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
1034 Janne Snabb (who went above and beyond: thank you).
1036 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
1037 string otherwise requires a connection and a bunch more work and it's
1038 relatively easy to get wrong. Should also expose TLS library linkage
1041 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
1042 64-bit ${eval} (JH/03).
1044 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
1045 GNU libc to support some of the 64-bit stuff, should not lead to
1046 conflicts. Defined before os.h is pulled in, so if a given platform
1047 needs to override this, it can.
1049 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
1050 protection layer was required, which is not implemented.
1051 Bugzilla 1254, patch from Wolfgang Breyha.
1053 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
1054 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
1055 tls_dhparam take prime identifiers. Also unbreak combination of
1056 OpenSSL+DH_params+TLSSNI.
1058 PP/39 Disable SSLv2 by default in OpenSSL support.
1064 PP/01 Solaris build fix for Oracle's LDAP libraries.
1065 Bugzilla 1109, patch from Stephen Usher.
1067 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
1069 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
1072 TF/02 Fix a couple more cases where we did not log the error message
1073 when unlink() failed. See also change 4.74-TF/03.
1075 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
1076 lock up or crash if it happened to be inside a call to libc when it
1077 got a SIGUSR1 from exiwhat.
1079 The SIGUSR1 handler appends the current process status to the process
1080 log which is later printed by exiwhat. It used to use the general
1081 purpose logging code to do this, but several functions it calls are
1082 not safe for signals.
1084 The new output code in the SIGUSR1 handler is specific to the process
1085 log, and simple enough that it's easy to inspect for signal safety.
1086 Removing some special cases also simplifies the general logging code.
1087 Removing the spurious timestamps from the process log simplifies
1090 TF/04 Improved ratelimit ACL condition.
1092 The /noupdate option has been deprecated in favour of /readonly which
1093 has clearer semantics. The /leaky, /strict, and /readonly update modes
1094 are mutually exclusive. The update mode is no longer included in the
1095 database key; it just determines when the database is updated. (This
1096 means that when you upgrde Exim will forget old rate measurements.)
1098 Exim now checks that the per_* options are used with an update mode that
1099 makes sense for the current ACL. For example, when Exim is processing a
1100 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
1101 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
1102 must specify per_mail/readonly. If you omit the update mode it defaults to
1103 /leaky where that makes sense (as before) or /readonly where required.
1105 The /noupdate option is now undocumented but still supported for
1106 backwards compatibility. It is equivalent to /readonly except that in
1107 ACLs where /readonly is required you may specify /leaky/noupdate or
1108 /strict/noupdate which are treated the same as /readonly.
1110 A useful new feature is the /count= option. This is a generalization
1111 of the per_byte option, so that you can measure the throughput of other
1112 aggregate values. For example, the per_byte option is now equivalent
1113 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
1115 The per_rcpt option has been generalized using the /count= mechanism
1116 (though it's more complicated than the per_byte equivalence). When it is
1117 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
1118 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
1119 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
1120 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
1121 Note that using per_rcpt with a non-readonly update mode in more than
1122 one ACL will cause the recipients to be double-counted. (The per_mail
1123 and per_byte options don't have this problem.)
1125 The handling of very low rates has changed slightly. If the computed rate
1126 is less than the event's count (usually one) then this event is the first
1127 after a long gap. In this case the rate is set to the same as this event's
1128 count, so that the first message of a spam run is counted properly.
1130 The major new feature is a mechanism for counting the rate of unique
1131 events. The new per_addr option counts the number of different
1132 recipients that someone has sent messages to in the last time period. It
1133 behaves like per_rcpt if all the recipient addresses are different, but
1134 duplicate recipient addresses do not increase the measured rate. Like
1135 the /count= option this is a general mechanism, so the per_addr option
1136 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
1137 example, measure the rate that a client uses different sender addresses
1138 with the options per_mail/unique=$sender_address. There are further
1139 details in the main documentation.
1141 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
1143 TF/06 Removed a few PCRE remnants.
1145 TF/07 Automatically extract Exim's version number from tags in the git
1146 repository when doing development or release builds.
1148 PP/02 Raise smtp_cmd_buffer_size to 16kB.
1149 Bugzilla 879. Patch from Paul Fisher.
1151 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
1152 Heavily based on revision 40f9a89a from Simon Arlott's tree.
1155 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
1157 PP/05 Variable $av_failed, true if the AV scanner deferred.
1158 Bugzilla 1078. Patch from John Horne.
1160 PP/06 Stop make process more reliably on build failure.
1161 Bugzilla 1087. Patch from Heiko Schlittermann.
1163 PP/07 Make maildir_use_size_file an _expandable_ boolean.
1164 Bugzilla 1089. Patch from Heiko Schlittermann.
1166 PP/08 Handle ${run} returning more data than OS pipe buffer size.
1167 Bugzilla 1131. Patch from Holger Weiß.
1169 PP/09 Handle IPv6 addresses with SPF.
1170 Bugzilla 860. Patch from Wolfgang Breyha.
1172 PP/10 GnuTLS: support TLS 1.2 & 1.1.
1174 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
1177 PP/11 match_* no longer expand right-hand-side by default.
1178 New compile-time build option, EXPAND_LISTMATCH_RHS.
1179 New expansion conditions, "inlist", "inlisti".
1181 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
1183 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
1185 PP/14 fix log_write() format string regression from TF/03.
1186 Bugzilla 1152. Patch from Dmitry Isaikin.
1192 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
1194 PP/02 Harmonised TLS library version reporting; only show if debugging.
1195 Layout now matches that introduced for other libraries in 4.74 PP/03.
1197 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
1199 PP/04 New "dns_use_edns0" global option.
1201 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
1204 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
1205 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
1207 TK/01 Updated PolarSSL code to 0.14.2.
1208 Bugzilla 1097. Patch from Andreas Metzler.
1210 PP/07 Catch divide-by-zero in ${eval:...}.
1211 Fixes bugzilla 1102.
1213 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
1216 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
1217 format-string attack -- SECURITY: remote arbitrary code execution.
1219 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
1220 time unintentionally subject to list matching rules, letting the header
1221 cause arbitrary Exim lookups (of items which can occur in lists, *not*
1222 arbitrary string expansion). This allowed for information disclosure.
1224 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
1225 INT_MIN/-1 -- value coerced to INT_MAX.
1231 NM/01 Workround for PCRE version dependency in version reporting
1234 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
1235 This fixes portability to compilers other than gcc, notably
1236 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
1238 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
1239 makefiles for portability to HP-UX and POSIX correctness.
1241 PP/01 Permit LOOKUP_foo enabling on the make command-line.
1242 Also via indented variable definition in the Makefile.
1243 (Debugging by Oliver Heesakkers).
1245 PP/02 Restore caching of spamd results with expanded spamd_address.
1246 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
1248 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
1249 Improves build reliability. Fix from: Frank Elsner
1251 NM/02 Fix wide character breakage in the rfc2047 coding
1252 Fixes bug 1064. Patch from Andrey N. Oktyabrski
1254 NM/03 Allow underscore in dnslist lookups
1255 Fixes bug 1026. Patch from Graeme Fowler
1257 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
1258 Code patches from Adam Ciarcinski of NetBSD.
1260 NM/04 Fixed exiqgrep to cope with mailq missing size issue
1263 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
1264 is logged, to avoid truncation. Patch from John Horne.
1266 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
1267 Patch from Jakob Hirsch.
1269 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
1270 SQL string expansion failure details.
1271 Patch from Andrey Oktyabrski.
1273 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
1274 Patch from Simon Arlott.
1276 PP/09 New lookups functionality failed to compile on old gcc which rejects
1277 extern declarations in function scope.
1278 Patch from Oliver Fleischmann
1280 PP/10 Use sig_atomic_t for flags set from signal handlers.
1281 Check getgroups() return and improve debugging.
1282 Fixed developed for diagnosis in bug 927 (which turned out to be
1285 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
1286 Patch from Mark Zealey.
1288 PP/12 Bugzilla 1056: Improved spamd server selection.
1289 Patch from Mark Zealey.
1291 PP/13 Bugzilla 1086: Deal with maildir quota file races.
1292 Based on patch from Heiko Schlittermann.
1294 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
1295 Patch from Uwe Doering, sign-off by Michael Haardt.
1297 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
1298 variable declaration deep within a block. Bug and patch from
1301 PP/15 lookups-Makefile IRIX compatibilty coercion.
1303 PP/16 Make DISABLE_DKIM build knob functional.
1305 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
1306 Patch by Simon Arlott
1308 TF/03 Fix valgrind.h portability to C89 compilers that do not support
1309 variable argument macros. Our copy now differs from upstream.
1315 TF/01 Failure to get a lock on a hints database can have serious
1316 consequences so log it to the panic log.
1318 TF/02 Log LMTP confirmation messages in the same way as SMTP,
1319 controlled using the smtp_confirmation log selector.
1321 TF/03 Include the error message when we fail to unlink a spool file.
1323 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
1324 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
1325 for maintaining out-of-tree patches for some time.
1327 PP/01 Bugzilla 139: Documentation and portability issues.
1328 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
1329 Handle per-OS dynamic-module compilation flags.
1331 PP/02 Let /dev/null have normal permissions.
1332 The 4.73 fixes were a little too stringent and complained about the
1333 permissions on /dev/null. Exempt it from some checks.
1334 Reported by Andreas M. Kirchwitz.
1336 PP/03 Report version information for many libraries, including
1337 Exim version information for dynamically loaded libraries. Created
1338 version.h, now support a version extension string for distributors
1339 who patch heavily. Dynamic module ABI change.
1341 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
1342 privilege escalation vulnerability whereby the Exim run-time user
1343 can cause root to append content of the attacker's choosing to
1346 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
1349 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
1350 If dropping privileges for untrusted macros, we disabled normal logging
1351 on the basis that it would fail; for the Exim run-time user, this is not
1352 the case, and it resulted in successful deliveries going unlogged.
1353 Fixed. Reported by Andreas Metzler.
1359 PP/01 Date: & Message-Id: revert to normally being appended to a message,
1360 only prepend for the Resent-* case. Fixes regression introduced in
1361 Exim 4.70 by NM/22 for Bugzilla 607.
1363 PP/02 Include check_rfc2047_length in configure.default because we're seeing
1364 increasing numbers of administrators be bitten by this.
1366 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
1368 PP/03 Bugzilla 994: added openssl_options main configuration option.
1370 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
1372 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
1374 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
1376 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
1377 without a peer certificate, leading to a segfault because of an
1378 assumption that peers always have certificates. Be a little more
1379 paranoid. Problem reported by Martin Tscholak.
1381 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
1382 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
1383 NB: ClamAV planning to remove STREAM in "middle of 2010".
1384 CL also introduces -bmalware, various -d+acl logging additions and
1385 more caution in buffer sizes.
1387 PP/09 Implemented reverse_ip expansion operator.
1389 PP/10 Bugzilla 937: provide a "debug" ACL control.
1391 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
1393 PP/12 Bugzilla 973: Implement --version.
1395 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
1397 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
1399 PP/15 Bugzilla 816: support multiple condition rules on Routers.
1401 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
1402 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
1403 ignore trailing whitespace.
1405 JJ/02 prevent non-panic DKIM error from being sent to paniclog
1407 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
1410 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
1411 Notification from Dr Andrew Aitchison.
1413 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
1414 ExtendedDetectionInfo response format.
1415 Notification from John Horne.
1417 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
1420 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
1421 XSL and documented dependency on system catalogs, with examples of how
1424 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
1427 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
1428 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
1429 configuration file which is writeable by the Exim user or group.
1431 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
1432 of configuration files to cover files specified with the -C option if
1433 they are going to be used with root privileges, not just the default
1436 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
1437 option (effectively making it always true).
1439 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
1440 files to be used while preserving root privileges.
1442 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
1443 that rogue child processes cannot use them.
1445 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
1446 run-time user, instead of root.
1448 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
1449 Exim run-time user without dropping privileges.
1451 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
1452 result string, instead of calling string_vformat() twice with the same
1455 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
1456 for other users. Others should always drop root privileges if they use
1457 -C on the command line, even for a whitelisted configure file.
1459 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
1461 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
1467 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
1468 $data_path, and $header_path variables; fixed documentation bugs and
1471 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
1472 exipick to access non-standard spools, including the "frozen" queue
1475 NM/01 Bugzilla 965: Support mysql stored procedures.
1476 Patch from Alain Williams
1478 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
1480 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
1481 Patch from Andreas Metzler
1483 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
1484 Patch from Kirill Miazine
1486 NM/05 Bugzilla 671: Added umask to procmail example.
1488 JJ/03 installed exipick 20100323.0, fixing doc bug
1490 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
1491 directory. Notification and patch from Dan Rosenberg.
1493 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
1495 TK/02 Improve log output when DKIM signing operation fails.
1497 MH/01 Treat the transport option dkim_domain as a colon separated
1498 list, not as a single string, and sign the message with each element,
1499 omitting multiple occurences of the same signer.
1501 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
1502 Bugzilla 985, 986. Patch by Simon Arlott
1504 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
1505 Patch by Simon Arlott
1507 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
1508 MBX locking. Notification from Dan Rosenberg.
1514 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
1516 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
1518 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
1520 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
1522 NM/04 Bugzilla 915: Flag broken perl installation during build.
1528 TK/01 Added patch by Johannes Berg that expands the main option
1529 "spamd_address" if it starts with a dollar sign.
1531 TK/02 Write list of recipients to X-Envelope-Sender header when building
1532 the mbox-format spool file for content scanning (suggested by Jakob
1535 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
1536 (http://www.dcc-servers.net/) support via dccifd. Activated by
1537 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
1539 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
1540 by Mark Daniel Reidel <mr@df.eu>.
1542 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
1543 When building exim an external PCRE library is now needed -
1544 PCRE is a system library on the majority of modern systems.
1545 See entry on PCRE_LIBS in EDITME file.
1547 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
1548 conversation. Added nologin parameter to request.
1549 Patch contributed by Kirill Miazine.
1551 TF/01 Do not log submission mode rewrites if they do not change the address.
1553 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
1555 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
1556 log files in place. Contributed by Roberto Lima.
1558 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
1560 TF/03 Bugzilla 615: When checking the local_parts router precondition
1561 after a local_part_suffix or local_part_prefix option, Exim now
1562 does not use the address's named list lookup cache, since this
1563 contains cached lookups for the whole local part.
1565 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
1566 Robert Millan. Documentation is in experimental-spec.txt.
1568 TF/04 Bugzilla 668: Fix parallel build (make -j).
1570 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
1572 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
1573 Patch provided by Jan Srzednicki.
1575 TF/05 Leading white space used to be stripped from $spam_report which
1576 wrecked the formatting. Now it is preserved.
1578 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
1579 that they are available at delivery time.
1581 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
1583 TF/08 TLS error reporting now respects the incoming_interface and
1584 incoming_port log selectors.
1586 TF/09 Produce a more useful error message if an SMTP transport's hosts
1587 setting expands to an empty string.
1589 NM/06 Bugzilla 744: EXPN did not work under TLS.
1590 Patch provided by Phil Pennock.
1592 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
1593 Patch provided by Richard Godbee.
1595 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
1596 acl_smtp_notquit, added index entry.
1598 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
1599 Patch provided by Eugene Bujak.
1601 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
1602 accept(). Patch provided by Maxim Dounin.
1604 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
1605 Patch provided by Phil Pennock.
1607 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
1609 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
1610 Patch provided by Brad "anomie" Jorsch.
1612 NM/14 Bugzilla 622: Added timeout setting to transport filter.
1613 Patch provided by Dean Brooks.
1615 TK/05 Add native DKIM support (does not depend on external libraries).
1617 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
1618 Patch provided by Graeme Fowler.
1620 NM/16 Bugzilla 851: Documentation example syntax fix.
1622 NM/17 Changed NOTICE file to remove references to embedded PCRE.
1624 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
1627 NM/19 Bugzilla 745: TLS version reporting.
1628 Patch provided by Phil Pennock.
1630 NM/20 Bugzilla 167: bool: condition support.
1631 Patch provided by Phil Pennock.
1633 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
1634 clients. Patch provided by Phil Pennock.
1636 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
1637 Patch provided by Brad "anomie" Jorsch.
1639 NM/23 Bugzilla 687: Fix misparses in eximstats.
1640 Patch provided by Heiko Schlittermann.
1642 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
1643 Patch provided by Heiko Schlittermann.
1645 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
1646 plus update to original patch.
1648 NM/26 Bugzilla 799: Documentation correction for ratelimit.
1650 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
1651 Patch provided by David Brownlee.
1653 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
1655 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
1657 NM/30 Bugzilla 888: TLS documentation bugfixes.
1659 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
1661 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
1662 Unlike the original bugzilla I have changed all shell scripts in src tree.
1664 NM/33 Bugzilla 898: Transport filter timeout fix.
1665 Patch by Todd Rinaldo.
1667 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
1668 Patch by Serge Demonchaux.
1670 NM/35 Bugzilla 39: Base64 decode bug fixes.
1671 Patch by Jakob Hirsch.
1673 NM/36 Bugzilla 909: Correct connect() call in dcc code.
1675 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
1677 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
1679 NM/39 Bugzilla 911: Fixed MakeLinks build script.
1685 TK/01 Add preliminary DKIM support. Currently requires a forked version of
1686 ALT-N's libdkim that I have put here:
1687 http://duncanthrax.net/exim-experimental/
1689 Note to Michael Haardt: I had to rename some vars in sieve.c. They
1690 were called 'true' and it seems that C99 defines that as a reserved
1691 keyword to be used with 'bool' variable types. That means you could
1692 not include C99-style headers which use bools without triggering
1693 build errors in sieve.c.
1695 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
1696 as mailq or other aliases. Changed the --help handling significantly
1697 to do whats expected. exim_usage() emits usage/help information.
1699 SC/01 Added the -bylocaldomain option to eximstats.
1701 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
1703 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
1705 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
1711 PH/01 Another patch from the Sieve maintainer.
1713 PH/02 When an IPv6 address is converted to a string for single-key lookup
1714 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
1715 dots are used instead of colons so that keys in lsearch files need not
1716 contain colons. This was done some time before quoting was made available
1717 in lsearch files. However, iplsearch files do require colons in IPv6 keys
1718 (notated using the quote facility) so as to distinguish them from IPv4
1719 keys. This meant that lookups for IP addresses in host lists did not work
1720 for iplsearch lookups.
1722 This has been fixed by arranging for IPv6 addresses to be expressed with
1723 colons if the lookup type is iplsearch. This is not incompatible, because
1724 previously such lookups could never work.
1726 The situation is now rather anomolous, since one *can* have colons in
1727 ordinary lsearch keys. However, making the change in all cases is
1728 incompatible and would probably break a number of configurations.
1730 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
1733 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
1734 conversion specification without a maximum field width, thereby enabling
1735 a rogue spamd server to cause a buffer overflow. While nobody in their
1736 right mind would setup Exim to query an untrusted spamd server, an
1737 attacker that gains access to a server running spamd could potentially
1738 exploit this vulnerability to run arbitrary code as the Exim user.
1740 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
1741 $primary_hostname instead of what libspf2 thinks the hosts name is.
1743 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
1744 a directory entry by the name of the lookup key. Previously, if a
1745 symlink pointed to a non-existing file or a file in a directory that
1746 Exim lacked permissions to read, a lookup for a key matching that
1747 symlink would fail. Now it is enough that a matching directory entry
1748 exists, symlink or not. (Bugzilla 503.)
1750 PH/03 The body_linecount and body_zerocount variables are now exported in the
1753 PH/04 Added the $dnslist_matched variable.
1755 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
1756 This means they are set thereafter only if the connection becomes
1759 PH/06 Added the client_condition to authenticators so that some can be skipped
1760 by clients under certain conditions.
1762 PH/07 The error message for a badly-placed control=no_multiline_responses left
1763 "_responses" off the end of the name.
1765 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
1767 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
1768 (without spaces) instead of just copying the configuration text.
1770 PH/10 Added the /noupdate option to the ratelimit ACL condition.
1772 PH/11 Added $max_received_linelength.
1774 PH/12 Added +ignore_defer and +include_defer to host lists.
1776 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
1777 way in which PCRE > 7.0 is built.
1779 PH/14 Implemented queue_only_load_latch.
1781 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
1782 MAIL command. The effect was to mangle the value on 64-bit systems.
1784 PH/16 Another patch from the Sieve maintainer.
1786 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
1788 PH/18 If a system quota error occurred while trying to create the file for
1789 a maildir delivery, the message "Mailbox is full" was not appended to the
1790 bounce if the delivery eventually timed out. Change 4.67/27 below applied
1791 only to a quota excession during the actual writing of the file.
1793 PH/19 It seems that peer DN values may contain newlines (and other non-printing
1794 characters?) which causes problems in log lines. The DN values are now
1795 passed through string_printing() before being added to log lines.
1797 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
1798 and InterBase are left for another time.)
1800 PH/21 Added message_body_newlines option.
1802 PH/22 Guard against possible overflow in moan_check_errorcopy().
1804 PH/23 POSIX allows open() to be a macro; guard against that.
1806 PH/24 If the recipient of an error message contained an @ in the local part
1807 (suitably quoted, of course), incorrect values were put in $domain and
1808 $local_part during the evaluation of errors_copy.
1814 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
1815 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
1818 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
1819 issue a MAIL command.
1821 PH/02 In an ACL statement such as
1823 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
1825 if a client was not listed at all, or was listed with a value other than
1826 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
1827 the condition was not true (as it should be), so access was not denied.
1828 The bug was that the ! inversion was incorrectly passed on to the second
1829 item. This has been fixed.
1831 PH/03 Added additional dnslists conditions == and =& which are different from
1832 = and & when the dns lookup returns more than one IP address.
1834 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
1835 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
1837 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
1838 FSYNC, which compiles an option called disable_fsync that allows for
1839 bypassing fsync(). The documentation is heavily laced with warnings.
1841 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
1843 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
1844 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
1845 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
1846 including adding "make clean"; (3) Added -fPIC when compiling the test
1847 dynamically loaded module, to get rid of a warning.
1849 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
1850 message fails, move_frozen_messages = true and ignore_bounce_errors_after
1851 = 0s. The bug is otherwise harmless.
1853 PH/07 There was a bug in the dovecot authenticator such that the value of
1854 $auth1 could be overwritten, and so not correctly preserved, after a
1855 successful authentication. This usually meant that the value preserved by
1856 the server_setid option was incorrect.
1858 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
1860 PH/09 Installed PCRE release 7.0.
1862 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
1863 run for batched SMTP input. It is now run at the start of every message
1864 in the batch. While fixing this I discovered that the process information
1865 (output by running exiwhat) was not always getting set for -bs and -bS
1866 input. This is fixed, and it now also says "batched" for BSMTP.
1868 PH/11 Added control=no_pipelining.
1870 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
1871 patch, slightly modified), and move the expansion of helo_data till after
1872 the connection is made in the smtp transport (so it can use these
1875 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
1877 PH/14 Added log_selector = +pid.
1879 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
1881 PH/16 Add ${if forany and ${if forall.
1883 PH/17 Added dsn_from option to vary the From: line in DSNs.
1885 PH/18 Flush SMTP output before performing a callout, unless control =
1886 no_callout_flush is set.
1888 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
1889 was true (the default) a successful delivery failed to delete the retry
1890 item, thus causing premature timeout of the address. The bug is now
1893 PH/20 Added hosts_avoid_pipelining to the smtp transport.
1895 PH/21 Long custom messages for fakedefer and fakereject are now split up
1896 into multiline reponses in the same way that messages for "deny" and
1897 other ACL rejections are.
1899 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
1900 with slight modification.
1902 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
1903 draft, changing the syntax and factoring some duplicate code".
1905 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
1906 for deliveries of the second and subsequent messages over the same SMTP
1909 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
1910 ${reduce, with only minor "tidies".
1912 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
1914 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
1915 expansion side effects.
1917 PH/27 When a message times out after an over-quota error from an Exim-imposed
1918 quota, the bounce message says "mailbox is full". This message was not
1919 being given when it was a system quota that was exceeded. It now should
1922 MH/03 Made $recipients available in local_scan(). local_scan() already has
1923 better access to the recipient list through recipients_list[], but
1924 $recipients can be useful in postmaster-provided expansion strings.
1926 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
1927 in the case of a MAIL command with additional options following the
1928 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
1929 were accidentally chopped off.
1931 PH/29 SMTP synchronization checks are implemented when a command is read -
1932 there is a check that no more input is waiting when there shouldn't be
1933 any. However, for some commands, a delay in an ACL can mean that it is
1934 some time before the response is written. In this time, more input might
1935 arrive, invalidly. So now there are extra checks after an ACL has run for
1936 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
1937 pipelining has not been advertised.
1939 PH/30 MH's patch to allow iscntrl() characters to be list separators.
1941 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
1942 returned in the SMTP response when smtp_return_error_details was false.
1943 This has been fixed.
1945 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
1946 instead of the C I/O that was originally supplied, because problems were
1947 reported on Solaris.
1949 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
1950 Exim which did not show up earlier: it was assuming that a call to
1951 SSL_CTX_set_info_callback() might give an error value. In fact, there is
1952 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
1953 was a macro that became an assignment, so it seemed to work. This has
1954 changed to a proper function call with a void return, hence the compile
1955 error. Exim's code has been fixed.
1957 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
1960 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
1962 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
1964 PH/37 If a message is not accepted after it has had an id assigned (e.g.
1965 because it turns out to be too big or there is a timeout) there is no
1966 "Completed" line in the log. When some messages of this type were
1967 selected by exigrep, they were listed as "not completed". Others were
1968 picked up by some special patterns. I have improved the selection
1969 criteria to be more general.
1971 PH/38 The host_find_failed option in the manualroute router can now be set
1972 to "ignore", to completely ignore a host whose IP address cannot be
1973 found. If all hosts are ignored, the behaviour is controlled by the new
1974 host_all_ignored option.
1976 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
1977 homing or because of multiple MX records with /mx) generated more than
1978 one IP address, and the following item turned out to be the local host,
1979 all the secondary addresses of the first item were incorrectly removed
1980 from the list, along with the local host and any following hosts (which
1981 is what is supposed to happen).
1983 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
1984 whoever called Exim into the -H file. In the case of the daemon it was
1985 behaving confusingly. When first started, it used values for whoever
1986 started the daemon, but after a SIGHUP it used the Exim user (because it
1987 calls itself on a restart). I have changed the code so that it now always
1990 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
1991 message are rejected with the same error (e.g. no authentication or bad
1992 sender address), and a DATA command is nevertheless sent (as can happen
1993 with PIPELINING or a stupid MUA), the error message that was given to the
1994 RCPT commands is included in the rejection of the DATA command. This is
1995 intended to be helpful for MUAs that show only the final error to their
1998 PH/42 Another patch from the Sieve maintainer.
2000 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
2001 Eximstats - Fixed some broken HTML links and added missing column headers
2003 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
2004 columns for Rejects, Temp Rejects, Ham, and Spam rows.
2006 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
2008 PH/43 Yet another patch from the Sieve maintainer.
2010 PH/44 I found a way to check for a TCP/IP connection going away before sending
2011 the response to the final '.' that terminates a message, but only in the
2012 case where the client has not sent further data following the '.'
2013 (unfortunately, this is allowed). However, in many cases there won't be
2014 any further data because there won't be any more messages to send. A call
2015 to select() can be used: if it shows that the input is "ready", there is
2016 either input waiting, or the socket has been closed. An attempt to read
2017 the next input character can distinguish the two cases. Previously, Exim
2018 would have sent an OK response which the client would never have see.
2019 This could lead to message repetition. This fix should cure that, at
2020 least in a lot of common cases.
2022 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
2023 advertised in response to EHLO.
2029 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
2030 fixed by 4.65/MH/01 (is this a record?) are fixed:
2032 (i) An empty string was always treated as zero by the numeric comparison
2033 operators. This behaviour has been restored.
2035 (ii) It is documented that the numeric comparison operators always treat
2036 their arguments as decimal numbers. This was broken in that numbers
2037 starting with 0 were being interpreted as octal.
2039 While fixing these problems I realized that there was another issue that
2040 hadn't been noticed. Values of message_size_limit (both the global option
2041 and the transport option) were treated as octal if they started with 0.
2042 The documentation was vague. These values are now always treated as
2043 decimal, and I will make that clear in the documentation.
2049 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
2050 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
2053 MH/01 Don't check that the operands of numeric comparison operators are
2054 integers when their expansion is in "skipping" mode (fixes bug
2055 introduced by 4.64-PH/07).
2057 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
2058 child addresses, Exim now panics and dies. Previously, because the count
2059 is held in a short int, deliveries were likely to be lost. As such a
2060 large number of recipients for a single message is ridiculous
2061 (performance will be very, very poor), I have chosen to impose a limit
2062 rather than extend the field.
2068 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
2069 leftover -K file (the existence of which was triggered by #402).
2070 While we were at it, introduced process PID as part of the -K
2071 filename. This should rule out race conditions when creating
2074 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
2075 processing considerably. Previous code took too long for large mails,
2076 triggering a timeout which in turn triggers #401.
2078 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
2079 in the DK code in transports.c. sendfile() is not really portable,
2080 hence the _LINUX specificness.
2082 TF/01 In the add_headers option to the mail command in an Exim filter,
2083 there was a bug that Exim would claim a syntax error in any
2084 header after the first one which had an odd number of characters
2087 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
2088 callout verification, Exim cached a "reject" for the entire domain. This
2089 is correct for most verifications, but it is not correct for a recipient
2090 verification with use_sender or use_postmaster set, because in that case
2091 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
2092 case of MAIL FROM:<> rejection from other early rejections (e.g.
2093 rejection of HELO). When verifying a recipient using a non-null MAIL
2094 address, the cache is ignored if it shows MAIL FROM:<> rejection.
2095 Whatever the result of the callout, the value of the domain cache is
2096 left unchanged (for any other kind of callout, getting as far as trying
2097 RCPT means that the domain itself is ok).
2099 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
2102 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
2103 manifest itself as EPIPE rather than ECONNECT. When tidying away a
2104 session, the daemon ignores ECONNECT errors and logs others; it now
2105 ignores EPIPE as well.
2107 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
2108 (quoted-printable decoding).
2110 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
2111 later the small subsequent patch to fix an introduced bug.
2113 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
2115 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
2117 PH/08 An error is now given if message_size_limit is specified negative.
2119 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
2120 to be given (somewhat) arbitrary names.
2122 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
2125 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
2126 miscellaneous code fixes
2128 PH/10 Added the log_reject_target ACL modifier to specify where to log
2131 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
2132 hostname. This is wrong, because it relates to the incoming message (and
2133 probably the interface on which it is arriving) and not to the outgoing
2134 callout (which could be using a different interface). This has been
2135 changed to use the value of the helo_data option from the smtp transport
2136 instead - this is what is used when a message is actually being sent. If
2137 there is no remote transport (possible with a router that sets up host
2138 addresses), $smtp_active_hostname is used.
2140 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
2141 tweaks were necessary in order to get it to work (see also 21 below):
2142 (a) The code assumed that strncpy() returns a negative number on buffer
2143 overflow, which isn't the case. Replaced with Exim's string_format()
2145 (b) There were several signed/unsigned issues. I just did the minimum
2146 hacking in of casts. There is scope for a larger refactoring.
2147 (c) The code used strcasecmp() which is not a standard C function.
2148 Replaced with Exim's strcmpic() function.
2149 (d) The code set only $1; it now sets $auth1 as well.
2150 (e) A simple test gave the error "authentication client didn't specify
2151 service in request". It would seem that Dovecot has changed its
2152 interface. Fortunately there's a specification; I followed it and
2153 changed what the client sends and it appears to be working now.
2155 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
2158 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
2159 address A is aliased to B and C, where B exists and C does not. Without
2160 -v the output is "A verified" because verification stops after a
2161 successful redirection if more than one address is generated. However,
2162 with -v the child addresses are also verified. Exim was outputting "A
2163 failed to verify" and then showing the successful verification for C,
2164 with its parentage. It now outputs "B failed to verify", showing B's
2165 parentage before showing the successful verification of C.
2167 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
2168 look up a TXT record in a specific list after matching in a combined
2171 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
2172 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
2173 they consult the DNS. I had assumed they would set it the way they
2174 wanted; and indeed my experiments on Linux seem to show that in some
2175 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
2176 To be on the safe side, however, I have now made the interface to
2177 host_find_byname() similar to host_find_bydns(), with an argument
2178 containing the DNS resolver options. The host_find_byname() function now
2179 sets these options at its start, just as host_find_bydns() does. The smtp
2180 transport options dns_qualify_single and dns_search_parents are passed to
2181 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
2182 of host_find_byname() use the default settings of RES_DEFNAMES
2183 (qualify_single) but not RES_DNSRCH (search_parents).
2185 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
2186 spool_read_header() do less string testing, by means of a preliminary
2187 switch on the second character of optional "-foo" lines. (This is
2188 overdue, caused by the large number of possibilities that now exist.
2189 Originally there were few.) While I was there, I also converted the
2190 str(n)cmp tests so they don't re-test the leading "-" and the first
2191 character, in the hope this might squeeze out yet more improvement.
2193 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
2194 flag allowing group syntax was set by the header_syntax check but not
2195 turned off, possible causing trouble later; (2) The flag was not being
2196 set at all for the header_verify test, causing "group"-style headers to
2197 be rejected. I have now set it in this case, and also caused header_
2198 verify to ignore an empty address taken from a group. While doing this, I
2199 came across some other cases where the code for allowing group syntax
2200 while scanning a header line wasn't quite right (mostly, not resetting
2201 the flag correctly in the right place). These bugs could have caused
2202 trouble for malformed header lines. I hope it is now all correct.
2204 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
2205 with the "reply" argument non-NULL. The code, however (which originally
2206 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
2207 but it didn't always do it. This confused somebody who was copying the
2208 code for some other use. I have removed all the tests.
2210 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
2211 feature that was used to support insecure browsers during the U.S. crypto
2212 embargo. It requires special client support, and Exim is probably the
2213 only MTA that supported it -- and would never use it because real RSA is
2214 always available. This code has been removed, because it had the bad
2215 effect of slowing Exim down by computing (never used) parameters for the
2216 RSA_EXPORT functionality.
2218 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
2219 authenticator to fail if there's a tab character in the incoming data
2220 (there should never be unless someone is messing about, as it's supposed
2221 to be base64-encoded). Also added, on Timo's advice, the "secured" option
2222 if the connection is using TLS or if the remote IP is the same as the
2223 local IP, and the "valid-client-cert option" if a client certificate has
2226 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
2227 authenticators. This can be used for authorization after authentication
2228 succeeds. (In the case of plaintext, it servers for both authentication
2231 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
2232 if any retry times were supplied.
2234 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
2235 connection, where there is no client IP address to check. In this
2236 situation, the verify now always succeeds.
2238 PH/25 Applied John Jetmore's -Mset patch.
2240 PH/26 Added -bem to be like -Mset, but loading a message from a file.
2242 PH/27 In a string expansion for a processed (not raw) header when multiple
2243 headers of the same name were present, leading whitespace was being
2244 removed from all of them, but trailing whitespace was being removed only
2245 from the last one. Now trailing whitespace is removed from each header
2246 before concatenation. Completely empty headers in a concatenation (as
2247 before) are ignored.
2249 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
2250 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
2252 PH/29 [Removed. This was a change that I later backed out, and forgot to
2253 correct the ChangeLog entry (that I had efficiently created) before
2254 committing the later change.]
2256 PH/30 Exim was sometimes attempting to deliver messages that had suffered
2257 address errors (4xx response to RCPT) over the same connection as other
2258 messages routed to the same hosts. Such deliveries are always "forced",
2259 so retry times are not inspected. This resulted in far too many retries
2260 for the affected addresses. The effect occurred only when there were more
2261 hosts than the hosts_max_try setting in the smtp transport when it had
2262 the 4xx errors. Those hosts that it had tried were not added to the list
2263 of hosts for which the message was waiting, so if all were tried, there
2264 was no problem. Two fixes have been applied:
2266 (i) If there are any address or message errors in an SMTP delivery, none
2267 of the hosts (tried or untried) are now added to the list of hosts
2268 for which the message is waiting, so the message should not be a
2269 candidate for sending over the same connection that was used for a
2270 successful delivery of some other message. This seems entirely
2271 reasonable: after all the message is NOT "waiting for some host".
2272 This is so "obvious" that I'm not sure why it wasn't done
2273 previously. Hope I haven't missed anything, but it can't do any
2274 harm, as the worst effect is to miss an optimization.
2276 (ii) If, despite (i), such a delivery is accidentally attempted, the
2277 routing retry time is respected, so at least it doesn't keep
2278 hammering the server.
2280 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
2281 in ${readsocket because some servers need this prod.
2283 PH/32 Added some extra debug output when updating a wait-xxx database.
2285 PH/33 The hint "could be header name not terminated by colon", which has been
2286 given for certain expansion errors for a long time, was not being given
2287 for the ${if def:h_colon_omitted{... case.
2289 PH/34 The spec says: "With one important exception, whenever a domain list is
2290 being scanned, $domain contains the subject domain." There was at least
2291 one case where this was not true.
2293 PH/35 The error "getsockname() failed: connection reset by peer" was being
2294 written to the panic log as well as the main log, but it isn't really
2295 panic-worthy as it just means the connection died rather early on. I have
2296 removed the panic log writing for the ECONNRESET error when getsockname()
2299 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
2300 runs only) independently of the message's sender address. This meant
2301 that, if the 4xx error was in fact related to the sender, a different
2302 message to the same recipient with a different sender could confuse
2303 things. In particualar, this can happen when sending to a greylisting
2304 server, but other circumstances could also provoke similar problems.
2305 I have changed the default so that the retry time for these errors is now
2306 based a combination of the sender and recipient addresses. This change
2307 can be overridden by setting address_retry_include_sender=false in the
2310 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
2311 remote server are returned as part of bounce messages. This was not
2312 happening for LMTP over a pipe (the lmtp transport), but now it is the
2313 same for both kinds of LMTP.
2315 PH/38 Despite being documented as not happening, Exim was rewriting addresses
2316 in header lines that were in fact CNAMEs. This is no longer the case.
2318 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
2319 and queue runs started by the daemon processed all messages. This has
2320 been fixed so that -R and -S can now usefully be given with -q<time>.
2322 PH/40 Import PCRE release 6.7 (fixes some bugs).
2324 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
2326 PH/42 Give an error if -q is specified more than once.
2328 PH/43 Renamed the variables $interface_address and $interface_port as
2329 $received_ip_address and $received_port, to make it clear that these
2330 values apply to message reception, and not to the outgoing interface when
2331 a message is delivered. (The old names remain recognized, of course.)
2333 PH/44 There was no timeout on the connect() call when using a Unix domain
2334 socket in the ${readsocket expansion. There now is.
2336 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
2337 be meaningful with "accept".
2339 SC/01 Eximstats V1.43
2340 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
2342 SC/02 Eximstats V1.44
2343 Use a glob alias rather than an array ref in the generated
2344 parser. This improves both readability and performance.
2346 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
2347 Collect SpamAssassin and rejection statistics.
2348 Don't display local sender or destination tables unless
2349 there is data to show.
2350 Added average volumes into the top table text output.
2352 SC/04 Eximstats V1.46
2353 Collect data on the number of addresses (recipients)
2354 as well as the number of messages.
2356 SC/05 Eximstats V1.47
2357 Added 'Message too big' to the list of mail rejection
2358 reasons (thanks to Marco Gaiarin).
2360 SC/06 Eximstats V1.48
2361 Mainlog lines which have GMT offsets and are too short to
2362 have a flag are now skipped.
2364 SC/07 Eximstats V1.49 (Alain Williams)
2365 Added the -emptyok flag.
2367 SC/08 Eximstats V1.50
2368 Fixes for obtaining the IP address from reject messages.
2370 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
2371 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
2372 whitesspace changes from 4.64-PH/27
2374 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
2377 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
2378 are found, allow negative numbers in numeric criteria)
2380 JJ/06 exipick.20061117.2, added new $message_body_missing variable
2382 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
2383 to match changes made in 4.64-PH/43
2385 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
2387 PH/47 Put in an explicit test for a DNS lookup of an address record where the
2388 "domain" is actually an IP address, and force a failure. This locks out
2389 those revolvers/nameservers that support "A-for-A" lookups, in
2390 contravention of the specifications.
2392 PH/48 When a host name was looked up from an IP address, and the subsequent
2393 forward lookup of the name timed out, the host name was left in
2394 $sender_host_name, contrary to the specification.
2396 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
2397 restricted to single-key lookups, Exim was not diagnosing an error if
2398 * or *@ was used with a query-style lookup.
2400 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
2402 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
2403 long ago, but noone interested enough thought of it. Let's just say that
2404 the "1.1" means that there are some new functions that weren't there at
2405 some point in the past.
2407 PH/51 Error processing for expansion failure of helo_data from an smtp
2408 transport during callout processing was broken.
2410 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
2411 tested/used via the -bh/-bhc/-bs options.
2413 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
2414 bug, fixed in subsequent PCRE releases).
2416 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
2417 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
2419 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
2425 SC/01 Use a glob alias rather than an array ref in eximstats generated
2426 parser. This improves both readability and performance.
2428 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
2429 Don't display local sender or destination tables in eximstats unless
2430 there is data to show.
2431 Added average volumes into the eximstats top table text output.
2433 SC/03 Collect data on the number of addresses (recipients) as well
2434 as the number of messages in eximstats.
2436 TF/01 Correct an error in the documentation for the redirect router. Exim
2437 does (usually) call initgroups() when daemonizing.
2439 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
2440 with consistent privilege compared to when running as a daemon.
2442 TF/03 Note in the spec that $authenticated_id is not set for local
2443 submissions from trusted users.
2445 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
2446 Thanks to Dean Brooks <dean@iglou.com> for the patch.
2448 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
2449 by adding some example configuration directives to the default
2450 configuration file. A little bit of work is required to uncomment the
2451 directives and define how usernames and passwords are checked, but
2452 there is now a framework to start from.
2454 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
2455 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
2456 without this. I don't know how relevant this is to other LDAP libraries.
2458 PH/02 Add the verb name to the "unknown ACL verb" error.
2460 PH/03 Magnus Holmgren's patch for filter_prepend_home.
2462 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
2464 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
2465 directory not expanded when it should be if an expanded home directory
2466 was set for the address (which is overridden by the transport).
2468 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
2471 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
2472 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
2473 because it is too late at that time, and has no effect.
2475 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
2476 security issue with \' (bugzilla #107). I could not use the
2477 PQescapeStringConn() function, because it needs a PGconn value as one of
2480 PH/08 When testing addresses using -bt, indicate those final addresses that
2481 are duplicates that would not cause an additional delivery. At least one
2482 person was confused, thinking that -bt output corresponded to deliveries.
2483 (Suppressing duplicates isn't a good idea as you lose the information
2484 about possibly different redirections that led to the duplicates.)
2486 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
2487 systems where poll() doesn't work, in particular OS X.
2489 PH/10 Added more information to debugging output for retry time not reached.
2491 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
2492 operations in malware.c.
2494 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
2497 PH/13 If write_rejectlog was set false when logging was sent to syslog with
2498 syslog_duplication set false, log lines that would normally be written
2499 both the the main log and to the reject log were not written to syslog at
2502 PH/14 In the default configuration, change the use of "message" in ACL warn
2503 statements to "add_header".
2505 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
2506 not followed by a command (e.g. "seen endif").
2508 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
2509 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
2512 PH/17 Added extra conditions to the default value of delay_warning_condition
2516 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
2517 { match{$h_precedence:}{(?i)bulk|list|junk} } \
2518 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
2521 The Auto-Submitted: and various List- headers are standardised, whereas I
2522 don't think Precedence: ever was.
2524 PH/18 Refactored debugging code in route_finduser() to show more information,
2525 in particular, the error code if getpwnam() issues one.
2527 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
2528 This is apparently needed in addition to the PH/07 change above to avoid
2529 any possible encoding problems.
2531 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
2532 but not after initializing Perl.
2534 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
2535 output them only if debugging. By default they are written stderr,
2536 apparently, which is not desirable.
2538 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
2541 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
2544 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
2546 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
2547 authenticated or an ident call has been made. Suppress the default
2548 values for $authenticated_id and $authenticated_sender (but permit -oMai
2549 and -oMas) when testing with -bh.
2551 PH/24 Re-jigged the order of the tests in the default configuration so that the
2552 tests for valid domains and recipients precede the DNS black list and CSA
2553 tests, on the grounds that those ones are more expensive.
2555 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
2556 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
2557 This bug exists in every version of Exim that I still have, right back to
2560 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
2561 However, an attempt to turn on case-sensitivity in a regex key by
2562 including (?-i) didn't work because the subject string was already
2563 lowercased, and the effects were non-intuitive. It turns out that a
2564 one-line patch can be used to allow (?-i) to work as expected.
2570 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
2571 other effects) broke the use of negated acl sub-conditions.
2573 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
2576 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
2577 "Deny" causes Exim to reject the incoming connection with a 554 error.
2578 Unfortunately, if there is a major crisis, such as a disk failure,
2579 tcp-wrappers gives "deny", whereas what one would like would be some
2580 kind of temporary error. A kludge has been added to help with this.
2581 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
2582 554 error is used if errno is still zero or contains ENOENT (which occurs
2583 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
2586 PH/03 Add -lutil to the default FreeBSD LIBS setting.
2588 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
2589 errors. Otherwise a message that provokes a temporary error (when other
2590 messages do not) can cause a whole host to time out.
2592 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
2593 addresses were routed directly to files or pipes from a redirect router.
2594 File deliveries just didn't batch; pipe deliveries might have suffered
2597 PH/06 A failure to get a lock for a hints database would erroneously always say
2598 "Failed to get write lock", even when it was really a read lock.
2600 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
2601 of 0600. This has been changed to use the value of the lockfile_mode
2602 option (which defaults to 0600).
2604 PH/08 Applied small patch from the Sieve maintainer.
2606 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
2607 folder from quota calculations, a direct delivery into this folder messed
2608 up the contents of the maildirsize file. This was because the regex was
2609 used only to exclude .Trash (or whatever) when the size of the mailbox
2610 was calculated. There was no check that a delivery was happening into an
2611 excluded directory. This bug has been fixed by ignoring all quota
2612 processing for deliveries into excluded directories.
2614 PH/10 Added the maildirfolder_create_regex option to appendfile.
2620 PH/01 The code for finding all the local interface addresses on a FreeBSD
2621 system running IPv6 was broken. This may well have applied to all BSD
2622 systems, as well as to others that have similar system calls. The broken
2623 code found IPv4 interfaces correctly, but gave incorrect values for the
2624 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
2625 that it would not match correctly against @[] and not recognize the IPv6
2628 PH/02 The ipliteral router was not recognizing addresses of the form user@
2629 [ipv6:....] because it didn't know about the "ipv6:" prefix.
2631 PH/03 Added disable_ipv6.
2633 PH/04 Changed $reply_address to use the raw form of the headers instead of the
2634 decoded form, because it is most often used to construct To: headers
2635 lines in autoreplies, and the decoded form may well be syntactically
2636 invalid. However, $reply_address has leading white space removed, and all
2637 newlines turned into spaces so that the autoreply transport does not
2640 PH/05 If group was specified without a user on a router, and no group or user
2641 was specified on a transport, the group from the router was ignored.
2643 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
2644 for visible compile-time settings that can be used to change these
2645 numbers, for those that want even more. Backwards compatibility with old
2646 spool files has been maintained. However, going back to a previous Exim
2647 release will lost any variables that are in spool files.
2649 PH/07 Two small changes when running in the test harness: increase delay when
2650 passing a TCP/IP connection to a new process, in case the original
2651 process has to generate a bounce, and remove special handling of
2652 127.0.0.2 (sic), which is no longer necessary.
2654 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
2655 be the same on different OS.
2657 PH/09 Moved a debug statement in filter processing to avoid a race problem when
2660 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
2661 whether --show-vars was specified or not
2663 JJ/02 exipick: Added support for new ACL variable spool format introduced
2666 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
2667 syntactically invalid From: or Reply-to: line, and a filter used this to
2668 generate an autoreply, and therefore failed to obtain an address for the
2669 autoreply, Exim could try to deliver to a non-existent relative file
2670 name, causing unrelated and misleading errors. What now happens is that
2671 it logs this as a hard delivery error, but does not attempt to create a
2674 PH/11 The exinext utility has a -C option for testing purposes, but although
2675 the given file was scanned by exinext itself; it wasn't being passed on
2676 when Exim was called.
2678 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
2679 an end-of-file indication when reading a command response.
2681 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
2682 compiled. In many other places in Exim, IPv6 addresses are always
2683 recognized, so I have changed this. It also means that IPv4 domain
2684 literals of the form [IPV4:n.n.n.n] are now always recognized.
2686 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
2687 used if the router is not running as root, for example, when verifying at
2688 ACL time, or when using -bh. The debugging output from this situation was
2689 non-existent - all you got was a failure to exec. I have made two
2692 (a) Failures to set uid/gid, the current directory, or a process leader
2693 in a subprocess such as that created by queryprogram now generate
2694 suitable debugging ouput when -d is set.
2696 (b) The queryprogram router detects when it is not running as root,
2697 outputs suitable debugging information if -d is set, and then runs
2698 the subprocess without attempting to change uid/gid.
2700 PH/15 Minor change to Makefile for building test_host (undocumented testing
2703 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
2704 additional section of a DNS packet that returns MX or SRV records.
2705 Instead, it always explicitly searches for A/AAAA records. This avoids
2706 major problems that occur when a DNS server includes only records of one
2707 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
2708 fixed another bug: if SRV records were looked up and the corresponding
2709 address records were *not* found in the additional section, the port
2710 values from the SRV records were lost.
2712 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
2713 using the correct key (the original address) when searching the retry
2714 rules in order to find which one to use for generating the retry hint.
2716 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
2717 adding the default one. Similarly, if it contains a Reply-To: header, the
2718 errors_reply_to option, if set, is not used.
2720 PH/19 When calculating a retry time, Exim used to measure the "time since
2721 failure" by looking at the "first failed" field in the retry record. Now
2722 it does not use this if it is later than than the arrival time of the
2723 message. Instead it uses the arrival time. This makes for better
2724 behaviour in cases where some deliveries succeed, thus re-setting the
2725 "first failed" field. An example is a quota failure for a huge message
2726 when small messages continue to be delivered. Without this change, the
2727 "time since failure" will always be short, possible causing more frequent
2728 delivery attempts for the huge message than are intended.
2729 [Note: This change was subsequently modified - see PH/04 for 4.62.]
2731 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
2732 $1, $2, $3) because the numerical variables can be reset during some
2733 expansion items (e.g. "match"), thereby losing the authentication data.
2735 PH/21 Make -bV show the size of off_t variables so that the test suite can
2736 decide whether to run tests for quotas > 2G.
2738 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
2739 mailbox_size, and mailbox_filecount in the appendfile transport. If a
2740 filecount value is greater than 2G or if a quota value is greater than 2G
2741 on a system where the size of off_t is not greater than 4, a panic error
2744 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
2745 never match. The debug and -bh output now contains an explicit error
2746 message indicating a malformed IPv4 address or mask.
2748 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
2749 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
2750 PH/23 above applies.
2752 PH/25 Do not write to syslog when running in the test harness. The only
2753 occasion when this arises is a failure to open the main or panic logs
2754 (for which there is an explicit test).
2756 PH/26 Added the /no_tell option to "control=freeze".
2758 PH/27 If a host name lookup failed very early in a connection, for example, if
2759 the IP address matched host_lookup and the reverse lookup yielded a name
2760 that did not have a forward lookup, an error message of the form "no IP
2761 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
2762 could be logged. Now it outputs the IP address instead of "NULL".
2764 PH/28 An enabling patch from MH: add new function child_open_exim2() which
2765 allows the sender and the authenticated sender to be set when
2766 submitting a message from within Exim. Since child_open_exim() is
2767 documented for local_scan(), the new function should be too.
2769 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
2770 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
2771 results in an empty string is now treated as unset.
2773 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
2775 PH/31 Added sender_verify_fail logging option.
2777 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
2778 needed by RFC 822 but not by RFC 2822 was commented out. I have now
2779 tidied the source and removed it altogether.
2781 PH/33 When a queue run was abandoned because the load average was too high, a
2782 log line was always written; now it is written only if the queue_run log
2783 selector is set. In addition, the log line for abandonment now contains
2784 information about the queue run such as the pid. This is always present
2785 in "start" and "stop" lines but was omitted from the "abandon" line.
2787 PH/34 Omit spaces between a header name and the colon in the error message that
2788 is given when verify = headers_syntax fails (if there are lots of them,
2789 the message gets confusing).
2791 PH/35 Change the default for dns_check_names_pattern to allow slashes within
2792 names, as there are now some PTR records that contain slashes. This check
2793 is only to protect against broken name servers that fall over on strange
2794 characters, so the fact that it applies to all lookups doesn't matter.
2796 PH/36 Now that the new test suite is complete, we can remove some of the
2797 special code in Exim that was needed for the old test suite. For example,
2798 sorting DNS records because real resolvers return them in an arbitrary
2799 order. The new test suite's fake resolver always returns records in the
2802 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
2803 bounces) except when queue_only is set, to avoid logging races between
2804 the different processes.
2806 PH/38 Panic-die if .include specifies a non-absolute path.
2808 PH/39 A tweak to the "H" retry rule from its user.
2810 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
2811 a label. They prevented compilation on older perls.
2813 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
2814 a warning to be raised on newish perls.
2816 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
2817 on queue. Changes to match documented behaviour of showing count of
2818 messages matching specified criteria.
2820 PH/40 Changed the default ident timeout from 30s to 5s.
2822 PH/41 Added support for the use of login_cap features, on those BSD systems
2823 that have them, for controlling the resources used by pipe deliveries.
2825 PH/42 The content-scanning code uses fopen() to create files in which to put
2826 message data. Previously it was not paying any attention to the mode of
2827 the files. Exim runs with umask(0) because the rest of the code creates
2828 files with open(), and sets the required mode explicitly. Thus, these
2829 files were ending up world-writeable. This was not a big issue, because,
2830 being within the spool directory, they were not world-accessible. I have
2831 created a function called modefopen, which takes an additional mode
2832 argument. It sets umask(777), creates the file, chmods it to the required
2833 mode, then resets the umask. All the relevant calls to fopen() in the
2834 content scanning code have been changed to use this function.
2836 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
2837 to 24 hours. This avoids potential overflow problems when processing G
2838 and H retry rules. I suspect nobody ever tinkers with this value.
2840 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
2842 PH/45 When the plaintext authenticator is running as a client, the server's
2843 challenges are checked to ensure they are valid base64 strings. By
2844 default, the authentication attempt is cancelled if an invalid string is
2845 received. Setting client_ignore_invalid_base64 true ignores these errors.
2846 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
2847 they are received. Thus, the responses can be made to depend on the
2848 challenges. If an invalid string is ignored, an empty string is placed in
2851 PH/46 Messages that are created by the autoreply transport now contains a
2852 References: header, in accordance with RFCs 2822 and 3834.
2854 PH/47 Added authenticated_sender_force to the smtp transport.
2856 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
2858 PH/49 Installed latest patch from the Sieve maintainer.
2860 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
2861 was also set, the appendfile transport was unnecessarily scanning a
2862 directory of message files (e.g. for maildir delivery) to find the count
2863 of files (along with the size), even though it did not need this
2864 information. It now does the scan only if it needs to find either the
2865 size of the count of files.
2867 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
2869 PH/52 Two bugs concerned with error handling when the smtp transport is
2872 (i) Exim was not creating retry information for temporary errors given
2873 for individual recipients after the DATA command when the smtp transport
2874 was used in LMTP mode. This meant that they could be retried too
2875 frequently, and not timed out correctly.
2877 (ii) Exim was setting the flag that allows error details to be returned
2878 for LMTP errors on RCPT commands, but not for LMTP errors for individual
2879 recipients that were returned after the DATA command.
2881 PH/53 This is related to PH/52, but is more general: for any failing address,
2882 when detailed error information was permitted to be returned to the
2883 sender, but the error was temporary, then after the final timeout, only
2884 "retry timeout exceeded" was returned. Now it returns the full error as
2885 well as "retry timeout exceeded".
2887 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
2888 do this, and (what is worse) MTAs that accept it.
2890 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
2891 will now be deprecated.
2893 PH/56 New os.c-cygwin from the Cygwin maintainer.
2895 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
2896 formats (previously only available in exim formats via -bpr, -bpru,
2897 and -bpra. Now also available in native and exiqgrep formats)
2899 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
2900 with very large, slow to parse queues
2902 JJ/08 exipick: added ! as generic prefix to negate any criteria format
2904 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
2906 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
2907 responses to authentication challenges, though it was showing the
2908 challenges; (ii) I've removed the CR characters from the debug output for
2911 PH/58 Allow for the insertion of a newline as well as a space when a string
2912 is turned into more than one encoded-word during RFC 2047 encoding. The
2913 Sieve code now uses this.
2915 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
2916 data_4xx, lost_connection, tls_required.
2918 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
2919 message was being sent as an SMTP response.
2921 PH/61 Add -l and -k options to exicyclog.
2923 PH/62 When verifying, if an address was redirected to one new address, so that
2924 verification continued, and the new address failed or deferred after
2925 having set something in $address_data, the value of $address_data was not
2926 passed back to the ACL. This was different to the case when no
2927 redirection occurred. The value is now passed back in both cases.
2929 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
2930 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
2931 use login_cap.h, so on its own it isn't the distinguishing feature. The
2932 new name refers directly to the setclassresources() function.
2934 PH/65 Added configuration files for NetBSD3.
2936 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
2938 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
2939 is preferred over IPv4.
2941 PH/68 The bounce_return_message and bounce_return_body options were not being
2942 honoured for bounces generated during the reception of non-SMTP messages.
2943 In particular, this applied to messages rejected by the ACL. This bug has
2944 been fixed. However, if bounce_return_message is true and bounce_return_
2945 body is false, the headers that are returned for a non-SMTP message
2946 include only those that have been read before the error was detected.
2947 (In the case of an ACL rejection, they have all been read.)
2949 PH/69 The HTML version of the specification is now built in a directory called
2950 spec_html instead of spec.html, because the latter looks like a path with
2951 a MIME-type, and this confuses some software.
2953 PH/70 Catch two compiler warnings in sieve.c.
2955 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
2956 function verify_get_ident() calls ip_connect() to connect a socket, but
2957 if the "connect()" function timed out, ip_connect() used to close the
2958 socket. However, verify_get_ident() also closes the socket later, and in
2959 between Exim writes to the log, which may get opened at this point. When
2960 the socket was closed in ip_connect(), the log could get the same file
2961 descriptor number as the socket. This naturally causes chaos. The fix is
2962 not to close the socket in ip_connect(); the socket should be closed by
2963 the function that creates it. There was only one place in the code where
2964 this was missing, in the iplookup router, which I don't think anybody now
2965 uses, but I've fixed it anyway.
2967 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
2968 well as to direct DNS lookups. Otherwise the handling of names in host
2969 lists is inconsistent and therefore confusing.
2975 PH/01 Two changes to the default runtime configuration:
2977 (1) Move the checks for relay_from_hosts and authenticated clients from
2978 after to before the (commented out) DNS black list checks.
2980 (2) Add control=submission to the relay_from_hosts and authenticated
2981 clients checks, on the grounds that messages accepted by these
2982 statements are most likely to be submissions.
2984 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
2986 (1) Generate an error if the third argument for the ${prvs expansion is
2989 (2) Treat a missing third argument of ${prvscheck as if it were an empty
2992 (3) Reset the variables that are obtained from the first argument of
2993 ${prvscheck and used in the second argument before leaving the code,
2994 because their memory is reclaimed, so using them afterwards may do
2997 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
2998 one (it's much easier than Tom thought :-).
3000 (5) Because of (4), we can now allow for the use of $prvscheck_result
3001 inside the third argument.
3003 PH/03 For some reason, the default setting of PATH when running a command from
3004 a pipe transport was just "/usr/bin". I have changed it to
3007 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
3008 anything to be listed in the output from -bV.
3010 PH/05 When a filter generated an autoreply, the entire To: header line was
3011 quoted in the delivery log line, like this:
3013 => >A.N.Other <ano@some.domain> <original@ddress> ...
3015 This has been changed so that it extracts the operative address. There
3016 may be more than one such address. If so, they are comma-separated, like
3019 => >ano@some.domain,ona@other.domain <original@ddress> ...
3021 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
3022 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
3023 not being looked up in the rDNS to get a host name, Exim was showing the
3024 IP address twice in Received: lines, even though the IP addresses were
3025 identical. For example:
3027 Received: from [1.2.3.4] (helo=[1.2.3.4])
3029 However, if the real host name was known, it was omitting the HELO data
3030 if it matched the actual IP address. This has been tidied up so that it
3031 doesn't show the same IP address twice.
3033 PH/07 When both +timestamp and +memory debugging was on, the value given by
3034 $tod_xxx expansions could be wrong, because the tod_stamp() function was
3035 called by the debug printing, thereby overwriting the timestamp buffer.
3036 Debugging no longer uses the tod_stamp() function when +timestamp is set.
3038 PH/08 When the original message was included in an autoreply transport, it
3039 always said "this is a copy of the message, including all the headers",
3040 even if body_only or headers_only was set. It now gives an appropriate
3043 PH/09 Applied a patch from the Sieve maintainer which:
3045 o fixes some comments
3046 o adds the (disabled) notify extension core
3047 o adds some debug output for the result of if/elsif tests
3048 o points to the current vacation draft in the documentation
3049 and documents the missing references header update
3053 o fixes a bug in processing the envelope test (when testing
3054 multiple envelope elements, the last element determinted the
3057 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
3058 Electronic Mail") by including:
3060 Auto-submitted: auto-generated
3062 in the messages that it generates (bounce messages and others, such as
3063 warnings). In the case of bounce messages for non-SMTP mesages, there was
3064 also a typo: it was using "Auto_submitted" (underscore instead of
3065 hyphen). Since every message generated by Exim is necessarily in response
3066 to another message, thes have all been changed to:
3068 Auto-Submitted: auto-replied
3070 in accordance with these statements in the RFC:
3072 The auto-replied keyword:
3074 - SHOULD be used on messages sent in direct response to another
3075 message by an automatic process,
3077 - MUST NOT be used on manually-generated messages,
3079 - MAY be used on Delivery Status Notifications (DSNs) and Message
3080 Disposition Notifications (MDNs),
3082 - MUST NOT be used on messages generated by automatic or periodic
3083 processes, except for messages which are automatic responses to
3086 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
3087 to the default Received: header definition.
3089 PH/12 Added log selector acl_warn_skipped (default on).
3091 PH/13 After a successful wildlsearch lookup, discard the values of numeric
3092 variables because (a) they are in the wrong storage pool and (b) even if
3093 they were copied, it wouldn't work properly because of the caching.
3095 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
3096 checking when decoding. Apparently there are clients that generate
3097 overlong encoded strings. Why am I not surprised?
3099 PH/15 If the first argument of "${if match_address" was not empty, but did not
3100 contain an "@" character, Exim crashed. Now it writes a panic log message
3101 and treats the condition as false.
3103 PH/16 In autoreply, treat an empty string for "once" the same as unset.
3105 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
3106 extension "envelope-auth". The code is finished and in agreement with
3107 other implementations, but there is no documentation so far and in fact,
3108 nobody wrote the draft yet. This extension is currently #undef'ed, thus
3109 not changing the active code.
3111 Print executed "if" and "elsif" statements when debugging is used. This
3112 helps a great deal to understand what a filter does.
3114 Document more things not specified clearly in RFC3028. I had all this
3115 sorted out, when out of a sudden new issues came to my mind. Oops."
3117 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
3120 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
3121 canonical form (as documented). However, after a host name lookup from
3122 the IP address, check_host() was doing a simple string comparison with
3123 addresses acquired from the DNS when checking that the found name did
3124 have the original IP as one of its addresses. Since any found IPv6
3125 addresses are likely to be in abbreviated form, the comparison could
3126 fail. Luckily, there already exists a function for doing the comparison
3127 by converting both addresses to binary, so now that is used instead of
3128 the text comparison.
3130 PH/20 There was another similar case to PH/19, when a complete host name was
3131 given in a host list; looking up its IP address could give an abbreviated
3132 form, whereas the current host's name might or might not be abbreviated.
3133 The same fix has been applied.
3139 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
3140 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
3143 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
3144 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
3146 PH/03 Typo: missing ".o" in src/pcre/Makefile.
3148 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
3149 header line, restrict the check to what is listed in RFCs 2369 and 2929.
3150 Also, for "Auto-Submitted", treat anything other than "no" as
3151 non-personal, in accordance with RFC 3834. (Previously it treated
3152 anything starting "auto-" as non-personal.)
3154 TF/01 The control=submission/name=... option had a problem with syntax
3155 errors if the name included a slash character. The /name= option
3156 now slurps the rest of the string, so it can include any characters
3157 but it must come last in the list of options (after /sender_retain
3160 PH/05 Some modifications to the interface to the fake nameserver for the new
3168 TK/01 Added the "success_on_redirect" address verification option. See
3169 NewStuff for rationale and an example.
3171 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
3173 PH/02 Patch to exigrep to allow it to work on syslog lines.
3175 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
3176 fread() to skip over the body file's header line, because in Cygwin the
3177 header line is locked and is inaccessible.
3179 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
3180 co-exist for some time) to make it clear that it is the Exim ID that is
3181 referenced, not the Message-ID: header line.
3183 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
3184 string_format() function, because snprintf() does not exist on all
3187 PH/06 The use of forbid_filter_existstest now also locks out the use of the
3188 ${stat: expansion item.
3190 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
3191 protocol synchronization error", to keep the pedants happy.
3193 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
3194 well as for IRIX systems, when gcc is being used. See the host.c source
3197 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
3199 PH/10 Named domain lists were not working if used in a queue_smtp_domains
3202 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
3203 transport and to the smtp transport in LMTP mode.
3205 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
3207 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
3208 run a filter in a subprocess. This could lead to confusion in subsequent
3209 lookups in the parent process. There should also be a search_tidyup() at
3210 the end of the subprocess.
3212 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
3213 only if the host matched helo_try_verify_hosts, which caused the
3214 verification to occur when the EHLO/HELO command was issued. The ACL just
3215 tested the remembered result. Now, if a previous verification attempt has
3216 not happened, "verify = helo" does it there and then.
3218 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
3220 TK/03 Fix log output including CR from clamd.
3222 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
3223 exist provoked a memory error which could cause a segfault.
3225 PH/15 Installed PCRE 6.2
3227 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
3229 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
3230 of the problem. Specifically, suggested +O2 rather than +O1 for the
3233 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
3235 PH/20 If a delivery was routed to a non-standard port by means of an SRV
3236 record, the port was not correctly logged when the outgoing_port log
3237 selector was set (it logged the transort's default port).
3239 PH/21 Added support for host-specific ports to manualroute, queryprogram,
3240 fallback_hosts, and "hosts" in the smtp transport.
3242 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
3243 host errors such as "Connection refused".
3245 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
3246 authentication with radiusclient 0.4.9:
3248 - Error returned from rc_read_config was caught wrongly
3249 - Username/password not passed on to radius server due to wrong length.
3251 The presumption is that some radiusclient API changes for 4.51/PH/17
3252 were not taken care of correctly. The code is still untested by me (my
3253 Linux distribution still has 0.3.2 of radiusclient), but it was
3254 contributed by a Radius user.
3256 PH/24 When doing a callout, the value of $domain wasn't set correctly when
3257 expanding the "port" option of the smtp transport.
3259 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
3260 while reading a MIME header. Thanks to Tom Hughes for a patch.
3262 PH/24 Include config.h inside local_scan.h so that configuration settings are
3265 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
3266 that in an ACL for RCPT (for example), you can examine exactly what was
3269 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
3270 commands, but it was not correctly comparing the address with the actual
3271 client host address. Thus, it would show the EHLO address in Received:
3272 header lines when this was not necessary.
3274 PH/27 Added the % operator to ${eval:}.
3276 PH/28 Exim tries to create and chdir to its spool directory when it starts;
3277 it should be ignoring failures (because with -C, for example, it has lost
3278 privilege). It wasn't ignoring creation failures other than "already
3281 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
3284 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
3285 because an input line was too long, either on its own, or by virtue of
3286 too many continuations, the temporary file was not being removed, and the
3287 return code was incorrect.
3289 PH/31 Missing "BOOL" in function definition in filtertest.c.
3291 PH/32 Applied Sieve patches from the maintainer.
3293 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
3295 PH/33 Added "verify = not_blind".
3297 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
3298 Local/Makefile (with some defaults set). These are used in built scripts
3299 such as exicyclog, but they have never been used in the exim_install
3300 script (though there are many overriding facilities there). I have
3301 arranged that the exim_install script now takes note of these two
3304 PH/35 Installed configuration files for Dragonfly.
3306 PH/36 When a locally submitted message by a trusted user did not contain a
3307 From: header, and the sender address was obtained from -f or from an SMTP
3308 MAIL command, and the trusted user did not use -F to supply a sender
3309 name, $originator_name was incorrectly used when constructing a From:
3310 header. Furthermore, $originator_name was used for submission mode
3311 messages from external hosts without From: headers in a similar way,
3312 which is clearly wrong.
3314 PH/37 Added control=suppress_local_fixups.
3316 PH/38 When log_selector = +received_sender was set, and the addition of the
3317 sender made the log line's construction buffer exactly full, or one byte
3318 less than full, an overflow happened when the terminating "\n" was
3321 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
3322 when the result of a list match is failure because a DNS lookup failed.
3324 PH/40 RM_COMMAND is now used in the building process.
3326 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
3327 the "build-* directories that it finds.
3329 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
3330 address in a domain literal was a prefix of an interface address.
3332 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
3333 when verifying a sender address, unless rewrite_headers is false.
3335 PH/44 Wrote a long comment about why errors_to addresses are verified as
3336 recipients, not senders.
3338 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
3339 the ratelimit ACL was added.
3341 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
3343 PH/46 Added extra information about PostgreSQL errors to the error string.
3345 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
3346 avoiding the need to install special zones in a real server. This is
3347 backwards compatible; if it can't find the fake resolver, it drops back.
3348 Thus, both old and new test suites can be run.
3350 TF/02 Added util/ratelimit.pl
3352 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
3353 clock is set back in time.
3355 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
3356 Candler <B.Candler@pobox.com>.
3358 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
3359 OK for addresses that are the result of redirections.
3361 PH/48 A number of further additions for the benefit of the new test suite,
3362 including a fake gethostbyname() that interfaces to the fake DNS resolver
3365 TF/06 The fix for widen_domains has also been applied to qualify_single and
3366 search_parents which are the other dnslookup options that can cause
3369 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
3372 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
3374 TF/07 Exim produced the error message "an SRV record indicated no SMTP
3375 service" if it encountered an MX record with an empty target hostname.
3376 The message is now "an MX or SRV record indicated no SMTP service".
3378 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
3379 if the DNS of the sending site is misconfigured. This is quite a
3380 common situation. This change restores the behaviour of treating a
3381 helo verification defer as a failure.
3383 PH/51 If self=fail was set on a router, the bounce message did not include the
3384 actual error message.
3390 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
3392 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
3393 command itself ended in error, the underlying message about the transport
3394 filter timeout was being overwritten with the pipe command error. Now the
3395 underlying error message should be appended to the second error message.
3397 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
3399 PH/02 Reduce the amount of output that "make" produces by default. Full output
3400 can still be requested.
3402 PH/03 The warning log line about a condition test deferring for a "warn" verb
3403 was being output only once per connection, rather than after each
3404 occurrence (because it was using the same function as for successful
3405 "warn" verbs). This seems wrong, so I have changed it.
3407 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
3408 it should not have, which might have caused a crash in the right
3409 circumstances, but probably never did.
3411 PH/04 Installed a modified version of Tony Finch's patch to make submission
3412 mode fix the return path as well as the Sender: header line, and to
3413 add a /name= option so that you can make the user's friendly name appear
3416 TF/03 Added the control = fakedefer ACL modifier.
3418 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
3419 Mark Lowes for thorough testing.
3421 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
3423 TK/03 Merged latest SRS patch from Miles Wilton.
3425 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
3426 with the definition in sysexits.h (which is #included earlier).
3427 Fortunately, Exim does not actually use EX_OK. The code used to try to
3428 preserve the sysexits.h value, by assumimg that macro definitions were
3429 scanned for macro replacements. I have been disabused of this notion,
3430 so now the code just undefines EX_OK before #including unistd.h.
3432 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
3433 in the smtp transport. When a block could not be written in a single
3434 write() function, the timeout was being re-applied to each part-write.
3435 This seems wrong - if the receiver was accepting one byte at a time it
3436 would take for ever. The timeout is now adjusted when this happens. It
3437 doesn't have to be particularly precise.
3439 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
3440 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
3442 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
3443 without a domain if the check to <postmaster@domain> fails.
3445 SC/01 Eximstats: added -xls and the ability to specify output files
3446 (patch written by Frank Heydlauf).
3448 SC/02 Eximstats: use FileHandles for outputing results.
3450 SC/03 Eximstats: allow any combination of xls, txt, and html output.
3452 SC/04 Eximstats: fixed display of large numbers with -nvr option
3454 SC/05 Eximstats: fixed merging of reports with empty tables.
3456 SC/06 Eximstats: added the -include_original_destination flag
3458 SC/07 Eximstats: removed tabs and trailing whitespace.
3460 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
3462 TK/06 MBOX spool code: Add real "From " MBOX separator line
3463 so the .eml file is really in mbox format (even though
3464 most programs do not really care). Patch from Alex Miller.
3466 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
3467 The latter is generated from $received_to and is only set if the
3468 message has one envelope recipient. SA can use these headers,
3469 obviously out-of-the-box. Patch from Alex Miller.
3471 PH/08 The ${def test on a variable was returning false if the variable's
3472 value was "0", contrary to what the specification has always said!
3473 The result should be true unless the variable is empty.
3475 PH/09 The syntax error of a character other than { following "${if
3476 def:variable_name" (after optional whitespace) was not being diagnosed.
3477 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
3478 accidental colon was present, for example, could give incorrect results.
3480 PH/10 Tidied the code in a number of places where the st_size field of a stat()
3481 result is used (not including appendfile, where other changes are about
3484 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
3485 This involved changing a lot of size variables from int to off_t. It
3486 should work with maildirs and everything.
3488 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
3489 spamd dying while we are connected to it.
3491 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
3494 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
3495 function for BDB 4.3.
3497 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
3499 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
3500 the sources for PCRE have been re-arranged and more files are now
3503 PH/15 The code I had for printing potentially long long variables in PH/11
3504 above was not the best (it lost precision). The length of off_t variables
3505 is now inspected at build time, and an appropriate printing format (%ld
3506 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
3507 to be "long long int" or "long int". This is needed for the internal
3508 formatting function string_vformat().
3510 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
3511 the configuration file to be ":syslog", then the script "guesses" where
3512 the logs files are, rather than using the compiled in default. In our
3513 case the guess is not the same as the compiled default, so the script
3514 suddenly stopped working when I started to use syslog. The patch checks
3515 to see if log_file_path is "". If so, it attempts to read it from exim
3516 with no configuration file to get the compiled in version, before it
3517 falls back to the previous guessing code."
3519 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
3520 implementing BATV in an Exim configuration. See NewStuff for the gory
3523 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
3524 Makefile that are specific to HP-UX.
3526 PH/18 If the "use_postmaster" option was set for a recipient callout together
3527 with the "random" option, the postmaster address was used as the MAIL
3528 FROM address for the random test, but not for the subsequent recipient
3529 test. It is now used for both.
3531 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
3532 patch removes a few documentation additions to RFC 3028, because the
3533 latest draft now contains them. It adds the new en;ascii-case comparator
3534 and a new error check for 8bit text in MIME parts. Comparator and
3535 require names are now matched exactly. I enabled the subaddress
3536 extension, but it is not well tested yet (read: it works for me)."
3538 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
3539 rework some of the code of TK/09 above to avoid the hardwired use of
3540 "%lld" and "long long". Replaced the call to snprintf() with a call to
3543 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
3544 records point to non-existent hosts", "retry timeout exceeded", and
3545 "retry time not reached for any host after a long failure period".
3547 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
3548 experimental DomainKeys support:
3550 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
3551 (2) On an error such as an illegally used "control", the wrong name for
3552 the control was given.
3554 These problems did NOT occur unless DomainKeys support was compiled.
3556 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
3558 PH/24 Added ${if match_ip condition.
3560 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
3561 databases so that it will be absolutely obvious if a crash occurs in the
3562 DB library. This is a regular occurrence (often caused by mis-matched
3565 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
3566 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
3567 picked up on a user's system that detects such things. There doesn't seem
3568 to be a gcc warning option for this - only an attribute that has to be
3569 put on the function's prototype. It seems that in Fedora Core 4 they have
3570 set this on a number of new functions. No doubt there will be more in due
3573 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
3574 specify a transport. However, if an address that was verified by such a
3575 router was the subject of a callout, Exim crashed because it tried to
3576 read the rcpt_include_affixes from the non-existent transport. Now it
3577 just assumes that the setting of that option is false. This bug was
3578 introduced by 4.51/PH/31.
3580 PH/28 Changed -d+all to exclude +memory, because that information is very
3581 rarely of interest, but it makes the output a lot bigger. People tend to
3582 do -d+all out of habit.
3584 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
3585 code in os-type was giving problems when libc.so lives in lib64, like on
3588 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
3589 aren't the modern standard, and it seems that some systems' include files
3590 don't always have them. Exim was already checking for some of the newer
3591 ones like T_AAAA, and defining it itself. I've added checks for all the
3592 record types that Exim uses.
3594 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
3595 not automatically generating a new one, as it is supposed to. This
3596 prevented TLS from working. If the file did exist, but contained invalid
3597 data, a new version was generated, as expected. It was only the case of a
3598 non-existent file that was broken.
3600 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
3601 with a change in libdomainkeys > 0.64.
3603 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
3604 from DNS. If the selector record carries the flag, it now has
3605 precedence over the domain-wide flag.
3607 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
3609 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
3610 the use of a port name that isn't defined in /etc/services) occurred, the
3611 message was deferred as in a normal delivery, and thus remained on the
3612 spool, instead of being failed because of the mua_wrapper setting. This
3613 is now fixed, and I tidied up some of the mua_wrapper messages at the
3616 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
3617 the messages in a hash of arrays rather than using individual hashes.
3618 This is a bit cleaner and results in dramatic memory savings, albeit
3619 at a slight CPU cost.
3621 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
3622 as requested by Marc Sherman.
3624 SC/10 Eximstats: added histograms for user specified patterns as requested
3627 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
3629 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
3630 fopen() in the content-scanning modules that did not already have it.
3636 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
3637 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
3639 TK/02 Fix ACL "control" statement not being available in MIME ACL.
3641 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
3643 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
3644 to test Sieve filters that use "vacation".
3646 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
3647 that changes the way the GnuTLS parameters are stored in the cache file.
3648 The new format can be generated externally. For backward compatibility,
3649 if the data in the cache doesn't make sense, Exim assumes it has read an
3650 old-format file, and it generates new data and writes a new file. This
3651 means that you can't go back to an older release without removing the
3654 PH/03 A redirect router that has both "unseen" and "one_time" set does not
3655 work if there are any delivery delays because "one_time" forces the
3656 parent to be marked "delivered", so its unseen clone is never tried
3657 again. For this reason, Exim now forbids the simultaneous setting of
3660 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
3661 redirected to themselves ("homonym" addresses). Read the long ChangeLog
3662 entry if you want to know the details. The fix, however, neglected to
3663 consider the case when local delivery batching is involved. The test for
3664 "previously delivered" was not happening when checking to see if an
3665 address could be batched with a previous (undelivered) one; under
3666 certain circumstances this could lead to multiple deliveries to the same
3669 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
3670 in its include files, and this causes problems building Exim.
3672 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
3673 header_syntax) but Exim was just ignoring anything given after a slash.
3674 In particular, this caused confusion with an attempt to use "verify =
3675 reverse_host_lookup/defer_ok". An error is now given when options are
3676 supplied for verify items that do not have them. (Maybe reverse_host_
3677 lookup should have a defer_ok option, but that's a different point.)
3679 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
3680 defined by RFC 821) to 2048, because there were problems with some AUTH
3681 commands, and RFC 1869 says the size should be increased for extended
3682 SMTP commands that take arguments.
3684 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
3687 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
3688 "unknown" error; now it says that the functionality isn't in the binary.
3690 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
3691 an address' error message when a string expansion fails (syntax or
3692 whatever). Otherwise the password may appear in the log. Following change
3693 PH/42 below, there is no longer a chance of it appearing in a bounce
3696 PH/11 Installed exipick version 20050225.0 from John Jetmore.
3698 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
3699 of its addresses was ever tried. (Bugzilla bug #2.)
3701 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
3702 the result incorrectly in the debug output. (It correctly added a newline
3703 to what was transported.)
3705 TF/01 Added $received_time.
3707 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
3708 commented out examples of how to interface to a virus scanner and to
3709 SpamAssassin. Also added commented examples of av_scanner and
3710 spamd_address settings.
3712 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
3713 and controls are allowed in which ACLs. There were a couple of minor
3714 errors. Some of the entries in the conditions table (which is a table of
3715 where they are NOT allowed) were getting very unwieldy; rewrote them as a
3716 negation of where the condition IS allowed.
3718 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
3720 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
3721 header file does not have a version number, so I've had to invent a new
3722 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
3723 API. The code is untested by me (my Linux distribution still has 0.3.2 of
3724 radiusclient), but it was contributed by a Radius user.
3726 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
3727 files or directories, for OpenSSL.
3729 PH/19 When an Exim process that is running as root has to create an Exim log
3730 file, it does so in a subprocess that runs as exim:exim so as to get the
3731 ownership right at creation (otherwise, other Exim processes might see
3732 the file with the wrong ownership). There was no test for failure of this
3733 fork() call, which would lead to the process getting stuck as it waited
3734 for a non-existent subprocess. Forks do occasionally fail when resources
3735 run out. I reviewed all the other calls to fork(); they all seem to check
3738 PH/20 When checking for unexpected SMTP input at connect time (before writing
3739 the banner), Exim was not dealing correctly with a non-positive return
3740 from the read() function. If the client had disconnected by this time,
3741 the result was a log entry for a synchronization error with an empty
3742 string after "input=" when read() returned zero. If read() returned -1
3743 (an event I could not check), uninitialized data bytes were printed.
3744 There were reports of junk text (parts of files, etc) appearing after
3747 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
3749 PH/22 Added support for macro redefinition, and (re)definition in between
3750 driver and ACL definitions.
3752 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
3753 forgetting to use the resulting value; it was using the unexpanded value.
3755 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
3756 hadn't been configured. The fix is from Juergen Kreileder, who
3757 understands it better than I do:
3759 "Here's what I see happening with three configured cyrus_sasl
3760 authenticators configured (plain, login, cram-md5):
3762 On startup auth_cyrus_sasl_init() gets called for each of these.
3763 This means three calls to sasl_listmech() without a specified mech_list.
3764 => SASL tests which mechs of all available mechs actually work
3765 => three warnings about OTP not working
3766 => the returned list contains: plain, login, cram-md5, digest-md5, ...
3768 With the patch, sasl_listmech() also gets called three times. But now
3769 SASL's mech_list option is set to the server_mech specified in the the
3770 authenticator. Or in other words, the answer from sasl_listmech()
3771 gets limited to just the mech you're testing for (which is different
3773 => the return list contains just 'plain' or 'login', 'cram-md5' or
3774 nothing depending on the value of ob->server_mech.
3776 I've just tested the patch: Authentication still works fine,
3777 unavailable mechs specified in the exim configuration are still
3778 caught, and the auth.log warnings about OTP are gone."
3780 PH/25 When debugging is enabled, the contents of the command line are added
3781 to the debugging output, even when log_selector=+arguments is not
3784 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
3785 answer is "GNU", and only if the return is "GNU/something" is the answer
3788 PH/27 $acl_verify_message is now set immediately after the failure of a
3789 verification in an ACL, and so is available in subsequent modifiers. In
3790 particular, the message can be preserved by coding like this:
3792 warn !verify = sender
3793 set acl_m0 = $acl_verify_message
3795 Previously, $acl_verify_message was set only while expanding "message"
3796 and "log_message" when a very denied access.
3798 PH/28 Modified OS/os.c-Linux with
3800 -#ifndef OS_LOAD_AVERAGE
3801 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
3803 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
3804 nomenclature these days.)
3806 PH/29 Installed patch from the Sieve maintainer that adds the options
3807 sieve_useraddress and sieve_subaddress to the redirect router.
3809 PH/30 In these circumstances:
3810 . Two addresses routed to the same list of hosts;
3811 . First host does not offer TLS;
3812 . First host accepts first address;
3813 . First host gives temporary error to second address;
3814 . Second host offers TLS and a TLS session is established;
3815 . Second host accepts second address.
3816 Exim incorrectly logged both deliveries with the TLS parameters (cipher
3817 and peerdn, if requested) that were in fact used only for the second
3820 PH/31 When doing a callout as part of verifying an address, Exim was not paying
3821 attention to any local part prefix or suffix that was matched by the
3822 router that accepted the address. It now behaves in the same way as it
3823 does for delivery: the affixes are removed from the local part unless
3824 rcpt_include_affixes is set on the transport.
3826 PH/32 Add the sender address, as F=<...>, to the log line when logging a
3827 timeout during the DATA phase of an incoming message.
3829 PH/33 Sieve envelope tests were broken for match types other than :is. I have
3830 applied a patch sanctioned by the Sieve maintainer.
3832 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
3833 the uid or gid is negative. A case of a negative gid caused this to be
3834 noticed. The fix allows for either to be negative.
3836 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
3837 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
3840 PH/36 The change PH/12 above was broken. Fixed it.
3842 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
3843 the grounds that routing the same address twice would always produce the
3844 same answer. This might have been true once, but it is certainly no
3845 longer true now. Routing a child address may depend on the previous
3846 routing that produced that child. Some complicated redirection strategies
3847 went wrong when messages had multiple recipients, and made Exim's
3848 behaviour dependent on the order in which the addresses were given.
3850 I have moved the duplicate checking until after the routing is complete.
3851 Exim scans the addresses that are assigned to local and remote
3852 transports, and removes any duplicates. This means that more work will be
3853 done, as duplicates will always all be routed, but duplicates are
3854 presumably rare, so I don't expect this is of any significance.
3856 For deliveries to pipes, files, and autoreplies, the duplicate checking
3857 still happens during the routing process, since they are not going to be
3860 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
3861 It corrects a timeout issue with spamd. This is Ian's comment: "The
3862 background is that sometimes spamd either never reads data from a
3863 connection it has accepted, or it never writes response data. The exiscan
3864 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
3865 blindly assumes that writes won't block so it may never time out."
3867 PH/39 Allow G after quota size as well as K and M.
3869 PH/40 The value set for $authenticated_id in an authenticator may not contain
3870 binary zeroes or newlines because the value is written to log lines and
3871 to spool files. There was no check on this. Now the value is run through
3872 the string_printing() function so that such characters are converted to
3873 printable escape sequences.
3875 PH/41 $message_linecount is a new variable that contains the total number of
3876 lines in the message. Compare $body_linecount, which is the count for the
3879 PH/42 Exim no longer gives details of delivery errors for specific addresses in
3880 bounce and delay warning messages, except in certain special cases, which
3883 (a) An SMTP error message from a remote host;
3884 (b) A message specified in a :fail: redirection;
3885 (c) A message specified in a "fail" command in a system filter;
3886 (d) A message specified in a FAIL return from the queryprogram router;
3887 (e) A message specified by the cannot_route_message router option.
3889 In these cases only, Exim does include the error details in bounce and
3890 warning messages. There are also a few cases where bland messages such
3891 as "unrouteable address" or "local delivery error" are given.
3893 PH/43 $value is now also set for the "else" part of a ${run expansion.
3895 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
3896 being worked on, but at least Exim now implements the latest version to
3899 PH/45 In a pipe transport, although a timeout while waiting for the pipe
3900 process to complete was treated as a delivery failure, a timeout while
3901 writing the message to the pipe was logged, but erroneously treated as a
3902 successful delivery. Such timeouts include transport filter timeouts. For
3903 consistency with the overall process timeout, these timeouts are now
3904 treated as errors, giving rise to delivery failures by default. However,
3905 there is now a new Boolean option for the pipe transport called
3906 timeout_defer, which, if set TRUE, converts the failures into defers for
3907 both kinds of timeout. A transport filter timeout is now identified in
3910 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
3911 systems where "make" and "gmake" are different, calling "gmake" at top
3912 level broke things. I've arranged for the value of $(MAKE) to be passed
3913 from the Makefile to this script so that it can call the same version of
3917 A note about Exim versions 4.44 and 4.50
3918 ----------------------------------------
3920 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
3921 changes of various kinds. As a consequence, a big documentation update was
3922 needed. This delayed the release for rather longer than seemed good, especially
3923 in the light of a couple of (minor) security issues. Therefore, the changes
3924 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
3925 release. So 4.44 and 4.50 are in effect two different branches that both start
3928 I have left the 4.50 change log unchanged; it contains all the changes since
3929 4.43. The change log for 4.44 is below; many of its items are identical to
3930 those for 4.50. This seems to be the most sensible way to preserve the
3931 historical information.
3937 1. Minor wording change to the doc/README.SIEVE file.
3939 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
3940 computation of the current number of files was incorrect.
3942 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
3943 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
3946 4. Give more explanation in the error message when the command for a transport
3947 filter fails to execute.
3949 5. There are several places where Exim runs a non-Exim command in a
3950 subprocess. The SIGUSR1 signal should be disabled for these processes. This
3951 was being done only for the command run by the queryprogram router. It is
3952 now done for all such subprocesses. The other cases are: ${run, transport
3953 filters, and the commands run by the lmtp and pipe transports.
3955 6. Added CONFIGURE_GROUP build-time option.
3957 7. Some older OS have a limit of 256 on the maximum number of file
3958 descriptors. Exim was using setrlimit() to set 1000 as a large value
3959 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
3960 systems. I've change it so that if it can't get 1000, it tries for 256.
3962 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
3963 was an oversight, and furthermore, ever since the addition of extra
3964 controls (e.g. 4.43/32), the checks on when to allow different forms of
3965 "control" were broken. There should now be diagnostics for all cases when a
3966 control that does not make sense is encountered.
3968 9. Added the /retain_sender option to "control=submission".
3970 10. $recipients is now available in the predata ACL (oversight).
3972 11. Tidy the search cache before the fork to do a delivery from a message
3973 received from the command line. Otherwise the child will trigger a lookup
3974 failure and thereby defer the delivery if it tries to use (for example) a
3975 cached ldap connection that the parent has called unbind on.
3977 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
3978 of $address_data from the recipient verification was clobbered by the
3979 sender verification.
3981 13. The value of address_data from a sender verification is now available in
3982 $sender_address_data in subsequent conditions in the ACL statement.
3984 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
3986 15. Added a new option "connect=<time>" to callout options, to set a different
3989 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
3990 was its contents. (It was OK if the option was not defined at all.)
3992 17. A "Completed" log line is now written for messages that are removed from
3993 the spool by the -Mrm option.
3995 18. New variables $sender_verify_failure and $recipient_verify_failure contain
3996 information about exactly what failed.
3998 19. Added -dd to debug only the daemon process.
4000 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
4001 handles timeouts, both on the server side and network timeouts. Renamed the
4002 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
4004 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
4005 It is now set to "smtps".
4007 22. $host_address is now set to the target address during the checking of
4008 ignore_target_hosts.
4010 23. When checking ignore_target_hosts for an ipliteral router, no host name was
4011 being passed; this would have caused $sender_host_name to have been used if
4012 matching the list had actually called for a host name (not very likely,
4013 since this list is usually IP addresses). A host name is now passed as
4016 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4017 code that specifies a non-restarting handler (typically sigaction() in
4018 modern systems) in an attempt to fix a rare and obscure crash bug.
4020 25. Narrowed the window for a race in the daemon that could cause it to ignore
4021 SIGCHLD signals. This is not a major problem, because they are used only to
4022 wake it up if nothing else does.
4024 26. A malformed maildirsize file could cause Exim to calculate negative values
4025 for the mailbox size or file count. Odd effects could occur as a result.
4026 The maildirsize information is now recalculated if the size or filecount
4029 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4030 support for a long time. Removed HAVE_SYS_VFS_H.
4032 28. Installed the latest version of exipick from John Jetmore.
4034 29. In an address list, if the pattern was not a regular expression, an empty
4035 subject address (from a bounce message) matched only if the pattern was an
4036 empty string. Non-empty patterns were not even tested. This was the wrong
4037 because it is perfectly reasonable to use an empty address as part of a
4038 database query. An empty address is now tested by patterns that are
4039 lookups. However, all the other forms of pattern expect the subject to
4040 contain a local part and a domain, and therefore, for them, an empty
4041 address still always fails if the pattern is not itself empty.
4043 30. Exim went into a mad DNS loop when attempting to do a callout where the
4044 host was specified on an smtp transport, and looking it up yielded more
4045 than one IP address.
4047 31. Re-factored the code for checking spool and log partition space into a
4048 function that finds that data and another that does the check. The former
4049 is then used to implement four new variables: $spool_space, $log_space,
4050 $spool_inodes, and $log_inodes.
4052 32. The RFC2047 encoding function was originally intended for short strings
4053 such as real names; it was not keeping to the 75-character limit for
4054 encoded words that the RFC imposes. It now respects the limit, and
4055 generates multiple encoded words if necessary. To be on the safe side, I
4056 have increased the buffer size for the ${rfc2047: expansion operator from
4059 33. It is now permitted to omit both strings after an "if" condition; if the
4060 condition is true, the result is "true". As before, when the second string
4061 is omitted, a false condition yields an empty string. This makes it less
4062 cumbersome to write custom ACL and router conditions.
4064 34. Failure to deliver a bounce message always caused it to be frozen, even if
4065 there was an errors_to setting on the router. The errors_to setting is now
4068 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4069 canonical form (fully expanded) before being placed in
4070 $sender_host_address.
4072 36. The table in the code that translates DNS record types into text (T_A to
4073 "A" for instance) was missing entries for NS and CNAME. It is just possible
4074 that this could have caused confusion if both these types were looked up
4075 for the same domain, because the text type is used as part of Exim's
4076 per-process caching. But the chance of anyone hitting this buglet seems
4079 37. The dnsdb lookup has been extended in a number of ways.
4081 (1) There is a new type, "zns", which walks up the domain tree until it
4082 finds some nameserver records. It should be used with care.
4084 (2) There is a new type, "mxh", which is like "mx" except that it returns
4085 just the host names, not the priorities.
4087 (3) It is now possible to give a list of domains (or IP addresses) to be
4088 looked up. The behaviour when one of the lookups defers can be
4089 controlled by a keyword.
4091 (4) It is now possible to specify the separator character for use when
4092 multiple records are returned.
4094 38. The dnslists ACL condition has been extended: it is now possible to supply
4095 a list of IP addresses and/or domains to be looked up in a particular DNS
4098 39. Added log_selector=+queue_time_overall.
4100 40. When running the queue in the test harness, wait just a tad after forking a
4101 delivery process, to get repeatability of debugging output.
4103 41. Include certificate and key file names in error message when GnuTLS fails
4104 to set them up, because the GnuTLS error message doesn't include the name
4105 of the failing file when there is a problem reading it.
4107 42. Allow both -bf and -bF in the same test run.
4109 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
4111 44. The "Exiscan patch" is now merged into the mainline Exim source.
4113 45. Sometimes the final signoff response after QUIT could fail to get
4114 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4115 < 0 before doing a fflush(). This bug looks as though it goes back to the
4116 introduction of TLS in release 3.20, but "sometimes" must have been rare
4117 because the tests only now provoked it.
4119 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
4120 (this can affect the format of dates).
4122 47. exim_tidydb, when checking for the continued existence of a message for
4123 which it has found a message-specific retry record, was not finding
4124 messages that were in split spool directories. Consequently, it was
4125 deleting retry records that should have stayed in existence.
4127 48. Steve fixed some bugs in eximstats.
4129 49. The SPA authentication driver was not abandoning authentication and moving
4130 on to the next authenticator when an expansion was forced to fail,
4131 contradicting the general specification for all authenticators. Instead it
4132 was generating a temporary error. It now behaves as specified.
4134 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
4135 (the order specifies the preference for clients). The order is now AES256,
4136 AES128, 3DES, ARCFOUR128.
4138 51. Small patch to Sieve code - explicitly set From: when generating an
4141 52. Exim crashed if a remote delivery caused a very long error message to be
4142 recorded - for instance if somebody sent an entire SpamAssassin report back
4143 as a large number of 550 error lines. This bug was coincidentally fixed by
4144 increasing the size of one of Exim's internal buffers (big_buffer) that
4145 happened as part of the Exiscan merge. However, to be on the safe side, I
4146 have made the code more robust (and fixed the comments that describe what
4149 53. Now that there can be additional text after "Completed" in log lines (if
4150 the queue_time_overall log selector is set), a one-byte patch to exigrep
4151 was needed to allow it to recognize "Completed" as not the last thing in
4154 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
4155 patch that reportedly fixes this has been added. I am not expert enough to
4156 create a test for it. This is what the patch creator wrote:
4158 "I found a little strange behaviour of ldap code when working with
4159 Windows 2003 AD Domain, where users was placed in more than one
4160 Organization Units. When I tried to give exim partial DN, the exit code
4161 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
4162 But simultaneously result of request was absolutely normal ldap result,
4163 so I produce this patch..."
4165 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
4166 REFERENCE, so I have modified the code to exclude the patch when that macro
4169 55. Some experimental protocols are using DNS PTR records for new purposes. The
4170 keys for these records are domain names, not reversed IP addresses. The
4171 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4172 leaves it alone. Component reversal etc. now happens only for IP addresses.
4175 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
4177 57. Double the size of the debug message buffer (to 2048) so that more of very
4178 long debug lines gets shown.
4180 58. The exicyclog utility now does better if the number of log files to keep
4181 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
4183 59. Two changes related to the smtp_active_hostname option:
4185 (1) $smtp_active_hostname is now available as a variable.
4186 (2) The default for smtp_banner uses $smtp_active_hostname instead
4187 of $primary_hostname.
4189 60. The host_aton() function is supposed to be passed a string that is known
4190 to be a valid IP address. However, in the case of IPv6 addresses, it was
4191 not checking this. This is a hostage to fortune. Exim now panics and dies
4192 if the condition is not met. A case was found where this could be provoked
4193 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4194 components; fortuitously, this particular loophole had already been fixed
4195 by change 4.50/55 above.
4197 If there are any other similar loopholes, the new check in host_aton()
4198 itself should stop them being exploited. The report I received stated that
4199 data on the command line could provoke the exploit when Exim was running as
4200 exim, but did not say which command line option was involved. All I could
4201 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4202 running as the user.
4205 61. There was a buffer overflow vulnerability in the SPA authentication code
4206 (which came originally from the Samba project). I have added a test to the
4207 spa_base64_to_bits() function which I hope fixes it.
4210 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
4211 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
4213 63. The daemon start-up calls getloadavg() while still root for those OS that
4214 need the first call to be done as root, but it missed one case: when
4215 deliver_queue_load_max is set with deliver_drop_privilege. This is
4216 necessary for the benefit of the queue runner, because there is no re-exec
4217 when deliver_drop_privilege is set.
4219 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
4220 This has been fixed.
4222 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4223 was in use, was not putting the data itself into the right store pool;
4224 consequently, it could be overwritten for a subsequent message in the same
4225 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4228 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
4230 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
4231 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
4232 the function were treating the return as a boolean value, which happened to
4233 work because 0=false and not-0=true, but is not correct code.
4235 68. The host_aton() function was not handling scoped IPv6 addresses (those
4236 with, for example, "%eth0" on the end) correctly.
4238 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
4239 negated items (that is, ~something) in unsigned ints. Some compilers
4240 apparently mutter when there is no cast.
4242 70. If an address verification called from an ACL failed, and did not produce a
4243 user-specific message (i.e. there was only a "system" message), nothing was
4244 put in $acl_verify_message. In this situation, it now puts the system
4247 71. Change 4.23/11 added synchronization checking at the start of an SMTP
4248 session; change 4.31/43 added the unwanted input to the log line - except
4249 that it did not do this in the start of session case. It now does.
4251 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
4252 This is wrong and can cause the other end to generate a synchronization
4253 error if it is another Exim or anything else that does the synchronization
4254 check. A QUIT command is no longer sent after a timeout.
4256 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
4257 during host lookups.
4259 74. The defer_ok option of callout verification was not working if it was used
4260 when verifying addresses in header lines, that is, for this case:
4262 verify = header_sender/callout=defer_ok
4264 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
4265 those file descriptors could be used for SMTP connections. If anything
4266 wrote to stderr (the example that came up was "warn" in embedded Perl), it
4267 could be sent to the SMTP client, causing chaos. The daemon now opens
4268 stdin, stdout, and stderr to /dev/null when it puts itself into the
4271 76. Arrange for output from Perl's "warn" command to be written to Exim's main
4272 log by default. The user can override this with suitable Perl magic.
4274 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
4275 the log message when discard triggers, was not working for the DATA ACL or
4276 for the non-SMTP ACL.
4278 78. Error message wording change in sieve.c.
4280 79. If smtp_accept_max_per_host was set, the number of connections could be
4281 restricted to fewer than expected, because the daemon was trying to set up
4282 a new connection before checking whether the processes handling previous
4283 connections had finished. The check for completed processes is now done
4284 earlier. On busy systems, this bug wouldn't be noticed because something
4285 else would have woken the daemon, and it would have reaped the completed
4288 80. If a message was submitted locally by a user whose login name contained one
4289 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
4290 It caused a spool format error. I have fixed the spool reading code. A
4291 related problem was that the "from" clause in the Received: line became
4292 illegal because of the space(s). It is now covered by ${quote_local_part.
4294 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
4297 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
4299 83. Patch from Sieve maintainer to fix unterminated string problem in
4300 "vacation" handling.
4302 84. Some minor changes to the Linux configuration files to help with other
4303 OS variants using glibc.
4305 85. One more patch for Sieve to update vacation handling to latest spec.
4308 ----------------------------------------------------
4309 See the note above about the 4.44 and 4.50 releases.
4310 ----------------------------------------------------
4316 1. Change 4.43/35 introduced a bug that caused file counts to be
4317 incorrectly computed when quota_filecount was set in an appendfile
4320 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4321 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4324 3. Give more explanation in the error message when the command for a transport
4325 filter fails to execute.
4327 4. There are several places where Exim runs a non-Exim command in a
4328 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4329 was being done only for the command run by the queryprogram router. It is
4330 now done for all such subprocesses. The other cases are: ${run, transport
4331 filters, and the commands run by the lmtp and pipe transports.
4333 5. Some older OS have a limit of 256 on the maximum number of file
4334 descriptors. Exim was using setrlimit() to set 1000 as a large value
4335 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4336 systems. I've change it so that if it can't get 1000, it tries for 256.
4338 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4339 was an oversight, and furthermore, ever since the addition of extra
4340 controls (e.g. 4.43/32), the checks on when to allow different forms of
4341 "control" were broken. There should now be diagnostics for all cases when a
4342 control that does not make sense is encountered.
4344 7. $recipients is now available in the predata ACL (oversight).
4346 8. Tidy the search cache before the fork to do a delivery from a message
4347 received from the command line. Otherwise the child will trigger a lookup
4348 failure and thereby defer the delivery if it tries to use (for example) a
4349 cached ldap connection that the parent has called unbind on.
4351 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4352 of $address_data from the recipient verification was clobbered by the
4353 sender verification.
4355 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4356 was its contents. (It was OK if the option was not defined at all.)
4358 11. A "Completed" log line is now written for messages that are removed from
4359 the spool by the -Mrm option.
4361 12. $host_address is now set to the target address during the checking of
4362 ignore_target_hosts.
4364 13. When checking ignore_target_hosts for an ipliteral router, no host name was
4365 being passed; this would have caused $sender_host_name to have been used if
4366 matching the list had actually called for a host name (not very likely,
4367 since this list is usually IP addresses). A host name is now passed as
4370 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4371 code that specifies a non-restarting handler (typically sigaction() in
4372 modern systems) in an attempt to fix a rare and obscure crash bug.
4374 15. Narrowed the window for a race in the daemon that could cause it to ignore
4375 SIGCHLD signals. This is not a major problem, because they are used only to
4376 wake it up if nothing else does.
4378 16. A malformed maildirsize file could cause Exim to calculate negative values
4379 for the mailbox size or file count. Odd effects could occur as a result.
4380 The maildirsize information is now recalculated if the size or filecount
4383 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4384 support for a long time. Removed HAVE_SYS_VFS_H.
4386 18. Updated exipick to current release from John Jetmore.
4388 19. Allow an empty sender to be matched against a lookup in an address list.
4389 Previously the only cases considered were a regular expression, or an
4392 20. Exim went into a mad DNS lookup loop when doing a callout where the
4393 host was specified on the transport, if the DNS lookup yielded more than
4396 21. The RFC2047 encoding function was originally intended for short strings
4397 such as real names; it was not keeping to the 75-character limit for
4398 encoded words that the RFC imposes. It now respects the limit, and
4399 generates multiple encoded words if necessary. To be on the safe side, I
4400 have increased the buffer size for the ${rfc2047: expansion operator from
4403 22. Failure to deliver a bounce message always caused it to be frozen, even if
4404 there was an errors_to setting on the router. The errors_to setting is now
4407 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4408 canonical form (fully expanded) before being placed in
4409 $sender_host_address.
4411 24. Updated eximstats to version 1.33
4413 25. Include certificate and key file names in error message when GnuTLS fails
4414 to set them up, because the GnuTLS error message doesn't include the name
4415 of the failing file when there is a problem reading it.
4417 26. Expand error message when OpenSSL has problems setting up cert/key files.
4420 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
4421 (this can affect the format of dates).
4423 28. exim_tidydb, when checking for the continued existence of a message for
4424 which it has found a message-specific retry record, was not finding
4425 messages that were in split spool directories. Consequently, it was
4426 deleting retry records that should have stayed in existence.
4428 29. eximstats updated to version 1.35
4429 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
4430 1.35 - bugfix such that pie charts by volume are generated correctly
4432 30. The SPA authentication driver was not abandoning authentication and moving
4433 on to the next authenticator when an expansion was forced to fail,
4434 contradicting the general specification for all authenticators. Instead it
4435 was generating a temporary error. It now behaves as specified.
4437 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
4438 (the order specifies the preference for clients). The order is now AES256,
4439 AES128, 3DES, ARCFOUR128.
4441 31. Small patch to Sieve code - explicitly set From: when generating an
4444 32. Exim crashed if a remote delivery caused a very long error message to be
4445 recorded - for instance if somebody sent an entire SpamAssassin report back
4446 as a large number of 550 error lines. This bug was coincidentally fixed by
4447 increasing the size of one of Exim's internal buffers (big_buffer) that
4448 happened as part of the Exiscan merge. However, to be on the safe side, I
4449 have made the code more robust (and fixed the comments that describe what
4452 33. Some experimental protocols are using DNS PTR records for new purposes. The
4453 keys for these records are domain names, not reversed IP addresses. The
4454 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4455 leaves it alone. Component reversal etc. now happens only for IP addresses.
4458 34. The host_aton() function is supposed to be passed a string that is known
4459 to be a valid IP address. However, in the case of IPv6 addresses, it was
4460 not checking this. This is a hostage to fortune. Exim now panics and dies
4461 if the condition is not met. A case was found where this could be provoked
4462 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4463 components; fortuitously, this particular loophole had already been fixed
4464 by change 4.50/55 or 4.44/33 above.
4466 If there are any other similar loopholes, the new check in host_aton()
4467 itself should stop them being exploited. The report I received stated that
4468 data on the command line could provoke the exploit when Exim was running as
4469 exim, but did not say which command line option was involved. All I could
4470 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4471 running as the user.
4474 35. There was a buffer overflow vulnerability in the SPA authentication code
4475 (which came originally from the Samba project). I have added a test to the
4476 spa_base64_to_bits() function which I hope fixes it.
4479 36. The daemon start-up calls getloadavg() while still root for those OS that
4480 need the first call to be done as root, but it missed one case: when
4481 deliver_queue_load_max is set with deliver_drop_privilege. This is
4482 necessary for the benefit of the queue runner, because there is no re-exec
4483 when deliver_drop_privilege is set.
4485 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4486 was in use, was not putting the data itself into the right store pool;
4487 consequently, it could be overwritten for a subsequent message in the same
4488 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4491 38. Sometimes the final signoff response after QUIT could fail to get
4492 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4493 < 0 before doing a fflush(). This bug looks as though it goes back to the
4494 introduction of TLS in release 3.20, but "sometimes" must have been rare
4495 because the tests only now provoked it.
4501 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
4502 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
4503 Now it returns an integer. A number of calls were still expecting a T/F
4504 return. Fortuitously, in all cases, the tests worked in OK situations,
4505 which is the norm. However, things would have gone wrong on any write
4506 failures on the smtp file descriptor. This function is used when sending
4507 messages over SMTP and also when doing verify callouts.
4509 2. When Exim is called to do synchronous delivery of a locally submitted
4510 message (the -odf or -odi options), it no longer closes stderr before doing
4513 3. Implemented the mua_wrapper option.
4515 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
4517 5. Implemented the functions header_remove(), header_testname(),
4518 header_add_at_position(), and receive_remove_recipient(), and exported them
4521 6. If an ACL "warn" statement specified the addition of headers, Exim already
4522 inserted X-ACL-Warn: at the start if there was no header name. However, it
4523 was not making this test for the second and subsequent header lines if
4524 there were newlines in the string. This meant that an invalid header could
4525 be inserted if Exim was badly configured.
4527 7. Allow an ACL "warn" statement to add header lines at the start or after all
4528 the Received: headers, as well as at the end.
4530 8. Added the rcpt_4xx retry error code.
4532 9. Added postmaster_mailfrom=xxx to callout verification option.
4534 10. Added mailfrom=xxxx to the callout verification option, for verify=
4537 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
4538 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
4540 12. Inserted some casts to stop certain compilers warning when using pointer
4541 differences as field lengths or precisions in printf-type calls (mostly
4542 affecting debugging statements).
4544 13. Added optional readline() support for -be (dynamically loaded).
4546 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
4547 same clock tick as a message's arrival, so that its received time was the
4548 same as the "first fail" time on the retry record, and that message
4549 remained on the queue past the ultimate address timeout, every queue runner
4550 would try a delivery (because it was past the ultimate address timeout) but
4551 after another failure, the ultimate address timeout, which should have then
4552 bounced the address, did not kick in. This was a "< instead of <=" error;
4553 in most cases the first failure would have been in the next clock tick
4554 after the received time, and all would be well.
4556 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
4557 being recognized when the domain list was tested by the match_domain
4558 condition in an expansion string.
4560 16. Added the ${str2b64: operator.
4562 17. Exim was always calling setrlimit() to set a large limit for the number of
4563 processes, without checking whether the existing limit was already
4564 adequate. (It did check for the limit on file descriptors.) Furthermore,
4565 errors from getrlimit() and setrlimit() were being ignored. Now they are
4566 logged to the main and panic logs, but Exim does carry on, to try to do its
4567 job under whatever limits there are.
4569 18. Imported PCRE 5.0.
4571 19. Trivial typo in log message " temporarily refused connection" (the leading
4574 20. If the log selector return_path_on_delivery was set and an address was
4575 redirected to /dev/null, the delivery process crashed because it assumed
4576 that a return path would always be set for a "successful" delivery. In this
4577 case, the whole delivery is bypassed as an optimization, and therefore no
4580 21. Internal re-arrangement: the function for sending a challenge and reading
4581 a response while authentication was assuming a zero-terminated challenge
4582 string. It's now changed to take a pointer and a length, to allow for
4583 binary data in such strings.
4585 22. Added the cyrus_sasl authenticator (code supplied by MBM).
4587 23. Exim was not respecting finduser_retries when seeking the login of the
4588 uid under which it was called; it was always trying 10 times. (The default
4589 setting of finduser_retries is zero.) Also, it was sleeping after the final
4590 failure, which is pointless.
4592 24. Implemented tls_on_connect_ports.
4594 25. Implemented acl_smtp_predata.
4596 26. If the domain in control=submission is set empty, Exim assumes that the
4597 authenticated id is a complete email address when it generates From: or
4598 Sender: header lines.
4600 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
4601 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
4602 chown and chgrp in /bin and hostname in /usr/bin.
4604 28. Exim was keeping the "process log" file open after each use, just as it
4605 does for the main log. This opens the possibility of it remaining open for
4606 long periods when the USR1 signal hits a daemon. Occasional processlog
4607 errors were reported, that could have been caused by this. Anyway, it seems
4608 much more sensible not to leave this file open at all, so that is what now
4611 29. The long-running daemon process does not normally write to the log once it
4612 has entered its main loop, and it closes the log before doing so. This is
4613 so that log files can straightforwardly be renamed and moved. However,
4614 there are a couple of unusual error situations where the daemon does write
4615 log entries, and I had neglected to close the log afterwards.
4617 30. The text of an SMTP error response that was received during a remote
4618 delivery was being truncated at 512 bytes. This is too short for some of
4619 the long messages that one sometimes sees. I've increased the limit to
4622 31. It is now possible to make retry rules that apply only when a message has a
4623 specific sender, in particular, an empty sender.
4625 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
4626 it possible to be selective about when SMTP synchronization is enforced.
4628 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
4630 32. Implemented hosts_connection_nolog.
4632 33. Added an ACL for QUIT.
4634 34. Setting "delay_warning=" to disable warnings was not working; it gave a
4637 35. Added mailbox_size and mailbox_filecount to appendfile.
4639 36. Added control = no_multiline_responses to ACLs.
4641 37. There was a bug in the logic of the code that waits for the clock to tick
4642 in the case where the clock went backwards by a substantial amount such
4643 that the microsecond fraction of "now" was more than the microsecond
4644 fraction of "then" (but the whole seconds number was less).
4646 38. Added support for the libradius Radius client library this is found on
4647 FreeBSD (previously only the radiusclient library was supported).
4653 1. When certain lookups returned multiple values in the form name=value, the
4654 quoting of the values was not always being done properly. Specifically:
4655 (a) If the value started with a double quote, but contained no whitespace,
4657 (b) If the value contained whitespace other than a space character (i.e.
4658 tabs or newlines or carriage returns) it was not quoted.
4659 This fix has been applied to the mysql and pgsql lookups by writing a
4660 separate quoting function and calling it from the lookup code. The fix
4661 should probably also be applied to nisplus, ibase and oracle lookups, but
4662 since I cannot test any of those, I have not disturbed their existing code.
4664 2. A hit in the callout cache for a specific address caused a log line with no
4665 reason for rejecting RCPT. Now it says "Previous (cached) callout
4666 verification failure".
4668 3. There was an off-by-one bug in the queryprogram router. An over-long
4669 return line was truncated at 256 instead of 255 characters, thereby
4670 overflowing its buffer with the terminating zero. As well as fixing this, I
4671 have increased the buffer size to 1024 (and made a note to document this).
4673 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
4674 when Exim is waiting for an SMTP response from a remote server, Exim
4675 restarts its select() call on the socket, thereby resetting its timeout.
4676 This is not a problem when such interrupts are rare. Somebody set up a cron
4677 job to run exiwhat every 2 minutes, which is less than the normal select()
4678 timeout (5 or 10 minutes). This meant that the select() timeout never
4679 kicked in because it was always reset. I have fixed this by comparing the
4680 time when an interrupt arrives with the time at the start of the first call
4681 to select(). If more time than the timeout has elapsed, the interrupt is
4682 treated as a timeout.
4684 5. Some internal re-factoring in preparation for the addition of Sieve
4685 extensions (by MH). In particular, the "personal" test is moved to a
4686 separate function, and given an option for scanning Cc: and Bcc: (which is
4687 not set for Exim filters).
4689 6. When Exim created an email address using the login of the caller as the
4690 local part (e.g. when creating a From: or Sender: header line), it was not
4691 quoting the local part when it contained special characters such as @.
4693 7. Installed new OpenBSD configuration files.
4695 8. Reworded some messages for syntax errors in "and" and "or" conditions to
4696 try to make them clearer.
4698 9. Callout options, other than the timeout value, were being ignored when
4699 verifying sender addresses in header lines. For example, when using
4701 verify = header_sender/callout=no_cache
4703 the cache was (incorrectly) being used.
4705 10. Added a missing instance of ${EXE} to the exim_install script; this affects
4706 only the Cygwin environment.
4708 11. When return_path_on_delivery was set as a log selector, if different remote
4709 addresses in the same message used different return paths and parallel
4710 remote delivery occurred, the wrong values would sometimes be logged.
4711 (Whenever a remote delivery process finished, the return path value from
4712 the most recently started remote delivery process was logged.)
4714 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
4715 lines when AUTH and/or TLS are in use. This is the "received protocol"
4716 field. Exim used to use "asmtp" for authenticated SMTP, without any
4717 indication (in the protocol name) for TLS use. Now it follows the RFC and
4718 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
4719 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
4720 names appear in log lines as well as in Received: header lines.
4722 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
4723 extensions, and comparison tests, and to fix some bugs.
4725 14. Changes to the "personal" filter test:
4727 (1) The test was buggy in that it was just doing the equivalent of
4728 "contains" tests on header lines. For example, if a user's address was
4729 anne@some.where, the "personal" test would incorrectly be true for
4731 To: susanne@some.where
4733 This test is now done by extracting each address from the header in turn,
4734 and checking the entire address. Other tests that are part of "personal"
4735 are now done using regular expressions (for example, to check local parts
4736 of addresses in From: header lines).
4738 (2) The list of non-personal local parts in From: addresses has been
4739 extended to include "listserv", "majordomo", "*-request", and "owner-*",
4740 taken from the Sieve specification recommendations.
4742 (3) If the message contains any header line starting with "List-" it is
4743 treated as non-personal.
4745 (4) The test for "circular" in the Subject: header line has been removed
4746 because it now seems ill-conceived.
4748 15. Minor typos in src/EDITME comments corrected.
4750 16. Installed latest exipick from John Jetmore.
4752 17. If headers_add on a router specified a text string that was too long for
4753 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
4754 of string_sprintf() is now avoided.
4756 18. $message_body_size was not set (it was always zero) when running the DATA
4757 ACL and the local_scan() function.
4759 19. For the "mail" command in an Exim filter, no default was being set for
4760 the once_repeat time, causing a random time value to be used if "once" was
4761 specified. (If the value happened to be <= 0, no repeat happened.) The
4762 default is now 0s, meaning "never repeat". The "vacation" command was OK
4763 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
4764 (I found it when inspecting the code).
4766 20. There is now an overall timeout for performing a callout verification. It
4767 defaults to 4 times the callout timeout, which applies to individual SMTP
4768 commands during the callout. The overall timeout applies when there is more
4769 than one host that can be tried. The timeout is checked before trying the
4770 next host. This prevents very long delays if there are a large number of
4771 hosts and all are timing out (e.g. when the network connections are timing
4772 out). The value of the overall timeout can be changed by specifying an
4773 additional sub-option for "callout", called "maxwait". For example:
4775 verify = sender/callout=5s,maxwait=20s
4777 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
4778 to the end before writing, but this should make it even safer).
4780 22. Exim was forgetting that it had advertised PIPELINING for the second and
4781 subsequent messages on an SMTP connection. It was also not resetting its
4782 memory on STARTTLS and an internal HELO.
4784 23. When Exim logs an SMTP synchronization error within a session, it now
4785 records whether PIPELINING has been advertised or not.
4787 24. Added 3 instances of "(long int)" casts to time_t variables that were being
4788 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
4789 rather than long int.
4791 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
4793 26. Added the never_mail option to autoreply.
4799 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
4800 crash if the getsockname() call failed; this can happen if a connection is
4801 closed very soon after it is established. The problem was simply in the
4802 order in which certain operations were done, causing Exim to try to write
4803 to the SMTP stream before it had set up the file descriptor. The bug has
4804 been fixed by making things happen in the correct order.
4810 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
4811 before the connection was closed, thus losing the rejection response.
4813 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
4814 some early Solaris releases, but causes trouble in current releases where
4815 socklen_t is defined.
4817 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
4820 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
4823 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
4824 that releasing the top of it at the end releases what was used for sub-
4825 expansions (unless the block got too big). However, discard this block if
4826 the first thing is a variable or header, so that we can use its block when
4827 it is dynamic (useful for very large $message_headers, for example).
4829 6. Lookups now cache *every* query, not just the most recent. A new, separate
4830 store pool is used for this. It can be recovered when all lookup caches are
4831 flushed. Lookups now release memory at the end of their result strings.
4832 This has involved some general refactoring of the lookup sources.
4834 7. Some code has been added to the store_xxx() functions to reduce the amount
4835 of flapping under certain conditions.
4837 8. log_incoming_interface used to affect only the <= reception log lines. Now
4838 it causes the local interface and port to be added to several more SMTP log
4839 lines, for example "SMTP connection from", and rejection lines.
4841 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
4843 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
4845 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
4846 could be overwritten at the end of the current message (or the start of a
4847 new message if it was set in a HELO ACL). The value is now preserved for
4848 the duration of the SMTP connection.
4850 12. If a transport had a headers_rewrite setting, and a matching header line
4851 contained an unqualified address, that address was qualified, even if it
4852 did not match any rewriting rules. The underlying bug was that the values
4853 of the flags that permit the existence of unqualified sender and recipient
4854 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
4855 non-local messages, and by -bnq for local messages) were not being
4856 preserved with the message after it was received.
4858 13. When Exim was logging an SMTP synchronization error, it could sometimes log
4859 "next input=" as part of the text comprising the host identity instead of
4860 the correct text. The code was using the same buffer for two different
4861 strings. However, depending on which order the printing function evaluated
4862 its arguments, the bug did not always show up. Under Linux, for example, my
4863 test suite worked just fine.
4865 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
4866 doesn't work with some older versions of Perl. It has been changed to "my",
4867 which in any case is probably the better facility to use.
4869 15. A really picky compiler found some instances of statements for creating
4870 error messages that either had too many or two few arguments for the format
4873 16. The size of the buffer for calls to the DNS resolver has been increased
4874 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
4875 for addresses that have a lot of PTR records. This alleviates a problem; it
4876 does not fully solve it.
4878 17. A dnsdb lookup for PTR records that receives more data than will fit in the
4879 buffer now truncates the list and logs the incident, which is the same
4880 action as happens when Exim is looking up a host name and its aliases.
4881 Previously in this situation something unpredictable would happen;
4882 sometimes it was "internal error: store_reset failed".
4884 18. If a server dropped the connection unexpectedly when an Exim client was
4885 using GnuTLS and trying to read a response, the client delivery process
4886 crashed while trying to generate an error log message.
4888 19. If a "warn" verb in an ACL added multiple headers to a message in a single
4889 string, for example:
4891 warn message = H1: something\nH2: something
4893 the text was added as a single header line from Exim's point of view
4894 though it ended up OK in the delivered message. However, searching for the
4895 second and subsequent header lines using $h_h2: did not work. This has been
4896 fixed. Similarly, if a system filter added multiple headers in this way,
4897 the routers could not see them.
4899 20. Expanded the error message when iplsearch is called with an invalid key to
4900 suggest using net-iplsearch in a host list.
4902 21. When running tests using -bh, any delays imposed by "delay" modifiers in
4903 ACLs are no longer actually imposed (and a message to that effect is
4906 22. If a "gecos" field in a passwd entry contained escaped characters, in
4907 particular, if it contained a \" sequence, Exim got it wrong when building
4908 a From: or a Sender: header from that name. A second bug also caused
4909 incorrect handling when an unquoted " was present following a character
4910 that needed quoting.
4912 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
4913 was not being matched caselessly.
4915 24. Arranged for all hyphens in the exim.8 source to be escaped with
4918 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
4919 properly. Recipient callout cache records were still being keyed to include
4920 the sender, even when use_sender was set false. This led to far more
4921 callouts that were necessary. The sender is no longer included in the key
4922 when use_sender is false.
4924 26. Added "control = submission" modifier to ACLs.
4926 27. Added the ${base62d: operator to decode base 62 numbers.
4928 28. dnsdb lookups can now access SRV records.
4930 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
4931 the configuration file.
4933 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
4934 (-v) mode. This makes the output for a verbose queue run more intelligible.
4936 31. Added a use_postmaster feature to recipient callouts.
4938 32. Added the $body_zerocount variable, containing the number of binary zero
4939 bytes in the message body.
4941 33. The time of last modification of the "new" subdirectory is now used as the
4942 "mailbox time last read" when there is a quota error for a maildir
4945 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
4947 35. Added +ignore_unknown as a special item in host lists.
4949 36. Code for decoding IPv6 addresses in host lists is now included, even if
4950 IPv6 support is not being compiled. This fixes a bug in which an IPv6
4951 address was recognized as an IP address, but was then not correctly decoded
4952 into binary, causing unexpected and incorrect effects when compared with
4959 1. Very minor rewording of debugging text in manualroute to say "list of
4960 hosts" instead of "hostlist".
4962 2. If verify=header_syntax was set, and a header line with an unqualified
4963 address (no domain) and a large number of spaces between the end of the
4964 name and the colon was received, the reception process suffered a buffer
4965 overflow, and (when I tested it) crashed. This was caused by some obsolete
4966 code that should have been removed. The fix is to remove it!
4968 3. When running in the test harness, delay a bit after writing a bounce
4969 message to get a bit more predictability in the log output.
4971 4. Added a call to search_tidyup() just before forking a reception process. In
4972 theory, someone could use a lookup in the expansion of smtp_accept_max_
4973 per_host which, without the tidyup, could leave open a database connection.
4975 5. Added the variables $recipient_data and $sender_data which get set from a
4976 lookup success in an ACL "recipients" or "senders" condition, or a router
4977 "senders" option, similar to $domain_data and $local_part_data.
4979 6. Moved the writing of debug_print from before to after the "senders" test
4982 7. Change 4.31/66 (moving the time when the Received: is generated) caused
4983 problems for message scanning, either using a data ACL, or using
4984 local_scan() because the Received: header was not generated till after they
4985 were called (in order to set the time as the time of reception completion).
4986 I have revised the way this works. The header is now generated after the
4987 body is received, but before the ACL or local_scan() are called. After they
4988 are run, the timestamp in the header is updated.
4994 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
4995 before starting a queue runner without re-exec. This happened only when
4996 deliver_drop_privilege was set or when the Exim user was set to root. The
4997 effect of the bug was that timeouts during subsequent deliveries caused
4998 crashes instead of being properly handled. The handler is now left at its
4999 default (and expected) setting.
5001 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
5002 message, again when deliver_drop_privilege is set or Exim is run as root.
5003 The bug described in (1) was not present in this case, but the tidying up
5004 of the other signals was missing. I have made the two cases consistent.
5006 3. The ignore_target_hosts setting on a manualroute router was being ignored
5007 for hosts that were looked up using the /MX notation.
5009 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
5012 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
5013 operated on the sender address. After changing the $sender_address to <>
5014 for the sender address verify, Exim was re-instated it as the original
5015 (before rewriting) address, but remembering that it had rewritten it, so it
5016 wasn't rewriting it again. This bug also had the effect of breaking the
5017 sender address verification caching when the sender address was rewritten.
5019 6. The ignore_target_hosts option was being ignored by the ipliteral router.
5020 This has been changed so that if the ip literal address matches
5021 ignore_target_hosts, the router declines.
5023 7. Added expansion conditions match_domain, match_address, and match_local_
5024 part (NOT match_host).
5026 8. The placeholder for the Received: header didn't have a length field set.
5028 9. Added code to Exim itself and to exim_lock to test for a specific race
5029 condition that could lead to file corruption when using MBX delivery. The
5030 issue is with the lockfile that is created in /tmp. If this file is removed
5031 after a process has opened it but before that process has acquired a lock,
5032 there is the potential for a second process to recreate the file and also
5033 acquire a lock. This could lead to two Exim processes writing to the file
5034 at the same time. The added code performs the same test as UW imapd; it
5035 checks after acquiring the lock that its file descriptor still refers to
5036 the same named file.
5038 10. The buffer for building added header lines was of fixed size, 8192 bytes.
5039 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
5042 11. Added the smtp_active_hostname option. If used, this will typically be made
5043 to depend on the incoming interface address. Because $interface_address is
5044 not set up until the daemon has forked a reception process, error responses
5045 that can happen earlier (such as "too many connections") no longer contain
5048 12. If an expansion in a condition on a "warn" statement fails because a lookup
5049 defers, the "warn" statement is abandoned, and the next ACL statement is
5050 processed. Previously this caused the whole ACL to be aborted.
5052 13. Added the iplsearch lookup type.
5054 14. Added ident_timeout as a log selector.
5056 15. Added tls_certificate_verified as a log selector.
5058 16. Added a global option tls_require_ciphers (compare the smtp transport
5059 option of the same name). This controls incoming TLS connections.
5061 17. I finally figured out how to make tls_require_ciphers do a similar thing
5062 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
5063 before starting the TLS session.
5065 18. Tabs are now shown as \t in -bP output.
5067 19. If the log selector return_path_on_delivery was set, Exim crashed when
5068 bouncing a message because it had too many Received: header lines.
5070 20. If two routers both had headers_remove settings, and the first one included
5071 a superfluous trailing colon, the final name in the first list and the
5072 first name in the second list were incorrectly joined into one item (with a
5073 colon in the middle).
5079 1. Added -C and -D options to the exinext utility, mainly to make it easier
5080 to include in the automated testing, but these could be helpful when
5081 multiple configurations are in use.
5083 2. The exinext utility was not formatting the output nicely when there was
5084 an alternate port involved in the retry record key, nor when there was a
5085 message id as well (for retries that were specific to a specific message
5086 and a specific host). It was also confused by IPv6 addresses, because of
5087 the additional colons they contain. I have fixed the IPv4 problem, and
5088 patched it up to do a reasonable job for IPv6.
5090 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
5091 delivery, the log line now contains "pipelined" if PIPELINING was used.
5093 4. An SMTP transport process used to panic and die if the bind() call to set
5094 an explicit outgoing interface failed. This has been changed; it is now
5095 treated in the same way as a connect() failure.
5097 5. A reference to $sender_host_name in the part of a conditional expansion
5098 that was being skipped was still causing a DNS lookup. This no longer
5101 6. The def: expansion condition was not recognizing references to header lines
5102 that used bh_ and bheader_.
5104 7. Added the _cache feature to named lists.
5106 8. The code for checking quota_filecount in the appendfile transport was
5107 allowing one more file than it should have been.
5109 9. For compatibility with Sendmail, the command line option
5117 and sets the incoming protocol and host name (for trusted callers). The
5118 host name and its colon can be omitted when only the protocol is to be set.
5119 Note the Exim already has two private options, -pd and -ps, that refer to
5120 embedded Perl. It is therefore impossible to set a protocol value of "d" or
5121 "s", but I don't think that's a major issue.
5123 10. A number of refactoring changes to the code, none of which should affect
5126 (a) The number of logging options was getting close to filling up the
5127 32-bit word that was used as a bit map. I have split them into two classes:
5128 those that are passed in the argument to log_write(), and those that are
5129 only ever tested independently outside of that function. These are now in
5130 separate 32-bit words, so there is plenty of room for expansion again.
5131 There is no change in the user interface or the logging behaviour.
5133 (b) When building, for example, log lines, the code previously used a
5134 macro that called string_cat() twice, in order to add two strings. This is
5135 not really sufficiently general. Furthermore, there was one instance where
5136 it was actually wrong because one of the argument was used twice, and in
5137 one call a function was used. (As it happened, calling the function twice
5138 did not affect the overall behaviour.) The macro has been replaced by a
5139 function that can join an arbitrary number of extra strings onto a growing
5142 (c) The code for expansion conditions now uses a table and a binary chop
5143 instead of a serial search (which was left over from when there were very
5144 few conditions). Also, it now recognizes conditions like "pam" even when
5145 the relevant support is not compiled in: a suitably worded error message is
5146 given if an attempt is made to use such a condition.
5148 11. Added ${time_interval:xxxxx}.
5150 12. A bug was causing one of the ddress fields not to be passed back correctly
5151 from remote delivery subprocesses. The field in question was not being
5152 subsequently used, so this caused to problems in practice.
5154 13. Added new log selectors queue_time and deliver_time.
5156 14. Might have fixed a bug in maildirsizefile handling that threw up
5157 "unexpected character" debug warnings, and recalculated the data
5158 unnecessarily. In any case, I expanded the warning message to give more
5161 15. Added the message "Restricted characters in address" to the statements in
5162 the default ACL that block characters like @ and % in local parts.
5164 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
5165 Three changes have been made:
5167 (a) There was a serious bug; a negative response to MAIL caused the whole
5168 recipient domain to be cached as invalid, thereby blocking all messages
5169 to all local parts at the same domain, from all senders. This bug has
5170 been fixed. The domain is no longer cached after a negative response to
5171 MAIL if the sender used is not empty.
5173 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
5176 (c) A new callout option, "use_sender" has been added for people who want
5177 the modified behaviour.
5183 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
5186 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
5187 indeed breaks things for older releases.
5189 3. Added additional logging to the case where there is a problem reading data
5190 from a filter that is running in a subprocess using a pipe, in order to
5191 try to track down a specific problem.
5193 4. Testing facility fudge: when running in the test harness and attempting
5194 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
5195 getting "No route to host". Convert this to a timeout.
5197 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
5200 6. Some OS don't have socklen_t but use size_t instead. This affects the
5201 fifth argument of getsockopt() amongst other things. This is now
5202 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
5203 can be set for individual OS. I have set it for SunOS5, OSF1, and
5204 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
5205 some earlier ones do not.
5207 7. Change 4.30/15 was not doing the test caselessly.
5209 8. The standard form for an IPv6 address literal was being rejected by address
5210 parsing in, for example, MAIL and RCPT commands. An example of this kind of
5211 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5212 this, as well as the form without the "IPv6" on the front (but only when
5213 address literals are enabled, of course).
5215 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
5217 10. Exim crashed if a message with an empty sender address specified by -f
5218 encountered a router with an errors_to setting. This could be provoked only
5219 by a command such as
5223 where an empty string was supplied; "<>" did not hit this bug.
5225 11. Installed PCRE release 4.5.
5227 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
5228 remained set. It is now erased.
5230 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
5231 times from message ids (which are base 36 rather than the normal 62).
5233 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
5234 were being counted as actual protocol errors, and logged if the log
5235 selector +smtp_protocol_error was set. One cannot be perfect in this test,
5236 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
5237 and DATA following a set of rejected RCPTs do not count as protocol errors.
5238 In other words, Exim assumes they were pipelined, though this may not
5239 actually be the case. Of course, in all cases the client gets an
5240 appropriate error code.
5242 15. If a lookup fails in an ACL condition, a message about the failure may
5243 be available; it is used if testing the ACL cannot continue, because most
5244 such messages specify what the cause of the deferral is. However, some
5245 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
5246 that caused an old message to be retained and used if a later statement
5247 caused a defer, replacing the real cause of the deferral.
5249 16. If an IP address had so many PTR records that the DNS lookup buffer
5250 was not large enough to hold them, Exim could crash while trying to process
5251 the truncated data. It now detects and logs this case.
5253 17. Further to 4.21/58, another change has been made: if (and only if) the
5254 first line of a message (the first header line) ends with CRLF, a bare LF
5255 in a subsequent header line has a space inserted after it, so as not to
5256 terminate the header.
5258 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
5259 unnecessarily, used atoi() instead of strtol(), and didn't check the
5260 termination when getting file sizes from file names by regex.
5262 19. Completely re-implemented the support for maildirsize files, in the light
5263 of a number of problems with the previous contributed implementation
5264 (4.30/29). In particular:
5266 . If the quota is zero, the maildirsize file is maintained, but no quota is
5269 . If the maildir directory does not exist, it is created before any attempt
5270 to write a maildirsize file.
5272 . The quota value in the file is just a cache; if the quota is changed in
5273 the transport, the new value overrides.
5275 . A regular expression is available for excluding directories from the
5278 20. The autoreply transport checks the characters in options that define the
5279 message's headers; it allows continued headers, but it was checking with
5280 isspace() after an embedded newline instead of explicitly looking for a
5283 21. If all the "regular" hosts to which an address was routed had passed their
5284 expiry times, and had not reached their retry times, the address was
5285 bounced, even if fallback hosts were defined. Now Exim should go on to try
5288 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
5289 equivalent code in the SMTP transport. Some hosts send humungous responses
5290 to HELO/EHLO, more than 1024 it seems.
5292 23. Refactoring: code in filter.c used (void *) for "any old type" but this
5293 gives compiler warnings in some environments. I've now done it "properly",
5296 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
5297 (because of problems with the built-in one) was declared to return uschar *
5298 instead of char *, causing compiler failure.
5300 25. Fixed a file descriptor leak when processing alias/forward files.
5302 26. Fixed a minor format string issue in dbfn.c.
5304 27. Typo in exim.c: ("dmbnz" for "dbmnz").
5306 28. If a filter file refered to $h_xxx or $message_headers, and the headers
5307 contained RFC 2047 "words", Exim's memory could, under certain conditions,
5310 29. When a sender address is verified, it is cached, to save repeating the test
5311 when there is more than one recipient in a message. However, when the
5312 verification involves a callout, it is possible for different callout
5313 options to be set for different recipients. It is too complicated to keep
5314 track of this in the cache, so now Exim always runs a verification when a
5315 callout is required, relying on the callout cache for the optimization.
5316 The overhead is duplication of the address routing, but this should not be
5319 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
5320 to be verified with callout=postmaster, and the main callout worked but the
5321 postmaster check failed, the verification correctly failed. However, if a
5322 subsequent RCPT command asked for sender verification *without* the
5323 postmaster check, incorrect caching caused this verification also to fail,
5326 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
5327 it was not caching the DNS options (qualify_single, search_parents) that
5328 were used when the lookup failed. A subsequent lookup with different
5329 options therefore always gave the same answer, though there were cases
5330 where it should not have. (Example: a "domains = !$mx_any" option on a
5331 dnslookup router: the "domains" option is always processed without any
5332 widening, but the router might have qualify_single set.) Now Exim uses the
5333 cached value only when the same options are set.
5335 32. Added John Jetmore's "exipick" utility to the distribution.
5337 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
5338 than a timeout (e.g. a certificate is required, and is not provided), an
5339 Exim server now closes the connection immediately. Previously it waited for
5340 the client to close - but if the client is SSL, it seems that they each
5341 wait for each other, leading to a delay before one of them times out.
5343 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
5344 maintained 0.8.x compatibility because I don't think many are using it, and
5345 it is clearly obsolete.
5347 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
5350 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
5351 client certificate was expired. A simple patch fixes this, though I don't
5352 understand the full logic of why the verify callback is called multiple
5355 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
5356 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
5357 which causes problems with some clients (such as the Certicom SSL Plus
5358 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
5359 disables the coutermeasure allowing Eudora to connect."
5361 38. Exim was not checking that a write() to a log file succeeded. This could
5362 lead to Bad Things if a log got too big, in particular if it hit a file
5363 size limit. Exim now panics and dies if it cannot write to a log file, just
5364 as it does if it cannot open a log file.
5366 39. Modified OS/Makefile-Linux so that it now contains
5368 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
5370 The two -D definitions ensure that Exim is compiled with large file
5371 support, which makes it possible to handle log files that are bigger than
5374 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
5375 instance) a domain was checked against a named list that involved a lookup,
5376 causing $domain_data to be set, then another domain was checked against the
5377 same list, then the first domain was re-checked, the value of $domain_data
5378 after the final check could be wrong. In particular, if the second check
5379 failed, it could be set empty. This bug probably also applied to
5382 41. The strip_trailing_dot option was not being applied to the address given
5383 with the -f command-line option.
5385 42. The code for reading a message's header from the spool was incrementing
5386 $received_count, but never initializing it. This meant that the value was
5387 incorrect (doubled) while delivering a message in the same process in which
5388 it was received. In the most common configuration of Exim, this never
5389 happens - a fresh exec is done - but it can happen when
5390 deliver_drop_privilege is set.
5392 43. When Exim logs an SMTP synchronization error - client data sent too soon -
5393 it now includes up to 150 characters of the unexpected data in the log
5396 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
5397 and building data strings. The size of both of these buffers was 10 000
5398 bytes - far larger than anybody would *ever* want, thought I. Needless to
5399 say, somebody hit the limit. I have increased the maximum line length to
5400 20 000 and the maximum data length of concatenated lines to 100 000. I have
5401 also fixed two bugs, because there was no checking on these buffers. Tsk,
5402 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
5403 buffer is too small.
5405 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
5406 lsearch lookups. Now it does.
5408 46. When parsing a route_list item in a manualroute router, a fixed-length
5409 buffer was used for the list of hosts. I made this 1024 bytes long,
5410 thinking that nobody would ever have a list of hosts that long. Wrong.
5411 Somebody had a whole pile of complicated expansion conditions, and the
5412 string was silently truncated, leading to an expansion error. It turns out
5413 that it is easier to change to an unlimited length (owing to other changes
5414 that have happened since this code was originally written) than to build
5415 structure for giving a limitation error. The length of the item that
5416 expands into the list of hosts is now unlimited.
5418 47. The lsearch lookup could not handle data where the length of text line was
5419 more than 4095 characters. Such lines were truncated, leading to shortened
5420 data being returned. It should now handle lines of any length.
5422 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
5423 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
5426 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
5427 replacing strings such as BIN_DIRECTORY in a source file: the replacement
5428 no longer happens in comment lines. A list of replacements is now placed
5429 at the head of all of the source files, except those whose only change is
5430 to replace PERL_COMMAND in the very first #! line.
5432 50. Replaced the slow insertion sort in queue.c, for sorting the list of
5433 messages on the queue, with a bottom-up merge sort, using code contributed
5434 by Michael Haardt. This should make operations like -bp somewhat faster on
5435 large queues. It won't affect queue runners, except when queue_run_in_order
5438 51. Installed eximstats 1.31 in the distribution.
5440 52. Added support for SRV lookups to the dnslookup router.
5442 53. If an ACL referred to $message_body or $message_body_end, the value was not
5443 reset for any messages that followed in the same SMTP session.
5445 54. The store-handling optimization for building very long strings was not
5446 differentiating between the different store pools. I don't think this
5447 actually made any difference in practice, but I've tidied it.
5449 55. While running the routers to verify a sender address, $sender_address
5450 was still set to the sender address. This is wrong, because when routing to
5451 send a bounce to the sender, it would be empty. Therefore, I have changed
5452 it so that, while verifying a sender address, $sender_address is set to <>.
5453 (There is no change to what happens when verifying a recipient address.)
5455 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
5456 target A or AAAA records (if not already returned) without resetting the
5457 qualify_single or search_parents options of the DNS resolver. These are
5458 inappropriate in this case because the targets of MX and SRV records must
5459 be FQDNs. A broken DNS record could cause trouble if it happened to have a
5460 target that, when qualified, matched something in the local domain. These
5461 two options are now turned off when doing these lookups.
5463 57. It seems that at least some releases of Reiserfs (which does not have the
5464 concept of a fixed number of inodes) returns zero and not -1 for the
5465 number of available inodes. This interacted badly with check_spool_inodes,
5466 which assumed that -1 was the "no such thing" setting. What I have done is
5467 to check that the total number of inodes is greater than zero before doing
5468 the test of how many are available.
5470 58. When a "warn" ACL statement has a log_message modifier, the message is
5471 remembered, and not repeated. This is to avoid a lot of repetition when a
5472 message has many recipients that cause the same warning to be written.
5473 Howewer, Exim was preserving the list of already written lines for an
5474 entire SMTP session, which doesn't seem right. The memory is now reset if a
5475 new message is started.
5477 59. The "rewrite" debugging flag was not showing the result of rewriting in the
5478 debugging output unless log_rewrite was also set.
5480 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
5481 of (int)(handle) when we know that handle contains (void *)(-1).
5483 61. The Exim daemon panic-logs an error return when it closes the incoming
5484 connection. However "connection reset by peer" seems to be common, and
5485 isn't really an error worthy of noting specially, so that particular error
5488 62. When Exim is trying to find all the local interfaces, it used to panic and
5489 die if the ioctl to get the interface flags failed. However, it seems that
5490 on at least one OS (Solaris 9) it is possible to have an interface that is
5491 included in the list of interfaces, but for which you get a failure error
5492 for this call. This happens when the interface is not "plumbed" into a
5493 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
5494 failure of the "get flags" call assumes that the interface is down.
5496 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
5497 makes life easier for people who are doing arithmetic on fields extracted
5498 from dates, where you often get leading zeros that should not be
5499 interpreted as octal.
5501 64. Added qualify_domain to the redirect router, to override the global
5504 65. If a pathologically long header line contained very many addresses (the
5505 report of this problem mentioned 10 000) and each of them was rewritten,
5506 Exim could use up a very large amount of memory. (It kept on making new
5507 copies of the header line as it rewrote, and never released the old ones.)
5508 At the expense of a bit more processing, the header rewriting function has
5509 been changed so that it no longer eats memory in this way.
5511 66. The generation of the Received: header has been moved from the time that a
5512 message starts to be received, to the time that it finishes. The timestamp
5513 in the Received: header should now be very close to that of the <= log
5514 line. There are two side-effects of this change:
5516 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
5517 logged header lines no longer include the local Received: line, because
5518 it has not yet been created. The same applies to a copy of the message
5519 that is returned to a non-SMTP sender when a message is rejected.
5521 (b) When a filter file is tested using -bf, no additional Received: header
5522 is added to the test message. After some thought, I decided that this
5525 This change does not affect the value of $received_for. It is still set
5526 after address rewriting, but before local_scan() is called.
5528 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
5530 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
5531 gave an unhelpful panic error message, and a defer error. I have managed to
5532 change this behaviour so that it now rejects any supplied certificate,
5533 which seems right, as the list of acceptable certificates is empty.
5535 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
5536 gave an unhelpful defer error. I have not managed to make this reject any
5537 supplied certificates, but the error message it gives is "no certificate
5538 supplied", which is not helpful.
5540 70. exigrep's output now also includes lines that are not associated with any
5541 message, but which match the given pattern. Implemented by a patch from
5542 Martin Sluka, which also tidied up the Perl a bit.
5544 71. Recipient callout verification, like sender verification, was using <> in
5545 the MAIL FROM command. This isn't really the right thing, since the actual
5546 sender may affect whether the remote host accepts the recipient or not. I
5547 have changed it to use the actual sender in the callout; this means that
5548 the cache record is now keyed on a recipient/sender pair, not just the
5549 recipient address. There doesn't seem to be a real danger of callout loops,
5550 since a callout by the remote host to check the sender would use <>.
5551 [SEE ABOVE: changed after hitting problems.]
5553 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
5554 temporary errors. However, in the case of such a code being given after
5555 the end of a data transmission (i.e. after ".") Exim was failing to write
5556 a retry record for the message. (Yes, there was some broken host that was
5557 actually sending 8xx at this point.)
5559 73. An unknown lookup type in a host list could cause Exim to panic-die when
5560 the list was checked. (An example that provoked this was putting <; in the
5561 middle of a list instead of at the start.) If this happened during a DATA
5562 ACL check, a -D file could be left lying around. This kind of configuration
5563 error no longer causes Exim to die; instead it causes a defer errror. The
5564 incident is still logged to the main and panic logs.
5566 74. Buglet left over from Exim 3 conversion. The message "too many messages
5567 in one connection" was written to the rejectlog but not the mainlog, except
5568 when address rewriting (yes!) was being logged.
5570 75. Added write_rejectlog option.
5572 76. When a system filter was run not as root (that is, when system_filter_user
5573 was set), the values of the $n variables were not being returned to the
5574 main process; thus, they were not subsequently available in the $sn
5577 77. Added +return_path_on_delivery log selector.
5579 78. A connection timeout was being treated differently from recipients deferred
5580 when testing hosts_max_try with a message that was older than the host's
5581 retry timeout. (The host should not be counted, thus allowing all hosts to
5582 be tried at least once before bouncing.) This may have been the cause of an
5583 occasionally reported bug whereby a message would remain on the queue
5584 longer than the retry timeout, but would be bounced if a delivery was
5585 forced. I say "may" because I never totally pinned down the problem;
5586 setting up timeout/retry tests is difficult. See also the next item.
5588 79. The ultimate address timeout was not being applied to errors that involved
5589 a combination of host plus message (for example, a timeout on a MAIL
5590 command). When an address resolved to a number of possible hosts, and they
5591 were not all tried for each delivery (e.g. because of hosts_max_try), a
5592 message could remain on the queue longer than the retry timeout.
5594 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
5597 81. Fixed an obscure SMTP outgoing bug which required at least the following
5598 conditions: (a) there was another message waiting for the same server;
5599 (b) the server returned 5xx to all RCPT commands in the first message so
5600 that the message was not completed; (c) the server dropped the connection
5601 or gave a negative response to the RSET that Exim sends to abort the
5602 transaction. The observed case was a dropped connection after DATA that had
5603 been sent in pipelining mode. That is, the server had advertised PIPELINING
5604 but was not implementing it correctly. The effect of the bug was incorrect
5605 behaviour, such as trying another host, and this could lead to a crash.
5611 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
5612 and daemon.c were passed as pointers to ints; they should have been
5613 pointers to socklen_t variables (which are typically unsigned ints).
5615 2. Some signed/unsigned type warnings in the os.c file for Linux have been
5618 3. Fixed a really odd bug that affected only the testing scheme; patching a
5619 certain fixed string in the binary changed the value of another string that
5620 happened to be identical to the end of the original first string.
5622 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
5623 name", it returns that address as the IP address. On some operating
5624 systems (e.g. Solaris), it also passes back the IP address string as the
5625 "host name". However, on others (e.g. Linux), it passes back an empty
5626 string. Exim wasn't checking for this, and was changing the host name to an
5627 empty string, assuming it had been canonicized.
5629 5. Although rare, it is permitted to have more than one PTR record for a given
5630 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
5631 all the names associated with an address, because they do in Solaris.
5632 However, it seems that they do not in Linux for data that comes from the
5633 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
5634 I found this out when I moved to a new Linux workstation and tried to run
5635 the Exim test suite.
5637 To get round this problem I have changed the code so that it now does its
5638 own call to the DNS to look up PTR records when searching for a host name.
5639 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
5640 addresses that are only in /etc/hosts are still found.
5642 This behaviour is, however, controlled by an option called host_lookup_
5643 order, which defaults to "bydns:byaddr". If people want to use the other
5644 order, or indeed, just use one or the other means of lookup, they can
5645 specify it in this variable.
5647 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
5648 some operating systems, this comes back from gethostbyaddr() as an empty
5649 string, and this is what Exim used to test for. However, it seems that in
5650 other systems, "." is yielded. Exim now tests for this case too.
5652 7. The values of check_spool_space and check_log_space are now held internally
5653 as a number of kilobytes instead of an absolute number of bytes. If a
5654 numbers is specified without 'K' or 'M', it is rounded up to the nearest
5655 kilobyte. This means that much larger values can be stored.
5657 8. Exim monitor: an attempt to get the action menu when not actually pointing
5658 at a message produces an empty menu entitled "No message selected". This
5659 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
5660 no entries in it ("Shell widget menu has zero width and/or height"). So I
5661 have added a single, blank menu entry in this case.
5663 9. Added ${quote_local_part.
5665 10. MIME decoding is now applied to the contents of Subject: header lines when
5668 11. Now that a reference to $sender_host_address automatically causes a reverse
5669 lookup to occur if necessary (4.13/18), there is no need to arrange for a
5670 host lookup before query-style lookups in lists that might use this
5671 variable. This has therefore been abolished, and the "net-" prefix is no
5672 longer necessary for query-style lookups.
5674 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
5675 have been a typo for LFLAGS, so it has been changed.
5677 13. The install script calls Exim with "-C /dev/null" in order to find the
5678 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
5679 to be output. Howeve, since Exim outputs its version number before the
5680 error, it didn't break the script. It just looked ugly. I fixed this by
5681 always allowing "-C /dev/null" if the caller is root.
5683 14. Ignore overlarge ACL variable number when reading spool file - insurance
5684 against a later release with more variables having written the file.
5686 15. The standard form for an IPv6 address literal was being rejected by EHLO.
5687 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5688 this, as well as the form without the "IPv6" on the front.
5690 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
5691 OS/Makefile-Darwin file.
5693 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
5694 only to LDAP libraries that do not have LDAP_OPT_DEREF.
5696 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
5697 variable. However, some OS use ints for inodes. I've added cast to long int
5698 to get rid of the compiler warning.
5700 19. I had forgotten to lock out "/../" in configuration file names when
5701 ALT_CONFIG_PREFIX was set.
5703 20. Routers used for verification do not need to specify transports. However,
5704 if such a router generated a host list, and callout was configured, Exim
5705 crashed, because it could not find a port number from the (non-existent)
5706 transport. It now assumes port 25 in this circumstance.
5708 21. Added the -t option to exigrep.
5710 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
5711 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
5712 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
5713 them is set, LOOKUP_LSEARCH is forced.
5715 23. "exim -bV" now outputs a list of lookups that are included in the binary.
5717 24. Added sender and host information to the "rejected by local_scan()" log
5718 line; previously there was no indication of these.
5720 25. Added .include_if_exists.
5722 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
5723 Linux. It turns out that not all file systems support this. Apparently some
5724 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
5725 do it.) To cope with this, the error EINVAL, which means that sync-ing is
5726 not supported on the file descriptor, is now ignored when Exim is trying to
5727 sync a directory. This applies only to Linux.
5729 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
5731 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
5732 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
5735 29. Added support for maildirsize files from supplied patch (modified a bit).
5737 30. The use of :fail: followed by an empty string could lead Exim to respond to
5738 sender verification failures with (e.g.):
5740 550 Verification failed for <xxx>
5741 550 Sender verify failed
5743 where the first response line was missing the '-' that indicates it is not
5744 the final line of the response.
5746 31. The loop for finding the name of the user that called Exim had a hardwired
5747 limit of 10; it now uses the value of finduser_retries, which is used for
5748 all other user lookups.
5750 32. Added $received_count variable, available in data and not_smtp ACLs, and at
5753 33. Exim was neglecting to zero errno before one call of strtol() when
5754 expanding a string and expecting an integer value. On some systems this
5755 resulted in spurious "integer overflow" errors. Also, it was casting the
5756 result into an int without checking.
5758 34. Testing for a connection timeout using "timeout_connect" in the retry rules
5759 did not work. The code looks as if it has *never* worked, though it appears
5760 to have been documented since at least releast 1.62. I have made it work.
5762 35. The "timeout_DNS" error in retry rules, also documented since at least
5763 1.62, also never worked. As it isn't clear exactly what this means, and
5764 clearly it isn't a major issue, I have abolished the feature by treating it
5765 as "timeout", and writing a warning to the main and panic logs.
5767 36. The display of retry rules for -brt wasn't always showing the error code
5770 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
5771 timeout_connect_A, timeout_connect_MX.
5773 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
5774 to the empty sender.
5776 39. The daemon was not analysing the content of -oX till after it had closed
5777 stderr and disconnected from the controlling terminal. This meant that any
5778 syntax errors were only noted on the panic log, and the return code from
5779 the command was 0. By re-arranging the code a little, I've made the
5780 decoding happen first, so such errors now appear on stderr, and the return
5781 code is 1. However, the actual setting up of the sockets still happens in
5782 the disconnected process, so errors there are still only recorded on the
5785 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
5786 connections (as happens on some IP stacks) was logged at start up time as
5787 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
5788 it from "IPv6 and IPv4", which means that two separate sockets are being
5791 41. The debug output for gethostbyname2() or getipnodebyname() failures now
5792 says whether AF_INET or AF_INET6 was passed as an argument.
5794 42. Exiwhat output was messed up when time zones were included in log
5797 43. Exiwhat now gives more information about the daemon's listening ports,
5798 and whether -tls-on-connect was used.
5800 44. The "port" option of the smtp transport is now expanded.
5802 45. A "message" modifier in a "warn" statement in a non-message ACL was being
5803 silently ignored. Now an error message is written to the main and panic
5806 46. There's a new ACL modifier called "logwrite" which writes to a log file
5807 as soon as it is encountered.
5809 47. Added $local_user_uid and $local_user_gid at routing time.
5811 48. Exim crashed when trying to verify a sender address that was being
5814 49. Exim was recognizing only a space character after ".include". It now also
5815 recognizes a tab character.
5817 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
5818 extracting the relevant information from the specification. The man page no
5819 longer contains scrambled data for the -d option, and I've added a section
5820 at the front about calling Exim under different names.
5822 51. Added "extra_headers" argument to the "mail" command in filter files.
5824 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
5827 53. Installed eximstats 1.29.
5829 54. Added transport_filter_timeout as a generic transport option.
5831 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
5832 Cc: header lines. This was required by RFC 822, but it not required by RFC
5835 56. Exim used to add From:, Date:, and Message-Id: header lines to any
5836 incoming messages that did not have them. Now it does so only if the
5837 message originates locally, that is, if there is no associated remote host
5838 address. When Resent- header lines are present, this applies to the Resent-
5839 lines rather than the non-Resent- lines.
5841 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
5842 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
5844 58. Messages for configuration errors now include the name of the main
5845 configuration file - useful now that there may be more than one file in a
5846 list (.included file names were always shown).
5848 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
5849 for those rare installations that do not start the daemon as root or run it
5850 setuid root. I've cut out the call to initgroups() if the daemon is not
5853 60. The Exim user and group can now be bound into the binary as text strings
5854 that are looked up at the start of Exim's processing.
5856 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
5858 62. Added $mailstore_basename variable.
5860 63. Installed patch to sieve.c from Michael Haardt.
5862 64. When Exim failed to open the panic log after failing to open the main log,
5863 the original message it was trying to log was written to stderr and debug
5864 output, but if they were not available (the usual case in production), it
5865 was lost. Now it is written to syslog before the two lines that record the
5866 failures to open the logs.
5868 65. Users' Exim filters run in subprocesses under the user's uid. It is
5869 possible for a "deliver" command or an alias in a "personal" command to
5870 provoke an address rewrite. If logging of address rewriting is configured,
5871 this fails because the process is not running as root or exim. There may be
5872 a better way of dealing with this, but for the moment (because 4.30 needs
5873 to be released), I have disabled address rewrite logging when running a
5874 filter in a non-root, non-exim process.
5880 1. The buildconfig auxiliary program wasn't quoting the value set for
5881 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
5882 not defined. This bug was masked in 4.22 by the effect that was fixed in
5885 2. Some messages that were rejected after a message id was allocated were
5886 shown as "incomplete" by exigrep. It no longer does this for messages that
5887 are rejected by local_scan() or the DATA or non-SMTP ACLs.
5889 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
5890 have allow_domain_literals set, the ID did not get logged in the <= line.
5891 Domain literals are now always recognized in Message-ID: header lines.
5893 4. The first argument for a ${extract expansion item is the key name or field
5894 number. Leading and trailing spaces in this item were not being ignored,
5895 causing some misleading effects.
5897 5. When deliver_drop_privilege was set, single queue runner processes started
5898 manually (i.e. by the command "exim -q") or by the daemon (which uses the
5899 same command in the process it spins off) were not dropping privilege.
5901 6. When the daemon running as "exim" started a queue runner, it always
5902 re-executed Exim in the spun-off process. This is a waste of effort when
5903 deliver_drop_privilege is set. The new process now just calls the
5904 queue-runner function directly.
5910 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
5913 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
5914 ignored. Though the use of -oP was forcing the writing of a pid file, it
5915 was always written to the default place.
5917 3. If the message "no IP address found for host xxxx" is generated during
5918 incoming verification, it is now followed by identification of the incoming
5919 connection (so you can more easily find what provoked it).
5921 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
5923 5. Added some features to "harden" Exim a bit more against certain attacks:
5925 (a) There is now a build-time option called FIXED_NEVER_USERS that can
5926 be put in Local/Makefile. This is like the never_users runtime option,
5927 but it cannot be overridden. The default setting is "root".
5929 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
5930 prefix string with which any file named in a -C command line option
5933 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
5934 is retained for -C and -D only if the caller of Exim is root. Without
5935 it, the exim user may also use -C and -D and retain privilege.
5937 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
5938 command line option is disabled.
5940 6. Macro names set by the -D option must start with an upper case letter, just
5941 like macro names defined in the configuration file.
5943 7. Added "dereference=" facility to LDAP.
5945 8. Two instances of the typo "uknown" in the source files are fixed.
5947 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
5948 the Configure-Makefile script screwed up while processing it.
5950 10. Incorporated PCRE 4.4.
5952 11. The SMTP synchronization check was not operating right at the start of an
5953 SMTP session. For example, it could not catch a HELO sent before the client
5954 waited for the greeting. There is now a check for outstanding input at the
5955 point when the greeting is written. Because of the duplex, asynchronous
5956 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
5957 way, but not yet received, when the check is performed.
5959 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
5960 on TCP/IP sockets, because this apparently causes some broken clients to
5963 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
5964 unchanged) from the Cygwin maintainer.
5966 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
5967 was supported instead of testing for mbx. Effectively a typo.
5969 15. The spa authenticator server code was not checking that the input it
5970 received was valid base64.
5972 16. The debug output line for the "set" modifier in ACLs was not showing the
5973 name of the variable that was being set.
5975 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
5977 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
5978 lookup on its own. Something else was needed to trigger the lookup. For
5979 example, a match in host_lookup or the need for a host name in a host list.
5980 Now, if $sender_host_name is referenced and the host name has not yet been
5981 looked up, a lookup is performed. If the lookup fails, the variable remains
5982 empty, and $host_lookup_failed is set to "1".
5984 19. Added "eqi" as a case-independent comparison operator.
5986 20. The saslauthd authentication condition could segfault if neither service
5987 nor realm was specified.
5989 21. If an overflowing value such as "2048M" was set for message_size_limit, the
5990 error message that was logged was misleading, and incoming SMTP
5991 connections were dropped. The message is now more accurate, and temporary
5992 errors are given to SMTP connections.
5994 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
5995 (except RSET) with a 421 error, until QUIT is received. However, it was
5996 failing to send a response to QUIT.
5998 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
5999 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
6000 ACL is now run after the helo_try_verify_hosts code.
6002 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
6003 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
6004 case-independent, so other case variants are also recognized). Apparently
6005 some systems use these upper case variants.
6007 25. If more than two messages were waiting for the same host, and a transport
6008 filter was specified for the transport, Exim sent two messages over the
6009 same TCP/IP connection, and then failed with "socket operation on non-
6010 socket" when it tried to send the third.
6012 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
6014 27. The extern definition of crypt16() in expand.c was not being excluded when
6015 the OS had its own crypt16() function.
6017 28. Added bounce_return_body as a new option, and bounce_return_size_limit
6018 as a preferred synonym for return_size_limit, both as an option and as an
6021 29. Added LIBS=-liconv to OS/Makefile-OSF1.
6023 30. Changed the default configuration ACL to relax the local part checking rule
6024 for addresses that are not in any local domains. For these addresses,
6025 slashes and pipe symbols are allowed within local parts, but the sequence
6026 /../ is explicitly forbidden.
6028 31. SPA server authentication was not clearing the challenge buffer before
6031 32. log_message in a "warn" ACL statement was writing to the reject log as
6032 well as to the main log, which contradicts the documentation and doesn't
6033 seem right (because no rejection is happening). So I have stopped it.
6035 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
6036 However, I am unable to do any testing of this.
6038 34. Fixed an infelicity in the appendfile transport. When checking directories
6039 for a mailbox, to see if any needed to be created, it was accidentally
6040 using path names with one or more superfluous leading slashes; tracing
6041 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
6043 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
6044 contents are added to the log line that is written for every discarded
6045 recipient. (Previously a log_message setting was ignored.)
6047 36. The ${quote: operator now quotes the string if it is empty.
6049 37. The install script runs exim in order to find its version number. If for
6050 some reason other than non-existence or emptiness, which it checks, it
6051 could not run './exim', it was installing it with an empty version number,
6052 i.e. as "exim-". This error state is now caught, and the installation is
6055 38. An argument was missing from the function that creates an error message
6056 when Exim fails to connect to the socket for saslauthd authentication.
6057 This could cause Exim to crash, or give a corrupted message.
6059 39. Added isip, isip4, and isip6 to ${if conditions.
6061 40. The ACL variables $acl_xx are now saved with the message, and can be
6062 accessed later in routers, transports, and filters.
6064 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
6065 strings in the file are not string-expanded.
6067 42. If a MAIL command specified a SIZE value that was too large to fit into an
6068 int variable, the check against message_size_limit failed. Such values are
6069 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
6070 day this will have to be increased, but I don't think I want to be around
6071 when emails are that large.
6078 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
6079 iconv() is not standard in FreeBSD.
6081 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
6082 IPv6 enabled. The observed symptom was a segmentation fault on return from
6083 the function os_common_find_running_interfaces() in src/os.c.
6085 3. In the check_special_case() function in daemon.c I had used "errno" as an
6086 argument name, which causes warnings on some systems. This was basically a
6087 typo, since it was named "eno" in the comments!
6089 4. The code that waits for the clock to tick (at a resolution of some fraction
6090 of a second) so as to ensure message-id uniqueness was always waiting for
6091 at least one whole tick, when it could have waited for less. [This is
6092 almost certainly not relevant at current processor speeds, where it is
6093 unlikely to ever wait at all. But we try to future-proof.]
6095 5. The function that sleeps for a time interval that includes fractions of a
6096 second contained a race. It did not block SIGALRM between setting the
6097 timer, and suspending (a couple of lines later). If the interval was short
6098 and the sigsuspend() was delayed until after it had expired, the suspension
6099 never ended. On busy systems this could lead to processes getting stuck for
6102 6. Some uncommon configurations may cause a lookup to happen in a queue runner
6103 process, before it forks any delivery processes. The open lookup caching
6104 mechanism meant that the open file or database connection was passed into
6105 the delivery process. The problem was that delivery processes always tidy
6106 up cached lookup data. This could cause a problem for the next delivery
6107 process started by the queue runner, because the external queue runner
6108 process does not know about the closure. So the next delivery process
6109 still has data in the lookup cache. In the case of a file lookup, there was
6110 no problem because closing a file descriptor in a subprocess doesn't affect
6111 the parent. However, if the lookup was caching a connection to a database,
6112 the connection was closed, and the second delivery process was likely to
6113 see errors such as "PGSQL: query failed: server closed the connection
6114 unexpectedly". The problem has been fixed by closing all cached lookups
6115 in a queue runner before running a delivery process.
6117 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
6118 seem to have the "const" qualifier which it has on other OS. I've
6121 8. Change 4.21/2 was too strict. It is only if there are two authenticators
6122 *of the same type* (client or server) with the same public name that an
6123 error should be diagnosed.
6125 9. When Exim looked up a host name for an IP address, but failed to find the
6126 original IP address when looking up the host name (a safety check), it
6127 output the message "<ip address> does not match any IP for NULL", which was
6128 confusing, to say the least. The bug was that the host name should have
6129 appeared instead of "NULL".
6131 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
6132 user that is built into the binary, and the -C or -D options is used, root
6133 privilege is dropped before the configuration file is read. In addition,
6134 logging is switched to stderr instead of the normal log files. If the
6135 configuration then re-defines the Exim user, the unprivileged environment
6136 is probably not what is expected, so Exim logs a panic warning message (but
6139 However, if deliver_drop_privilege is set, the unprivileged state may well
6140 be exactly what is intended, so the warning has been cut out in that case,
6141 and Exim is allowed to try to write to its normal log files.
6147 1. smtp_return_error_details was not giving details for temporary sender
6148 or receiver verification errors.
6150 2. Diagnose a configuration error if two authenticators have the same public
6153 3. Exim used not to create the message log file for a message until the first
6154 delivery attempt. This could be confusing when incoming messages were held
6155 for policy or load reasons. The message log file is now created at the time
6156 the message is received, and an initial "Received" line is written to it.
6158 4. The automatically generated man page for command line options had a minor
6159 bug that caused no ill effects; however, a more serious problem was that
6160 the procedure for building the man page automatically didn't always
6161 operate. Consequently, release 4.20 contains an out-of-date version. This
6162 shouldn't happen again.
6164 5. When building Exim with embedded Perl support, the script that builds the
6165 Makefile was calling 'perl' to find its compile-time parameters, ignoring
6166 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
6168 6. The freeze_tell option was not being used for messages that were frozen on
6169 arrival, either by an ACL or by local_scan().
6171 7. Added the smtp_incomplete_transaction log selector.
6173 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
6174 was accepting AUTH without a new EHLO.
6176 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
6177 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
6180 10. Logging of TCP/IP connections (when configured) now happens in the main
6181 daemon process instead of the child process, so that the TCP/IP connection
6182 count is more accurate (but it can never be perfect).
6184 11. The use of "drop" in a nested ACL was not being handled correctly in the
6185 outer ACL. Now, if condition failure induced by the nested "drop" causes
6186 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
6187 or "require"), the connection is dropped.
6189 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
6190 that yield "discard" can now be used with an "accept" or a "discard" verb,
6191 but an error is generated for any others (because I can't see a useful way
6192 to define what should happen).
6194 13. When an ACL is read dynamically from a file (or anywhere else), the lines
6195 are now processed in the same way as lines in the Exim configuration file.
6196 In particular, continuation lines are supported.
6198 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
6200 15. Added -ti meaning -t -i.
6202 16. Check for letters, digits, hyphens, and dots in the names of dnslist
6203 domains, and warn by logging if others are found.
6205 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
6206 returned from GIFCONF when Exim is trying to find the list of interfaces on
6207 a host. The code in os.c has been modified to copy each ifreq to an aligned
6208 structure in all cases.
6210 Also, in some cases, the returned ifreq's were being copied to a 'struct
6211 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
6212 means the last couple of bytes of an IPv6 address could be chopped if the
6213 ifreq contained only a normal sockaddr (14 bytes storage).
6215 18. Named domain lists were not supported in the hosts_treat_as_local option.
6216 An entry such as +xxxx was not recognized, and was treated as a literal
6219 19. Ensure that header lines added by a DATA ACL are included in the reject log
6220 if the ACL subsequently rejects the message.
6222 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
6223 MD5 (which is deprecated).
6225 21. When testing a filter file using -bf, Exim was writing a message when it
6226 took the sender from a "From " line in the message, but it was not doing so
6227 when it took $return_path from a Return-Path: header line. It now does.
6229 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
6230 with a valid header line field name (a series of printing characters
6231 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
6233 23. Changed "disc" in the source to "disk" to conform to the documentation and
6234 the book and for uniformity.
6236 24. Ignore Sendmail's -Ooption=value command line item.
6238 25. When execve() failed while trying to run a command in a pipe transport,
6239 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
6240 could be confused with a return value of 69 from the command itself. This
6241 has been changed to 127, the value the shell returns if it is asked to run
6242 a non-existent command. The wording for the related log line suggests a
6243 non-existent command as the problem.
6245 26. If received_header_text expands to an empty string, do not add a Received:
6246 header line to the message. (Well, it adds a token one on the spool, but
6247 marks it "old" so that it doesn't get used or transmitted.)
6249 27. Installed eximstats 1.28 (addition of -nt option).
6251 28. There was no check for failure on the call to getsockname() in the daemon
6252 code. This can fail if there is a shortage of resources on the system, with
6253 ENOMEM, for example. A temporary error is now given on failure.
6255 29. Contrary to the C standard, it seems that in some environments, the
6256 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
6257 program. Exim now does this explicitly; it affects the formatting of
6258 timestamps using strftime().
6260 30. If exiqsumm was given junk data, it threw up some uninitialized variable
6261 complaints. I've now initialized all the variables, to avoid this.
6263 32. Header lines added by a system filter were not being "seen" during
6264 transport-time rewrites.
6266 33. The info_callback() function passed to OpenSSL is set up with type void
6267 (*)(SSL *, int, int), as described somewhere. However, when calling the
6268 function (actually a macro) that sets it up, the type void(*)() is
6269 expected. I've put in a cast to prevent warnings from picky compilers.
6271 34. If a DNS black list lookup found a CNAME record, but there were no A
6272 records associated with the domain it pointed at, Exim crashed.
6274 35. If a DNS black list lookup returned more than one A record, Exim ignored
6275 all but the first. It now scans all returned addresses if a particular IP
6276 value is being sought. In this situation, the contents of the
6277 $dnslist_value variable are a list of all the addresses, separated by a
6280 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
6281 to accept a host name and all its aliases if the forward lookup for any of
6282 them yielded the IP address of the incoming connection. Now it accepts only
6283 those names whose forward lookup yields the correct IP address. Any other
6284 names are discarded. This closes a loophole whereby a rogue DNS
6285 administrator could create reverse DNS records to break through a
6286 wildcarded host restriction in an ACL.
6288 37. If a user filter or a system filter that ran in a subprocess used any of
6289 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
6290 the wrong values were passed to the pipe command ($thisaddress had the
6291 value of $0, $0 had the value of $1, etc). This bug was introduced by
6292 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
6294 38. Improved the line breaking for long SMTP error messages from ACLs.
6295 Previously, if there was no break point between 40 and 75 characters, Exim
6296 left the rest of the message alone. Two changes have been made: (a) I've
6297 reduced the minimum length to 35 characters; (b) if it can't find a break
6298 point between 35 and 75 characters, it looks ahead and uses the first one
6299 that it finds. This may give the occasional overlong line, but at least the
6300 remaining text gets split now.
6302 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
6303 file descriptors might be low, and that setting 1000 would always raise it.
6304 It turns out that in some environments, the limit is already over 1000 and
6305 that lowering it causes trouble. So now Exim takes care not to decrease it.
6307 40. When delivering a message, the value of $return_path is set to $sender_
6308 address at the start of routing (routers may change the value). By an
6309 oversight, this default was not being set up when an address was tested by
6310 -bt or -bv, which affected the outcome if any router or filter referred to
6313 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
6314 log only when "message" or "log_message" are set. However, if one of the
6315 conditions was an address verification, or a call to a nested ACL, the
6316 messages generated by the underlying test were being passed through. This
6317 no longer happens. The underlying message is available in $acl_verify_
6318 message for both "message" and "log_message" expansions, so it can be
6319 passed through if needed.
6321 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
6322 new expansion $bh_ to give the encoded byte string without charset
6323 translation. Translation happens only if iconv() is available; HAVE_ICONV
6324 indicates this at build time. HEADERS_CHARSET gives the charset to
6325 translate to; headers_charset can change it in the configuration, and
6326 "headers charset" can change it in an individual filter file.
6328 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
6329 that creates RFC 2047 encoded "words" labels them as that charset instead
6330 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
6331 expansion operator; (ii) when Exim creates a From: line for a local
6332 message; (iii) when a header line is rewritten to include a "phrase" part.
6334 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
6335 buggy, causing it to skip the first lines of messages whose message ID
6336 ended in 'D'. This would not have bitten before Exim release 4.14, because
6337 message IDs were unlikely to end in 'D' before then. The effect was to have
6338 incorrect size information for certain domains.
6340 45. #include "config.h" was missing at the start of the crypt16.c module. This
6341 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
6344 46. If there was a timeout during a "random" callout check, Exim treated it as
6345 a failure of the random address, and carried on sending RSET and the real
6346 address. If the delay was just some slowness somewhere, the response to the
6347 original RCPT would be taken as a response to RSET and so on, causing
6348 mayhem of various kinds.
6350 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
6351 when I implemented it. It didn't allow for the fact that some option values
6352 may legitimatetly be negative (e.g. size_addition), and it didn't even do
6353 the right test for positive values.
6355 48. Domain names in DNS records are case-independent. Exim always looks them up
6356 in lower case. Some resolvers return domain names in exactly the case they
6357 appear in the zone file, that is, they may contain uppercase letters. Not
6358 all resolvers do this - some return always lower case. Exim was treating a
6359 change of case by a resolver as a change of domain, similar to a widening
6360 of a domain abbreviation. This triggered its re-routing code and so it was
6361 trying to route what was effectively the same domain again. This normally
6362 caused routing to fail (because the router wouldn't handle the domain
6363 twice). Now Exim checks for this case specially, and just changes the
6364 casing of the domain that it ultimately uses when it transmits the message
6367 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
6370 50. If a filter generated a file delivery with a non-absolute name (possible if
6371 no home directory exists for the router), the forbid_file option was not
6374 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
6375 the existing equality matching.
6377 52. Exim was using ints instead of ino_t variables in some places where it was
6378 dealing with inode numbers.
6380 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
6381 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
6382 called TMPDIR, and if it finds it is different, it changes its value.
6384 54. The smtp_printf() function is now made available to local_scan() so
6385 additional output lines can be written before returning. There is also an
6386 smtp_fflush() function to enable the detection of a dropped connection.
6387 The variables smtp_input and smtp_batched_input are exported to
6390 55. Changed the default runtime configuration: the message "Unknown user"
6391 has been removed from the ACL, and instead placed on the localuser router,
6392 using the cannot_route_message feature. This means that any verification
6393 failures that generate their own messages won't get overridden. Similarly,
6394 the "Unrouteable address" message that was in the ACL for unverifiable
6395 relay addresses has also been removed.
6397 56. Added hosts_avoid_esmtp to the smtp transport.
6399 57. The exicyclog script was not checking for the esoteric option
6400 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
6401 will work only if exicyclog is run under the appropriate euid.
6403 58. Following a discussion on the list, the rules by which Exim recognises line
6404 endings on incoming messages have been changed. The -dropcr and drop_cr
6405 options are now no-ops, retained only for backwards compatibility. The
6406 following line terminators are recognized: LF CRLF CR. However, special
6407 processing applies to CR:
6409 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
6410 nor a local message in the state where . is a terminator.
6412 (ii) If a bare CR is encountered in a header line, an extra space is added
6413 after the line terminator so as not to end the header. The reasoning
6414 behind this is that bare CRs in header lines are most likely either
6415 to be mistakes, or people trying to play silly games.
6417 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
6418 was being incorrectly given as 18 bytes larger than it should have been.
6419 This is a VOB (very old bug).
6421 60. This may never have affected anything current, but just in case it has:
6422 When the local host is found other than at the start of a list of hosts,
6423 the local host, those with the same MX, and any that follow, are discarded.
6424 When the list in question was part of a longer list of hosts, the following
6425 hosts (not currently being processed) were also being discarded. This no
6426 longer happens. I'm not sure if this situation could ever has previously
6429 61. Added the "/MX" feature to lists of hosts in the manualroute and query
6432 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
6433 header. This is something that is recommended in a new Internet Draft, and
6434 is something that is documented as being done by Sendmail. There are two
6435 possible values. For messages generated by the autoreply transport, Exim
6438 Auto-Submitted: auto-replied
6440 whereas for all other generated messages (e.g. bounces) it adds
6442 Auto-Submitted: auto-generated
6444 63. The "personal" condition in filters now includes a test for the
6445 Auto-Submitted: header. If it contains the string "auto-" the message it
6446 not considered personal.
6448 64. Added rcpt_include_affixes as a generic transport option.
6450 65. Added queue_only_override (default true).
6452 66. Added the syslog_duplication option.
6454 67. If what should have been the first header line of a message consisted of
6455 a space followed by a colon, Exim was mis-interpreting it as a header line.
6456 It isn't of course - it is syntactically invalid and should therefore be
6457 treated as the start of the message body. The misbehaviour could have
6458 caused a number of strange effects, including loss of data in subsequent
6459 header lines, and spool format errors.
6461 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
6462 client host had authenticated. This control can now be exercised by an ACL
6463 for more flexibility.
6465 69. By default, callouts do not happen when testing with -bh. There is now a
6466 variant, -bhc, which does actually run the callout code, including
6467 consulting and updating the callout cache.
6469 70. Added support for saslauthd authentication, courtesy of Alexander
6472 71. If statvfs() failed on the spool or log directories while checking their
6473 size for availability, Exim confusingly gave the error "space shortage".
6474 Furthermore, in debugging mode it crashed with a floating point exception.
6475 These checks are done if check_{spool,log}_{space,inodes} are set, and when
6476 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
6477 serious problem, Exim now writes to the main and panic logs when this
6478 happens, with details of the failure. It then refuses to accept the
6479 incoming message, giving the message "spool directory problem" or "log
6480 directory problem" with a 421 code for SMTP messages.
6482 72. When Exim is about to re-exec itself, it ensures that the file descriptors
6483 0, 1, and 2 exist, because some OS complain for execs without them (see
6484 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
6485 descriptors. However, the code omitted to check that the open succeeded,
6486 causing mysterious errors if for some reason the permissions on /dev/null
6487 got screwed. Now Exim writes a message to the main and panic logs, and
6488 bombs out if it can't open /dev/null.
6490 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
6491 interact so that it is all more flexible. It is supposed to remain
6492 backwards compatible. Also added extra_local_interfaces.
6494 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
6495 to bomb out with an assertion failure - to the client this appears as a
6496 connection drop. This problem occurs in the part of the code that was taken
6497 from the Samba project. Fortunately, the assertion is in a very simple
6498 function, so I have fixed this by reproducing the function inline in the
6499 one place where it is called, and arranging for authentication to fail
6500 instead of killing the process with assert().
6502 75. The SPA client code was not working when the server requested OEM rather
6503 than Unicode encoding.
6505 76. Added code to make require_files with a specific uid setting more usable in
6506 the case where statting the file as root fails - usually a non-root-mounted
6507 NFS file system. When this happens and the failure is EACCES, Exim now
6508 forks a subprocess and does the per-uid checking as the relevant uid.
6510 77. Added process_log_path.
6512 78. If log_file_path was not explicitly set, a setting of check_log_space or
6513 check_log_inodes was ignored.
6515 79. If a space check for the spool or log partitions fails, the incident is now
6516 logged. Of course, in the latter case the data may get lost...
6518 80. Added the %p formatting code to string_format() so that it can be used to
6519 print addresses in debug_print(). Adjusted all the address printing in the
6520 debugging in store.c to use %p rather than %d.
6522 81. There was a concern that a line of code in smtp_in.c could overflow a
6523 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
6524 initially expressed, the concern was not well-founded, because trailing
6525 spaces are removed early. However, if the trailing spaces were followed by
6526 a NULL, they did not get removed, so the overflow was possible. Two fixes
6529 (a) I re-wrote the offending code in a cleaner fashion.
6530 (b) If an incoming SMTP command contains a NULL character, it is rejected
6533 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
6534 runs initgroups(), so that if the Exim user is in any additional groups,
6535 they will be used during message reception.
6541 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob