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.
40 JH/10 Support ${sha256:} applied to a string (as well as the previous
43 JH/11 Cutthrough: avoid using the callout hints db on a verify callout when
44 a cutthrough deliver is pending, as we always want to make a connection.
45 This also avoids re-routing the message when later placing the cutthrough
46 connection after a verify cache hit.
47 Do not update it with the verify result either.
49 JH/12 Cutthrough: disable when verify option success_on_redirect is used, and
50 when routing results in more than one destination address.
55 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
56 and 3.4.4 - once the server is enabled to respond to an OCSP request
57 it does even when not requested, resulting in a stapling non-aware
58 client dropping the TLS connection.
60 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
61 support variable-length bit vectors. No functional change.
63 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
64 The I= interface field on outgoing lines is now after the H= remote
65 host field, same as incoming lines. There is a separate
66 outgoing_interface log selector which allows you to disable the
69 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
70 If not running log_selector +smtp_connection the mainlog would be held
71 open indefinitely after a "too many connections" event, including to a
72 deleted file after a log rotate. Leave the per net connection logging
73 leaving it open for efficiency as that will be quickly detected by the
74 check on the next write.
76 HS/01 Bug 1671: Fix post transport crash.
77 Processing the wait-<transport> messages could crash the delivery
78 process if the message IDs didn't exist for some reason. When
79 using 'split_spool_directory=yes' the construction of the spool
80 file name failed already, exposing the same netto behaviour.
82 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
83 mime_regex ACL conditions.
85 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
86 to DSN fail messages (bounces): remote IP, remote greeting, remote response
87 to HELO, local diagnostic string.
89 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
90 log line to debug. Even when configured with a tls authenticator many
91 client connections are expected to not authenticate in this way, so
92 an authenticate fail is not an error.
94 HS/02 Add the Exim version string to the process info. This way exiwhat
95 gives some more detail about the running daemon.
97 JH/06 Bug 1395: time-limit cacheing of DNS lookups, to the TTL value. This may
98 matter for fast-change records such as DNSBLs.
100 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
101 retry record, even if constant. There may be multiple transports with
102 different interface settings and the retry behaviour needs to be kept
105 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
107 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
109 JH/10 Bug 840: fix log_defer_output option of pipe transport
111 JH/11 Bug 830: use same host for all RCPTS of a message, even under
112 hosts_randomize. This matters a lot when combined with mua_wrapper.
114 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
115 ${quote_pgsql:<string>} operator.
117 JH/13 Bug 1708: avoid misaligned access in cached lookup.
119 JH/14 Change header file name for freeradius-client. Relevant if compiling
120 with Radius support; from the Gentoo tree and checked under Fedora.
122 JH/15 Bug 1712: Introduce $prdr_requested flag variable
124 JH/16 Bug 1714: Permit an empty string as expansion result for transport
125 option transport_filter, meaning no filtering.
127 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
129 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
130 defaults to "*" (all hosts). The variable is now available when not built
131 with TLS, default unset, mainly to enable keeping the testuite sane.
132 If a server certificate is not supplied (via tls_certificate) an error is
133 logged, and clients will find TLS connections fail on startup. Presumably
134 they will retry in-clear.
135 Packagers of Exim are strongly encouraged to create a server certificate
136 at installation time.
138 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
139 with the $config_file variable.
141 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
142 in transport context, after the attempt, and per-recipient. The latter type
143 is per host attempted. The event data is the error message, and the errno
144 information encodes the lookup type (A vs. MX) used for the (first) host,
145 and the trailing two digits of the smtp 4xx reponse.
147 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
148 to write to mainlog (or rejectlog, paniclog) in the window between file
149 creation and permissions/ownership being changed. Particularly affects
150 installations where exicyclog is run as root, rather than exim user;
151 result is that the running daemon panics and dies.
153 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
155 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
156 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
157 "pri" and "weight". Note that the previous implicit priority given by the
158 list order is no longer honoured.
160 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalisation
163 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
164 by defining SUPPORT_SOCKS.
166 JH/26 Move PROXY support from Experimental to mainline, enabled for a build
167 by defining SUPPORT_PROXY. Note that the proxy_required_hosts option
168 is renamed to hosts_proxy, and the proxy_{host,target}_{address,port}.
169 variables are renamed to proxy_{local,external}_{address,port}.
171 JH/27 Move Internationalisation support from Experimental to mainline, enabled
172 for a build by defining SUPPORT_I18N
174 JH/28 Bug 1745: Fix redis lookups to handle (quoted) spaces embedded in parts
175 of the query string, and make ${quote_redis:} do that quoting.
177 JH/29 Move Events support from Experimental to mainline, enabled by default
178 and removable for a build by defining DISABLE_EVENT.
180 JH/30 Updated DANE implementation code to current from Viktor Dukhovni.
182 JH/31 Fix bug with hosts_connection_nolog and named-lists which were wrongly
183 cached by the daemon.
185 JH/32 Move Redis support from Experimental to mainline, enabled for a build
186 by defining LOOKUP_REDIS. The libhiredis library is required.
188 JH/33 Bug 1748: Permit ACL dnslists= condition in non-smtp ACLs if explicit
189 keys are given for lookup.
191 JH/34 Bug 1192: replace the embedded copy of PolarSSL RSA routines in the DKIM
192 support, by using OpenSSL or GnuTLS library ones. This means DKIM is
193 only supported when built with TLS support. The PolarSSL SHA routines
194 are still used when the TLS library is too old for convenient support.
196 JH/35 Require SINGLE_DH_USE by default in OpenSSL (main config option
197 openssl_options), for security. OpenSSL forces this from version 1.1.0
198 server-side so match that on older versions.
200 JH/36 Bug 1778: longstanding bug in memory use by the ${run } expansion: A fresh
201 allocation for $value could be released as the expansion processing
202 concluded, but leaving the global pointer active for it.
204 JH/37 Bug 1769: Permit a VRFY ACL to override the default 252 response,
205 and to use the domains and local_parts ACL conditions.
207 JH/38 Fix cutthrough bug with body lines having a single dot. The dot was
208 incorrectly not doubled on cutthrough transmission, hence seen as a
209 body-termination at the receiving system - resulting in truncated mails.
210 Commonly the sender saw a TCP-level error, and retransmitted the message
211 via the normal store-and-forward channel. This could result in duplicates
212 received - but deduplicating mailstores were liable to retain only the
213 initial truncated version.
215 JH/39 Bug 1781: Fix use of DKIM private-keys having trailing '=' in the base-64.
217 JH/40 Fix crash in queryprogram router when compiled with EXPERIMENTAL_SRS.
219 JH/41 Bug 1792: Fix selection of headers to sign for DKIM: bottom-up. While
220 we're in there, support oversigning also; bug 1309.
222 JH/42 Bug 1796: Fix error logged on a malware scanner connection failure.
224 HS/04 Add support for keep_environment and add_environment options.
226 JH/43 Tidy coding issues detected by gcc --fsanitize=undefined. Some remain;
227 either intentional arithmetic overflow during PRNG, or testing config-
230 JH/44 Bug 1800: The combination of a -bhc commandline option and cutthrough
231 delivery resulted in actual delivery. Cancel cutthrough before DATA
234 JH/45 Fix cutthrough, when connection not opened by verify and target hard-
235 rejects a recipient: pass the reject to the originator.
237 JH/46 Multiple issues raised by Coverity. Some were obvious or plausible bugs.
238 Many were false-positives and ignorable, but it's worth fixing the
241 JH/47 Fix build on HP-UX and older Solaris, which need (un)setenv now also
242 for the new environment-manipulation done at startup. Move the routines
243 from being local to tls.c to being global via the os.c file.
245 JH/48 Bug 1807: Fix ${extract } for the numeric/3-string case. While preparsing
246 an extract embedded as result-arg for a map, the first arg for extract
247 is unavailable so we cannot tell if this is a numbered or keyed
248 extraction. Accept either.
254 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
257 JH/02 The smtp transport option "multi_domain" is now expanded.
259 JH/03 The smtp transport now requests PRDR by default, if the server offers
262 JH/04 Certificate name checking on server certificates, when exim is a client,
263 is now done by default. The transport option tls_verify_cert_hostnames
264 can be used to disable this per-host. The build option
265 EXPERIMENTAL_CERTNAMES is withdrawn.
267 JH/05 The value of the tls_verify_certificates smtp transport and main options
268 default to the word "system" to access the system default CA bundle.
269 For GnuTLS, only version 3.0.20 or later.
271 JH/06 Verification of the server certificate for a TLS connection is now tried
272 (but not required) by default. The verification status is now logged by
273 default, for both outbound TLS and client-certificate supplying inbound
276 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
279 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
280 Status Notification (bounce) messages are now MIME format per RFC 3464.
281 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
282 under the control of the dsn_advertise_hosts option, and routers may
283 have a dsn_lasthop option.
285 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
286 default, modifiable by a malware= option. The list separator for
287 the options can now be changed in the usual way. Bug 68.
289 JH/10 The smtp_receive_timeout main option is now expanded before use.
291 JH/11 The incoming_interface log option now also enables logging of the
292 local interface on delivery outgoing connections.
294 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
295 if the interface and destination host and port all match.
297 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
300 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
301 Patch from Andrew Lewis.
303 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
304 now supports optional time-restrictions, weighting, and priority
305 modifiers per server. Patch originally by <rommer@active.by>.
307 JH/16 The spamd_address main option now supports a mixed list of local
308 and remote servers. Remote servers can be IPv6 addresses, and
309 specify a port-range.
311 JH/17 Bug 68: The spamd_address main option now supports an optional
312 timeout value per server.
314 JH/18 Bug 1581: Router and transport options headers_add/remove can
315 now have the list separator specified.
317 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
320 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
323 JH/21 Support for the A6 type of dns record is withdrawn.
325 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
326 rather than the verbs used.
328 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
329 from 255 to 1024 chars.
331 JH/24 Verification callouts now attempt to use TLS by default.
333 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
334 are generic router options now. The defaults didn't change.
336 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
337 Original patch from Alexander Shikoff, worked over by JH.
339 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
340 files if ZCAT_COMMAND is not executable.
342 JH/26 Bug 1539: Add timout/retry options on dnsdb lookups.
344 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
346 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
347 Normally benign, it bites when the pair was led to by a CNAME;
348 modern usage is to not canoicalize the domain to a CNAME target
349 (and we were inconsistent anyway for A-only vs AAAA+A).
351 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
353 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
354 when evaluating $sender_host_dnssec.
356 JH/31 Check the HELO verification lookup for DNSSEC, adding new
357 $sender_helo_dnssec variable.
359 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
361 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
363 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
365 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
366 documented as working, but never had. Support all but $spam_report.
368 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
369 added for tls authenticator.
371 HS/03 Add perl_taintmode main config option
376 TL/01 When running the test suite, the README says that variables such as
377 no_msglog_check are global and can be placed anywhere in a specific
378 test's script, however it was observed that placement needed to be near
379 the beginning for it to behave that way. Changed the runtest perl
380 script to read through the entire script once to detect and set these
381 variables, reset to the beginning of the script, and then run through
382 the script parsing/test process like normal.
384 TL/02 The BSD's have an arc4random API. One of the functions to induce
385 adding randomness was arc4random_stir(), but it has been removed in
386 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
387 function when detected.
389 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
390 cause callback expansion.
392 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
393 syntax errors in an expansion can be treated as a string instead of
394 logging or causing an error, due to the internal use of bool_lax
395 instead of bool when processing it.
397 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
398 server certificates when making smtp deliveries.
400 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
402 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
404 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
406 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
407 Merged patch from Sebastian Wiedenroth.
409 JH/05 Fix results-pipe from transport process. Several recipients, combined
410 with certificate use, exposed issues where response data items split
411 over buffer boundaries were not parsed properly. This eventually
412 resulted in duplicates being sent. This issue only became common enough
413 to notice due to the introduction of conection certificate information,
414 the item size being so much larger. Found and fixed by Wolfgang Breyha.
416 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
417 size buffer was used, resulting in syntax errors when an expansion
420 JH/07 Add support for directories of certificates when compiled with a GnuTLS
421 version 3.3.6 or later.
423 JH/08 Rename the TPDA expermimental facility to Event Actions. The #ifdef
424 is EXPERIMENTAL_EVENT, the main-configuration and transport options
425 both become "event_action", the variables become $event_name, $event_data
426 and $event_defer_errno. There is a new variable $verify_mode, usable in
427 routers, transports and related events. The tls:cert event is now also
428 raised for inbound connections, if the main configuration event_action
431 TL/06 In test suite, disable OCSP for old versions of openssl which contained
432 early OCSP support, but no stapling (appears to be less than 1.0.0).
434 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
435 server certificate names available under the smtp transport option
436 "tls_verify_cert_hostname" now do not permit multi-component wildcard
439 JH/10 Time-related extraction expansions from certificates now use the main
440 option "timezone" setting for output formatting, and are consistent
441 between OpenSSL and GnuTLS compilations. Bug 1541.
443 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
444 encoded parameter in the incoming message. Bug 1558.
446 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
447 include certificate info, eximon was claiming there were spoolfile
450 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
452 JH/14 Log delivery-related information more consistently, using the sequence
453 "H=<name> [<ip>]" wherever possible.
455 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
456 are problematic for Debian distribution, omit them from the release
459 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
461 JH/16 Fix string representation of time values on 64bit time_t anchitectures.
464 JH/17 Fix a null-indirection in certextract expansions when a nondefault
465 output list separator was used.
470 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
471 checkers that were complaining about end of non-void function with no
474 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
475 This was a regression intruduced in 4.83 by another bugfix.
477 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
479 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
480 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
486 TF/01 Correctly close the server side of TLS when forking for delivery.
488 When a message was received over SMTP with TLS, Exim failed to clear up
489 the incoming connection properly after forking off the child process to
490 deliver the message. In some situations the subsequent outgoing
491 delivery connection happened to have the same fd number as the incoming
492 connection previously had. Exim would try to use TLS and fail, logging
493 a "Bad file descriptor" error.
495 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
496 utilities have not been installed.
498 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
499 temporary space as the ACL may create new global variables.
501 TL/01 LDAP support uses per connection or global context settings, depending
502 upon the detected version of the libraries at build time.
504 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
505 to extract and use the src ip:port in logging and expansions as if it
506 were a direct connection from the outside internet. PPv2 support was
507 updated based on HAProxy spec change in May 2014.
509 JH/02 Add ${listextract {number}{list}{success}{fail}}.
511 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
512 Properly escape header and check for NULL return.
514 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
517 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
519 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
520 characters in header names, implemented as a verify condition.
521 Contributed by Michael Fischer v. Mollard.
523 TL/05 Rename SPF condition results err_perm and err_temp to standardized
524 results permerror and temperror. Previous values are deprecated but
525 still accepted. In a future release, err_perm and err_temp will be
526 completely removed, which will be a backward incompatibility if the
527 ACL tests for either of these two old results. Patch contributed by
528 user bes-internal on the mailing list.
530 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
532 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
533 selectors, in both main and reject logs.
535 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
538 JH/07 Add malware type "sock" for talking to simple daemon.
540 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
542 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
543 routers/transports under cutthrough routing.
545 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
546 numbers. Touch up "bool" conditional to keep the same definition.
548 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
550 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
552 JH/12 Expand items in router/transport headers_add or headers_remove lists
553 individually rather than the list as a whole. Bug 1452.
555 Required for reasonable handling of multiple headers_ options when
556 they may be empty; requires that headers_remove items with embedded
557 colons must have them doubled (or the list-separator changed).
559 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
560 view the policy declared in the DMARC record. Currently, $dmarc_status
561 is a combined value of both the record presence and the result of the
564 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
566 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
567 dnslookup router and the smtp transport (applying to the forward
570 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
571 of ldap servers used for a specific lookup. Patch provided by Heiko
574 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
575 New variable $lookup_dnssec_authenticated for observability.
577 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
578 Patch submitted by Lars Timman.
580 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
582 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
583 Requires trusted mode and valid format message id, aborts otherwise.
584 Patch contributed by Heiko Schlichting.
586 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
587 certextract with support for various fields. Bug 1358.
589 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
590 is requested by default, modifiable by smtp transport option
593 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
594 operate on certificate variables to give certificate fingerprints
595 Also new ${sha256:cert_variable}.
597 JH/23 The PRDR feature is moved from being Experimental into the mainline.
599 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
600 Christian Aistleitner.
602 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
604 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
605 file. Patch from Wolfgang Breyha.
607 JH/25 Expand the coverage of the delivery $host and $host_address to
608 client authenticators run in verify callout. Bug 1476.
610 JH/26 Port service names are now accepted for tls_on_connect_ports, to
611 align with daemon_smtp_ports. Bug 72.
613 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
614 support and error reporting did not work properly.
616 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
617 and is readable. Patch from Andrew Colin Kissa.
619 TL/14 Enhance documentation of ${run expansion and how it parses the
620 commandline after expansion, particularly in the case when an
621 unquoted variable expansion results in an empty value.
623 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
625 PP/02 Fix internal collision of T_APL on systems which support RFC3123
626 by renaming away from it. Addresses GH issue 15, reported by
629 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
631 TL/15 SECURITY: prevent double expansion in math comparison functions
632 (can expand unsanitized data). Not remotely exploitable.
639 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
641 PP/02 Make -n do something, by making it not do something.
642 When combined with -bP, the name of an option is not output.
644 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
647 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
648 $sender_host_name and config options to manage this, and basic check
651 PP/05 DSCP support for outbound connections and control modifier for inbound.
653 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
654 (Only plugin which currently uses this is kerberos4, which nobody should
655 be using, but we should make it available and other future plugins might
656 conceivably use it, even though it would break NAT; stuff *should* be
657 using channel bindings instead).
659 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
660 name; added for Sendmail compatibility; requires admin caller.
661 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
662 just ignore it); requires trusted caller.
663 Also parse but ignore: -Ac -Am -X<logfile>
666 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
668 TL/02 Add +smtp_confirmation as a default logging option.
670 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
671 Patch by Magnus Holmgren from 2007-02-20.
673 TL/04 Bugzilla 1281 - Spec typo.
674 Bugzilla 1283 - Spec typo.
675 Bugzilla 1290 - Spec grammar fixes.
677 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
679 TL/06 Add Experimental DMARC support using libopendmarc libraries.
681 TL/07 Fix an out of order global option causing a segfault. Reported to dev
682 mailing list by by Dmitry Isaikin.
684 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
686 JH/02 Support "G" suffix to numbers in ${if comparisons.
688 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
690 NM/01 Bugzilla 1197 - Spec typo
691 Bugzilla 1196 - Spec examples corrections
693 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
695 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
696 gnutls_enable_pkcs11, but renamed to more accurately indicate its
699 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
700 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
702 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
703 "acl {{name}{arg}...}", and optional args on acl condition
706 JH/05 Permit multiple router/transport headers_add/remove lines.
708 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
710 JH/07 Avoid using a waiting database for a single-message-only transport.
711 Performance patch from Paul Fisher. Bugzilla 1262.
713 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
716 JH/09 Add $headers_added variable, with content from use of ACL modifier
717 add_header (but not yet added to the message). Bugzilla 199.
719 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
720 Pulled from Bugzilla 817 by Wolfgang Breyha.
722 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
724 (nb: this is the same fix as in Exim 4.80.1)
726 JH/11 Add A= logging on delivery lines, and a client_set_id option on
729 JH/12 Add optional authenticated_sender logging to A= and a log_selector
732 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
734 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
735 advertise SMTP AUTH mechanism to us, instead of a generic
736 protocol violation error. Also, make Exim more robust to bad
737 data from the Dovecot auth socket.
739 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
741 When a queue runner is handling a message, Exim first routes the
742 recipient addresses, during which it prunes them based on the retry
743 hints database. After that it attempts to deliver the message to
744 any remaining recipients. It then updates the hints database using
747 So if a recipient address works intermittently, it can get repeatedly
748 deferred at routing time. The retry hints record remains fresh so the
749 address never reaches the final cutoff time.
751 This is a fairly common occurrence when a user is bumping up against
752 their storage quota. Exim had some logic in its local delivery code
753 to deal with this. However it did not apply to per-recipient defers
754 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
756 This change adds a proper retry rule check during routing so that the
757 final cutoff time is checked against the message's age. We only do
758 this check if there is an address retry record and there is not a
759 domain retry record; this implies that previous attempts to handle
760 the address had the retry_use_local_parts option turned on. We use
761 this as an approximation for the destination being like a local
762 delivery, as in LMTP.
764 I suspect this new check makes the old local delivery cutoff check
765 redundant, but I have not verified this so I left the code in place.
767 TF/02 Correct gecos expansion when From: is a prefix of the username.
769 Test 0254 submits a message to Exim with the header
773 When I ran the test suite under the user fanf2, Exim expanded
774 the header to contain my full name, whereas it should have added
775 a Resent-Sender: header. It erroneously treats any prefix of the
776 username as equal to the username.
778 This change corrects that bug.
780 GF/01 DCC debug and logging tidyup
781 Error conditions log to paniclog rather than rejectlog.
782 Debug lines prefixed by "DCC: " to remove any ambiguity.
784 TF/03 Avoid unnecessary rebuilds of lookup-related code.
786 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
787 Bug spotted by Jeremy Harris; was flawed since initial commit.
788 Would have resulted in OCSP responses post-SNI triggering an Exim
789 NULL dereference and crash.
791 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
793 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
794 Bug detection, analysis and fix by Samuel Thibault.
795 Bugzilla 1331, Debian bug #698092.
797 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
799 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
800 Server implementation by Todd Lyons, client by JH.
801 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
802 config variable "prdr_enable" controls whether the server
803 advertises the facility. If the client requests PRDR a new
804 acl_data_smtp_prdr ACL is called once for each recipient, after
805 the body content is received and before the acl_smtp_data ACL.
806 The client is controlled by bolth of: a hosts_try_prdr option
807 on the smtp transport, and the server advertisement.
808 Default client logging of deliveries and rejections involving
809 PRDR are flagged with the string "PRDR".
811 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
812 fclose(). Diagnosis by Todd Lyons.
814 PP/17 Update configure.default to handle IPv6 localhost better.
815 Patch by Alain Williams (plus minor tweaks).
818 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
819 This is now consistent with GnuTLS, and is now documented: the
820 previous undocumented portable approach to treating the option as
821 unset was to force an expansion failure. That still works, and
822 an empty string is now equivalent.
824 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
825 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
826 not performing validation itself.
828 PP/20 Added force_command boolean option to pipe transport.
829 Patch from Nick Koston, of cPanel Inc.
831 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
834 TF/04 Added udpsend ACL modifer and hexquote expansion operator
836 PP/21 Fix eximon continuous updating with timestamped log-files.
837 Broken in a format-string cleanup in 4.80, missed when I repaired the
838 other false fix of the same issue.
839 Report and fix from Heiko Schlichting.
842 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
843 Report from Prashanth Katuri.
845 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
846 It's SecureTransport, so affects any MacOS clients which use the
847 system-integrated TLS libraries, including email clients.
849 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
850 using a MIME ACL for non-SMTP local injection.
851 Report and assistance in diagnosis by Warren Baker.
853 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
855 JH/16 Fix comparisons for 64b. Bugzilla 1385.
857 TL/09 Add expansion variable $authenticated_fail_id to keep track of
858 last id that failed so it may be referenced in subsequent ACL's.
860 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
863 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
864 ldap library initialization, allowing self-signed CA's to be
865 used. Also properly sets require_cert option later in code by
866 using NULL (global ldap config) instead of ldap handle (per
867 session). Bug diagnosis and testing by alxgomz.
869 TL/12 Enhanced documentation in the ratelimit.pl script provided in
870 the src/util/ subdirectory.
872 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
873 renamed to Transport Post Delivery Action by Jeremy Harris, as
876 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
877 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
878 redis_servers = needs to be configured which will be used by the redis
879 lookup. Patch from Warren Baker, of The Packet Hub.
881 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
883 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
884 hostname or reverse DNS when processing a host list. Used suggestions
885 from multiple comments on this bug.
887 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
889 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
890 Missed a few lines, added it to make the runtest require no keyboard
893 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
894 contains upper case chars. Make router use caseful_local_part.
896 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
897 support when GnuTLS has been built with p11-kit.
903 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
905 This, or similar/improved, will also be change PP/11 of 4.82.
911 PP/01 Handle short writes when writing local log-files.
912 In practice, only affects FreeBSD (8 onwards).
913 Bugzilla 1053, with thanks to Dmitry Isaikin.
915 NM/01 Bugzilla 949 - Documentation tweak
917 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
920 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
922 PP/02 Implemented gsasl authenticator.
924 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
926 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
927 `pkg-config foo` for cflags/libs.
929 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
930 with rest of GSASL and with heimdal_gssapi.
932 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
933 `pkg-config foo` for cflags/libs for the TLS implementation.
935 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
936 properties get this fed in as external SSF. A number of robustness
937 and debugging improvements to the cyrus_sasl authenticator.
939 PP/08 cyrus_sasl server now expands the server_realm option.
941 PP/09 Bugzilla 1214 - Log authentication information in reject log.
942 Patch by Jeremy Harris.
944 PP/10 Added dbmjz lookup type.
946 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
948 PP/12 MAIL args handles TAB as well as SP, for better interop with
949 non-compliant senders.
950 Analysis and variant patch by Todd Lyons.
952 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
953 Bug report from Lars Müller <lars@samba.org> (via SUSE),
954 Patch from Dirk Mueller <dmueller@suse.com>
956 PP/13 tls_peerdn now print-escaped for spool files.
957 Observed some $tls_peerdn in wild which contained \n, which resulted
958 in spool file corruption.
960 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
961 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
962 or write after TLS renegotiation, which otherwise led to messages
965 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
966 as a tracking header (ie: a signed header comes before the signature).
967 Patch from Wolfgang Breyha.
969 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
970 comma-sep list; embedded commas doubled.
972 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
974 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
976 Report and patch from Dmitry Banschikov.
978 PP/16 Removed "dont_insert_empty_fragments" fron "openssl_options".
979 Removed SSL_clear() after SSL_new() which led to protocol negotiation
980 failures. We appear to now support TLS1.1+ with Exim.
982 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
983 lets Exim select keys and certificates based upon TLS SNI from client.
984 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
985 before an outbound SMTP session. New log_selector, +tls_sni.
987 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
988 NULL dereference. Report and patch from Alun Jones.
990 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
991 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
992 Not seeing resolver debug output on NetBSD, but suspect this is a
993 resolver implementation change.
995 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
996 Left warnings. Added "eximon gdb" invocation mode.
998 PP/21 Defaulting "accept_8bitmime" to true, not false.
1000 PP/22 Added -bw for inetd wait mode support.
1002 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
1003 locate the relevant includes and libraries. Made this the default.
1005 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
1006 Bugzilla 1246, report and most of solution from Tomasz Kusy.
1008 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
1009 This may cause build issues on older platforms.
1011 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
1012 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
1013 gnutls_require_mac & gnutls_require_protocols (no longer supported).
1014 Added SNI support via GnuTLS too.
1015 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
1017 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
1019 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
1020 Applied second patch from Janne, implementing suggestion to default
1021 multiple-strings-in-record handling to match SPF spec.
1023 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
1025 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
1026 read-only, out of scope).
1027 Patch from Wolfgang Breyha, report from Stuart Northfield.
1029 PP/29 Fix three issues highlighted by clang analyser static analysis.
1030 Only crash-plausible issue would require the Cambridge-specific
1031 iplookup router and a misconfiguration.
1032 Report from Marcin Mirosław.
1034 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
1036 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
1037 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
1038 As part of this, removing so much warning spew let me fix some minor
1039 real issues in debug logging.
1041 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
1042 assignment on my part. Fixed.
1044 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
1045 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
1046 Janne Snabb (who went above and beyond: thank you).
1048 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
1049 string otherwise requires a connection and a bunch more work and it's
1050 relatively easy to get wrong. Should also expose TLS library linkage
1053 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
1054 64-bit ${eval} (JH/03).
1056 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
1057 GNU libc to support some of the 64-bit stuff, should not lead to
1058 conflicts. Defined before os.h is pulled in, so if a given platform
1059 needs to override this, it can.
1061 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
1062 protection layer was required, which is not implemented.
1063 Bugzilla 1254, patch from Wolfgang Breyha.
1065 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
1066 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
1067 tls_dhparam take prime identifiers. Also unbreak combination of
1068 OpenSSL+DH_params+TLSSNI.
1070 PP/39 Disable SSLv2 by default in OpenSSL support.
1076 PP/01 Solaris build fix for Oracle's LDAP libraries.
1077 Bugzilla 1109, patch from Stephen Usher.
1079 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
1081 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
1084 TF/02 Fix a couple more cases where we did not log the error message
1085 when unlink() failed. See also change 4.74-TF/03.
1087 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
1088 lock up or crash if it happened to be inside a call to libc when it
1089 got a SIGUSR1 from exiwhat.
1091 The SIGUSR1 handler appends the current process status to the process
1092 log which is later printed by exiwhat. It used to use the general
1093 purpose logging code to do this, but several functions it calls are
1094 not safe for signals.
1096 The new output code in the SIGUSR1 handler is specific to the process
1097 log, and simple enough that it's easy to inspect for signal safety.
1098 Removing some special cases also simplifies the general logging code.
1099 Removing the spurious timestamps from the process log simplifies
1102 TF/04 Improved ratelimit ACL condition.
1104 The /noupdate option has been deprecated in favour of /readonly which
1105 has clearer semantics. The /leaky, /strict, and /readonly update modes
1106 are mutually exclusive. The update mode is no longer included in the
1107 database key; it just determines when the database is updated. (This
1108 means that when you upgrde Exim will forget old rate measurements.)
1110 Exim now checks that the per_* options are used with an update mode that
1111 makes sense for the current ACL. For example, when Exim is processing a
1112 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
1113 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
1114 must specify per_mail/readonly. If you omit the update mode it defaults to
1115 /leaky where that makes sense (as before) or /readonly where required.
1117 The /noupdate option is now undocumented but still supported for
1118 backwards compatibility. It is equivalent to /readonly except that in
1119 ACLs where /readonly is required you may specify /leaky/noupdate or
1120 /strict/noupdate which are treated the same as /readonly.
1122 A useful new feature is the /count= option. This is a generalization
1123 of the per_byte option, so that you can measure the throughput of other
1124 aggregate values. For example, the per_byte option is now equivalent
1125 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
1127 The per_rcpt option has been generalized using the /count= mechanism
1128 (though it's more complicated than the per_byte equivalence). When it is
1129 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
1130 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
1131 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
1132 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
1133 Note that using per_rcpt with a non-readonly update mode in more than
1134 one ACL will cause the recipients to be double-counted. (The per_mail
1135 and per_byte options don't have this problem.)
1137 The handling of very low rates has changed slightly. If the computed rate
1138 is less than the event's count (usually one) then this event is the first
1139 after a long gap. In this case the rate is set to the same as this event's
1140 count, so that the first message of a spam run is counted properly.
1142 The major new feature is a mechanism for counting the rate of unique
1143 events. The new per_addr option counts the number of different
1144 recipients that someone has sent messages to in the last time period. It
1145 behaves like per_rcpt if all the recipient addresses are different, but
1146 duplicate recipient addresses do not increase the measured rate. Like
1147 the /count= option this is a general mechanism, so the per_addr option
1148 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
1149 example, measure the rate that a client uses different sender addresses
1150 with the options per_mail/unique=$sender_address. There are further
1151 details in the main documentation.
1153 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
1155 TF/06 Removed a few PCRE remnants.
1157 TF/07 Automatically extract Exim's version number from tags in the git
1158 repository when doing development or release builds.
1160 PP/02 Raise smtp_cmd_buffer_size to 16kB.
1161 Bugzilla 879. Patch from Paul Fisher.
1163 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
1164 Heavily based on revision 40f9a89a from Simon Arlott's tree.
1167 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
1169 PP/05 Variable $av_failed, true if the AV scanner deferred.
1170 Bugzilla 1078. Patch from John Horne.
1172 PP/06 Stop make process more reliably on build failure.
1173 Bugzilla 1087. Patch from Heiko Schlittermann.
1175 PP/07 Make maildir_use_size_file an _expandable_ boolean.
1176 Bugzilla 1089. Patch from Heiko Schlittermann.
1178 PP/08 Handle ${run} returning more data than OS pipe buffer size.
1179 Bugzilla 1131. Patch from Holger Weiß.
1181 PP/09 Handle IPv6 addresses with SPF.
1182 Bugzilla 860. Patch from Wolfgang Breyha.
1184 PP/10 GnuTLS: support TLS 1.2 & 1.1.
1186 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
1189 PP/11 match_* no longer expand right-hand-side by default.
1190 New compile-time build option, EXPAND_LISTMATCH_RHS.
1191 New expansion conditions, "inlist", "inlisti".
1193 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
1195 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
1197 PP/14 fix log_write() format string regression from TF/03.
1198 Bugzilla 1152. Patch from Dmitry Isaikin.
1204 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
1206 PP/02 Harmonised TLS library version reporting; only show if debugging.
1207 Layout now matches that introduced for other libraries in 4.74 PP/03.
1209 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
1211 PP/04 New "dns_use_edns0" global option.
1213 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
1216 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
1217 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
1219 TK/01 Updated PolarSSL code to 0.14.2.
1220 Bugzilla 1097. Patch from Andreas Metzler.
1222 PP/07 Catch divide-by-zero in ${eval:...}.
1223 Fixes bugzilla 1102.
1225 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
1228 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
1229 format-string attack -- SECURITY: remote arbitrary code execution.
1231 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
1232 time unintentionally subject to list matching rules, letting the header
1233 cause arbitrary Exim lookups (of items which can occur in lists, *not*
1234 arbitrary string expansion). This allowed for information disclosure.
1236 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
1237 INT_MIN/-1 -- value coerced to INT_MAX.
1243 NM/01 Workround for PCRE version dependency in version reporting
1246 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
1247 This fixes portability to compilers other than gcc, notably
1248 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
1250 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
1251 makefiles for portability to HP-UX and POSIX correctness.
1253 PP/01 Permit LOOKUP_foo enabling on the make command-line.
1254 Also via indented variable definition in the Makefile.
1255 (Debugging by Oliver Heesakkers).
1257 PP/02 Restore caching of spamd results with expanded spamd_address.
1258 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
1260 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
1261 Improves build reliability. Fix from: Frank Elsner
1263 NM/02 Fix wide character breakage in the rfc2047 coding
1264 Fixes bug 1064. Patch from Andrey N. Oktyabrski
1266 NM/03 Allow underscore in dnslist lookups
1267 Fixes bug 1026. Patch from Graeme Fowler
1269 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
1270 Code patches from Adam Ciarcinski of NetBSD.
1272 NM/04 Fixed exiqgrep to cope with mailq missing size issue
1275 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
1276 is logged, to avoid truncation. Patch from John Horne.
1278 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
1279 Patch from Jakob Hirsch.
1281 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
1282 SQL string expansion failure details.
1283 Patch from Andrey Oktyabrski.
1285 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
1286 Patch from Simon Arlott.
1288 PP/09 New lookups functionality failed to compile on old gcc which rejects
1289 extern declarations in function scope.
1290 Patch from Oliver Fleischmann
1292 PP/10 Use sig_atomic_t for flags set from signal handlers.
1293 Check getgroups() return and improve debugging.
1294 Fixed developed for diagnosis in bug 927 (which turned out to be
1297 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
1298 Patch from Mark Zealey.
1300 PP/12 Bugzilla 1056: Improved spamd server selection.
1301 Patch from Mark Zealey.
1303 PP/13 Bugzilla 1086: Deal with maildir quota file races.
1304 Based on patch from Heiko Schlittermann.
1306 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
1307 Patch from Uwe Doering, sign-off by Michael Haardt.
1309 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
1310 variable declaration deep within a block. Bug and patch from
1313 PP/15 lookups-Makefile IRIX compatibilty coercion.
1315 PP/16 Make DISABLE_DKIM build knob functional.
1317 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
1318 Patch by Simon Arlott
1320 TF/03 Fix valgrind.h portability to C89 compilers that do not support
1321 variable argument macros. Our copy now differs from upstream.
1327 TF/01 Failure to get a lock on a hints database can have serious
1328 consequences so log it to the panic log.
1330 TF/02 Log LMTP confirmation messages in the same way as SMTP,
1331 controlled using the smtp_confirmation log selector.
1333 TF/03 Include the error message when we fail to unlink a spool file.
1335 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
1336 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
1337 for maintaining out-of-tree patches for some time.
1339 PP/01 Bugzilla 139: Documentation and portability issues.
1340 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
1341 Handle per-OS dynamic-module compilation flags.
1343 PP/02 Let /dev/null have normal permissions.
1344 The 4.73 fixes were a little too stringent and complained about the
1345 permissions on /dev/null. Exempt it from some checks.
1346 Reported by Andreas M. Kirchwitz.
1348 PP/03 Report version information for many libraries, including
1349 Exim version information for dynamically loaded libraries. Created
1350 version.h, now support a version extension string for distributors
1351 who patch heavily. Dynamic module ABI change.
1353 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
1354 privilege escalation vulnerability whereby the Exim run-time user
1355 can cause root to append content of the attacker's choosing to
1358 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
1361 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
1362 If dropping privileges for untrusted macros, we disabled normal logging
1363 on the basis that it would fail; for the Exim run-time user, this is not
1364 the case, and it resulted in successful deliveries going unlogged.
1365 Fixed. Reported by Andreas Metzler.
1371 PP/01 Date: & Message-Id: revert to normally being appended to a message,
1372 only prepend for the Resent-* case. Fixes regression introduced in
1373 Exim 4.70 by NM/22 for Bugzilla 607.
1375 PP/02 Include check_rfc2047_length in configure.default because we're seeing
1376 increasing numbers of administrators be bitten by this.
1378 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
1380 PP/03 Bugzilla 994: added openssl_options main configuration option.
1382 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
1384 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
1386 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
1388 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
1389 without a peer certificate, leading to a segfault because of an
1390 assumption that peers always have certificates. Be a little more
1391 paranoid. Problem reported by Martin Tscholak.
1393 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
1394 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
1395 NB: ClamAV planning to remove STREAM in "middle of 2010".
1396 CL also introduces -bmalware, various -d+acl logging additions and
1397 more caution in buffer sizes.
1399 PP/09 Implemented reverse_ip expansion operator.
1401 PP/10 Bugzilla 937: provide a "debug" ACL control.
1403 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
1405 PP/12 Bugzilla 973: Implement --version.
1407 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
1409 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
1411 PP/15 Bugzilla 816: support multiple condition rules on Routers.
1413 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
1414 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
1415 ignore trailing whitespace.
1417 JJ/02 prevent non-panic DKIM error from being sent to paniclog
1419 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
1422 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
1423 Notification from Dr Andrew Aitchison.
1425 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
1426 ExtendedDetectionInfo response format.
1427 Notification from John Horne.
1429 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
1432 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
1433 XSL and documented dependency on system catalogs, with examples of how
1436 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
1439 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
1440 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
1441 configuration file which is writeable by the Exim user or group.
1443 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
1444 of configuration files to cover files specified with the -C option if
1445 they are going to be used with root privileges, not just the default
1448 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
1449 option (effectively making it always true).
1451 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
1452 files to be used while preserving root privileges.
1454 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
1455 that rogue child processes cannot use them.
1457 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
1458 run-time user, instead of root.
1460 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
1461 Exim run-time user without dropping privileges.
1463 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
1464 result string, instead of calling string_vformat() twice with the same
1467 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
1468 for other users. Others should always drop root privileges if they use
1469 -C on the command line, even for a whitelisted configure file.
1471 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
1473 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
1479 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
1480 $data_path, and $header_path variables; fixed documentation bugs and
1483 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
1484 exipick to access non-standard spools, including the "frozen" queue
1487 NM/01 Bugzilla 965: Support mysql stored procedures.
1488 Patch from Alain Williams
1490 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
1492 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
1493 Patch from Andreas Metzler
1495 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
1496 Patch from Kirill Miazine
1498 NM/05 Bugzilla 671: Added umask to procmail example.
1500 JJ/03 installed exipick 20100323.0, fixing doc bug
1502 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
1503 directory. Notification and patch from Dan Rosenberg.
1505 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
1507 TK/02 Improve log output when DKIM signing operation fails.
1509 MH/01 Treat the transport option dkim_domain as a colon separated
1510 list, not as a single string, and sign the message with each element,
1511 omitting multiple occurences of the same signer.
1513 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
1514 Bugzilla 985, 986. Patch by Simon Arlott
1516 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
1517 Patch by Simon Arlott
1519 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
1520 MBX locking. Notification from Dan Rosenberg.
1526 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
1528 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
1530 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
1532 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
1534 NM/04 Bugzilla 915: Flag broken perl installation during build.
1540 TK/01 Added patch by Johannes Berg that expands the main option
1541 "spamd_address" if it starts with a dollar sign.
1543 TK/02 Write list of recipients to X-Envelope-Sender header when building
1544 the mbox-format spool file for content scanning (suggested by Jakob
1547 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
1548 (http://www.dcc-servers.net/) support via dccifd. Activated by
1549 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
1551 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
1552 by Mark Daniel Reidel <mr@df.eu>.
1554 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
1555 When building exim an external PCRE library is now needed -
1556 PCRE is a system library on the majority of modern systems.
1557 See entry on PCRE_LIBS in EDITME file.
1559 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
1560 conversation. Added nologin parameter to request.
1561 Patch contributed by Kirill Miazine.
1563 TF/01 Do not log submission mode rewrites if they do not change the address.
1565 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
1567 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
1568 log files in place. Contributed by Roberto Lima.
1570 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
1572 TF/03 Bugzilla 615: When checking the local_parts router precondition
1573 after a local_part_suffix or local_part_prefix option, Exim now
1574 does not use the address's named list lookup cache, since this
1575 contains cached lookups for the whole local part.
1577 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
1578 Robert Millan. Documentation is in experimental-spec.txt.
1580 TF/04 Bugzilla 668: Fix parallel build (make -j).
1582 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
1584 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
1585 Patch provided by Jan Srzednicki.
1587 TF/05 Leading white space used to be stripped from $spam_report which
1588 wrecked the formatting. Now it is preserved.
1590 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
1591 that they are available at delivery time.
1593 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
1595 TF/08 TLS error reporting now respects the incoming_interface and
1596 incoming_port log selectors.
1598 TF/09 Produce a more useful error message if an SMTP transport's hosts
1599 setting expands to an empty string.
1601 NM/06 Bugzilla 744: EXPN did not work under TLS.
1602 Patch provided by Phil Pennock.
1604 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
1605 Patch provided by Richard Godbee.
1607 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
1608 acl_smtp_notquit, added index entry.
1610 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
1611 Patch provided by Eugene Bujak.
1613 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
1614 accept(). Patch provided by Maxim Dounin.
1616 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
1617 Patch provided by Phil Pennock.
1619 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
1621 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
1622 Patch provided by Brad "anomie" Jorsch.
1624 NM/14 Bugzilla 622: Added timeout setting to transport filter.
1625 Patch provided by Dean Brooks.
1627 TK/05 Add native DKIM support (does not depend on external libraries).
1629 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
1630 Patch provided by Graeme Fowler.
1632 NM/16 Bugzilla 851: Documentation example syntax fix.
1634 NM/17 Changed NOTICE file to remove references to embedded PCRE.
1636 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
1639 NM/19 Bugzilla 745: TLS version reporting.
1640 Patch provided by Phil Pennock.
1642 NM/20 Bugzilla 167: bool: condition support.
1643 Patch provided by Phil Pennock.
1645 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
1646 clients. Patch provided by Phil Pennock.
1648 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
1649 Patch provided by Brad "anomie" Jorsch.
1651 NM/23 Bugzilla 687: Fix misparses in eximstats.
1652 Patch provided by Heiko Schlittermann.
1654 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
1655 Patch provided by Heiko Schlittermann.
1657 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
1658 plus update to original patch.
1660 NM/26 Bugzilla 799: Documentation correction for ratelimit.
1662 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
1663 Patch provided by David Brownlee.
1665 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
1667 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
1669 NM/30 Bugzilla 888: TLS documentation bugfixes.
1671 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
1673 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
1674 Unlike the original bugzilla I have changed all shell scripts in src tree.
1676 NM/33 Bugzilla 898: Transport filter timeout fix.
1677 Patch by Todd Rinaldo.
1679 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
1680 Patch by Serge Demonchaux.
1682 NM/35 Bugzilla 39: Base64 decode bug fixes.
1683 Patch by Jakob Hirsch.
1685 NM/36 Bugzilla 909: Correct connect() call in dcc code.
1687 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
1689 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
1691 NM/39 Bugzilla 911: Fixed MakeLinks build script.
1697 TK/01 Add preliminary DKIM support. Currently requires a forked version of
1698 ALT-N's libdkim that I have put here:
1699 http://duncanthrax.net/exim-experimental/
1701 Note to Michael Haardt: I had to rename some vars in sieve.c. They
1702 were called 'true' and it seems that C99 defines that as a reserved
1703 keyword to be used with 'bool' variable types. That means you could
1704 not include C99-style headers which use bools without triggering
1705 build errors in sieve.c.
1707 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
1708 as mailq or other aliases. Changed the --help handling significantly
1709 to do whats expected. exim_usage() emits usage/help information.
1711 SC/01 Added the -bylocaldomain option to eximstats.
1713 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
1715 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
1717 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
1723 PH/01 Another patch from the Sieve maintainer.
1725 PH/02 When an IPv6 address is converted to a string for single-key lookup
1726 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
1727 dots are used instead of colons so that keys in lsearch files need not
1728 contain colons. This was done some time before quoting was made available
1729 in lsearch files. However, iplsearch files do require colons in IPv6 keys
1730 (notated using the quote facility) so as to distinguish them from IPv4
1731 keys. This meant that lookups for IP addresses in host lists did not work
1732 for iplsearch lookups.
1734 This has been fixed by arranging for IPv6 addresses to be expressed with
1735 colons if the lookup type is iplsearch. This is not incompatible, because
1736 previously such lookups could never work.
1738 The situation is now rather anomolous, since one *can* have colons in
1739 ordinary lsearch keys. However, making the change in all cases is
1740 incompatible and would probably break a number of configurations.
1742 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
1745 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
1746 conversion specification without a maximum field width, thereby enabling
1747 a rogue spamd server to cause a buffer overflow. While nobody in their
1748 right mind would setup Exim to query an untrusted spamd server, an
1749 attacker that gains access to a server running spamd could potentially
1750 exploit this vulnerability to run arbitrary code as the Exim user.
1752 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
1753 $primary_hostname instead of what libspf2 thinks the hosts name is.
1755 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
1756 a directory entry by the name of the lookup key. Previously, if a
1757 symlink pointed to a non-existing file or a file in a directory that
1758 Exim lacked permissions to read, a lookup for a key matching that
1759 symlink would fail. Now it is enough that a matching directory entry
1760 exists, symlink or not. (Bugzilla 503.)
1762 PH/03 The body_linecount and body_zerocount variables are now exported in the
1765 PH/04 Added the $dnslist_matched variable.
1767 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
1768 This means they are set thereafter only if the connection becomes
1771 PH/06 Added the client_condition to authenticators so that some can be skipped
1772 by clients under certain conditions.
1774 PH/07 The error message for a badly-placed control=no_multiline_responses left
1775 "_responses" off the end of the name.
1777 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
1779 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
1780 (without spaces) instead of just copying the configuration text.
1782 PH/10 Added the /noupdate option to the ratelimit ACL condition.
1784 PH/11 Added $max_received_linelength.
1786 PH/12 Added +ignore_defer and +include_defer to host lists.
1788 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
1789 way in which PCRE > 7.0 is built.
1791 PH/14 Implemented queue_only_load_latch.
1793 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
1794 MAIL command. The effect was to mangle the value on 64-bit systems.
1796 PH/16 Another patch from the Sieve maintainer.
1798 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
1800 PH/18 If a system quota error occurred while trying to create the file for
1801 a maildir delivery, the message "Mailbox is full" was not appended to the
1802 bounce if the delivery eventually timed out. Change 4.67/27 below applied
1803 only to a quota excession during the actual writing of the file.
1805 PH/19 It seems that peer DN values may contain newlines (and other non-printing
1806 characters?) which causes problems in log lines. The DN values are now
1807 passed through string_printing() before being added to log lines.
1809 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
1810 and InterBase are left for another time.)
1812 PH/21 Added message_body_newlines option.
1814 PH/22 Guard against possible overflow in moan_check_errorcopy().
1816 PH/23 POSIX allows open() to be a macro; guard against that.
1818 PH/24 If the recipient of an error message contained an @ in the local part
1819 (suitably quoted, of course), incorrect values were put in $domain and
1820 $local_part during the evaluation of errors_copy.
1826 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
1827 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
1830 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
1831 issue a MAIL command.
1833 PH/02 In an ACL statement such as
1835 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
1837 if a client was not listed at all, or was listed with a value other than
1838 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
1839 the condition was not true (as it should be), so access was not denied.
1840 The bug was that the ! inversion was incorrectly passed on to the second
1841 item. This has been fixed.
1843 PH/03 Added additional dnslists conditions == and =& which are different from
1844 = and & when the dns lookup returns more than one IP address.
1846 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
1847 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
1849 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
1850 FSYNC, which compiles an option called disable_fsync that allows for
1851 bypassing fsync(). The documentation is heavily laced with warnings.
1853 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
1855 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
1856 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
1857 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
1858 including adding "make clean"; (3) Added -fPIC when compiling the test
1859 dynamically loaded module, to get rid of a warning.
1861 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
1862 message fails, move_frozen_messages = true and ignore_bounce_errors_after
1863 = 0s. The bug is otherwise harmless.
1865 PH/07 There was a bug in the dovecot authenticator such that the value of
1866 $auth1 could be overwritten, and so not correctly preserved, after a
1867 successful authentication. This usually meant that the value preserved by
1868 the server_setid option was incorrect.
1870 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
1872 PH/09 Installed PCRE release 7.0.
1874 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
1875 run for batched SMTP input. It is now run at the start of every message
1876 in the batch. While fixing this I discovered that the process information
1877 (output by running exiwhat) was not always getting set for -bs and -bS
1878 input. This is fixed, and it now also says "batched" for BSMTP.
1880 PH/11 Added control=no_pipelining.
1882 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
1883 patch, slightly modified), and move the expansion of helo_data till after
1884 the connection is made in the smtp transport (so it can use these
1887 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
1889 PH/14 Added log_selector = +pid.
1891 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
1893 PH/16 Add ${if forany and ${if forall.
1895 PH/17 Added dsn_from option to vary the From: line in DSNs.
1897 PH/18 Flush SMTP output before performing a callout, unless control =
1898 no_callout_flush is set.
1900 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
1901 was true (the default) a successful delivery failed to delete the retry
1902 item, thus causing premature timeout of the address. The bug is now
1905 PH/20 Added hosts_avoid_pipelining to the smtp transport.
1907 PH/21 Long custom messages for fakedefer and fakereject are now split up
1908 into multiline reponses in the same way that messages for "deny" and
1909 other ACL rejections are.
1911 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
1912 with slight modification.
1914 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
1915 draft, changing the syntax and factoring some duplicate code".
1917 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
1918 for deliveries of the second and subsequent messages over the same SMTP
1921 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
1922 ${reduce, with only minor "tidies".
1924 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
1926 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
1927 expansion side effects.
1929 PH/27 When a message times out after an over-quota error from an Exim-imposed
1930 quota, the bounce message says "mailbox is full". This message was not
1931 being given when it was a system quota that was exceeded. It now should
1934 MH/03 Made $recipients available in local_scan(). local_scan() already has
1935 better access to the recipient list through recipients_list[], but
1936 $recipients can be useful in postmaster-provided expansion strings.
1938 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
1939 in the case of a MAIL command with additional options following the
1940 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
1941 were accidentally chopped off.
1943 PH/29 SMTP synchronization checks are implemented when a command is read -
1944 there is a check that no more input is waiting when there shouldn't be
1945 any. However, for some commands, a delay in an ACL can mean that it is
1946 some time before the response is written. In this time, more input might
1947 arrive, invalidly. So now there are extra checks after an ACL has run for
1948 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
1949 pipelining has not been advertised.
1951 PH/30 MH's patch to allow iscntrl() characters to be list separators.
1953 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
1954 returned in the SMTP response when smtp_return_error_details was false.
1955 This has been fixed.
1957 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
1958 instead of the C I/O that was originally supplied, because problems were
1959 reported on Solaris.
1961 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
1962 Exim which did not show up earlier: it was assuming that a call to
1963 SSL_CTX_set_info_callback() might give an error value. In fact, there is
1964 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
1965 was a macro that became an assignment, so it seemed to work. This has
1966 changed to a proper function call with a void return, hence the compile
1967 error. Exim's code has been fixed.
1969 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
1972 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
1974 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
1976 PH/37 If a message is not accepted after it has had an id assigned (e.g.
1977 because it turns out to be too big or there is a timeout) there is no
1978 "Completed" line in the log. When some messages of this type were
1979 selected by exigrep, they were listed as "not completed". Others were
1980 picked up by some special patterns. I have improved the selection
1981 criteria to be more general.
1983 PH/38 The host_find_failed option in the manualroute router can now be set
1984 to "ignore", to completely ignore a host whose IP address cannot be
1985 found. If all hosts are ignored, the behaviour is controlled by the new
1986 host_all_ignored option.
1988 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
1989 homing or because of multiple MX records with /mx) generated more than
1990 one IP address, and the following item turned out to be the local host,
1991 all the secondary addresses of the first item were incorrectly removed
1992 from the list, along with the local host and any following hosts (which
1993 is what is supposed to happen).
1995 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
1996 whoever called Exim into the -H file. In the case of the daemon it was
1997 behaving confusingly. When first started, it used values for whoever
1998 started the daemon, but after a SIGHUP it used the Exim user (because it
1999 calls itself on a restart). I have changed the code so that it now always
2002 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
2003 message are rejected with the same error (e.g. no authentication or bad
2004 sender address), and a DATA command is nevertheless sent (as can happen
2005 with PIPELINING or a stupid MUA), the error message that was given to the
2006 RCPT commands is included in the rejection of the DATA command. This is
2007 intended to be helpful for MUAs that show only the final error to their
2010 PH/42 Another patch from the Sieve maintainer.
2012 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
2013 Eximstats - Fixed some broken HTML links and added missing column headers
2015 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
2016 columns for Rejects, Temp Rejects, Ham, and Spam rows.
2018 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
2020 PH/43 Yet another patch from the Sieve maintainer.
2022 PH/44 I found a way to check for a TCP/IP connection going away before sending
2023 the response to the final '.' that terminates a message, but only in the
2024 case where the client has not sent further data following the '.'
2025 (unfortunately, this is allowed). However, in many cases there won't be
2026 any further data because there won't be any more messages to send. A call
2027 to select() can be used: if it shows that the input is "ready", there is
2028 either input waiting, or the socket has been closed. An attempt to read
2029 the next input character can distinguish the two cases. Previously, Exim
2030 would have sent an OK response which the client would never have see.
2031 This could lead to message repetition. This fix should cure that, at
2032 least in a lot of common cases.
2034 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
2035 advertised in response to EHLO.
2041 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
2042 fixed by 4.65/MH/01 (is this a record?) are fixed:
2044 (i) An empty string was always treated as zero by the numeric comparison
2045 operators. This behaviour has been restored.
2047 (ii) It is documented that the numeric comparison operators always treat
2048 their arguments as decimal numbers. This was broken in that numbers
2049 starting with 0 were being interpreted as octal.
2051 While fixing these problems I realized that there was another issue that
2052 hadn't been noticed. Values of message_size_limit (both the global option
2053 and the transport option) were treated as octal if they started with 0.
2054 The documentation was vague. These values are now always treated as
2055 decimal, and I will make that clear in the documentation.
2061 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
2062 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
2065 MH/01 Don't check that the operands of numeric comparison operators are
2066 integers when their expansion is in "skipping" mode (fixes bug
2067 introduced by 4.64-PH/07).
2069 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
2070 child addresses, Exim now panics and dies. Previously, because the count
2071 is held in a short int, deliveries were likely to be lost. As such a
2072 large number of recipients for a single message is ridiculous
2073 (performance will be very, very poor), I have chosen to impose a limit
2074 rather than extend the field.
2080 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
2081 leftover -K file (the existence of which was triggered by #402).
2082 While we were at it, introduced process PID as part of the -K
2083 filename. This should rule out race conditions when creating
2086 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
2087 processing considerably. Previous code took too long for large mails,
2088 triggering a timeout which in turn triggers #401.
2090 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
2091 in the DK code in transports.c. sendfile() is not really portable,
2092 hence the _LINUX specificness.
2094 TF/01 In the add_headers option to the mail command in an Exim filter,
2095 there was a bug that Exim would claim a syntax error in any
2096 header after the first one which had an odd number of characters
2099 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
2100 callout verification, Exim cached a "reject" for the entire domain. This
2101 is correct for most verifications, but it is not correct for a recipient
2102 verification with use_sender or use_postmaster set, because in that case
2103 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
2104 case of MAIL FROM:<> rejection from other early rejections (e.g.
2105 rejection of HELO). When verifying a recipient using a non-null MAIL
2106 address, the cache is ignored if it shows MAIL FROM:<> rejection.
2107 Whatever the result of the callout, the value of the domain cache is
2108 left unchanged (for any other kind of callout, getting as far as trying
2109 RCPT means that the domain itself is ok).
2111 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
2114 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
2115 manifest itself as EPIPE rather than ECONNECT. When tidying away a
2116 session, the daemon ignores ECONNECT errors and logs others; it now
2117 ignores EPIPE as well.
2119 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
2120 (quoted-printable decoding).
2122 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
2123 later the small subsequent patch to fix an introduced bug.
2125 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
2127 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
2129 PH/08 An error is now given if message_size_limit is specified negative.
2131 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
2132 to be given (somewhat) arbitrary names.
2134 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
2137 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
2138 miscellaneous code fixes
2140 PH/10 Added the log_reject_target ACL modifier to specify where to log
2143 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
2144 hostname. This is wrong, because it relates to the incoming message (and
2145 probably the interface on which it is arriving) and not to the outgoing
2146 callout (which could be using a different interface). This has been
2147 changed to use the value of the helo_data option from the smtp transport
2148 instead - this is what is used when a message is actually being sent. If
2149 there is no remote transport (possible with a router that sets up host
2150 addresses), $smtp_active_hostname is used.
2152 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
2153 tweaks were necessary in order to get it to work (see also 21 below):
2154 (a) The code assumed that strncpy() returns a negative number on buffer
2155 overflow, which isn't the case. Replaced with Exim's string_format()
2157 (b) There were several signed/unsigned issues. I just did the minimum
2158 hacking in of casts. There is scope for a larger refactoring.
2159 (c) The code used strcasecmp() which is not a standard C function.
2160 Replaced with Exim's strcmpic() function.
2161 (d) The code set only $1; it now sets $auth1 as well.
2162 (e) A simple test gave the error "authentication client didn't specify
2163 service in request". It would seem that Dovecot has changed its
2164 interface. Fortunately there's a specification; I followed it and
2165 changed what the client sends and it appears to be working now.
2167 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
2170 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
2171 address A is aliased to B and C, where B exists and C does not. Without
2172 -v the output is "A verified" because verification stops after a
2173 successful redirection if more than one address is generated. However,
2174 with -v the child addresses are also verified. Exim was outputting "A
2175 failed to verify" and then showing the successful verification for C,
2176 with its parentage. It now outputs "B failed to verify", showing B's
2177 parentage before showing the successful verification of C.
2179 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
2180 look up a TXT record in a specific list after matching in a combined
2183 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
2184 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
2185 they consult the DNS. I had assumed they would set it the way they
2186 wanted; and indeed my experiments on Linux seem to show that in some
2187 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
2188 To be on the safe side, however, I have now made the interface to
2189 host_find_byname() similar to host_find_bydns(), with an argument
2190 containing the DNS resolver options. The host_find_byname() function now
2191 sets these options at its start, just as host_find_bydns() does. The smtp
2192 transport options dns_qualify_single and dns_search_parents are passed to
2193 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
2194 of host_find_byname() use the default settings of RES_DEFNAMES
2195 (qualify_single) but not RES_DNSRCH (search_parents).
2197 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
2198 spool_read_header() do less string testing, by means of a preliminary
2199 switch on the second character of optional "-foo" lines. (This is
2200 overdue, caused by the large number of possibilities that now exist.
2201 Originally there were few.) While I was there, I also converted the
2202 str(n)cmp tests so they don't re-test the leading "-" and the first
2203 character, in the hope this might squeeze out yet more improvement.
2205 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
2206 flag allowing group syntax was set by the header_syntax check but not
2207 turned off, possible causing trouble later; (2) The flag was not being
2208 set at all for the header_verify test, causing "group"-style headers to
2209 be rejected. I have now set it in this case, and also caused header_
2210 verify to ignore an empty address taken from a group. While doing this, I
2211 came across some other cases where the code for allowing group syntax
2212 while scanning a header line wasn't quite right (mostly, not resetting
2213 the flag correctly in the right place). These bugs could have caused
2214 trouble for malformed header lines. I hope it is now all correct.
2216 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
2217 with the "reply" argument non-NULL. The code, however (which originally
2218 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
2219 but it didn't always do it. This confused somebody who was copying the
2220 code for some other use. I have removed all the tests.
2222 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
2223 feature that was used to support insecure browsers during the U.S. crypto
2224 embargo. It requires special client support, and Exim is probably the
2225 only MTA that supported it -- and would never use it because real RSA is
2226 always available. This code has been removed, because it had the bad
2227 effect of slowing Exim down by computing (never used) parameters for the
2228 RSA_EXPORT functionality.
2230 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
2231 authenticator to fail if there's a tab character in the incoming data
2232 (there should never be unless someone is messing about, as it's supposed
2233 to be base64-encoded). Also added, on Timo's advice, the "secured" option
2234 if the connection is using TLS or if the remote IP is the same as the
2235 local IP, and the "valid-client-cert option" if a client certificate has
2238 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
2239 authenticators. This can be used for authorization after authentication
2240 succeeds. (In the case of plaintext, it servers for both authentication
2243 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
2244 if any retry times were supplied.
2246 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
2247 connection, where there is no client IP address to check. In this
2248 situation, the verify now always succeeds.
2250 PH/25 Applied John Jetmore's -Mset patch.
2252 PH/26 Added -bem to be like -Mset, but loading a message from a file.
2254 PH/27 In a string expansion for a processed (not raw) header when multiple
2255 headers of the same name were present, leading whitespace was being
2256 removed from all of them, but trailing whitespace was being removed only
2257 from the last one. Now trailing whitespace is removed from each header
2258 before concatenation. Completely empty headers in a concatenation (as
2259 before) are ignored.
2261 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
2262 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
2264 PH/29 [Removed. This was a change that I later backed out, and forgot to
2265 correct the ChangeLog entry (that I had efficiently created) before
2266 committing the later change.]
2268 PH/30 Exim was sometimes attempting to deliver messages that had suffered
2269 address errors (4xx response to RCPT) over the same connection as other
2270 messages routed to the same hosts. Such deliveries are always "forced",
2271 so retry times are not inspected. This resulted in far too many retries
2272 for the affected addresses. The effect occurred only when there were more
2273 hosts than the hosts_max_try setting in the smtp transport when it had
2274 the 4xx errors. Those hosts that it had tried were not added to the list
2275 of hosts for which the message was waiting, so if all were tried, there
2276 was no problem. Two fixes have been applied:
2278 (i) If there are any address or message errors in an SMTP delivery, none
2279 of the hosts (tried or untried) are now added to the list of hosts
2280 for which the message is waiting, so the message should not be a
2281 candidate for sending over the same connection that was used for a
2282 successful delivery of some other message. This seems entirely
2283 reasonable: after all the message is NOT "waiting for some host".
2284 This is so "obvious" that I'm not sure why it wasn't done
2285 previously. Hope I haven't missed anything, but it can't do any
2286 harm, as the worst effect is to miss an optimization.
2288 (ii) If, despite (i), such a delivery is accidentally attempted, the
2289 routing retry time is respected, so at least it doesn't keep
2290 hammering the server.
2292 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
2293 in ${readsocket because some servers need this prod.
2295 PH/32 Added some extra debug output when updating a wait-xxx database.
2297 PH/33 The hint "could be header name not terminated by colon", which has been
2298 given for certain expansion errors for a long time, was not being given
2299 for the ${if def:h_colon_omitted{... case.
2301 PH/34 The spec says: "With one important exception, whenever a domain list is
2302 being scanned, $domain contains the subject domain." There was at least
2303 one case where this was not true.
2305 PH/35 The error "getsockname() failed: connection reset by peer" was being
2306 written to the panic log as well as the main log, but it isn't really
2307 panic-worthy as it just means the connection died rather early on. I have
2308 removed the panic log writing for the ECONNRESET error when getsockname()
2311 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
2312 runs only) independently of the message's sender address. This meant
2313 that, if the 4xx error was in fact related to the sender, a different
2314 message to the same recipient with a different sender could confuse
2315 things. In particualar, this can happen when sending to a greylisting
2316 server, but other circumstances could also provoke similar problems.
2317 I have changed the default so that the retry time for these errors is now
2318 based a combination of the sender and recipient addresses. This change
2319 can be overridden by setting address_retry_include_sender=false in the
2322 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
2323 remote server are returned as part of bounce messages. This was not
2324 happening for LMTP over a pipe (the lmtp transport), but now it is the
2325 same for both kinds of LMTP.
2327 PH/38 Despite being documented as not happening, Exim was rewriting addresses
2328 in header lines that were in fact CNAMEs. This is no longer the case.
2330 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
2331 and queue runs started by the daemon processed all messages. This has
2332 been fixed so that -R and -S can now usefully be given with -q<time>.
2334 PH/40 Import PCRE release 6.7 (fixes some bugs).
2336 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
2338 PH/42 Give an error if -q is specified more than once.
2340 PH/43 Renamed the variables $interface_address and $interface_port as
2341 $received_ip_address and $received_port, to make it clear that these
2342 values apply to message reception, and not to the outgoing interface when
2343 a message is delivered. (The old names remain recognized, of course.)
2345 PH/44 There was no timeout on the connect() call when using a Unix domain
2346 socket in the ${readsocket expansion. There now is.
2348 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
2349 be meaningful with "accept".
2351 SC/01 Eximstats V1.43
2352 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
2354 SC/02 Eximstats V1.44
2355 Use a glob alias rather than an array ref in the generated
2356 parser. This improves both readability and performance.
2358 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
2359 Collect SpamAssassin and rejection statistics.
2360 Don't display local sender or destination tables unless
2361 there is data to show.
2362 Added average volumes into the top table text output.
2364 SC/04 Eximstats V1.46
2365 Collect data on the number of addresses (recipients)
2366 as well as the number of messages.
2368 SC/05 Eximstats V1.47
2369 Added 'Message too big' to the list of mail rejection
2370 reasons (thanks to Marco Gaiarin).
2372 SC/06 Eximstats V1.48
2373 Mainlog lines which have GMT offsets and are too short to
2374 have a flag are now skipped.
2376 SC/07 Eximstats V1.49 (Alain Williams)
2377 Added the -emptyok flag.
2379 SC/08 Eximstats V1.50
2380 Fixes for obtaining the IP address from reject messages.
2382 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
2383 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
2384 whitesspace changes from 4.64-PH/27
2386 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
2389 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
2390 are found, allow negative numbers in numeric criteria)
2392 JJ/06 exipick.20061117.2, added new $message_body_missing variable
2394 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
2395 to match changes made in 4.64-PH/43
2397 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
2399 PH/47 Put in an explicit test for a DNS lookup of an address record where the
2400 "domain" is actually an IP address, and force a failure. This locks out
2401 those revolvers/nameservers that support "A-for-A" lookups, in
2402 contravention of the specifications.
2404 PH/48 When a host name was looked up from an IP address, and the subsequent
2405 forward lookup of the name timed out, the host name was left in
2406 $sender_host_name, contrary to the specification.
2408 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
2409 restricted to single-key lookups, Exim was not diagnosing an error if
2410 * or *@ was used with a query-style lookup.
2412 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
2414 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
2415 long ago, but noone interested enough thought of it. Let's just say that
2416 the "1.1" means that there are some new functions that weren't there at
2417 some point in the past.
2419 PH/51 Error processing for expansion failure of helo_data from an smtp
2420 transport during callout processing was broken.
2422 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
2423 tested/used via the -bh/-bhc/-bs options.
2425 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
2426 bug, fixed in subsequent PCRE releases).
2428 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
2429 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
2431 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
2437 SC/01 Use a glob alias rather than an array ref in eximstats generated
2438 parser. This improves both readability and performance.
2440 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
2441 Don't display local sender or destination tables in eximstats unless
2442 there is data to show.
2443 Added average volumes into the eximstats top table text output.
2445 SC/03 Collect data on the number of addresses (recipients) as well
2446 as the number of messages in eximstats.
2448 TF/01 Correct an error in the documentation for the redirect router. Exim
2449 does (usually) call initgroups() when daemonizing.
2451 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
2452 with consistent privilege compared to when running as a daemon.
2454 TF/03 Note in the spec that $authenticated_id is not set for local
2455 submissions from trusted users.
2457 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
2458 Thanks to Dean Brooks <dean@iglou.com> for the patch.
2460 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
2461 by adding some example configuration directives to the default
2462 configuration file. A little bit of work is required to uncomment the
2463 directives and define how usernames and passwords are checked, but
2464 there is now a framework to start from.
2466 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
2467 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
2468 without this. I don't know how relevant this is to other LDAP libraries.
2470 PH/02 Add the verb name to the "unknown ACL verb" error.
2472 PH/03 Magnus Holmgren's patch for filter_prepend_home.
2474 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
2476 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
2477 directory not expanded when it should be if an expanded home directory
2478 was set for the address (which is overridden by the transport).
2480 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
2483 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
2484 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
2485 because it is too late at that time, and has no effect.
2487 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
2488 security issue with \' (bugzilla #107). I could not use the
2489 PQescapeStringConn() function, because it needs a PGconn value as one of
2492 PH/08 When testing addresses using -bt, indicate those final addresses that
2493 are duplicates that would not cause an additional delivery. At least one
2494 person was confused, thinking that -bt output corresponded to deliveries.
2495 (Suppressing duplicates isn't a good idea as you lose the information
2496 about possibly different redirections that led to the duplicates.)
2498 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
2499 systems where poll() doesn't work, in particular OS X.
2501 PH/10 Added more information to debugging output for retry time not reached.
2503 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
2504 operations in malware.c.
2506 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
2509 PH/13 If write_rejectlog was set false when logging was sent to syslog with
2510 syslog_duplication set false, log lines that would normally be written
2511 both the the main log and to the reject log were not written to syslog at
2514 PH/14 In the default configuration, change the use of "message" in ACL warn
2515 statements to "add_header".
2517 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
2518 not followed by a command (e.g. "seen endif").
2520 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
2521 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
2524 PH/17 Added extra conditions to the default value of delay_warning_condition
2528 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
2529 { match{$h_precedence:}{(?i)bulk|list|junk} } \
2530 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
2533 The Auto-Submitted: and various List- headers are standardised, whereas I
2534 don't think Precedence: ever was.
2536 PH/18 Refactored debugging code in route_finduser() to show more information,
2537 in particular, the error code if getpwnam() issues one.
2539 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
2540 This is apparently needed in addition to the PH/07 change above to avoid
2541 any possible encoding problems.
2543 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
2544 but not after initializing Perl.
2546 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
2547 output them only if debugging. By default they are written stderr,
2548 apparently, which is not desirable.
2550 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
2553 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
2556 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
2558 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
2559 authenticated or an ident call has been made. Suppress the default
2560 values for $authenticated_id and $authenticated_sender (but permit -oMai
2561 and -oMas) when testing with -bh.
2563 PH/24 Re-jigged the order of the tests in the default configuration so that the
2564 tests for valid domains and recipients precede the DNS black list and CSA
2565 tests, on the grounds that those ones are more expensive.
2567 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
2568 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
2569 This bug exists in every version of Exim that I still have, right back to
2572 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
2573 However, an attempt to turn on case-sensitivity in a regex key by
2574 including (?-i) didn't work because the subject string was already
2575 lowercased, and the effects were non-intuitive. It turns out that a
2576 one-line patch can be used to allow (?-i) to work as expected.
2582 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
2583 other effects) broke the use of negated acl sub-conditions.
2585 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
2588 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
2589 "Deny" causes Exim to reject the incoming connection with a 554 error.
2590 Unfortunately, if there is a major crisis, such as a disk failure,
2591 tcp-wrappers gives "deny", whereas what one would like would be some
2592 kind of temporary error. A kludge has been added to help with this.
2593 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
2594 554 error is used if errno is still zero or contains ENOENT (which occurs
2595 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
2598 PH/03 Add -lutil to the default FreeBSD LIBS setting.
2600 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
2601 errors. Otherwise a message that provokes a temporary error (when other
2602 messages do not) can cause a whole host to time out.
2604 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
2605 addresses were routed directly to files or pipes from a redirect router.
2606 File deliveries just didn't batch; pipe deliveries might have suffered
2609 PH/06 A failure to get a lock for a hints database would erroneously always say
2610 "Failed to get write lock", even when it was really a read lock.
2612 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
2613 of 0600. This has been changed to use the value of the lockfile_mode
2614 option (which defaults to 0600).
2616 PH/08 Applied small patch from the Sieve maintainer.
2618 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
2619 folder from quota calculations, a direct delivery into this folder messed
2620 up the contents of the maildirsize file. This was because the regex was
2621 used only to exclude .Trash (or whatever) when the size of the mailbox
2622 was calculated. There was no check that a delivery was happening into an
2623 excluded directory. This bug has been fixed by ignoring all quota
2624 processing for deliveries into excluded directories.
2626 PH/10 Added the maildirfolder_create_regex option to appendfile.
2632 PH/01 The code for finding all the local interface addresses on a FreeBSD
2633 system running IPv6 was broken. This may well have applied to all BSD
2634 systems, as well as to others that have similar system calls. The broken
2635 code found IPv4 interfaces correctly, but gave incorrect values for the
2636 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
2637 that it would not match correctly against @[] and not recognize the IPv6
2640 PH/02 The ipliteral router was not recognizing addresses of the form user@
2641 [ipv6:....] because it didn't know about the "ipv6:" prefix.
2643 PH/03 Added disable_ipv6.
2645 PH/04 Changed $reply_address to use the raw form of the headers instead of the
2646 decoded form, because it is most often used to construct To: headers
2647 lines in autoreplies, and the decoded form may well be syntactically
2648 invalid. However, $reply_address has leading white space removed, and all
2649 newlines turned into spaces so that the autoreply transport does not
2652 PH/05 If group was specified without a user on a router, and no group or user
2653 was specified on a transport, the group from the router was ignored.
2655 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
2656 for visible compile-time settings that can be used to change these
2657 numbers, for those that want even more. Backwards compatibility with old
2658 spool files has been maintained. However, going back to a previous Exim
2659 release will lost any variables that are in spool files.
2661 PH/07 Two small changes when running in the test harness: increase delay when
2662 passing a TCP/IP connection to a new process, in case the original
2663 process has to generate a bounce, and remove special handling of
2664 127.0.0.2 (sic), which is no longer necessary.
2666 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
2667 be the same on different OS.
2669 PH/09 Moved a debug statement in filter processing to avoid a race problem when
2672 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
2673 whether --show-vars was specified or not
2675 JJ/02 exipick: Added support for new ACL variable spool format introduced
2678 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
2679 syntactically invalid From: or Reply-to: line, and a filter used this to
2680 generate an autoreply, and therefore failed to obtain an address for the
2681 autoreply, Exim could try to deliver to a non-existent relative file
2682 name, causing unrelated and misleading errors. What now happens is that
2683 it logs this as a hard delivery error, but does not attempt to create a
2686 PH/11 The exinext utility has a -C option for testing purposes, but although
2687 the given file was scanned by exinext itself; it wasn't being passed on
2688 when Exim was called.
2690 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
2691 an end-of-file indication when reading a command response.
2693 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
2694 compiled. In many other places in Exim, IPv6 addresses are always
2695 recognized, so I have changed this. It also means that IPv4 domain
2696 literals of the form [IPV4:n.n.n.n] are now always recognized.
2698 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
2699 used if the router is not running as root, for example, when verifying at
2700 ACL time, or when using -bh. The debugging output from this situation was
2701 non-existent - all you got was a failure to exec. I have made two
2704 (a) Failures to set uid/gid, the current directory, or a process leader
2705 in a subprocess such as that created by queryprogram now generate
2706 suitable debugging ouput when -d is set.
2708 (b) The queryprogram router detects when it is not running as root,
2709 outputs suitable debugging information if -d is set, and then runs
2710 the subprocess without attempting to change uid/gid.
2712 PH/15 Minor change to Makefile for building test_host (undocumented testing
2715 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
2716 additional section of a DNS packet that returns MX or SRV records.
2717 Instead, it always explicitly searches for A/AAAA records. This avoids
2718 major problems that occur when a DNS server includes only records of one
2719 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
2720 fixed another bug: if SRV records were looked up and the corresponding
2721 address records were *not* found in the additional section, the port
2722 values from the SRV records were lost.
2724 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
2725 using the correct key (the original address) when searching the retry
2726 rules in order to find which one to use for generating the retry hint.
2728 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
2729 adding the default one. Similarly, if it contains a Reply-To: header, the
2730 errors_reply_to option, if set, is not used.
2732 PH/19 When calculating a retry time, Exim used to measure the "time since
2733 failure" by looking at the "first failed" field in the retry record. Now
2734 it does not use this if it is later than than the arrival time of the
2735 message. Instead it uses the arrival time. This makes for better
2736 behaviour in cases where some deliveries succeed, thus re-setting the
2737 "first failed" field. An example is a quota failure for a huge message
2738 when small messages continue to be delivered. Without this change, the
2739 "time since failure" will always be short, possible causing more frequent
2740 delivery attempts for the huge message than are intended.
2741 [Note: This change was subsequently modified - see PH/04 for 4.62.]
2743 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
2744 $1, $2, $3) because the numerical variables can be reset during some
2745 expansion items (e.g. "match"), thereby losing the authentication data.
2747 PH/21 Make -bV show the size of off_t variables so that the test suite can
2748 decide whether to run tests for quotas > 2G.
2750 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
2751 mailbox_size, and mailbox_filecount in the appendfile transport. If a
2752 filecount value is greater than 2G or if a quota value is greater than 2G
2753 on a system where the size of off_t is not greater than 4, a panic error
2756 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
2757 never match. The debug and -bh output now contains an explicit error
2758 message indicating a malformed IPv4 address or mask.
2760 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
2761 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
2762 PH/23 above applies.
2764 PH/25 Do not write to syslog when running in the test harness. The only
2765 occasion when this arises is a failure to open the main or panic logs
2766 (for which there is an explicit test).
2768 PH/26 Added the /no_tell option to "control=freeze".
2770 PH/27 If a host name lookup failed very early in a connection, for example, if
2771 the IP address matched host_lookup and the reverse lookup yielded a name
2772 that did not have a forward lookup, an error message of the form "no IP
2773 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
2774 could be logged. Now it outputs the IP address instead of "NULL".
2776 PH/28 An enabling patch from MH: add new function child_open_exim2() which
2777 allows the sender and the authenticated sender to be set when
2778 submitting a message from within Exim. Since child_open_exim() is
2779 documented for local_scan(), the new function should be too.
2781 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
2782 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
2783 results in an empty string is now treated as unset.
2785 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
2787 PH/31 Added sender_verify_fail logging option.
2789 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
2790 needed by RFC 822 but not by RFC 2822 was commented out. I have now
2791 tidied the source and removed it altogether.
2793 PH/33 When a queue run was abandoned because the load average was too high, a
2794 log line was always written; now it is written only if the queue_run log
2795 selector is set. In addition, the log line for abandonment now contains
2796 information about the queue run such as the pid. This is always present
2797 in "start" and "stop" lines but was omitted from the "abandon" line.
2799 PH/34 Omit spaces between a header name and the colon in the error message that
2800 is given when verify = headers_syntax fails (if there are lots of them,
2801 the message gets confusing).
2803 PH/35 Change the default for dns_check_names_pattern to allow slashes within
2804 names, as there are now some PTR records that contain slashes. This check
2805 is only to protect against broken name servers that fall over on strange
2806 characters, so the fact that it applies to all lookups doesn't matter.
2808 PH/36 Now that the new test suite is complete, we can remove some of the
2809 special code in Exim that was needed for the old test suite. For example,
2810 sorting DNS records because real resolvers return them in an arbitrary
2811 order. The new test suite's fake resolver always returns records in the
2814 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
2815 bounces) except when queue_only is set, to avoid logging races between
2816 the different processes.
2818 PH/38 Panic-die if .include specifies a non-absolute path.
2820 PH/39 A tweak to the "H" retry rule from its user.
2822 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
2823 a label. They prevented compilation on older perls.
2825 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
2826 a warning to be raised on newish perls.
2828 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
2829 on queue. Changes to match documented behaviour of showing count of
2830 messages matching specified criteria.
2832 PH/40 Changed the default ident timeout from 30s to 5s.
2834 PH/41 Added support for the use of login_cap features, on those BSD systems
2835 that have them, for controlling the resources used by pipe deliveries.
2837 PH/42 The content-scanning code uses fopen() to create files in which to put
2838 message data. Previously it was not paying any attention to the mode of
2839 the files. Exim runs with umask(0) because the rest of the code creates
2840 files with open(), and sets the required mode explicitly. Thus, these
2841 files were ending up world-writeable. This was not a big issue, because,
2842 being within the spool directory, they were not world-accessible. I have
2843 created a function called modefopen, which takes an additional mode
2844 argument. It sets umask(777), creates the file, chmods it to the required
2845 mode, then resets the umask. All the relevant calls to fopen() in the
2846 content scanning code have been changed to use this function.
2848 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
2849 to 24 hours. This avoids potential overflow problems when processing G
2850 and H retry rules. I suspect nobody ever tinkers with this value.
2852 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
2854 PH/45 When the plaintext authenticator is running as a client, the server's
2855 challenges are checked to ensure they are valid base64 strings. By
2856 default, the authentication attempt is cancelled if an invalid string is
2857 received. Setting client_ignore_invalid_base64 true ignores these errors.
2858 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
2859 they are received. Thus, the responses can be made to depend on the
2860 challenges. If an invalid string is ignored, an empty string is placed in
2863 PH/46 Messages that are created by the autoreply transport now contains a
2864 References: header, in accordance with RFCs 2822 and 3834.
2866 PH/47 Added authenticated_sender_force to the smtp transport.
2868 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
2870 PH/49 Installed latest patch from the Sieve maintainer.
2872 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
2873 was also set, the appendfile transport was unnecessarily scanning a
2874 directory of message files (e.g. for maildir delivery) to find the count
2875 of files (along with the size), even though it did not need this
2876 information. It now does the scan only if it needs to find either the
2877 size of the count of files.
2879 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
2881 PH/52 Two bugs concerned with error handling when the smtp transport is
2884 (i) Exim was not creating retry information for temporary errors given
2885 for individual recipients after the DATA command when the smtp transport
2886 was used in LMTP mode. This meant that they could be retried too
2887 frequently, and not timed out correctly.
2889 (ii) Exim was setting the flag that allows error details to be returned
2890 for LMTP errors on RCPT commands, but not for LMTP errors for individual
2891 recipients that were returned after the DATA command.
2893 PH/53 This is related to PH/52, but is more general: for any failing address,
2894 when detailed error information was permitted to be returned to the
2895 sender, but the error was temporary, then after the final timeout, only
2896 "retry timeout exceeded" was returned. Now it returns the full error as
2897 well as "retry timeout exceeded".
2899 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
2900 do this, and (what is worse) MTAs that accept it.
2902 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
2903 will now be deprecated.
2905 PH/56 New os.c-cygwin from the Cygwin maintainer.
2907 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
2908 formats (previously only available in exim formats via -bpr, -bpru,
2909 and -bpra. Now also available in native and exiqgrep formats)
2911 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
2912 with very large, slow to parse queues
2914 JJ/08 exipick: added ! as generic prefix to negate any criteria format
2916 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
2918 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
2919 responses to authentication challenges, though it was showing the
2920 challenges; (ii) I've removed the CR characters from the debug output for
2923 PH/58 Allow for the insertion of a newline as well as a space when a string
2924 is turned into more than one encoded-word during RFC 2047 encoding. The
2925 Sieve code now uses this.
2927 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
2928 data_4xx, lost_connection, tls_required.
2930 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
2931 message was being sent as an SMTP response.
2933 PH/61 Add -l and -k options to exicyclog.
2935 PH/62 When verifying, if an address was redirected to one new address, so that
2936 verification continued, and the new address failed or deferred after
2937 having set something in $address_data, the value of $address_data was not
2938 passed back to the ACL. This was different to the case when no
2939 redirection occurred. The value is now passed back in both cases.
2941 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
2942 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
2943 use login_cap.h, so on its own it isn't the distinguishing feature. The
2944 new name refers directly to the setclassresources() function.
2946 PH/65 Added configuration files for NetBSD3.
2948 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
2950 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
2951 is preferred over IPv4.
2953 PH/68 The bounce_return_message and bounce_return_body options were not being
2954 honoured for bounces generated during the reception of non-SMTP messages.
2955 In particular, this applied to messages rejected by the ACL. This bug has
2956 been fixed. However, if bounce_return_message is true and bounce_return_
2957 body is false, the headers that are returned for a non-SMTP message
2958 include only those that have been read before the error was detected.
2959 (In the case of an ACL rejection, they have all been read.)
2961 PH/69 The HTML version of the specification is now built in a directory called
2962 spec_html instead of spec.html, because the latter looks like a path with
2963 a MIME-type, and this confuses some software.
2965 PH/70 Catch two compiler warnings in sieve.c.
2967 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
2968 function verify_get_ident() calls ip_connect() to connect a socket, but
2969 if the "connect()" function timed out, ip_connect() used to close the
2970 socket. However, verify_get_ident() also closes the socket later, and in
2971 between Exim writes to the log, which may get opened at this point. When
2972 the socket was closed in ip_connect(), the log could get the same file
2973 descriptor number as the socket. This naturally causes chaos. The fix is
2974 not to close the socket in ip_connect(); the socket should be closed by
2975 the function that creates it. There was only one place in the code where
2976 this was missing, in the iplookup router, which I don't think anybody now
2977 uses, but I've fixed it anyway.
2979 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
2980 well as to direct DNS lookups. Otherwise the handling of names in host
2981 lists is inconsistent and therefore confusing.
2987 PH/01 Two changes to the default runtime configuration:
2989 (1) Move the checks for relay_from_hosts and authenticated clients from
2990 after to before the (commented out) DNS black list checks.
2992 (2) Add control=submission to the relay_from_hosts and authenticated
2993 clients checks, on the grounds that messages accepted by these
2994 statements are most likely to be submissions.
2996 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
2998 (1) Generate an error if the third argument for the ${prvs expansion is
3001 (2) Treat a missing third argument of ${prvscheck as if it were an empty
3004 (3) Reset the variables that are obtained from the first argument of
3005 ${prvscheck and used in the second argument before leaving the code,
3006 because their memory is reclaimed, so using them afterwards may do
3009 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
3010 one (it's much easier than Tom thought :-).
3012 (5) Because of (4), we can now allow for the use of $prvscheck_result
3013 inside the third argument.
3015 PH/03 For some reason, the default setting of PATH when running a command from
3016 a pipe transport was just "/usr/bin". I have changed it to
3019 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
3020 anything to be listed in the output from -bV.
3022 PH/05 When a filter generated an autoreply, the entire To: header line was
3023 quoted in the delivery log line, like this:
3025 => >A.N.Other <ano@some.domain> <original@ddress> ...
3027 This has been changed so that it extracts the operative address. There
3028 may be more than one such address. If so, they are comma-separated, like
3031 => >ano@some.domain,ona@other.domain <original@ddress> ...
3033 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
3034 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
3035 not being looked up in the rDNS to get a host name, Exim was showing the
3036 IP address twice in Received: lines, even though the IP addresses were
3037 identical. For example:
3039 Received: from [1.2.3.4] (helo=[1.2.3.4])
3041 However, if the real host name was known, it was omitting the HELO data
3042 if it matched the actual IP address. This has been tidied up so that it
3043 doesn't show the same IP address twice.
3045 PH/07 When both +timestamp and +memory debugging was on, the value given by
3046 $tod_xxx expansions could be wrong, because the tod_stamp() function was
3047 called by the debug printing, thereby overwriting the timestamp buffer.
3048 Debugging no longer uses the tod_stamp() function when +timestamp is set.
3050 PH/08 When the original message was included in an autoreply transport, it
3051 always said "this is a copy of the message, including all the headers",
3052 even if body_only or headers_only was set. It now gives an appropriate
3055 PH/09 Applied a patch from the Sieve maintainer which:
3057 o fixes some comments
3058 o adds the (disabled) notify extension core
3059 o adds some debug output for the result of if/elsif tests
3060 o points to the current vacation draft in the documentation
3061 and documents the missing references header update
3065 o fixes a bug in processing the envelope test (when testing
3066 multiple envelope elements, the last element determinted the
3069 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
3070 Electronic Mail") by including:
3072 Auto-submitted: auto-generated
3074 in the messages that it generates (bounce messages and others, such as
3075 warnings). In the case of bounce messages for non-SMTP mesages, there was
3076 also a typo: it was using "Auto_submitted" (underscore instead of
3077 hyphen). Since every message generated by Exim is necessarily in response
3078 to another message, thes have all been changed to:
3080 Auto-Submitted: auto-replied
3082 in accordance with these statements in the RFC:
3084 The auto-replied keyword:
3086 - SHOULD be used on messages sent in direct response to another
3087 message by an automatic process,
3089 - MUST NOT be used on manually-generated messages,
3091 - MAY be used on Delivery Status Notifications (DSNs) and Message
3092 Disposition Notifications (MDNs),
3094 - MUST NOT be used on messages generated by automatic or periodic
3095 processes, except for messages which are automatic responses to
3098 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
3099 to the default Received: header definition.
3101 PH/12 Added log selector acl_warn_skipped (default on).
3103 PH/13 After a successful wildlsearch lookup, discard the values of numeric
3104 variables because (a) they are in the wrong storage pool and (b) even if
3105 they were copied, it wouldn't work properly because of the caching.
3107 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
3108 checking when decoding. Apparently there are clients that generate
3109 overlong encoded strings. Why am I not surprised?
3111 PH/15 If the first argument of "${if match_address" was not empty, but did not
3112 contain an "@" character, Exim crashed. Now it writes a panic log message
3113 and treats the condition as false.
3115 PH/16 In autoreply, treat an empty string for "once" the same as unset.
3117 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
3118 extension "envelope-auth". The code is finished and in agreement with
3119 other implementations, but there is no documentation so far and in fact,
3120 nobody wrote the draft yet. This extension is currently #undef'ed, thus
3121 not changing the active code.
3123 Print executed "if" and "elsif" statements when debugging is used. This
3124 helps a great deal to understand what a filter does.
3126 Document more things not specified clearly in RFC3028. I had all this
3127 sorted out, when out of a sudden new issues came to my mind. Oops."
3129 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
3132 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
3133 canonical form (as documented). However, after a host name lookup from
3134 the IP address, check_host() was doing a simple string comparison with
3135 addresses acquired from the DNS when checking that the found name did
3136 have the original IP as one of its addresses. Since any found IPv6
3137 addresses are likely to be in abbreviated form, the comparison could
3138 fail. Luckily, there already exists a function for doing the comparison
3139 by converting both addresses to binary, so now that is used instead of
3140 the text comparison.
3142 PH/20 There was another similar case to PH/19, when a complete host name was
3143 given in a host list; looking up its IP address could give an abbreviated
3144 form, whereas the current host's name might or might not be abbreviated.
3145 The same fix has been applied.
3151 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
3152 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
3155 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
3156 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
3158 PH/03 Typo: missing ".o" in src/pcre/Makefile.
3160 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
3161 header line, restrict the check to what is listed in RFCs 2369 and 2929.
3162 Also, for "Auto-Submitted", treat anything other than "no" as
3163 non-personal, in accordance with RFC 3834. (Previously it treated
3164 anything starting "auto-" as non-personal.)
3166 TF/01 The control=submission/name=... option had a problem with syntax
3167 errors if the name included a slash character. The /name= option
3168 now slurps the rest of the string, so it can include any characters
3169 but it must come last in the list of options (after /sender_retain
3172 PH/05 Some modifications to the interface to the fake nameserver for the new
3180 TK/01 Added the "success_on_redirect" address verification option. See
3181 NewStuff for rationale and an example.
3183 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
3185 PH/02 Patch to exigrep to allow it to work on syslog lines.
3187 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
3188 fread() to skip over the body file's header line, because in Cygwin the
3189 header line is locked and is inaccessible.
3191 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
3192 co-exist for some time) to make it clear that it is the Exim ID that is
3193 referenced, not the Message-ID: header line.
3195 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
3196 string_format() function, because snprintf() does not exist on all
3199 PH/06 The use of forbid_filter_existstest now also locks out the use of the
3200 ${stat: expansion item.
3202 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
3203 protocol synchronization error", to keep the pedants happy.
3205 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
3206 well as for IRIX systems, when gcc is being used. See the host.c source
3209 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
3211 PH/10 Named domain lists were not working if used in a queue_smtp_domains
3214 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
3215 transport and to the smtp transport in LMTP mode.
3217 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
3219 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
3220 run a filter in a subprocess. This could lead to confusion in subsequent
3221 lookups in the parent process. There should also be a search_tidyup() at
3222 the end of the subprocess.
3224 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
3225 only if the host matched helo_try_verify_hosts, which caused the
3226 verification to occur when the EHLO/HELO command was issued. The ACL just
3227 tested the remembered result. Now, if a previous verification attempt has
3228 not happened, "verify = helo" does it there and then.
3230 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
3232 TK/03 Fix log output including CR from clamd.
3234 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
3235 exist provoked a memory error which could cause a segfault.
3237 PH/15 Installed PCRE 6.2
3239 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
3241 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
3242 of the problem. Specifically, suggested +O2 rather than +O1 for the
3245 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
3247 PH/20 If a delivery was routed to a non-standard port by means of an SRV
3248 record, the port was not correctly logged when the outgoing_port log
3249 selector was set (it logged the transort's default port).
3251 PH/21 Added support for host-specific ports to manualroute, queryprogram,
3252 fallback_hosts, and "hosts" in the smtp transport.
3254 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
3255 host errors such as "Connection refused".
3257 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
3258 authentication with radiusclient 0.4.9:
3260 - Error returned from rc_read_config was caught wrongly
3261 - Username/password not passed on to radius server due to wrong length.
3263 The presumption is that some radiusclient API changes for 4.51/PH/17
3264 were not taken care of correctly. The code is still untested by me (my
3265 Linux distribution still has 0.3.2 of radiusclient), but it was
3266 contributed by a Radius user.
3268 PH/24 When doing a callout, the value of $domain wasn't set correctly when
3269 expanding the "port" option of the smtp transport.
3271 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
3272 while reading a MIME header. Thanks to Tom Hughes for a patch.
3274 PH/24 Include config.h inside local_scan.h so that configuration settings are
3277 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
3278 that in an ACL for RCPT (for example), you can examine exactly what was
3281 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
3282 commands, but it was not correctly comparing the address with the actual
3283 client host address. Thus, it would show the EHLO address in Received:
3284 header lines when this was not necessary.
3286 PH/27 Added the % operator to ${eval:}.
3288 PH/28 Exim tries to create and chdir to its spool directory when it starts;
3289 it should be ignoring failures (because with -C, for example, it has lost
3290 privilege). It wasn't ignoring creation failures other than "already
3293 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
3296 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
3297 because an input line was too long, either on its own, or by virtue of
3298 too many continuations, the temporary file was not being removed, and the
3299 return code was incorrect.
3301 PH/31 Missing "BOOL" in function definition in filtertest.c.
3303 PH/32 Applied Sieve patches from the maintainer.
3305 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
3307 PH/33 Added "verify = not_blind".
3309 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
3310 Local/Makefile (with some defaults set). These are used in built scripts
3311 such as exicyclog, but they have never been used in the exim_install
3312 script (though there are many overriding facilities there). I have
3313 arranged that the exim_install script now takes note of these two
3316 PH/35 Installed configuration files for Dragonfly.
3318 PH/36 When a locally submitted message by a trusted user did not contain a
3319 From: header, and the sender address was obtained from -f or from an SMTP
3320 MAIL command, and the trusted user did not use -F to supply a sender
3321 name, $originator_name was incorrectly used when constructing a From:
3322 header. Furthermore, $originator_name was used for submission mode
3323 messages from external hosts without From: headers in a similar way,
3324 which is clearly wrong.
3326 PH/37 Added control=suppress_local_fixups.
3328 PH/38 When log_selector = +received_sender was set, and the addition of the
3329 sender made the log line's construction buffer exactly full, or one byte
3330 less than full, an overflow happened when the terminating "\n" was
3333 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
3334 when the result of a list match is failure because a DNS lookup failed.
3336 PH/40 RM_COMMAND is now used in the building process.
3338 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
3339 the "build-* directories that it finds.
3341 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
3342 address in a domain literal was a prefix of an interface address.
3344 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
3345 when verifying a sender address, unless rewrite_headers is false.
3347 PH/44 Wrote a long comment about why errors_to addresses are verified as
3348 recipients, not senders.
3350 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
3351 the ratelimit ACL was added.
3353 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
3355 PH/46 Added extra information about PostgreSQL errors to the error string.
3357 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
3358 avoiding the need to install special zones in a real server. This is
3359 backwards compatible; if it can't find the fake resolver, it drops back.
3360 Thus, both old and new test suites can be run.
3362 TF/02 Added util/ratelimit.pl
3364 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
3365 clock is set back in time.
3367 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
3368 Candler <B.Candler@pobox.com>.
3370 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
3371 OK for addresses that are the result of redirections.
3373 PH/48 A number of further additions for the benefit of the new test suite,
3374 including a fake gethostbyname() that interfaces to the fake DNS resolver
3377 TF/06 The fix for widen_domains has also been applied to qualify_single and
3378 search_parents which are the other dnslookup options that can cause
3381 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
3384 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
3386 TF/07 Exim produced the error message "an SRV record indicated no SMTP
3387 service" if it encountered an MX record with an empty target hostname.
3388 The message is now "an MX or SRV record indicated no SMTP service".
3390 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
3391 if the DNS of the sending site is misconfigured. This is quite a
3392 common situation. This change restores the behaviour of treating a
3393 helo verification defer as a failure.
3395 PH/51 If self=fail was set on a router, the bounce message did not include the
3396 actual error message.
3402 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
3404 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
3405 command itself ended in error, the underlying message about the transport
3406 filter timeout was being overwritten with the pipe command error. Now the
3407 underlying error message should be appended to the second error message.
3409 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
3411 PH/02 Reduce the amount of output that "make" produces by default. Full output
3412 can still be requested.
3414 PH/03 The warning log line about a condition test deferring for a "warn" verb
3415 was being output only once per connection, rather than after each
3416 occurrence (because it was using the same function as for successful
3417 "warn" verbs). This seems wrong, so I have changed it.
3419 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
3420 it should not have, which might have caused a crash in the right
3421 circumstances, but probably never did.
3423 PH/04 Installed a modified version of Tony Finch's patch to make submission
3424 mode fix the return path as well as the Sender: header line, and to
3425 add a /name= option so that you can make the user's friendly name appear
3428 TF/03 Added the control = fakedefer ACL modifier.
3430 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
3431 Mark Lowes for thorough testing.
3433 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
3435 TK/03 Merged latest SRS patch from Miles Wilton.
3437 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
3438 with the definition in sysexits.h (which is #included earlier).
3439 Fortunately, Exim does not actually use EX_OK. The code used to try to
3440 preserve the sysexits.h value, by assumimg that macro definitions were
3441 scanned for macro replacements. I have been disabused of this notion,
3442 so now the code just undefines EX_OK before #including unistd.h.
3444 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
3445 in the smtp transport. When a block could not be written in a single
3446 write() function, the timeout was being re-applied to each part-write.
3447 This seems wrong - if the receiver was accepting one byte at a time it
3448 would take for ever. The timeout is now adjusted when this happens. It
3449 doesn't have to be particularly precise.
3451 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
3452 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
3454 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
3455 without a domain if the check to <postmaster@domain> fails.
3457 SC/01 Eximstats: added -xls and the ability to specify output files
3458 (patch written by Frank Heydlauf).
3460 SC/02 Eximstats: use FileHandles for outputing results.
3462 SC/03 Eximstats: allow any combination of xls, txt, and html output.
3464 SC/04 Eximstats: fixed display of large numbers with -nvr option
3466 SC/05 Eximstats: fixed merging of reports with empty tables.
3468 SC/06 Eximstats: added the -include_original_destination flag
3470 SC/07 Eximstats: removed tabs and trailing whitespace.
3472 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
3474 TK/06 MBOX spool code: Add real "From " MBOX separator line
3475 so the .eml file is really in mbox format (even though
3476 most programs do not really care). Patch from Alex Miller.
3478 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
3479 The latter is generated from $received_to and is only set if the
3480 message has one envelope recipient. SA can use these headers,
3481 obviously out-of-the-box. Patch from Alex Miller.
3483 PH/08 The ${def test on a variable was returning false if the variable's
3484 value was "0", contrary to what the specification has always said!
3485 The result should be true unless the variable is empty.
3487 PH/09 The syntax error of a character other than { following "${if
3488 def:variable_name" (after optional whitespace) was not being diagnosed.
3489 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
3490 accidental colon was present, for example, could give incorrect results.
3492 PH/10 Tidied the code in a number of places where the st_size field of a stat()
3493 result is used (not including appendfile, where other changes are about
3496 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
3497 This involved changing a lot of size variables from int to off_t. It
3498 should work with maildirs and everything.
3500 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
3501 spamd dying while we are connected to it.
3503 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
3506 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
3507 function for BDB 4.3.
3509 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
3511 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
3512 the sources for PCRE have been re-arranged and more files are now
3515 PH/15 The code I had for printing potentially long long variables in PH/11
3516 above was not the best (it lost precision). The length of off_t variables
3517 is now inspected at build time, and an appropriate printing format (%ld
3518 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
3519 to be "long long int" or "long int". This is needed for the internal
3520 formatting function string_vformat().
3522 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
3523 the configuration file to be ":syslog", then the script "guesses" where
3524 the logs files are, rather than using the compiled in default. In our
3525 case the guess is not the same as the compiled default, so the script
3526 suddenly stopped working when I started to use syslog. The patch checks
3527 to see if log_file_path is "". If so, it attempts to read it from exim
3528 with no configuration file to get the compiled in version, before it
3529 falls back to the previous guessing code."
3531 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
3532 implementing BATV in an Exim configuration. See NewStuff for the gory
3535 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
3536 Makefile that are specific to HP-UX.
3538 PH/18 If the "use_postmaster" option was set for a recipient callout together
3539 with the "random" option, the postmaster address was used as the MAIL
3540 FROM address for the random test, but not for the subsequent recipient
3541 test. It is now used for both.
3543 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
3544 patch removes a few documentation additions to RFC 3028, because the
3545 latest draft now contains them. It adds the new en;ascii-case comparator
3546 and a new error check for 8bit text in MIME parts. Comparator and
3547 require names are now matched exactly. I enabled the subaddress
3548 extension, but it is not well tested yet (read: it works for me)."
3550 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
3551 rework some of the code of TK/09 above to avoid the hardwired use of
3552 "%lld" and "long long". Replaced the call to snprintf() with a call to
3555 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
3556 records point to non-existent hosts", "retry timeout exceeded", and
3557 "retry time not reached for any host after a long failure period".
3559 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
3560 experimental DomainKeys support:
3562 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
3563 (2) On an error such as an illegally used "control", the wrong name for
3564 the control was given.
3566 These problems did NOT occur unless DomainKeys support was compiled.
3568 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
3570 PH/24 Added ${if match_ip condition.
3572 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
3573 databases so that it will be absolutely obvious if a crash occurs in the
3574 DB library. This is a regular occurrence (often caused by mis-matched
3577 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
3578 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
3579 picked up on a user's system that detects such things. There doesn't seem
3580 to be a gcc warning option for this - only an attribute that has to be
3581 put on the function's prototype. It seems that in Fedora Core 4 they have
3582 set this on a number of new functions. No doubt there will be more in due
3585 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
3586 specify a transport. However, if an address that was verified by such a
3587 router was the subject of a callout, Exim crashed because it tried to
3588 read the rcpt_include_affixes from the non-existent transport. Now it
3589 just assumes that the setting of that option is false. This bug was
3590 introduced by 4.51/PH/31.
3592 PH/28 Changed -d+all to exclude +memory, because that information is very
3593 rarely of interest, but it makes the output a lot bigger. People tend to
3594 do -d+all out of habit.
3596 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
3597 code in os-type was giving problems when libc.so lives in lib64, like on
3600 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
3601 aren't the modern standard, and it seems that some systems' include files
3602 don't always have them. Exim was already checking for some of the newer
3603 ones like T_AAAA, and defining it itself. I've added checks for all the
3604 record types that Exim uses.
3606 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
3607 not automatically generating a new one, as it is supposed to. This
3608 prevented TLS from working. If the file did exist, but contained invalid
3609 data, a new version was generated, as expected. It was only the case of a
3610 non-existent file that was broken.
3612 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
3613 with a change in libdomainkeys > 0.64.
3615 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
3616 from DNS. If the selector record carries the flag, it now has
3617 precedence over the domain-wide flag.
3619 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
3621 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
3622 the use of a port name that isn't defined in /etc/services) occurred, the
3623 message was deferred as in a normal delivery, and thus remained on the
3624 spool, instead of being failed because of the mua_wrapper setting. This
3625 is now fixed, and I tidied up some of the mua_wrapper messages at the
3628 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
3629 the messages in a hash of arrays rather than using individual hashes.
3630 This is a bit cleaner and results in dramatic memory savings, albeit
3631 at a slight CPU cost.
3633 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
3634 as requested by Marc Sherman.
3636 SC/10 Eximstats: added histograms for user specified patterns as requested
3639 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
3641 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
3642 fopen() in the content-scanning modules that did not already have it.
3648 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
3649 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
3651 TK/02 Fix ACL "control" statement not being available in MIME ACL.
3653 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
3655 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
3656 to test Sieve filters that use "vacation".
3658 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
3659 that changes the way the GnuTLS parameters are stored in the cache file.
3660 The new format can be generated externally. For backward compatibility,
3661 if the data in the cache doesn't make sense, Exim assumes it has read an
3662 old-format file, and it generates new data and writes a new file. This
3663 means that you can't go back to an older release without removing the
3666 PH/03 A redirect router that has both "unseen" and "one_time" set does not
3667 work if there are any delivery delays because "one_time" forces the
3668 parent to be marked "delivered", so its unseen clone is never tried
3669 again. For this reason, Exim now forbids the simultaneous setting of
3672 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
3673 redirected to themselves ("homonym" addresses). Read the long ChangeLog
3674 entry if you want to know the details. The fix, however, neglected to
3675 consider the case when local delivery batching is involved. The test for
3676 "previously delivered" was not happening when checking to see if an
3677 address could be batched with a previous (undelivered) one; under
3678 certain circumstances this could lead to multiple deliveries to the same
3681 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
3682 in its include files, and this causes problems building Exim.
3684 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
3685 header_syntax) but Exim was just ignoring anything given after a slash.
3686 In particular, this caused confusion with an attempt to use "verify =
3687 reverse_host_lookup/defer_ok". An error is now given when options are
3688 supplied for verify items that do not have them. (Maybe reverse_host_
3689 lookup should have a defer_ok option, but that's a different point.)
3691 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
3692 defined by RFC 821) to 2048, because there were problems with some AUTH
3693 commands, and RFC 1869 says the size should be increased for extended
3694 SMTP commands that take arguments.
3696 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
3699 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
3700 "unknown" error; now it says that the functionality isn't in the binary.
3702 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
3703 an address' error message when a string expansion fails (syntax or
3704 whatever). Otherwise the password may appear in the log. Following change
3705 PH/42 below, there is no longer a chance of it appearing in a bounce
3708 PH/11 Installed exipick version 20050225.0 from John Jetmore.
3710 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
3711 of its addresses was ever tried. (Bugzilla bug #2.)
3713 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
3714 the result incorrectly in the debug output. (It correctly added a newline
3715 to what was transported.)
3717 TF/01 Added $received_time.
3719 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
3720 commented out examples of how to interface to a virus scanner and to
3721 SpamAssassin. Also added commented examples of av_scanner and
3722 spamd_address settings.
3724 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
3725 and controls are allowed in which ACLs. There were a couple of minor
3726 errors. Some of the entries in the conditions table (which is a table of
3727 where they are NOT allowed) were getting very unwieldy; rewrote them as a
3728 negation of where the condition IS allowed.
3730 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
3732 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
3733 header file does not have a version number, so I've had to invent a new
3734 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
3735 API. The code is untested by me (my Linux distribution still has 0.3.2 of
3736 radiusclient), but it was contributed by a Radius user.
3738 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
3739 files or directories, for OpenSSL.
3741 PH/19 When an Exim process that is running as root has to create an Exim log
3742 file, it does so in a subprocess that runs as exim:exim so as to get the
3743 ownership right at creation (otherwise, other Exim processes might see
3744 the file with the wrong ownership). There was no test for failure of this
3745 fork() call, which would lead to the process getting stuck as it waited
3746 for a non-existent subprocess. Forks do occasionally fail when resources
3747 run out. I reviewed all the other calls to fork(); they all seem to check
3750 PH/20 When checking for unexpected SMTP input at connect time (before writing
3751 the banner), Exim was not dealing correctly with a non-positive return
3752 from the read() function. If the client had disconnected by this time,
3753 the result was a log entry for a synchronization error with an empty
3754 string after "input=" when read() returned zero. If read() returned -1
3755 (an event I could not check), uninitialized data bytes were printed.
3756 There were reports of junk text (parts of files, etc) appearing after
3759 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
3761 PH/22 Added support for macro redefinition, and (re)definition in between
3762 driver and ACL definitions.
3764 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
3765 forgetting to use the resulting value; it was using the unexpanded value.
3767 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
3768 hadn't been configured. The fix is from Juergen Kreileder, who
3769 understands it better than I do:
3771 "Here's what I see happening with three configured cyrus_sasl
3772 authenticators configured (plain, login, cram-md5):
3774 On startup auth_cyrus_sasl_init() gets called for each of these.
3775 This means three calls to sasl_listmech() without a specified mech_list.
3776 => SASL tests which mechs of all available mechs actually work
3777 => three warnings about OTP not working
3778 => the returned list contains: plain, login, cram-md5, digest-md5, ...
3780 With the patch, sasl_listmech() also gets called three times. But now
3781 SASL's mech_list option is set to the server_mech specified in the the
3782 authenticator. Or in other words, the answer from sasl_listmech()
3783 gets limited to just the mech you're testing for (which is different
3785 => the return list contains just 'plain' or 'login', 'cram-md5' or
3786 nothing depending on the value of ob->server_mech.
3788 I've just tested the patch: Authentication still works fine,
3789 unavailable mechs specified in the exim configuration are still
3790 caught, and the auth.log warnings about OTP are gone."
3792 PH/25 When debugging is enabled, the contents of the command line are added
3793 to the debugging output, even when log_selector=+arguments is not
3796 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
3797 answer is "GNU", and only if the return is "GNU/something" is the answer
3800 PH/27 $acl_verify_message is now set immediately after the failure of a
3801 verification in an ACL, and so is available in subsequent modifiers. In
3802 particular, the message can be preserved by coding like this:
3804 warn !verify = sender
3805 set acl_m0 = $acl_verify_message
3807 Previously, $acl_verify_message was set only while expanding "message"
3808 and "log_message" when a very denied access.
3810 PH/28 Modified OS/os.c-Linux with
3812 -#ifndef OS_LOAD_AVERAGE
3813 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
3815 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
3816 nomenclature these days.)
3818 PH/29 Installed patch from the Sieve maintainer that adds the options
3819 sieve_useraddress and sieve_subaddress to the redirect router.
3821 PH/30 In these circumstances:
3822 . Two addresses routed to the same list of hosts;
3823 . First host does not offer TLS;
3824 . First host accepts first address;
3825 . First host gives temporary error to second address;
3826 . Second host offers TLS and a TLS session is established;
3827 . Second host accepts second address.
3828 Exim incorrectly logged both deliveries with the TLS parameters (cipher
3829 and peerdn, if requested) that were in fact used only for the second
3832 PH/31 When doing a callout as part of verifying an address, Exim was not paying
3833 attention to any local part prefix or suffix that was matched by the
3834 router that accepted the address. It now behaves in the same way as it
3835 does for delivery: the affixes are removed from the local part unless
3836 rcpt_include_affixes is set on the transport.
3838 PH/32 Add the sender address, as F=<...>, to the log line when logging a
3839 timeout during the DATA phase of an incoming message.
3841 PH/33 Sieve envelope tests were broken for match types other than :is. I have
3842 applied a patch sanctioned by the Sieve maintainer.
3844 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
3845 the uid or gid is negative. A case of a negative gid caused this to be
3846 noticed. The fix allows for either to be negative.
3848 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
3849 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
3852 PH/36 The change PH/12 above was broken. Fixed it.
3854 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
3855 the grounds that routing the same address twice would always produce the
3856 same answer. This might have been true once, but it is certainly no
3857 longer true now. Routing a child address may depend on the previous
3858 routing that produced that child. Some complicated redirection strategies
3859 went wrong when messages had multiple recipients, and made Exim's
3860 behaviour dependent on the order in which the addresses were given.
3862 I have moved the duplicate checking until after the routing is complete.
3863 Exim scans the addresses that are assigned to local and remote
3864 transports, and removes any duplicates. This means that more work will be
3865 done, as duplicates will always all be routed, but duplicates are
3866 presumably rare, so I don't expect this is of any significance.
3868 For deliveries to pipes, files, and autoreplies, the duplicate checking
3869 still happens during the routing process, since they are not going to be
3872 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
3873 It corrects a timeout issue with spamd. This is Ian's comment: "The
3874 background is that sometimes spamd either never reads data from a
3875 connection it has accepted, or it never writes response data. The exiscan
3876 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
3877 blindly assumes that writes won't block so it may never time out."
3879 PH/39 Allow G after quota size as well as K and M.
3881 PH/40 The value set for $authenticated_id in an authenticator may not contain
3882 binary zeroes or newlines because the value is written to log lines and
3883 to spool files. There was no check on this. Now the value is run through
3884 the string_printing() function so that such characters are converted to
3885 printable escape sequences.
3887 PH/41 $message_linecount is a new variable that contains the total number of
3888 lines in the message. Compare $body_linecount, which is the count for the
3891 PH/42 Exim no longer gives details of delivery errors for specific addresses in
3892 bounce and delay warning messages, except in certain special cases, which
3895 (a) An SMTP error message from a remote host;
3896 (b) A message specified in a :fail: redirection;
3897 (c) A message specified in a "fail" command in a system filter;
3898 (d) A message specified in a FAIL return from the queryprogram router;
3899 (e) A message specified by the cannot_route_message router option.
3901 In these cases only, Exim does include the error details in bounce and
3902 warning messages. There are also a few cases where bland messages such
3903 as "unrouteable address" or "local delivery error" are given.
3905 PH/43 $value is now also set for the "else" part of a ${run expansion.
3907 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
3908 being worked on, but at least Exim now implements the latest version to
3911 PH/45 In a pipe transport, although a timeout while waiting for the pipe
3912 process to complete was treated as a delivery failure, a timeout while
3913 writing the message to the pipe was logged, but erroneously treated as a
3914 successful delivery. Such timeouts include transport filter timeouts. For
3915 consistency with the overall process timeout, these timeouts are now
3916 treated as errors, giving rise to delivery failures by default. However,
3917 there is now a new Boolean option for the pipe transport called
3918 timeout_defer, which, if set TRUE, converts the failures into defers for
3919 both kinds of timeout. A transport filter timeout is now identified in
3922 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
3923 systems where "make" and "gmake" are different, calling "gmake" at top
3924 level broke things. I've arranged for the value of $(MAKE) to be passed
3925 from the Makefile to this script so that it can call the same version of
3929 A note about Exim versions 4.44 and 4.50
3930 ----------------------------------------
3932 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
3933 changes of various kinds. As a consequence, a big documentation update was
3934 needed. This delayed the release for rather longer than seemed good, especially
3935 in the light of a couple of (minor) security issues. Therefore, the changes
3936 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
3937 release. So 4.44 and 4.50 are in effect two different branches that both start
3940 I have left the 4.50 change log unchanged; it contains all the changes since
3941 4.43. The change log for 4.44 is below; many of its items are identical to
3942 those for 4.50. This seems to be the most sensible way to preserve the
3943 historical information.
3949 1. Minor wording change to the doc/README.SIEVE file.
3951 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
3952 computation of the current number of files was incorrect.
3954 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
3955 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
3958 4. Give more explanation in the error message when the command for a transport
3959 filter fails to execute.
3961 5. There are several places where Exim runs a non-Exim command in a
3962 subprocess. The SIGUSR1 signal should be disabled for these processes. This
3963 was being done only for the command run by the queryprogram router. It is
3964 now done for all such subprocesses. The other cases are: ${run, transport
3965 filters, and the commands run by the lmtp and pipe transports.
3967 6. Added CONFIGURE_GROUP build-time option.
3969 7. Some older OS have a limit of 256 on the maximum number of file
3970 descriptors. Exim was using setrlimit() to set 1000 as a large value
3971 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
3972 systems. I've change it so that if it can't get 1000, it tries for 256.
3974 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
3975 was an oversight, and furthermore, ever since the addition of extra
3976 controls (e.g. 4.43/32), the checks on when to allow different forms of
3977 "control" were broken. There should now be diagnostics for all cases when a
3978 control that does not make sense is encountered.
3980 9. Added the /retain_sender option to "control=submission".
3982 10. $recipients is now available in the predata ACL (oversight).
3984 11. Tidy the search cache before the fork to do a delivery from a message
3985 received from the command line. Otherwise the child will trigger a lookup
3986 failure and thereby defer the delivery if it tries to use (for example) a
3987 cached ldap connection that the parent has called unbind on.
3989 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
3990 of $address_data from the recipient verification was clobbered by the
3991 sender verification.
3993 13. The value of address_data from a sender verification is now available in
3994 $sender_address_data in subsequent conditions in the ACL statement.
3996 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
3998 15. Added a new option "connect=<time>" to callout options, to set a different
4001 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4002 was its contents. (It was OK if the option was not defined at all.)
4004 17. A "Completed" log line is now written for messages that are removed from
4005 the spool by the -Mrm option.
4007 18. New variables $sender_verify_failure and $recipient_verify_failure contain
4008 information about exactly what failed.
4010 19. Added -dd to debug only the daemon process.
4012 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
4013 handles timeouts, both on the server side and network timeouts. Renamed the
4014 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
4016 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
4017 It is now set to "smtps".
4019 22. $host_address is now set to the target address during the checking of
4020 ignore_target_hosts.
4022 23. When checking ignore_target_hosts for an ipliteral router, no host name was
4023 being passed; this would have caused $sender_host_name to have been used if
4024 matching the list had actually called for a host name (not very likely,
4025 since this list is usually IP addresses). A host name is now passed as
4028 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4029 code that specifies a non-restarting handler (typically sigaction() in
4030 modern systems) in an attempt to fix a rare and obscure crash bug.
4032 25. Narrowed the window for a race in the daemon that could cause it to ignore
4033 SIGCHLD signals. This is not a major problem, because they are used only to
4034 wake it up if nothing else does.
4036 26. A malformed maildirsize file could cause Exim to calculate negative values
4037 for the mailbox size or file count. Odd effects could occur as a result.
4038 The maildirsize information is now recalculated if the size or filecount
4041 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4042 support for a long time. Removed HAVE_SYS_VFS_H.
4044 28. Installed the latest version of exipick from John Jetmore.
4046 29. In an address list, if the pattern was not a regular expression, an empty
4047 subject address (from a bounce message) matched only if the pattern was an
4048 empty string. Non-empty patterns were not even tested. This was the wrong
4049 because it is perfectly reasonable to use an empty address as part of a
4050 database query. An empty address is now tested by patterns that are
4051 lookups. However, all the other forms of pattern expect the subject to
4052 contain a local part and a domain, and therefore, for them, an empty
4053 address still always fails if the pattern is not itself empty.
4055 30. Exim went into a mad DNS loop when attempting to do a callout where the
4056 host was specified on an smtp transport, and looking it up yielded more
4057 than one IP address.
4059 31. Re-factored the code for checking spool and log partition space into a
4060 function that finds that data and another that does the check. The former
4061 is then used to implement four new variables: $spool_space, $log_space,
4062 $spool_inodes, and $log_inodes.
4064 32. The RFC2047 encoding function was originally intended for short strings
4065 such as real names; it was not keeping to the 75-character limit for
4066 encoded words that the RFC imposes. It now respects the limit, and
4067 generates multiple encoded words if necessary. To be on the safe side, I
4068 have increased the buffer size for the ${rfc2047: expansion operator from
4071 33. It is now permitted to omit both strings after an "if" condition; if the
4072 condition is true, the result is "true". As before, when the second string
4073 is omitted, a false condition yields an empty string. This makes it less
4074 cumbersome to write custom ACL and router conditions.
4076 34. Failure to deliver a bounce message always caused it to be frozen, even if
4077 there was an errors_to setting on the router. The errors_to setting is now
4080 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4081 canonical form (fully expanded) before being placed in
4082 $sender_host_address.
4084 36. The table in the code that translates DNS record types into text (T_A to
4085 "A" for instance) was missing entries for NS and CNAME. It is just possible
4086 that this could have caused confusion if both these types were looked up
4087 for the same domain, because the text type is used as part of Exim's
4088 per-process caching. But the chance of anyone hitting this buglet seems
4091 37. The dnsdb lookup has been extended in a number of ways.
4093 (1) There is a new type, "zns", which walks up the domain tree until it
4094 finds some nameserver records. It should be used with care.
4096 (2) There is a new type, "mxh", which is like "mx" except that it returns
4097 just the host names, not the priorities.
4099 (3) It is now possible to give a list of domains (or IP addresses) to be
4100 looked up. The behaviour when one of the lookups defers can be
4101 controlled by a keyword.
4103 (4) It is now possible to specify the separator character for use when
4104 multiple records are returned.
4106 38. The dnslists ACL condition has been extended: it is now possible to supply
4107 a list of IP addresses and/or domains to be looked up in a particular DNS
4110 39. Added log_selector=+queue_time_overall.
4112 40. When running the queue in the test harness, wait just a tad after forking a
4113 delivery process, to get repeatability of debugging output.
4115 41. Include certificate and key file names in error message when GnuTLS fails
4116 to set them up, because the GnuTLS error message doesn't include the name
4117 of the failing file when there is a problem reading it.
4119 42. Allow both -bf and -bF in the same test run.
4121 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
4123 44. The "Exiscan patch" is now merged into the mainline Exim source.
4125 45. Sometimes the final signoff response after QUIT could fail to get
4126 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4127 < 0 before doing a fflush(). This bug looks as though it goes back to the
4128 introduction of TLS in release 3.20, but "sometimes" must have been rare
4129 because the tests only now provoked it.
4131 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
4132 (this can affect the format of dates).
4134 47. exim_tidydb, when checking for the continued existence of a message for
4135 which it has found a message-specific retry record, was not finding
4136 messages that were in split spool directories. Consequently, it was
4137 deleting retry records that should have stayed in existence.
4139 48. Steve fixed some bugs in eximstats.
4141 49. The SPA authentication driver was not abandoning authentication and moving
4142 on to the next authenticator when an expansion was forced to fail,
4143 contradicting the general specification for all authenticators. Instead it
4144 was generating a temporary error. It now behaves as specified.
4146 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
4147 (the order specifies the preference for clients). The order is now AES256,
4148 AES128, 3DES, ARCFOUR128.
4150 51. Small patch to Sieve code - explicitly set From: when generating an
4153 52. Exim crashed if a remote delivery caused a very long error message to be
4154 recorded - for instance if somebody sent an entire SpamAssassin report back
4155 as a large number of 550 error lines. This bug was coincidentally fixed by
4156 increasing the size of one of Exim's internal buffers (big_buffer) that
4157 happened as part of the Exiscan merge. However, to be on the safe side, I
4158 have made the code more robust (and fixed the comments that describe what
4161 53. Now that there can be additional text after "Completed" in log lines (if
4162 the queue_time_overall log selector is set), a one-byte patch to exigrep
4163 was needed to allow it to recognize "Completed" as not the last thing in
4166 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
4167 patch that reportedly fixes this has been added. I am not expert enough to
4168 create a test for it. This is what the patch creator wrote:
4170 "I found a little strange behaviour of ldap code when working with
4171 Windows 2003 AD Domain, where users was placed in more than one
4172 Organization Units. When I tried to give exim partial DN, the exit code
4173 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
4174 But simultaneously result of request was absolutely normal ldap result,
4175 so I produce this patch..."
4177 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
4178 REFERENCE, so I have modified the code to exclude the patch when that macro
4181 55. Some experimental protocols are using DNS PTR records for new purposes. The
4182 keys for these records are domain names, not reversed IP addresses. The
4183 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4184 leaves it alone. Component reversal etc. now happens only for IP addresses.
4187 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
4189 57. Double the size of the debug message buffer (to 2048) so that more of very
4190 long debug lines gets shown.
4192 58. The exicyclog utility now does better if the number of log files to keep
4193 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
4195 59. Two changes related to the smtp_active_hostname option:
4197 (1) $smtp_active_hostname is now available as a variable.
4198 (2) The default for smtp_banner uses $smtp_active_hostname instead
4199 of $primary_hostname.
4201 60. The host_aton() function is supposed to be passed a string that is known
4202 to be a valid IP address. However, in the case of IPv6 addresses, it was
4203 not checking this. This is a hostage to fortune. Exim now panics and dies
4204 if the condition is not met. A case was found where this could be provoked
4205 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4206 components; fortuitously, this particular loophole had already been fixed
4207 by change 4.50/55 above.
4209 If there are any other similar loopholes, the new check in host_aton()
4210 itself should stop them being exploited. The report I received stated that
4211 data on the command line could provoke the exploit when Exim was running as
4212 exim, but did not say which command line option was involved. All I could
4213 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4214 running as the user.
4217 61. There was a buffer overflow vulnerability in the SPA authentication code
4218 (which came originally from the Samba project). I have added a test to the
4219 spa_base64_to_bits() function which I hope fixes it.
4222 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
4223 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
4225 63. The daemon start-up calls getloadavg() while still root for those OS that
4226 need the first call to be done as root, but it missed one case: when
4227 deliver_queue_load_max is set with deliver_drop_privilege. This is
4228 necessary for the benefit of the queue runner, because there is no re-exec
4229 when deliver_drop_privilege is set.
4231 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
4232 This has been fixed.
4234 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4235 was in use, was not putting the data itself into the right store pool;
4236 consequently, it could be overwritten for a subsequent message in the same
4237 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4240 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
4242 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
4243 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
4244 the function were treating the return as a boolean value, which happened to
4245 work because 0=false and not-0=true, but is not correct code.
4247 68. The host_aton() function was not handling scoped IPv6 addresses (those
4248 with, for example, "%eth0" on the end) correctly.
4250 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
4251 negated items (that is, ~something) in unsigned ints. Some compilers
4252 apparently mutter when there is no cast.
4254 70. If an address verification called from an ACL failed, and did not produce a
4255 user-specific message (i.e. there was only a "system" message), nothing was
4256 put in $acl_verify_message. In this situation, it now puts the system
4259 71. Change 4.23/11 added synchronization checking at the start of an SMTP
4260 session; change 4.31/43 added the unwanted input to the log line - except
4261 that it did not do this in the start of session case. It now does.
4263 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
4264 This is wrong and can cause the other end to generate a synchronization
4265 error if it is another Exim or anything else that does the synchronization
4266 check. A QUIT command is no longer sent after a timeout.
4268 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
4269 during host lookups.
4271 74. The defer_ok option of callout verification was not working if it was used
4272 when verifying addresses in header lines, that is, for this case:
4274 verify = header_sender/callout=defer_ok
4276 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
4277 those file descriptors could be used for SMTP connections. If anything
4278 wrote to stderr (the example that came up was "warn" in embedded Perl), it
4279 could be sent to the SMTP client, causing chaos. The daemon now opens
4280 stdin, stdout, and stderr to /dev/null when it puts itself into the
4283 76. Arrange for output from Perl's "warn" command to be written to Exim's main
4284 log by default. The user can override this with suitable Perl magic.
4286 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
4287 the log message when discard triggers, was not working for the DATA ACL or
4288 for the non-SMTP ACL.
4290 78. Error message wording change in sieve.c.
4292 79. If smtp_accept_max_per_host was set, the number of connections could be
4293 restricted to fewer than expected, because the daemon was trying to set up
4294 a new connection before checking whether the processes handling previous
4295 connections had finished. The check for completed processes is now done
4296 earlier. On busy systems, this bug wouldn't be noticed because something
4297 else would have woken the daemon, and it would have reaped the completed
4300 80. If a message was submitted locally by a user whose login name contained one
4301 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
4302 It caused a spool format error. I have fixed the spool reading code. A
4303 related problem was that the "from" clause in the Received: line became
4304 illegal because of the space(s). It is now covered by ${quote_local_part.
4306 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
4309 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
4311 83. Patch from Sieve maintainer to fix unterminated string problem in
4312 "vacation" handling.
4314 84. Some minor changes to the Linux configuration files to help with other
4315 OS variants using glibc.
4317 85. One more patch for Sieve to update vacation handling to latest spec.
4320 ----------------------------------------------------
4321 See the note above about the 4.44 and 4.50 releases.
4322 ----------------------------------------------------
4328 1. Change 4.43/35 introduced a bug that caused file counts to be
4329 incorrectly computed when quota_filecount was set in an appendfile
4332 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4333 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4336 3. Give more explanation in the error message when the command for a transport
4337 filter fails to execute.
4339 4. There are several places where Exim runs a non-Exim command in a
4340 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4341 was being done only for the command run by the queryprogram router. It is
4342 now done for all such subprocesses. The other cases are: ${run, transport
4343 filters, and the commands run by the lmtp and pipe transports.
4345 5. Some older OS have a limit of 256 on the maximum number of file
4346 descriptors. Exim was using setrlimit() to set 1000 as a large value
4347 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4348 systems. I've change it so that if it can't get 1000, it tries for 256.
4350 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4351 was an oversight, and furthermore, ever since the addition of extra
4352 controls (e.g. 4.43/32), the checks on when to allow different forms of
4353 "control" were broken. There should now be diagnostics for all cases when a
4354 control that does not make sense is encountered.
4356 7. $recipients is now available in the predata ACL (oversight).
4358 8. Tidy the search cache before the fork to do a delivery from a message
4359 received from the command line. Otherwise the child will trigger a lookup
4360 failure and thereby defer the delivery if it tries to use (for example) a
4361 cached ldap connection that the parent has called unbind on.
4363 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4364 of $address_data from the recipient verification was clobbered by the
4365 sender verification.
4367 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4368 was its contents. (It was OK if the option was not defined at all.)
4370 11. A "Completed" log line is now written for messages that are removed from
4371 the spool by the -Mrm option.
4373 12. $host_address is now set to the target address during the checking of
4374 ignore_target_hosts.
4376 13. When checking ignore_target_hosts for an ipliteral router, no host name was
4377 being passed; this would have caused $sender_host_name to have been used if
4378 matching the list had actually called for a host name (not very likely,
4379 since this list is usually IP addresses). A host name is now passed as
4382 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4383 code that specifies a non-restarting handler (typically sigaction() in
4384 modern systems) in an attempt to fix a rare and obscure crash bug.
4386 15. Narrowed the window for a race in the daemon that could cause it to ignore
4387 SIGCHLD signals. This is not a major problem, because they are used only to
4388 wake it up if nothing else does.
4390 16. A malformed maildirsize file could cause Exim to calculate negative values
4391 for the mailbox size or file count. Odd effects could occur as a result.
4392 The maildirsize information is now recalculated if the size or filecount
4395 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4396 support for a long time. Removed HAVE_SYS_VFS_H.
4398 18. Updated exipick to current release from John Jetmore.
4400 19. Allow an empty sender to be matched against a lookup in an address list.
4401 Previously the only cases considered were a regular expression, or an
4404 20. Exim went into a mad DNS lookup loop when doing a callout where the
4405 host was specified on the transport, if the DNS lookup yielded more than
4408 21. The RFC2047 encoding function was originally intended for short strings
4409 such as real names; it was not keeping to the 75-character limit for
4410 encoded words that the RFC imposes. It now respects the limit, and
4411 generates multiple encoded words if necessary. To be on the safe side, I
4412 have increased the buffer size for the ${rfc2047: expansion operator from
4415 22. Failure to deliver a bounce message always caused it to be frozen, even if
4416 there was an errors_to setting on the router. The errors_to setting is now
4419 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4420 canonical form (fully expanded) before being placed in
4421 $sender_host_address.
4423 24. Updated eximstats to version 1.33
4425 25. Include certificate and key file names in error message when GnuTLS fails
4426 to set them up, because the GnuTLS error message doesn't include the name
4427 of the failing file when there is a problem reading it.
4429 26. Expand error message when OpenSSL has problems setting up cert/key files.
4432 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
4433 (this can affect the format of dates).
4435 28. exim_tidydb, when checking for the continued existence of a message for
4436 which it has found a message-specific retry record, was not finding
4437 messages that were in split spool directories. Consequently, it was
4438 deleting retry records that should have stayed in existence.
4440 29. eximstats updated to version 1.35
4441 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
4442 1.35 - bugfix such that pie charts by volume are generated correctly
4444 30. The SPA authentication driver was not abandoning authentication and moving
4445 on to the next authenticator when an expansion was forced to fail,
4446 contradicting the general specification for all authenticators. Instead it
4447 was generating a temporary error. It now behaves as specified.
4449 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
4450 (the order specifies the preference for clients). The order is now AES256,
4451 AES128, 3DES, ARCFOUR128.
4453 31. Small patch to Sieve code - explicitly set From: when generating an
4456 32. Exim crashed if a remote delivery caused a very long error message to be
4457 recorded - for instance if somebody sent an entire SpamAssassin report back
4458 as a large number of 550 error lines. This bug was coincidentally fixed by
4459 increasing the size of one of Exim's internal buffers (big_buffer) that
4460 happened as part of the Exiscan merge. However, to be on the safe side, I
4461 have made the code more robust (and fixed the comments that describe what
4464 33. Some experimental protocols are using DNS PTR records for new purposes. The
4465 keys for these records are domain names, not reversed IP addresses. The
4466 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4467 leaves it alone. Component reversal etc. now happens only for IP addresses.
4470 34. The host_aton() function is supposed to be passed a string that is known
4471 to be a valid IP address. However, in the case of IPv6 addresses, it was
4472 not checking this. This is a hostage to fortune. Exim now panics and dies
4473 if the condition is not met. A case was found where this could be provoked
4474 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4475 components; fortuitously, this particular loophole had already been fixed
4476 by change 4.50/55 or 4.44/33 above.
4478 If there are any other similar loopholes, the new check in host_aton()
4479 itself should stop them being exploited. The report I received stated that
4480 data on the command line could provoke the exploit when Exim was running as
4481 exim, but did not say which command line option was involved. All I could
4482 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4483 running as the user.
4486 35. There was a buffer overflow vulnerability in the SPA authentication code
4487 (which came originally from the Samba project). I have added a test to the
4488 spa_base64_to_bits() function which I hope fixes it.
4491 36. The daemon start-up calls getloadavg() while still root for those OS that
4492 need the first call to be done as root, but it missed one case: when
4493 deliver_queue_load_max is set with deliver_drop_privilege. This is
4494 necessary for the benefit of the queue runner, because there is no re-exec
4495 when deliver_drop_privilege is set.
4497 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4498 was in use, was not putting the data itself into the right store pool;
4499 consequently, it could be overwritten for a subsequent message in the same
4500 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4503 38. Sometimes the final signoff response after QUIT could fail to get
4504 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4505 < 0 before doing a fflush(). This bug looks as though it goes back to the
4506 introduction of TLS in release 3.20, but "sometimes" must have been rare
4507 because the tests only now provoked it.
4513 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
4514 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
4515 Now it returns an integer. A number of calls were still expecting a T/F
4516 return. Fortuitously, in all cases, the tests worked in OK situations,
4517 which is the norm. However, things would have gone wrong on any write
4518 failures on the smtp file descriptor. This function is used when sending
4519 messages over SMTP and also when doing verify callouts.
4521 2. When Exim is called to do synchronous delivery of a locally submitted
4522 message (the -odf or -odi options), it no longer closes stderr before doing
4525 3. Implemented the mua_wrapper option.
4527 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
4529 5. Implemented the functions header_remove(), header_testname(),
4530 header_add_at_position(), and receive_remove_recipient(), and exported them
4533 6. If an ACL "warn" statement specified the addition of headers, Exim already
4534 inserted X-ACL-Warn: at the start if there was no header name. However, it
4535 was not making this test for the second and subsequent header lines if
4536 there were newlines in the string. This meant that an invalid header could
4537 be inserted if Exim was badly configured.
4539 7. Allow an ACL "warn" statement to add header lines at the start or after all
4540 the Received: headers, as well as at the end.
4542 8. Added the rcpt_4xx retry error code.
4544 9. Added postmaster_mailfrom=xxx to callout verification option.
4546 10. Added mailfrom=xxxx to the callout verification option, for verify=
4549 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
4550 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
4552 12. Inserted some casts to stop certain compilers warning when using pointer
4553 differences as field lengths or precisions in printf-type calls (mostly
4554 affecting debugging statements).
4556 13. Added optional readline() support for -be (dynamically loaded).
4558 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
4559 same clock tick as a message's arrival, so that its received time was the
4560 same as the "first fail" time on the retry record, and that message
4561 remained on the queue past the ultimate address timeout, every queue runner
4562 would try a delivery (because it was past the ultimate address timeout) but
4563 after another failure, the ultimate address timeout, which should have then
4564 bounced the address, did not kick in. This was a "< instead of <=" error;
4565 in most cases the first failure would have been in the next clock tick
4566 after the received time, and all would be well.
4568 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
4569 being recognized when the domain list was tested by the match_domain
4570 condition in an expansion string.
4572 16. Added the ${str2b64: operator.
4574 17. Exim was always calling setrlimit() to set a large limit for the number of
4575 processes, without checking whether the existing limit was already
4576 adequate. (It did check for the limit on file descriptors.) Furthermore,
4577 errors from getrlimit() and setrlimit() were being ignored. Now they are
4578 logged to the main and panic logs, but Exim does carry on, to try to do its
4579 job under whatever limits there are.
4581 18. Imported PCRE 5.0.
4583 19. Trivial typo in log message " temporarily refused connection" (the leading
4586 20. If the log selector return_path_on_delivery was set and an address was
4587 redirected to /dev/null, the delivery process crashed because it assumed
4588 that a return path would always be set for a "successful" delivery. In this
4589 case, the whole delivery is bypassed as an optimization, and therefore no
4592 21. Internal re-arrangement: the function for sending a challenge and reading
4593 a response while authentication was assuming a zero-terminated challenge
4594 string. It's now changed to take a pointer and a length, to allow for
4595 binary data in such strings.
4597 22. Added the cyrus_sasl authenticator (code supplied by MBM).
4599 23. Exim was not respecting finduser_retries when seeking the login of the
4600 uid under which it was called; it was always trying 10 times. (The default
4601 setting of finduser_retries is zero.) Also, it was sleeping after the final
4602 failure, which is pointless.
4604 24. Implemented tls_on_connect_ports.
4606 25. Implemented acl_smtp_predata.
4608 26. If the domain in control=submission is set empty, Exim assumes that the
4609 authenticated id is a complete email address when it generates From: or
4610 Sender: header lines.
4612 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
4613 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
4614 chown and chgrp in /bin and hostname in /usr/bin.
4616 28. Exim was keeping the "process log" file open after each use, just as it
4617 does for the main log. This opens the possibility of it remaining open for
4618 long periods when the USR1 signal hits a daemon. Occasional processlog
4619 errors were reported, that could have been caused by this. Anyway, it seems
4620 much more sensible not to leave this file open at all, so that is what now
4623 29. The long-running daemon process does not normally write to the log once it
4624 has entered its main loop, and it closes the log before doing so. This is
4625 so that log files can straightforwardly be renamed and moved. However,
4626 there are a couple of unusual error situations where the daemon does write
4627 log entries, and I had neglected to close the log afterwards.
4629 30. The text of an SMTP error response that was received during a remote
4630 delivery was being truncated at 512 bytes. This is too short for some of
4631 the long messages that one sometimes sees. I've increased the limit to
4634 31. It is now possible to make retry rules that apply only when a message has a
4635 specific sender, in particular, an empty sender.
4637 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
4638 it possible to be selective about when SMTP synchronization is enforced.
4640 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
4642 32. Implemented hosts_connection_nolog.
4644 33. Added an ACL for QUIT.
4646 34. Setting "delay_warning=" to disable warnings was not working; it gave a
4649 35. Added mailbox_size and mailbox_filecount to appendfile.
4651 36. Added control = no_multiline_responses to ACLs.
4653 37. There was a bug in the logic of the code that waits for the clock to tick
4654 in the case where the clock went backwards by a substantial amount such
4655 that the microsecond fraction of "now" was more than the microsecond
4656 fraction of "then" (but the whole seconds number was less).
4658 38. Added support for the libradius Radius client library this is found on
4659 FreeBSD (previously only the radiusclient library was supported).
4665 1. When certain lookups returned multiple values in the form name=value, the
4666 quoting of the values was not always being done properly. Specifically:
4667 (a) If the value started with a double quote, but contained no whitespace,
4669 (b) If the value contained whitespace other than a space character (i.e.
4670 tabs or newlines or carriage returns) it was not quoted.
4671 This fix has been applied to the mysql and pgsql lookups by writing a
4672 separate quoting function and calling it from the lookup code. The fix
4673 should probably also be applied to nisplus, ibase and oracle lookups, but
4674 since I cannot test any of those, I have not disturbed their existing code.
4676 2. A hit in the callout cache for a specific address caused a log line with no
4677 reason for rejecting RCPT. Now it says "Previous (cached) callout
4678 verification failure".
4680 3. There was an off-by-one bug in the queryprogram router. An over-long
4681 return line was truncated at 256 instead of 255 characters, thereby
4682 overflowing its buffer with the terminating zero. As well as fixing this, I
4683 have increased the buffer size to 1024 (and made a note to document this).
4685 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
4686 when Exim is waiting for an SMTP response from a remote server, Exim
4687 restarts its select() call on the socket, thereby resetting its timeout.
4688 This is not a problem when such interrupts are rare. Somebody set up a cron
4689 job to run exiwhat every 2 minutes, which is less than the normal select()
4690 timeout (5 or 10 minutes). This meant that the select() timeout never
4691 kicked in because it was always reset. I have fixed this by comparing the
4692 time when an interrupt arrives with the time at the start of the first call
4693 to select(). If more time than the timeout has elapsed, the interrupt is
4694 treated as a timeout.
4696 5. Some internal re-factoring in preparation for the addition of Sieve
4697 extensions (by MH). In particular, the "personal" test is moved to a
4698 separate function, and given an option for scanning Cc: and Bcc: (which is
4699 not set for Exim filters).
4701 6. When Exim created an email address using the login of the caller as the
4702 local part (e.g. when creating a From: or Sender: header line), it was not
4703 quoting the local part when it contained special characters such as @.
4705 7. Installed new OpenBSD configuration files.
4707 8. Reworded some messages for syntax errors in "and" and "or" conditions to
4708 try to make them clearer.
4710 9. Callout options, other than the timeout value, were being ignored when
4711 verifying sender addresses in header lines. For example, when using
4713 verify = header_sender/callout=no_cache
4715 the cache was (incorrectly) being used.
4717 10. Added a missing instance of ${EXE} to the exim_install script; this affects
4718 only the Cygwin environment.
4720 11. When return_path_on_delivery was set as a log selector, if different remote
4721 addresses in the same message used different return paths and parallel
4722 remote delivery occurred, the wrong values would sometimes be logged.
4723 (Whenever a remote delivery process finished, the return path value from
4724 the most recently started remote delivery process was logged.)
4726 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
4727 lines when AUTH and/or TLS are in use. This is the "received protocol"
4728 field. Exim used to use "asmtp" for authenticated SMTP, without any
4729 indication (in the protocol name) for TLS use. Now it follows the RFC and
4730 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
4731 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
4732 names appear in log lines as well as in Received: header lines.
4734 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
4735 extensions, and comparison tests, and to fix some bugs.
4737 14. Changes to the "personal" filter test:
4739 (1) The test was buggy in that it was just doing the equivalent of
4740 "contains" tests on header lines. For example, if a user's address was
4741 anne@some.where, the "personal" test would incorrectly be true for
4743 To: susanne@some.where
4745 This test is now done by extracting each address from the header in turn,
4746 and checking the entire address. Other tests that are part of "personal"
4747 are now done using regular expressions (for example, to check local parts
4748 of addresses in From: header lines).
4750 (2) The list of non-personal local parts in From: addresses has been
4751 extended to include "listserv", "majordomo", "*-request", and "owner-*",
4752 taken from the Sieve specification recommendations.
4754 (3) If the message contains any header line starting with "List-" it is
4755 treated as non-personal.
4757 (4) The test for "circular" in the Subject: header line has been removed
4758 because it now seems ill-conceived.
4760 15. Minor typos in src/EDITME comments corrected.
4762 16. Installed latest exipick from John Jetmore.
4764 17. If headers_add on a router specified a text string that was too long for
4765 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
4766 of string_sprintf() is now avoided.
4768 18. $message_body_size was not set (it was always zero) when running the DATA
4769 ACL and the local_scan() function.
4771 19. For the "mail" command in an Exim filter, no default was being set for
4772 the once_repeat time, causing a random time value to be used if "once" was
4773 specified. (If the value happened to be <= 0, no repeat happened.) The
4774 default is now 0s, meaning "never repeat". The "vacation" command was OK
4775 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
4776 (I found it when inspecting the code).
4778 20. There is now an overall timeout for performing a callout verification. It
4779 defaults to 4 times the callout timeout, which applies to individual SMTP
4780 commands during the callout. The overall timeout applies when there is more
4781 than one host that can be tried. The timeout is checked before trying the
4782 next host. This prevents very long delays if there are a large number of
4783 hosts and all are timing out (e.g. when the network connections are timing
4784 out). The value of the overall timeout can be changed by specifying an
4785 additional sub-option for "callout", called "maxwait". For example:
4787 verify = sender/callout=5s,maxwait=20s
4789 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
4790 to the end before writing, but this should make it even safer).
4792 22. Exim was forgetting that it had advertised PIPELINING for the second and
4793 subsequent messages on an SMTP connection. It was also not resetting its
4794 memory on STARTTLS and an internal HELO.
4796 23. When Exim logs an SMTP synchronization error within a session, it now
4797 records whether PIPELINING has been advertised or not.
4799 24. Added 3 instances of "(long int)" casts to time_t variables that were being
4800 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
4801 rather than long int.
4803 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
4805 26. Added the never_mail option to autoreply.
4811 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
4812 crash if the getsockname() call failed; this can happen if a connection is
4813 closed very soon after it is established. The problem was simply in the
4814 order in which certain operations were done, causing Exim to try to write
4815 to the SMTP stream before it had set up the file descriptor. The bug has
4816 been fixed by making things happen in the correct order.
4822 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
4823 before the connection was closed, thus losing the rejection response.
4825 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
4826 some early Solaris releases, but causes trouble in current releases where
4827 socklen_t is defined.
4829 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
4832 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
4835 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
4836 that releasing the top of it at the end releases what was used for sub-
4837 expansions (unless the block got too big). However, discard this block if
4838 the first thing is a variable or header, so that we can use its block when
4839 it is dynamic (useful for very large $message_headers, for example).
4841 6. Lookups now cache *every* query, not just the most recent. A new, separate
4842 store pool is used for this. It can be recovered when all lookup caches are
4843 flushed. Lookups now release memory at the end of their result strings.
4844 This has involved some general refactoring of the lookup sources.
4846 7. Some code has been added to the store_xxx() functions to reduce the amount
4847 of flapping under certain conditions.
4849 8. log_incoming_interface used to affect only the <= reception log lines. Now
4850 it causes the local interface and port to be added to several more SMTP log
4851 lines, for example "SMTP connection from", and rejection lines.
4853 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
4855 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
4857 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
4858 could be overwritten at the end of the current message (or the start of a
4859 new message if it was set in a HELO ACL). The value is now preserved for
4860 the duration of the SMTP connection.
4862 12. If a transport had a headers_rewrite setting, and a matching header line
4863 contained an unqualified address, that address was qualified, even if it
4864 did not match any rewriting rules. The underlying bug was that the values
4865 of the flags that permit the existence of unqualified sender and recipient
4866 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
4867 non-local messages, and by -bnq for local messages) were not being
4868 preserved with the message after it was received.
4870 13. When Exim was logging an SMTP synchronization error, it could sometimes log
4871 "next input=" as part of the text comprising the host identity instead of
4872 the correct text. The code was using the same buffer for two different
4873 strings. However, depending on which order the printing function evaluated
4874 its arguments, the bug did not always show up. Under Linux, for example, my
4875 test suite worked just fine.
4877 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
4878 doesn't work with some older versions of Perl. It has been changed to "my",
4879 which in any case is probably the better facility to use.
4881 15. A really picky compiler found some instances of statements for creating
4882 error messages that either had too many or two few arguments for the format
4885 16. The size of the buffer for calls to the DNS resolver has been increased
4886 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
4887 for addresses that have a lot of PTR records. This alleviates a problem; it
4888 does not fully solve it.
4890 17. A dnsdb lookup for PTR records that receives more data than will fit in the
4891 buffer now truncates the list and logs the incident, which is the same
4892 action as happens when Exim is looking up a host name and its aliases.
4893 Previously in this situation something unpredictable would happen;
4894 sometimes it was "internal error: store_reset failed".
4896 18. If a server dropped the connection unexpectedly when an Exim client was
4897 using GnuTLS and trying to read a response, the client delivery process
4898 crashed while trying to generate an error log message.
4900 19. If a "warn" verb in an ACL added multiple headers to a message in a single
4901 string, for example:
4903 warn message = H1: something\nH2: something
4905 the text was added as a single header line from Exim's point of view
4906 though it ended up OK in the delivered message. However, searching for the
4907 second and subsequent header lines using $h_h2: did not work. This has been
4908 fixed. Similarly, if a system filter added multiple headers in this way,
4909 the routers could not see them.
4911 20. Expanded the error message when iplsearch is called with an invalid key to
4912 suggest using net-iplsearch in a host list.
4914 21. When running tests using -bh, any delays imposed by "delay" modifiers in
4915 ACLs are no longer actually imposed (and a message to that effect is
4918 22. If a "gecos" field in a passwd entry contained escaped characters, in
4919 particular, if it contained a \" sequence, Exim got it wrong when building
4920 a From: or a Sender: header from that name. A second bug also caused
4921 incorrect handling when an unquoted " was present following a character
4922 that needed quoting.
4924 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
4925 was not being matched caselessly.
4927 24. Arranged for all hyphens in the exim.8 source to be escaped with
4930 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
4931 properly. Recipient callout cache records were still being keyed to include
4932 the sender, even when use_sender was set false. This led to far more
4933 callouts that were necessary. The sender is no longer included in the key
4934 when use_sender is false.
4936 26. Added "control = submission" modifier to ACLs.
4938 27. Added the ${base62d: operator to decode base 62 numbers.
4940 28. dnsdb lookups can now access SRV records.
4942 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
4943 the configuration file.
4945 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
4946 (-v) mode. This makes the output for a verbose queue run more intelligible.
4948 31. Added a use_postmaster feature to recipient callouts.
4950 32. Added the $body_zerocount variable, containing the number of binary zero
4951 bytes in the message body.
4953 33. The time of last modification of the "new" subdirectory is now used as the
4954 "mailbox time last read" when there is a quota error for a maildir
4957 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
4959 35. Added +ignore_unknown as a special item in host lists.
4961 36. Code for decoding IPv6 addresses in host lists is now included, even if
4962 IPv6 support is not being compiled. This fixes a bug in which an IPv6
4963 address was recognized as an IP address, but was then not correctly decoded
4964 into binary, causing unexpected and incorrect effects when compared with
4971 1. Very minor rewording of debugging text in manualroute to say "list of
4972 hosts" instead of "hostlist".
4974 2. If verify=header_syntax was set, and a header line with an unqualified
4975 address (no domain) and a large number of spaces between the end of the
4976 name and the colon was received, the reception process suffered a buffer
4977 overflow, and (when I tested it) crashed. This was caused by some obsolete
4978 code that should have been removed. The fix is to remove it!
4980 3. When running in the test harness, delay a bit after writing a bounce
4981 message to get a bit more predictability in the log output.
4983 4. Added a call to search_tidyup() just before forking a reception process. In
4984 theory, someone could use a lookup in the expansion of smtp_accept_max_
4985 per_host which, without the tidyup, could leave open a database connection.
4987 5. Added the variables $recipient_data and $sender_data which get set from a
4988 lookup success in an ACL "recipients" or "senders" condition, or a router
4989 "senders" option, similar to $domain_data and $local_part_data.
4991 6. Moved the writing of debug_print from before to after the "senders" test
4994 7. Change 4.31/66 (moving the time when the Received: is generated) caused
4995 problems for message scanning, either using a data ACL, or using
4996 local_scan() because the Received: header was not generated till after they
4997 were called (in order to set the time as the time of reception completion).
4998 I have revised the way this works. The header is now generated after the
4999 body is received, but before the ACL or local_scan() are called. After they
5000 are run, the timestamp in the header is updated.
5006 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
5007 before starting a queue runner without re-exec. This happened only when
5008 deliver_drop_privilege was set or when the Exim user was set to root. The
5009 effect of the bug was that timeouts during subsequent deliveries caused
5010 crashes instead of being properly handled. The handler is now left at its
5011 default (and expected) setting.
5013 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
5014 message, again when deliver_drop_privilege is set or Exim is run as root.
5015 The bug described in (1) was not present in this case, but the tidying up
5016 of the other signals was missing. I have made the two cases consistent.
5018 3. The ignore_target_hosts setting on a manualroute router was being ignored
5019 for hosts that were looked up using the /MX notation.
5021 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
5024 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
5025 operated on the sender address. After changing the $sender_address to <>
5026 for the sender address verify, Exim was re-instated it as the original
5027 (before rewriting) address, but remembering that it had rewritten it, so it
5028 wasn't rewriting it again. This bug also had the effect of breaking the
5029 sender address verification caching when the sender address was rewritten.
5031 6. The ignore_target_hosts option was being ignored by the ipliteral router.
5032 This has been changed so that if the ip literal address matches
5033 ignore_target_hosts, the router declines.
5035 7. Added expansion conditions match_domain, match_address, and match_local_
5036 part (NOT match_host).
5038 8. The placeholder for the Received: header didn't have a length field set.
5040 9. Added code to Exim itself and to exim_lock to test for a specific race
5041 condition that could lead to file corruption when using MBX delivery. The
5042 issue is with the lockfile that is created in /tmp. If this file is removed
5043 after a process has opened it but before that process has acquired a lock,
5044 there is the potential for a second process to recreate the file and also
5045 acquire a lock. This could lead to two Exim processes writing to the file
5046 at the same time. The added code performs the same test as UW imapd; it
5047 checks after acquiring the lock that its file descriptor still refers to
5048 the same named file.
5050 10. The buffer for building added header lines was of fixed size, 8192 bytes.
5051 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
5054 11. Added the smtp_active_hostname option. If used, this will typically be made
5055 to depend on the incoming interface address. Because $interface_address is
5056 not set up until the daemon has forked a reception process, error responses
5057 that can happen earlier (such as "too many connections") no longer contain
5060 12. If an expansion in a condition on a "warn" statement fails because a lookup
5061 defers, the "warn" statement is abandoned, and the next ACL statement is
5062 processed. Previously this caused the whole ACL to be aborted.
5064 13. Added the iplsearch lookup type.
5066 14. Added ident_timeout as a log selector.
5068 15. Added tls_certificate_verified as a log selector.
5070 16. Added a global option tls_require_ciphers (compare the smtp transport
5071 option of the same name). This controls incoming TLS connections.
5073 17. I finally figured out how to make tls_require_ciphers do a similar thing
5074 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
5075 before starting the TLS session.
5077 18. Tabs are now shown as \t in -bP output.
5079 19. If the log selector return_path_on_delivery was set, Exim crashed when
5080 bouncing a message because it had too many Received: header lines.
5082 20. If two routers both had headers_remove settings, and the first one included
5083 a superfluous trailing colon, the final name in the first list and the
5084 first name in the second list were incorrectly joined into one item (with a
5085 colon in the middle).
5091 1. Added -C and -D options to the exinext utility, mainly to make it easier
5092 to include in the automated testing, but these could be helpful when
5093 multiple configurations are in use.
5095 2. The exinext utility was not formatting the output nicely when there was
5096 an alternate port involved in the retry record key, nor when there was a
5097 message id as well (for retries that were specific to a specific message
5098 and a specific host). It was also confused by IPv6 addresses, because of
5099 the additional colons they contain. I have fixed the IPv4 problem, and
5100 patched it up to do a reasonable job for IPv6.
5102 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
5103 delivery, the log line now contains "pipelined" if PIPELINING was used.
5105 4. An SMTP transport process used to panic and die if the bind() call to set
5106 an explicit outgoing interface failed. This has been changed; it is now
5107 treated in the same way as a connect() failure.
5109 5. A reference to $sender_host_name in the part of a conditional expansion
5110 that was being skipped was still causing a DNS lookup. This no longer
5113 6. The def: expansion condition was not recognizing references to header lines
5114 that used bh_ and bheader_.
5116 7. Added the _cache feature to named lists.
5118 8. The code for checking quota_filecount in the appendfile transport was
5119 allowing one more file than it should have been.
5121 9. For compatibility with Sendmail, the command line option
5129 and sets the incoming protocol and host name (for trusted callers). The
5130 host name and its colon can be omitted when only the protocol is to be set.
5131 Note the Exim already has two private options, -pd and -ps, that refer to
5132 embedded Perl. It is therefore impossible to set a protocol value of "d" or
5133 "s", but I don't think that's a major issue.
5135 10. A number of refactoring changes to the code, none of which should affect
5138 (a) The number of logging options was getting close to filling up the
5139 32-bit word that was used as a bit map. I have split them into two classes:
5140 those that are passed in the argument to log_write(), and those that are
5141 only ever tested independently outside of that function. These are now in
5142 separate 32-bit words, so there is plenty of room for expansion again.
5143 There is no change in the user interface or the logging behaviour.
5145 (b) When building, for example, log lines, the code previously used a
5146 macro that called string_cat() twice, in order to add two strings. This is
5147 not really sufficiently general. Furthermore, there was one instance where
5148 it was actually wrong because one of the argument was used twice, and in
5149 one call a function was used. (As it happened, calling the function twice
5150 did not affect the overall behaviour.) The macro has been replaced by a
5151 function that can join an arbitrary number of extra strings onto a growing
5154 (c) The code for expansion conditions now uses a table and a binary chop
5155 instead of a serial search (which was left over from when there were very
5156 few conditions). Also, it now recognizes conditions like "pam" even when
5157 the relevant support is not compiled in: a suitably worded error message is
5158 given if an attempt is made to use such a condition.
5160 11. Added ${time_interval:xxxxx}.
5162 12. A bug was causing one of the ddress fields not to be passed back correctly
5163 from remote delivery subprocesses. The field in question was not being
5164 subsequently used, so this caused to problems in practice.
5166 13. Added new log selectors queue_time and deliver_time.
5168 14. Might have fixed a bug in maildirsizefile handling that threw up
5169 "unexpected character" debug warnings, and recalculated the data
5170 unnecessarily. In any case, I expanded the warning message to give more
5173 15. Added the message "Restricted characters in address" to the statements in
5174 the default ACL that block characters like @ and % in local parts.
5176 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
5177 Three changes have been made:
5179 (a) There was a serious bug; a negative response to MAIL caused the whole
5180 recipient domain to be cached as invalid, thereby blocking all messages
5181 to all local parts at the same domain, from all senders. This bug has
5182 been fixed. The domain is no longer cached after a negative response to
5183 MAIL if the sender used is not empty.
5185 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
5188 (c) A new callout option, "use_sender" has been added for people who want
5189 the modified behaviour.
5195 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
5198 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
5199 indeed breaks things for older releases.
5201 3. Added additional logging to the case where there is a problem reading data
5202 from a filter that is running in a subprocess using a pipe, in order to
5203 try to track down a specific problem.
5205 4. Testing facility fudge: when running in the test harness and attempting
5206 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
5207 getting "No route to host". Convert this to a timeout.
5209 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
5212 6. Some OS don't have socklen_t but use size_t instead. This affects the
5213 fifth argument of getsockopt() amongst other things. This is now
5214 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
5215 can be set for individual OS. I have set it for SunOS5, OSF1, and
5216 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
5217 some earlier ones do not.
5219 7. Change 4.30/15 was not doing the test caselessly.
5221 8. The standard form for an IPv6 address literal was being rejected by address
5222 parsing in, for example, MAIL and RCPT commands. An example of this kind of
5223 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5224 this, as well as the form without the "IPv6" on the front (but only when
5225 address literals are enabled, of course).
5227 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
5229 10. Exim crashed if a message with an empty sender address specified by -f
5230 encountered a router with an errors_to setting. This could be provoked only
5231 by a command such as
5235 where an empty string was supplied; "<>" did not hit this bug.
5237 11. Installed PCRE release 4.5.
5239 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
5240 remained set. It is now erased.
5242 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
5243 times from message ids (which are base 36 rather than the normal 62).
5245 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
5246 were being counted as actual protocol errors, and logged if the log
5247 selector +smtp_protocol_error was set. One cannot be perfect in this test,
5248 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
5249 and DATA following a set of rejected RCPTs do not count as protocol errors.
5250 In other words, Exim assumes they were pipelined, though this may not
5251 actually be the case. Of course, in all cases the client gets an
5252 appropriate error code.
5254 15. If a lookup fails in an ACL condition, a message about the failure may
5255 be available; it is used if testing the ACL cannot continue, because most
5256 such messages specify what the cause of the deferral is. However, some
5257 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
5258 that caused an old message to be retained and used if a later statement
5259 caused a defer, replacing the real cause of the deferral.
5261 16. If an IP address had so many PTR records that the DNS lookup buffer
5262 was not large enough to hold them, Exim could crash while trying to process
5263 the truncated data. It now detects and logs this case.
5265 17. Further to 4.21/58, another change has been made: if (and only if) the
5266 first line of a message (the first header line) ends with CRLF, a bare LF
5267 in a subsequent header line has a space inserted after it, so as not to
5268 terminate the header.
5270 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
5271 unnecessarily, used atoi() instead of strtol(), and didn't check the
5272 termination when getting file sizes from file names by regex.
5274 19. Completely re-implemented the support for maildirsize files, in the light
5275 of a number of problems with the previous contributed implementation
5276 (4.30/29). In particular:
5278 . If the quota is zero, the maildirsize file is maintained, but no quota is
5281 . If the maildir directory does not exist, it is created before any attempt
5282 to write a maildirsize file.
5284 . The quota value in the file is just a cache; if the quota is changed in
5285 the transport, the new value overrides.
5287 . A regular expression is available for excluding directories from the
5290 20. The autoreply transport checks the characters in options that define the
5291 message's headers; it allows continued headers, but it was checking with
5292 isspace() after an embedded newline instead of explicitly looking for a
5295 21. If all the "regular" hosts to which an address was routed had passed their
5296 expiry times, and had not reached their retry times, the address was
5297 bounced, even if fallback hosts were defined. Now Exim should go on to try
5300 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
5301 equivalent code in the SMTP transport. Some hosts send humungous responses
5302 to HELO/EHLO, more than 1024 it seems.
5304 23. Refactoring: code in filter.c used (void *) for "any old type" but this
5305 gives compiler warnings in some environments. I've now done it "properly",
5308 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
5309 (because of problems with the built-in one) was declared to return uschar *
5310 instead of char *, causing compiler failure.
5312 25. Fixed a file descriptor leak when processing alias/forward files.
5314 26. Fixed a minor format string issue in dbfn.c.
5316 27. Typo in exim.c: ("dmbnz" for "dbmnz").
5318 28. If a filter file refered to $h_xxx or $message_headers, and the headers
5319 contained RFC 2047 "words", Exim's memory could, under certain conditions,
5322 29. When a sender address is verified, it is cached, to save repeating the test
5323 when there is more than one recipient in a message. However, when the
5324 verification involves a callout, it is possible for different callout
5325 options to be set for different recipients. It is too complicated to keep
5326 track of this in the cache, so now Exim always runs a verification when a
5327 callout is required, relying on the callout cache for the optimization.
5328 The overhead is duplication of the address routing, but this should not be
5331 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
5332 to be verified with callout=postmaster, and the main callout worked but the
5333 postmaster check failed, the verification correctly failed. However, if a
5334 subsequent RCPT command asked for sender verification *without* the
5335 postmaster check, incorrect caching caused this verification also to fail,
5338 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
5339 it was not caching the DNS options (qualify_single, search_parents) that
5340 were used when the lookup failed. A subsequent lookup with different
5341 options therefore always gave the same answer, though there were cases
5342 where it should not have. (Example: a "domains = !$mx_any" option on a
5343 dnslookup router: the "domains" option is always processed without any
5344 widening, but the router might have qualify_single set.) Now Exim uses the
5345 cached value only when the same options are set.
5347 32. Added John Jetmore's "exipick" utility to the distribution.
5349 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
5350 than a timeout (e.g. a certificate is required, and is not provided), an
5351 Exim server now closes the connection immediately. Previously it waited for
5352 the client to close - but if the client is SSL, it seems that they each
5353 wait for each other, leading to a delay before one of them times out.
5355 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
5356 maintained 0.8.x compatibility because I don't think many are using it, and
5357 it is clearly obsolete.
5359 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
5362 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
5363 client certificate was expired. A simple patch fixes this, though I don't
5364 understand the full logic of why the verify callback is called multiple
5367 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
5368 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
5369 which causes problems with some clients (such as the Certicom SSL Plus
5370 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
5371 disables the coutermeasure allowing Eudora to connect."
5373 38. Exim was not checking that a write() to a log file succeeded. This could
5374 lead to Bad Things if a log got too big, in particular if it hit a file
5375 size limit. Exim now panics and dies if it cannot write to a log file, just
5376 as it does if it cannot open a log file.
5378 39. Modified OS/Makefile-Linux so that it now contains
5380 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
5382 The two -D definitions ensure that Exim is compiled with large file
5383 support, which makes it possible to handle log files that are bigger than
5386 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
5387 instance) a domain was checked against a named list that involved a lookup,
5388 causing $domain_data to be set, then another domain was checked against the
5389 same list, then the first domain was re-checked, the value of $domain_data
5390 after the final check could be wrong. In particular, if the second check
5391 failed, it could be set empty. This bug probably also applied to
5394 41. The strip_trailing_dot option was not being applied to the address given
5395 with the -f command-line option.
5397 42. The code for reading a message's header from the spool was incrementing
5398 $received_count, but never initializing it. This meant that the value was
5399 incorrect (doubled) while delivering a message in the same process in which
5400 it was received. In the most common configuration of Exim, this never
5401 happens - a fresh exec is done - but it can happen when
5402 deliver_drop_privilege is set.
5404 43. When Exim logs an SMTP synchronization error - client data sent too soon -
5405 it now includes up to 150 characters of the unexpected data in the log
5408 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
5409 and building data strings. The size of both of these buffers was 10 000
5410 bytes - far larger than anybody would *ever* want, thought I. Needless to
5411 say, somebody hit the limit. I have increased the maximum line length to
5412 20 000 and the maximum data length of concatenated lines to 100 000. I have
5413 also fixed two bugs, because there was no checking on these buffers. Tsk,
5414 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
5415 buffer is too small.
5417 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
5418 lsearch lookups. Now it does.
5420 46. When parsing a route_list item in a manualroute router, a fixed-length
5421 buffer was used for the list of hosts. I made this 1024 bytes long,
5422 thinking that nobody would ever have a list of hosts that long. Wrong.
5423 Somebody had a whole pile of complicated expansion conditions, and the
5424 string was silently truncated, leading to an expansion error. It turns out
5425 that it is easier to change to an unlimited length (owing to other changes
5426 that have happened since this code was originally written) than to build
5427 structure for giving a limitation error. The length of the item that
5428 expands into the list of hosts is now unlimited.
5430 47. The lsearch lookup could not handle data where the length of text line was
5431 more than 4095 characters. Such lines were truncated, leading to shortened
5432 data being returned. It should now handle lines of any length.
5434 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
5435 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
5438 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
5439 replacing strings such as BIN_DIRECTORY in a source file: the replacement
5440 no longer happens in comment lines. A list of replacements is now placed
5441 at the head of all of the source files, except those whose only change is
5442 to replace PERL_COMMAND in the very first #! line.
5444 50. Replaced the slow insertion sort in queue.c, for sorting the list of
5445 messages on the queue, with a bottom-up merge sort, using code contributed
5446 by Michael Haardt. This should make operations like -bp somewhat faster on
5447 large queues. It won't affect queue runners, except when queue_run_in_order
5450 51. Installed eximstats 1.31 in the distribution.
5452 52. Added support for SRV lookups to the dnslookup router.
5454 53. If an ACL referred to $message_body or $message_body_end, the value was not
5455 reset for any messages that followed in the same SMTP session.
5457 54. The store-handling optimization for building very long strings was not
5458 differentiating between the different store pools. I don't think this
5459 actually made any difference in practice, but I've tidied it.
5461 55. While running the routers to verify a sender address, $sender_address
5462 was still set to the sender address. This is wrong, because when routing to
5463 send a bounce to the sender, it would be empty. Therefore, I have changed
5464 it so that, while verifying a sender address, $sender_address is set to <>.
5465 (There is no change to what happens when verifying a recipient address.)
5467 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
5468 target A or AAAA records (if not already returned) without resetting the
5469 qualify_single or search_parents options of the DNS resolver. These are
5470 inappropriate in this case because the targets of MX and SRV records must
5471 be FQDNs. A broken DNS record could cause trouble if it happened to have a
5472 target that, when qualified, matched something in the local domain. These
5473 two options are now turned off when doing these lookups.
5475 57. It seems that at least some releases of Reiserfs (which does not have the
5476 concept of a fixed number of inodes) returns zero and not -1 for the
5477 number of available inodes. This interacted badly with check_spool_inodes,
5478 which assumed that -1 was the "no such thing" setting. What I have done is
5479 to check that the total number of inodes is greater than zero before doing
5480 the test of how many are available.
5482 58. When a "warn" ACL statement has a log_message modifier, the message is
5483 remembered, and not repeated. This is to avoid a lot of repetition when a
5484 message has many recipients that cause the same warning to be written.
5485 Howewer, Exim was preserving the list of already written lines for an
5486 entire SMTP session, which doesn't seem right. The memory is now reset if a
5487 new message is started.
5489 59. The "rewrite" debugging flag was not showing the result of rewriting in the
5490 debugging output unless log_rewrite was also set.
5492 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
5493 of (int)(handle) when we know that handle contains (void *)(-1).
5495 61. The Exim daemon panic-logs an error return when it closes the incoming
5496 connection. However "connection reset by peer" seems to be common, and
5497 isn't really an error worthy of noting specially, so that particular error
5500 62. When Exim is trying to find all the local interfaces, it used to panic and
5501 die if the ioctl to get the interface flags failed. However, it seems that
5502 on at least one OS (Solaris 9) it is possible to have an interface that is
5503 included in the list of interfaces, but for which you get a failure error
5504 for this call. This happens when the interface is not "plumbed" into a
5505 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
5506 failure of the "get flags" call assumes that the interface is down.
5508 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
5509 makes life easier for people who are doing arithmetic on fields extracted
5510 from dates, where you often get leading zeros that should not be
5511 interpreted as octal.
5513 64. Added qualify_domain to the redirect router, to override the global
5516 65. If a pathologically long header line contained very many addresses (the
5517 report of this problem mentioned 10 000) and each of them was rewritten,
5518 Exim could use up a very large amount of memory. (It kept on making new
5519 copies of the header line as it rewrote, and never released the old ones.)
5520 At the expense of a bit more processing, the header rewriting function has
5521 been changed so that it no longer eats memory in this way.
5523 66. The generation of the Received: header has been moved from the time that a
5524 message starts to be received, to the time that it finishes. The timestamp
5525 in the Received: header should now be very close to that of the <= log
5526 line. There are two side-effects of this change:
5528 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
5529 logged header lines no longer include the local Received: line, because
5530 it has not yet been created. The same applies to a copy of the message
5531 that is returned to a non-SMTP sender when a message is rejected.
5533 (b) When a filter file is tested using -bf, no additional Received: header
5534 is added to the test message. After some thought, I decided that this
5537 This change does not affect the value of $received_for. It is still set
5538 after address rewriting, but before local_scan() is called.
5540 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
5542 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
5543 gave an unhelpful panic error message, and a defer error. I have managed to
5544 change this behaviour so that it now rejects any supplied certificate,
5545 which seems right, as the list of acceptable certificates is empty.
5547 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
5548 gave an unhelpful defer error. I have not managed to make this reject any
5549 supplied certificates, but the error message it gives is "no certificate
5550 supplied", which is not helpful.
5552 70. exigrep's output now also includes lines that are not associated with any
5553 message, but which match the given pattern. Implemented by a patch from
5554 Martin Sluka, which also tidied up the Perl a bit.
5556 71. Recipient callout verification, like sender verification, was using <> in
5557 the MAIL FROM command. This isn't really the right thing, since the actual
5558 sender may affect whether the remote host accepts the recipient or not. I
5559 have changed it to use the actual sender in the callout; this means that
5560 the cache record is now keyed on a recipient/sender pair, not just the
5561 recipient address. There doesn't seem to be a real danger of callout loops,
5562 since a callout by the remote host to check the sender would use <>.
5563 [SEE ABOVE: changed after hitting problems.]
5565 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
5566 temporary errors. However, in the case of such a code being given after
5567 the end of a data transmission (i.e. after ".") Exim was failing to write
5568 a retry record for the message. (Yes, there was some broken host that was
5569 actually sending 8xx at this point.)
5571 73. An unknown lookup type in a host list could cause Exim to panic-die when
5572 the list was checked. (An example that provoked this was putting <; in the
5573 middle of a list instead of at the start.) If this happened during a DATA
5574 ACL check, a -D file could be left lying around. This kind of configuration
5575 error no longer causes Exim to die; instead it causes a defer errror. The
5576 incident is still logged to the main and panic logs.
5578 74. Buglet left over from Exim 3 conversion. The message "too many messages
5579 in one connection" was written to the rejectlog but not the mainlog, except
5580 when address rewriting (yes!) was being logged.
5582 75. Added write_rejectlog option.
5584 76. When a system filter was run not as root (that is, when system_filter_user
5585 was set), the values of the $n variables were not being returned to the
5586 main process; thus, they were not subsequently available in the $sn
5589 77. Added +return_path_on_delivery log selector.
5591 78. A connection timeout was being treated differently from recipients deferred
5592 when testing hosts_max_try with a message that was older than the host's
5593 retry timeout. (The host should not be counted, thus allowing all hosts to
5594 be tried at least once before bouncing.) This may have been the cause of an
5595 occasionally reported bug whereby a message would remain on the queue
5596 longer than the retry timeout, but would be bounced if a delivery was
5597 forced. I say "may" because I never totally pinned down the problem;
5598 setting up timeout/retry tests is difficult. See also the next item.
5600 79. The ultimate address timeout was not being applied to errors that involved
5601 a combination of host plus message (for example, a timeout on a MAIL
5602 command). When an address resolved to a number of possible hosts, and they
5603 were not all tried for each delivery (e.g. because of hosts_max_try), a
5604 message could remain on the queue longer than the retry timeout.
5606 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
5609 81. Fixed an obscure SMTP outgoing bug which required at least the following
5610 conditions: (a) there was another message waiting for the same server;
5611 (b) the server returned 5xx to all RCPT commands in the first message so
5612 that the message was not completed; (c) the server dropped the connection
5613 or gave a negative response to the RSET that Exim sends to abort the
5614 transaction. The observed case was a dropped connection after DATA that had
5615 been sent in pipelining mode. That is, the server had advertised PIPELINING
5616 but was not implementing it correctly. The effect of the bug was incorrect
5617 behaviour, such as trying another host, and this could lead to a crash.
5623 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
5624 and daemon.c were passed as pointers to ints; they should have been
5625 pointers to socklen_t variables (which are typically unsigned ints).
5627 2. Some signed/unsigned type warnings in the os.c file for Linux have been
5630 3. Fixed a really odd bug that affected only the testing scheme; patching a
5631 certain fixed string in the binary changed the value of another string that
5632 happened to be identical to the end of the original first string.
5634 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
5635 name", it returns that address as the IP address. On some operating
5636 systems (e.g. Solaris), it also passes back the IP address string as the
5637 "host name". However, on others (e.g. Linux), it passes back an empty
5638 string. Exim wasn't checking for this, and was changing the host name to an
5639 empty string, assuming it had been canonicized.
5641 5. Although rare, it is permitted to have more than one PTR record for a given
5642 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
5643 all the names associated with an address, because they do in Solaris.
5644 However, it seems that they do not in Linux for data that comes from the
5645 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
5646 I found this out when I moved to a new Linux workstation and tried to run
5647 the Exim test suite.
5649 To get round this problem I have changed the code so that it now does its
5650 own call to the DNS to look up PTR records when searching for a host name.
5651 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
5652 addresses that are only in /etc/hosts are still found.
5654 This behaviour is, however, controlled by an option called host_lookup_
5655 order, which defaults to "bydns:byaddr". If people want to use the other
5656 order, or indeed, just use one or the other means of lookup, they can
5657 specify it in this variable.
5659 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
5660 some operating systems, this comes back from gethostbyaddr() as an empty
5661 string, and this is what Exim used to test for. However, it seems that in
5662 other systems, "." is yielded. Exim now tests for this case too.
5664 7. The values of check_spool_space and check_log_space are now held internally
5665 as a number of kilobytes instead of an absolute number of bytes. If a
5666 numbers is specified without 'K' or 'M', it is rounded up to the nearest
5667 kilobyte. This means that much larger values can be stored.
5669 8. Exim monitor: an attempt to get the action menu when not actually pointing
5670 at a message produces an empty menu entitled "No message selected". This
5671 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
5672 no entries in it ("Shell widget menu has zero width and/or height"). So I
5673 have added a single, blank menu entry in this case.
5675 9. Added ${quote_local_part.
5677 10. MIME decoding is now applied to the contents of Subject: header lines when
5680 11. Now that a reference to $sender_host_address automatically causes a reverse
5681 lookup to occur if necessary (4.13/18), there is no need to arrange for a
5682 host lookup before query-style lookups in lists that might use this
5683 variable. This has therefore been abolished, and the "net-" prefix is no
5684 longer necessary for query-style lookups.
5686 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
5687 have been a typo for LFLAGS, so it has been changed.
5689 13. The install script calls Exim with "-C /dev/null" in order to find the
5690 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
5691 to be output. Howeve, since Exim outputs its version number before the
5692 error, it didn't break the script. It just looked ugly. I fixed this by
5693 always allowing "-C /dev/null" if the caller is root.
5695 14. Ignore overlarge ACL variable number when reading spool file - insurance
5696 against a later release with more variables having written the file.
5698 15. The standard form for an IPv6 address literal was being rejected by EHLO.
5699 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5700 this, as well as the form without the "IPv6" on the front.
5702 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
5703 OS/Makefile-Darwin file.
5705 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
5706 only to LDAP libraries that do not have LDAP_OPT_DEREF.
5708 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
5709 variable. However, some OS use ints for inodes. I've added cast to long int
5710 to get rid of the compiler warning.
5712 19. I had forgotten to lock out "/../" in configuration file names when
5713 ALT_CONFIG_PREFIX was set.
5715 20. Routers used for verification do not need to specify transports. However,
5716 if such a router generated a host list, and callout was configured, Exim
5717 crashed, because it could not find a port number from the (non-existent)
5718 transport. It now assumes port 25 in this circumstance.
5720 21. Added the -t option to exigrep.
5722 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
5723 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
5724 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
5725 them is set, LOOKUP_LSEARCH is forced.
5727 23. "exim -bV" now outputs a list of lookups that are included in the binary.
5729 24. Added sender and host information to the "rejected by local_scan()" log
5730 line; previously there was no indication of these.
5732 25. Added .include_if_exists.
5734 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
5735 Linux. It turns out that not all file systems support this. Apparently some
5736 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
5737 do it.) To cope with this, the error EINVAL, which means that sync-ing is
5738 not supported on the file descriptor, is now ignored when Exim is trying to
5739 sync a directory. This applies only to Linux.
5741 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
5743 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
5744 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
5747 29. Added support for maildirsize files from supplied patch (modified a bit).
5749 30. The use of :fail: followed by an empty string could lead Exim to respond to
5750 sender verification failures with (e.g.):
5752 550 Verification failed for <xxx>
5753 550 Sender verify failed
5755 where the first response line was missing the '-' that indicates it is not
5756 the final line of the response.
5758 31. The loop for finding the name of the user that called Exim had a hardwired
5759 limit of 10; it now uses the value of finduser_retries, which is used for
5760 all other user lookups.
5762 32. Added $received_count variable, available in data and not_smtp ACLs, and at
5765 33. Exim was neglecting to zero errno before one call of strtol() when
5766 expanding a string and expecting an integer value. On some systems this
5767 resulted in spurious "integer overflow" errors. Also, it was casting the
5768 result into an int without checking.
5770 34. Testing for a connection timeout using "timeout_connect" in the retry rules
5771 did not work. The code looks as if it has *never* worked, though it appears
5772 to have been documented since at least releast 1.62. I have made it work.
5774 35. The "timeout_DNS" error in retry rules, also documented since at least
5775 1.62, also never worked. As it isn't clear exactly what this means, and
5776 clearly it isn't a major issue, I have abolished the feature by treating it
5777 as "timeout", and writing a warning to the main and panic logs.
5779 36. The display of retry rules for -brt wasn't always showing the error code
5782 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
5783 timeout_connect_A, timeout_connect_MX.
5785 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
5786 to the empty sender.
5788 39. The daemon was not analysing the content of -oX till after it had closed
5789 stderr and disconnected from the controlling terminal. This meant that any
5790 syntax errors were only noted on the panic log, and the return code from
5791 the command was 0. By re-arranging the code a little, I've made the
5792 decoding happen first, so such errors now appear on stderr, and the return
5793 code is 1. However, the actual setting up of the sockets still happens in
5794 the disconnected process, so errors there are still only recorded on the
5797 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
5798 connections (as happens on some IP stacks) was logged at start up time as
5799 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
5800 it from "IPv6 and IPv4", which means that two separate sockets are being
5803 41. The debug output for gethostbyname2() or getipnodebyname() failures now
5804 says whether AF_INET or AF_INET6 was passed as an argument.
5806 42. Exiwhat output was messed up when time zones were included in log
5809 43. Exiwhat now gives more information about the daemon's listening ports,
5810 and whether -tls-on-connect was used.
5812 44. The "port" option of the smtp transport is now expanded.
5814 45. A "message" modifier in a "warn" statement in a non-message ACL was being
5815 silently ignored. Now an error message is written to the main and panic
5818 46. There's a new ACL modifier called "logwrite" which writes to a log file
5819 as soon as it is encountered.
5821 47. Added $local_user_uid and $local_user_gid at routing time.
5823 48. Exim crashed when trying to verify a sender address that was being
5826 49. Exim was recognizing only a space character after ".include". It now also
5827 recognizes a tab character.
5829 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
5830 extracting the relevant information from the specification. The man page no
5831 longer contains scrambled data for the -d option, and I've added a section
5832 at the front about calling Exim under different names.
5834 51. Added "extra_headers" argument to the "mail" command in filter files.
5836 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
5839 53. Installed eximstats 1.29.
5841 54. Added transport_filter_timeout as a generic transport option.
5843 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
5844 Cc: header lines. This was required by RFC 822, but it not required by RFC
5847 56. Exim used to add From:, Date:, and Message-Id: header lines to any
5848 incoming messages that did not have them. Now it does so only if the
5849 message originates locally, that is, if there is no associated remote host
5850 address. When Resent- header lines are present, this applies to the Resent-
5851 lines rather than the non-Resent- lines.
5853 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
5854 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
5856 58. Messages for configuration errors now include the name of the main
5857 configuration file - useful now that there may be more than one file in a
5858 list (.included file names were always shown).
5860 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
5861 for those rare installations that do not start the daemon as root or run it
5862 setuid root. I've cut out the call to initgroups() if the daemon is not
5865 60. The Exim user and group can now be bound into the binary as text strings
5866 that are looked up at the start of Exim's processing.
5868 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
5870 62. Added $mailstore_basename variable.
5872 63. Installed patch to sieve.c from Michael Haardt.
5874 64. When Exim failed to open the panic log after failing to open the main log,
5875 the original message it was trying to log was written to stderr and debug
5876 output, but if they were not available (the usual case in production), it
5877 was lost. Now it is written to syslog before the two lines that record the
5878 failures to open the logs.
5880 65. Users' Exim filters run in subprocesses under the user's uid. It is
5881 possible for a "deliver" command or an alias in a "personal" command to
5882 provoke an address rewrite. If logging of address rewriting is configured,
5883 this fails because the process is not running as root or exim. There may be
5884 a better way of dealing with this, but for the moment (because 4.30 needs
5885 to be released), I have disabled address rewrite logging when running a
5886 filter in a non-root, non-exim process.
5892 1. The buildconfig auxiliary program wasn't quoting the value set for
5893 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
5894 not defined. This bug was masked in 4.22 by the effect that was fixed in
5897 2. Some messages that were rejected after a message id was allocated were
5898 shown as "incomplete" by exigrep. It no longer does this for messages that
5899 are rejected by local_scan() or the DATA or non-SMTP ACLs.
5901 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
5902 have allow_domain_literals set, the ID did not get logged in the <= line.
5903 Domain literals are now always recognized in Message-ID: header lines.
5905 4. The first argument for a ${extract expansion item is the key name or field
5906 number. Leading and trailing spaces in this item were not being ignored,
5907 causing some misleading effects.
5909 5. When deliver_drop_privilege was set, single queue runner processes started
5910 manually (i.e. by the command "exim -q") or by the daemon (which uses the
5911 same command in the process it spins off) were not dropping privilege.
5913 6. When the daemon running as "exim" started a queue runner, it always
5914 re-executed Exim in the spun-off process. This is a waste of effort when
5915 deliver_drop_privilege is set. The new process now just calls the
5916 queue-runner function directly.
5922 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
5925 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
5926 ignored. Though the use of -oP was forcing the writing of a pid file, it
5927 was always written to the default place.
5929 3. If the message "no IP address found for host xxxx" is generated during
5930 incoming verification, it is now followed by identification of the incoming
5931 connection (so you can more easily find what provoked it).
5933 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
5935 5. Added some features to "harden" Exim a bit more against certain attacks:
5937 (a) There is now a build-time option called FIXED_NEVER_USERS that can
5938 be put in Local/Makefile. This is like the never_users runtime option,
5939 but it cannot be overridden. The default setting is "root".
5941 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
5942 prefix string with which any file named in a -C command line option
5945 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
5946 is retained for -C and -D only if the caller of Exim is root. Without
5947 it, the exim user may also use -C and -D and retain privilege.
5949 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
5950 command line option is disabled.
5952 6. Macro names set by the -D option must start with an upper case letter, just
5953 like macro names defined in the configuration file.
5955 7. Added "dereference=" facility to LDAP.
5957 8. Two instances of the typo "uknown" in the source files are fixed.
5959 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
5960 the Configure-Makefile script screwed up while processing it.
5962 10. Incorporated PCRE 4.4.
5964 11. The SMTP synchronization check was not operating right at the start of an
5965 SMTP session. For example, it could not catch a HELO sent before the client
5966 waited for the greeting. There is now a check for outstanding input at the
5967 point when the greeting is written. Because of the duplex, asynchronous
5968 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
5969 way, but not yet received, when the check is performed.
5971 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
5972 on TCP/IP sockets, because this apparently causes some broken clients to
5975 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
5976 unchanged) from the Cygwin maintainer.
5978 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
5979 was supported instead of testing for mbx. Effectively a typo.
5981 15. The spa authenticator server code was not checking that the input it
5982 received was valid base64.
5984 16. The debug output line for the "set" modifier in ACLs was not showing the
5985 name of the variable that was being set.
5987 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
5989 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
5990 lookup on its own. Something else was needed to trigger the lookup. For
5991 example, a match in host_lookup or the need for a host name in a host list.
5992 Now, if $sender_host_name is referenced and the host name has not yet been
5993 looked up, a lookup is performed. If the lookup fails, the variable remains
5994 empty, and $host_lookup_failed is set to "1".
5996 19. Added "eqi" as a case-independent comparison operator.
5998 20. The saslauthd authentication condition could segfault if neither service
5999 nor realm was specified.
6001 21. If an overflowing value such as "2048M" was set for message_size_limit, the
6002 error message that was logged was misleading, and incoming SMTP
6003 connections were dropped. The message is now more accurate, and temporary
6004 errors are given to SMTP connections.
6006 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
6007 (except RSET) with a 421 error, until QUIT is received. However, it was
6008 failing to send a response to QUIT.
6010 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
6011 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
6012 ACL is now run after the helo_try_verify_hosts code.
6014 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
6015 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
6016 case-independent, so other case variants are also recognized). Apparently
6017 some systems use these upper case variants.
6019 25. If more than two messages were waiting for the same host, and a transport
6020 filter was specified for the transport, Exim sent two messages over the
6021 same TCP/IP connection, and then failed with "socket operation on non-
6022 socket" when it tried to send the third.
6024 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
6026 27. The extern definition of crypt16() in expand.c was not being excluded when
6027 the OS had its own crypt16() function.
6029 28. Added bounce_return_body as a new option, and bounce_return_size_limit
6030 as a preferred synonym for return_size_limit, both as an option and as an
6033 29. Added LIBS=-liconv to OS/Makefile-OSF1.
6035 30. Changed the default configuration ACL to relax the local part checking rule
6036 for addresses that are not in any local domains. For these addresses,
6037 slashes and pipe symbols are allowed within local parts, but the sequence
6038 /../ is explicitly forbidden.
6040 31. SPA server authentication was not clearing the challenge buffer before
6043 32. log_message in a "warn" ACL statement was writing to the reject log as
6044 well as to the main log, which contradicts the documentation and doesn't
6045 seem right (because no rejection is happening). So I have stopped it.
6047 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
6048 However, I am unable to do any testing of this.
6050 34. Fixed an infelicity in the appendfile transport. When checking directories
6051 for a mailbox, to see if any needed to be created, it was accidentally
6052 using path names with one or more superfluous leading slashes; tracing
6053 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
6055 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
6056 contents are added to the log line that is written for every discarded
6057 recipient. (Previously a log_message setting was ignored.)
6059 36. The ${quote: operator now quotes the string if it is empty.
6061 37. The install script runs exim in order to find its version number. If for
6062 some reason other than non-existence or emptiness, which it checks, it
6063 could not run './exim', it was installing it with an empty version number,
6064 i.e. as "exim-". This error state is now caught, and the installation is
6067 38. An argument was missing from the function that creates an error message
6068 when Exim fails to connect to the socket for saslauthd authentication.
6069 This could cause Exim to crash, or give a corrupted message.
6071 39. Added isip, isip4, and isip6 to ${if conditions.
6073 40. The ACL variables $acl_xx are now saved with the message, and can be
6074 accessed later in routers, transports, and filters.
6076 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
6077 strings in the file are not string-expanded.
6079 42. If a MAIL command specified a SIZE value that was too large to fit into an
6080 int variable, the check against message_size_limit failed. Such values are
6081 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
6082 day this will have to be increased, but I don't think I want to be around
6083 when emails are that large.
6090 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
6091 iconv() is not standard in FreeBSD.
6093 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
6094 IPv6 enabled. The observed symptom was a segmentation fault on return from
6095 the function os_common_find_running_interfaces() in src/os.c.
6097 3. In the check_special_case() function in daemon.c I had used "errno" as an
6098 argument name, which causes warnings on some systems. This was basically a
6099 typo, since it was named "eno" in the comments!
6101 4. The code that waits for the clock to tick (at a resolution of some fraction
6102 of a second) so as to ensure message-id uniqueness was always waiting for
6103 at least one whole tick, when it could have waited for less. [This is
6104 almost certainly not relevant at current processor speeds, where it is
6105 unlikely to ever wait at all. But we try to future-proof.]
6107 5. The function that sleeps for a time interval that includes fractions of a
6108 second contained a race. It did not block SIGALRM between setting the
6109 timer, and suspending (a couple of lines later). If the interval was short
6110 and the sigsuspend() was delayed until after it had expired, the suspension
6111 never ended. On busy systems this could lead to processes getting stuck for
6114 6. Some uncommon configurations may cause a lookup to happen in a queue runner
6115 process, before it forks any delivery processes. The open lookup caching
6116 mechanism meant that the open file or database connection was passed into
6117 the delivery process. The problem was that delivery processes always tidy
6118 up cached lookup data. This could cause a problem for the next delivery
6119 process started by the queue runner, because the external queue runner
6120 process does not know about the closure. So the next delivery process
6121 still has data in the lookup cache. In the case of a file lookup, there was
6122 no problem because closing a file descriptor in a subprocess doesn't affect
6123 the parent. However, if the lookup was caching a connection to a database,
6124 the connection was closed, and the second delivery process was likely to
6125 see errors such as "PGSQL: query failed: server closed the connection
6126 unexpectedly". The problem has been fixed by closing all cached lookups
6127 in a queue runner before running a delivery process.
6129 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
6130 seem to have the "const" qualifier which it has on other OS. I've
6133 8. Change 4.21/2 was too strict. It is only if there are two authenticators
6134 *of the same type* (client or server) with the same public name that an
6135 error should be diagnosed.
6137 9. When Exim looked up a host name for an IP address, but failed to find the
6138 original IP address when looking up the host name (a safety check), it
6139 output the message "<ip address> does not match any IP for NULL", which was
6140 confusing, to say the least. The bug was that the host name should have
6141 appeared instead of "NULL".
6143 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
6144 user that is built into the binary, and the -C or -D options is used, root
6145 privilege is dropped before the configuration file is read. In addition,
6146 logging is switched to stderr instead of the normal log files. If the
6147 configuration then re-defines the Exim user, the unprivileged environment
6148 is probably not what is expected, so Exim logs a panic warning message (but
6151 However, if deliver_drop_privilege is set, the unprivileged state may well
6152 be exactly what is intended, so the warning has been cut out in that case,
6153 and Exim is allowed to try to write to its normal log files.
6159 1. smtp_return_error_details was not giving details for temporary sender
6160 or receiver verification errors.
6162 2. Diagnose a configuration error if two authenticators have the same public
6165 3. Exim used not to create the message log file for a message until the first
6166 delivery attempt. This could be confusing when incoming messages were held
6167 for policy or load reasons. The message log file is now created at the time
6168 the message is received, and an initial "Received" line is written to it.
6170 4. The automatically generated man page for command line options had a minor
6171 bug that caused no ill effects; however, a more serious problem was that
6172 the procedure for building the man page automatically didn't always
6173 operate. Consequently, release 4.20 contains an out-of-date version. This
6174 shouldn't happen again.
6176 5. When building Exim with embedded Perl support, the script that builds the
6177 Makefile was calling 'perl' to find its compile-time parameters, ignoring
6178 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
6180 6. The freeze_tell option was not being used for messages that were frozen on
6181 arrival, either by an ACL or by local_scan().
6183 7. Added the smtp_incomplete_transaction log selector.
6185 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
6186 was accepting AUTH without a new EHLO.
6188 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
6189 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
6192 10. Logging of TCP/IP connections (when configured) now happens in the main
6193 daemon process instead of the child process, so that the TCP/IP connection
6194 count is more accurate (but it can never be perfect).
6196 11. The use of "drop" in a nested ACL was not being handled correctly in the
6197 outer ACL. Now, if condition failure induced by the nested "drop" causes
6198 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
6199 or "require"), the connection is dropped.
6201 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
6202 that yield "discard" can now be used with an "accept" or a "discard" verb,
6203 but an error is generated for any others (because I can't see a useful way
6204 to define what should happen).
6206 13. When an ACL is read dynamically from a file (or anywhere else), the lines
6207 are now processed in the same way as lines in the Exim configuration file.
6208 In particular, continuation lines are supported.
6210 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
6212 15. Added -ti meaning -t -i.
6214 16. Check for letters, digits, hyphens, and dots in the names of dnslist
6215 domains, and warn by logging if others are found.
6217 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
6218 returned from GIFCONF when Exim is trying to find the list of interfaces on
6219 a host. The code in os.c has been modified to copy each ifreq to an aligned
6220 structure in all cases.
6222 Also, in some cases, the returned ifreq's were being copied to a 'struct
6223 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
6224 means the last couple of bytes of an IPv6 address could be chopped if the
6225 ifreq contained only a normal sockaddr (14 bytes storage).
6227 18. Named domain lists were not supported in the hosts_treat_as_local option.
6228 An entry such as +xxxx was not recognized, and was treated as a literal
6231 19. Ensure that header lines added by a DATA ACL are included in the reject log
6232 if the ACL subsequently rejects the message.
6234 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
6235 MD5 (which is deprecated).
6237 21. When testing a filter file using -bf, Exim was writing a message when it
6238 took the sender from a "From " line in the message, but it was not doing so
6239 when it took $return_path from a Return-Path: header line. It now does.
6241 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
6242 with a valid header line field name (a series of printing characters
6243 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
6245 23. Changed "disc" in the source to "disk" to conform to the documentation and
6246 the book and for uniformity.
6248 24. Ignore Sendmail's -Ooption=value command line item.
6250 25. When execve() failed while trying to run a command in a pipe transport,
6251 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
6252 could be confused with a return value of 69 from the command itself. This
6253 has been changed to 127, the value the shell returns if it is asked to run
6254 a non-existent command. The wording for the related log line suggests a
6255 non-existent command as the problem.
6257 26. If received_header_text expands to an empty string, do not add a Received:
6258 header line to the message. (Well, it adds a token one on the spool, but
6259 marks it "old" so that it doesn't get used or transmitted.)
6261 27. Installed eximstats 1.28 (addition of -nt option).
6263 28. There was no check for failure on the call to getsockname() in the daemon
6264 code. This can fail if there is a shortage of resources on the system, with
6265 ENOMEM, for example. A temporary error is now given on failure.
6267 29. Contrary to the C standard, it seems that in some environments, the
6268 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
6269 program. Exim now does this explicitly; it affects the formatting of
6270 timestamps using strftime().
6272 30. If exiqsumm was given junk data, it threw up some uninitialized variable
6273 complaints. I've now initialized all the variables, to avoid this.
6275 32. Header lines added by a system filter were not being "seen" during
6276 transport-time rewrites.
6278 33. The info_callback() function passed to OpenSSL is set up with type void
6279 (*)(SSL *, int, int), as described somewhere. However, when calling the
6280 function (actually a macro) that sets it up, the type void(*)() is
6281 expected. I've put in a cast to prevent warnings from picky compilers.
6283 34. If a DNS black list lookup found a CNAME record, but there were no A
6284 records associated with the domain it pointed at, Exim crashed.
6286 35. If a DNS black list lookup returned more than one A record, Exim ignored
6287 all but the first. It now scans all returned addresses if a particular IP
6288 value is being sought. In this situation, the contents of the
6289 $dnslist_value variable are a list of all the addresses, separated by a
6292 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
6293 to accept a host name and all its aliases if the forward lookup for any of
6294 them yielded the IP address of the incoming connection. Now it accepts only
6295 those names whose forward lookup yields the correct IP address. Any other
6296 names are discarded. This closes a loophole whereby a rogue DNS
6297 administrator could create reverse DNS records to break through a
6298 wildcarded host restriction in an ACL.
6300 37. If a user filter or a system filter that ran in a subprocess used any of
6301 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
6302 the wrong values were passed to the pipe command ($thisaddress had the
6303 value of $0, $0 had the value of $1, etc). This bug was introduced by
6304 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
6306 38. Improved the line breaking for long SMTP error messages from ACLs.
6307 Previously, if there was no break point between 40 and 75 characters, Exim
6308 left the rest of the message alone. Two changes have been made: (a) I've
6309 reduced the minimum length to 35 characters; (b) if it can't find a break
6310 point between 35 and 75 characters, it looks ahead and uses the first one
6311 that it finds. This may give the occasional overlong line, but at least the
6312 remaining text gets split now.
6314 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
6315 file descriptors might be low, and that setting 1000 would always raise it.
6316 It turns out that in some environments, the limit is already over 1000 and
6317 that lowering it causes trouble. So now Exim takes care not to decrease it.
6319 40. When delivering a message, the value of $return_path is set to $sender_
6320 address at the start of routing (routers may change the value). By an
6321 oversight, this default was not being set up when an address was tested by
6322 -bt or -bv, which affected the outcome if any router or filter referred to
6325 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
6326 log only when "message" or "log_message" are set. However, if one of the
6327 conditions was an address verification, or a call to a nested ACL, the
6328 messages generated by the underlying test were being passed through. This
6329 no longer happens. The underlying message is available in $acl_verify_
6330 message for both "message" and "log_message" expansions, so it can be
6331 passed through if needed.
6333 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
6334 new expansion $bh_ to give the encoded byte string without charset
6335 translation. Translation happens only if iconv() is available; HAVE_ICONV
6336 indicates this at build time. HEADERS_CHARSET gives the charset to
6337 translate to; headers_charset can change it in the configuration, and
6338 "headers charset" can change it in an individual filter file.
6340 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
6341 that creates RFC 2047 encoded "words" labels them as that charset instead
6342 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
6343 expansion operator; (ii) when Exim creates a From: line for a local
6344 message; (iii) when a header line is rewritten to include a "phrase" part.
6346 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
6347 buggy, causing it to skip the first lines of messages whose message ID
6348 ended in 'D'. This would not have bitten before Exim release 4.14, because
6349 message IDs were unlikely to end in 'D' before then. The effect was to have
6350 incorrect size information for certain domains.
6352 45. #include "config.h" was missing at the start of the crypt16.c module. This
6353 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
6356 46. If there was a timeout during a "random" callout check, Exim treated it as
6357 a failure of the random address, and carried on sending RSET and the real
6358 address. If the delay was just some slowness somewhere, the response to the
6359 original RCPT would be taken as a response to RSET and so on, causing
6360 mayhem of various kinds.
6362 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
6363 when I implemented it. It didn't allow for the fact that some option values
6364 may legitimatetly be negative (e.g. size_addition), and it didn't even do
6365 the right test for positive values.
6367 48. Domain names in DNS records are case-independent. Exim always looks them up
6368 in lower case. Some resolvers return domain names in exactly the case they
6369 appear in the zone file, that is, they may contain uppercase letters. Not
6370 all resolvers do this - some return always lower case. Exim was treating a
6371 change of case by a resolver as a change of domain, similar to a widening
6372 of a domain abbreviation. This triggered its re-routing code and so it was
6373 trying to route what was effectively the same domain again. This normally
6374 caused routing to fail (because the router wouldn't handle the domain
6375 twice). Now Exim checks for this case specially, and just changes the
6376 casing of the domain that it ultimately uses when it transmits the message
6379 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
6382 50. If a filter generated a file delivery with a non-absolute name (possible if
6383 no home directory exists for the router), the forbid_file option was not
6386 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
6387 the existing equality matching.
6389 52. Exim was using ints instead of ino_t variables in some places where it was
6390 dealing with inode numbers.
6392 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
6393 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
6394 called TMPDIR, and if it finds it is different, it changes its value.
6396 54. The smtp_printf() function is now made available to local_scan() so
6397 additional output lines can be written before returning. There is also an
6398 smtp_fflush() function to enable the detection of a dropped connection.
6399 The variables smtp_input and smtp_batched_input are exported to
6402 55. Changed the default runtime configuration: the message "Unknown user"
6403 has been removed from the ACL, and instead placed on the localuser router,
6404 using the cannot_route_message feature. This means that any verification
6405 failures that generate their own messages won't get overridden. Similarly,
6406 the "Unrouteable address" message that was in the ACL for unverifiable
6407 relay addresses has also been removed.
6409 56. Added hosts_avoid_esmtp to the smtp transport.
6411 57. The exicyclog script was not checking for the esoteric option
6412 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
6413 will work only if exicyclog is run under the appropriate euid.
6415 58. Following a discussion on the list, the rules by which Exim recognises line
6416 endings on incoming messages have been changed. The -dropcr and drop_cr
6417 options are now no-ops, retained only for backwards compatibility. The
6418 following line terminators are recognized: LF CRLF CR. However, special
6419 processing applies to CR:
6421 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
6422 nor a local message in the state where . is a terminator.
6424 (ii) If a bare CR is encountered in a header line, an extra space is added
6425 after the line terminator so as not to end the header. The reasoning
6426 behind this is that bare CRs in header lines are most likely either
6427 to be mistakes, or people trying to play silly games.
6429 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
6430 was being incorrectly given as 18 bytes larger than it should have been.
6431 This is a VOB (very old bug).
6433 60. This may never have affected anything current, but just in case it has:
6434 When the local host is found other than at the start of a list of hosts,
6435 the local host, those with the same MX, and any that follow, are discarded.
6436 When the list in question was part of a longer list of hosts, the following
6437 hosts (not currently being processed) were also being discarded. This no
6438 longer happens. I'm not sure if this situation could ever has previously
6441 61. Added the "/MX" feature to lists of hosts in the manualroute and query
6444 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
6445 header. This is something that is recommended in a new Internet Draft, and
6446 is something that is documented as being done by Sendmail. There are two
6447 possible values. For messages generated by the autoreply transport, Exim
6450 Auto-Submitted: auto-replied
6452 whereas for all other generated messages (e.g. bounces) it adds
6454 Auto-Submitted: auto-generated
6456 63. The "personal" condition in filters now includes a test for the
6457 Auto-Submitted: header. If it contains the string "auto-" the message it
6458 not considered personal.
6460 64. Added rcpt_include_affixes as a generic transport option.
6462 65. Added queue_only_override (default true).
6464 66. Added the syslog_duplication option.
6466 67. If what should have been the first header line of a message consisted of
6467 a space followed by a colon, Exim was mis-interpreting it as a header line.
6468 It isn't of course - it is syntactically invalid and should therefore be
6469 treated as the start of the message body. The misbehaviour could have
6470 caused a number of strange effects, including loss of data in subsequent
6471 header lines, and spool format errors.
6473 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
6474 client host had authenticated. This control can now be exercised by an ACL
6475 for more flexibility.
6477 69. By default, callouts do not happen when testing with -bh. There is now a
6478 variant, -bhc, which does actually run the callout code, including
6479 consulting and updating the callout cache.
6481 70. Added support for saslauthd authentication, courtesy of Alexander
6484 71. If statvfs() failed on the spool or log directories while checking their
6485 size for availability, Exim confusingly gave the error "space shortage".
6486 Furthermore, in debugging mode it crashed with a floating point exception.
6487 These checks are done if check_{spool,log}_{space,inodes} are set, and when
6488 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
6489 serious problem, Exim now writes to the main and panic logs when this
6490 happens, with details of the failure. It then refuses to accept the
6491 incoming message, giving the message "spool directory problem" or "log
6492 directory problem" with a 421 code for SMTP messages.
6494 72. When Exim is about to re-exec itself, it ensures that the file descriptors
6495 0, 1, and 2 exist, because some OS complain for execs without them (see
6496 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
6497 descriptors. However, the code omitted to check that the open succeeded,
6498 causing mysterious errors if for some reason the permissions on /dev/null
6499 got screwed. Now Exim writes a message to the main and panic logs, and
6500 bombs out if it can't open /dev/null.
6502 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
6503 interact so that it is all more flexible. It is supposed to remain
6504 backwards compatible. Also added extra_local_interfaces.
6506 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
6507 to bomb out with an assertion failure - to the client this appears as a
6508 connection drop. This problem occurs in the part of the code that was taken
6509 from the Samba project. Fortunately, the assertion is in a very simple
6510 function, so I have fixed this by reproducing the function inline in the
6511 one place where it is called, and arranging for authentication to fail
6512 instead of killing the process with assert().
6514 75. The SPA client code was not working when the server requested OEM rather
6515 than Unicode encoding.
6517 76. Added code to make require_files with a specific uid setting more usable in
6518 the case where statting the file as root fails - usually a non-root-mounted
6519 NFS file system. When this happens and the failure is EACCES, Exim now
6520 forks a subprocess and does the per-uid checking as the relevant uid.
6522 77. Added process_log_path.
6524 78. If log_file_path was not explicitly set, a setting of check_log_space or
6525 check_log_inodes was ignored.
6527 79. If a space check for the spool or log partitions fails, the incident is now
6528 logged. Of course, in the latter case the data may get lost...
6530 80. Added the %p formatting code to string_format() so that it can be used to
6531 print addresses in debug_print(). Adjusted all the address printing in the
6532 debugging in store.c to use %p rather than %d.
6534 81. There was a concern that a line of code in smtp_in.c could overflow a
6535 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
6536 initially expressed, the concern was not well-founded, because trailing
6537 spaces are removed early. However, if the trailing spaces were followed by
6538 a NULL, they did not get removed, so the overflow was possible. Two fixes
6541 (a) I re-wrote the offending code in a cleaner fashion.
6542 (b) If an incoming SMTP command contains a NULL character, it is rejected
6545 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
6546 runs initgroups(), so that if the Exim user is in any additional groups,
6547 they will be used during message reception.
6553 The change log for 4.20 and earlier releases has been archived.