1 Change log file for Exim from version 4.21
2 -------------------------------------------
7 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
8 and 3.4.4 - once the server is enabled to respond to an OCSP request
9 it does even when not requested, resulting in a stapling non-aware
10 client dropping the TLS connection.
12 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
13 support variable-length bit vectors. No functional change.
15 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
16 The I= interface field on outgoing lines is now after the H= remote
17 host field, same as incoming lines. There is a separate
18 outgoing_interface log selector which allows you to disable the
21 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
22 If not running log_selector +smtp_connection the mainlog would be held
23 open indefinitely after a "too many connections" event, including to a
24 deleted file after a log rotate. Leave the per net connection logging
25 leaving it open for efficiency as that will be quickly detected by the
26 check on the next write.
28 HS/01 Bug 1671: Fix post transport crash.
29 Processing the wait-<transport> messages could crash the delivery
30 process if the message IDs didn't exist for some reason. When
31 using 'split_spool_directory=yes' the construction of the spool
32 file name failed already, exposing the same netto behaviour.
34 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
35 mime_regex ACL conditions.
37 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
38 to DSN fail messages (bounces): remote IP, remote greeting, remote response
39 to HELO, local diagnostic string.
41 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
42 log line to debug. Even when configured with a tls authenticator many
43 client connections are expected to not authenticate in this way, so
44 an authenticate fail is not an error.
46 HS/02 Add the Exim version string to the process info. This way exiwhat
47 gives some more detail about the running daemon.
49 JH/06 Bug 1395: time-limit cacheing of DNS lookups, to the TTL value. This may
50 matter for fast-change records such as DNSBLs.
52 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
53 retry record, even if constant. There may be multiple transports with
54 different interface settings and the retry behaviour needs to be kept
57 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
59 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
61 JH/10 Bug 840: fix log_defer_output option of pipe transport
63 JH/11 Bug 830: use same host for all RCPTS of a message, even under
64 hosts_randomize. This matters a lot when combined with mua_wrapper.
66 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
67 ${quote_pgsql:<string>} operator.
69 JH/13 Bug 1708: avoid misaligned access in cached lookup.
71 JH/14 Change header file name for freeradius-client. Relevant if compiling
72 with Radius support; from the Gentoo tree and checked under Fedora.
74 JH/15 Bug 1712: Introduce $prdr_requested flag variable
76 JH/16 Bug 1714: Permit an empty string as expansion result for transport
77 option transport_filter, meaning no filtering.
79 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
81 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
82 defaults to "*" (all hosts). The variable is now available when not built
83 with TLS, default unset, mainly to enable keeping the testuite sane.
84 If a server certificate is not supplied (via tls_certificate) an error is
85 logged, and clients will find TLS connections fail on startup. Presumably
86 they will retry in-clear.
87 Packagers of Exim are strongly encouraged to create a server certificate
90 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
91 with the $config_file variable.
93 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
94 in transport context, after the attempt, and per-recipient. The latter type
95 is per host attempted. The event data is the error message, and the errno
96 information encodes the lookup type (A vs. MX) used for the (first) host,
97 and the trailing two digits of the smtp 4xx reponse.
99 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
100 to write to mainlog (or rejectlog, paniclog) in the window between file
101 creation and permissions/ownership being changed. Particularly affects
102 installations where exicyclog is run as root, rather than exim user;
103 result is that the running daemon panics and dies.
105 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
107 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
108 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
109 "pri" and "weight". Note that the previous implicit priority given by the
110 list order is no longer honoured.
112 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalisation
115 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
116 by defining SUPPORT_SOCKS.
118 JH/26 Move PROXY support from Experimental to mainline, enabled for a build
119 by defining SUPPORT_PROXY. Note that the proxy_required_hosts option
120 is renamed to hosts_proxy, and the proxy_{host,target}_{address,port}.
121 variables are renamed to proxy_{local,external}_{address,port}.
123 JH/27 Move Internationalisation support from Experimental to mainline, enabled
124 for a build by defining SUPPORT_I18N
126 JH/28 Bug 1745: Fix redis lookups to handle (quoted) spaces embedded in parts
127 of the query string, and make ${quote_redis:} do that quoting.
129 JH/29 Move Events support from Experimental to mainline, enabled by default
130 and removable for a build by defining DISABLE_EVENT.
132 JH/30 Updated DANE implementation code to current from Viktor Dukhovni.
134 JH/31 Fix bug with hosts_connection_nolog and named-lists which were wrongly
135 cached by the daemon.
137 JH/32 Move Redis support from Experimental to mainline, enabled for a build
138 by defining LOOKUP_REDIS. The libhiredis library is required.
140 JH/33 Bug 1748: Permit ACL dnslists= condition in non-smtp ACLs if explicit
141 keys are given for lookup.
143 JH/34 Bug 1192: replace the embedded copy of PolarSSL RSA routines in the DKIM
144 support, by using OpenSSL or GnuTLS library ones. This means DKIM is
145 only supported when built with TLS support. The PolarSSL SHA routines
146 are still used when the TLS library is too old for convenient support.
148 JH/35 Require SINGLE_DH_USE by default in OpenSSL (main config option
149 openssl_options), for security. OpenSSL forces this from version 1.1.0
150 server-side so match that on older versions.
152 JH/36 Bug 1778: longstanding bug in memory use by the ${run } expansion: A fresh
153 allocation for $value could be released as the expansion processing
154 concluded, but leaving the global pointer active for it.
156 JH/37 Bug 1769: Permit a VRFY ACL to override the default 252 response,
157 and to use the domains and local_parts ACL conditions.
159 JH/38 Fix cutthrough bug with body lines having a single dot. The dot was
160 incorrectly not doubled on cutthrough transmission, hence seen as a
161 body-termination at the receiving system - resulting in truncated mails.
162 Commonly the sender saw a TCP-level error, and retransmitted the message
163 via the normal store-and-forward channel. This could result in duplicates
164 received - but deduplicating mailstores were liable to retain only the
165 initial truncated version.
167 JH/39 Bug 1781: Fix use of DKIM private-keys having trailing '=' in the base-64.
169 JH/40 Fix crash in queryprogram router when compiled with EXPERIMENTAL_SRS.
171 JH/41 Bug 1792: Fix selection of headers to sign for DKIM: bottom-up. While
172 we're in there, support oversigning also; bug 1309.
174 JH/42 Bug 1796: Fix error logged on a malware scanner connection failure.
176 HS/04 Add support for keep_environment and add_environment options.
178 JH/43 Tidy coding issues detected by gcc --fsanitize=undefined. Some remain;
179 either intentional arithmetic overflow during PRNG, or testing config-
182 JH/44 Bug 1800: The combination of a -bhc commandline option and cutthrough
183 delivery resulted in actual delivery. Cancel cutthrough before DATA
190 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
193 JH/02 The smtp transport option "multi_domain" is now expanded.
195 JH/03 The smtp transport now requests PRDR by default, if the server offers
198 JH/04 Certificate name checking on server certificates, when exim is a client,
199 is now done by default. The transport option tls_verify_cert_hostnames
200 can be used to disable this per-host. The build option
201 EXPERIMENTAL_CERTNAMES is withdrawn.
203 JH/05 The value of the tls_verify_certificates smtp transport and main options
204 default to the word "system" to access the system default CA bundle.
205 For GnuTLS, only version 3.0.20 or later.
207 JH/06 Verification of the server certificate for a TLS connection is now tried
208 (but not required) by default. The verification status is now logged by
209 default, for both outbound TLS and client-certificate supplying inbound
212 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
215 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
216 Status Notification (bounce) messages are now MIME format per RFC 3464.
217 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
218 under the control of the dsn_advertise_hosts option, and routers may
219 have a dsn_lasthop option.
221 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
222 default, modifiable by a malware= option. The list separator for
223 the options can now be changed in the usual way. Bug 68.
225 JH/10 The smtp_receive_timeout main option is now expanded before use.
227 JH/11 The incoming_interface log option now also enables logging of the
228 local interface on delivery outgoing connections.
230 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
231 if the interface and destination host and port all match.
233 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
236 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
237 Patch from Andrew Lewis.
239 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
240 now supports optional time-restrictions, weighting, and priority
241 modifiers per server. Patch originally by <rommer@active.by>.
243 JH/16 The spamd_address main option now supports a mixed list of local
244 and remote servers. Remote servers can be IPv6 addresses, and
245 specify a port-range.
247 JH/17 Bug 68: The spamd_address main option now supports an optional
248 timeout value per server.
250 JH/18 Bug 1581: Router and transport options headers_add/remove can
251 now have the list separator specified.
253 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
256 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
259 JH/21 Support for the A6 type of dns record is withdrawn.
261 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
262 rather than the verbs used.
264 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
265 from 255 to 1024 chars.
267 JH/24 Verification callouts now attempt to use TLS by default.
269 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
270 are generic router options now. The defaults didn't change.
272 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
273 Original patch from Alexander Shikoff, worked over by JH.
275 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
276 files if ZCAT_COMMAND is not executable.
278 JH/26 Bug 1539: Add timout/retry options on dnsdb lookups.
280 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
282 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
283 Normally benign, it bites when the pair was led to by a CNAME;
284 modern usage is to not canoicalize the domain to a CNAME target
285 (and we were inconsistent anyway for A-only vs AAAA+A).
287 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
289 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
290 when evaluating $sender_host_dnssec.
292 JH/31 Check the HELO verification lookup for DNSSEC, adding new
293 $sender_helo_dnssec variable.
295 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
297 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
299 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
301 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
302 documented as working, but never had. Support all but $spam_report.
304 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
305 added for tls authenticator.
310 TL/01 When running the test suite, the README says that variables such as
311 no_msglog_check are global and can be placed anywhere in a specific
312 test's script, however it was observed that placement needed to be near
313 the beginning for it to behave that way. Changed the runtest perl
314 script to read through the entire script once to detect and set these
315 variables, reset to the beginning of the script, and then run through
316 the script parsing/test process like normal.
318 TL/02 The BSD's have an arc4random API. One of the functions to induce
319 adding randomness was arc4random_stir(), but it has been removed in
320 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
321 function when detected.
323 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
324 cause callback expansion.
326 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
327 syntax errors in an expansion can be treated as a string instead of
328 logging or causing an error, due to the internal use of bool_lax
329 instead of bool when processing it.
331 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
332 server certificates when making smtp deliveries.
334 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
336 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
338 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
340 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
341 Merged patch from Sebastian Wiedenroth.
343 JH/05 Fix results-pipe from transport process. Several recipients, combined
344 with certificate use, exposed issues where response data items split
345 over buffer boundaries were not parsed properly. This eventually
346 resulted in duplicates being sent. This issue only became common enough
347 to notice due to the introduction of conection certificate information,
348 the item size being so much larger. Found and fixed by Wolfgang Breyha.
350 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
351 size buffer was used, resulting in syntax errors when an expansion
354 JH/07 Add support for directories of certificates when compiled with a GnuTLS
355 version 3.3.6 or later.
357 JH/08 Rename the TPDA expermimental facility to Event Actions. The #ifdef
358 is EXPERIMENTAL_EVENT, the main-configuration and transport options
359 both become "event_action", the variables become $event_name, $event_data
360 and $event_defer_errno. There is a new variable $verify_mode, usable in
361 routers, transports and related events. The tls:cert event is now also
362 raised for inbound connections, if the main configuration event_action
365 TL/06 In test suite, disable OCSP for old versions of openssl which contained
366 early OCSP support, but no stapling (appears to be less than 1.0.0).
368 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
369 server certificate names available under the smtp transport option
370 "tls_verify_cert_hostname" now do not permit multi-component wildcard
373 JH/10 Time-related extraction expansions from certificates now use the main
374 option "timezone" setting for output formatting, and are consistent
375 between OpenSSL and GnuTLS compilations. Bug 1541.
377 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
378 encoded parameter in the incoming message. Bug 1558.
380 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
381 include certificate info, eximon was claiming there were spoolfile
384 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
386 JH/14 Log delivery-related information more consistently, using the sequence
387 "H=<name> [<ip>]" wherever possible.
389 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
390 are problematic for Debian distribution, omit them from the release
393 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
395 JH/16 Fix string representation of time values on 64bit time_t anchitectures.
398 JH/17 Fix a null-indirection in certextract expansions when a nondefault
399 output list separator was used.
404 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
405 checkers that were complaining about end of non-void function with no
408 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
409 This was a regression intruduced in 4.83 by another bugfix.
411 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
413 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
414 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
420 TF/01 Correctly close the server side of TLS when forking for delivery.
422 When a message was received over SMTP with TLS, Exim failed to clear up
423 the incoming connection properly after forking off the child process to
424 deliver the message. In some situations the subsequent outgoing
425 delivery connection happened to have the same fd number as the incoming
426 connection previously had. Exim would try to use TLS and fail, logging
427 a "Bad file descriptor" error.
429 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
430 utilities have not been installed.
432 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
433 temporary space as the ACL may create new global variables.
435 TL/01 LDAP support uses per connection or global context settings, depending
436 upon the detected version of the libraries at build time.
438 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
439 to extract and use the src ip:port in logging and expansions as if it
440 were a direct connection from the outside internet. PPv2 support was
441 updated based on HAProxy spec change in May 2014.
443 JH/02 Add ${listextract {number}{list}{success}{fail}}.
445 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
446 Properly escape header and check for NULL return.
448 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
451 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
453 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
454 characters in header names, implemented as a verify condition.
455 Contributed by Michael Fischer v. Mollard.
457 TL/05 Rename SPF condition results err_perm and err_temp to standardized
458 results permerror and temperror. Previous values are deprecated but
459 still accepted. In a future release, err_perm and err_temp will be
460 completely removed, which will be a backward incompatibility if the
461 ACL tests for either of these two old results. Patch contributed by
462 user bes-internal on the mailing list.
464 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
466 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
467 selectors, in both main and reject logs.
469 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
472 JH/07 Add malware type "sock" for talking to simple daemon.
474 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
476 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
477 routers/transports under cutthrough routing.
479 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
480 numbers. Touch up "bool" conditional to keep the same definition.
482 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
484 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
486 JH/12 Expand items in router/transport headers_add or headers_remove lists
487 individually rather than the list as a whole. Bug 1452.
489 Required for reasonable handling of multiple headers_ options when
490 they may be empty; requires that headers_remove items with embedded
491 colons must have them doubled (or the list-separator changed).
493 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
494 view the policy declared in the DMARC record. Currently, $dmarc_status
495 is a combined value of both the record presence and the result of the
498 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
500 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
501 dnslookup router and the smtp transport (applying to the forward
504 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
505 of ldap servers used for a specific lookup. Patch provided by Heiko
508 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
509 New variable $lookup_dnssec_authenticated for observability.
511 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
512 Patch submitted by Lars Timman.
514 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
516 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
517 Requires trusted mode and valid format message id, aborts otherwise.
518 Patch contributed by Heiko Schlichting.
520 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
521 certextract with support for various fields. Bug 1358.
523 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
524 is requested by default, modifiable by smtp transport option
527 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
528 operate on certificate variables to give certificate fingerprints
529 Also new ${sha256:cert_variable}.
531 JH/23 The PRDR feature is moved from being Experimental into the mainline.
533 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
534 Christian Aistleitner.
536 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
538 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
539 file. Patch from Wolfgang Breyha.
541 JH/25 Expand the coverage of the delivery $host and $host_address to
542 client authenticators run in verify callout. Bug 1476.
544 JH/26 Port service names are now accepted for tls_on_connect_ports, to
545 align with daemon_smtp_ports. Bug 72.
547 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
548 support and error reporting did not work properly.
550 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
551 and is readable. Patch from Andrew Colin Kissa.
553 TL/14 Enhance documentation of ${run expansion and how it parses the
554 commandline after expansion, particularly in the case when an
555 unquoted variable expansion results in an empty value.
557 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
559 PP/02 Fix internal collision of T_APL on systems which support RFC3123
560 by renaming away from it. Addresses GH issue 15, reported by
563 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
565 TL/15 SECURITY: prevent double expansion in math comparison functions
566 (can expand unsanitized data). Not remotely exploitable.
573 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
575 PP/02 Make -n do something, by making it not do something.
576 When combined with -bP, the name of an option is not output.
578 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
581 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
582 $sender_host_name and config options to manage this, and basic check
585 PP/05 DSCP support for outbound connections and control modifier for inbound.
587 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
588 (Only plugin which currently uses this is kerberos4, which nobody should
589 be using, but we should make it available and other future plugins might
590 conceivably use it, even though it would break NAT; stuff *should* be
591 using channel bindings instead).
593 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
594 name; added for Sendmail compatibility; requires admin caller.
595 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
596 just ignore it); requires trusted caller.
597 Also parse but ignore: -Ac -Am -X<logfile>
600 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
602 TL/02 Add +smtp_confirmation as a default logging option.
604 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
605 Patch by Magnus Holmgren from 2007-02-20.
607 TL/04 Bugzilla 1281 - Spec typo.
608 Bugzilla 1283 - Spec typo.
609 Bugzilla 1290 - Spec grammar fixes.
611 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
613 TL/06 Add Experimental DMARC support using libopendmarc libraries.
615 TL/07 Fix an out of order global option causing a segfault. Reported to dev
616 mailing list by by Dmitry Isaikin.
618 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
620 JH/02 Support "G" suffix to numbers in ${if comparisons.
622 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
624 NM/01 Bugzilla 1197 - Spec typo
625 Bugzilla 1196 - Spec examples corrections
627 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
629 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
630 gnutls_enable_pkcs11, but renamed to more accurately indicate its
633 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
634 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
636 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
637 "acl {{name}{arg}...}", and optional args on acl condition
640 JH/05 Permit multiple router/transport headers_add/remove lines.
642 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
644 JH/07 Avoid using a waiting database for a single-message-only transport.
645 Performance patch from Paul Fisher. Bugzilla 1262.
647 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
650 JH/09 Add $headers_added variable, with content from use of ACL modifier
651 add_header (but not yet added to the message). Bugzilla 199.
653 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
654 Pulled from Bugzilla 817 by Wolfgang Breyha.
656 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
658 (nb: this is the same fix as in Exim 4.80.1)
660 JH/11 Add A= logging on delivery lines, and a client_set_id option on
663 JH/12 Add optional authenticated_sender logging to A= and a log_selector
666 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
668 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
669 advertise SMTP AUTH mechanism to us, instead of a generic
670 protocol violation error. Also, make Exim more robust to bad
671 data from the Dovecot auth socket.
673 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
675 When a queue runner is handling a message, Exim first routes the
676 recipient addresses, during which it prunes them based on the retry
677 hints database. After that it attempts to deliver the message to
678 any remaining recipients. It then updates the hints database using
681 So if a recipient address works intermittently, it can get repeatedly
682 deferred at routing time. The retry hints record remains fresh so the
683 address never reaches the final cutoff time.
685 This is a fairly common occurrence when a user is bumping up against
686 their storage quota. Exim had some logic in its local delivery code
687 to deal with this. However it did not apply to per-recipient defers
688 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
690 This change adds a proper retry rule check during routing so that the
691 final cutoff time is checked against the message's age. We only do
692 this check if there is an address retry record and there is not a
693 domain retry record; this implies that previous attempts to handle
694 the address had the retry_use_local_parts option turned on. We use
695 this as an approximation for the destination being like a local
696 delivery, as in LMTP.
698 I suspect this new check makes the old local delivery cutoff check
699 redundant, but I have not verified this so I left the code in place.
701 TF/02 Correct gecos expansion when From: is a prefix of the username.
703 Test 0254 submits a message to Exim with the header
707 When I ran the test suite under the user fanf2, Exim expanded
708 the header to contain my full name, whereas it should have added
709 a Resent-Sender: header. It erroneously treats any prefix of the
710 username as equal to the username.
712 This change corrects that bug.
714 GF/01 DCC debug and logging tidyup
715 Error conditions log to paniclog rather than rejectlog.
716 Debug lines prefixed by "DCC: " to remove any ambiguity.
718 TF/03 Avoid unnecessary rebuilds of lookup-related code.
720 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
721 Bug spotted by Jeremy Harris; was flawed since initial commit.
722 Would have resulted in OCSP responses post-SNI triggering an Exim
723 NULL dereference and crash.
725 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
727 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
728 Bug detection, analysis and fix by Samuel Thibault.
729 Bugzilla 1331, Debian bug #698092.
731 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
733 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
734 Server implementation by Todd Lyons, client by JH.
735 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
736 config variable "prdr_enable" controls whether the server
737 advertises the facility. If the client requests PRDR a new
738 acl_data_smtp_prdr ACL is called once for each recipient, after
739 the body content is received and before the acl_smtp_data ACL.
740 The client is controlled by bolth of: a hosts_try_prdr option
741 on the smtp transport, and the server advertisement.
742 Default client logging of deliveries and rejections involving
743 PRDR are flagged with the string "PRDR".
745 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
746 fclose(). Diagnosis by Todd Lyons.
748 PP/17 Update configure.default to handle IPv6 localhost better.
749 Patch by Alain Williams (plus minor tweaks).
752 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
753 This is now consistent with GnuTLS, and is now documented: the
754 previous undocumented portable approach to treating the option as
755 unset was to force an expansion failure. That still works, and
756 an empty string is now equivalent.
758 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
759 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
760 not performing validation itself.
762 PP/20 Added force_command boolean option to pipe transport.
763 Patch from Nick Koston, of cPanel Inc.
765 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
768 TF/04 Added udpsend ACL modifer and hexquote expansion operator
770 PP/21 Fix eximon continuous updating with timestamped log-files.
771 Broken in a format-string cleanup in 4.80, missed when I repaired the
772 other false fix of the same issue.
773 Report and fix from Heiko Schlichting.
776 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
777 Report from Prashanth Katuri.
779 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
780 It's SecureTransport, so affects any MacOS clients which use the
781 system-integrated TLS libraries, including email clients.
783 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
784 using a MIME ACL for non-SMTP local injection.
785 Report and assistance in diagnosis by Warren Baker.
787 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
789 JH/16 Fix comparisons for 64b. Bugzilla 1385.
791 TL/09 Add expansion variable $authenticated_fail_id to keep track of
792 last id that failed so it may be referenced in subsequent ACL's.
794 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
797 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
798 ldap library initialization, allowing self-signed CA's to be
799 used. Also properly sets require_cert option later in code by
800 using NULL (global ldap config) instead of ldap handle (per
801 session). Bug diagnosis and testing by alxgomz.
803 TL/12 Enhanced documentation in the ratelimit.pl script provided in
804 the src/util/ subdirectory.
806 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
807 renamed to Transport Post Delivery Action by Jeremy Harris, as
810 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
811 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
812 redis_servers = needs to be configured which will be used by the redis
813 lookup. Patch from Warren Baker, of The Packet Hub.
815 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
817 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
818 hostname or reverse DNS when processing a host list. Used suggestions
819 from multiple comments on this bug.
821 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
823 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
824 Missed a few lines, added it to make the runtest require no keyboard
827 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
828 contains upper case chars. Make router use caseful_local_part.
830 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
831 support when GnuTLS has been built with p11-kit.
837 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
839 This, or similar/improved, will also be change PP/11 of 4.82.
845 PP/01 Handle short writes when writing local log-files.
846 In practice, only affects FreeBSD (8 onwards).
847 Bugzilla 1053, with thanks to Dmitry Isaikin.
849 NM/01 Bugzilla 949 - Documentation tweak
851 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
854 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
856 PP/02 Implemented gsasl authenticator.
858 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
860 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
861 `pkg-config foo` for cflags/libs.
863 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
864 with rest of GSASL and with heimdal_gssapi.
866 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
867 `pkg-config foo` for cflags/libs for the TLS implementation.
869 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
870 properties get this fed in as external SSF. A number of robustness
871 and debugging improvements to the cyrus_sasl authenticator.
873 PP/08 cyrus_sasl server now expands the server_realm option.
875 PP/09 Bugzilla 1214 - Log authentication information in reject log.
876 Patch by Jeremy Harris.
878 PP/10 Added dbmjz lookup type.
880 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
882 PP/12 MAIL args handles TAB as well as SP, for better interop with
883 non-compliant senders.
884 Analysis and variant patch by Todd Lyons.
886 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
887 Bug report from Lars Müller <lars@samba.org> (via SUSE),
888 Patch from Dirk Mueller <dmueller@suse.com>
890 PP/13 tls_peerdn now print-escaped for spool files.
891 Observed some $tls_peerdn in wild which contained \n, which resulted
892 in spool file corruption.
894 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
895 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
896 or write after TLS renegotiation, which otherwise led to messages
899 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
900 as a tracking header (ie: a signed header comes before the signature).
901 Patch from Wolfgang Breyha.
903 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
904 comma-sep list; embedded commas doubled.
906 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
908 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
910 Report and patch from Dmitry Banschikov.
912 PP/16 Removed "dont_insert_empty_fragments" fron "openssl_options".
913 Removed SSL_clear() after SSL_new() which led to protocol negotiation
914 failures. We appear to now support TLS1.1+ with Exim.
916 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
917 lets Exim select keys and certificates based upon TLS SNI from client.
918 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
919 before an outbound SMTP session. New log_selector, +tls_sni.
921 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
922 NULL dereference. Report and patch from Alun Jones.
924 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
925 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
926 Not seeing resolver debug output on NetBSD, but suspect this is a
927 resolver implementation change.
929 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
930 Left warnings. Added "eximon gdb" invocation mode.
932 PP/21 Defaulting "accept_8bitmime" to true, not false.
934 PP/22 Added -bw for inetd wait mode support.
936 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
937 locate the relevant includes and libraries. Made this the default.
939 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
940 Bugzilla 1246, report and most of solution from Tomasz Kusy.
942 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
943 This may cause build issues on older platforms.
945 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
946 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
947 gnutls_require_mac & gnutls_require_protocols (no longer supported).
948 Added SNI support via GnuTLS too.
949 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
951 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
953 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
954 Applied second patch from Janne, implementing suggestion to default
955 multiple-strings-in-record handling to match SPF spec.
957 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
959 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
960 read-only, out of scope).
961 Patch from Wolfgang Breyha, report from Stuart Northfield.
963 PP/29 Fix three issues highlighted by clang analyser static analysis.
964 Only crash-plausible issue would require the Cambridge-specific
965 iplookup router and a misconfiguration.
966 Report from Marcin Mirosław.
968 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
970 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
971 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
972 As part of this, removing so much warning spew let me fix some minor
973 real issues in debug logging.
975 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
976 assignment on my part. Fixed.
978 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
979 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
980 Janne Snabb (who went above and beyond: thank you).
982 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
983 string otherwise requires a connection and a bunch more work and it's
984 relatively easy to get wrong. Should also expose TLS library linkage
987 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
988 64-bit ${eval} (JH/03).
990 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
991 GNU libc to support some of the 64-bit stuff, should not lead to
992 conflicts. Defined before os.h is pulled in, so if a given platform
993 needs to override this, it can.
995 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
996 protection layer was required, which is not implemented.
997 Bugzilla 1254, patch from Wolfgang Breyha.
999 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
1000 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
1001 tls_dhparam take prime identifiers. Also unbreak combination of
1002 OpenSSL+DH_params+TLSSNI.
1004 PP/39 Disable SSLv2 by default in OpenSSL support.
1010 PP/01 Solaris build fix for Oracle's LDAP libraries.
1011 Bugzilla 1109, patch from Stephen Usher.
1013 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
1015 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
1018 TF/02 Fix a couple more cases where we did not log the error message
1019 when unlink() failed. See also change 4.74-TF/03.
1021 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
1022 lock up or crash if it happened to be inside a call to libc when it
1023 got a SIGUSR1 from exiwhat.
1025 The SIGUSR1 handler appends the current process status to the process
1026 log which is later printed by exiwhat. It used to use the general
1027 purpose logging code to do this, but several functions it calls are
1028 not safe for signals.
1030 The new output code in the SIGUSR1 handler is specific to the process
1031 log, and simple enough that it's easy to inspect for signal safety.
1032 Removing some special cases also simplifies the general logging code.
1033 Removing the spurious timestamps from the process log simplifies
1036 TF/04 Improved ratelimit ACL condition.
1038 The /noupdate option has been deprecated in favour of /readonly which
1039 has clearer semantics. The /leaky, /strict, and /readonly update modes
1040 are mutually exclusive. The update mode is no longer included in the
1041 database key; it just determines when the database is updated. (This
1042 means that when you upgrde Exim will forget old rate measurements.)
1044 Exim now checks that the per_* options are used with an update mode that
1045 makes sense for the current ACL. For example, when Exim is processing a
1046 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
1047 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
1048 must specify per_mail/readonly. If you omit the update mode it defaults to
1049 /leaky where that makes sense (as before) or /readonly where required.
1051 The /noupdate option is now undocumented but still supported for
1052 backwards compatibility. It is equivalent to /readonly except that in
1053 ACLs where /readonly is required you may specify /leaky/noupdate or
1054 /strict/noupdate which are treated the same as /readonly.
1056 A useful new feature is the /count= option. This is a generalization
1057 of the per_byte option, so that you can measure the throughput of other
1058 aggregate values. For example, the per_byte option is now equivalent
1059 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
1061 The per_rcpt option has been generalized using the /count= mechanism
1062 (though it's more complicated than the per_byte equivalence). When it is
1063 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
1064 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
1065 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
1066 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
1067 Note that using per_rcpt with a non-readonly update mode in more than
1068 one ACL will cause the recipients to be double-counted. (The per_mail
1069 and per_byte options don't have this problem.)
1071 The handling of very low rates has changed slightly. If the computed rate
1072 is less than the event's count (usually one) then this event is the first
1073 after a long gap. In this case the rate is set to the same as this event's
1074 count, so that the first message of a spam run is counted properly.
1076 The major new feature is a mechanism for counting the rate of unique
1077 events. The new per_addr option counts the number of different
1078 recipients that someone has sent messages to in the last time period. It
1079 behaves like per_rcpt if all the recipient addresses are different, but
1080 duplicate recipient addresses do not increase the measured rate. Like
1081 the /count= option this is a general mechanism, so the per_addr option
1082 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
1083 example, measure the rate that a client uses different sender addresses
1084 with the options per_mail/unique=$sender_address. There are further
1085 details in the main documentation.
1087 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
1089 TF/06 Removed a few PCRE remnants.
1091 TF/07 Automatically extract Exim's version number from tags in the git
1092 repository when doing development or release builds.
1094 PP/02 Raise smtp_cmd_buffer_size to 16kB.
1095 Bugzilla 879. Patch from Paul Fisher.
1097 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
1098 Heavily based on revision 40f9a89a from Simon Arlott's tree.
1101 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
1103 PP/05 Variable $av_failed, true if the AV scanner deferred.
1104 Bugzilla 1078. Patch from John Horne.
1106 PP/06 Stop make process more reliably on build failure.
1107 Bugzilla 1087. Patch from Heiko Schlittermann.
1109 PP/07 Make maildir_use_size_file an _expandable_ boolean.
1110 Bugzilla 1089. Patch from Heiko Schlittermann.
1112 PP/08 Handle ${run} returning more data than OS pipe buffer size.
1113 Bugzilla 1131. Patch from Holger Weiß.
1115 PP/09 Handle IPv6 addresses with SPF.
1116 Bugzilla 860. Patch from Wolfgang Breyha.
1118 PP/10 GnuTLS: support TLS 1.2 & 1.1.
1120 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
1123 PP/11 match_* no longer expand right-hand-side by default.
1124 New compile-time build option, EXPAND_LISTMATCH_RHS.
1125 New expansion conditions, "inlist", "inlisti".
1127 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
1129 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
1131 PP/14 fix log_write() format string regression from TF/03.
1132 Bugzilla 1152. Patch from Dmitry Isaikin.
1138 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
1140 PP/02 Harmonised TLS library version reporting; only show if debugging.
1141 Layout now matches that introduced for other libraries in 4.74 PP/03.
1143 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
1145 PP/04 New "dns_use_edns0" global option.
1147 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
1150 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
1151 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
1153 TK/01 Updated PolarSSL code to 0.14.2.
1154 Bugzilla 1097. Patch from Andreas Metzler.
1156 PP/07 Catch divide-by-zero in ${eval:...}.
1157 Fixes bugzilla 1102.
1159 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
1162 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
1163 format-string attack -- SECURITY: remote arbitrary code execution.
1165 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
1166 time unintentionally subject to list matching rules, letting the header
1167 cause arbitrary Exim lookups (of items which can occur in lists, *not*
1168 arbitrary string expansion). This allowed for information disclosure.
1170 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
1171 INT_MIN/-1 -- value coerced to INT_MAX.
1177 NM/01 Workround for PCRE version dependency in version reporting
1180 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
1181 This fixes portability to compilers other than gcc, notably
1182 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
1184 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
1185 makefiles for portability to HP-UX and POSIX correctness.
1187 PP/01 Permit LOOKUP_foo enabling on the make command-line.
1188 Also via indented variable definition in the Makefile.
1189 (Debugging by Oliver Heesakkers).
1191 PP/02 Restore caching of spamd results with expanded spamd_address.
1192 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
1194 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
1195 Improves build reliability. Fix from: Frank Elsner
1197 NM/02 Fix wide character breakage in the rfc2047 coding
1198 Fixes bug 1064. Patch from Andrey N. Oktyabrski
1200 NM/03 Allow underscore in dnslist lookups
1201 Fixes bug 1026. Patch from Graeme Fowler
1203 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
1204 Code patches from Adam Ciarcinski of NetBSD.
1206 NM/04 Fixed exiqgrep to cope with mailq missing size issue
1209 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
1210 is logged, to avoid truncation. Patch from John Horne.
1212 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
1213 Patch from Jakob Hirsch.
1215 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
1216 SQL string expansion failure details.
1217 Patch from Andrey Oktyabrski.
1219 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
1220 Patch from Simon Arlott.
1222 PP/09 New lookups functionality failed to compile on old gcc which rejects
1223 extern declarations in function scope.
1224 Patch from Oliver Fleischmann
1226 PP/10 Use sig_atomic_t for flags set from signal handlers.
1227 Check getgroups() return and improve debugging.
1228 Fixed developed for diagnosis in bug 927 (which turned out to be
1231 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
1232 Patch from Mark Zealey.
1234 PP/12 Bugzilla 1056: Improved spamd server selection.
1235 Patch from Mark Zealey.
1237 PP/13 Bugzilla 1086: Deal with maildir quota file races.
1238 Based on patch from Heiko Schlittermann.
1240 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
1241 Patch from Uwe Doering, sign-off by Michael Haardt.
1243 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
1244 variable declaration deep within a block. Bug and patch from
1247 PP/15 lookups-Makefile IRIX compatibilty coercion.
1249 PP/16 Make DISABLE_DKIM build knob functional.
1251 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
1252 Patch by Simon Arlott
1254 TF/03 Fix valgrind.h portability to C89 compilers that do not support
1255 variable argument macros. Our copy now differs from upstream.
1261 TF/01 Failure to get a lock on a hints database can have serious
1262 consequences so log it to the panic log.
1264 TF/02 Log LMTP confirmation messages in the same way as SMTP,
1265 controlled using the smtp_confirmation log selector.
1267 TF/03 Include the error message when we fail to unlink a spool file.
1269 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
1270 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
1271 for maintaining out-of-tree patches for some time.
1273 PP/01 Bugzilla 139: Documentation and portability issues.
1274 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
1275 Handle per-OS dynamic-module compilation flags.
1277 PP/02 Let /dev/null have normal permissions.
1278 The 4.73 fixes were a little too stringent and complained about the
1279 permissions on /dev/null. Exempt it from some checks.
1280 Reported by Andreas M. Kirchwitz.
1282 PP/03 Report version information for many libraries, including
1283 Exim version information for dynamically loaded libraries. Created
1284 version.h, now support a version extension string for distributors
1285 who patch heavily. Dynamic module ABI change.
1287 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
1288 privilege escalation vulnerability whereby the Exim run-time user
1289 can cause root to append content of the attacker's choosing to
1292 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
1295 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
1296 If dropping privileges for untrusted macros, we disabled normal logging
1297 on the basis that it would fail; for the Exim run-time user, this is not
1298 the case, and it resulted in successful deliveries going unlogged.
1299 Fixed. Reported by Andreas Metzler.
1305 PP/01 Date: & Message-Id: revert to normally being appended to a message,
1306 only prepend for the Resent-* case. Fixes regression introduced in
1307 Exim 4.70 by NM/22 for Bugzilla 607.
1309 PP/02 Include check_rfc2047_length in configure.default because we're seeing
1310 increasing numbers of administrators be bitten by this.
1312 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
1314 PP/03 Bugzilla 994: added openssl_options main configuration option.
1316 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
1318 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
1320 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
1322 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
1323 without a peer certificate, leading to a segfault because of an
1324 assumption that peers always have certificates. Be a little more
1325 paranoid. Problem reported by Martin Tscholak.
1327 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
1328 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
1329 NB: ClamAV planning to remove STREAM in "middle of 2010".
1330 CL also introduces -bmalware, various -d+acl logging additions and
1331 more caution in buffer sizes.
1333 PP/09 Implemented reverse_ip expansion operator.
1335 PP/10 Bugzilla 937: provide a "debug" ACL control.
1337 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
1339 PP/12 Bugzilla 973: Implement --version.
1341 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
1343 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
1345 PP/15 Bugzilla 816: support multiple condition rules on Routers.
1347 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
1348 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
1349 ignore trailing whitespace.
1351 JJ/02 prevent non-panic DKIM error from being sent to paniclog
1353 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
1356 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
1357 Notification from Dr Andrew Aitchison.
1359 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
1360 ExtendedDetectionInfo response format.
1361 Notification from John Horne.
1363 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
1366 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
1367 XSL and documented dependency on system catalogs, with examples of how
1370 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
1373 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
1374 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
1375 configuration file which is writeable by the Exim user or group.
1377 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
1378 of configuration files to cover files specified with the -C option if
1379 they are going to be used with root privileges, not just the default
1382 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
1383 option (effectively making it always true).
1385 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
1386 files to be used while preserving root privileges.
1388 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
1389 that rogue child processes cannot use them.
1391 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
1392 run-time user, instead of root.
1394 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
1395 Exim run-time user without dropping privileges.
1397 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
1398 result string, instead of calling string_vformat() twice with the same
1401 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
1402 for other users. Others should always drop root privileges if they use
1403 -C on the command line, even for a whitelisted configure file.
1405 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
1407 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
1413 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
1414 $data_path, and $header_path variables; fixed documentation bugs and
1417 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
1418 exipick to access non-standard spools, including the "frozen" queue
1421 NM/01 Bugzilla 965: Support mysql stored procedures.
1422 Patch from Alain Williams
1424 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
1426 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
1427 Patch from Andreas Metzler
1429 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
1430 Patch from Kirill Miazine
1432 NM/05 Bugzilla 671: Added umask to procmail example.
1434 JJ/03 installed exipick 20100323.0, fixing doc bug
1436 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
1437 directory. Notification and patch from Dan Rosenberg.
1439 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
1441 TK/02 Improve log output when DKIM signing operation fails.
1443 MH/01 Treat the transport option dkim_domain as a colon separated
1444 list, not as a single string, and sign the message with each element,
1445 omitting multiple occurences of the same signer.
1447 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
1448 Bugzilla 985, 986. Patch by Simon Arlott
1450 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
1451 Patch by Simon Arlott
1453 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
1454 MBX locking. Notification from Dan Rosenberg.
1460 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
1462 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
1464 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
1466 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
1468 NM/04 Bugzilla 915: Flag broken perl installation during build.
1474 TK/01 Added patch by Johannes Berg that expands the main option
1475 "spamd_address" if it starts with a dollar sign.
1477 TK/02 Write list of recipients to X-Envelope-Sender header when building
1478 the mbox-format spool file for content scanning (suggested by Jakob
1481 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
1482 (http://www.dcc-servers.net/) support via dccifd. Activated by
1483 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
1485 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
1486 by Mark Daniel Reidel <mr@df.eu>.
1488 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
1489 When building exim an external PCRE library is now needed -
1490 PCRE is a system library on the majority of modern systems.
1491 See entry on PCRE_LIBS in EDITME file.
1493 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
1494 conversation. Added nologin parameter to request.
1495 Patch contributed by Kirill Miazine.
1497 TF/01 Do not log submission mode rewrites if they do not change the address.
1499 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
1501 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
1502 log files in place. Contributed by Roberto Lima.
1504 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
1506 TF/03 Bugzilla 615: When checking the local_parts router precondition
1507 after a local_part_suffix or local_part_prefix option, Exim now
1508 does not use the address's named list lookup cache, since this
1509 contains cached lookups for the whole local part.
1511 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
1512 Robert Millan. Documentation is in experimental-spec.txt.
1514 TF/04 Bugzilla 668: Fix parallel build (make -j).
1516 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
1518 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
1519 Patch provided by Jan Srzednicki.
1521 TF/05 Leading white space used to be stripped from $spam_report which
1522 wrecked the formatting. Now it is preserved.
1524 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
1525 that they are available at delivery time.
1527 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
1529 TF/08 TLS error reporting now respects the incoming_interface and
1530 incoming_port log selectors.
1532 TF/09 Produce a more useful error message if an SMTP transport's hosts
1533 setting expands to an empty string.
1535 NM/06 Bugzilla 744: EXPN did not work under TLS.
1536 Patch provided by Phil Pennock.
1538 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
1539 Patch provided by Richard Godbee.
1541 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
1542 acl_smtp_notquit, added index entry.
1544 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
1545 Patch provided by Eugene Bujak.
1547 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
1548 accept(). Patch provided by Maxim Dounin.
1550 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
1551 Patch provided by Phil Pennock.
1553 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
1555 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
1556 Patch provided by Brad "anomie" Jorsch.
1558 NM/14 Bugzilla 622: Added timeout setting to transport filter.
1559 Patch provided by Dean Brooks.
1561 TK/05 Add native DKIM support (does not depend on external libraries).
1563 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
1564 Patch provided by Graeme Fowler.
1566 NM/16 Bugzilla 851: Documentation example syntax fix.
1568 NM/17 Changed NOTICE file to remove references to embedded PCRE.
1570 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
1573 NM/19 Bugzilla 745: TLS version reporting.
1574 Patch provided by Phil Pennock.
1576 NM/20 Bugzilla 167: bool: condition support.
1577 Patch provided by Phil Pennock.
1579 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
1580 clients. Patch provided by Phil Pennock.
1582 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
1583 Patch provided by Brad "anomie" Jorsch.
1585 NM/23 Bugzilla 687: Fix misparses in eximstats.
1586 Patch provided by Heiko Schlittermann.
1588 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
1589 Patch provided by Heiko Schlittermann.
1591 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
1592 plus update to original patch.
1594 NM/26 Bugzilla 799: Documentation correction for ratelimit.
1596 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
1597 Patch provided by David Brownlee.
1599 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
1601 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
1603 NM/30 Bugzilla 888: TLS documentation bugfixes.
1605 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
1607 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
1608 Unlike the original bugzilla I have changed all shell scripts in src tree.
1610 NM/33 Bugzilla 898: Transport filter timeout fix.
1611 Patch by Todd Rinaldo.
1613 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
1614 Patch by Serge Demonchaux.
1616 NM/35 Bugzilla 39: Base64 decode bug fixes.
1617 Patch by Jakob Hirsch.
1619 NM/36 Bugzilla 909: Correct connect() call in dcc code.
1621 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
1623 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
1625 NM/39 Bugzilla 911: Fixed MakeLinks build script.
1631 TK/01 Add preliminary DKIM support. Currently requires a forked version of
1632 ALT-N's libdkim that I have put here:
1633 http://duncanthrax.net/exim-experimental/
1635 Note to Michael Haardt: I had to rename some vars in sieve.c. They
1636 were called 'true' and it seems that C99 defines that as a reserved
1637 keyword to be used with 'bool' variable types. That means you could
1638 not include C99-style headers which use bools without triggering
1639 build errors in sieve.c.
1641 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
1642 as mailq or other aliases. Changed the --help handling significantly
1643 to do whats expected. exim_usage() emits usage/help information.
1645 SC/01 Added the -bylocaldomain option to eximstats.
1647 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
1649 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
1651 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
1657 PH/01 Another patch from the Sieve maintainer.
1659 PH/02 When an IPv6 address is converted to a string for single-key lookup
1660 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
1661 dots are used instead of colons so that keys in lsearch files need not
1662 contain colons. This was done some time before quoting was made available
1663 in lsearch files. However, iplsearch files do require colons in IPv6 keys
1664 (notated using the quote facility) so as to distinguish them from IPv4
1665 keys. This meant that lookups for IP addresses in host lists did not work
1666 for iplsearch lookups.
1668 This has been fixed by arranging for IPv6 addresses to be expressed with
1669 colons if the lookup type is iplsearch. This is not incompatible, because
1670 previously such lookups could never work.
1672 The situation is now rather anomolous, since one *can* have colons in
1673 ordinary lsearch keys. However, making the change in all cases is
1674 incompatible and would probably break a number of configurations.
1676 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
1679 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
1680 conversion specification without a maximum field width, thereby enabling
1681 a rogue spamd server to cause a buffer overflow. While nobody in their
1682 right mind would setup Exim to query an untrusted spamd server, an
1683 attacker that gains access to a server running spamd could potentially
1684 exploit this vulnerability to run arbitrary code as the Exim user.
1686 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
1687 $primary_hostname instead of what libspf2 thinks the hosts name is.
1689 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
1690 a directory entry by the name of the lookup key. Previously, if a
1691 symlink pointed to a non-existing file or a file in a directory that
1692 Exim lacked permissions to read, a lookup for a key matching that
1693 symlink would fail. Now it is enough that a matching directory entry
1694 exists, symlink or not. (Bugzilla 503.)
1696 PH/03 The body_linecount and body_zerocount variables are now exported in the
1699 PH/04 Added the $dnslist_matched variable.
1701 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
1702 This means they are set thereafter only if the connection becomes
1705 PH/06 Added the client_condition to authenticators so that some can be skipped
1706 by clients under certain conditions.
1708 PH/07 The error message for a badly-placed control=no_multiline_responses left
1709 "_responses" off the end of the name.
1711 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
1713 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
1714 (without spaces) instead of just copying the configuration text.
1716 PH/10 Added the /noupdate option to the ratelimit ACL condition.
1718 PH/11 Added $max_received_linelength.
1720 PH/12 Added +ignore_defer and +include_defer to host lists.
1722 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
1723 way in which PCRE > 7.0 is built.
1725 PH/14 Implemented queue_only_load_latch.
1727 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
1728 MAIL command. The effect was to mangle the value on 64-bit systems.
1730 PH/16 Another patch from the Sieve maintainer.
1732 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
1734 PH/18 If a system quota error occurred while trying to create the file for
1735 a maildir delivery, the message "Mailbox is full" was not appended to the
1736 bounce if the delivery eventually timed out. Change 4.67/27 below applied
1737 only to a quota excession during the actual writing of the file.
1739 PH/19 It seems that peer DN values may contain newlines (and other non-printing
1740 characters?) which causes problems in log lines. The DN values are now
1741 passed through string_printing() before being added to log lines.
1743 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
1744 and InterBase are left for another time.)
1746 PH/21 Added message_body_newlines option.
1748 PH/22 Guard against possible overflow in moan_check_errorcopy().
1750 PH/23 POSIX allows open() to be a macro; guard against that.
1752 PH/24 If the recipient of an error message contained an @ in the local part
1753 (suitably quoted, of course), incorrect values were put in $domain and
1754 $local_part during the evaluation of errors_copy.
1760 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
1761 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
1764 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
1765 issue a MAIL command.
1767 PH/02 In an ACL statement such as
1769 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
1771 if a client was not listed at all, or was listed with a value other than
1772 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
1773 the condition was not true (as it should be), so access was not denied.
1774 The bug was that the ! inversion was incorrectly passed on to the second
1775 item. This has been fixed.
1777 PH/03 Added additional dnslists conditions == and =& which are different from
1778 = and & when the dns lookup returns more than one IP address.
1780 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
1781 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
1783 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
1784 FSYNC, which compiles an option called disable_fsync that allows for
1785 bypassing fsync(). The documentation is heavily laced with warnings.
1787 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
1789 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
1790 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
1791 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
1792 including adding "make clean"; (3) Added -fPIC when compiling the test
1793 dynamically loaded module, to get rid of a warning.
1795 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
1796 message fails, move_frozen_messages = true and ignore_bounce_errors_after
1797 = 0s. The bug is otherwise harmless.
1799 PH/07 There was a bug in the dovecot authenticator such that the value of
1800 $auth1 could be overwritten, and so not correctly preserved, after a
1801 successful authentication. This usually meant that the value preserved by
1802 the server_setid option was incorrect.
1804 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
1806 PH/09 Installed PCRE release 7.0.
1808 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
1809 run for batched SMTP input. It is now run at the start of every message
1810 in the batch. While fixing this I discovered that the process information
1811 (output by running exiwhat) was not always getting set for -bs and -bS
1812 input. This is fixed, and it now also says "batched" for BSMTP.
1814 PH/11 Added control=no_pipelining.
1816 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
1817 patch, slightly modified), and move the expansion of helo_data till after
1818 the connection is made in the smtp transport (so it can use these
1821 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
1823 PH/14 Added log_selector = +pid.
1825 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
1827 PH/16 Add ${if forany and ${if forall.
1829 PH/17 Added dsn_from option to vary the From: line in DSNs.
1831 PH/18 Flush SMTP output before performing a callout, unless control =
1832 no_callout_flush is set.
1834 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
1835 was true (the default) a successful delivery failed to delete the retry
1836 item, thus causing premature timeout of the address. The bug is now
1839 PH/20 Added hosts_avoid_pipelining to the smtp transport.
1841 PH/21 Long custom messages for fakedefer and fakereject are now split up
1842 into multiline reponses in the same way that messages for "deny" and
1843 other ACL rejections are.
1845 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
1846 with slight modification.
1848 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
1849 draft, changing the syntax and factoring some duplicate code".
1851 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
1852 for deliveries of the second and subsequent messages over the same SMTP
1855 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
1856 ${reduce, with only minor "tidies".
1858 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
1860 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
1861 expansion side effects.
1863 PH/27 When a message times out after an over-quota error from an Exim-imposed
1864 quota, the bounce message says "mailbox is full". This message was not
1865 being given when it was a system quota that was exceeded. It now should
1868 MH/03 Made $recipients available in local_scan(). local_scan() already has
1869 better access to the recipient list through recipients_list[], but
1870 $recipients can be useful in postmaster-provided expansion strings.
1872 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
1873 in the case of a MAIL command with additional options following the
1874 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
1875 were accidentally chopped off.
1877 PH/29 SMTP synchronization checks are implemented when a command is read -
1878 there is a check that no more input is waiting when there shouldn't be
1879 any. However, for some commands, a delay in an ACL can mean that it is
1880 some time before the response is written. In this time, more input might
1881 arrive, invalidly. So now there are extra checks after an ACL has run for
1882 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
1883 pipelining has not been advertised.
1885 PH/30 MH's patch to allow iscntrl() characters to be list separators.
1887 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
1888 returned in the SMTP response when smtp_return_error_details was false.
1889 This has been fixed.
1891 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
1892 instead of the C I/O that was originally supplied, because problems were
1893 reported on Solaris.
1895 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
1896 Exim which did not show up earlier: it was assuming that a call to
1897 SSL_CTX_set_info_callback() might give an error value. In fact, there is
1898 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
1899 was a macro that became an assignment, so it seemed to work. This has
1900 changed to a proper function call with a void return, hence the compile
1901 error. Exim's code has been fixed.
1903 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
1906 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
1908 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
1910 PH/37 If a message is not accepted after it has had an id assigned (e.g.
1911 because it turns out to be too big or there is a timeout) there is no
1912 "Completed" line in the log. When some messages of this type were
1913 selected by exigrep, they were listed as "not completed". Others were
1914 picked up by some special patterns. I have improved the selection
1915 criteria to be more general.
1917 PH/38 The host_find_failed option in the manualroute router can now be set
1918 to "ignore", to completely ignore a host whose IP address cannot be
1919 found. If all hosts are ignored, the behaviour is controlled by the new
1920 host_all_ignored option.
1922 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
1923 homing or because of multiple MX records with /mx) generated more than
1924 one IP address, and the following item turned out to be the local host,
1925 all the secondary addresses of the first item were incorrectly removed
1926 from the list, along with the local host and any following hosts (which
1927 is what is supposed to happen).
1929 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
1930 whoever called Exim into the -H file. In the case of the daemon it was
1931 behaving confusingly. When first started, it used values for whoever
1932 started the daemon, but after a SIGHUP it used the Exim user (because it
1933 calls itself on a restart). I have changed the code so that it now always
1936 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
1937 message are rejected with the same error (e.g. no authentication or bad
1938 sender address), and a DATA command is nevertheless sent (as can happen
1939 with PIPELINING or a stupid MUA), the error message that was given to the
1940 RCPT commands is included in the rejection of the DATA command. This is
1941 intended to be helpful for MUAs that show only the final error to their
1944 PH/42 Another patch from the Sieve maintainer.
1946 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
1947 Eximstats - Fixed some broken HTML links and added missing column headers
1949 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
1950 columns for Rejects, Temp Rejects, Ham, and Spam rows.
1952 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
1954 PH/43 Yet another patch from the Sieve maintainer.
1956 PH/44 I found a way to check for a TCP/IP connection going away before sending
1957 the response to the final '.' that terminates a message, but only in the
1958 case where the client has not sent further data following the '.'
1959 (unfortunately, this is allowed). However, in many cases there won't be
1960 any further data because there won't be any more messages to send. A call
1961 to select() can be used: if it shows that the input is "ready", there is
1962 either input waiting, or the socket has been closed. An attempt to read
1963 the next input character can distinguish the two cases. Previously, Exim
1964 would have sent an OK response which the client would never have see.
1965 This could lead to message repetition. This fix should cure that, at
1966 least in a lot of common cases.
1968 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
1969 advertised in response to EHLO.
1975 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
1976 fixed by 4.65/MH/01 (is this a record?) are fixed:
1978 (i) An empty string was always treated as zero by the numeric comparison
1979 operators. This behaviour has been restored.
1981 (ii) It is documented that the numeric comparison operators always treat
1982 their arguments as decimal numbers. This was broken in that numbers
1983 starting with 0 were being interpreted as octal.
1985 While fixing these problems I realized that there was another issue that
1986 hadn't been noticed. Values of message_size_limit (both the global option
1987 and the transport option) were treated as octal if they started with 0.
1988 The documentation was vague. These values are now always treated as
1989 decimal, and I will make that clear in the documentation.
1995 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
1996 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
1999 MH/01 Don't check that the operands of numeric comparison operators are
2000 integers when their expansion is in "skipping" mode (fixes bug
2001 introduced by 4.64-PH/07).
2003 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
2004 child addresses, Exim now panics and dies. Previously, because the count
2005 is held in a short int, deliveries were likely to be lost. As such a
2006 large number of recipients for a single message is ridiculous
2007 (performance will be very, very poor), I have chosen to impose a limit
2008 rather than extend the field.
2014 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
2015 leftover -K file (the existence of which was triggered by #402).
2016 While we were at it, introduced process PID as part of the -K
2017 filename. This should rule out race conditions when creating
2020 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
2021 processing considerably. Previous code took too long for large mails,
2022 triggering a timeout which in turn triggers #401.
2024 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
2025 in the DK code in transports.c. sendfile() is not really portable,
2026 hence the _LINUX specificness.
2028 TF/01 In the add_headers option to the mail command in an Exim filter,
2029 there was a bug that Exim would claim a syntax error in any
2030 header after the first one which had an odd number of characters
2033 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
2034 callout verification, Exim cached a "reject" for the entire domain. This
2035 is correct for most verifications, but it is not correct for a recipient
2036 verification with use_sender or use_postmaster set, because in that case
2037 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
2038 case of MAIL FROM:<> rejection from other early rejections (e.g.
2039 rejection of HELO). When verifying a recipient using a non-null MAIL
2040 address, the cache is ignored if it shows MAIL FROM:<> rejection.
2041 Whatever the result of the callout, the value of the domain cache is
2042 left unchanged (for any other kind of callout, getting as far as trying
2043 RCPT means that the domain itself is ok).
2045 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
2048 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
2049 manifest itself as EPIPE rather than ECONNECT. When tidying away a
2050 session, the daemon ignores ECONNECT errors and logs others; it now
2051 ignores EPIPE as well.
2053 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
2054 (quoted-printable decoding).
2056 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
2057 later the small subsequent patch to fix an introduced bug.
2059 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
2061 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
2063 PH/08 An error is now given if message_size_limit is specified negative.
2065 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
2066 to be given (somewhat) arbitrary names.
2068 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
2071 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
2072 miscellaneous code fixes
2074 PH/10 Added the log_reject_target ACL modifier to specify where to log
2077 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
2078 hostname. This is wrong, because it relates to the incoming message (and
2079 probably the interface on which it is arriving) and not to the outgoing
2080 callout (which could be using a different interface). This has been
2081 changed to use the value of the helo_data option from the smtp transport
2082 instead - this is what is used when a message is actually being sent. If
2083 there is no remote transport (possible with a router that sets up host
2084 addresses), $smtp_active_hostname is used.
2086 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
2087 tweaks were necessary in order to get it to work (see also 21 below):
2088 (a) The code assumed that strncpy() returns a negative number on buffer
2089 overflow, which isn't the case. Replaced with Exim's string_format()
2091 (b) There were several signed/unsigned issues. I just did the minimum
2092 hacking in of casts. There is scope for a larger refactoring.
2093 (c) The code used strcasecmp() which is not a standard C function.
2094 Replaced with Exim's strcmpic() function.
2095 (d) The code set only $1; it now sets $auth1 as well.
2096 (e) A simple test gave the error "authentication client didn't specify
2097 service in request". It would seem that Dovecot has changed its
2098 interface. Fortunately there's a specification; I followed it and
2099 changed what the client sends and it appears to be working now.
2101 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
2104 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
2105 address A is aliased to B and C, where B exists and C does not. Without
2106 -v the output is "A verified" because verification stops after a
2107 successful redirection if more than one address is generated. However,
2108 with -v the child addresses are also verified. Exim was outputting "A
2109 failed to verify" and then showing the successful verification for C,
2110 with its parentage. It now outputs "B failed to verify", showing B's
2111 parentage before showing the successful verification of C.
2113 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
2114 look up a TXT record in a specific list after matching in a combined
2117 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
2118 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
2119 they consult the DNS. I had assumed they would set it the way they
2120 wanted; and indeed my experiments on Linux seem to show that in some
2121 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
2122 To be on the safe side, however, I have now made the interface to
2123 host_find_byname() similar to host_find_bydns(), with an argument
2124 containing the DNS resolver options. The host_find_byname() function now
2125 sets these options at its start, just as host_find_bydns() does. The smtp
2126 transport options dns_qualify_single and dns_search_parents are passed to
2127 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
2128 of host_find_byname() use the default settings of RES_DEFNAMES
2129 (qualify_single) but not RES_DNSRCH (search_parents).
2131 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
2132 spool_read_header() do less string testing, by means of a preliminary
2133 switch on the second character of optional "-foo" lines. (This is
2134 overdue, caused by the large number of possibilities that now exist.
2135 Originally there were few.) While I was there, I also converted the
2136 str(n)cmp tests so they don't re-test the leading "-" and the first
2137 character, in the hope this might squeeze out yet more improvement.
2139 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
2140 flag allowing group syntax was set by the header_syntax check but not
2141 turned off, possible causing trouble later; (2) The flag was not being
2142 set at all for the header_verify test, causing "group"-style headers to
2143 be rejected. I have now set it in this case, and also caused header_
2144 verify to ignore an empty address taken from a group. While doing this, I
2145 came across some other cases where the code for allowing group syntax
2146 while scanning a header line wasn't quite right (mostly, not resetting
2147 the flag correctly in the right place). These bugs could have caused
2148 trouble for malformed header lines. I hope it is now all correct.
2150 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
2151 with the "reply" argument non-NULL. The code, however (which originally
2152 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
2153 but it didn't always do it. This confused somebody who was copying the
2154 code for some other use. I have removed all the tests.
2156 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
2157 feature that was used to support insecure browsers during the U.S. crypto
2158 embargo. It requires special client support, and Exim is probably the
2159 only MTA that supported it -- and would never use it because real RSA is
2160 always available. This code has been removed, because it had the bad
2161 effect of slowing Exim down by computing (never used) parameters for the
2162 RSA_EXPORT functionality.
2164 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
2165 authenticator to fail if there's a tab character in the incoming data
2166 (there should never be unless someone is messing about, as it's supposed
2167 to be base64-encoded). Also added, on Timo's advice, the "secured" option
2168 if the connection is using TLS or if the remote IP is the same as the
2169 local IP, and the "valid-client-cert option" if a client certificate has
2172 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
2173 authenticators. This can be used for authorization after authentication
2174 succeeds. (In the case of plaintext, it servers for both authentication
2177 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
2178 if any retry times were supplied.
2180 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
2181 connection, where there is no client IP address to check. In this
2182 situation, the verify now always succeeds.
2184 PH/25 Applied John Jetmore's -Mset patch.
2186 PH/26 Added -bem to be like -Mset, but loading a message from a file.
2188 PH/27 In a string expansion for a processed (not raw) header when multiple
2189 headers of the same name were present, leading whitespace was being
2190 removed from all of them, but trailing whitespace was being removed only
2191 from the last one. Now trailing whitespace is removed from each header
2192 before concatenation. Completely empty headers in a concatenation (as
2193 before) are ignored.
2195 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
2196 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
2198 PH/29 [Removed. This was a change that I later backed out, and forgot to
2199 correct the ChangeLog entry (that I had efficiently created) before
2200 committing the later change.]
2202 PH/30 Exim was sometimes attempting to deliver messages that had suffered
2203 address errors (4xx response to RCPT) over the same connection as other
2204 messages routed to the same hosts. Such deliveries are always "forced",
2205 so retry times are not inspected. This resulted in far too many retries
2206 for the affected addresses. The effect occurred only when there were more
2207 hosts than the hosts_max_try setting in the smtp transport when it had
2208 the 4xx errors. Those hosts that it had tried were not added to the list
2209 of hosts for which the message was waiting, so if all were tried, there
2210 was no problem. Two fixes have been applied:
2212 (i) If there are any address or message errors in an SMTP delivery, none
2213 of the hosts (tried or untried) are now added to the list of hosts
2214 for which the message is waiting, so the message should not be a
2215 candidate for sending over the same connection that was used for a
2216 successful delivery of some other message. This seems entirely
2217 reasonable: after all the message is NOT "waiting for some host".
2218 This is so "obvious" that I'm not sure why it wasn't done
2219 previously. Hope I haven't missed anything, but it can't do any
2220 harm, as the worst effect is to miss an optimization.
2222 (ii) If, despite (i), such a delivery is accidentally attempted, the
2223 routing retry time is respected, so at least it doesn't keep
2224 hammering the server.
2226 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
2227 in ${readsocket because some servers need this prod.
2229 PH/32 Added some extra debug output when updating a wait-xxx database.
2231 PH/33 The hint "could be header name not terminated by colon", which has been
2232 given for certain expansion errors for a long time, was not being given
2233 for the ${if def:h_colon_omitted{... case.
2235 PH/34 The spec says: "With one important exception, whenever a domain list is
2236 being scanned, $domain contains the subject domain." There was at least
2237 one case where this was not true.
2239 PH/35 The error "getsockname() failed: connection reset by peer" was being
2240 written to the panic log as well as the main log, but it isn't really
2241 panic-worthy as it just means the connection died rather early on. I have
2242 removed the panic log writing for the ECONNRESET error when getsockname()
2245 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
2246 runs only) independently of the message's sender address. This meant
2247 that, if the 4xx error was in fact related to the sender, a different
2248 message to the same recipient with a different sender could confuse
2249 things. In particualar, this can happen when sending to a greylisting
2250 server, but other circumstances could also provoke similar problems.
2251 I have changed the default so that the retry time for these errors is now
2252 based a combination of the sender and recipient addresses. This change
2253 can be overridden by setting address_retry_include_sender=false in the
2256 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
2257 remote server are returned as part of bounce messages. This was not
2258 happening for LMTP over a pipe (the lmtp transport), but now it is the
2259 same for both kinds of LMTP.
2261 PH/38 Despite being documented as not happening, Exim was rewriting addresses
2262 in header lines that were in fact CNAMEs. This is no longer the case.
2264 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
2265 and queue runs started by the daemon processed all messages. This has
2266 been fixed so that -R and -S can now usefully be given with -q<time>.
2268 PH/40 Import PCRE release 6.7 (fixes some bugs).
2270 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
2272 PH/42 Give an error if -q is specified more than once.
2274 PH/43 Renamed the variables $interface_address and $interface_port as
2275 $received_ip_address and $received_port, to make it clear that these
2276 values apply to message reception, and not to the outgoing interface when
2277 a message is delivered. (The old names remain recognized, of course.)
2279 PH/44 There was no timeout on the connect() call when using a Unix domain
2280 socket in the ${readsocket expansion. There now is.
2282 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
2283 be meaningful with "accept".
2285 SC/01 Eximstats V1.43
2286 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
2288 SC/02 Eximstats V1.44
2289 Use a glob alias rather than an array ref in the generated
2290 parser. This improves both readability and performance.
2292 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
2293 Collect SpamAssassin and rejection statistics.
2294 Don't display local sender or destination tables unless
2295 there is data to show.
2296 Added average volumes into the top table text output.
2298 SC/04 Eximstats V1.46
2299 Collect data on the number of addresses (recipients)
2300 as well as the number of messages.
2302 SC/05 Eximstats V1.47
2303 Added 'Message too big' to the list of mail rejection
2304 reasons (thanks to Marco Gaiarin).
2306 SC/06 Eximstats V1.48
2307 Mainlog lines which have GMT offsets and are too short to
2308 have a flag are now skipped.
2310 SC/07 Eximstats V1.49 (Alain Williams)
2311 Added the -emptyok flag.
2313 SC/08 Eximstats V1.50
2314 Fixes for obtaining the IP address from reject messages.
2316 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
2317 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
2318 whitesspace changes from 4.64-PH/27
2320 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
2323 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
2324 are found, allow negative numbers in numeric criteria)
2326 JJ/06 exipick.20061117.2, added new $message_body_missing variable
2328 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
2329 to match changes made in 4.64-PH/43
2331 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
2333 PH/47 Put in an explicit test for a DNS lookup of an address record where the
2334 "domain" is actually an IP address, and force a failure. This locks out
2335 those revolvers/nameservers that support "A-for-A" lookups, in
2336 contravention of the specifications.
2338 PH/48 When a host name was looked up from an IP address, and the subsequent
2339 forward lookup of the name timed out, the host name was left in
2340 $sender_host_name, contrary to the specification.
2342 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
2343 restricted to single-key lookups, Exim was not diagnosing an error if
2344 * or *@ was used with a query-style lookup.
2346 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
2348 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
2349 long ago, but noone interested enough thought of it. Let's just say that
2350 the "1.1" means that there are some new functions that weren't there at
2351 some point in the past.
2353 PH/51 Error processing for expansion failure of helo_data from an smtp
2354 transport during callout processing was broken.
2356 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
2357 tested/used via the -bh/-bhc/-bs options.
2359 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
2360 bug, fixed in subsequent PCRE releases).
2362 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
2363 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
2365 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
2371 SC/01 Use a glob alias rather than an array ref in eximstats generated
2372 parser. This improves both readability and performance.
2374 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
2375 Don't display local sender or destination tables in eximstats unless
2376 there is data to show.
2377 Added average volumes into the eximstats top table text output.
2379 SC/03 Collect data on the number of addresses (recipients) as well
2380 as the number of messages in eximstats.
2382 TF/01 Correct an error in the documentation for the redirect router. Exim
2383 does (usually) call initgroups() when daemonizing.
2385 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
2386 with consistent privilege compared to when running as a daemon.
2388 TF/03 Note in the spec that $authenticated_id is not set for local
2389 submissions from trusted users.
2391 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
2392 Thanks to Dean Brooks <dean@iglou.com> for the patch.
2394 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
2395 by adding some example configuration directives to the default
2396 configuration file. A little bit of work is required to uncomment the
2397 directives and define how usernames and passwords are checked, but
2398 there is now a framework to start from.
2400 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
2401 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
2402 without this. I don't know how relevant this is to other LDAP libraries.
2404 PH/02 Add the verb name to the "unknown ACL verb" error.
2406 PH/03 Magnus Holmgren's patch for filter_prepend_home.
2408 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
2410 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
2411 directory not expanded when it should be if an expanded home directory
2412 was set for the address (which is overridden by the transport).
2414 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
2417 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
2418 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
2419 because it is too late at that time, and has no effect.
2421 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
2422 security issue with \' (bugzilla #107). I could not use the
2423 PQescapeStringConn() function, because it needs a PGconn value as one of
2426 PH/08 When testing addresses using -bt, indicate those final addresses that
2427 are duplicates that would not cause an additional delivery. At least one
2428 person was confused, thinking that -bt output corresponded to deliveries.
2429 (Suppressing duplicates isn't a good idea as you lose the information
2430 about possibly different redirections that led to the duplicates.)
2432 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
2433 systems where poll() doesn't work, in particular OS X.
2435 PH/10 Added more information to debugging output for retry time not reached.
2437 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
2438 operations in malware.c.
2440 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
2443 PH/13 If write_rejectlog was set false when logging was sent to syslog with
2444 syslog_duplication set false, log lines that would normally be written
2445 both the the main log and to the reject log were not written to syslog at
2448 PH/14 In the default configuration, change the use of "message" in ACL warn
2449 statements to "add_header".
2451 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
2452 not followed by a command (e.g. "seen endif").
2454 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
2455 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
2458 PH/17 Added extra conditions to the default value of delay_warning_condition
2462 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
2463 { match{$h_precedence:}{(?i)bulk|list|junk} } \
2464 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
2467 The Auto-Submitted: and various List- headers are standardised, whereas I
2468 don't think Precedence: ever was.
2470 PH/18 Refactored debugging code in route_finduser() to show more information,
2471 in particular, the error code if getpwnam() issues one.
2473 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
2474 This is apparently needed in addition to the PH/07 change above to avoid
2475 any possible encoding problems.
2477 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
2478 but not after initializing Perl.
2480 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
2481 output them only if debugging. By default they are written stderr,
2482 apparently, which is not desirable.
2484 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
2487 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
2490 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
2492 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
2493 authenticated or an ident call has been made. Suppress the default
2494 values for $authenticated_id and $authenticated_sender (but permit -oMai
2495 and -oMas) when testing with -bh.
2497 PH/24 Re-jigged the order of the tests in the default configuration so that the
2498 tests for valid domains and recipients precede the DNS black list and CSA
2499 tests, on the grounds that those ones are more expensive.
2501 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
2502 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
2503 This bug exists in every version of Exim that I still have, right back to
2506 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
2507 However, an attempt to turn on case-sensitivity in a regex key by
2508 including (?-i) didn't work because the subject string was already
2509 lowercased, and the effects were non-intuitive. It turns out that a
2510 one-line patch can be used to allow (?-i) to work as expected.
2516 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
2517 other effects) broke the use of negated acl sub-conditions.
2519 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
2522 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
2523 "Deny" causes Exim to reject the incoming connection with a 554 error.
2524 Unfortunately, if there is a major crisis, such as a disk failure,
2525 tcp-wrappers gives "deny", whereas what one would like would be some
2526 kind of temporary error. A kludge has been added to help with this.
2527 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
2528 554 error is used if errno is still zero or contains ENOENT (which occurs
2529 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
2532 PH/03 Add -lutil to the default FreeBSD LIBS setting.
2534 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
2535 errors. Otherwise a message that provokes a temporary error (when other
2536 messages do not) can cause a whole host to time out.
2538 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
2539 addresses were routed directly to files or pipes from a redirect router.
2540 File deliveries just didn't batch; pipe deliveries might have suffered
2543 PH/06 A failure to get a lock for a hints database would erroneously always say
2544 "Failed to get write lock", even when it was really a read lock.
2546 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
2547 of 0600. This has been changed to use the value of the lockfile_mode
2548 option (which defaults to 0600).
2550 PH/08 Applied small patch from the Sieve maintainer.
2552 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
2553 folder from quota calculations, a direct delivery into this folder messed
2554 up the contents of the maildirsize file. This was because the regex was
2555 used only to exclude .Trash (or whatever) when the size of the mailbox
2556 was calculated. There was no check that a delivery was happening into an
2557 excluded directory. This bug has been fixed by ignoring all quota
2558 processing for deliveries into excluded directories.
2560 PH/10 Added the maildirfolder_create_regex option to appendfile.
2566 PH/01 The code for finding all the local interface addresses on a FreeBSD
2567 system running IPv6 was broken. This may well have applied to all BSD
2568 systems, as well as to others that have similar system calls. The broken
2569 code found IPv4 interfaces correctly, but gave incorrect values for the
2570 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
2571 that it would not match correctly against @[] and not recognize the IPv6
2574 PH/02 The ipliteral router was not recognizing addresses of the form user@
2575 [ipv6:....] because it didn't know about the "ipv6:" prefix.
2577 PH/03 Added disable_ipv6.
2579 PH/04 Changed $reply_address to use the raw form of the headers instead of the
2580 decoded form, because it is most often used to construct To: headers
2581 lines in autoreplies, and the decoded form may well be syntactically
2582 invalid. However, $reply_address has leading white space removed, and all
2583 newlines turned into spaces so that the autoreply transport does not
2586 PH/05 If group was specified without a user on a router, and no group or user
2587 was specified on a transport, the group from the router was ignored.
2589 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
2590 for visible compile-time settings that can be used to change these
2591 numbers, for those that want even more. Backwards compatibility with old
2592 spool files has been maintained. However, going back to a previous Exim
2593 release will lost any variables that are in spool files.
2595 PH/07 Two small changes when running in the test harness: increase delay when
2596 passing a TCP/IP connection to a new process, in case the original
2597 process has to generate a bounce, and remove special handling of
2598 127.0.0.2 (sic), which is no longer necessary.
2600 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
2601 be the same on different OS.
2603 PH/09 Moved a debug statement in filter processing to avoid a race problem when
2606 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
2607 whether --show-vars was specified or not
2609 JJ/02 exipick: Added support for new ACL variable spool format introduced
2612 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
2613 syntactically invalid From: or Reply-to: line, and a filter used this to
2614 generate an autoreply, and therefore failed to obtain an address for the
2615 autoreply, Exim could try to deliver to a non-existent relative file
2616 name, causing unrelated and misleading errors. What now happens is that
2617 it logs this as a hard delivery error, but does not attempt to create a
2620 PH/11 The exinext utility has a -C option for testing purposes, but although
2621 the given file was scanned by exinext itself; it wasn't being passed on
2622 when Exim was called.
2624 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
2625 an end-of-file indication when reading a command response.
2627 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
2628 compiled. In many other places in Exim, IPv6 addresses are always
2629 recognized, so I have changed this. It also means that IPv4 domain
2630 literals of the form [IPV4:n.n.n.n] are now always recognized.
2632 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
2633 used if the router is not running as root, for example, when verifying at
2634 ACL time, or when using -bh. The debugging output from this situation was
2635 non-existent - all you got was a failure to exec. I have made two
2638 (a) Failures to set uid/gid, the current directory, or a process leader
2639 in a subprocess such as that created by queryprogram now generate
2640 suitable debugging ouput when -d is set.
2642 (b) The queryprogram router detects when it is not running as root,
2643 outputs suitable debugging information if -d is set, and then runs
2644 the subprocess without attempting to change uid/gid.
2646 PH/15 Minor change to Makefile for building test_host (undocumented testing
2649 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
2650 additional section of a DNS packet that returns MX or SRV records.
2651 Instead, it always explicitly searches for A/AAAA records. This avoids
2652 major problems that occur when a DNS server includes only records of one
2653 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
2654 fixed another bug: if SRV records were looked up and the corresponding
2655 address records were *not* found in the additional section, the port
2656 values from the SRV records were lost.
2658 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
2659 using the correct key (the original address) when searching the retry
2660 rules in order to find which one to use for generating the retry hint.
2662 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
2663 adding the default one. Similarly, if it contains a Reply-To: header, the
2664 errors_reply_to option, if set, is not used.
2666 PH/19 When calculating a retry time, Exim used to measure the "time since
2667 failure" by looking at the "first failed" field in the retry record. Now
2668 it does not use this if it is later than than the arrival time of the
2669 message. Instead it uses the arrival time. This makes for better
2670 behaviour in cases where some deliveries succeed, thus re-setting the
2671 "first failed" field. An example is a quota failure for a huge message
2672 when small messages continue to be delivered. Without this change, the
2673 "time since failure" will always be short, possible causing more frequent
2674 delivery attempts for the huge message than are intended.
2675 [Note: This change was subsequently modified - see PH/04 for 4.62.]
2677 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
2678 $1, $2, $3) because the numerical variables can be reset during some
2679 expansion items (e.g. "match"), thereby losing the authentication data.
2681 PH/21 Make -bV show the size of off_t variables so that the test suite can
2682 decide whether to run tests for quotas > 2G.
2684 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
2685 mailbox_size, and mailbox_filecount in the appendfile transport. If a
2686 filecount value is greater than 2G or if a quota value is greater than 2G
2687 on a system where the size of off_t is not greater than 4, a panic error
2690 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
2691 never match. The debug and -bh output now contains an explicit error
2692 message indicating a malformed IPv4 address or mask.
2694 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
2695 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
2696 PH/23 above applies.
2698 PH/25 Do not write to syslog when running in the test harness. The only
2699 occasion when this arises is a failure to open the main or panic logs
2700 (for which there is an explicit test).
2702 PH/26 Added the /no_tell option to "control=freeze".
2704 PH/27 If a host name lookup failed very early in a connection, for example, if
2705 the IP address matched host_lookup and the reverse lookup yielded a name
2706 that did not have a forward lookup, an error message of the form "no IP
2707 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
2708 could be logged. Now it outputs the IP address instead of "NULL".
2710 PH/28 An enabling patch from MH: add new function child_open_exim2() which
2711 allows the sender and the authenticated sender to be set when
2712 submitting a message from within Exim. Since child_open_exim() is
2713 documented for local_scan(), the new function should be too.
2715 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
2716 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
2717 results in an empty string is now treated as unset.
2719 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
2721 PH/31 Added sender_verify_fail logging option.
2723 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
2724 needed by RFC 822 but not by RFC 2822 was commented out. I have now
2725 tidied the source and removed it altogether.
2727 PH/33 When a queue run was abandoned because the load average was too high, a
2728 log line was always written; now it is written only if the queue_run log
2729 selector is set. In addition, the log line for abandonment now contains
2730 information about the queue run such as the pid. This is always present
2731 in "start" and "stop" lines but was omitted from the "abandon" line.
2733 PH/34 Omit spaces between a header name and the colon in the error message that
2734 is given when verify = headers_syntax fails (if there are lots of them,
2735 the message gets confusing).
2737 PH/35 Change the default for dns_check_names_pattern to allow slashes within
2738 names, as there are now some PTR records that contain slashes. This check
2739 is only to protect against broken name servers that fall over on strange
2740 characters, so the fact that it applies to all lookups doesn't matter.
2742 PH/36 Now that the new test suite is complete, we can remove some of the
2743 special code in Exim that was needed for the old test suite. For example,
2744 sorting DNS records because real resolvers return them in an arbitrary
2745 order. The new test suite's fake resolver always returns records in the
2748 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
2749 bounces) except when queue_only is set, to avoid logging races between
2750 the different processes.
2752 PH/38 Panic-die if .include specifies a non-absolute path.
2754 PH/39 A tweak to the "H" retry rule from its user.
2756 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
2757 a label. They prevented compilation on older perls.
2759 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
2760 a warning to be raised on newish perls.
2762 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
2763 on queue. Changes to match documented behaviour of showing count of
2764 messages matching specified criteria.
2766 PH/40 Changed the default ident timeout from 30s to 5s.
2768 PH/41 Added support for the use of login_cap features, on those BSD systems
2769 that have them, for controlling the resources used by pipe deliveries.
2771 PH/42 The content-scanning code uses fopen() to create files in which to put
2772 message data. Previously it was not paying any attention to the mode of
2773 the files. Exim runs with umask(0) because the rest of the code creates
2774 files with open(), and sets the required mode explicitly. Thus, these
2775 files were ending up world-writeable. This was not a big issue, because,
2776 being within the spool directory, they were not world-accessible. I have
2777 created a function called modefopen, which takes an additional mode
2778 argument. It sets umask(777), creates the file, chmods it to the required
2779 mode, then resets the umask. All the relevant calls to fopen() in the
2780 content scanning code have been changed to use this function.
2782 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
2783 to 24 hours. This avoids potential overflow problems when processing G
2784 and H retry rules. I suspect nobody ever tinkers with this value.
2786 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
2788 PH/45 When the plaintext authenticator is running as a client, the server's
2789 challenges are checked to ensure they are valid base64 strings. By
2790 default, the authentication attempt is cancelled if an invalid string is
2791 received. Setting client_ignore_invalid_base64 true ignores these errors.
2792 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
2793 they are received. Thus, the responses can be made to depend on the
2794 challenges. If an invalid string is ignored, an empty string is placed in
2797 PH/46 Messages that are created by the autoreply transport now contains a
2798 References: header, in accordance with RFCs 2822 and 3834.
2800 PH/47 Added authenticated_sender_force to the smtp transport.
2802 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
2804 PH/49 Installed latest patch from the Sieve maintainer.
2806 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
2807 was also set, the appendfile transport was unnecessarily scanning a
2808 directory of message files (e.g. for maildir delivery) to find the count
2809 of files (along with the size), even though it did not need this
2810 information. It now does the scan only if it needs to find either the
2811 size of the count of files.
2813 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
2815 PH/52 Two bugs concerned with error handling when the smtp transport is
2818 (i) Exim was not creating retry information for temporary errors given
2819 for individual recipients after the DATA command when the smtp transport
2820 was used in LMTP mode. This meant that they could be retried too
2821 frequently, and not timed out correctly.
2823 (ii) Exim was setting the flag that allows error details to be returned
2824 for LMTP errors on RCPT commands, but not for LMTP errors for individual
2825 recipients that were returned after the DATA command.
2827 PH/53 This is related to PH/52, but is more general: for any failing address,
2828 when detailed error information was permitted to be returned to the
2829 sender, but the error was temporary, then after the final timeout, only
2830 "retry timeout exceeded" was returned. Now it returns the full error as
2831 well as "retry timeout exceeded".
2833 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
2834 do this, and (what is worse) MTAs that accept it.
2836 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
2837 will now be deprecated.
2839 PH/56 New os.c-cygwin from the Cygwin maintainer.
2841 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
2842 formats (previously only available in exim formats via -bpr, -bpru,
2843 and -bpra. Now also available in native and exiqgrep formats)
2845 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
2846 with very large, slow to parse queues
2848 JJ/08 exipick: added ! as generic prefix to negate any criteria format
2850 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
2852 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
2853 responses to authentication challenges, though it was showing the
2854 challenges; (ii) I've removed the CR characters from the debug output for
2857 PH/58 Allow for the insertion of a newline as well as a space when a string
2858 is turned into more than one encoded-word during RFC 2047 encoding. The
2859 Sieve code now uses this.
2861 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
2862 data_4xx, lost_connection, tls_required.
2864 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
2865 message was being sent as an SMTP response.
2867 PH/61 Add -l and -k options to exicyclog.
2869 PH/62 When verifying, if an address was redirected to one new address, so that
2870 verification continued, and the new address failed or deferred after
2871 having set something in $address_data, the value of $address_data was not
2872 passed back to the ACL. This was different to the case when no
2873 redirection occurred. The value is now passed back in both cases.
2875 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
2876 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
2877 use login_cap.h, so on its own it isn't the distinguishing feature. The
2878 new name refers directly to the setclassresources() function.
2880 PH/65 Added configuration files for NetBSD3.
2882 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
2884 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
2885 is preferred over IPv4.
2887 PH/68 The bounce_return_message and bounce_return_body options were not being
2888 honoured for bounces generated during the reception of non-SMTP messages.
2889 In particular, this applied to messages rejected by the ACL. This bug has
2890 been fixed. However, if bounce_return_message is true and bounce_return_
2891 body is false, the headers that are returned for a non-SMTP message
2892 include only those that have been read before the error was detected.
2893 (In the case of an ACL rejection, they have all been read.)
2895 PH/69 The HTML version of the specification is now built in a directory called
2896 spec_html instead of spec.html, because the latter looks like a path with
2897 a MIME-type, and this confuses some software.
2899 PH/70 Catch two compiler warnings in sieve.c.
2901 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
2902 function verify_get_ident() calls ip_connect() to connect a socket, but
2903 if the "connect()" function timed out, ip_connect() used to close the
2904 socket. However, verify_get_ident() also closes the socket later, and in
2905 between Exim writes to the log, which may get opened at this point. When
2906 the socket was closed in ip_connect(), the log could get the same file
2907 descriptor number as the socket. This naturally causes chaos. The fix is
2908 not to close the socket in ip_connect(); the socket should be closed by
2909 the function that creates it. There was only one place in the code where
2910 this was missing, in the iplookup router, which I don't think anybody now
2911 uses, but I've fixed it anyway.
2913 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
2914 well as to direct DNS lookups. Otherwise the handling of names in host
2915 lists is inconsistent and therefore confusing.
2921 PH/01 Two changes to the default runtime configuration:
2923 (1) Move the checks for relay_from_hosts and authenticated clients from
2924 after to before the (commented out) DNS black list checks.
2926 (2) Add control=submission to the relay_from_hosts and authenticated
2927 clients checks, on the grounds that messages accepted by these
2928 statements are most likely to be submissions.
2930 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
2932 (1) Generate an error if the third argument for the ${prvs expansion is
2935 (2) Treat a missing third argument of ${prvscheck as if it were an empty
2938 (3) Reset the variables that are obtained from the first argument of
2939 ${prvscheck and used in the second argument before leaving the code,
2940 because their memory is reclaimed, so using them afterwards may do
2943 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
2944 one (it's much easier than Tom thought :-).
2946 (5) Because of (4), we can now allow for the use of $prvscheck_result
2947 inside the third argument.
2949 PH/03 For some reason, the default setting of PATH when running a command from
2950 a pipe transport was just "/usr/bin". I have changed it to
2953 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
2954 anything to be listed in the output from -bV.
2956 PH/05 When a filter generated an autoreply, the entire To: header line was
2957 quoted in the delivery log line, like this:
2959 => >A.N.Other <ano@some.domain> <original@ddress> ...
2961 This has been changed so that it extracts the operative address. There
2962 may be more than one such address. If so, they are comma-separated, like
2965 => >ano@some.domain,ona@other.domain <original@ddress> ...
2967 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
2968 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
2969 not being looked up in the rDNS to get a host name, Exim was showing the
2970 IP address twice in Received: lines, even though the IP addresses were
2971 identical. For example:
2973 Received: from [1.2.3.4] (helo=[1.2.3.4])
2975 However, if the real host name was known, it was omitting the HELO data
2976 if it matched the actual IP address. This has been tidied up so that it
2977 doesn't show the same IP address twice.
2979 PH/07 When both +timestamp and +memory debugging was on, the value given by
2980 $tod_xxx expansions could be wrong, because the tod_stamp() function was
2981 called by the debug printing, thereby overwriting the timestamp buffer.
2982 Debugging no longer uses the tod_stamp() function when +timestamp is set.
2984 PH/08 When the original message was included in an autoreply transport, it
2985 always said "this is a copy of the message, including all the headers",
2986 even if body_only or headers_only was set. It now gives an appropriate
2989 PH/09 Applied a patch from the Sieve maintainer which:
2991 o fixes some comments
2992 o adds the (disabled) notify extension core
2993 o adds some debug output for the result of if/elsif tests
2994 o points to the current vacation draft in the documentation
2995 and documents the missing references header update
2999 o fixes a bug in processing the envelope test (when testing
3000 multiple envelope elements, the last element determinted the
3003 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
3004 Electronic Mail") by including:
3006 Auto-submitted: auto-generated
3008 in the messages that it generates (bounce messages and others, such as
3009 warnings). In the case of bounce messages for non-SMTP mesages, there was
3010 also a typo: it was using "Auto_submitted" (underscore instead of
3011 hyphen). Since every message generated by Exim is necessarily in response
3012 to another message, thes have all been changed to:
3014 Auto-Submitted: auto-replied
3016 in accordance with these statements in the RFC:
3018 The auto-replied keyword:
3020 - SHOULD be used on messages sent in direct response to another
3021 message by an automatic process,
3023 - MUST NOT be used on manually-generated messages,
3025 - MAY be used on Delivery Status Notifications (DSNs) and Message
3026 Disposition Notifications (MDNs),
3028 - MUST NOT be used on messages generated by automatic or periodic
3029 processes, except for messages which are automatic responses to
3032 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
3033 to the default Received: header definition.
3035 PH/12 Added log selector acl_warn_skipped (default on).
3037 PH/13 After a successful wildlsearch lookup, discard the values of numeric
3038 variables because (a) they are in the wrong storage pool and (b) even if
3039 they were copied, it wouldn't work properly because of the caching.
3041 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
3042 checking when decoding. Apparently there are clients that generate
3043 overlong encoded strings. Why am I not surprised?
3045 PH/15 If the first argument of "${if match_address" was not empty, but did not
3046 contain an "@" character, Exim crashed. Now it writes a panic log message
3047 and treats the condition as false.
3049 PH/16 In autoreply, treat an empty string for "once" the same as unset.
3051 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
3052 extension "envelope-auth". The code is finished and in agreement with
3053 other implementations, but there is no documentation so far and in fact,
3054 nobody wrote the draft yet. This extension is currently #undef'ed, thus
3055 not changing the active code.
3057 Print executed "if" and "elsif" statements when debugging is used. This
3058 helps a great deal to understand what a filter does.
3060 Document more things not specified clearly in RFC3028. I had all this
3061 sorted out, when out of a sudden new issues came to my mind. Oops."
3063 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
3066 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
3067 canonical form (as documented). However, after a host name lookup from
3068 the IP address, check_host() was doing a simple string comparison with
3069 addresses acquired from the DNS when checking that the found name did
3070 have the original IP as one of its addresses. Since any found IPv6
3071 addresses are likely to be in abbreviated form, the comparison could
3072 fail. Luckily, there already exists a function for doing the comparison
3073 by converting both addresses to binary, so now that is used instead of
3074 the text comparison.
3076 PH/20 There was another similar case to PH/19, when a complete host name was
3077 given in a host list; looking up its IP address could give an abbreviated
3078 form, whereas the current host's name might or might not be abbreviated.
3079 The same fix has been applied.
3085 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
3086 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
3089 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
3090 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
3092 PH/03 Typo: missing ".o" in src/pcre/Makefile.
3094 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
3095 header line, restrict the check to what is listed in RFCs 2369 and 2929.
3096 Also, for "Auto-Submitted", treat anything other than "no" as
3097 non-personal, in accordance with RFC 3834. (Previously it treated
3098 anything starting "auto-" as non-personal.)
3100 TF/01 The control=submission/name=... option had a problem with syntax
3101 errors if the name included a slash character. The /name= option
3102 now slurps the rest of the string, so it can include any characters
3103 but it must come last in the list of options (after /sender_retain
3106 PH/05 Some modifications to the interface to the fake nameserver for the new
3114 TK/01 Added the "success_on_redirect" address verification option. See
3115 NewStuff for rationale and an example.
3117 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
3119 PH/02 Patch to exigrep to allow it to work on syslog lines.
3121 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
3122 fread() to skip over the body file's header line, because in Cygwin the
3123 header line is locked and is inaccessible.
3125 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
3126 co-exist for some time) to make it clear that it is the Exim ID that is
3127 referenced, not the Message-ID: header line.
3129 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
3130 string_format() function, because snprintf() does not exist on all
3133 PH/06 The use of forbid_filter_existstest now also locks out the use of the
3134 ${stat: expansion item.
3136 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
3137 protocol synchronization error", to keep the pedants happy.
3139 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
3140 well as for IRIX systems, when gcc is being used. See the host.c source
3143 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
3145 PH/10 Named domain lists were not working if used in a queue_smtp_domains
3148 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
3149 transport and to the smtp transport in LMTP mode.
3151 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
3153 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
3154 run a filter in a subprocess. This could lead to confusion in subsequent
3155 lookups in the parent process. There should also be a search_tidyup() at
3156 the end of the subprocess.
3158 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
3159 only if the host matched helo_try_verify_hosts, which caused the
3160 verification to occur when the EHLO/HELO command was issued. The ACL just
3161 tested the remembered result. Now, if a previous verification attempt has
3162 not happened, "verify = helo" does it there and then.
3164 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
3166 TK/03 Fix log output including CR from clamd.
3168 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
3169 exist provoked a memory error which could cause a segfault.
3171 PH/15 Installed PCRE 6.2
3173 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
3175 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
3176 of the problem. Specifically, suggested +O2 rather than +O1 for the
3179 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
3181 PH/20 If a delivery was routed to a non-standard port by means of an SRV
3182 record, the port was not correctly logged when the outgoing_port log
3183 selector was set (it logged the transort's default port).
3185 PH/21 Added support for host-specific ports to manualroute, queryprogram,
3186 fallback_hosts, and "hosts" in the smtp transport.
3188 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
3189 host errors such as "Connection refused".
3191 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
3192 authentication with radiusclient 0.4.9:
3194 - Error returned from rc_read_config was caught wrongly
3195 - Username/password not passed on to radius server due to wrong length.
3197 The presumption is that some radiusclient API changes for 4.51/PH/17
3198 were not taken care of correctly. The code is still untested by me (my
3199 Linux distribution still has 0.3.2 of radiusclient), but it was
3200 contributed by a Radius user.
3202 PH/24 When doing a callout, the value of $domain wasn't set correctly when
3203 expanding the "port" option of the smtp transport.
3205 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
3206 while reading a MIME header. Thanks to Tom Hughes for a patch.
3208 PH/24 Include config.h inside local_scan.h so that configuration settings are
3211 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
3212 that in an ACL for RCPT (for example), you can examine exactly what was
3215 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
3216 commands, but it was not correctly comparing the address with the actual
3217 client host address. Thus, it would show the EHLO address in Received:
3218 header lines when this was not necessary.
3220 PH/27 Added the % operator to ${eval:}.
3222 PH/28 Exim tries to create and chdir to its spool directory when it starts;
3223 it should be ignoring failures (because with -C, for example, it has lost
3224 privilege). It wasn't ignoring creation failures other than "already
3227 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
3230 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
3231 because an input line was too long, either on its own, or by virtue of
3232 too many continuations, the temporary file was not being removed, and the
3233 return code was incorrect.
3235 PH/31 Missing "BOOL" in function definition in filtertest.c.
3237 PH/32 Applied Sieve patches from the maintainer.
3239 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
3241 PH/33 Added "verify = not_blind".
3243 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
3244 Local/Makefile (with some defaults set). These are used in built scripts
3245 such as exicyclog, but they have never been used in the exim_install
3246 script (though there are many overriding facilities there). I have
3247 arranged that the exim_install script now takes note of these two
3250 PH/35 Installed configuration files for Dragonfly.
3252 PH/36 When a locally submitted message by a trusted user did not contain a
3253 From: header, and the sender address was obtained from -f or from an SMTP
3254 MAIL command, and the trusted user did not use -F to supply a sender
3255 name, $originator_name was incorrectly used when constructing a From:
3256 header. Furthermore, $originator_name was used for submission mode
3257 messages from external hosts without From: headers in a similar way,
3258 which is clearly wrong.
3260 PH/37 Added control=suppress_local_fixups.
3262 PH/38 When log_selector = +received_sender was set, and the addition of the
3263 sender made the log line's construction buffer exactly full, or one byte
3264 less than full, an overflow happened when the terminating "\n" was
3267 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
3268 when the result of a list match is failure because a DNS lookup failed.
3270 PH/40 RM_COMMAND is now used in the building process.
3272 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
3273 the "build-* directories that it finds.
3275 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
3276 address in a domain literal was a prefix of an interface address.
3278 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
3279 when verifying a sender address, unless rewrite_headers is false.
3281 PH/44 Wrote a long comment about why errors_to addresses are verified as
3282 recipients, not senders.
3284 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
3285 the ratelimit ACL was added.
3287 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
3289 PH/46 Added extra information about PostgreSQL errors to the error string.
3291 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
3292 avoiding the need to install special zones in a real server. This is
3293 backwards compatible; if it can't find the fake resolver, it drops back.
3294 Thus, both old and new test suites can be run.
3296 TF/02 Added util/ratelimit.pl
3298 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
3299 clock is set back in time.
3301 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
3302 Candler <B.Candler@pobox.com>.
3304 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
3305 OK for addresses that are the result of redirections.
3307 PH/48 A number of further additions for the benefit of the new test suite,
3308 including a fake gethostbyname() that interfaces to the fake DNS resolver
3311 TF/06 The fix for widen_domains has also been applied to qualify_single and
3312 search_parents which are the other dnslookup options that can cause
3315 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
3318 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
3320 TF/07 Exim produced the error message "an SRV record indicated no SMTP
3321 service" if it encountered an MX record with an empty target hostname.
3322 The message is now "an MX or SRV record indicated no SMTP service".
3324 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
3325 if the DNS of the sending site is misconfigured. This is quite a
3326 common situation. This change restores the behaviour of treating a
3327 helo verification defer as a failure.
3329 PH/51 If self=fail was set on a router, the bounce message did not include the
3330 actual error message.
3336 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
3338 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
3339 command itself ended in error, the underlying message about the transport
3340 filter timeout was being overwritten with the pipe command error. Now the
3341 underlying error message should be appended to the second error message.
3343 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
3345 PH/02 Reduce the amount of output that "make" produces by default. Full output
3346 can still be requested.
3348 PH/03 The warning log line about a condition test deferring for a "warn" verb
3349 was being output only once per connection, rather than after each
3350 occurrence (because it was using the same function as for successful
3351 "warn" verbs). This seems wrong, so I have changed it.
3353 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
3354 it should not have, which might have caused a crash in the right
3355 circumstances, but probably never did.
3357 PH/04 Installed a modified version of Tony Finch's patch to make submission
3358 mode fix the return path as well as the Sender: header line, and to
3359 add a /name= option so that you can make the user's friendly name appear
3362 TF/03 Added the control = fakedefer ACL modifier.
3364 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
3365 Mark Lowes for thorough testing.
3367 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
3369 TK/03 Merged latest SRS patch from Miles Wilton.
3371 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
3372 with the definition in sysexits.h (which is #included earlier).
3373 Fortunately, Exim does not actually use EX_OK. The code used to try to
3374 preserve the sysexits.h value, by assumimg that macro definitions were
3375 scanned for macro replacements. I have been disabused of this notion,
3376 so now the code just undefines EX_OK before #including unistd.h.
3378 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
3379 in the smtp transport. When a block could not be written in a single
3380 write() function, the timeout was being re-applied to each part-write.
3381 This seems wrong - if the receiver was accepting one byte at a time it
3382 would take for ever. The timeout is now adjusted when this happens. It
3383 doesn't have to be particularly precise.
3385 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
3386 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
3388 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
3389 without a domain if the check to <postmaster@domain> fails.
3391 SC/01 Eximstats: added -xls and the ability to specify output files
3392 (patch written by Frank Heydlauf).
3394 SC/02 Eximstats: use FileHandles for outputing results.
3396 SC/03 Eximstats: allow any combination of xls, txt, and html output.
3398 SC/04 Eximstats: fixed display of large numbers with -nvr option
3400 SC/05 Eximstats: fixed merging of reports with empty tables.
3402 SC/06 Eximstats: added the -include_original_destination flag
3404 SC/07 Eximstats: removed tabs and trailing whitespace.
3406 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
3408 TK/06 MBOX spool code: Add real "From " MBOX separator line
3409 so the .eml file is really in mbox format (even though
3410 most programs do not really care). Patch from Alex Miller.
3412 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
3413 The latter is generated from $received_to and is only set if the
3414 message has one envelope recipient. SA can use these headers,
3415 obviously out-of-the-box. Patch from Alex Miller.
3417 PH/08 The ${def test on a variable was returning false if the variable's
3418 value was "0", contrary to what the specification has always said!
3419 The result should be true unless the variable is empty.
3421 PH/09 The syntax error of a character other than { following "${if
3422 def:variable_name" (after optional whitespace) was not being diagnosed.
3423 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
3424 accidental colon was present, for example, could give incorrect results.
3426 PH/10 Tidied the code in a number of places where the st_size field of a stat()
3427 result is used (not including appendfile, where other changes are about
3430 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
3431 This involved changing a lot of size variables from int to off_t. It
3432 should work with maildirs and everything.
3434 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
3435 spamd dying while we are connected to it.
3437 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
3440 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
3441 function for BDB 4.3.
3443 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
3445 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
3446 the sources for PCRE have been re-arranged and more files are now
3449 PH/15 The code I had for printing potentially long long variables in PH/11
3450 above was not the best (it lost precision). The length of off_t variables
3451 is now inspected at build time, and an appropriate printing format (%ld
3452 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
3453 to be "long long int" or "long int". This is needed for the internal
3454 formatting function string_vformat().
3456 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
3457 the configuration file to be ":syslog", then the script "guesses" where
3458 the logs files are, rather than using the compiled in default. In our
3459 case the guess is not the same as the compiled default, so the script
3460 suddenly stopped working when I started to use syslog. The patch checks
3461 to see if log_file_path is "". If so, it attempts to read it from exim
3462 with no configuration file to get the compiled in version, before it
3463 falls back to the previous guessing code."
3465 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
3466 implementing BATV in an Exim configuration. See NewStuff for the gory
3469 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
3470 Makefile that are specific to HP-UX.
3472 PH/18 If the "use_postmaster" option was set for a recipient callout together
3473 with the "random" option, the postmaster address was used as the MAIL
3474 FROM address for the random test, but not for the subsequent recipient
3475 test. It is now used for both.
3477 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
3478 patch removes a few documentation additions to RFC 3028, because the
3479 latest draft now contains them. It adds the new en;ascii-case comparator
3480 and a new error check for 8bit text in MIME parts. Comparator and
3481 require names are now matched exactly. I enabled the subaddress
3482 extension, but it is not well tested yet (read: it works for me)."
3484 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
3485 rework some of the code of TK/09 above to avoid the hardwired use of
3486 "%lld" and "long long". Replaced the call to snprintf() with a call to
3489 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
3490 records point to non-existent hosts", "retry timeout exceeded", and
3491 "retry time not reached for any host after a long failure period".
3493 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
3494 experimental DomainKeys support:
3496 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
3497 (2) On an error such as an illegally used "control", the wrong name for
3498 the control was given.
3500 These problems did NOT occur unless DomainKeys support was compiled.
3502 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
3504 PH/24 Added ${if match_ip condition.
3506 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
3507 databases so that it will be absolutely obvious if a crash occurs in the
3508 DB library. This is a regular occurrence (often caused by mis-matched
3511 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
3512 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
3513 picked up on a user's system that detects such things. There doesn't seem
3514 to be a gcc warning option for this - only an attribute that has to be
3515 put on the function's prototype. It seems that in Fedora Core 4 they have
3516 set this on a number of new functions. No doubt there will be more in due
3519 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
3520 specify a transport. However, if an address that was verified by such a
3521 router was the subject of a callout, Exim crashed because it tried to
3522 read the rcpt_include_affixes from the non-existent transport. Now it
3523 just assumes that the setting of that option is false. This bug was
3524 introduced by 4.51/PH/31.
3526 PH/28 Changed -d+all to exclude +memory, because that information is very
3527 rarely of interest, but it makes the output a lot bigger. People tend to
3528 do -d+all out of habit.
3530 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
3531 code in os-type was giving problems when libc.so lives in lib64, like on
3534 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
3535 aren't the modern standard, and it seems that some systems' include files
3536 don't always have them. Exim was already checking for some of the newer
3537 ones like T_AAAA, and defining it itself. I've added checks for all the
3538 record types that Exim uses.
3540 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
3541 not automatically generating a new one, as it is supposed to. This
3542 prevented TLS from working. If the file did exist, but contained invalid
3543 data, a new version was generated, as expected. It was only the case of a
3544 non-existent file that was broken.
3546 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
3547 with a change in libdomainkeys > 0.64.
3549 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
3550 from DNS. If the selector record carries the flag, it now has
3551 precedence over the domain-wide flag.
3553 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
3555 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
3556 the use of a port name that isn't defined in /etc/services) occurred, the
3557 message was deferred as in a normal delivery, and thus remained on the
3558 spool, instead of being failed because of the mua_wrapper setting. This
3559 is now fixed, and I tidied up some of the mua_wrapper messages at the
3562 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
3563 the messages in a hash of arrays rather than using individual hashes.
3564 This is a bit cleaner and results in dramatic memory savings, albeit
3565 at a slight CPU cost.
3567 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
3568 as requested by Marc Sherman.
3570 SC/10 Eximstats: added histograms for user specified patterns as requested
3573 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
3575 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
3576 fopen() in the content-scanning modules that did not already have it.
3582 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
3583 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
3585 TK/02 Fix ACL "control" statement not being available in MIME ACL.
3587 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
3589 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
3590 to test Sieve filters that use "vacation".
3592 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
3593 that changes the way the GnuTLS parameters are stored in the cache file.
3594 The new format can be generated externally. For backward compatibility,
3595 if the data in the cache doesn't make sense, Exim assumes it has read an
3596 old-format file, and it generates new data and writes a new file. This
3597 means that you can't go back to an older release without removing the
3600 PH/03 A redirect router that has both "unseen" and "one_time" set does not
3601 work if there are any delivery delays because "one_time" forces the
3602 parent to be marked "delivered", so its unseen clone is never tried
3603 again. For this reason, Exim now forbids the simultaneous setting of
3606 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
3607 redirected to themselves ("homonym" addresses). Read the long ChangeLog
3608 entry if you want to know the details. The fix, however, neglected to
3609 consider the case when local delivery batching is involved. The test for
3610 "previously delivered" was not happening when checking to see if an
3611 address could be batched with a previous (undelivered) one; under
3612 certain circumstances this could lead to multiple deliveries to the same
3615 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
3616 in its include files, and this causes problems building Exim.
3618 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
3619 header_syntax) but Exim was just ignoring anything given after a slash.
3620 In particular, this caused confusion with an attempt to use "verify =
3621 reverse_host_lookup/defer_ok". An error is now given when options are
3622 supplied for verify items that do not have them. (Maybe reverse_host_
3623 lookup should have a defer_ok option, but that's a different point.)
3625 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
3626 defined by RFC 821) to 2048, because there were problems with some AUTH
3627 commands, and RFC 1869 says the size should be increased for extended
3628 SMTP commands that take arguments.
3630 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
3633 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
3634 "unknown" error; now it says that the functionality isn't in the binary.
3636 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
3637 an address' error message when a string expansion fails (syntax or
3638 whatever). Otherwise the password may appear in the log. Following change
3639 PH/42 below, there is no longer a chance of it appearing in a bounce
3642 PH/11 Installed exipick version 20050225.0 from John Jetmore.
3644 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
3645 of its addresses was ever tried. (Bugzilla bug #2.)
3647 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
3648 the result incorrectly in the debug output. (It correctly added a newline
3649 to what was transported.)
3651 TF/01 Added $received_time.
3653 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
3654 commented out examples of how to interface to a virus scanner and to
3655 SpamAssassin. Also added commented examples of av_scanner and
3656 spamd_address settings.
3658 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
3659 and controls are allowed in which ACLs. There were a couple of minor
3660 errors. Some of the entries in the conditions table (which is a table of
3661 where they are NOT allowed) were getting very unwieldy; rewrote them as a
3662 negation of where the condition IS allowed.
3664 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
3666 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
3667 header file does not have a version number, so I've had to invent a new
3668 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
3669 API. The code is untested by me (my Linux distribution still has 0.3.2 of
3670 radiusclient), but it was contributed by a Radius user.
3672 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
3673 files or directories, for OpenSSL.
3675 PH/19 When an Exim process that is running as root has to create an Exim log
3676 file, it does so in a subprocess that runs as exim:exim so as to get the
3677 ownership right at creation (otherwise, other Exim processes might see
3678 the file with the wrong ownership). There was no test for failure of this
3679 fork() call, which would lead to the process getting stuck as it waited
3680 for a non-existent subprocess. Forks do occasionally fail when resources
3681 run out. I reviewed all the other calls to fork(); they all seem to check
3684 PH/20 When checking for unexpected SMTP input at connect time (before writing
3685 the banner), Exim was not dealing correctly with a non-positive return
3686 from the read() function. If the client had disconnected by this time,
3687 the result was a log entry for a synchronization error with an empty
3688 string after "input=" when read() returned zero. If read() returned -1
3689 (an event I could not check), uninitialized data bytes were printed.
3690 There were reports of junk text (parts of files, etc) appearing after
3693 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
3695 PH/22 Added support for macro redefinition, and (re)definition in between
3696 driver and ACL definitions.
3698 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
3699 forgetting to use the resulting value; it was using the unexpanded value.
3701 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
3702 hadn't been configured. The fix is from Juergen Kreileder, who
3703 understands it better than I do:
3705 "Here's what I see happening with three configured cyrus_sasl
3706 authenticators configured (plain, login, cram-md5):
3708 On startup auth_cyrus_sasl_init() gets called for each of these.
3709 This means three calls to sasl_listmech() without a specified mech_list.
3710 => SASL tests which mechs of all available mechs actually work
3711 => three warnings about OTP not working
3712 => the returned list contains: plain, login, cram-md5, digest-md5, ...
3714 With the patch, sasl_listmech() also gets called three times. But now
3715 SASL's mech_list option is set to the server_mech specified in the the
3716 authenticator. Or in other words, the answer from sasl_listmech()
3717 gets limited to just the mech you're testing for (which is different
3719 => the return list contains just 'plain' or 'login', 'cram-md5' or
3720 nothing depending on the value of ob->server_mech.
3722 I've just tested the patch: Authentication still works fine,
3723 unavailable mechs specified in the exim configuration are still
3724 caught, and the auth.log warnings about OTP are gone."
3726 PH/25 When debugging is enabled, the contents of the command line are added
3727 to the debugging output, even when log_selector=+arguments is not
3730 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
3731 answer is "GNU", and only if the return is "GNU/something" is the answer
3734 PH/27 $acl_verify_message is now set immediately after the failure of a
3735 verification in an ACL, and so is available in subsequent modifiers. In
3736 particular, the message can be preserved by coding like this:
3738 warn !verify = sender
3739 set acl_m0 = $acl_verify_message
3741 Previously, $acl_verify_message was set only while expanding "message"
3742 and "log_message" when a very denied access.
3744 PH/28 Modified OS/os.c-Linux with
3746 -#ifndef OS_LOAD_AVERAGE
3747 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
3749 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
3750 nomenclature these days.)
3752 PH/29 Installed patch from the Sieve maintainer that adds the options
3753 sieve_useraddress and sieve_subaddress to the redirect router.
3755 PH/30 In these circumstances:
3756 . Two addresses routed to the same list of hosts;
3757 . First host does not offer TLS;
3758 . First host accepts first address;
3759 . First host gives temporary error to second address;
3760 . Second host offers TLS and a TLS session is established;
3761 . Second host accepts second address.
3762 Exim incorrectly logged both deliveries with the TLS parameters (cipher
3763 and peerdn, if requested) that were in fact used only for the second
3766 PH/31 When doing a callout as part of verifying an address, Exim was not paying
3767 attention to any local part prefix or suffix that was matched by the
3768 router that accepted the address. It now behaves in the same way as it
3769 does for delivery: the affixes are removed from the local part unless
3770 rcpt_include_affixes is set on the transport.
3772 PH/32 Add the sender address, as F=<...>, to the log line when logging a
3773 timeout during the DATA phase of an incoming message.
3775 PH/33 Sieve envelope tests were broken for match types other than :is. I have
3776 applied a patch sanctioned by the Sieve maintainer.
3778 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
3779 the uid or gid is negative. A case of a negative gid caused this to be
3780 noticed. The fix allows for either to be negative.
3782 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
3783 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
3786 PH/36 The change PH/12 above was broken. Fixed it.
3788 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
3789 the grounds that routing the same address twice would always produce the
3790 same answer. This might have been true once, but it is certainly no
3791 longer true now. Routing a child address may depend on the previous
3792 routing that produced that child. Some complicated redirection strategies
3793 went wrong when messages had multiple recipients, and made Exim's
3794 behaviour dependent on the order in which the addresses were given.
3796 I have moved the duplicate checking until after the routing is complete.
3797 Exim scans the addresses that are assigned to local and remote
3798 transports, and removes any duplicates. This means that more work will be
3799 done, as duplicates will always all be routed, but duplicates are
3800 presumably rare, so I don't expect this is of any significance.
3802 For deliveries to pipes, files, and autoreplies, the duplicate checking
3803 still happens during the routing process, since they are not going to be
3806 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
3807 It corrects a timeout issue with spamd. This is Ian's comment: "The
3808 background is that sometimes spamd either never reads data from a
3809 connection it has accepted, or it never writes response data. The exiscan
3810 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
3811 blindly assumes that writes won't block so it may never time out."
3813 PH/39 Allow G after quota size as well as K and M.
3815 PH/40 The value set for $authenticated_id in an authenticator may not contain
3816 binary zeroes or newlines because the value is written to log lines and
3817 to spool files. There was no check on this. Now the value is run through
3818 the string_printing() function so that such characters are converted to
3819 printable escape sequences.
3821 PH/41 $message_linecount is a new variable that contains the total number of
3822 lines in the message. Compare $body_linecount, which is the count for the
3825 PH/42 Exim no longer gives details of delivery errors for specific addresses in
3826 bounce and delay warning messages, except in certain special cases, which
3829 (a) An SMTP error message from a remote host;
3830 (b) A message specified in a :fail: redirection;
3831 (c) A message specified in a "fail" command in a system filter;
3832 (d) A message specified in a FAIL return from the queryprogram router;
3833 (e) A message specified by the cannot_route_message router option.
3835 In these cases only, Exim does include the error details in bounce and
3836 warning messages. There are also a few cases where bland messages such
3837 as "unrouteable address" or "local delivery error" are given.
3839 PH/43 $value is now also set for the "else" part of a ${run expansion.
3841 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
3842 being worked on, but at least Exim now implements the latest version to
3845 PH/45 In a pipe transport, although a timeout while waiting for the pipe
3846 process to complete was treated as a delivery failure, a timeout while
3847 writing the message to the pipe was logged, but erroneously treated as a
3848 successful delivery. Such timeouts include transport filter timeouts. For
3849 consistency with the overall process timeout, these timeouts are now
3850 treated as errors, giving rise to delivery failures by default. However,
3851 there is now a new Boolean option for the pipe transport called
3852 timeout_defer, which, if set TRUE, converts the failures into defers for
3853 both kinds of timeout. A transport filter timeout is now identified in
3856 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
3857 systems where "make" and "gmake" are different, calling "gmake" at top
3858 level broke things. I've arranged for the value of $(MAKE) to be passed
3859 from the Makefile to this script so that it can call the same version of
3863 A note about Exim versions 4.44 and 4.50
3864 ----------------------------------------
3866 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
3867 changes of various kinds. As a consequence, a big documentation update was
3868 needed. This delayed the release for rather longer than seemed good, especially
3869 in the light of a couple of (minor) security issues. Therefore, the changes
3870 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
3871 release. So 4.44 and 4.50 are in effect two different branches that both start
3874 I have left the 4.50 change log unchanged; it contains all the changes since
3875 4.43. The change log for 4.44 is below; many of its items are identical to
3876 those for 4.50. This seems to be the most sensible way to preserve the
3877 historical information.
3883 1. Minor wording change to the doc/README.SIEVE file.
3885 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
3886 computation of the current number of files was incorrect.
3888 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
3889 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
3892 4. Give more explanation in the error message when the command for a transport
3893 filter fails to execute.
3895 5. There are several places where Exim runs a non-Exim command in a
3896 subprocess. The SIGUSR1 signal should be disabled for these processes. This
3897 was being done only for the command run by the queryprogram router. It is
3898 now done for all such subprocesses. The other cases are: ${run, transport
3899 filters, and the commands run by the lmtp and pipe transports.
3901 6. Added CONFIGURE_GROUP build-time option.
3903 7. Some older OS have a limit of 256 on the maximum number of file
3904 descriptors. Exim was using setrlimit() to set 1000 as a large value
3905 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
3906 systems. I've change it so that if it can't get 1000, it tries for 256.
3908 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
3909 was an oversight, and furthermore, ever since the addition of extra
3910 controls (e.g. 4.43/32), the checks on when to allow different forms of
3911 "control" were broken. There should now be diagnostics for all cases when a
3912 control that does not make sense is encountered.
3914 9. Added the /retain_sender option to "control=submission".
3916 10. $recipients is now available in the predata ACL (oversight).
3918 11. Tidy the search cache before the fork to do a delivery from a message
3919 received from the command line. Otherwise the child will trigger a lookup
3920 failure and thereby defer the delivery if it tries to use (for example) a
3921 cached ldap connection that the parent has called unbind on.
3923 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
3924 of $address_data from the recipient verification was clobbered by the
3925 sender verification.
3927 13. The value of address_data from a sender verification is now available in
3928 $sender_address_data in subsequent conditions in the ACL statement.
3930 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
3932 15. Added a new option "connect=<time>" to callout options, to set a different
3935 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
3936 was its contents. (It was OK if the option was not defined at all.)
3938 17. A "Completed" log line is now written for messages that are removed from
3939 the spool by the -Mrm option.
3941 18. New variables $sender_verify_failure and $recipient_verify_failure contain
3942 information about exactly what failed.
3944 19. Added -dd to debug only the daemon process.
3946 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
3947 handles timeouts, both on the server side and network timeouts. Renamed the
3948 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
3950 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
3951 It is now set to "smtps".
3953 22. $host_address is now set to the target address during the checking of
3954 ignore_target_hosts.
3956 23. When checking ignore_target_hosts for an ipliteral router, no host name was
3957 being passed; this would have caused $sender_host_name to have been used if
3958 matching the list had actually called for a host name (not very likely,
3959 since this list is usually IP addresses). A host name is now passed as
3962 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
3963 code that specifies a non-restarting handler (typically sigaction() in
3964 modern systems) in an attempt to fix a rare and obscure crash bug.
3966 25. Narrowed the window for a race in the daemon that could cause it to ignore
3967 SIGCHLD signals. This is not a major problem, because they are used only to
3968 wake it up if nothing else does.
3970 26. A malformed maildirsize file could cause Exim to calculate negative values
3971 for the mailbox size or file count. Odd effects could occur as a result.
3972 The maildirsize information is now recalculated if the size or filecount
3975 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
3976 support for a long time. Removed HAVE_SYS_VFS_H.
3978 28. Installed the latest version of exipick from John Jetmore.
3980 29. In an address list, if the pattern was not a regular expression, an empty
3981 subject address (from a bounce message) matched only if the pattern was an
3982 empty string. Non-empty patterns were not even tested. This was the wrong
3983 because it is perfectly reasonable to use an empty address as part of a
3984 database query. An empty address is now tested by patterns that are
3985 lookups. However, all the other forms of pattern expect the subject to
3986 contain a local part and a domain, and therefore, for them, an empty
3987 address still always fails if the pattern is not itself empty.
3989 30. Exim went into a mad DNS loop when attempting to do a callout where the
3990 host was specified on an smtp transport, and looking it up yielded more
3991 than one IP address.
3993 31. Re-factored the code for checking spool and log partition space into a
3994 function that finds that data and another that does the check. The former
3995 is then used to implement four new variables: $spool_space, $log_space,
3996 $spool_inodes, and $log_inodes.
3998 32. The RFC2047 encoding function was originally intended for short strings
3999 such as real names; it was not keeping to the 75-character limit for
4000 encoded words that the RFC imposes. It now respects the limit, and
4001 generates multiple encoded words if necessary. To be on the safe side, I
4002 have increased the buffer size for the ${rfc2047: expansion operator from
4005 33. It is now permitted to omit both strings after an "if" condition; if the
4006 condition is true, the result is "true". As before, when the second string
4007 is omitted, a false condition yields an empty string. This makes it less
4008 cumbersome to write custom ACL and router conditions.
4010 34. Failure to deliver a bounce message always caused it to be frozen, even if
4011 there was an errors_to setting on the router. The errors_to setting is now
4014 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4015 canonical form (fully expanded) before being placed in
4016 $sender_host_address.
4018 36. The table in the code that translates DNS record types into text (T_A to
4019 "A" for instance) was missing entries for NS and CNAME. It is just possible
4020 that this could have caused confusion if both these types were looked up
4021 for the same domain, because the text type is used as part of Exim's
4022 per-process caching. But the chance of anyone hitting this buglet seems
4025 37. The dnsdb lookup has been extended in a number of ways.
4027 (1) There is a new type, "zns", which walks up the domain tree until it
4028 finds some nameserver records. It should be used with care.
4030 (2) There is a new type, "mxh", which is like "mx" except that it returns
4031 just the host names, not the priorities.
4033 (3) It is now possible to give a list of domains (or IP addresses) to be
4034 looked up. The behaviour when one of the lookups defers can be
4035 controlled by a keyword.
4037 (4) It is now possible to specify the separator character for use when
4038 multiple records are returned.
4040 38. The dnslists ACL condition has been extended: it is now possible to supply
4041 a list of IP addresses and/or domains to be looked up in a particular DNS
4044 39. Added log_selector=+queue_time_overall.
4046 40. When running the queue in the test harness, wait just a tad after forking a
4047 delivery process, to get repeatability of debugging output.
4049 41. Include certificate and key file names in error message when GnuTLS fails
4050 to set them up, because the GnuTLS error message doesn't include the name
4051 of the failing file when there is a problem reading it.
4053 42. Allow both -bf and -bF in the same test run.
4055 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
4057 44. The "Exiscan patch" is now merged into the mainline Exim source.
4059 45. Sometimes the final signoff response after QUIT could fail to get
4060 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4061 < 0 before doing a fflush(). This bug looks as though it goes back to the
4062 introduction of TLS in release 3.20, but "sometimes" must have been rare
4063 because the tests only now provoked it.
4065 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
4066 (this can affect the format of dates).
4068 47. exim_tidydb, when checking for the continued existence of a message for
4069 which it has found a message-specific retry record, was not finding
4070 messages that were in split spool directories. Consequently, it was
4071 deleting retry records that should have stayed in existence.
4073 48. Steve fixed some bugs in eximstats.
4075 49. The SPA authentication driver was not abandoning authentication and moving
4076 on to the next authenticator when an expansion was forced to fail,
4077 contradicting the general specification for all authenticators. Instead it
4078 was generating a temporary error. It now behaves as specified.
4080 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
4081 (the order specifies the preference for clients). The order is now AES256,
4082 AES128, 3DES, ARCFOUR128.
4084 51. Small patch to Sieve code - explicitly set From: when generating an
4087 52. Exim crashed if a remote delivery caused a very long error message to be
4088 recorded - for instance if somebody sent an entire SpamAssassin report back
4089 as a large number of 550 error lines. This bug was coincidentally fixed by
4090 increasing the size of one of Exim's internal buffers (big_buffer) that
4091 happened as part of the Exiscan merge. However, to be on the safe side, I
4092 have made the code more robust (and fixed the comments that describe what
4095 53. Now that there can be additional text after "Completed" in log lines (if
4096 the queue_time_overall log selector is set), a one-byte patch to exigrep
4097 was needed to allow it to recognize "Completed" as not the last thing in
4100 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
4101 patch that reportedly fixes this has been added. I am not expert enough to
4102 create a test for it. This is what the patch creator wrote:
4104 "I found a little strange behaviour of ldap code when working with
4105 Windows 2003 AD Domain, where users was placed in more than one
4106 Organization Units. When I tried to give exim partial DN, the exit code
4107 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
4108 But simultaneously result of request was absolutely normal ldap result,
4109 so I produce this patch..."
4111 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
4112 REFERENCE, so I have modified the code to exclude the patch when that macro
4115 55. Some experimental protocols are using DNS PTR records for new purposes. The
4116 keys for these records are domain names, not reversed IP addresses. The
4117 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4118 leaves it alone. Component reversal etc. now happens only for IP addresses.
4121 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
4123 57. Double the size of the debug message buffer (to 2048) so that more of very
4124 long debug lines gets shown.
4126 58. The exicyclog utility now does better if the number of log files to keep
4127 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
4129 59. Two changes related to the smtp_active_hostname option:
4131 (1) $smtp_active_hostname is now available as a variable.
4132 (2) The default for smtp_banner uses $smtp_active_hostname instead
4133 of $primary_hostname.
4135 60. The host_aton() function is supposed to be passed a string that is known
4136 to be a valid IP address. However, in the case of IPv6 addresses, it was
4137 not checking this. This is a hostage to fortune. Exim now panics and dies
4138 if the condition is not met. A case was found where this could be provoked
4139 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4140 components; fortuitously, this particular loophole had already been fixed
4141 by change 4.50/55 above.
4143 If there are any other similar loopholes, the new check in host_aton()
4144 itself should stop them being exploited. The report I received stated that
4145 data on the command line could provoke the exploit when Exim was running as
4146 exim, but did not say which command line option was involved. All I could
4147 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4148 running as the user.
4151 61. There was a buffer overflow vulnerability in the SPA authentication code
4152 (which came originally from the Samba project). I have added a test to the
4153 spa_base64_to_bits() function which I hope fixes it.
4156 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
4157 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
4159 63. The daemon start-up calls getloadavg() while still root for those OS that
4160 need the first call to be done as root, but it missed one case: when
4161 deliver_queue_load_max is set with deliver_drop_privilege. This is
4162 necessary for the benefit of the queue runner, because there is no re-exec
4163 when deliver_drop_privilege is set.
4165 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
4166 This has been fixed.
4168 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4169 was in use, was not putting the data itself into the right store pool;
4170 consequently, it could be overwritten for a subsequent message in the same
4171 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4174 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
4176 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
4177 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
4178 the function were treating the return as a boolean value, which happened to
4179 work because 0=false and not-0=true, but is not correct code.
4181 68. The host_aton() function was not handling scoped IPv6 addresses (those
4182 with, for example, "%eth0" on the end) correctly.
4184 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
4185 negated items (that is, ~something) in unsigned ints. Some compilers
4186 apparently mutter when there is no cast.
4188 70. If an address verification called from an ACL failed, and did not produce a
4189 user-specific message (i.e. there was only a "system" message), nothing was
4190 put in $acl_verify_message. In this situation, it now puts the system
4193 71. Change 4.23/11 added synchronization checking at the start of an SMTP
4194 session; change 4.31/43 added the unwanted input to the log line - except
4195 that it did not do this in the start of session case. It now does.
4197 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
4198 This is wrong and can cause the other end to generate a synchronization
4199 error if it is another Exim or anything else that does the synchronization
4200 check. A QUIT command is no longer sent after a timeout.
4202 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
4203 during host lookups.
4205 74. The defer_ok option of callout verification was not working if it was used
4206 when verifying addresses in header lines, that is, for this case:
4208 verify = header_sender/callout=defer_ok
4210 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
4211 those file descriptors could be used for SMTP connections. If anything
4212 wrote to stderr (the example that came up was "warn" in embedded Perl), it
4213 could be sent to the SMTP client, causing chaos. The daemon now opens
4214 stdin, stdout, and stderr to /dev/null when it puts itself into the
4217 76. Arrange for output from Perl's "warn" command to be written to Exim's main
4218 log by default. The user can override this with suitable Perl magic.
4220 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
4221 the log message when discard triggers, was not working for the DATA ACL or
4222 for the non-SMTP ACL.
4224 78. Error message wording change in sieve.c.
4226 79. If smtp_accept_max_per_host was set, the number of connections could be
4227 restricted to fewer than expected, because the daemon was trying to set up
4228 a new connection before checking whether the processes handling previous
4229 connections had finished. The check for completed processes is now done
4230 earlier. On busy systems, this bug wouldn't be noticed because something
4231 else would have woken the daemon, and it would have reaped the completed
4234 80. If a message was submitted locally by a user whose login name contained one
4235 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
4236 It caused a spool format error. I have fixed the spool reading code. A
4237 related problem was that the "from" clause in the Received: line became
4238 illegal because of the space(s). It is now covered by ${quote_local_part.
4240 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
4243 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
4245 83. Patch from Sieve maintainer to fix unterminated string problem in
4246 "vacation" handling.
4248 84. Some minor changes to the Linux configuration files to help with other
4249 OS variants using glibc.
4251 85. One more patch for Sieve to update vacation handling to latest spec.
4254 ----------------------------------------------------
4255 See the note above about the 4.44 and 4.50 releases.
4256 ----------------------------------------------------
4262 1. Change 4.43/35 introduced a bug that caused file counts to be
4263 incorrectly computed when quota_filecount was set in an appendfile
4266 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4267 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4270 3. Give more explanation in the error message when the command for a transport
4271 filter fails to execute.
4273 4. There are several places where Exim runs a non-Exim command in a
4274 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4275 was being done only for the command run by the queryprogram router. It is
4276 now done for all such subprocesses. The other cases are: ${run, transport
4277 filters, and the commands run by the lmtp and pipe transports.
4279 5. Some older OS have a limit of 256 on the maximum number of file
4280 descriptors. Exim was using setrlimit() to set 1000 as a large value
4281 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4282 systems. I've change it so that if it can't get 1000, it tries for 256.
4284 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4285 was an oversight, and furthermore, ever since the addition of extra
4286 controls (e.g. 4.43/32), the checks on when to allow different forms of
4287 "control" were broken. There should now be diagnostics for all cases when a
4288 control that does not make sense is encountered.
4290 7. $recipients is now available in the predata ACL (oversight).
4292 8. Tidy the search cache before the fork to do a delivery from a message
4293 received from the command line. Otherwise the child will trigger a lookup
4294 failure and thereby defer the delivery if it tries to use (for example) a
4295 cached ldap connection that the parent has called unbind on.
4297 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4298 of $address_data from the recipient verification was clobbered by the
4299 sender verification.
4301 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4302 was its contents. (It was OK if the option was not defined at all.)
4304 11. A "Completed" log line is now written for messages that are removed from
4305 the spool by the -Mrm option.
4307 12. $host_address is now set to the target address during the checking of
4308 ignore_target_hosts.
4310 13. When checking ignore_target_hosts for an ipliteral router, no host name was
4311 being passed; this would have caused $sender_host_name to have been used if
4312 matching the list had actually called for a host name (not very likely,
4313 since this list is usually IP addresses). A host name is now passed as
4316 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4317 code that specifies a non-restarting handler (typically sigaction() in
4318 modern systems) in an attempt to fix a rare and obscure crash bug.
4320 15. Narrowed the window for a race in the daemon that could cause it to ignore
4321 SIGCHLD signals. This is not a major problem, because they are used only to
4322 wake it up if nothing else does.
4324 16. A malformed maildirsize file could cause Exim to calculate negative values
4325 for the mailbox size or file count. Odd effects could occur as a result.
4326 The maildirsize information is now recalculated if the size or filecount
4329 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4330 support for a long time. Removed HAVE_SYS_VFS_H.
4332 18. Updated exipick to current release from John Jetmore.
4334 19. Allow an empty sender to be matched against a lookup in an address list.
4335 Previously the only cases considered were a regular expression, or an
4338 20. Exim went into a mad DNS lookup loop when doing a callout where the
4339 host was specified on the transport, if the DNS lookup yielded more than
4342 21. The RFC2047 encoding function was originally intended for short strings
4343 such as real names; it was not keeping to the 75-character limit for
4344 encoded words that the RFC imposes. It now respects the limit, and
4345 generates multiple encoded words if necessary. To be on the safe side, I
4346 have increased the buffer size for the ${rfc2047: expansion operator from
4349 22. Failure to deliver a bounce message always caused it to be frozen, even if
4350 there was an errors_to setting on the router. The errors_to setting is now
4353 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4354 canonical form (fully expanded) before being placed in
4355 $sender_host_address.
4357 24. Updated eximstats to version 1.33
4359 25. Include certificate and key file names in error message when GnuTLS fails
4360 to set them up, because the GnuTLS error message doesn't include the name
4361 of the failing file when there is a problem reading it.
4363 26. Expand error message when OpenSSL has problems setting up cert/key files.
4366 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
4367 (this can affect the format of dates).
4369 28. exim_tidydb, when checking for the continued existence of a message for
4370 which it has found a message-specific retry record, was not finding
4371 messages that were in split spool directories. Consequently, it was
4372 deleting retry records that should have stayed in existence.
4374 29. eximstats updated to version 1.35
4375 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
4376 1.35 - bugfix such that pie charts by volume are generated correctly
4378 30. The SPA authentication driver was not abandoning authentication and moving
4379 on to the next authenticator when an expansion was forced to fail,
4380 contradicting the general specification for all authenticators. Instead it
4381 was generating a temporary error. It now behaves as specified.
4383 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
4384 (the order specifies the preference for clients). The order is now AES256,
4385 AES128, 3DES, ARCFOUR128.
4387 31. Small patch to Sieve code - explicitly set From: when generating an
4390 32. Exim crashed if a remote delivery caused a very long error message to be
4391 recorded - for instance if somebody sent an entire SpamAssassin report back
4392 as a large number of 550 error lines. This bug was coincidentally fixed by
4393 increasing the size of one of Exim's internal buffers (big_buffer) that
4394 happened as part of the Exiscan merge. However, to be on the safe side, I
4395 have made the code more robust (and fixed the comments that describe what
4398 33. Some experimental protocols are using DNS PTR records for new purposes. The
4399 keys for these records are domain names, not reversed IP addresses. The
4400 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4401 leaves it alone. Component reversal etc. now happens only for IP addresses.
4404 34. The host_aton() function is supposed to be passed a string that is known
4405 to be a valid IP address. However, in the case of IPv6 addresses, it was
4406 not checking this. This is a hostage to fortune. Exim now panics and dies
4407 if the condition is not met. A case was found where this could be provoked
4408 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4409 components; fortuitously, this particular loophole had already been fixed
4410 by change 4.50/55 or 4.44/33 above.
4412 If there are any other similar loopholes, the new check in host_aton()
4413 itself should stop them being exploited. The report I received stated that
4414 data on the command line could provoke the exploit when Exim was running as
4415 exim, but did not say which command line option was involved. All I could
4416 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4417 running as the user.
4420 35. There was a buffer overflow vulnerability in the SPA authentication code
4421 (which came originally from the Samba project). I have added a test to the
4422 spa_base64_to_bits() function which I hope fixes it.
4425 36. The daemon start-up calls getloadavg() while still root for those OS that
4426 need the first call to be done as root, but it missed one case: when
4427 deliver_queue_load_max is set with deliver_drop_privilege. This is
4428 necessary for the benefit of the queue runner, because there is no re-exec
4429 when deliver_drop_privilege is set.
4431 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4432 was in use, was not putting the data itself into the right store pool;
4433 consequently, it could be overwritten for a subsequent message in the same
4434 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4437 38. Sometimes the final signoff response after QUIT could fail to get
4438 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4439 < 0 before doing a fflush(). This bug looks as though it goes back to the
4440 introduction of TLS in release 3.20, but "sometimes" must have been rare
4441 because the tests only now provoked it.
4447 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
4448 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
4449 Now it returns an integer. A number of calls were still expecting a T/F
4450 return. Fortuitously, in all cases, the tests worked in OK situations,
4451 which is the norm. However, things would have gone wrong on any write
4452 failures on the smtp file descriptor. This function is used when sending
4453 messages over SMTP and also when doing verify callouts.
4455 2. When Exim is called to do synchronous delivery of a locally submitted
4456 message (the -odf or -odi options), it no longer closes stderr before doing
4459 3. Implemented the mua_wrapper option.
4461 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
4463 5. Implemented the functions header_remove(), header_testname(),
4464 header_add_at_position(), and receive_remove_recipient(), and exported them
4467 6. If an ACL "warn" statement specified the addition of headers, Exim already
4468 inserted X-ACL-Warn: at the start if there was no header name. However, it
4469 was not making this test for the second and subsequent header lines if
4470 there were newlines in the string. This meant that an invalid header could
4471 be inserted if Exim was badly configured.
4473 7. Allow an ACL "warn" statement to add header lines at the start or after all
4474 the Received: headers, as well as at the end.
4476 8. Added the rcpt_4xx retry error code.
4478 9. Added postmaster_mailfrom=xxx to callout verification option.
4480 10. Added mailfrom=xxxx to the callout verification option, for verify=
4483 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
4484 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
4486 12. Inserted some casts to stop certain compilers warning when using pointer
4487 differences as field lengths or precisions in printf-type calls (mostly
4488 affecting debugging statements).
4490 13. Added optional readline() support for -be (dynamically loaded).
4492 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
4493 same clock tick as a message's arrival, so that its received time was the
4494 same as the "first fail" time on the retry record, and that message
4495 remained on the queue past the ultimate address timeout, every queue runner
4496 would try a delivery (because it was past the ultimate address timeout) but
4497 after another failure, the ultimate address timeout, which should have then
4498 bounced the address, did not kick in. This was a "< instead of <=" error;
4499 in most cases the first failure would have been in the next clock tick
4500 after the received time, and all would be well.
4502 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
4503 being recognized when the domain list was tested by the match_domain
4504 condition in an expansion string.
4506 16. Added the ${str2b64: operator.
4508 17. Exim was always calling setrlimit() to set a large limit for the number of
4509 processes, without checking whether the existing limit was already
4510 adequate. (It did check for the limit on file descriptors.) Furthermore,
4511 errors from getrlimit() and setrlimit() were being ignored. Now they are
4512 logged to the main and panic logs, but Exim does carry on, to try to do its
4513 job under whatever limits there are.
4515 18. Imported PCRE 5.0.
4517 19. Trivial typo in log message " temporarily refused connection" (the leading
4520 20. If the log selector return_path_on_delivery was set and an address was
4521 redirected to /dev/null, the delivery process crashed because it assumed
4522 that a return path would always be set for a "successful" delivery. In this
4523 case, the whole delivery is bypassed as an optimization, and therefore no
4526 21. Internal re-arrangement: the function for sending a challenge and reading
4527 a response while authentication was assuming a zero-terminated challenge
4528 string. It's now changed to take a pointer and a length, to allow for
4529 binary data in such strings.
4531 22. Added the cyrus_sasl authenticator (code supplied by MBM).
4533 23. Exim was not respecting finduser_retries when seeking the login of the
4534 uid under which it was called; it was always trying 10 times. (The default
4535 setting of finduser_retries is zero.) Also, it was sleeping after the final
4536 failure, which is pointless.
4538 24. Implemented tls_on_connect_ports.
4540 25. Implemented acl_smtp_predata.
4542 26. If the domain in control=submission is set empty, Exim assumes that the
4543 authenticated id is a complete email address when it generates From: or
4544 Sender: header lines.
4546 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
4547 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
4548 chown and chgrp in /bin and hostname in /usr/bin.
4550 28. Exim was keeping the "process log" file open after each use, just as it
4551 does for the main log. This opens the possibility of it remaining open for
4552 long periods when the USR1 signal hits a daemon. Occasional processlog
4553 errors were reported, that could have been caused by this. Anyway, it seems
4554 much more sensible not to leave this file open at all, so that is what now
4557 29. The long-running daemon process does not normally write to the log once it
4558 has entered its main loop, and it closes the log before doing so. This is
4559 so that log files can straightforwardly be renamed and moved. However,
4560 there are a couple of unusual error situations where the daemon does write
4561 log entries, and I had neglected to close the log afterwards.
4563 30. The text of an SMTP error response that was received during a remote
4564 delivery was being truncated at 512 bytes. This is too short for some of
4565 the long messages that one sometimes sees. I've increased the limit to
4568 31. It is now possible to make retry rules that apply only when a message has a
4569 specific sender, in particular, an empty sender.
4571 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
4572 it possible to be selective about when SMTP synchronization is enforced.
4574 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
4576 32. Implemented hosts_connection_nolog.
4578 33. Added an ACL for QUIT.
4580 34. Setting "delay_warning=" to disable warnings was not working; it gave a
4583 35. Added mailbox_size and mailbox_filecount to appendfile.
4585 36. Added control = no_multiline_responses to ACLs.
4587 37. There was a bug in the logic of the code that waits for the clock to tick
4588 in the case where the clock went backwards by a substantial amount such
4589 that the microsecond fraction of "now" was more than the microsecond
4590 fraction of "then" (but the whole seconds number was less).
4592 38. Added support for the libradius Radius client library this is found on
4593 FreeBSD (previously only the radiusclient library was supported).
4599 1. When certain lookups returned multiple values in the form name=value, the
4600 quoting of the values was not always being done properly. Specifically:
4601 (a) If the value started with a double quote, but contained no whitespace,
4603 (b) If the value contained whitespace other than a space character (i.e.
4604 tabs or newlines or carriage returns) it was not quoted.
4605 This fix has been applied to the mysql and pgsql lookups by writing a
4606 separate quoting function and calling it from the lookup code. The fix
4607 should probably also be applied to nisplus, ibase and oracle lookups, but
4608 since I cannot test any of those, I have not disturbed their existing code.
4610 2. A hit in the callout cache for a specific address caused a log line with no
4611 reason for rejecting RCPT. Now it says "Previous (cached) callout
4612 verification failure".
4614 3. There was an off-by-one bug in the queryprogram router. An over-long
4615 return line was truncated at 256 instead of 255 characters, thereby
4616 overflowing its buffer with the terminating zero. As well as fixing this, I
4617 have increased the buffer size to 1024 (and made a note to document this).
4619 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
4620 when Exim is waiting for an SMTP response from a remote server, Exim
4621 restarts its select() call on the socket, thereby resetting its timeout.
4622 This is not a problem when such interrupts are rare. Somebody set up a cron
4623 job to run exiwhat every 2 minutes, which is less than the normal select()
4624 timeout (5 or 10 minutes). This meant that the select() timeout never
4625 kicked in because it was always reset. I have fixed this by comparing the
4626 time when an interrupt arrives with the time at the start of the first call
4627 to select(). If more time than the timeout has elapsed, the interrupt is
4628 treated as a timeout.
4630 5. Some internal re-factoring in preparation for the addition of Sieve
4631 extensions (by MH). In particular, the "personal" test is moved to a
4632 separate function, and given an option for scanning Cc: and Bcc: (which is
4633 not set for Exim filters).
4635 6. When Exim created an email address using the login of the caller as the
4636 local part (e.g. when creating a From: or Sender: header line), it was not
4637 quoting the local part when it contained special characters such as @.
4639 7. Installed new OpenBSD configuration files.
4641 8. Reworded some messages for syntax errors in "and" and "or" conditions to
4642 try to make them clearer.
4644 9. Callout options, other than the timeout value, were being ignored when
4645 verifying sender addresses in header lines. For example, when using
4647 verify = header_sender/callout=no_cache
4649 the cache was (incorrectly) being used.
4651 10. Added a missing instance of ${EXE} to the exim_install script; this affects
4652 only the Cygwin environment.
4654 11. When return_path_on_delivery was set as a log selector, if different remote
4655 addresses in the same message used different return paths and parallel
4656 remote delivery occurred, the wrong values would sometimes be logged.
4657 (Whenever a remote delivery process finished, the return path value from
4658 the most recently started remote delivery process was logged.)
4660 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
4661 lines when AUTH and/or TLS are in use. This is the "received protocol"
4662 field. Exim used to use "asmtp" for authenticated SMTP, without any
4663 indication (in the protocol name) for TLS use. Now it follows the RFC and
4664 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
4665 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
4666 names appear in log lines as well as in Received: header lines.
4668 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
4669 extensions, and comparison tests, and to fix some bugs.
4671 14. Changes to the "personal" filter test:
4673 (1) The test was buggy in that it was just doing the equivalent of
4674 "contains" tests on header lines. For example, if a user's address was
4675 anne@some.where, the "personal" test would incorrectly be true for
4677 To: susanne@some.where
4679 This test is now done by extracting each address from the header in turn,
4680 and checking the entire address. Other tests that are part of "personal"
4681 are now done using regular expressions (for example, to check local parts
4682 of addresses in From: header lines).
4684 (2) The list of non-personal local parts in From: addresses has been
4685 extended to include "listserv", "majordomo", "*-request", and "owner-*",
4686 taken from the Sieve specification recommendations.
4688 (3) If the message contains any header line starting with "List-" it is
4689 treated as non-personal.
4691 (4) The test for "circular" in the Subject: header line has been removed
4692 because it now seems ill-conceived.
4694 15. Minor typos in src/EDITME comments corrected.
4696 16. Installed latest exipick from John Jetmore.
4698 17. If headers_add on a router specified a text string that was too long for
4699 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
4700 of string_sprintf() is now avoided.
4702 18. $message_body_size was not set (it was always zero) when running the DATA
4703 ACL and the local_scan() function.
4705 19. For the "mail" command in an Exim filter, no default was being set for
4706 the once_repeat time, causing a random time value to be used if "once" was
4707 specified. (If the value happened to be <= 0, no repeat happened.) The
4708 default is now 0s, meaning "never repeat". The "vacation" command was OK
4709 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
4710 (I found it when inspecting the code).
4712 20. There is now an overall timeout for performing a callout verification. It
4713 defaults to 4 times the callout timeout, which applies to individual SMTP
4714 commands during the callout. The overall timeout applies when there is more
4715 than one host that can be tried. The timeout is checked before trying the
4716 next host. This prevents very long delays if there are a large number of
4717 hosts and all are timing out (e.g. when the network connections are timing
4718 out). The value of the overall timeout can be changed by specifying an
4719 additional sub-option for "callout", called "maxwait". For example:
4721 verify = sender/callout=5s,maxwait=20s
4723 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
4724 to the end before writing, but this should make it even safer).
4726 22. Exim was forgetting that it had advertised PIPELINING for the second and
4727 subsequent messages on an SMTP connection. It was also not resetting its
4728 memory on STARTTLS and an internal HELO.
4730 23. When Exim logs an SMTP synchronization error within a session, it now
4731 records whether PIPELINING has been advertised or not.
4733 24. Added 3 instances of "(long int)" casts to time_t variables that were being
4734 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
4735 rather than long int.
4737 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
4739 26. Added the never_mail option to autoreply.
4745 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
4746 crash if the getsockname() call failed; this can happen if a connection is
4747 closed very soon after it is established. The problem was simply in the
4748 order in which certain operations were done, causing Exim to try to write
4749 to the SMTP stream before it had set up the file descriptor. The bug has
4750 been fixed by making things happen in the correct order.
4756 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
4757 before the connection was closed, thus losing the rejection response.
4759 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
4760 some early Solaris releases, but causes trouble in current releases where
4761 socklen_t is defined.
4763 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
4766 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
4769 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
4770 that releasing the top of it at the end releases what was used for sub-
4771 expansions (unless the block got too big). However, discard this block if
4772 the first thing is a variable or header, so that we can use its block when
4773 it is dynamic (useful for very large $message_headers, for example).
4775 6. Lookups now cache *every* query, not just the most recent. A new, separate
4776 store pool is used for this. It can be recovered when all lookup caches are
4777 flushed. Lookups now release memory at the end of their result strings.
4778 This has involved some general refactoring of the lookup sources.
4780 7. Some code has been added to the store_xxx() functions to reduce the amount
4781 of flapping under certain conditions.
4783 8. log_incoming_interface used to affect only the <= reception log lines. Now
4784 it causes the local interface and port to be added to several more SMTP log
4785 lines, for example "SMTP connection from", and rejection lines.
4787 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
4789 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
4791 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
4792 could be overwritten at the end of the current message (or the start of a
4793 new message if it was set in a HELO ACL). The value is now preserved for
4794 the duration of the SMTP connection.
4796 12. If a transport had a headers_rewrite setting, and a matching header line
4797 contained an unqualified address, that address was qualified, even if it
4798 did not match any rewriting rules. The underlying bug was that the values
4799 of the flags that permit the existence of unqualified sender and recipient
4800 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
4801 non-local messages, and by -bnq for local messages) were not being
4802 preserved with the message after it was received.
4804 13. When Exim was logging an SMTP synchronization error, it could sometimes log
4805 "next input=" as part of the text comprising the host identity instead of
4806 the correct text. The code was using the same buffer for two different
4807 strings. However, depending on which order the printing function evaluated
4808 its arguments, the bug did not always show up. Under Linux, for example, my
4809 test suite worked just fine.
4811 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
4812 doesn't work with some older versions of Perl. It has been changed to "my",
4813 which in any case is probably the better facility to use.
4815 15. A really picky compiler found some instances of statements for creating
4816 error messages that either had too many or two few arguments for the format
4819 16. The size of the buffer for calls to the DNS resolver has been increased
4820 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
4821 for addresses that have a lot of PTR records. This alleviates a problem; it
4822 does not fully solve it.
4824 17. A dnsdb lookup for PTR records that receives more data than will fit in the
4825 buffer now truncates the list and logs the incident, which is the same
4826 action as happens when Exim is looking up a host name and its aliases.
4827 Previously in this situation something unpredictable would happen;
4828 sometimes it was "internal error: store_reset failed".
4830 18. If a server dropped the connection unexpectedly when an Exim client was
4831 using GnuTLS and trying to read a response, the client delivery process
4832 crashed while trying to generate an error log message.
4834 19. If a "warn" verb in an ACL added multiple headers to a message in a single
4835 string, for example:
4837 warn message = H1: something\nH2: something
4839 the text was added as a single header line from Exim's point of view
4840 though it ended up OK in the delivered message. However, searching for the
4841 second and subsequent header lines using $h_h2: did not work. This has been
4842 fixed. Similarly, if a system filter added multiple headers in this way,
4843 the routers could not see them.
4845 20. Expanded the error message when iplsearch is called with an invalid key to
4846 suggest using net-iplsearch in a host list.
4848 21. When running tests using -bh, any delays imposed by "delay" modifiers in
4849 ACLs are no longer actually imposed (and a message to that effect is
4852 22. If a "gecos" field in a passwd entry contained escaped characters, in
4853 particular, if it contained a \" sequence, Exim got it wrong when building
4854 a From: or a Sender: header from that name. A second bug also caused
4855 incorrect handling when an unquoted " was present following a character
4856 that needed quoting.
4858 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
4859 was not being matched caselessly.
4861 24. Arranged for all hyphens in the exim.8 source to be escaped with
4864 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
4865 properly. Recipient callout cache records were still being keyed to include
4866 the sender, even when use_sender was set false. This led to far more
4867 callouts that were necessary. The sender is no longer included in the key
4868 when use_sender is false.
4870 26. Added "control = submission" modifier to ACLs.
4872 27. Added the ${base62d: operator to decode base 62 numbers.
4874 28. dnsdb lookups can now access SRV records.
4876 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
4877 the configuration file.
4879 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
4880 (-v) mode. This makes the output for a verbose queue run more intelligible.
4882 31. Added a use_postmaster feature to recipient callouts.
4884 32. Added the $body_zerocount variable, containing the number of binary zero
4885 bytes in the message body.
4887 33. The time of last modification of the "new" subdirectory is now used as the
4888 "mailbox time last read" when there is a quota error for a maildir
4891 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
4893 35. Added +ignore_unknown as a special item in host lists.
4895 36. Code for decoding IPv6 addresses in host lists is now included, even if
4896 IPv6 support is not being compiled. This fixes a bug in which an IPv6
4897 address was recognized as an IP address, but was then not correctly decoded
4898 into binary, causing unexpected and incorrect effects when compared with
4905 1. Very minor rewording of debugging text in manualroute to say "list of
4906 hosts" instead of "hostlist".
4908 2. If verify=header_syntax was set, and a header line with an unqualified
4909 address (no domain) and a large number of spaces between the end of the
4910 name and the colon was received, the reception process suffered a buffer
4911 overflow, and (when I tested it) crashed. This was caused by some obsolete
4912 code that should have been removed. The fix is to remove it!
4914 3. When running in the test harness, delay a bit after writing a bounce
4915 message to get a bit more predictability in the log output.
4917 4. Added a call to search_tidyup() just before forking a reception process. In
4918 theory, someone could use a lookup in the expansion of smtp_accept_max_
4919 per_host which, without the tidyup, could leave open a database connection.
4921 5. Added the variables $recipient_data and $sender_data which get set from a
4922 lookup success in an ACL "recipients" or "senders" condition, or a router
4923 "senders" option, similar to $domain_data and $local_part_data.
4925 6. Moved the writing of debug_print from before to after the "senders" test
4928 7. Change 4.31/66 (moving the time when the Received: is generated) caused
4929 problems for message scanning, either using a data ACL, or using
4930 local_scan() because the Received: header was not generated till after they
4931 were called (in order to set the time as the time of reception completion).
4932 I have revised the way this works. The header is now generated after the
4933 body is received, but before the ACL or local_scan() are called. After they
4934 are run, the timestamp in the header is updated.
4940 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
4941 before starting a queue runner without re-exec. This happened only when
4942 deliver_drop_privilege was set or when the Exim user was set to root. The
4943 effect of the bug was that timeouts during subsequent deliveries caused
4944 crashes instead of being properly handled. The handler is now left at its
4945 default (and expected) setting.
4947 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
4948 message, again when deliver_drop_privilege is set or Exim is run as root.
4949 The bug described in (1) was not present in this case, but the tidying up
4950 of the other signals was missing. I have made the two cases consistent.
4952 3. The ignore_target_hosts setting on a manualroute router was being ignored
4953 for hosts that were looked up using the /MX notation.
4955 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
4958 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
4959 operated on the sender address. After changing the $sender_address to <>
4960 for the sender address verify, Exim was re-instated it as the original
4961 (before rewriting) address, but remembering that it had rewritten it, so it
4962 wasn't rewriting it again. This bug also had the effect of breaking the
4963 sender address verification caching when the sender address was rewritten.
4965 6. The ignore_target_hosts option was being ignored by the ipliteral router.
4966 This has been changed so that if the ip literal address matches
4967 ignore_target_hosts, the router declines.
4969 7. Added expansion conditions match_domain, match_address, and match_local_
4970 part (NOT match_host).
4972 8. The placeholder for the Received: header didn't have a length field set.
4974 9. Added code to Exim itself and to exim_lock to test for a specific race
4975 condition that could lead to file corruption when using MBX delivery. The
4976 issue is with the lockfile that is created in /tmp. If this file is removed
4977 after a process has opened it but before that process has acquired a lock,
4978 there is the potential for a second process to recreate the file and also
4979 acquire a lock. This could lead to two Exim processes writing to the file
4980 at the same time. The added code performs the same test as UW imapd; it
4981 checks after acquiring the lock that its file descriptor still refers to
4982 the same named file.
4984 10. The buffer for building added header lines was of fixed size, 8192 bytes.
4985 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
4988 11. Added the smtp_active_hostname option. If used, this will typically be made
4989 to depend on the incoming interface address. Because $interface_address is
4990 not set up until the daemon has forked a reception process, error responses
4991 that can happen earlier (such as "too many connections") no longer contain
4994 12. If an expansion in a condition on a "warn" statement fails because a lookup
4995 defers, the "warn" statement is abandoned, and the next ACL statement is
4996 processed. Previously this caused the whole ACL to be aborted.
4998 13. Added the iplsearch lookup type.
5000 14. Added ident_timeout as a log selector.
5002 15. Added tls_certificate_verified as a log selector.
5004 16. Added a global option tls_require_ciphers (compare the smtp transport
5005 option of the same name). This controls incoming TLS connections.
5007 17. I finally figured out how to make tls_require_ciphers do a similar thing
5008 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
5009 before starting the TLS session.
5011 18. Tabs are now shown as \t in -bP output.
5013 19. If the log selector return_path_on_delivery was set, Exim crashed when
5014 bouncing a message because it had too many Received: header lines.
5016 20. If two routers both had headers_remove settings, and the first one included
5017 a superfluous trailing colon, the final name in the first list and the
5018 first name in the second list were incorrectly joined into one item (with a
5019 colon in the middle).
5025 1. Added -C and -D options to the exinext utility, mainly to make it easier
5026 to include in the automated testing, but these could be helpful when
5027 multiple configurations are in use.
5029 2. The exinext utility was not formatting the output nicely when there was
5030 an alternate port involved in the retry record key, nor when there was a
5031 message id as well (for retries that were specific to a specific message
5032 and a specific host). It was also confused by IPv6 addresses, because of
5033 the additional colons they contain. I have fixed the IPv4 problem, and
5034 patched it up to do a reasonable job for IPv6.
5036 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
5037 delivery, the log line now contains "pipelined" if PIPELINING was used.
5039 4. An SMTP transport process used to panic and die if the bind() call to set
5040 an explicit outgoing interface failed. This has been changed; it is now
5041 treated in the same way as a connect() failure.
5043 5. A reference to $sender_host_name in the part of a conditional expansion
5044 that was being skipped was still causing a DNS lookup. This no longer
5047 6. The def: expansion condition was not recognizing references to header lines
5048 that used bh_ and bheader_.
5050 7. Added the _cache feature to named lists.
5052 8. The code for checking quota_filecount in the appendfile transport was
5053 allowing one more file than it should have been.
5055 9. For compatibility with Sendmail, the command line option
5063 and sets the incoming protocol and host name (for trusted callers). The
5064 host name and its colon can be omitted when only the protocol is to be set.
5065 Note the Exim already has two private options, -pd and -ps, that refer to
5066 embedded Perl. It is therefore impossible to set a protocol value of "d" or
5067 "s", but I don't think that's a major issue.
5069 10. A number of refactoring changes to the code, none of which should affect
5072 (a) The number of logging options was getting close to filling up the
5073 32-bit word that was used as a bit map. I have split them into two classes:
5074 those that are passed in the argument to log_write(), and those that are
5075 only ever tested independently outside of that function. These are now in
5076 separate 32-bit words, so there is plenty of room for expansion again.
5077 There is no change in the user interface or the logging behaviour.
5079 (b) When building, for example, log lines, the code previously used a
5080 macro that called string_cat() twice, in order to add two strings. This is
5081 not really sufficiently general. Furthermore, there was one instance where
5082 it was actually wrong because one of the argument was used twice, and in
5083 one call a function was used. (As it happened, calling the function twice
5084 did not affect the overall behaviour.) The macro has been replaced by a
5085 function that can join an arbitrary number of extra strings onto a growing
5088 (c) The code for expansion conditions now uses a table and a binary chop
5089 instead of a serial search (which was left over from when there were very
5090 few conditions). Also, it now recognizes conditions like "pam" even when
5091 the relevant support is not compiled in: a suitably worded error message is
5092 given if an attempt is made to use such a condition.
5094 11. Added ${time_interval:xxxxx}.
5096 12. A bug was causing one of the ddress fields not to be passed back correctly
5097 from remote delivery subprocesses. The field in question was not being
5098 subsequently used, so this caused to problems in practice.
5100 13. Added new log selectors queue_time and deliver_time.
5102 14. Might have fixed a bug in maildirsizefile handling that threw up
5103 "unexpected character" debug warnings, and recalculated the data
5104 unnecessarily. In any case, I expanded the warning message to give more
5107 15. Added the message "Restricted characters in address" to the statements in
5108 the default ACL that block characters like @ and % in local parts.
5110 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
5111 Three changes have been made:
5113 (a) There was a serious bug; a negative response to MAIL caused the whole
5114 recipient domain to be cached as invalid, thereby blocking all messages
5115 to all local parts at the same domain, from all senders. This bug has
5116 been fixed. The domain is no longer cached after a negative response to
5117 MAIL if the sender used is not empty.
5119 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
5122 (c) A new callout option, "use_sender" has been added for people who want
5123 the modified behaviour.
5129 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
5132 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
5133 indeed breaks things for older releases.
5135 3. Added additional logging to the case where there is a problem reading data
5136 from a filter that is running in a subprocess using a pipe, in order to
5137 try to track down a specific problem.
5139 4. Testing facility fudge: when running in the test harness and attempting
5140 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
5141 getting "No route to host". Convert this to a timeout.
5143 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
5146 6. Some OS don't have socklen_t but use size_t instead. This affects the
5147 fifth argument of getsockopt() amongst other things. This is now
5148 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
5149 can be set for individual OS. I have set it for SunOS5, OSF1, and
5150 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
5151 some earlier ones do not.
5153 7. Change 4.30/15 was not doing the test caselessly.
5155 8. The standard form for an IPv6 address literal was being rejected by address
5156 parsing in, for example, MAIL and RCPT commands. An example of this kind of
5157 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5158 this, as well as the form without the "IPv6" on the front (but only when
5159 address literals are enabled, of course).
5161 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
5163 10. Exim crashed if a message with an empty sender address specified by -f
5164 encountered a router with an errors_to setting. This could be provoked only
5165 by a command such as
5169 where an empty string was supplied; "<>" did not hit this bug.
5171 11. Installed PCRE release 4.5.
5173 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
5174 remained set. It is now erased.
5176 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
5177 times from message ids (which are base 36 rather than the normal 62).
5179 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
5180 were being counted as actual protocol errors, and logged if the log
5181 selector +smtp_protocol_error was set. One cannot be perfect in this test,
5182 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
5183 and DATA following a set of rejected RCPTs do not count as protocol errors.
5184 In other words, Exim assumes they were pipelined, though this may not
5185 actually be the case. Of course, in all cases the client gets an
5186 appropriate error code.
5188 15. If a lookup fails in an ACL condition, a message about the failure may
5189 be available; it is used if testing the ACL cannot continue, because most
5190 such messages specify what the cause of the deferral is. However, some
5191 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
5192 that caused an old message to be retained and used if a later statement
5193 caused a defer, replacing the real cause of the deferral.
5195 16. If an IP address had so many PTR records that the DNS lookup buffer
5196 was not large enough to hold them, Exim could crash while trying to process
5197 the truncated data. It now detects and logs this case.
5199 17. Further to 4.21/58, another change has been made: if (and only if) the
5200 first line of a message (the first header line) ends with CRLF, a bare LF
5201 in a subsequent header line has a space inserted after it, so as not to
5202 terminate the header.
5204 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
5205 unnecessarily, used atoi() instead of strtol(), and didn't check the
5206 termination when getting file sizes from file names by regex.
5208 19. Completely re-implemented the support for maildirsize files, in the light
5209 of a number of problems with the previous contributed implementation
5210 (4.30/29). In particular:
5212 . If the quota is zero, the maildirsize file is maintained, but no quota is
5215 . If the maildir directory does not exist, it is created before any attempt
5216 to write a maildirsize file.
5218 . The quota value in the file is just a cache; if the quota is changed in
5219 the transport, the new value overrides.
5221 . A regular expression is available for excluding directories from the
5224 20. The autoreply transport checks the characters in options that define the
5225 message's headers; it allows continued headers, but it was checking with
5226 isspace() after an embedded newline instead of explicitly looking for a
5229 21. If all the "regular" hosts to which an address was routed had passed their
5230 expiry times, and had not reached their retry times, the address was
5231 bounced, even if fallback hosts were defined. Now Exim should go on to try
5234 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
5235 equivalent code in the SMTP transport. Some hosts send humungous responses
5236 to HELO/EHLO, more than 1024 it seems.
5238 23. Refactoring: code in filter.c used (void *) for "any old type" but this
5239 gives compiler warnings in some environments. I've now done it "properly",
5242 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
5243 (because of problems with the built-in one) was declared to return uschar *
5244 instead of char *, causing compiler failure.
5246 25. Fixed a file descriptor leak when processing alias/forward files.
5248 26. Fixed a minor format string issue in dbfn.c.
5250 27. Typo in exim.c: ("dmbnz" for "dbmnz").
5252 28. If a filter file refered to $h_xxx or $message_headers, and the headers
5253 contained RFC 2047 "words", Exim's memory could, under certain conditions,
5256 29. When a sender address is verified, it is cached, to save repeating the test
5257 when there is more than one recipient in a message. However, when the
5258 verification involves a callout, it is possible for different callout
5259 options to be set for different recipients. It is too complicated to keep
5260 track of this in the cache, so now Exim always runs a verification when a
5261 callout is required, relying on the callout cache for the optimization.
5262 The overhead is duplication of the address routing, but this should not be
5265 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
5266 to be verified with callout=postmaster, and the main callout worked but the
5267 postmaster check failed, the verification correctly failed. However, if a
5268 subsequent RCPT command asked for sender verification *without* the
5269 postmaster check, incorrect caching caused this verification also to fail,
5272 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
5273 it was not caching the DNS options (qualify_single, search_parents) that
5274 were used when the lookup failed. A subsequent lookup with different
5275 options therefore always gave the same answer, though there were cases
5276 where it should not have. (Example: a "domains = !$mx_any" option on a
5277 dnslookup router: the "domains" option is always processed without any
5278 widening, but the router might have qualify_single set.) Now Exim uses the
5279 cached value only when the same options are set.
5281 32. Added John Jetmore's "exipick" utility to the distribution.
5283 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
5284 than a timeout (e.g. a certificate is required, and is not provided), an
5285 Exim server now closes the connection immediately. Previously it waited for
5286 the client to close - but if the client is SSL, it seems that they each
5287 wait for each other, leading to a delay before one of them times out.
5289 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
5290 maintained 0.8.x compatibility because I don't think many are using it, and
5291 it is clearly obsolete.
5293 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
5296 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
5297 client certificate was expired. A simple patch fixes this, though I don't
5298 understand the full logic of why the verify callback is called multiple
5301 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
5302 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
5303 which causes problems with some clients (such as the Certicom SSL Plus
5304 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
5305 disables the coutermeasure allowing Eudora to connect."
5307 38. Exim was not checking that a write() to a log file succeeded. This could
5308 lead to Bad Things if a log got too big, in particular if it hit a file
5309 size limit. Exim now panics and dies if it cannot write to a log file, just
5310 as it does if it cannot open a log file.
5312 39. Modified OS/Makefile-Linux so that it now contains
5314 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
5316 The two -D definitions ensure that Exim is compiled with large file
5317 support, which makes it possible to handle log files that are bigger than
5320 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
5321 instance) a domain was checked against a named list that involved a lookup,
5322 causing $domain_data to be set, then another domain was checked against the
5323 same list, then the first domain was re-checked, the value of $domain_data
5324 after the final check could be wrong. In particular, if the second check
5325 failed, it could be set empty. This bug probably also applied to
5328 41. The strip_trailing_dot option was not being applied to the address given
5329 with the -f command-line option.
5331 42. The code for reading a message's header from the spool was incrementing
5332 $received_count, but never initializing it. This meant that the value was
5333 incorrect (doubled) while delivering a message in the same process in which
5334 it was received. In the most common configuration of Exim, this never
5335 happens - a fresh exec is done - but it can happen when
5336 deliver_drop_privilege is set.
5338 43. When Exim logs an SMTP synchronization error - client data sent too soon -
5339 it now includes up to 150 characters of the unexpected data in the log
5342 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
5343 and building data strings. The size of both of these buffers was 10 000
5344 bytes - far larger than anybody would *ever* want, thought I. Needless to
5345 say, somebody hit the limit. I have increased the maximum line length to
5346 20 000 and the maximum data length of concatenated lines to 100 000. I have
5347 also fixed two bugs, because there was no checking on these buffers. Tsk,
5348 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
5349 buffer is too small.
5351 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
5352 lsearch lookups. Now it does.
5354 46. When parsing a route_list item in a manualroute router, a fixed-length
5355 buffer was used for the list of hosts. I made this 1024 bytes long,
5356 thinking that nobody would ever have a list of hosts that long. Wrong.
5357 Somebody had a whole pile of complicated expansion conditions, and the
5358 string was silently truncated, leading to an expansion error. It turns out
5359 that it is easier to change to an unlimited length (owing to other changes
5360 that have happened since this code was originally written) than to build
5361 structure for giving a limitation error. The length of the item that
5362 expands into the list of hosts is now unlimited.
5364 47. The lsearch lookup could not handle data where the length of text line was
5365 more than 4095 characters. Such lines were truncated, leading to shortened
5366 data being returned. It should now handle lines of any length.
5368 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
5369 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
5372 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
5373 replacing strings such as BIN_DIRECTORY in a source file: the replacement
5374 no longer happens in comment lines. A list of replacements is now placed
5375 at the head of all of the source files, except those whose only change is
5376 to replace PERL_COMMAND in the very first #! line.
5378 50. Replaced the slow insertion sort in queue.c, for sorting the list of
5379 messages on the queue, with a bottom-up merge sort, using code contributed
5380 by Michael Haardt. This should make operations like -bp somewhat faster on
5381 large queues. It won't affect queue runners, except when queue_run_in_order
5384 51. Installed eximstats 1.31 in the distribution.
5386 52. Added support for SRV lookups to the dnslookup router.
5388 53. If an ACL referred to $message_body or $message_body_end, the value was not
5389 reset for any messages that followed in the same SMTP session.
5391 54. The store-handling optimization for building very long strings was not
5392 differentiating between the different store pools. I don't think this
5393 actually made any difference in practice, but I've tidied it.
5395 55. While running the routers to verify a sender address, $sender_address
5396 was still set to the sender address. This is wrong, because when routing to
5397 send a bounce to the sender, it would be empty. Therefore, I have changed
5398 it so that, while verifying a sender address, $sender_address is set to <>.
5399 (There is no change to what happens when verifying a recipient address.)
5401 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
5402 target A or AAAA records (if not already returned) without resetting the
5403 qualify_single or search_parents options of the DNS resolver. These are
5404 inappropriate in this case because the targets of MX and SRV records must
5405 be FQDNs. A broken DNS record could cause trouble if it happened to have a
5406 target that, when qualified, matched something in the local domain. These
5407 two options are now turned off when doing these lookups.
5409 57. It seems that at least some releases of Reiserfs (which does not have the
5410 concept of a fixed number of inodes) returns zero and not -1 for the
5411 number of available inodes. This interacted badly with check_spool_inodes,
5412 which assumed that -1 was the "no such thing" setting. What I have done is
5413 to check that the total number of inodes is greater than zero before doing
5414 the test of how many are available.
5416 58. When a "warn" ACL statement has a log_message modifier, the message is
5417 remembered, and not repeated. This is to avoid a lot of repetition when a
5418 message has many recipients that cause the same warning to be written.
5419 Howewer, Exim was preserving the list of already written lines for an
5420 entire SMTP session, which doesn't seem right. The memory is now reset if a
5421 new message is started.
5423 59. The "rewrite" debugging flag was not showing the result of rewriting in the
5424 debugging output unless log_rewrite was also set.
5426 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
5427 of (int)(handle) when we know that handle contains (void *)(-1).
5429 61. The Exim daemon panic-logs an error return when it closes the incoming
5430 connection. However "connection reset by peer" seems to be common, and
5431 isn't really an error worthy of noting specially, so that particular error
5434 62. When Exim is trying to find all the local interfaces, it used to panic and
5435 die if the ioctl to get the interface flags failed. However, it seems that
5436 on at least one OS (Solaris 9) it is possible to have an interface that is
5437 included in the list of interfaces, but for which you get a failure error
5438 for this call. This happens when the interface is not "plumbed" into a
5439 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
5440 failure of the "get flags" call assumes that the interface is down.
5442 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
5443 makes life easier for people who are doing arithmetic on fields extracted
5444 from dates, where you often get leading zeros that should not be
5445 interpreted as octal.
5447 64. Added qualify_domain to the redirect router, to override the global
5450 65. If a pathologically long header line contained very many addresses (the
5451 report of this problem mentioned 10 000) and each of them was rewritten,
5452 Exim could use up a very large amount of memory. (It kept on making new
5453 copies of the header line as it rewrote, and never released the old ones.)
5454 At the expense of a bit more processing, the header rewriting function has
5455 been changed so that it no longer eats memory in this way.
5457 66. The generation of the Received: header has been moved from the time that a
5458 message starts to be received, to the time that it finishes. The timestamp
5459 in the Received: header should now be very close to that of the <= log
5460 line. There are two side-effects of this change:
5462 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
5463 logged header lines no longer include the local Received: line, because
5464 it has not yet been created. The same applies to a copy of the message
5465 that is returned to a non-SMTP sender when a message is rejected.
5467 (b) When a filter file is tested using -bf, no additional Received: header
5468 is added to the test message. After some thought, I decided that this
5471 This change does not affect the value of $received_for. It is still set
5472 after address rewriting, but before local_scan() is called.
5474 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
5476 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
5477 gave an unhelpful panic error message, and a defer error. I have managed to
5478 change this behaviour so that it now rejects any supplied certificate,
5479 which seems right, as the list of acceptable certificates is empty.
5481 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
5482 gave an unhelpful defer error. I have not managed to make this reject any
5483 supplied certificates, but the error message it gives is "no certificate
5484 supplied", which is not helpful.
5486 70. exigrep's output now also includes lines that are not associated with any
5487 message, but which match the given pattern. Implemented by a patch from
5488 Martin Sluka, which also tidied up the Perl a bit.
5490 71. Recipient callout verification, like sender verification, was using <> in
5491 the MAIL FROM command. This isn't really the right thing, since the actual
5492 sender may affect whether the remote host accepts the recipient or not. I
5493 have changed it to use the actual sender in the callout; this means that
5494 the cache record is now keyed on a recipient/sender pair, not just the
5495 recipient address. There doesn't seem to be a real danger of callout loops,
5496 since a callout by the remote host to check the sender would use <>.
5497 [SEE ABOVE: changed after hitting problems.]
5499 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
5500 temporary errors. However, in the case of such a code being given after
5501 the end of a data transmission (i.e. after ".") Exim was failing to write
5502 a retry record for the message. (Yes, there was some broken host that was
5503 actually sending 8xx at this point.)
5505 73. An unknown lookup type in a host list could cause Exim to panic-die when
5506 the list was checked. (An example that provoked this was putting <; in the
5507 middle of a list instead of at the start.) If this happened during a DATA
5508 ACL check, a -D file could be left lying around. This kind of configuration
5509 error no longer causes Exim to die; instead it causes a defer errror. The
5510 incident is still logged to the main and panic logs.
5512 74. Buglet left over from Exim 3 conversion. The message "too many messages
5513 in one connection" was written to the rejectlog but not the mainlog, except
5514 when address rewriting (yes!) was being logged.
5516 75. Added write_rejectlog option.
5518 76. When a system filter was run not as root (that is, when system_filter_user
5519 was set), the values of the $n variables were not being returned to the
5520 main process; thus, they were not subsequently available in the $sn
5523 77. Added +return_path_on_delivery log selector.
5525 78. A connection timeout was being treated differently from recipients deferred
5526 when testing hosts_max_try with a message that was older than the host's
5527 retry timeout. (The host should not be counted, thus allowing all hosts to
5528 be tried at least once before bouncing.) This may have been the cause of an
5529 occasionally reported bug whereby a message would remain on the queue
5530 longer than the retry timeout, but would be bounced if a delivery was
5531 forced. I say "may" because I never totally pinned down the problem;
5532 setting up timeout/retry tests is difficult. See also the next item.
5534 79. The ultimate address timeout was not being applied to errors that involved
5535 a combination of host plus message (for example, a timeout on a MAIL
5536 command). When an address resolved to a number of possible hosts, and they
5537 were not all tried for each delivery (e.g. because of hosts_max_try), a
5538 message could remain on the queue longer than the retry timeout.
5540 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
5543 81. Fixed an obscure SMTP outgoing bug which required at least the following
5544 conditions: (a) there was another message waiting for the same server;
5545 (b) the server returned 5xx to all RCPT commands in the first message so
5546 that the message was not completed; (c) the server dropped the connection
5547 or gave a negative response to the RSET that Exim sends to abort the
5548 transaction. The observed case was a dropped connection after DATA that had
5549 been sent in pipelining mode. That is, the server had advertised PIPELINING
5550 but was not implementing it correctly. The effect of the bug was incorrect
5551 behaviour, such as trying another host, and this could lead to a crash.
5557 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
5558 and daemon.c were passed as pointers to ints; they should have been
5559 pointers to socklen_t variables (which are typically unsigned ints).
5561 2. Some signed/unsigned type warnings in the os.c file for Linux have been
5564 3. Fixed a really odd bug that affected only the testing scheme; patching a
5565 certain fixed string in the binary changed the value of another string that
5566 happened to be identical to the end of the original first string.
5568 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
5569 name", it returns that address as the IP address. On some operating
5570 systems (e.g. Solaris), it also passes back the IP address string as the
5571 "host name". However, on others (e.g. Linux), it passes back an empty
5572 string. Exim wasn't checking for this, and was changing the host name to an
5573 empty string, assuming it had been canonicized.
5575 5. Although rare, it is permitted to have more than one PTR record for a given
5576 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
5577 all the names associated with an address, because they do in Solaris.
5578 However, it seems that they do not in Linux for data that comes from the
5579 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
5580 I found this out when I moved to a new Linux workstation and tried to run
5581 the Exim test suite.
5583 To get round this problem I have changed the code so that it now does its
5584 own call to the DNS to look up PTR records when searching for a host name.
5585 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
5586 addresses that are only in /etc/hosts are still found.
5588 This behaviour is, however, controlled by an option called host_lookup_
5589 order, which defaults to "bydns:byaddr". If people want to use the other
5590 order, or indeed, just use one or the other means of lookup, they can
5591 specify it in this variable.
5593 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
5594 some operating systems, this comes back from gethostbyaddr() as an empty
5595 string, and this is what Exim used to test for. However, it seems that in
5596 other systems, "." is yielded. Exim now tests for this case too.
5598 7. The values of check_spool_space and check_log_space are now held internally
5599 as a number of kilobytes instead of an absolute number of bytes. If a
5600 numbers is specified without 'K' or 'M', it is rounded up to the nearest
5601 kilobyte. This means that much larger values can be stored.
5603 8. Exim monitor: an attempt to get the action menu when not actually pointing
5604 at a message produces an empty menu entitled "No message selected". This
5605 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
5606 no entries in it ("Shell widget menu has zero width and/or height"). So I
5607 have added a single, blank menu entry in this case.
5609 9. Added ${quote_local_part.
5611 10. MIME decoding is now applied to the contents of Subject: header lines when
5614 11. Now that a reference to $sender_host_address automatically causes a reverse
5615 lookup to occur if necessary (4.13/18), there is no need to arrange for a
5616 host lookup before query-style lookups in lists that might use this
5617 variable. This has therefore been abolished, and the "net-" prefix is no
5618 longer necessary for query-style lookups.
5620 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
5621 have been a typo for LFLAGS, so it has been changed.
5623 13. The install script calls Exim with "-C /dev/null" in order to find the
5624 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
5625 to be output. Howeve, since Exim outputs its version number before the
5626 error, it didn't break the script. It just looked ugly. I fixed this by
5627 always allowing "-C /dev/null" if the caller is root.
5629 14. Ignore overlarge ACL variable number when reading spool file - insurance
5630 against a later release with more variables having written the file.
5632 15. The standard form for an IPv6 address literal was being rejected by EHLO.
5633 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5634 this, as well as the form without the "IPv6" on the front.
5636 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
5637 OS/Makefile-Darwin file.
5639 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
5640 only to LDAP libraries that do not have LDAP_OPT_DEREF.
5642 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
5643 variable. However, some OS use ints for inodes. I've added cast to long int
5644 to get rid of the compiler warning.
5646 19. I had forgotten to lock out "/../" in configuration file names when
5647 ALT_CONFIG_PREFIX was set.
5649 20. Routers used for verification do not need to specify transports. However,
5650 if such a router generated a host list, and callout was configured, Exim
5651 crashed, because it could not find a port number from the (non-existent)
5652 transport. It now assumes port 25 in this circumstance.
5654 21. Added the -t option to exigrep.
5656 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
5657 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
5658 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
5659 them is set, LOOKUP_LSEARCH is forced.
5661 23. "exim -bV" now outputs a list of lookups that are included in the binary.
5663 24. Added sender and host information to the "rejected by local_scan()" log
5664 line; previously there was no indication of these.
5666 25. Added .include_if_exists.
5668 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
5669 Linux. It turns out that not all file systems support this. Apparently some
5670 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
5671 do it.) To cope with this, the error EINVAL, which means that sync-ing is
5672 not supported on the file descriptor, is now ignored when Exim is trying to
5673 sync a directory. This applies only to Linux.
5675 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
5677 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
5678 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
5681 29. Added support for maildirsize files from supplied patch (modified a bit).
5683 30. The use of :fail: followed by an empty string could lead Exim to respond to
5684 sender verification failures with (e.g.):
5686 550 Verification failed for <xxx>
5687 550 Sender verify failed
5689 where the first response line was missing the '-' that indicates it is not
5690 the final line of the response.
5692 31. The loop for finding the name of the user that called Exim had a hardwired
5693 limit of 10; it now uses the value of finduser_retries, which is used for
5694 all other user lookups.
5696 32. Added $received_count variable, available in data and not_smtp ACLs, and at
5699 33. Exim was neglecting to zero errno before one call of strtol() when
5700 expanding a string and expecting an integer value. On some systems this
5701 resulted in spurious "integer overflow" errors. Also, it was casting the
5702 result into an int without checking.
5704 34. Testing for a connection timeout using "timeout_connect" in the retry rules
5705 did not work. The code looks as if it has *never* worked, though it appears
5706 to have been documented since at least releast 1.62. I have made it work.
5708 35. The "timeout_DNS" error in retry rules, also documented since at least
5709 1.62, also never worked. As it isn't clear exactly what this means, and
5710 clearly it isn't a major issue, I have abolished the feature by treating it
5711 as "timeout", and writing a warning to the main and panic logs.
5713 36. The display of retry rules for -brt wasn't always showing the error code
5716 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
5717 timeout_connect_A, timeout_connect_MX.
5719 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
5720 to the empty sender.
5722 39. The daemon was not analysing the content of -oX till after it had closed
5723 stderr and disconnected from the controlling terminal. This meant that any
5724 syntax errors were only noted on the panic log, and the return code from
5725 the command was 0. By re-arranging the code a little, I've made the
5726 decoding happen first, so such errors now appear on stderr, and the return
5727 code is 1. However, the actual setting up of the sockets still happens in
5728 the disconnected process, so errors there are still only recorded on the
5731 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
5732 connections (as happens on some IP stacks) was logged at start up time as
5733 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
5734 it from "IPv6 and IPv4", which means that two separate sockets are being
5737 41. The debug output for gethostbyname2() or getipnodebyname() failures now
5738 says whether AF_INET or AF_INET6 was passed as an argument.
5740 42. Exiwhat output was messed up when time zones were included in log
5743 43. Exiwhat now gives more information about the daemon's listening ports,
5744 and whether -tls-on-connect was used.
5746 44. The "port" option of the smtp transport is now expanded.
5748 45. A "message" modifier in a "warn" statement in a non-message ACL was being
5749 silently ignored. Now an error message is written to the main and panic
5752 46. There's a new ACL modifier called "logwrite" which writes to a log file
5753 as soon as it is encountered.
5755 47. Added $local_user_uid and $local_user_gid at routing time.
5757 48. Exim crashed when trying to verify a sender address that was being
5760 49. Exim was recognizing only a space character after ".include". It now also
5761 recognizes a tab character.
5763 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
5764 extracting the relevant information from the specification. The man page no
5765 longer contains scrambled data for the -d option, and I've added a section
5766 at the front about calling Exim under different names.
5768 51. Added "extra_headers" argument to the "mail" command in filter files.
5770 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
5773 53. Installed eximstats 1.29.
5775 54. Added transport_filter_timeout as a generic transport option.
5777 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
5778 Cc: header lines. This was required by RFC 822, but it not required by RFC
5781 56. Exim used to add From:, Date:, and Message-Id: header lines to any
5782 incoming messages that did not have them. Now it does so only if the
5783 message originates locally, that is, if there is no associated remote host
5784 address. When Resent- header lines are present, this applies to the Resent-
5785 lines rather than the non-Resent- lines.
5787 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
5788 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
5790 58. Messages for configuration errors now include the name of the main
5791 configuration file - useful now that there may be more than one file in a
5792 list (.included file names were always shown).
5794 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
5795 for those rare installations that do not start the daemon as root or run it
5796 setuid root. I've cut out the call to initgroups() if the daemon is not
5799 60. The Exim user and group can now be bound into the binary as text strings
5800 that are looked up at the start of Exim's processing.
5802 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
5804 62. Added $mailstore_basename variable.
5806 63. Installed patch to sieve.c from Michael Haardt.
5808 64. When Exim failed to open the panic log after failing to open the main log,
5809 the original message it was trying to log was written to stderr and debug
5810 output, but if they were not available (the usual case in production), it
5811 was lost. Now it is written to syslog before the two lines that record the
5812 failures to open the logs.
5814 65. Users' Exim filters run in subprocesses under the user's uid. It is
5815 possible for a "deliver" command or an alias in a "personal" command to
5816 provoke an address rewrite. If logging of address rewriting is configured,
5817 this fails because the process is not running as root or exim. There may be
5818 a better way of dealing with this, but for the moment (because 4.30 needs
5819 to be released), I have disabled address rewrite logging when running a
5820 filter in a non-root, non-exim process.
5826 1. The buildconfig auxiliary program wasn't quoting the value set for
5827 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
5828 not defined. This bug was masked in 4.22 by the effect that was fixed in
5831 2. Some messages that were rejected after a message id was allocated were
5832 shown as "incomplete" by exigrep. It no longer does this for messages that
5833 are rejected by local_scan() or the DATA or non-SMTP ACLs.
5835 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
5836 have allow_domain_literals set, the ID did not get logged in the <= line.
5837 Domain literals are now always recognized in Message-ID: header lines.
5839 4. The first argument for a ${extract expansion item is the key name or field
5840 number. Leading and trailing spaces in this item were not being ignored,
5841 causing some misleading effects.
5843 5. When deliver_drop_privilege was set, single queue runner processes started
5844 manually (i.e. by the command "exim -q") or by the daemon (which uses the
5845 same command in the process it spins off) were not dropping privilege.
5847 6. When the daemon running as "exim" started a queue runner, it always
5848 re-executed Exim in the spun-off process. This is a waste of effort when
5849 deliver_drop_privilege is set. The new process now just calls the
5850 queue-runner function directly.
5856 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
5859 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
5860 ignored. Though the use of -oP was forcing the writing of a pid file, it
5861 was always written to the default place.
5863 3. If the message "no IP address found for host xxxx" is generated during
5864 incoming verification, it is now followed by identification of the incoming
5865 connection (so you can more easily find what provoked it).
5867 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
5869 5. Added some features to "harden" Exim a bit more against certain attacks:
5871 (a) There is now a build-time option called FIXED_NEVER_USERS that can
5872 be put in Local/Makefile. This is like the never_users runtime option,
5873 but it cannot be overridden. The default setting is "root".
5875 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
5876 prefix string with which any file named in a -C command line option
5879 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
5880 is retained for -C and -D only if the caller of Exim is root. Without
5881 it, the exim user may also use -C and -D and retain privilege.
5883 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
5884 command line option is disabled.
5886 6. Macro names set by the -D option must start with an upper case letter, just
5887 like macro names defined in the configuration file.
5889 7. Added "dereference=" facility to LDAP.
5891 8. Two instances of the typo "uknown" in the source files are fixed.
5893 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
5894 the Configure-Makefile script screwed up while processing it.
5896 10. Incorporated PCRE 4.4.
5898 11. The SMTP synchronization check was not operating right at the start of an
5899 SMTP session. For example, it could not catch a HELO sent before the client
5900 waited for the greeting. There is now a check for outstanding input at the
5901 point when the greeting is written. Because of the duplex, asynchronous
5902 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
5903 way, but not yet received, when the check is performed.
5905 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
5906 on TCP/IP sockets, because this apparently causes some broken clients to
5909 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
5910 unchanged) from the Cygwin maintainer.
5912 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
5913 was supported instead of testing for mbx. Effectively a typo.
5915 15. The spa authenticator server code was not checking that the input it
5916 received was valid base64.
5918 16. The debug output line for the "set" modifier in ACLs was not showing the
5919 name of the variable that was being set.
5921 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
5923 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
5924 lookup on its own. Something else was needed to trigger the lookup. For
5925 example, a match in host_lookup or the need for a host name in a host list.
5926 Now, if $sender_host_name is referenced and the host name has not yet been
5927 looked up, a lookup is performed. If the lookup fails, the variable remains
5928 empty, and $host_lookup_failed is set to "1".
5930 19. Added "eqi" as a case-independent comparison operator.
5932 20. The saslauthd authentication condition could segfault if neither service
5933 nor realm was specified.
5935 21. If an overflowing value such as "2048M" was set for message_size_limit, the
5936 error message that was logged was misleading, and incoming SMTP
5937 connections were dropped. The message is now more accurate, and temporary
5938 errors are given to SMTP connections.
5940 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
5941 (except RSET) with a 421 error, until QUIT is received. However, it was
5942 failing to send a response to QUIT.
5944 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
5945 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
5946 ACL is now run after the helo_try_verify_hosts code.
5948 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
5949 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
5950 case-independent, so other case variants are also recognized). Apparently
5951 some systems use these upper case variants.
5953 25. If more than two messages were waiting for the same host, and a transport
5954 filter was specified for the transport, Exim sent two messages over the
5955 same TCP/IP connection, and then failed with "socket operation on non-
5956 socket" when it tried to send the third.
5958 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
5960 27. The extern definition of crypt16() in expand.c was not being excluded when
5961 the OS had its own crypt16() function.
5963 28. Added bounce_return_body as a new option, and bounce_return_size_limit
5964 as a preferred synonym for return_size_limit, both as an option and as an
5967 29. Added LIBS=-liconv to OS/Makefile-OSF1.
5969 30. Changed the default configuration ACL to relax the local part checking rule
5970 for addresses that are not in any local domains. For these addresses,
5971 slashes and pipe symbols are allowed within local parts, but the sequence
5972 /../ is explicitly forbidden.
5974 31. SPA server authentication was not clearing the challenge buffer before
5977 32. log_message in a "warn" ACL statement was writing to the reject log as
5978 well as to the main log, which contradicts the documentation and doesn't
5979 seem right (because no rejection is happening). So I have stopped it.
5981 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
5982 However, I am unable to do any testing of this.
5984 34. Fixed an infelicity in the appendfile transport. When checking directories
5985 for a mailbox, to see if any needed to be created, it was accidentally
5986 using path names with one or more superfluous leading slashes; tracing
5987 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
5989 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
5990 contents are added to the log line that is written for every discarded
5991 recipient. (Previously a log_message setting was ignored.)
5993 36. The ${quote: operator now quotes the string if it is empty.
5995 37. The install script runs exim in order to find its version number. If for
5996 some reason other than non-existence or emptiness, which it checks, it
5997 could not run './exim', it was installing it with an empty version number,
5998 i.e. as "exim-". This error state is now caught, and the installation is
6001 38. An argument was missing from the function that creates an error message
6002 when Exim fails to connect to the socket for saslauthd authentication.
6003 This could cause Exim to crash, or give a corrupted message.
6005 39. Added isip, isip4, and isip6 to ${if conditions.
6007 40. The ACL variables $acl_xx are now saved with the message, and can be
6008 accessed later in routers, transports, and filters.
6010 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
6011 strings in the file are not string-expanded.
6013 42. If a MAIL command specified a SIZE value that was too large to fit into an
6014 int variable, the check against message_size_limit failed. Such values are
6015 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
6016 day this will have to be increased, but I don't think I want to be around
6017 when emails are that large.
6024 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
6025 iconv() is not standard in FreeBSD.
6027 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
6028 IPv6 enabled. The observed symptom was a segmentation fault on return from
6029 the function os_common_find_running_interfaces() in src/os.c.
6031 3. In the check_special_case() function in daemon.c I had used "errno" as an
6032 argument name, which causes warnings on some systems. This was basically a
6033 typo, since it was named "eno" in the comments!
6035 4. The code that waits for the clock to tick (at a resolution of some fraction
6036 of a second) so as to ensure message-id uniqueness was always waiting for
6037 at least one whole tick, when it could have waited for less. [This is
6038 almost certainly not relevant at current processor speeds, where it is
6039 unlikely to ever wait at all. But we try to future-proof.]
6041 5. The function that sleeps for a time interval that includes fractions of a
6042 second contained a race. It did not block SIGALRM between setting the
6043 timer, and suspending (a couple of lines later). If the interval was short
6044 and the sigsuspend() was delayed until after it had expired, the suspension
6045 never ended. On busy systems this could lead to processes getting stuck for
6048 6. Some uncommon configurations may cause a lookup to happen in a queue runner
6049 process, before it forks any delivery processes. The open lookup caching
6050 mechanism meant that the open file or database connection was passed into
6051 the delivery process. The problem was that delivery processes always tidy
6052 up cached lookup data. This could cause a problem for the next delivery
6053 process started by the queue runner, because the external queue runner
6054 process does not know about the closure. So the next delivery process
6055 still has data in the lookup cache. In the case of a file lookup, there was
6056 no problem because closing a file descriptor in a subprocess doesn't affect
6057 the parent. However, if the lookup was caching a connection to a database,
6058 the connection was closed, and the second delivery process was likely to
6059 see errors such as "PGSQL: query failed: server closed the connection
6060 unexpectedly". The problem has been fixed by closing all cached lookups
6061 in a queue runner before running a delivery process.
6063 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
6064 seem to have the "const" qualifier which it has on other OS. I've
6067 8. Change 4.21/2 was too strict. It is only if there are two authenticators
6068 *of the same type* (client or server) with the same public name that an
6069 error should be diagnosed.
6071 9. When Exim looked up a host name for an IP address, but failed to find the
6072 original IP address when looking up the host name (a safety check), it
6073 output the message "<ip address> does not match any IP for NULL", which was
6074 confusing, to say the least. The bug was that the host name should have
6075 appeared instead of "NULL".
6077 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
6078 user that is built into the binary, and the -C or -D options is used, root
6079 privilege is dropped before the configuration file is read. In addition,
6080 logging is switched to stderr instead of the normal log files. If the
6081 configuration then re-defines the Exim user, the unprivileged environment
6082 is probably not what is expected, so Exim logs a panic warning message (but
6085 However, if deliver_drop_privilege is set, the unprivileged state may well
6086 be exactly what is intended, so the warning has been cut out in that case,
6087 and Exim is allowed to try to write to its normal log files.
6093 1. smtp_return_error_details was not giving details for temporary sender
6094 or receiver verification errors.
6096 2. Diagnose a configuration error if two authenticators have the same public
6099 3. Exim used not to create the message log file for a message until the first
6100 delivery attempt. This could be confusing when incoming messages were held
6101 for policy or load reasons. The message log file is now created at the time
6102 the message is received, and an initial "Received" line is written to it.
6104 4. The automatically generated man page for command line options had a minor
6105 bug that caused no ill effects; however, a more serious problem was that
6106 the procedure for building the man page automatically didn't always
6107 operate. Consequently, release 4.20 contains an out-of-date version. This
6108 shouldn't happen again.
6110 5. When building Exim with embedded Perl support, the script that builds the
6111 Makefile was calling 'perl' to find its compile-time parameters, ignoring
6112 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
6114 6. The freeze_tell option was not being used for messages that were frozen on
6115 arrival, either by an ACL or by local_scan().
6117 7. Added the smtp_incomplete_transaction log selector.
6119 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
6120 was accepting AUTH without a new EHLO.
6122 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
6123 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
6126 10. Logging of TCP/IP connections (when configured) now happens in the main
6127 daemon process instead of the child process, so that the TCP/IP connection
6128 count is more accurate (but it can never be perfect).
6130 11. The use of "drop" in a nested ACL was not being handled correctly in the
6131 outer ACL. Now, if condition failure induced by the nested "drop" causes
6132 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
6133 or "require"), the connection is dropped.
6135 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
6136 that yield "discard" can now be used with an "accept" or a "discard" verb,
6137 but an error is generated for any others (because I can't see a useful way
6138 to define what should happen).
6140 13. When an ACL is read dynamically from a file (or anywhere else), the lines
6141 are now processed in the same way as lines in the Exim configuration file.
6142 In particular, continuation lines are supported.
6144 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
6146 15. Added -ti meaning -t -i.
6148 16. Check for letters, digits, hyphens, and dots in the names of dnslist
6149 domains, and warn by logging if others are found.
6151 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
6152 returned from GIFCONF when Exim is trying to find the list of interfaces on
6153 a host. The code in os.c has been modified to copy each ifreq to an aligned
6154 structure in all cases.
6156 Also, in some cases, the returned ifreq's were being copied to a 'struct
6157 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
6158 means the last couple of bytes of an IPv6 address could be chopped if the
6159 ifreq contained only a normal sockaddr (14 bytes storage).
6161 18. Named domain lists were not supported in the hosts_treat_as_local option.
6162 An entry such as +xxxx was not recognized, and was treated as a literal
6165 19. Ensure that header lines added by a DATA ACL are included in the reject log
6166 if the ACL subsequently rejects the message.
6168 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
6169 MD5 (which is deprecated).
6171 21. When testing a filter file using -bf, Exim was writing a message when it
6172 took the sender from a "From " line in the message, but it was not doing so
6173 when it took $return_path from a Return-Path: header line. It now does.
6175 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
6176 with a valid header line field name (a series of printing characters
6177 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
6179 23. Changed "disc" in the source to "disk" to conform to the documentation and
6180 the book and for uniformity.
6182 24. Ignore Sendmail's -Ooption=value command line item.
6184 25. When execve() failed while trying to run a command in a pipe transport,
6185 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
6186 could be confused with a return value of 69 from the command itself. This
6187 has been changed to 127, the value the shell returns if it is asked to run
6188 a non-existent command. The wording for the related log line suggests a
6189 non-existent command as the problem.
6191 26. If received_header_text expands to an empty string, do not add a Received:
6192 header line to the message. (Well, it adds a token one on the spool, but
6193 marks it "old" so that it doesn't get used or transmitted.)
6195 27. Installed eximstats 1.28 (addition of -nt option).
6197 28. There was no check for failure on the call to getsockname() in the daemon
6198 code. This can fail if there is a shortage of resources on the system, with
6199 ENOMEM, for example. A temporary error is now given on failure.
6201 29. Contrary to the C standard, it seems that in some environments, the
6202 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
6203 program. Exim now does this explicitly; it affects the formatting of
6204 timestamps using strftime().
6206 30. If exiqsumm was given junk data, it threw up some uninitialized variable
6207 complaints. I've now initialized all the variables, to avoid this.
6209 32. Header lines added by a system filter were not being "seen" during
6210 transport-time rewrites.
6212 33. The info_callback() function passed to OpenSSL is set up with type void
6213 (*)(SSL *, int, int), as described somewhere. However, when calling the
6214 function (actually a macro) that sets it up, the type void(*)() is
6215 expected. I've put in a cast to prevent warnings from picky compilers.
6217 34. If a DNS black list lookup found a CNAME record, but there were no A
6218 records associated with the domain it pointed at, Exim crashed.
6220 35. If a DNS black list lookup returned more than one A record, Exim ignored
6221 all but the first. It now scans all returned addresses if a particular IP
6222 value is being sought. In this situation, the contents of the
6223 $dnslist_value variable are a list of all the addresses, separated by a
6226 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
6227 to accept a host name and all its aliases if the forward lookup for any of
6228 them yielded the IP address of the incoming connection. Now it accepts only
6229 those names whose forward lookup yields the correct IP address. Any other
6230 names are discarded. This closes a loophole whereby a rogue DNS
6231 administrator could create reverse DNS records to break through a
6232 wildcarded host restriction in an ACL.
6234 37. If a user filter or a system filter that ran in a subprocess used any of
6235 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
6236 the wrong values were passed to the pipe command ($thisaddress had the
6237 value of $0, $0 had the value of $1, etc). This bug was introduced by
6238 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
6240 38. Improved the line breaking for long SMTP error messages from ACLs.
6241 Previously, if there was no break point between 40 and 75 characters, Exim
6242 left the rest of the message alone. Two changes have been made: (a) I've
6243 reduced the minimum length to 35 characters; (b) if it can't find a break
6244 point between 35 and 75 characters, it looks ahead and uses the first one
6245 that it finds. This may give the occasional overlong line, but at least the
6246 remaining text gets split now.
6248 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
6249 file descriptors might be low, and that setting 1000 would always raise it.
6250 It turns out that in some environments, the limit is already over 1000 and
6251 that lowering it causes trouble. So now Exim takes care not to decrease it.
6253 40. When delivering a message, the value of $return_path is set to $sender_
6254 address at the start of routing (routers may change the value). By an
6255 oversight, this default was not being set up when an address was tested by
6256 -bt or -bv, which affected the outcome if any router or filter referred to
6259 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
6260 log only when "message" or "log_message" are set. However, if one of the
6261 conditions was an address verification, or a call to a nested ACL, the
6262 messages generated by the underlying test were being passed through. This
6263 no longer happens. The underlying message is available in $acl_verify_
6264 message for both "message" and "log_message" expansions, so it can be
6265 passed through if needed.
6267 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
6268 new expansion $bh_ to give the encoded byte string without charset
6269 translation. Translation happens only if iconv() is available; HAVE_ICONV
6270 indicates this at build time. HEADERS_CHARSET gives the charset to
6271 translate to; headers_charset can change it in the configuration, and
6272 "headers charset" can change it in an individual filter file.
6274 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
6275 that creates RFC 2047 encoded "words" labels them as that charset instead
6276 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
6277 expansion operator; (ii) when Exim creates a From: line for a local
6278 message; (iii) when a header line is rewritten to include a "phrase" part.
6280 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
6281 buggy, causing it to skip the first lines of messages whose message ID
6282 ended in 'D'. This would not have bitten before Exim release 4.14, because
6283 message IDs were unlikely to end in 'D' before then. The effect was to have
6284 incorrect size information for certain domains.
6286 45. #include "config.h" was missing at the start of the crypt16.c module. This
6287 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
6290 46. If there was a timeout during a "random" callout check, Exim treated it as
6291 a failure of the random address, and carried on sending RSET and the real
6292 address. If the delay was just some slowness somewhere, the response to the
6293 original RCPT would be taken as a response to RSET and so on, causing
6294 mayhem of various kinds.
6296 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
6297 when I implemented it. It didn't allow for the fact that some option values
6298 may legitimatetly be negative (e.g. size_addition), and it didn't even do
6299 the right test for positive values.
6301 48. Domain names in DNS records are case-independent. Exim always looks them up
6302 in lower case. Some resolvers return domain names in exactly the case they
6303 appear in the zone file, that is, they may contain uppercase letters. Not
6304 all resolvers do this - some return always lower case. Exim was treating a
6305 change of case by a resolver as a change of domain, similar to a widening
6306 of a domain abbreviation. This triggered its re-routing code and so it was
6307 trying to route what was effectively the same domain again. This normally
6308 caused routing to fail (because the router wouldn't handle the domain
6309 twice). Now Exim checks for this case specially, and just changes the
6310 casing of the domain that it ultimately uses when it transmits the message
6313 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
6316 50. If a filter generated a file delivery with a non-absolute name (possible if
6317 no home directory exists for the router), the forbid_file option was not
6320 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
6321 the existing equality matching.
6323 52. Exim was using ints instead of ino_t variables in some places where it was
6324 dealing with inode numbers.
6326 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
6327 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
6328 called TMPDIR, and if it finds it is different, it changes its value.
6330 54. The smtp_printf() function is now made available to local_scan() so
6331 additional output lines can be written before returning. There is also an
6332 smtp_fflush() function to enable the detection of a dropped connection.
6333 The variables smtp_input and smtp_batched_input are exported to
6336 55. Changed the default runtime configuration: the message "Unknown user"
6337 has been removed from the ACL, and instead placed on the localuser router,
6338 using the cannot_route_message feature. This means that any verification
6339 failures that generate their own messages won't get overridden. Similarly,
6340 the "Unrouteable address" message that was in the ACL for unverifiable
6341 relay addresses has also been removed.
6343 56. Added hosts_avoid_esmtp to the smtp transport.
6345 57. The exicyclog script was not checking for the esoteric option
6346 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
6347 will work only if exicyclog is run under the appropriate euid.
6349 58. Following a discussion on the list, the rules by which Exim recognises line
6350 endings on incoming messages have been changed. The -dropcr and drop_cr
6351 options are now no-ops, retained only for backwards compatibility. The
6352 following line terminators are recognized: LF CRLF CR. However, special
6353 processing applies to CR:
6355 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
6356 nor a local message in the state where . is a terminator.
6358 (ii) If a bare CR is encountered in a header line, an extra space is added
6359 after the line terminator so as not to end the header. The reasoning
6360 behind this is that bare CRs in header lines are most likely either
6361 to be mistakes, or people trying to play silly games.
6363 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
6364 was being incorrectly given as 18 bytes larger than it should have been.
6365 This is a VOB (very old bug).
6367 60. This may never have affected anything current, but just in case it has:
6368 When the local host is found other than at the start of a list of hosts,
6369 the local host, those with the same MX, and any that follow, are discarded.
6370 When the list in question was part of a longer list of hosts, the following
6371 hosts (not currently being processed) were also being discarded. This no
6372 longer happens. I'm not sure if this situation could ever has previously
6375 61. Added the "/MX" feature to lists of hosts in the manualroute and query
6378 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
6379 header. This is something that is recommended in a new Internet Draft, and
6380 is something that is documented as being done by Sendmail. There are two
6381 possible values. For messages generated by the autoreply transport, Exim
6384 Auto-Submitted: auto-replied
6386 whereas for all other generated messages (e.g. bounces) it adds
6388 Auto-Submitted: auto-generated
6390 63. The "personal" condition in filters now includes a test for the
6391 Auto-Submitted: header. If it contains the string "auto-" the message it
6392 not considered personal.
6394 64. Added rcpt_include_affixes as a generic transport option.
6396 65. Added queue_only_override (default true).
6398 66. Added the syslog_duplication option.
6400 67. If what should have been the first header line of a message consisted of
6401 a space followed by a colon, Exim was mis-interpreting it as a header line.
6402 It isn't of course - it is syntactically invalid and should therefore be
6403 treated as the start of the message body. The misbehaviour could have
6404 caused a number of strange effects, including loss of data in subsequent
6405 header lines, and spool format errors.
6407 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
6408 client host had authenticated. This control can now be exercised by an ACL
6409 for more flexibility.
6411 69. By default, callouts do not happen when testing with -bh. There is now a
6412 variant, -bhc, which does actually run the callout code, including
6413 consulting and updating the callout cache.
6415 70. Added support for saslauthd authentication, courtesy of Alexander
6418 71. If statvfs() failed on the spool or log directories while checking their
6419 size for availability, Exim confusingly gave the error "space shortage".
6420 Furthermore, in debugging mode it crashed with a floating point exception.
6421 These checks are done if check_{spool,log}_{space,inodes} are set, and when
6422 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
6423 serious problem, Exim now writes to the main and panic logs when this
6424 happens, with details of the failure. It then refuses to accept the
6425 incoming message, giving the message "spool directory problem" or "log
6426 directory problem" with a 421 code for SMTP messages.
6428 72. When Exim is about to re-exec itself, it ensures that the file descriptors
6429 0, 1, and 2 exist, because some OS complain for execs without them (see
6430 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
6431 descriptors. However, the code omitted to check that the open succeeded,
6432 causing mysterious errors if for some reason the permissions on /dev/null
6433 got screwed. Now Exim writes a message to the main and panic logs, and
6434 bombs out if it can't open /dev/null.
6436 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
6437 interact so that it is all more flexible. It is supposed to remain
6438 backwards compatible. Also added extra_local_interfaces.
6440 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
6441 to bomb out with an assertion failure - to the client this appears as a
6442 connection drop. This problem occurs in the part of the code that was taken
6443 from the Samba project. Fortunately, the assertion is in a very simple
6444 function, so I have fixed this by reproducing the function inline in the
6445 one place where it is called, and arranging for authentication to fail
6446 instead of killing the process with assert().
6448 75. The SPA client code was not working when the server requested OEM rather
6449 than Unicode encoding.
6451 76. Added code to make require_files with a specific uid setting more usable in
6452 the case where statting the file as root fails - usually a non-root-mounted
6453 NFS file system. When this happens and the failure is EACCES, Exim now
6454 forks a subprocess and does the per-uid checking as the relevant uid.
6456 77. Added process_log_path.
6458 78. If log_file_path was not explicitly set, a setting of check_log_space or
6459 check_log_inodes was ignored.
6461 79. If a space check for the spool or log partitions fails, the incident is now
6462 logged. Of course, in the latter case the data may get lost...
6464 80. Added the %p formatting code to string_format() so that it can be used to
6465 print addresses in debug_print(). Adjusted all the address printing in the
6466 debugging in store.c to use %p rather than %d.
6468 81. There was a concern that a line of code in smtp_in.c could overflow a
6469 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
6470 initially expressed, the concern was not well-founded, because trailing
6471 spaces are removed early. However, if the trailing spaces were followed by
6472 a NULL, they did not get removed, so the overflow was possible. Two fixes
6475 (a) I re-wrote the offending code in a cleaner fashion.
6476 (b) If an incoming SMTP command contains a NULL character, it is rejected
6479 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
6480 runs initgroups(), so that if the Exim user is in any additional groups,
6481 they will be used during message reception.
6487 The change log for 4.20 and earlier releases has been archived.
projects / exim.git / blob