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 private-keys having trailing '=' in the base-64.
173 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
176 JH/02 The smtp transport option "multi_domain" is now expanded.
178 JH/03 The smtp transport now requests PRDR by default, if the server offers
181 JH/04 Certificate name checking on server certificates, when exim is a client,
182 is now done by default. The transport option tls_verify_cert_hostnames
183 can be used to disable this per-host. The build option
184 EXPERIMENTAL_CERTNAMES is withdrawn.
186 JH/05 The value of the tls_verify_certificates smtp transport and main options
187 default to the word "system" to access the system default CA bundle.
188 For GnuTLS, only version 3.0.20 or later.
190 JH/06 Verification of the server certificate for a TLS connection is now tried
191 (but not required) by default. The verification status is now logged by
192 default, for both outbound TLS and client-certificate supplying inbound
195 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
198 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
199 Status Notification (bounce) messages are now MIME format per RFC 3464.
200 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
201 under the control of the dsn_advertise_hosts option, and routers may
202 have a dsn_lasthop option.
204 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
205 default, modifiable by a malware= option. The list separator for
206 the options can now be changed in the usual way. Bug 68.
208 JH/10 The smtp_receive_timeout main option is now expanded before use.
210 JH/11 The incoming_interface log option now also enables logging of the
211 local interface on delivery outgoing connections.
213 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
214 if the interface and destination host and port all match.
216 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
219 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
220 Patch from Andrew Lewis.
222 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
223 now supports optional time-restrictions, weighting, and priority
224 modifiers per server. Patch originally by <rommer@active.by>.
226 JH/16 The spamd_address main option now supports a mixed list of local
227 and remote servers. Remote servers can be IPv6 addresses, and
228 specify a port-range.
230 JH/17 Bug 68: The spamd_address main option now supports an optional
231 timeout value per server.
233 JH/18 Bug 1581: Router and transport options headers_add/remove can
234 now have the list separator specified.
236 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
239 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
242 JH/21 Support for the A6 type of dns record is withdrawn.
244 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
245 rather than the verbs used.
247 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
248 from 255 to 1024 chars.
250 JH/24 Verification callouts now attempt to use TLS by default.
252 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
253 are generic router options now. The defaults didn't change.
255 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
256 Original patch from Alexander Shikoff, worked over by JH.
258 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
259 files if ZCAT_COMMAND is not executable.
261 JH/26 Bug 1539: Add timout/retry options on dnsdb lookups.
263 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
265 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
266 Normally benign, it bites when the pair was led to by a CNAME;
267 modern usage is to not canoicalize the domain to a CNAME target
268 (and we were inconsistent anyway for A-only vs AAAA+A).
270 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
272 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
273 when evaluating $sender_host_dnssec.
275 JH/31 Check the HELO verification lookup for DNSSEC, adding new
276 $sender_helo_dnssec variable.
278 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
280 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
282 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
284 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
285 documented as working, but never had. Support all but $spam_report.
287 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
288 added for tls authenticator.
293 TL/01 When running the test suite, the README says that variables such as
294 no_msglog_check are global and can be placed anywhere in a specific
295 test's script, however it was observed that placement needed to be near
296 the beginning for it to behave that way. Changed the runtest perl
297 script to read through the entire script once to detect and set these
298 variables, reset to the beginning of the script, and then run through
299 the script parsing/test process like normal.
301 TL/02 The BSD's have an arc4random API. One of the functions to induce
302 adding randomness was arc4random_stir(), but it has been removed in
303 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
304 function when detected.
306 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
307 cause callback expansion.
309 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
310 syntax errors in an expansion can be treated as a string instead of
311 logging or causing an error, due to the internal use of bool_lax
312 instead of bool when processing it.
314 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
315 server certificates when making smtp deliveries.
317 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
319 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
321 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
323 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
324 Merged patch from Sebastian Wiedenroth.
326 JH/05 Fix results-pipe from transport process. Several recipients, combined
327 with certificate use, exposed issues where response data items split
328 over buffer boundaries were not parsed properly. This eventually
329 resulted in duplicates being sent. This issue only became common enough
330 to notice due to the introduction of conection certificate information,
331 the item size being so much larger. Found and fixed by Wolfgang Breyha.
333 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
334 size buffer was used, resulting in syntax errors when an expansion
337 JH/07 Add support for directories of certificates when compiled with a GnuTLS
338 version 3.3.6 or later.
340 JH/08 Rename the TPDA expermimental facility to Event Actions. The #ifdef
341 is EXPERIMENTAL_EVENT, the main-configuration and transport options
342 both become "event_action", the variables become $event_name, $event_data
343 and $event_defer_errno. There is a new variable $verify_mode, usable in
344 routers, transports and related events. The tls:cert event is now also
345 raised for inbound connections, if the main configuration event_action
348 TL/06 In test suite, disable OCSP for old versions of openssl which contained
349 early OCSP support, but no stapling (appears to be less than 1.0.0).
351 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
352 server certificate names available under the smtp transport option
353 "tls_verify_cert_hostname" now do not permit multi-component wildcard
356 JH/10 Time-related extraction expansions from certificates now use the main
357 option "timezone" setting for output formatting, and are consistent
358 between OpenSSL and GnuTLS compilations. Bug 1541.
360 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
361 encoded parameter in the incoming message. Bug 1558.
363 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
364 include certificate info, eximon was claiming there were spoolfile
367 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
369 JH/14 Log delivery-related information more consistently, using the sequence
370 "H=<name> [<ip>]" wherever possible.
372 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
373 are problematic for Debian distribution, omit them from the release
376 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
378 JH/16 Fix string representation of time values on 64bit time_t anchitectures.
381 JH/17 Fix a null-indirection in certextract expansions when a nondefault
382 output list separator was used.
387 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
388 checkers that were complaining about end of non-void function with no
391 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
392 This was a regression intruduced in 4.83 by another bugfix.
394 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
396 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
397 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
403 TF/01 Correctly close the server side of TLS when forking for delivery.
405 When a message was received over SMTP with TLS, Exim failed to clear up
406 the incoming connection properly after forking off the child process to
407 deliver the message. In some situations the subsequent outgoing
408 delivery connection happened to have the same fd number as the incoming
409 connection previously had. Exim would try to use TLS and fail, logging
410 a "Bad file descriptor" error.
412 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
413 utilities have not been installed.
415 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
416 temporary space as the ACL may create new global variables.
418 TL/01 LDAP support uses per connection or global context settings, depending
419 upon the detected version of the libraries at build time.
421 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
422 to extract and use the src ip:port in logging and expansions as if it
423 were a direct connection from the outside internet. PPv2 support was
424 updated based on HAProxy spec change in May 2014.
426 JH/02 Add ${listextract {number}{list}{success}{fail}}.
428 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
429 Properly escape header and check for NULL return.
431 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
434 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
436 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
437 characters in header names, implemented as a verify condition.
438 Contributed by Michael Fischer v. Mollard.
440 TL/05 Rename SPF condition results err_perm and err_temp to standardized
441 results permerror and temperror. Previous values are deprecated but
442 still accepted. In a future release, err_perm and err_temp will be
443 completely removed, which will be a backward incompatibility if the
444 ACL tests for either of these two old results. Patch contributed by
445 user bes-internal on the mailing list.
447 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
449 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
450 selectors, in both main and reject logs.
452 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
455 JH/07 Add malware type "sock" for talking to simple daemon.
457 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
459 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
460 routers/transports under cutthrough routing.
462 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
463 numbers. Touch up "bool" conditional to keep the same definition.
465 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
467 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
469 JH/12 Expand items in router/transport headers_add or headers_remove lists
470 individually rather than the list as a whole. Bug 1452.
472 Required for reasonable handling of multiple headers_ options when
473 they may be empty; requires that headers_remove items with embedded
474 colons must have them doubled (or the list-separator changed).
476 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
477 view the policy declared in the DMARC record. Currently, $dmarc_status
478 is a combined value of both the record presence and the result of the
481 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
483 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
484 dnslookup router and the smtp transport (applying to the forward
487 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
488 of ldap servers used for a specific lookup. Patch provided by Heiko
491 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
492 New variable $lookup_dnssec_authenticated for observability.
494 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
495 Patch submitted by Lars Timman.
497 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
499 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
500 Requires trusted mode and valid format message id, aborts otherwise.
501 Patch contributed by Heiko Schlichting.
503 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
504 certextract with support for various fields. Bug 1358.
506 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
507 is requested by default, modifiable by smtp transport option
510 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
511 operate on certificate variables to give certificate fingerprints
512 Also new ${sha256:cert_variable}.
514 JH/23 The PRDR feature is moved from being Experimental into the mainline.
516 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
517 Christian Aistleitner.
519 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
521 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
522 file. Patch from Wolfgang Breyha.
524 JH/25 Expand the coverage of the delivery $host and $host_address to
525 client authenticators run in verify callout. Bug 1476.
527 JH/26 Port service names are now accepted for tls_on_connect_ports, to
528 align with daemon_smtp_ports. Bug 72.
530 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
531 support and error reporting did not work properly.
533 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
534 and is readable. Patch from Andrew Colin Kissa.
536 TL/14 Enhance documentation of ${run expansion and how it parses the
537 commandline after expansion, particularly in the case when an
538 unquoted variable expansion results in an empty value.
540 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
542 PP/02 Fix internal collision of T_APL on systems which support RFC3123
543 by renaming away from it. Addresses GH issue 15, reported by
546 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
548 TL/15 SECURITY: prevent double expansion in math comparison functions
549 (can expand unsanitized data). Not remotely exploitable.
556 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
558 PP/02 Make -n do something, by making it not do something.
559 When combined with -bP, the name of an option is not output.
561 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
564 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
565 $sender_host_name and config options to manage this, and basic check
568 PP/05 DSCP support for outbound connections and control modifier for inbound.
570 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
571 (Only plugin which currently uses this is kerberos4, which nobody should
572 be using, but we should make it available and other future plugins might
573 conceivably use it, even though it would break NAT; stuff *should* be
574 using channel bindings instead).
576 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
577 name; added for Sendmail compatibility; requires admin caller.
578 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
579 just ignore it); requires trusted caller.
580 Also parse but ignore: -Ac -Am -X<logfile>
583 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
585 TL/02 Add +smtp_confirmation as a default logging option.
587 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
588 Patch by Magnus Holmgren from 2007-02-20.
590 TL/04 Bugzilla 1281 - Spec typo.
591 Bugzilla 1283 - Spec typo.
592 Bugzilla 1290 - Spec grammar fixes.
594 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
596 TL/06 Add Experimental DMARC support using libopendmarc libraries.
598 TL/07 Fix an out of order global option causing a segfault. Reported to dev
599 mailing list by by Dmitry Isaikin.
601 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
603 JH/02 Support "G" suffix to numbers in ${if comparisons.
605 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
607 NM/01 Bugzilla 1197 - Spec typo
608 Bugzilla 1196 - Spec examples corrections
610 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
612 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
613 gnutls_enable_pkcs11, but renamed to more accurately indicate its
616 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
617 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
619 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
620 "acl {{name}{arg}...}", and optional args on acl condition
623 JH/05 Permit multiple router/transport headers_add/remove lines.
625 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
627 JH/07 Avoid using a waiting database for a single-message-only transport.
628 Performance patch from Paul Fisher. Bugzilla 1262.
630 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
633 JH/09 Add $headers_added variable, with content from use of ACL modifier
634 add_header (but not yet added to the message). Bugzilla 199.
636 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
637 Pulled from Bugzilla 817 by Wolfgang Breyha.
639 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
641 (nb: this is the same fix as in Exim 4.80.1)
643 JH/11 Add A= logging on delivery lines, and a client_set_id option on
646 JH/12 Add optional authenticated_sender logging to A= and a log_selector
649 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
651 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
652 advertise SMTP AUTH mechanism to us, instead of a generic
653 protocol violation error. Also, make Exim more robust to bad
654 data from the Dovecot auth socket.
656 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
658 When a queue runner is handling a message, Exim first routes the
659 recipient addresses, during which it prunes them based on the retry
660 hints database. After that it attempts to deliver the message to
661 any remaining recipients. It then updates the hints database using
664 So if a recipient address works intermittently, it can get repeatedly
665 deferred at routing time. The retry hints record remains fresh so the
666 address never reaches the final cutoff time.
668 This is a fairly common occurrence when a user is bumping up against
669 their storage quota. Exim had some logic in its local delivery code
670 to deal with this. However it did not apply to per-recipient defers
671 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
673 This change adds a proper retry rule check during routing so that the
674 final cutoff time is checked against the message's age. We only do
675 this check if there is an address retry record and there is not a
676 domain retry record; this implies that previous attempts to handle
677 the address had the retry_use_local_parts option turned on. We use
678 this as an approximation for the destination being like a local
679 delivery, as in LMTP.
681 I suspect this new check makes the old local delivery cutoff check
682 redundant, but I have not verified this so I left the code in place.
684 TF/02 Correct gecos expansion when From: is a prefix of the username.
686 Test 0254 submits a message to Exim with the header
690 When I ran the test suite under the user fanf2, Exim expanded
691 the header to contain my full name, whereas it should have added
692 a Resent-Sender: header. It erroneously treats any prefix of the
693 username as equal to the username.
695 This change corrects that bug.
697 GF/01 DCC debug and logging tidyup
698 Error conditions log to paniclog rather than rejectlog.
699 Debug lines prefixed by "DCC: " to remove any ambiguity.
701 TF/03 Avoid unnecessary rebuilds of lookup-related code.
703 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
704 Bug spotted by Jeremy Harris; was flawed since initial commit.
705 Would have resulted in OCSP responses post-SNI triggering an Exim
706 NULL dereference and crash.
708 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
710 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
711 Bug detection, analysis and fix by Samuel Thibault.
712 Bugzilla 1331, Debian bug #698092.
714 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
716 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
717 Server implementation by Todd Lyons, client by JH.
718 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
719 config variable "prdr_enable" controls whether the server
720 advertises the facility. If the client requests PRDR a new
721 acl_data_smtp_prdr ACL is called once for each recipient, after
722 the body content is received and before the acl_smtp_data ACL.
723 The client is controlled by bolth of: a hosts_try_prdr option
724 on the smtp transport, and the server advertisement.
725 Default client logging of deliveries and rejections involving
726 PRDR are flagged with the string "PRDR".
728 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
729 fclose(). Diagnosis by Todd Lyons.
731 PP/17 Update configure.default to handle IPv6 localhost better.
732 Patch by Alain Williams (plus minor tweaks).
735 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
736 This is now consistent with GnuTLS, and is now documented: the
737 previous undocumented portable approach to treating the option as
738 unset was to force an expansion failure. That still works, and
739 an empty string is now equivalent.
741 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
742 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
743 not performing validation itself.
745 PP/20 Added force_command boolean option to pipe transport.
746 Patch from Nick Koston, of cPanel Inc.
748 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
751 TF/04 Added udpsend ACL modifer and hexquote expansion operator
753 PP/21 Fix eximon continuous updating with timestamped log-files.
754 Broken in a format-string cleanup in 4.80, missed when I repaired the
755 other false fix of the same issue.
756 Report and fix from Heiko Schlichting.
759 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
760 Report from Prashanth Katuri.
762 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
763 It's SecureTransport, so affects any MacOS clients which use the
764 system-integrated TLS libraries, including email clients.
766 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
767 using a MIME ACL for non-SMTP local injection.
768 Report and assistance in diagnosis by Warren Baker.
770 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
772 JH/16 Fix comparisons for 64b. Bugzilla 1385.
774 TL/09 Add expansion variable $authenticated_fail_id to keep track of
775 last id that failed so it may be referenced in subsequent ACL's.
777 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
780 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
781 ldap library initialization, allowing self-signed CA's to be
782 used. Also properly sets require_cert option later in code by
783 using NULL (global ldap config) instead of ldap handle (per
784 session). Bug diagnosis and testing by alxgomz.
786 TL/12 Enhanced documentation in the ratelimit.pl script provided in
787 the src/util/ subdirectory.
789 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
790 renamed to Transport Post Delivery Action by Jeremy Harris, as
793 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
794 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
795 redis_servers = needs to be configured which will be used by the redis
796 lookup. Patch from Warren Baker, of The Packet Hub.
798 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
800 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
801 hostname or reverse DNS when processing a host list. Used suggestions
802 from multiple comments on this bug.
804 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
806 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
807 Missed a few lines, added it to make the runtest require no keyboard
810 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
811 contains upper case chars. Make router use caseful_local_part.
813 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
814 support when GnuTLS has been built with p11-kit.
820 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
822 This, or similar/improved, will also be change PP/11 of 4.82.
828 PP/01 Handle short writes when writing local log-files.
829 In practice, only affects FreeBSD (8 onwards).
830 Bugzilla 1053, with thanks to Dmitry Isaikin.
832 NM/01 Bugzilla 949 - Documentation tweak
834 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
837 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
839 PP/02 Implemented gsasl authenticator.
841 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
843 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
844 `pkg-config foo` for cflags/libs.
846 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
847 with rest of GSASL and with heimdal_gssapi.
849 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
850 `pkg-config foo` for cflags/libs for the TLS implementation.
852 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
853 properties get this fed in as external SSF. A number of robustness
854 and debugging improvements to the cyrus_sasl authenticator.
856 PP/08 cyrus_sasl server now expands the server_realm option.
858 PP/09 Bugzilla 1214 - Log authentication information in reject log.
859 Patch by Jeremy Harris.
861 PP/10 Added dbmjz lookup type.
863 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
865 PP/12 MAIL args handles TAB as well as SP, for better interop with
866 non-compliant senders.
867 Analysis and variant patch by Todd Lyons.
869 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
870 Bug report from Lars Müller <lars@samba.org> (via SUSE),
871 Patch from Dirk Mueller <dmueller@suse.com>
873 PP/13 tls_peerdn now print-escaped for spool files.
874 Observed some $tls_peerdn in wild which contained \n, which resulted
875 in spool file corruption.
877 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
878 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
879 or write after TLS renegotiation, which otherwise led to messages
882 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
883 as a tracking header (ie: a signed header comes before the signature).
884 Patch from Wolfgang Breyha.
886 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
887 comma-sep list; embedded commas doubled.
889 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
891 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
893 Report and patch from Dmitry Banschikov.
895 PP/16 Removed "dont_insert_empty_fragments" fron "openssl_options".
896 Removed SSL_clear() after SSL_new() which led to protocol negotiation
897 failures. We appear to now support TLS1.1+ with Exim.
899 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
900 lets Exim select keys and certificates based upon TLS SNI from client.
901 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
902 before an outbound SMTP session. New log_selector, +tls_sni.
904 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
905 NULL dereference. Report and patch from Alun Jones.
907 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
908 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
909 Not seeing resolver debug output on NetBSD, but suspect this is a
910 resolver implementation change.
912 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
913 Left warnings. Added "eximon gdb" invocation mode.
915 PP/21 Defaulting "accept_8bitmime" to true, not false.
917 PP/22 Added -bw for inetd wait mode support.
919 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
920 locate the relevant includes and libraries. Made this the default.
922 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
923 Bugzilla 1246, report and most of solution from Tomasz Kusy.
925 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
926 This may cause build issues on older platforms.
928 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
929 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
930 gnutls_require_mac & gnutls_require_protocols (no longer supported).
931 Added SNI support via GnuTLS too.
932 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
934 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
936 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
937 Applied second patch from Janne, implementing suggestion to default
938 multiple-strings-in-record handling to match SPF spec.
940 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
942 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
943 read-only, out of scope).
944 Patch from Wolfgang Breyha, report from Stuart Northfield.
946 PP/29 Fix three issues highlighted by clang analyser static analysis.
947 Only crash-plausible issue would require the Cambridge-specific
948 iplookup router and a misconfiguration.
949 Report from Marcin Mirosław.
951 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
953 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
954 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
955 As part of this, removing so much warning spew let me fix some minor
956 real issues in debug logging.
958 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
959 assignment on my part. Fixed.
961 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
962 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
963 Janne Snabb (who went above and beyond: thank you).
965 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
966 string otherwise requires a connection and a bunch more work and it's
967 relatively easy to get wrong. Should also expose TLS library linkage
970 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
971 64-bit ${eval} (JH/03).
973 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
974 GNU libc to support some of the 64-bit stuff, should not lead to
975 conflicts. Defined before os.h is pulled in, so if a given platform
976 needs to override this, it can.
978 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
979 protection layer was required, which is not implemented.
980 Bugzilla 1254, patch from Wolfgang Breyha.
982 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
983 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
984 tls_dhparam take prime identifiers. Also unbreak combination of
985 OpenSSL+DH_params+TLSSNI.
987 PP/39 Disable SSLv2 by default in OpenSSL support.
993 PP/01 Solaris build fix for Oracle's LDAP libraries.
994 Bugzilla 1109, patch from Stephen Usher.
996 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
998 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
1001 TF/02 Fix a couple more cases where we did not log the error message
1002 when unlink() failed. See also change 4.74-TF/03.
1004 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
1005 lock up or crash if it happened to be inside a call to libc when it
1006 got a SIGUSR1 from exiwhat.
1008 The SIGUSR1 handler appends the current process status to the process
1009 log which is later printed by exiwhat. It used to use the general
1010 purpose logging code to do this, but several functions it calls are
1011 not safe for signals.
1013 The new output code in the SIGUSR1 handler is specific to the process
1014 log, and simple enough that it's easy to inspect for signal safety.
1015 Removing some special cases also simplifies the general logging code.
1016 Removing the spurious timestamps from the process log simplifies
1019 TF/04 Improved ratelimit ACL condition.
1021 The /noupdate option has been deprecated in favour of /readonly which
1022 has clearer semantics. The /leaky, /strict, and /readonly update modes
1023 are mutually exclusive. The update mode is no longer included in the
1024 database key; it just determines when the database is updated. (This
1025 means that when you upgrde Exim will forget old rate measurements.)
1027 Exim now checks that the per_* options are used with an update mode that
1028 makes sense for the current ACL. For example, when Exim is processing a
1029 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
1030 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
1031 must specify per_mail/readonly. If you omit the update mode it defaults to
1032 /leaky where that makes sense (as before) or /readonly where required.
1034 The /noupdate option is now undocumented but still supported for
1035 backwards compatibility. It is equivalent to /readonly except that in
1036 ACLs where /readonly is required you may specify /leaky/noupdate or
1037 /strict/noupdate which are treated the same as /readonly.
1039 A useful new feature is the /count= option. This is a generalization
1040 of the per_byte option, so that you can measure the throughput of other
1041 aggregate values. For example, the per_byte option is now equivalent
1042 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
1044 The per_rcpt option has been generalized using the /count= mechanism
1045 (though it's more complicated than the per_byte equivalence). When it is
1046 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
1047 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
1048 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
1049 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
1050 Note that using per_rcpt with a non-readonly update mode in more than
1051 one ACL will cause the recipients to be double-counted. (The per_mail
1052 and per_byte options don't have this problem.)
1054 The handling of very low rates has changed slightly. If the computed rate
1055 is less than the event's count (usually one) then this event is the first
1056 after a long gap. In this case the rate is set to the same as this event's
1057 count, so that the first message of a spam run is counted properly.
1059 The major new feature is a mechanism for counting the rate of unique
1060 events. The new per_addr option counts the number of different
1061 recipients that someone has sent messages to in the last time period. It
1062 behaves like per_rcpt if all the recipient addresses are different, but
1063 duplicate recipient addresses do not increase the measured rate. Like
1064 the /count= option this is a general mechanism, so the per_addr option
1065 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
1066 example, measure the rate that a client uses different sender addresses
1067 with the options per_mail/unique=$sender_address. There are further
1068 details in the main documentation.
1070 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
1072 TF/06 Removed a few PCRE remnants.
1074 TF/07 Automatically extract Exim's version number from tags in the git
1075 repository when doing development or release builds.
1077 PP/02 Raise smtp_cmd_buffer_size to 16kB.
1078 Bugzilla 879. Patch from Paul Fisher.
1080 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
1081 Heavily based on revision 40f9a89a from Simon Arlott's tree.
1084 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
1086 PP/05 Variable $av_failed, true if the AV scanner deferred.
1087 Bugzilla 1078. Patch from John Horne.
1089 PP/06 Stop make process more reliably on build failure.
1090 Bugzilla 1087. Patch from Heiko Schlittermann.
1092 PP/07 Make maildir_use_size_file an _expandable_ boolean.
1093 Bugzilla 1089. Patch from Heiko Schlittermann.
1095 PP/08 Handle ${run} returning more data than OS pipe buffer size.
1096 Bugzilla 1131. Patch from Holger Weiß.
1098 PP/09 Handle IPv6 addresses with SPF.
1099 Bugzilla 860. Patch from Wolfgang Breyha.
1101 PP/10 GnuTLS: support TLS 1.2 & 1.1.
1103 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
1106 PP/11 match_* no longer expand right-hand-side by default.
1107 New compile-time build option, EXPAND_LISTMATCH_RHS.
1108 New expansion conditions, "inlist", "inlisti".
1110 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
1112 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
1114 PP/14 fix log_write() format string regression from TF/03.
1115 Bugzilla 1152. Patch from Dmitry Isaikin.
1121 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
1123 PP/02 Harmonised TLS library version reporting; only show if debugging.
1124 Layout now matches that introduced for other libraries in 4.74 PP/03.
1126 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
1128 PP/04 New "dns_use_edns0" global option.
1130 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
1133 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
1134 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
1136 TK/01 Updated PolarSSL code to 0.14.2.
1137 Bugzilla 1097. Patch from Andreas Metzler.
1139 PP/07 Catch divide-by-zero in ${eval:...}.
1140 Fixes bugzilla 1102.
1142 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
1145 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
1146 format-string attack -- SECURITY: remote arbitrary code execution.
1148 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
1149 time unintentionally subject to list matching rules, letting the header
1150 cause arbitrary Exim lookups (of items which can occur in lists, *not*
1151 arbitrary string expansion). This allowed for information disclosure.
1153 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
1154 INT_MIN/-1 -- value coerced to INT_MAX.
1160 NM/01 Workround for PCRE version dependency in version reporting
1163 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
1164 This fixes portability to compilers other than gcc, notably
1165 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
1167 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
1168 makefiles for portability to HP-UX and POSIX correctness.
1170 PP/01 Permit LOOKUP_foo enabling on the make command-line.
1171 Also via indented variable definition in the Makefile.
1172 (Debugging by Oliver Heesakkers).
1174 PP/02 Restore caching of spamd results with expanded spamd_address.
1175 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
1177 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
1178 Improves build reliability. Fix from: Frank Elsner
1180 NM/02 Fix wide character breakage in the rfc2047 coding
1181 Fixes bug 1064. Patch from Andrey N. Oktyabrski
1183 NM/03 Allow underscore in dnslist lookups
1184 Fixes bug 1026. Patch from Graeme Fowler
1186 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
1187 Code patches from Adam Ciarcinski of NetBSD.
1189 NM/04 Fixed exiqgrep to cope with mailq missing size issue
1192 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
1193 is logged, to avoid truncation. Patch from John Horne.
1195 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
1196 Patch from Jakob Hirsch.
1198 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
1199 SQL string expansion failure details.
1200 Patch from Andrey Oktyabrski.
1202 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
1203 Patch from Simon Arlott.
1205 PP/09 New lookups functionality failed to compile on old gcc which rejects
1206 extern declarations in function scope.
1207 Patch from Oliver Fleischmann
1209 PP/10 Use sig_atomic_t for flags set from signal handlers.
1210 Check getgroups() return and improve debugging.
1211 Fixed developed for diagnosis in bug 927 (which turned out to be
1214 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
1215 Patch from Mark Zealey.
1217 PP/12 Bugzilla 1056: Improved spamd server selection.
1218 Patch from Mark Zealey.
1220 PP/13 Bugzilla 1086: Deal with maildir quota file races.
1221 Based on patch from Heiko Schlittermann.
1223 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
1224 Patch from Uwe Doering, sign-off by Michael Haardt.
1226 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
1227 variable declaration deep within a block. Bug and patch from
1230 PP/15 lookups-Makefile IRIX compatibilty coercion.
1232 PP/16 Make DISABLE_DKIM build knob functional.
1234 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
1235 Patch by Simon Arlott
1237 TF/03 Fix valgrind.h portability to C89 compilers that do not support
1238 variable argument macros. Our copy now differs from upstream.
1244 TF/01 Failure to get a lock on a hints database can have serious
1245 consequences so log it to the panic log.
1247 TF/02 Log LMTP confirmation messages in the same way as SMTP,
1248 controlled using the smtp_confirmation log selector.
1250 TF/03 Include the error message when we fail to unlink a spool file.
1252 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
1253 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
1254 for maintaining out-of-tree patches for some time.
1256 PP/01 Bugzilla 139: Documentation and portability issues.
1257 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
1258 Handle per-OS dynamic-module compilation flags.
1260 PP/02 Let /dev/null have normal permissions.
1261 The 4.73 fixes were a little too stringent and complained about the
1262 permissions on /dev/null. Exempt it from some checks.
1263 Reported by Andreas M. Kirchwitz.
1265 PP/03 Report version information for many libraries, including
1266 Exim version information for dynamically loaded libraries. Created
1267 version.h, now support a version extension string for distributors
1268 who patch heavily. Dynamic module ABI change.
1270 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
1271 privilege escalation vulnerability whereby the Exim run-time user
1272 can cause root to append content of the attacker's choosing to
1275 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
1278 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
1279 If dropping privileges for untrusted macros, we disabled normal logging
1280 on the basis that it would fail; for the Exim run-time user, this is not
1281 the case, and it resulted in successful deliveries going unlogged.
1282 Fixed. Reported by Andreas Metzler.
1288 PP/01 Date: & Message-Id: revert to normally being appended to a message,
1289 only prepend for the Resent-* case. Fixes regression introduced in
1290 Exim 4.70 by NM/22 for Bugzilla 607.
1292 PP/02 Include check_rfc2047_length in configure.default because we're seeing
1293 increasing numbers of administrators be bitten by this.
1295 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
1297 PP/03 Bugzilla 994: added openssl_options main configuration option.
1299 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
1301 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
1303 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
1305 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
1306 without a peer certificate, leading to a segfault because of an
1307 assumption that peers always have certificates. Be a little more
1308 paranoid. Problem reported by Martin Tscholak.
1310 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
1311 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
1312 NB: ClamAV planning to remove STREAM in "middle of 2010".
1313 CL also introduces -bmalware, various -d+acl logging additions and
1314 more caution in buffer sizes.
1316 PP/09 Implemented reverse_ip expansion operator.
1318 PP/10 Bugzilla 937: provide a "debug" ACL control.
1320 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
1322 PP/12 Bugzilla 973: Implement --version.
1324 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
1326 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
1328 PP/15 Bugzilla 816: support multiple condition rules on Routers.
1330 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
1331 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
1332 ignore trailing whitespace.
1334 JJ/02 prevent non-panic DKIM error from being sent to paniclog
1336 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
1339 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
1340 Notification from Dr Andrew Aitchison.
1342 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
1343 ExtendedDetectionInfo response format.
1344 Notification from John Horne.
1346 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
1349 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
1350 XSL and documented dependency on system catalogs, with examples of how
1353 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
1356 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
1357 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
1358 configuration file which is writeable by the Exim user or group.
1360 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
1361 of configuration files to cover files specified with the -C option if
1362 they are going to be used with root privileges, not just the default
1365 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
1366 option (effectively making it always true).
1368 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
1369 files to be used while preserving root privileges.
1371 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
1372 that rogue child processes cannot use them.
1374 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
1375 run-time user, instead of root.
1377 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
1378 Exim run-time user without dropping privileges.
1380 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
1381 result string, instead of calling string_vformat() twice with the same
1384 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
1385 for other users. Others should always drop root privileges if they use
1386 -C on the command line, even for a whitelisted configure file.
1388 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
1390 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
1396 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
1397 $data_path, and $header_path variables; fixed documentation bugs and
1400 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
1401 exipick to access non-standard spools, including the "frozen" queue
1404 NM/01 Bugzilla 965: Support mysql stored procedures.
1405 Patch from Alain Williams
1407 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
1409 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
1410 Patch from Andreas Metzler
1412 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
1413 Patch from Kirill Miazine
1415 NM/05 Bugzilla 671: Added umask to procmail example.
1417 JJ/03 installed exipick 20100323.0, fixing doc bug
1419 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
1420 directory. Notification and patch from Dan Rosenberg.
1422 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
1424 TK/02 Improve log output when DKIM signing operation fails.
1426 MH/01 Treat the transport option dkim_domain as a colon separated
1427 list, not as a single string, and sign the message with each element,
1428 omitting multiple occurences of the same signer.
1430 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
1431 Bugzilla 985, 986. Patch by Simon Arlott
1433 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
1434 Patch by Simon Arlott
1436 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
1437 MBX locking. Notification from Dan Rosenberg.
1443 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
1445 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
1447 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
1449 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
1451 NM/04 Bugzilla 915: Flag broken perl installation during build.
1457 TK/01 Added patch by Johannes Berg that expands the main option
1458 "spamd_address" if it starts with a dollar sign.
1460 TK/02 Write list of recipients to X-Envelope-Sender header when building
1461 the mbox-format spool file for content scanning (suggested by Jakob
1464 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
1465 (http://www.dcc-servers.net/) support via dccifd. Activated by
1466 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
1468 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
1469 by Mark Daniel Reidel <mr@df.eu>.
1471 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
1472 When building exim an external PCRE library is now needed -
1473 PCRE is a system library on the majority of modern systems.
1474 See entry on PCRE_LIBS in EDITME file.
1476 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
1477 conversation. Added nologin parameter to request.
1478 Patch contributed by Kirill Miazine.
1480 TF/01 Do not log submission mode rewrites if they do not change the address.
1482 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
1484 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
1485 log files in place. Contributed by Roberto Lima.
1487 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
1489 TF/03 Bugzilla 615: When checking the local_parts router precondition
1490 after a local_part_suffix or local_part_prefix option, Exim now
1491 does not use the address's named list lookup cache, since this
1492 contains cached lookups for the whole local part.
1494 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
1495 Robert Millan. Documentation is in experimental-spec.txt.
1497 TF/04 Bugzilla 668: Fix parallel build (make -j).
1499 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
1501 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
1502 Patch provided by Jan Srzednicki.
1504 TF/05 Leading white space used to be stripped from $spam_report which
1505 wrecked the formatting. Now it is preserved.
1507 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
1508 that they are available at delivery time.
1510 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
1512 TF/08 TLS error reporting now respects the incoming_interface and
1513 incoming_port log selectors.
1515 TF/09 Produce a more useful error message if an SMTP transport's hosts
1516 setting expands to an empty string.
1518 NM/06 Bugzilla 744: EXPN did not work under TLS.
1519 Patch provided by Phil Pennock.
1521 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
1522 Patch provided by Richard Godbee.
1524 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
1525 acl_smtp_notquit, added index entry.
1527 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
1528 Patch provided by Eugene Bujak.
1530 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
1531 accept(). Patch provided by Maxim Dounin.
1533 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
1534 Patch provided by Phil Pennock.
1536 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
1538 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
1539 Patch provided by Brad "anomie" Jorsch.
1541 NM/14 Bugzilla 622: Added timeout setting to transport filter.
1542 Patch provided by Dean Brooks.
1544 TK/05 Add native DKIM support (does not depend on external libraries).
1546 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
1547 Patch provided by Graeme Fowler.
1549 NM/16 Bugzilla 851: Documentation example syntax fix.
1551 NM/17 Changed NOTICE file to remove references to embedded PCRE.
1553 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
1556 NM/19 Bugzilla 745: TLS version reporting.
1557 Patch provided by Phil Pennock.
1559 NM/20 Bugzilla 167: bool: condition support.
1560 Patch provided by Phil Pennock.
1562 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
1563 clients. Patch provided by Phil Pennock.
1565 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
1566 Patch provided by Brad "anomie" Jorsch.
1568 NM/23 Bugzilla 687: Fix misparses in eximstats.
1569 Patch provided by Heiko Schlittermann.
1571 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
1572 Patch provided by Heiko Schlittermann.
1574 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
1575 plus update to original patch.
1577 NM/26 Bugzilla 799: Documentation correction for ratelimit.
1579 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
1580 Patch provided by David Brownlee.
1582 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
1584 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
1586 NM/30 Bugzilla 888: TLS documentation bugfixes.
1588 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
1590 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
1591 Unlike the original bugzilla I have changed all shell scripts in src tree.
1593 NM/33 Bugzilla 898: Transport filter timeout fix.
1594 Patch by Todd Rinaldo.
1596 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
1597 Patch by Serge Demonchaux.
1599 NM/35 Bugzilla 39: Base64 decode bug fixes.
1600 Patch by Jakob Hirsch.
1602 NM/36 Bugzilla 909: Correct connect() call in dcc code.
1604 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
1606 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
1608 NM/39 Bugzilla 911: Fixed MakeLinks build script.
1614 TK/01 Add preliminary DKIM support. Currently requires a forked version of
1615 ALT-N's libdkim that I have put here:
1616 http://duncanthrax.net/exim-experimental/
1618 Note to Michael Haardt: I had to rename some vars in sieve.c. They
1619 were called 'true' and it seems that C99 defines that as a reserved
1620 keyword to be used with 'bool' variable types. That means you could
1621 not include C99-style headers which use bools without triggering
1622 build errors in sieve.c.
1624 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
1625 as mailq or other aliases. Changed the --help handling significantly
1626 to do whats expected. exim_usage() emits usage/help information.
1628 SC/01 Added the -bylocaldomain option to eximstats.
1630 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
1632 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
1634 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
1640 PH/01 Another patch from the Sieve maintainer.
1642 PH/02 When an IPv6 address is converted to a string for single-key lookup
1643 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
1644 dots are used instead of colons so that keys in lsearch files need not
1645 contain colons. This was done some time before quoting was made available
1646 in lsearch files. However, iplsearch files do require colons in IPv6 keys
1647 (notated using the quote facility) so as to distinguish them from IPv4
1648 keys. This meant that lookups for IP addresses in host lists did not work
1649 for iplsearch lookups.
1651 This has been fixed by arranging for IPv6 addresses to be expressed with
1652 colons if the lookup type is iplsearch. This is not incompatible, because
1653 previously such lookups could never work.
1655 The situation is now rather anomolous, since one *can* have colons in
1656 ordinary lsearch keys. However, making the change in all cases is
1657 incompatible and would probably break a number of configurations.
1659 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
1662 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
1663 conversion specification without a maximum field width, thereby enabling
1664 a rogue spamd server to cause a buffer overflow. While nobody in their
1665 right mind would setup Exim to query an untrusted spamd server, an
1666 attacker that gains access to a server running spamd could potentially
1667 exploit this vulnerability to run arbitrary code as the Exim user.
1669 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
1670 $primary_hostname instead of what libspf2 thinks the hosts name is.
1672 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
1673 a directory entry by the name of the lookup key. Previously, if a
1674 symlink pointed to a non-existing file or a file in a directory that
1675 Exim lacked permissions to read, a lookup for a key matching that
1676 symlink would fail. Now it is enough that a matching directory entry
1677 exists, symlink or not. (Bugzilla 503.)
1679 PH/03 The body_linecount and body_zerocount variables are now exported in the
1682 PH/04 Added the $dnslist_matched variable.
1684 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
1685 This means they are set thereafter only if the connection becomes
1688 PH/06 Added the client_condition to authenticators so that some can be skipped
1689 by clients under certain conditions.
1691 PH/07 The error message for a badly-placed control=no_multiline_responses left
1692 "_responses" off the end of the name.
1694 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
1696 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
1697 (without spaces) instead of just copying the configuration text.
1699 PH/10 Added the /noupdate option to the ratelimit ACL condition.
1701 PH/11 Added $max_received_linelength.
1703 PH/12 Added +ignore_defer and +include_defer to host lists.
1705 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
1706 way in which PCRE > 7.0 is built.
1708 PH/14 Implemented queue_only_load_latch.
1710 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
1711 MAIL command. The effect was to mangle the value on 64-bit systems.
1713 PH/16 Another patch from the Sieve maintainer.
1715 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
1717 PH/18 If a system quota error occurred while trying to create the file for
1718 a maildir delivery, the message "Mailbox is full" was not appended to the
1719 bounce if the delivery eventually timed out. Change 4.67/27 below applied
1720 only to a quota excession during the actual writing of the file.
1722 PH/19 It seems that peer DN values may contain newlines (and other non-printing
1723 characters?) which causes problems in log lines. The DN values are now
1724 passed through string_printing() before being added to log lines.
1726 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
1727 and InterBase are left for another time.)
1729 PH/21 Added message_body_newlines option.
1731 PH/22 Guard against possible overflow in moan_check_errorcopy().
1733 PH/23 POSIX allows open() to be a macro; guard against that.
1735 PH/24 If the recipient of an error message contained an @ in the local part
1736 (suitably quoted, of course), incorrect values were put in $domain and
1737 $local_part during the evaluation of errors_copy.
1743 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
1744 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
1747 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
1748 issue a MAIL command.
1750 PH/02 In an ACL statement such as
1752 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
1754 if a client was not listed at all, or was listed with a value other than
1755 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
1756 the condition was not true (as it should be), so access was not denied.
1757 The bug was that the ! inversion was incorrectly passed on to the second
1758 item. This has been fixed.
1760 PH/03 Added additional dnslists conditions == and =& which are different from
1761 = and & when the dns lookup returns more than one IP address.
1763 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
1764 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
1766 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
1767 FSYNC, which compiles an option called disable_fsync that allows for
1768 bypassing fsync(). The documentation is heavily laced with warnings.
1770 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
1772 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
1773 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
1774 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
1775 including adding "make clean"; (3) Added -fPIC when compiling the test
1776 dynamically loaded module, to get rid of a warning.
1778 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
1779 message fails, move_frozen_messages = true and ignore_bounce_errors_after
1780 = 0s. The bug is otherwise harmless.
1782 PH/07 There was a bug in the dovecot authenticator such that the value of
1783 $auth1 could be overwritten, and so not correctly preserved, after a
1784 successful authentication. This usually meant that the value preserved by
1785 the server_setid option was incorrect.
1787 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
1789 PH/09 Installed PCRE release 7.0.
1791 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
1792 run for batched SMTP input. It is now run at the start of every message
1793 in the batch. While fixing this I discovered that the process information
1794 (output by running exiwhat) was not always getting set for -bs and -bS
1795 input. This is fixed, and it now also says "batched" for BSMTP.
1797 PH/11 Added control=no_pipelining.
1799 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
1800 patch, slightly modified), and move the expansion of helo_data till after
1801 the connection is made in the smtp transport (so it can use these
1804 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
1806 PH/14 Added log_selector = +pid.
1808 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
1810 PH/16 Add ${if forany and ${if forall.
1812 PH/17 Added dsn_from option to vary the From: line in DSNs.
1814 PH/18 Flush SMTP output before performing a callout, unless control =
1815 no_callout_flush is set.
1817 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
1818 was true (the default) a successful delivery failed to delete the retry
1819 item, thus causing premature timeout of the address. The bug is now
1822 PH/20 Added hosts_avoid_pipelining to the smtp transport.
1824 PH/21 Long custom messages for fakedefer and fakereject are now split up
1825 into multiline reponses in the same way that messages for "deny" and
1826 other ACL rejections are.
1828 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
1829 with slight modification.
1831 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
1832 draft, changing the syntax and factoring some duplicate code".
1834 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
1835 for deliveries of the second and subsequent messages over the same SMTP
1838 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
1839 ${reduce, with only minor "tidies".
1841 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
1843 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
1844 expansion side effects.
1846 PH/27 When a message times out after an over-quota error from an Exim-imposed
1847 quota, the bounce message says "mailbox is full". This message was not
1848 being given when it was a system quota that was exceeded. It now should
1851 MH/03 Made $recipients available in local_scan(). local_scan() already has
1852 better access to the recipient list through recipients_list[], but
1853 $recipients can be useful in postmaster-provided expansion strings.
1855 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
1856 in the case of a MAIL command with additional options following the
1857 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
1858 were accidentally chopped off.
1860 PH/29 SMTP synchronization checks are implemented when a command is read -
1861 there is a check that no more input is waiting when there shouldn't be
1862 any. However, for some commands, a delay in an ACL can mean that it is
1863 some time before the response is written. In this time, more input might
1864 arrive, invalidly. So now there are extra checks after an ACL has run for
1865 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
1866 pipelining has not been advertised.
1868 PH/30 MH's patch to allow iscntrl() characters to be list separators.
1870 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
1871 returned in the SMTP response when smtp_return_error_details was false.
1872 This has been fixed.
1874 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
1875 instead of the C I/O that was originally supplied, because problems were
1876 reported on Solaris.
1878 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
1879 Exim which did not show up earlier: it was assuming that a call to
1880 SSL_CTX_set_info_callback() might give an error value. In fact, there is
1881 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
1882 was a macro that became an assignment, so it seemed to work. This has
1883 changed to a proper function call with a void return, hence the compile
1884 error. Exim's code has been fixed.
1886 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
1889 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
1891 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
1893 PH/37 If a message is not accepted after it has had an id assigned (e.g.
1894 because it turns out to be too big or there is a timeout) there is no
1895 "Completed" line in the log. When some messages of this type were
1896 selected by exigrep, they were listed as "not completed". Others were
1897 picked up by some special patterns. I have improved the selection
1898 criteria to be more general.
1900 PH/38 The host_find_failed option in the manualroute router can now be set
1901 to "ignore", to completely ignore a host whose IP address cannot be
1902 found. If all hosts are ignored, the behaviour is controlled by the new
1903 host_all_ignored option.
1905 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
1906 homing or because of multiple MX records with /mx) generated more than
1907 one IP address, and the following item turned out to be the local host,
1908 all the secondary addresses of the first item were incorrectly removed
1909 from the list, along with the local host and any following hosts (which
1910 is what is supposed to happen).
1912 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
1913 whoever called Exim into the -H file. In the case of the daemon it was
1914 behaving confusingly. When first started, it used values for whoever
1915 started the daemon, but after a SIGHUP it used the Exim user (because it
1916 calls itself on a restart). I have changed the code so that it now always
1919 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
1920 message are rejected with the same error (e.g. no authentication or bad
1921 sender address), and a DATA command is nevertheless sent (as can happen
1922 with PIPELINING or a stupid MUA), the error message that was given to the
1923 RCPT commands is included in the rejection of the DATA command. This is
1924 intended to be helpful for MUAs that show only the final error to their
1927 PH/42 Another patch from the Sieve maintainer.
1929 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
1930 Eximstats - Fixed some broken HTML links and added missing column headers
1932 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
1933 columns for Rejects, Temp Rejects, Ham, and Spam rows.
1935 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
1937 PH/43 Yet another patch from the Sieve maintainer.
1939 PH/44 I found a way to check for a TCP/IP connection going away before sending
1940 the response to the final '.' that terminates a message, but only in the
1941 case where the client has not sent further data following the '.'
1942 (unfortunately, this is allowed). However, in many cases there won't be
1943 any further data because there won't be any more messages to send. A call
1944 to select() can be used: if it shows that the input is "ready", there is
1945 either input waiting, or the socket has been closed. An attempt to read
1946 the next input character can distinguish the two cases. Previously, Exim
1947 would have sent an OK response which the client would never have see.
1948 This could lead to message repetition. This fix should cure that, at
1949 least in a lot of common cases.
1951 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
1952 advertised in response to EHLO.
1958 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
1959 fixed by 4.65/MH/01 (is this a record?) are fixed:
1961 (i) An empty string was always treated as zero by the numeric comparison
1962 operators. This behaviour has been restored.
1964 (ii) It is documented that the numeric comparison operators always treat
1965 their arguments as decimal numbers. This was broken in that numbers
1966 starting with 0 were being interpreted as octal.
1968 While fixing these problems I realized that there was another issue that
1969 hadn't been noticed. Values of message_size_limit (both the global option
1970 and the transport option) were treated as octal if they started with 0.
1971 The documentation was vague. These values are now always treated as
1972 decimal, and I will make that clear in the documentation.
1978 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
1979 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
1982 MH/01 Don't check that the operands of numeric comparison operators are
1983 integers when their expansion is in "skipping" mode (fixes bug
1984 introduced by 4.64-PH/07).
1986 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
1987 child addresses, Exim now panics and dies. Previously, because the count
1988 is held in a short int, deliveries were likely to be lost. As such a
1989 large number of recipients for a single message is ridiculous
1990 (performance will be very, very poor), I have chosen to impose a limit
1991 rather than extend the field.
1997 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
1998 leftover -K file (the existence of which was triggered by #402).
1999 While we were at it, introduced process PID as part of the -K
2000 filename. This should rule out race conditions when creating
2003 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
2004 processing considerably. Previous code took too long for large mails,
2005 triggering a timeout which in turn triggers #401.
2007 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
2008 in the DK code in transports.c. sendfile() is not really portable,
2009 hence the _LINUX specificness.
2011 TF/01 In the add_headers option to the mail command in an Exim filter,
2012 there was a bug that Exim would claim a syntax error in any
2013 header after the first one which had an odd number of characters
2016 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
2017 callout verification, Exim cached a "reject" for the entire domain. This
2018 is correct for most verifications, but it is not correct for a recipient
2019 verification with use_sender or use_postmaster set, because in that case
2020 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
2021 case of MAIL FROM:<> rejection from other early rejections (e.g.
2022 rejection of HELO). When verifying a recipient using a non-null MAIL
2023 address, the cache is ignored if it shows MAIL FROM:<> rejection.
2024 Whatever the result of the callout, the value of the domain cache is
2025 left unchanged (for any other kind of callout, getting as far as trying
2026 RCPT means that the domain itself is ok).
2028 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
2031 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
2032 manifest itself as EPIPE rather than ECONNECT. When tidying away a
2033 session, the daemon ignores ECONNECT errors and logs others; it now
2034 ignores EPIPE as well.
2036 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
2037 (quoted-printable decoding).
2039 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
2040 later the small subsequent patch to fix an introduced bug.
2042 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
2044 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
2046 PH/08 An error is now given if message_size_limit is specified negative.
2048 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
2049 to be given (somewhat) arbitrary names.
2051 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
2054 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
2055 miscellaneous code fixes
2057 PH/10 Added the log_reject_target ACL modifier to specify where to log
2060 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
2061 hostname. This is wrong, because it relates to the incoming message (and
2062 probably the interface on which it is arriving) and not to the outgoing
2063 callout (which could be using a different interface). This has been
2064 changed to use the value of the helo_data option from the smtp transport
2065 instead - this is what is used when a message is actually being sent. If
2066 there is no remote transport (possible with a router that sets up host
2067 addresses), $smtp_active_hostname is used.
2069 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
2070 tweaks were necessary in order to get it to work (see also 21 below):
2071 (a) The code assumed that strncpy() returns a negative number on buffer
2072 overflow, which isn't the case. Replaced with Exim's string_format()
2074 (b) There were several signed/unsigned issues. I just did the minimum
2075 hacking in of casts. There is scope for a larger refactoring.
2076 (c) The code used strcasecmp() which is not a standard C function.
2077 Replaced with Exim's strcmpic() function.
2078 (d) The code set only $1; it now sets $auth1 as well.
2079 (e) A simple test gave the error "authentication client didn't specify
2080 service in request". It would seem that Dovecot has changed its
2081 interface. Fortunately there's a specification; I followed it and
2082 changed what the client sends and it appears to be working now.
2084 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
2087 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
2088 address A is aliased to B and C, where B exists and C does not. Without
2089 -v the output is "A verified" because verification stops after a
2090 successful redirection if more than one address is generated. However,
2091 with -v the child addresses are also verified. Exim was outputting "A
2092 failed to verify" and then showing the successful verification for C,
2093 with its parentage. It now outputs "B failed to verify", showing B's
2094 parentage before showing the successful verification of C.
2096 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
2097 look up a TXT record in a specific list after matching in a combined
2100 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
2101 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
2102 they consult the DNS. I had assumed they would set it the way they
2103 wanted; and indeed my experiments on Linux seem to show that in some
2104 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
2105 To be on the safe side, however, I have now made the interface to
2106 host_find_byname() similar to host_find_bydns(), with an argument
2107 containing the DNS resolver options. The host_find_byname() function now
2108 sets these options at its start, just as host_find_bydns() does. The smtp
2109 transport options dns_qualify_single and dns_search_parents are passed to
2110 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
2111 of host_find_byname() use the default settings of RES_DEFNAMES
2112 (qualify_single) but not RES_DNSRCH (search_parents).
2114 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
2115 spool_read_header() do less string testing, by means of a preliminary
2116 switch on the second character of optional "-foo" lines. (This is
2117 overdue, caused by the large number of possibilities that now exist.
2118 Originally there were few.) While I was there, I also converted the
2119 str(n)cmp tests so they don't re-test the leading "-" and the first
2120 character, in the hope this might squeeze out yet more improvement.
2122 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
2123 flag allowing group syntax was set by the header_syntax check but not
2124 turned off, possible causing trouble later; (2) The flag was not being
2125 set at all for the header_verify test, causing "group"-style headers to
2126 be rejected. I have now set it in this case, and also caused header_
2127 verify to ignore an empty address taken from a group. While doing this, I
2128 came across some other cases where the code for allowing group syntax
2129 while scanning a header line wasn't quite right (mostly, not resetting
2130 the flag correctly in the right place). These bugs could have caused
2131 trouble for malformed header lines. I hope it is now all correct.
2133 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
2134 with the "reply" argument non-NULL. The code, however (which originally
2135 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
2136 but it didn't always do it. This confused somebody who was copying the
2137 code for some other use. I have removed all the tests.
2139 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
2140 feature that was used to support insecure browsers during the U.S. crypto
2141 embargo. It requires special client support, and Exim is probably the
2142 only MTA that supported it -- and would never use it because real RSA is
2143 always available. This code has been removed, because it had the bad
2144 effect of slowing Exim down by computing (never used) parameters for the
2145 RSA_EXPORT functionality.
2147 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
2148 authenticator to fail if there's a tab character in the incoming data
2149 (there should never be unless someone is messing about, as it's supposed
2150 to be base64-encoded). Also added, on Timo's advice, the "secured" option
2151 if the connection is using TLS or if the remote IP is the same as the
2152 local IP, and the "valid-client-cert option" if a client certificate has
2155 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
2156 authenticators. This can be used for authorization after authentication
2157 succeeds. (In the case of plaintext, it servers for both authentication
2160 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
2161 if any retry times were supplied.
2163 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
2164 connection, where there is no client IP address to check. In this
2165 situation, the verify now always succeeds.
2167 PH/25 Applied John Jetmore's -Mset patch.
2169 PH/26 Added -bem to be like -Mset, but loading a message from a file.
2171 PH/27 In a string expansion for a processed (not raw) header when multiple
2172 headers of the same name were present, leading whitespace was being
2173 removed from all of them, but trailing whitespace was being removed only
2174 from the last one. Now trailing whitespace is removed from each header
2175 before concatenation. Completely empty headers in a concatenation (as
2176 before) are ignored.
2178 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
2179 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
2181 PH/29 [Removed. This was a change that I later backed out, and forgot to
2182 correct the ChangeLog entry (that I had efficiently created) before
2183 committing the later change.]
2185 PH/30 Exim was sometimes attempting to deliver messages that had suffered
2186 address errors (4xx response to RCPT) over the same connection as other
2187 messages routed to the same hosts. Such deliveries are always "forced",
2188 so retry times are not inspected. This resulted in far too many retries
2189 for the affected addresses. The effect occurred only when there were more
2190 hosts than the hosts_max_try setting in the smtp transport when it had
2191 the 4xx errors. Those hosts that it had tried were not added to the list
2192 of hosts for which the message was waiting, so if all were tried, there
2193 was no problem. Two fixes have been applied:
2195 (i) If there are any address or message errors in an SMTP delivery, none
2196 of the hosts (tried or untried) are now added to the list of hosts
2197 for which the message is waiting, so the message should not be a
2198 candidate for sending over the same connection that was used for a
2199 successful delivery of some other message. This seems entirely
2200 reasonable: after all the message is NOT "waiting for some host".
2201 This is so "obvious" that I'm not sure why it wasn't done
2202 previously. Hope I haven't missed anything, but it can't do any
2203 harm, as the worst effect is to miss an optimization.
2205 (ii) If, despite (i), such a delivery is accidentally attempted, the
2206 routing retry time is respected, so at least it doesn't keep
2207 hammering the server.
2209 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
2210 in ${readsocket because some servers need this prod.
2212 PH/32 Added some extra debug output when updating a wait-xxx database.
2214 PH/33 The hint "could be header name not terminated by colon", which has been
2215 given for certain expansion errors for a long time, was not being given
2216 for the ${if def:h_colon_omitted{... case.
2218 PH/34 The spec says: "With one important exception, whenever a domain list is
2219 being scanned, $domain contains the subject domain." There was at least
2220 one case where this was not true.
2222 PH/35 The error "getsockname() failed: connection reset by peer" was being
2223 written to the panic log as well as the main log, but it isn't really
2224 panic-worthy as it just means the connection died rather early on. I have
2225 removed the panic log writing for the ECONNRESET error when getsockname()
2228 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
2229 runs only) independently of the message's sender address. This meant
2230 that, if the 4xx error was in fact related to the sender, a different
2231 message to the same recipient with a different sender could confuse
2232 things. In particualar, this can happen when sending to a greylisting
2233 server, but other circumstances could also provoke similar problems.
2234 I have changed the default so that the retry time for these errors is now
2235 based a combination of the sender and recipient addresses. This change
2236 can be overridden by setting address_retry_include_sender=false in the
2239 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
2240 remote server are returned as part of bounce messages. This was not
2241 happening for LMTP over a pipe (the lmtp transport), but now it is the
2242 same for both kinds of LMTP.
2244 PH/38 Despite being documented as not happening, Exim was rewriting addresses
2245 in header lines that were in fact CNAMEs. This is no longer the case.
2247 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
2248 and queue runs started by the daemon processed all messages. This has
2249 been fixed so that -R and -S can now usefully be given with -q<time>.
2251 PH/40 Import PCRE release 6.7 (fixes some bugs).
2253 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
2255 PH/42 Give an error if -q is specified more than once.
2257 PH/43 Renamed the variables $interface_address and $interface_port as
2258 $received_ip_address and $received_port, to make it clear that these
2259 values apply to message reception, and not to the outgoing interface when
2260 a message is delivered. (The old names remain recognized, of course.)
2262 PH/44 There was no timeout on the connect() call when using a Unix domain
2263 socket in the ${readsocket expansion. There now is.
2265 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
2266 be meaningful with "accept".
2268 SC/01 Eximstats V1.43
2269 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
2271 SC/02 Eximstats V1.44
2272 Use a glob alias rather than an array ref in the generated
2273 parser. This improves both readability and performance.
2275 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
2276 Collect SpamAssassin and rejection statistics.
2277 Don't display local sender or destination tables unless
2278 there is data to show.
2279 Added average volumes into the top table text output.
2281 SC/04 Eximstats V1.46
2282 Collect data on the number of addresses (recipients)
2283 as well as the number of messages.
2285 SC/05 Eximstats V1.47
2286 Added 'Message too big' to the list of mail rejection
2287 reasons (thanks to Marco Gaiarin).
2289 SC/06 Eximstats V1.48
2290 Mainlog lines which have GMT offsets and are too short to
2291 have a flag are now skipped.
2293 SC/07 Eximstats V1.49 (Alain Williams)
2294 Added the -emptyok flag.
2296 SC/08 Eximstats V1.50
2297 Fixes for obtaining the IP address from reject messages.
2299 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
2300 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
2301 whitesspace changes from 4.64-PH/27
2303 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
2306 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
2307 are found, allow negative numbers in numeric criteria)
2309 JJ/06 exipick.20061117.2, added new $message_body_missing variable
2311 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
2312 to match changes made in 4.64-PH/43
2314 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
2316 PH/47 Put in an explicit test for a DNS lookup of an address record where the
2317 "domain" is actually an IP address, and force a failure. This locks out
2318 those revolvers/nameservers that support "A-for-A" lookups, in
2319 contravention of the specifications.
2321 PH/48 When a host name was looked up from an IP address, and the subsequent
2322 forward lookup of the name timed out, the host name was left in
2323 $sender_host_name, contrary to the specification.
2325 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
2326 restricted to single-key lookups, Exim was not diagnosing an error if
2327 * or *@ was used with a query-style lookup.
2329 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
2331 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
2332 long ago, but noone interested enough thought of it. Let's just say that
2333 the "1.1" means that there are some new functions that weren't there at
2334 some point in the past.
2336 PH/51 Error processing for expansion failure of helo_data from an smtp
2337 transport during callout processing was broken.
2339 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
2340 tested/used via the -bh/-bhc/-bs options.
2342 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
2343 bug, fixed in subsequent PCRE releases).
2345 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
2346 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
2348 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
2354 SC/01 Use a glob alias rather than an array ref in eximstats generated
2355 parser. This improves both readability and performance.
2357 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
2358 Don't display local sender or destination tables in eximstats unless
2359 there is data to show.
2360 Added average volumes into the eximstats top table text output.
2362 SC/03 Collect data on the number of addresses (recipients) as well
2363 as the number of messages in eximstats.
2365 TF/01 Correct an error in the documentation for the redirect router. Exim
2366 does (usually) call initgroups() when daemonizing.
2368 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
2369 with consistent privilege compared to when running as a daemon.
2371 TF/03 Note in the spec that $authenticated_id is not set for local
2372 submissions from trusted users.
2374 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
2375 Thanks to Dean Brooks <dean@iglou.com> for the patch.
2377 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
2378 by adding some example configuration directives to the default
2379 configuration file. A little bit of work is required to uncomment the
2380 directives and define how usernames and passwords are checked, but
2381 there is now a framework to start from.
2383 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
2384 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
2385 without this. I don't know how relevant this is to other LDAP libraries.
2387 PH/02 Add the verb name to the "unknown ACL verb" error.
2389 PH/03 Magnus Holmgren's patch for filter_prepend_home.
2391 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
2393 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
2394 directory not expanded when it should be if an expanded home directory
2395 was set for the address (which is overridden by the transport).
2397 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
2400 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
2401 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
2402 because it is too late at that time, and has no effect.
2404 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
2405 security issue with \' (bugzilla #107). I could not use the
2406 PQescapeStringConn() function, because it needs a PGconn value as one of
2409 PH/08 When testing addresses using -bt, indicate those final addresses that
2410 are duplicates that would not cause an additional delivery. At least one
2411 person was confused, thinking that -bt output corresponded to deliveries.
2412 (Suppressing duplicates isn't a good idea as you lose the information
2413 about possibly different redirections that led to the duplicates.)
2415 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
2416 systems where poll() doesn't work, in particular OS X.
2418 PH/10 Added more information to debugging output for retry time not reached.
2420 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
2421 operations in malware.c.
2423 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
2426 PH/13 If write_rejectlog was set false when logging was sent to syslog with
2427 syslog_duplication set false, log lines that would normally be written
2428 both the the main log and to the reject log were not written to syslog at
2431 PH/14 In the default configuration, change the use of "message" in ACL warn
2432 statements to "add_header".
2434 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
2435 not followed by a command (e.g. "seen endif").
2437 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
2438 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
2441 PH/17 Added extra conditions to the default value of delay_warning_condition
2445 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
2446 { match{$h_precedence:}{(?i)bulk|list|junk} } \
2447 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
2450 The Auto-Submitted: and various List- headers are standardised, whereas I
2451 don't think Precedence: ever was.
2453 PH/18 Refactored debugging code in route_finduser() to show more information,
2454 in particular, the error code if getpwnam() issues one.
2456 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
2457 This is apparently needed in addition to the PH/07 change above to avoid
2458 any possible encoding problems.
2460 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
2461 but not after initializing Perl.
2463 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
2464 output them only if debugging. By default they are written stderr,
2465 apparently, which is not desirable.
2467 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
2470 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
2473 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
2475 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
2476 authenticated or an ident call has been made. Suppress the default
2477 values for $authenticated_id and $authenticated_sender (but permit -oMai
2478 and -oMas) when testing with -bh.
2480 PH/24 Re-jigged the order of the tests in the default configuration so that the
2481 tests for valid domains and recipients precede the DNS black list and CSA
2482 tests, on the grounds that those ones are more expensive.
2484 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
2485 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
2486 This bug exists in every version of Exim that I still have, right back to
2489 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
2490 However, an attempt to turn on case-sensitivity in a regex key by
2491 including (?-i) didn't work because the subject string was already
2492 lowercased, and the effects were non-intuitive. It turns out that a
2493 one-line patch can be used to allow (?-i) to work as expected.
2499 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
2500 other effects) broke the use of negated acl sub-conditions.
2502 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
2505 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
2506 "Deny" causes Exim to reject the incoming connection with a 554 error.
2507 Unfortunately, if there is a major crisis, such as a disk failure,
2508 tcp-wrappers gives "deny", whereas what one would like would be some
2509 kind of temporary error. A kludge has been added to help with this.
2510 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
2511 554 error is used if errno is still zero or contains ENOENT (which occurs
2512 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
2515 PH/03 Add -lutil to the default FreeBSD LIBS setting.
2517 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
2518 errors. Otherwise a message that provokes a temporary error (when other
2519 messages do not) can cause a whole host to time out.
2521 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
2522 addresses were routed directly to files or pipes from a redirect router.
2523 File deliveries just didn't batch; pipe deliveries might have suffered
2526 PH/06 A failure to get a lock for a hints database would erroneously always say
2527 "Failed to get write lock", even when it was really a read lock.
2529 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
2530 of 0600. This has been changed to use the value of the lockfile_mode
2531 option (which defaults to 0600).
2533 PH/08 Applied small patch from the Sieve maintainer.
2535 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
2536 folder from quota calculations, a direct delivery into this folder messed
2537 up the contents of the maildirsize file. This was because the regex was
2538 used only to exclude .Trash (or whatever) when the size of the mailbox
2539 was calculated. There was no check that a delivery was happening into an
2540 excluded directory. This bug has been fixed by ignoring all quota
2541 processing for deliveries into excluded directories.
2543 PH/10 Added the maildirfolder_create_regex option to appendfile.
2549 PH/01 The code for finding all the local interface addresses on a FreeBSD
2550 system running IPv6 was broken. This may well have applied to all BSD
2551 systems, as well as to others that have similar system calls. The broken
2552 code found IPv4 interfaces correctly, but gave incorrect values for the
2553 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
2554 that it would not match correctly against @[] and not recognize the IPv6
2557 PH/02 The ipliteral router was not recognizing addresses of the form user@
2558 [ipv6:....] because it didn't know about the "ipv6:" prefix.
2560 PH/03 Added disable_ipv6.
2562 PH/04 Changed $reply_address to use the raw form of the headers instead of the
2563 decoded form, because it is most often used to construct To: headers
2564 lines in autoreplies, and the decoded form may well be syntactically
2565 invalid. However, $reply_address has leading white space removed, and all
2566 newlines turned into spaces so that the autoreply transport does not
2569 PH/05 If group was specified without a user on a router, and no group or user
2570 was specified on a transport, the group from the router was ignored.
2572 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
2573 for visible compile-time settings that can be used to change these
2574 numbers, for those that want even more. Backwards compatibility with old
2575 spool files has been maintained. However, going back to a previous Exim
2576 release will lost any variables that are in spool files.
2578 PH/07 Two small changes when running in the test harness: increase delay when
2579 passing a TCP/IP connection to a new process, in case the original
2580 process has to generate a bounce, and remove special handling of
2581 127.0.0.2 (sic), which is no longer necessary.
2583 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
2584 be the same on different OS.
2586 PH/09 Moved a debug statement in filter processing to avoid a race problem when
2589 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
2590 whether --show-vars was specified or not
2592 JJ/02 exipick: Added support for new ACL variable spool format introduced
2595 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
2596 syntactically invalid From: or Reply-to: line, and a filter used this to
2597 generate an autoreply, and therefore failed to obtain an address for the
2598 autoreply, Exim could try to deliver to a non-existent relative file
2599 name, causing unrelated and misleading errors. What now happens is that
2600 it logs this as a hard delivery error, but does not attempt to create a
2603 PH/11 The exinext utility has a -C option for testing purposes, but although
2604 the given file was scanned by exinext itself; it wasn't being passed on
2605 when Exim was called.
2607 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
2608 an end-of-file indication when reading a command response.
2610 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
2611 compiled. In many other places in Exim, IPv6 addresses are always
2612 recognized, so I have changed this. It also means that IPv4 domain
2613 literals of the form [IPV4:n.n.n.n] are now always recognized.
2615 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
2616 used if the router is not running as root, for example, when verifying at
2617 ACL time, or when using -bh. The debugging output from this situation was
2618 non-existent - all you got was a failure to exec. I have made two
2621 (a) Failures to set uid/gid, the current directory, or a process leader
2622 in a subprocess such as that created by queryprogram now generate
2623 suitable debugging ouput when -d is set.
2625 (b) The queryprogram router detects when it is not running as root,
2626 outputs suitable debugging information if -d is set, and then runs
2627 the subprocess without attempting to change uid/gid.
2629 PH/15 Minor change to Makefile for building test_host (undocumented testing
2632 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
2633 additional section of a DNS packet that returns MX or SRV records.
2634 Instead, it always explicitly searches for A/AAAA records. This avoids
2635 major problems that occur when a DNS server includes only records of one
2636 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
2637 fixed another bug: if SRV records were looked up and the corresponding
2638 address records were *not* found in the additional section, the port
2639 values from the SRV records were lost.
2641 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
2642 using the correct key (the original address) when searching the retry
2643 rules in order to find which one to use for generating the retry hint.
2645 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
2646 adding the default one. Similarly, if it contains a Reply-To: header, the
2647 errors_reply_to option, if set, is not used.
2649 PH/19 When calculating a retry time, Exim used to measure the "time since
2650 failure" by looking at the "first failed" field in the retry record. Now
2651 it does not use this if it is later than than the arrival time of the
2652 message. Instead it uses the arrival time. This makes for better
2653 behaviour in cases where some deliveries succeed, thus re-setting the
2654 "first failed" field. An example is a quota failure for a huge message
2655 when small messages continue to be delivered. Without this change, the
2656 "time since failure" will always be short, possible causing more frequent
2657 delivery attempts for the huge message than are intended.
2658 [Note: This change was subsequently modified - see PH/04 for 4.62.]
2660 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
2661 $1, $2, $3) because the numerical variables can be reset during some
2662 expansion items (e.g. "match"), thereby losing the authentication data.
2664 PH/21 Make -bV show the size of off_t variables so that the test suite can
2665 decide whether to run tests for quotas > 2G.
2667 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
2668 mailbox_size, and mailbox_filecount in the appendfile transport. If a
2669 filecount value is greater than 2G or if a quota value is greater than 2G
2670 on a system where the size of off_t is not greater than 4, a panic error
2673 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
2674 never match. The debug and -bh output now contains an explicit error
2675 message indicating a malformed IPv4 address or mask.
2677 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
2678 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
2679 PH/23 above applies.
2681 PH/25 Do not write to syslog when running in the test harness. The only
2682 occasion when this arises is a failure to open the main or panic logs
2683 (for which there is an explicit test).
2685 PH/26 Added the /no_tell option to "control=freeze".
2687 PH/27 If a host name lookup failed very early in a connection, for example, if
2688 the IP address matched host_lookup and the reverse lookup yielded a name
2689 that did not have a forward lookup, an error message of the form "no IP
2690 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
2691 could be logged. Now it outputs the IP address instead of "NULL".
2693 PH/28 An enabling patch from MH: add new function child_open_exim2() which
2694 allows the sender and the authenticated sender to be set when
2695 submitting a message from within Exim. Since child_open_exim() is
2696 documented for local_scan(), the new function should be too.
2698 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
2699 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
2700 results in an empty string is now treated as unset.
2702 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
2704 PH/31 Added sender_verify_fail logging option.
2706 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
2707 needed by RFC 822 but not by RFC 2822 was commented out. I have now
2708 tidied the source and removed it altogether.
2710 PH/33 When a queue run was abandoned because the load average was too high, a
2711 log line was always written; now it is written only if the queue_run log
2712 selector is set. In addition, the log line for abandonment now contains
2713 information about the queue run such as the pid. This is always present
2714 in "start" and "stop" lines but was omitted from the "abandon" line.
2716 PH/34 Omit spaces between a header name and the colon in the error message that
2717 is given when verify = headers_syntax fails (if there are lots of them,
2718 the message gets confusing).
2720 PH/35 Change the default for dns_check_names_pattern to allow slashes within
2721 names, as there are now some PTR records that contain slashes. This check
2722 is only to protect against broken name servers that fall over on strange
2723 characters, so the fact that it applies to all lookups doesn't matter.
2725 PH/36 Now that the new test suite is complete, we can remove some of the
2726 special code in Exim that was needed for the old test suite. For example,
2727 sorting DNS records because real resolvers return them in an arbitrary
2728 order. The new test suite's fake resolver always returns records in the
2731 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
2732 bounces) except when queue_only is set, to avoid logging races between
2733 the different processes.
2735 PH/38 Panic-die if .include specifies a non-absolute path.
2737 PH/39 A tweak to the "H" retry rule from its user.
2739 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
2740 a label. They prevented compilation on older perls.
2742 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
2743 a warning to be raised on newish perls.
2745 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
2746 on queue. Changes to match documented behaviour of showing count of
2747 messages matching specified criteria.
2749 PH/40 Changed the default ident timeout from 30s to 5s.
2751 PH/41 Added support for the use of login_cap features, on those BSD systems
2752 that have them, for controlling the resources used by pipe deliveries.
2754 PH/42 The content-scanning code uses fopen() to create files in which to put
2755 message data. Previously it was not paying any attention to the mode of
2756 the files. Exim runs with umask(0) because the rest of the code creates
2757 files with open(), and sets the required mode explicitly. Thus, these
2758 files were ending up world-writeable. This was not a big issue, because,
2759 being within the spool directory, they were not world-accessible. I have
2760 created a function called modefopen, which takes an additional mode
2761 argument. It sets umask(777), creates the file, chmods it to the required
2762 mode, then resets the umask. All the relevant calls to fopen() in the
2763 content scanning code have been changed to use this function.
2765 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
2766 to 24 hours. This avoids potential overflow problems when processing G
2767 and H retry rules. I suspect nobody ever tinkers with this value.
2769 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
2771 PH/45 When the plaintext authenticator is running as a client, the server's
2772 challenges are checked to ensure they are valid base64 strings. By
2773 default, the authentication attempt is cancelled if an invalid string is
2774 received. Setting client_ignore_invalid_base64 true ignores these errors.
2775 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
2776 they are received. Thus, the responses can be made to depend on the
2777 challenges. If an invalid string is ignored, an empty string is placed in
2780 PH/46 Messages that are created by the autoreply transport now contains a
2781 References: header, in accordance with RFCs 2822 and 3834.
2783 PH/47 Added authenticated_sender_force to the smtp transport.
2785 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
2787 PH/49 Installed latest patch from the Sieve maintainer.
2789 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
2790 was also set, the appendfile transport was unnecessarily scanning a
2791 directory of message files (e.g. for maildir delivery) to find the count
2792 of files (along with the size), even though it did not need this
2793 information. It now does the scan only if it needs to find either the
2794 size of the count of files.
2796 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
2798 PH/52 Two bugs concerned with error handling when the smtp transport is
2801 (i) Exim was not creating retry information for temporary errors given
2802 for individual recipients after the DATA command when the smtp transport
2803 was used in LMTP mode. This meant that they could be retried too
2804 frequently, and not timed out correctly.
2806 (ii) Exim was setting the flag that allows error details to be returned
2807 for LMTP errors on RCPT commands, but not for LMTP errors for individual
2808 recipients that were returned after the DATA command.
2810 PH/53 This is related to PH/52, but is more general: for any failing address,
2811 when detailed error information was permitted to be returned to the
2812 sender, but the error was temporary, then after the final timeout, only
2813 "retry timeout exceeded" was returned. Now it returns the full error as
2814 well as "retry timeout exceeded".
2816 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
2817 do this, and (what is worse) MTAs that accept it.
2819 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
2820 will now be deprecated.
2822 PH/56 New os.c-cygwin from the Cygwin maintainer.
2824 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
2825 formats (previously only available in exim formats via -bpr, -bpru,
2826 and -bpra. Now also available in native and exiqgrep formats)
2828 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
2829 with very large, slow to parse queues
2831 JJ/08 exipick: added ! as generic prefix to negate any criteria format
2833 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
2835 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
2836 responses to authentication challenges, though it was showing the
2837 challenges; (ii) I've removed the CR characters from the debug output for
2840 PH/58 Allow for the insertion of a newline as well as a space when a string
2841 is turned into more than one encoded-word during RFC 2047 encoding. The
2842 Sieve code now uses this.
2844 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
2845 data_4xx, lost_connection, tls_required.
2847 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
2848 message was being sent as an SMTP response.
2850 PH/61 Add -l and -k options to exicyclog.
2852 PH/62 When verifying, if an address was redirected to one new address, so that
2853 verification continued, and the new address failed or deferred after
2854 having set something in $address_data, the value of $address_data was not
2855 passed back to the ACL. This was different to the case when no
2856 redirection occurred. The value is now passed back in both cases.
2858 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
2859 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
2860 use login_cap.h, so on its own it isn't the distinguishing feature. The
2861 new name refers directly to the setclassresources() function.
2863 PH/65 Added configuration files for NetBSD3.
2865 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
2867 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
2868 is preferred over IPv4.
2870 PH/68 The bounce_return_message and bounce_return_body options were not being
2871 honoured for bounces generated during the reception of non-SMTP messages.
2872 In particular, this applied to messages rejected by the ACL. This bug has
2873 been fixed. However, if bounce_return_message is true and bounce_return_
2874 body is false, the headers that are returned for a non-SMTP message
2875 include only those that have been read before the error was detected.
2876 (In the case of an ACL rejection, they have all been read.)
2878 PH/69 The HTML version of the specification is now built in a directory called
2879 spec_html instead of spec.html, because the latter looks like a path with
2880 a MIME-type, and this confuses some software.
2882 PH/70 Catch two compiler warnings in sieve.c.
2884 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
2885 function verify_get_ident() calls ip_connect() to connect a socket, but
2886 if the "connect()" function timed out, ip_connect() used to close the
2887 socket. However, verify_get_ident() also closes the socket later, and in
2888 between Exim writes to the log, which may get opened at this point. When
2889 the socket was closed in ip_connect(), the log could get the same file
2890 descriptor number as the socket. This naturally causes chaos. The fix is
2891 not to close the socket in ip_connect(); the socket should be closed by
2892 the function that creates it. There was only one place in the code where
2893 this was missing, in the iplookup router, which I don't think anybody now
2894 uses, but I've fixed it anyway.
2896 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
2897 well as to direct DNS lookups. Otherwise the handling of names in host
2898 lists is inconsistent and therefore confusing.
2904 PH/01 Two changes to the default runtime configuration:
2906 (1) Move the checks for relay_from_hosts and authenticated clients from
2907 after to before the (commented out) DNS black list checks.
2909 (2) Add control=submission to the relay_from_hosts and authenticated
2910 clients checks, on the grounds that messages accepted by these
2911 statements are most likely to be submissions.
2913 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
2915 (1) Generate an error if the third argument for the ${prvs expansion is
2918 (2) Treat a missing third argument of ${prvscheck as if it were an empty
2921 (3) Reset the variables that are obtained from the first argument of
2922 ${prvscheck and used in the second argument before leaving the code,
2923 because their memory is reclaimed, so using them afterwards may do
2926 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
2927 one (it's much easier than Tom thought :-).
2929 (5) Because of (4), we can now allow for the use of $prvscheck_result
2930 inside the third argument.
2932 PH/03 For some reason, the default setting of PATH when running a command from
2933 a pipe transport was just "/usr/bin". I have changed it to
2936 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
2937 anything to be listed in the output from -bV.
2939 PH/05 When a filter generated an autoreply, the entire To: header line was
2940 quoted in the delivery log line, like this:
2942 => >A.N.Other <ano@some.domain> <original@ddress> ...
2944 This has been changed so that it extracts the operative address. There
2945 may be more than one such address. If so, they are comma-separated, like
2948 => >ano@some.domain,ona@other.domain <original@ddress> ...
2950 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
2951 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
2952 not being looked up in the rDNS to get a host name, Exim was showing the
2953 IP address twice in Received: lines, even though the IP addresses were
2954 identical. For example:
2956 Received: from [1.2.3.4] (helo=[1.2.3.4])
2958 However, if the real host name was known, it was omitting the HELO data
2959 if it matched the actual IP address. This has been tidied up so that it
2960 doesn't show the same IP address twice.
2962 PH/07 When both +timestamp and +memory debugging was on, the value given by
2963 $tod_xxx expansions could be wrong, because the tod_stamp() function was
2964 called by the debug printing, thereby overwriting the timestamp buffer.
2965 Debugging no longer uses the tod_stamp() function when +timestamp is set.
2967 PH/08 When the original message was included in an autoreply transport, it
2968 always said "this is a copy of the message, including all the headers",
2969 even if body_only or headers_only was set. It now gives an appropriate
2972 PH/09 Applied a patch from the Sieve maintainer which:
2974 o fixes some comments
2975 o adds the (disabled) notify extension core
2976 o adds some debug output for the result of if/elsif tests
2977 o points to the current vacation draft in the documentation
2978 and documents the missing references header update
2982 o fixes a bug in processing the envelope test (when testing
2983 multiple envelope elements, the last element determinted the
2986 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
2987 Electronic Mail") by including:
2989 Auto-submitted: auto-generated
2991 in the messages that it generates (bounce messages and others, such as
2992 warnings). In the case of bounce messages for non-SMTP mesages, there was
2993 also a typo: it was using "Auto_submitted" (underscore instead of
2994 hyphen). Since every message generated by Exim is necessarily in response
2995 to another message, thes have all been changed to:
2997 Auto-Submitted: auto-replied
2999 in accordance with these statements in the RFC:
3001 The auto-replied keyword:
3003 - SHOULD be used on messages sent in direct response to another
3004 message by an automatic process,
3006 - MUST NOT be used on manually-generated messages,
3008 - MAY be used on Delivery Status Notifications (DSNs) and Message
3009 Disposition Notifications (MDNs),
3011 - MUST NOT be used on messages generated by automatic or periodic
3012 processes, except for messages which are automatic responses to
3015 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
3016 to the default Received: header definition.
3018 PH/12 Added log selector acl_warn_skipped (default on).
3020 PH/13 After a successful wildlsearch lookup, discard the values of numeric
3021 variables because (a) they are in the wrong storage pool and (b) even if
3022 they were copied, it wouldn't work properly because of the caching.
3024 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
3025 checking when decoding. Apparently there are clients that generate
3026 overlong encoded strings. Why am I not surprised?
3028 PH/15 If the first argument of "${if match_address" was not empty, but did not
3029 contain an "@" character, Exim crashed. Now it writes a panic log message
3030 and treats the condition as false.
3032 PH/16 In autoreply, treat an empty string for "once" the same as unset.
3034 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
3035 extension "envelope-auth". The code is finished and in agreement with
3036 other implementations, but there is no documentation so far and in fact,
3037 nobody wrote the draft yet. This extension is currently #undef'ed, thus
3038 not changing the active code.
3040 Print executed "if" and "elsif" statements when debugging is used. This
3041 helps a great deal to understand what a filter does.
3043 Document more things not specified clearly in RFC3028. I had all this
3044 sorted out, when out of a sudden new issues came to my mind. Oops."
3046 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
3049 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
3050 canonical form (as documented). However, after a host name lookup from
3051 the IP address, check_host() was doing a simple string comparison with
3052 addresses acquired from the DNS when checking that the found name did
3053 have the original IP as one of its addresses. Since any found IPv6
3054 addresses are likely to be in abbreviated form, the comparison could
3055 fail. Luckily, there already exists a function for doing the comparison
3056 by converting both addresses to binary, so now that is used instead of
3057 the text comparison.
3059 PH/20 There was another similar case to PH/19, when a complete host name was
3060 given in a host list; looking up its IP address could give an abbreviated
3061 form, whereas the current host's name might or might not be abbreviated.
3062 The same fix has been applied.
3068 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
3069 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
3072 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
3073 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
3075 PH/03 Typo: missing ".o" in src/pcre/Makefile.
3077 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
3078 header line, restrict the check to what is listed in RFCs 2369 and 2929.
3079 Also, for "Auto-Submitted", treat anything other than "no" as
3080 non-personal, in accordance with RFC 3834. (Previously it treated
3081 anything starting "auto-" as non-personal.)
3083 TF/01 The control=submission/name=... option had a problem with syntax
3084 errors if the name included a slash character. The /name= option
3085 now slurps the rest of the string, so it can include any characters
3086 but it must come last in the list of options (after /sender_retain
3089 PH/05 Some modifications to the interface to the fake nameserver for the new
3097 TK/01 Added the "success_on_redirect" address verification option. See
3098 NewStuff for rationale and an example.
3100 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
3102 PH/02 Patch to exigrep to allow it to work on syslog lines.
3104 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
3105 fread() to skip over the body file's header line, because in Cygwin the
3106 header line is locked and is inaccessible.
3108 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
3109 co-exist for some time) to make it clear that it is the Exim ID that is
3110 referenced, not the Message-ID: header line.
3112 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
3113 string_format() function, because snprintf() does not exist on all
3116 PH/06 The use of forbid_filter_existstest now also locks out the use of the
3117 ${stat: expansion item.
3119 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
3120 protocol synchronization error", to keep the pedants happy.
3122 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
3123 well as for IRIX systems, when gcc is being used. See the host.c source
3126 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
3128 PH/10 Named domain lists were not working if used in a queue_smtp_domains
3131 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
3132 transport and to the smtp transport in LMTP mode.
3134 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
3136 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
3137 run a filter in a subprocess. This could lead to confusion in subsequent
3138 lookups in the parent process. There should also be a search_tidyup() at
3139 the end of the subprocess.
3141 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
3142 only if the host matched helo_try_verify_hosts, which caused the
3143 verification to occur when the EHLO/HELO command was issued. The ACL just
3144 tested the remembered result. Now, if a previous verification attempt has
3145 not happened, "verify = helo" does it there and then.
3147 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
3149 TK/03 Fix log output including CR from clamd.
3151 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
3152 exist provoked a memory error which could cause a segfault.
3154 PH/15 Installed PCRE 6.2
3156 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
3158 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
3159 of the problem. Specifically, suggested +O2 rather than +O1 for the
3162 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
3164 PH/20 If a delivery was routed to a non-standard port by means of an SRV
3165 record, the port was not correctly logged when the outgoing_port log
3166 selector was set (it logged the transort's default port).
3168 PH/21 Added support for host-specific ports to manualroute, queryprogram,
3169 fallback_hosts, and "hosts" in the smtp transport.
3171 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
3172 host errors such as "Connection refused".
3174 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
3175 authentication with radiusclient 0.4.9:
3177 - Error returned from rc_read_config was caught wrongly
3178 - Username/password not passed on to radius server due to wrong length.
3180 The presumption is that some radiusclient API changes for 4.51/PH/17
3181 were not taken care of correctly. The code is still untested by me (my
3182 Linux distribution still has 0.3.2 of radiusclient), but it was
3183 contributed by a Radius user.
3185 PH/24 When doing a callout, the value of $domain wasn't set correctly when
3186 expanding the "port" option of the smtp transport.
3188 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
3189 while reading a MIME header. Thanks to Tom Hughes for a patch.
3191 PH/24 Include config.h inside local_scan.h so that configuration settings are
3194 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
3195 that in an ACL for RCPT (for example), you can examine exactly what was
3198 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
3199 commands, but it was not correctly comparing the address with the actual
3200 client host address. Thus, it would show the EHLO address in Received:
3201 header lines when this was not necessary.
3203 PH/27 Added the % operator to ${eval:}.
3205 PH/28 Exim tries to create and chdir to its spool directory when it starts;
3206 it should be ignoring failures (because with -C, for example, it has lost
3207 privilege). It wasn't ignoring creation failures other than "already
3210 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
3213 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
3214 because an input line was too long, either on its own, or by virtue of
3215 too many continuations, the temporary file was not being removed, and the
3216 return code was incorrect.
3218 PH/31 Missing "BOOL" in function definition in filtertest.c.
3220 PH/32 Applied Sieve patches from the maintainer.
3222 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
3224 PH/33 Added "verify = not_blind".
3226 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
3227 Local/Makefile (with some defaults set). These are used in built scripts
3228 such as exicyclog, but they have never been used in the exim_install
3229 script (though there are many overriding facilities there). I have
3230 arranged that the exim_install script now takes note of these two
3233 PH/35 Installed configuration files for Dragonfly.
3235 PH/36 When a locally submitted message by a trusted user did not contain a
3236 From: header, and the sender address was obtained from -f or from an SMTP
3237 MAIL command, and the trusted user did not use -F to supply a sender
3238 name, $originator_name was incorrectly used when constructing a From:
3239 header. Furthermore, $originator_name was used for submission mode
3240 messages from external hosts without From: headers in a similar way,
3241 which is clearly wrong.
3243 PH/37 Added control=suppress_local_fixups.
3245 PH/38 When log_selector = +received_sender was set, and the addition of the
3246 sender made the log line's construction buffer exactly full, or one byte
3247 less than full, an overflow happened when the terminating "\n" was
3250 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
3251 when the result of a list match is failure because a DNS lookup failed.
3253 PH/40 RM_COMMAND is now used in the building process.
3255 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
3256 the "build-* directories that it finds.
3258 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
3259 address in a domain literal was a prefix of an interface address.
3261 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
3262 when verifying a sender address, unless rewrite_headers is false.
3264 PH/44 Wrote a long comment about why errors_to addresses are verified as
3265 recipients, not senders.
3267 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
3268 the ratelimit ACL was added.
3270 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
3272 PH/46 Added extra information about PostgreSQL errors to the error string.
3274 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
3275 avoiding the need to install special zones in a real server. This is
3276 backwards compatible; if it can't find the fake resolver, it drops back.
3277 Thus, both old and new test suites can be run.
3279 TF/02 Added util/ratelimit.pl
3281 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
3282 clock is set back in time.
3284 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
3285 Candler <B.Candler@pobox.com>.
3287 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
3288 OK for addresses that are the result of redirections.
3290 PH/48 A number of further additions for the benefit of the new test suite,
3291 including a fake gethostbyname() that interfaces to the fake DNS resolver
3294 TF/06 The fix for widen_domains has also been applied to qualify_single and
3295 search_parents which are the other dnslookup options that can cause
3298 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
3301 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
3303 TF/07 Exim produced the error message "an SRV record indicated no SMTP
3304 service" if it encountered an MX record with an empty target hostname.
3305 The message is now "an MX or SRV record indicated no SMTP service".
3307 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
3308 if the DNS of the sending site is misconfigured. This is quite a
3309 common situation. This change restores the behaviour of treating a
3310 helo verification defer as a failure.
3312 PH/51 If self=fail was set on a router, the bounce message did not include the
3313 actual error message.
3319 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
3321 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
3322 command itself ended in error, the underlying message about the transport
3323 filter timeout was being overwritten with the pipe command error. Now the
3324 underlying error message should be appended to the second error message.
3326 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
3328 PH/02 Reduce the amount of output that "make" produces by default. Full output
3329 can still be requested.
3331 PH/03 The warning log line about a condition test deferring for a "warn" verb
3332 was being output only once per connection, rather than after each
3333 occurrence (because it was using the same function as for successful
3334 "warn" verbs). This seems wrong, so I have changed it.
3336 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
3337 it should not have, which might have caused a crash in the right
3338 circumstances, but probably never did.
3340 PH/04 Installed a modified version of Tony Finch's patch to make submission
3341 mode fix the return path as well as the Sender: header line, and to
3342 add a /name= option so that you can make the user's friendly name appear
3345 TF/03 Added the control = fakedefer ACL modifier.
3347 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
3348 Mark Lowes for thorough testing.
3350 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
3352 TK/03 Merged latest SRS patch from Miles Wilton.
3354 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
3355 with the definition in sysexits.h (which is #included earlier).
3356 Fortunately, Exim does not actually use EX_OK. The code used to try to
3357 preserve the sysexits.h value, by assumimg that macro definitions were
3358 scanned for macro replacements. I have been disabused of this notion,
3359 so now the code just undefines EX_OK before #including unistd.h.
3361 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
3362 in the smtp transport. When a block could not be written in a single
3363 write() function, the timeout was being re-applied to each part-write.
3364 This seems wrong - if the receiver was accepting one byte at a time it
3365 would take for ever. The timeout is now adjusted when this happens. It
3366 doesn't have to be particularly precise.
3368 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
3369 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
3371 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
3372 without a domain if the check to <postmaster@domain> fails.
3374 SC/01 Eximstats: added -xls and the ability to specify output files
3375 (patch written by Frank Heydlauf).
3377 SC/02 Eximstats: use FileHandles for outputing results.
3379 SC/03 Eximstats: allow any combination of xls, txt, and html output.
3381 SC/04 Eximstats: fixed display of large numbers with -nvr option
3383 SC/05 Eximstats: fixed merging of reports with empty tables.
3385 SC/06 Eximstats: added the -include_original_destination flag
3387 SC/07 Eximstats: removed tabs and trailing whitespace.
3389 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
3391 TK/06 MBOX spool code: Add real "From " MBOX separator line
3392 so the .eml file is really in mbox format (even though
3393 most programs do not really care). Patch from Alex Miller.
3395 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
3396 The latter is generated from $received_to and is only set if the
3397 message has one envelope recipient. SA can use these headers,
3398 obviously out-of-the-box. Patch from Alex Miller.
3400 PH/08 The ${def test on a variable was returning false if the variable's
3401 value was "0", contrary to what the specification has always said!
3402 The result should be true unless the variable is empty.
3404 PH/09 The syntax error of a character other than { following "${if
3405 def:variable_name" (after optional whitespace) was not being diagnosed.
3406 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
3407 accidental colon was present, for example, could give incorrect results.
3409 PH/10 Tidied the code in a number of places where the st_size field of a stat()
3410 result is used (not including appendfile, where other changes are about
3413 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
3414 This involved changing a lot of size variables from int to off_t. It
3415 should work with maildirs and everything.
3417 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
3418 spamd dying while we are connected to it.
3420 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
3423 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
3424 function for BDB 4.3.
3426 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
3428 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
3429 the sources for PCRE have been re-arranged and more files are now
3432 PH/15 The code I had for printing potentially long long variables in PH/11
3433 above was not the best (it lost precision). The length of off_t variables
3434 is now inspected at build time, and an appropriate printing format (%ld
3435 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
3436 to be "long long int" or "long int". This is needed for the internal
3437 formatting function string_vformat().
3439 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
3440 the configuration file to be ":syslog", then the script "guesses" where
3441 the logs files are, rather than using the compiled in default. In our
3442 case the guess is not the same as the compiled default, so the script
3443 suddenly stopped working when I started to use syslog. The patch checks
3444 to see if log_file_path is "". If so, it attempts to read it from exim
3445 with no configuration file to get the compiled in version, before it
3446 falls back to the previous guessing code."
3448 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
3449 implementing BATV in an Exim configuration. See NewStuff for the gory
3452 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
3453 Makefile that are specific to HP-UX.
3455 PH/18 If the "use_postmaster" option was set for a recipient callout together
3456 with the "random" option, the postmaster address was used as the MAIL
3457 FROM address for the random test, but not for the subsequent recipient
3458 test. It is now used for both.
3460 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
3461 patch removes a few documentation additions to RFC 3028, because the
3462 latest draft now contains them. It adds the new en;ascii-case comparator
3463 and a new error check for 8bit text in MIME parts. Comparator and
3464 require names are now matched exactly. I enabled the subaddress
3465 extension, but it is not well tested yet (read: it works for me)."
3467 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
3468 rework some of the code of TK/09 above to avoid the hardwired use of
3469 "%lld" and "long long". Replaced the call to snprintf() with a call to
3472 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
3473 records point to non-existent hosts", "retry timeout exceeded", and
3474 "retry time not reached for any host after a long failure period".
3476 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
3477 experimental DomainKeys support:
3479 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
3480 (2) On an error such as an illegally used "control", the wrong name for
3481 the control was given.
3483 These problems did NOT occur unless DomainKeys support was compiled.
3485 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
3487 PH/24 Added ${if match_ip condition.
3489 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
3490 databases so that it will be absolutely obvious if a crash occurs in the
3491 DB library. This is a regular occurrence (often caused by mis-matched
3494 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
3495 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
3496 picked up on a user's system that detects such things. There doesn't seem
3497 to be a gcc warning option for this - only an attribute that has to be
3498 put on the function's prototype. It seems that in Fedora Core 4 they have
3499 set this on a number of new functions. No doubt there will be more in due
3502 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
3503 specify a transport. However, if an address that was verified by such a
3504 router was the subject of a callout, Exim crashed because it tried to
3505 read the rcpt_include_affixes from the non-existent transport. Now it
3506 just assumes that the setting of that option is false. This bug was
3507 introduced by 4.51/PH/31.
3509 PH/28 Changed -d+all to exclude +memory, because that information is very
3510 rarely of interest, but it makes the output a lot bigger. People tend to
3511 do -d+all out of habit.
3513 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
3514 code in os-type was giving problems when libc.so lives in lib64, like on
3517 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
3518 aren't the modern standard, and it seems that some systems' include files
3519 don't always have them. Exim was already checking for some of the newer
3520 ones like T_AAAA, and defining it itself. I've added checks for all the
3521 record types that Exim uses.
3523 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
3524 not automatically generating a new one, as it is supposed to. This
3525 prevented TLS from working. If the file did exist, but contained invalid
3526 data, a new version was generated, as expected. It was only the case of a
3527 non-existent file that was broken.
3529 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
3530 with a change in libdomainkeys > 0.64.
3532 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
3533 from DNS. If the selector record carries the flag, it now has
3534 precedence over the domain-wide flag.
3536 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
3538 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
3539 the use of a port name that isn't defined in /etc/services) occurred, the
3540 message was deferred as in a normal delivery, and thus remained on the
3541 spool, instead of being failed because of the mua_wrapper setting. This
3542 is now fixed, and I tidied up some of the mua_wrapper messages at the
3545 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
3546 the messages in a hash of arrays rather than using individual hashes.
3547 This is a bit cleaner and results in dramatic memory savings, albeit
3548 at a slight CPU cost.
3550 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
3551 as requested by Marc Sherman.
3553 SC/10 Eximstats: added histograms for user specified patterns as requested
3556 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
3558 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
3559 fopen() in the content-scanning modules that did not already have it.
3565 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
3566 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
3568 TK/02 Fix ACL "control" statement not being available in MIME ACL.
3570 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
3572 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
3573 to test Sieve filters that use "vacation".
3575 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
3576 that changes the way the GnuTLS parameters are stored in the cache file.
3577 The new format can be generated externally. For backward compatibility,
3578 if the data in the cache doesn't make sense, Exim assumes it has read an
3579 old-format file, and it generates new data and writes a new file. This
3580 means that you can't go back to an older release without removing the
3583 PH/03 A redirect router that has both "unseen" and "one_time" set does not
3584 work if there are any delivery delays because "one_time" forces the
3585 parent to be marked "delivered", so its unseen clone is never tried
3586 again. For this reason, Exim now forbids the simultaneous setting of
3589 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
3590 redirected to themselves ("homonym" addresses). Read the long ChangeLog
3591 entry if you want to know the details. The fix, however, neglected to
3592 consider the case when local delivery batching is involved. The test for
3593 "previously delivered" was not happening when checking to see if an
3594 address could be batched with a previous (undelivered) one; under
3595 certain circumstances this could lead to multiple deliveries to the same
3598 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
3599 in its include files, and this causes problems building Exim.
3601 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
3602 header_syntax) but Exim was just ignoring anything given after a slash.
3603 In particular, this caused confusion with an attempt to use "verify =
3604 reverse_host_lookup/defer_ok". An error is now given when options are
3605 supplied for verify items that do not have them. (Maybe reverse_host_
3606 lookup should have a defer_ok option, but that's a different point.)
3608 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
3609 defined by RFC 821) to 2048, because there were problems with some AUTH
3610 commands, and RFC 1869 says the size should be increased for extended
3611 SMTP commands that take arguments.
3613 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
3616 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
3617 "unknown" error; now it says that the functionality isn't in the binary.
3619 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
3620 an address' error message when a string expansion fails (syntax or
3621 whatever). Otherwise the password may appear in the log. Following change
3622 PH/42 below, there is no longer a chance of it appearing in a bounce
3625 PH/11 Installed exipick version 20050225.0 from John Jetmore.
3627 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
3628 of its addresses was ever tried. (Bugzilla bug #2.)
3630 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
3631 the result incorrectly in the debug output. (It correctly added a newline
3632 to what was transported.)
3634 TF/01 Added $received_time.
3636 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
3637 commented out examples of how to interface to a virus scanner and to
3638 SpamAssassin. Also added commented examples of av_scanner and
3639 spamd_address settings.
3641 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
3642 and controls are allowed in which ACLs. There were a couple of minor
3643 errors. Some of the entries in the conditions table (which is a table of
3644 where they are NOT allowed) were getting very unwieldy; rewrote them as a
3645 negation of where the condition IS allowed.
3647 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
3649 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
3650 header file does not have a version number, so I've had to invent a new
3651 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
3652 API. The code is untested by me (my Linux distribution still has 0.3.2 of
3653 radiusclient), but it was contributed by a Radius user.
3655 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
3656 files or directories, for OpenSSL.
3658 PH/19 When an Exim process that is running as root has to create an Exim log
3659 file, it does so in a subprocess that runs as exim:exim so as to get the
3660 ownership right at creation (otherwise, other Exim processes might see
3661 the file with the wrong ownership). There was no test for failure of this
3662 fork() call, which would lead to the process getting stuck as it waited
3663 for a non-existent subprocess. Forks do occasionally fail when resources
3664 run out. I reviewed all the other calls to fork(); they all seem to check
3667 PH/20 When checking for unexpected SMTP input at connect time (before writing
3668 the banner), Exim was not dealing correctly with a non-positive return
3669 from the read() function. If the client had disconnected by this time,
3670 the result was a log entry for a synchronization error with an empty
3671 string after "input=" when read() returned zero. If read() returned -1
3672 (an event I could not check), uninitialized data bytes were printed.
3673 There were reports of junk text (parts of files, etc) appearing after
3676 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
3678 PH/22 Added support for macro redefinition, and (re)definition in between
3679 driver and ACL definitions.
3681 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
3682 forgetting to use the resulting value; it was using the unexpanded value.
3684 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
3685 hadn't been configured. The fix is from Juergen Kreileder, who
3686 understands it better than I do:
3688 "Here's what I see happening with three configured cyrus_sasl
3689 authenticators configured (plain, login, cram-md5):
3691 On startup auth_cyrus_sasl_init() gets called for each of these.
3692 This means three calls to sasl_listmech() without a specified mech_list.
3693 => SASL tests which mechs of all available mechs actually work
3694 => three warnings about OTP not working
3695 => the returned list contains: plain, login, cram-md5, digest-md5, ...
3697 With the patch, sasl_listmech() also gets called three times. But now
3698 SASL's mech_list option is set to the server_mech specified in the the
3699 authenticator. Or in other words, the answer from sasl_listmech()
3700 gets limited to just the mech you're testing for (which is different
3702 => the return list contains just 'plain' or 'login', 'cram-md5' or
3703 nothing depending on the value of ob->server_mech.
3705 I've just tested the patch: Authentication still works fine,
3706 unavailable mechs specified in the exim configuration are still
3707 caught, and the auth.log warnings about OTP are gone."
3709 PH/25 When debugging is enabled, the contents of the command line are added
3710 to the debugging output, even when log_selector=+arguments is not
3713 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
3714 answer is "GNU", and only if the return is "GNU/something" is the answer
3717 PH/27 $acl_verify_message is now set immediately after the failure of a
3718 verification in an ACL, and so is available in subsequent modifiers. In
3719 particular, the message can be preserved by coding like this:
3721 warn !verify = sender
3722 set acl_m0 = $acl_verify_message
3724 Previously, $acl_verify_message was set only while expanding "message"
3725 and "log_message" when a very denied access.
3727 PH/28 Modified OS/os.c-Linux with
3729 -#ifndef OS_LOAD_AVERAGE
3730 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
3732 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
3733 nomenclature these days.)
3735 PH/29 Installed patch from the Sieve maintainer that adds the options
3736 sieve_useraddress and sieve_subaddress to the redirect router.
3738 PH/30 In these circumstances:
3739 . Two addresses routed to the same list of hosts;
3740 . First host does not offer TLS;
3741 . First host accepts first address;
3742 . First host gives temporary error to second address;
3743 . Second host offers TLS and a TLS session is established;
3744 . Second host accepts second address.
3745 Exim incorrectly logged both deliveries with the TLS parameters (cipher
3746 and peerdn, if requested) that were in fact used only for the second
3749 PH/31 When doing a callout as part of verifying an address, Exim was not paying
3750 attention to any local part prefix or suffix that was matched by the
3751 router that accepted the address. It now behaves in the same way as it
3752 does for delivery: the affixes are removed from the local part unless
3753 rcpt_include_affixes is set on the transport.
3755 PH/32 Add the sender address, as F=<...>, to the log line when logging a
3756 timeout during the DATA phase of an incoming message.
3758 PH/33 Sieve envelope tests were broken for match types other than :is. I have
3759 applied a patch sanctioned by the Sieve maintainer.
3761 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
3762 the uid or gid is negative. A case of a negative gid caused this to be
3763 noticed. The fix allows for either to be negative.
3765 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
3766 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
3769 PH/36 The change PH/12 above was broken. Fixed it.
3771 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
3772 the grounds that routing the same address twice would always produce the
3773 same answer. This might have been true once, but it is certainly no
3774 longer true now. Routing a child address may depend on the previous
3775 routing that produced that child. Some complicated redirection strategies
3776 went wrong when messages had multiple recipients, and made Exim's
3777 behaviour dependent on the order in which the addresses were given.
3779 I have moved the duplicate checking until after the routing is complete.
3780 Exim scans the addresses that are assigned to local and remote
3781 transports, and removes any duplicates. This means that more work will be
3782 done, as duplicates will always all be routed, but duplicates are
3783 presumably rare, so I don't expect this is of any significance.
3785 For deliveries to pipes, files, and autoreplies, the duplicate checking
3786 still happens during the routing process, since they are not going to be
3789 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
3790 It corrects a timeout issue with spamd. This is Ian's comment: "The
3791 background is that sometimes spamd either never reads data from a
3792 connection it has accepted, or it never writes response data. The exiscan
3793 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
3794 blindly assumes that writes won't block so it may never time out."
3796 PH/39 Allow G after quota size as well as K and M.
3798 PH/40 The value set for $authenticated_id in an authenticator may not contain
3799 binary zeroes or newlines because the value is written to log lines and
3800 to spool files. There was no check on this. Now the value is run through
3801 the string_printing() function so that such characters are converted to
3802 printable escape sequences.
3804 PH/41 $message_linecount is a new variable that contains the total number of
3805 lines in the message. Compare $body_linecount, which is the count for the
3808 PH/42 Exim no longer gives details of delivery errors for specific addresses in
3809 bounce and delay warning messages, except in certain special cases, which
3812 (a) An SMTP error message from a remote host;
3813 (b) A message specified in a :fail: redirection;
3814 (c) A message specified in a "fail" command in a system filter;
3815 (d) A message specified in a FAIL return from the queryprogram router;
3816 (e) A message specified by the cannot_route_message router option.
3818 In these cases only, Exim does include the error details in bounce and
3819 warning messages. There are also a few cases where bland messages such
3820 as "unrouteable address" or "local delivery error" are given.
3822 PH/43 $value is now also set for the "else" part of a ${run expansion.
3824 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
3825 being worked on, but at least Exim now implements the latest version to
3828 PH/45 In a pipe transport, although a timeout while waiting for the pipe
3829 process to complete was treated as a delivery failure, a timeout while
3830 writing the message to the pipe was logged, but erroneously treated as a
3831 successful delivery. Such timeouts include transport filter timeouts. For
3832 consistency with the overall process timeout, these timeouts are now
3833 treated as errors, giving rise to delivery failures by default. However,
3834 there is now a new Boolean option for the pipe transport called
3835 timeout_defer, which, if set TRUE, converts the failures into defers for
3836 both kinds of timeout. A transport filter timeout is now identified in
3839 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
3840 systems where "make" and "gmake" are different, calling "gmake" at top
3841 level broke things. I've arranged for the value of $(MAKE) to be passed
3842 from the Makefile to this script so that it can call the same version of
3846 A note about Exim versions 4.44 and 4.50
3847 ----------------------------------------
3849 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
3850 changes of various kinds. As a consequence, a big documentation update was
3851 needed. This delayed the release for rather longer than seemed good, especially
3852 in the light of a couple of (minor) security issues. Therefore, the changes
3853 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
3854 release. So 4.44 and 4.50 are in effect two different branches that both start
3857 I have left the 4.50 change log unchanged; it contains all the changes since
3858 4.43. The change log for 4.44 is below; many of its items are identical to
3859 those for 4.50. This seems to be the most sensible way to preserve the
3860 historical information.
3866 1. Minor wording change to the doc/README.SIEVE file.
3868 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
3869 computation of the current number of files was incorrect.
3871 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
3872 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
3875 4. Give more explanation in the error message when the command for a transport
3876 filter fails to execute.
3878 5. There are several places where Exim runs a non-Exim command in a
3879 subprocess. The SIGUSR1 signal should be disabled for these processes. This
3880 was being done only for the command run by the queryprogram router. It is
3881 now done for all such subprocesses. The other cases are: ${run, transport
3882 filters, and the commands run by the lmtp and pipe transports.
3884 6. Added CONFIGURE_GROUP build-time option.
3886 7. Some older OS have a limit of 256 on the maximum number of file
3887 descriptors. Exim was using setrlimit() to set 1000 as a large value
3888 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
3889 systems. I've change it so that if it can't get 1000, it tries for 256.
3891 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
3892 was an oversight, and furthermore, ever since the addition of extra
3893 controls (e.g. 4.43/32), the checks on when to allow different forms of
3894 "control" were broken. There should now be diagnostics for all cases when a
3895 control that does not make sense is encountered.
3897 9. Added the /retain_sender option to "control=submission".
3899 10. $recipients is now available in the predata ACL (oversight).
3901 11. Tidy the search cache before the fork to do a delivery from a message
3902 received from the command line. Otherwise the child will trigger a lookup
3903 failure and thereby defer the delivery if it tries to use (for example) a
3904 cached ldap connection that the parent has called unbind on.
3906 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
3907 of $address_data from the recipient verification was clobbered by the
3908 sender verification.
3910 13. The value of address_data from a sender verification is now available in
3911 $sender_address_data in subsequent conditions in the ACL statement.
3913 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
3915 15. Added a new option "connect=<time>" to callout options, to set a different
3918 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
3919 was its contents. (It was OK if the option was not defined at all.)
3921 17. A "Completed" log line is now written for messages that are removed from
3922 the spool by the -Mrm option.
3924 18. New variables $sender_verify_failure and $recipient_verify_failure contain
3925 information about exactly what failed.
3927 19. Added -dd to debug only the daemon process.
3929 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
3930 handles timeouts, both on the server side and network timeouts. Renamed the
3931 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
3933 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
3934 It is now set to "smtps".
3936 22. $host_address is now set to the target address during the checking of
3937 ignore_target_hosts.
3939 23. When checking ignore_target_hosts for an ipliteral router, no host name was
3940 being passed; this would have caused $sender_host_name to have been used if
3941 matching the list had actually called for a host name (not very likely,
3942 since this list is usually IP addresses). A host name is now passed as
3945 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
3946 code that specifies a non-restarting handler (typically sigaction() in
3947 modern systems) in an attempt to fix a rare and obscure crash bug.
3949 25. Narrowed the window for a race in the daemon that could cause it to ignore
3950 SIGCHLD signals. This is not a major problem, because they are used only to
3951 wake it up if nothing else does.
3953 26. A malformed maildirsize file could cause Exim to calculate negative values
3954 for the mailbox size or file count. Odd effects could occur as a result.
3955 The maildirsize information is now recalculated if the size or filecount
3958 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
3959 support for a long time. Removed HAVE_SYS_VFS_H.
3961 28. Installed the latest version of exipick from John Jetmore.
3963 29. In an address list, if the pattern was not a regular expression, an empty
3964 subject address (from a bounce message) matched only if the pattern was an
3965 empty string. Non-empty patterns were not even tested. This was the wrong
3966 because it is perfectly reasonable to use an empty address as part of a
3967 database query. An empty address is now tested by patterns that are
3968 lookups. However, all the other forms of pattern expect the subject to
3969 contain a local part and a domain, and therefore, for them, an empty
3970 address still always fails if the pattern is not itself empty.
3972 30. Exim went into a mad DNS loop when attempting to do a callout where the
3973 host was specified on an smtp transport, and looking it up yielded more
3974 than one IP address.
3976 31. Re-factored the code for checking spool and log partition space into a
3977 function that finds that data and another that does the check. The former
3978 is then used to implement four new variables: $spool_space, $log_space,
3979 $spool_inodes, and $log_inodes.
3981 32. The RFC2047 encoding function was originally intended for short strings
3982 such as real names; it was not keeping to the 75-character limit for
3983 encoded words that the RFC imposes. It now respects the limit, and
3984 generates multiple encoded words if necessary. To be on the safe side, I
3985 have increased the buffer size for the ${rfc2047: expansion operator from
3988 33. It is now permitted to omit both strings after an "if" condition; if the
3989 condition is true, the result is "true". As before, when the second string
3990 is omitted, a false condition yields an empty string. This makes it less
3991 cumbersome to write custom ACL and router conditions.
3993 34. Failure to deliver a bounce message always caused it to be frozen, even if
3994 there was an errors_to setting on the router. The errors_to setting is now
3997 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
3998 canonical form (fully expanded) before being placed in
3999 $sender_host_address.
4001 36. The table in the code that translates DNS record types into text (T_A to
4002 "A" for instance) was missing entries for NS and CNAME. It is just possible
4003 that this could have caused confusion if both these types were looked up
4004 for the same domain, because the text type is used as part of Exim's
4005 per-process caching. But the chance of anyone hitting this buglet seems
4008 37. The dnsdb lookup has been extended in a number of ways.
4010 (1) There is a new type, "zns", which walks up the domain tree until it
4011 finds some nameserver records. It should be used with care.
4013 (2) There is a new type, "mxh", which is like "mx" except that it returns
4014 just the host names, not the priorities.
4016 (3) It is now possible to give a list of domains (or IP addresses) to be
4017 looked up. The behaviour when one of the lookups defers can be
4018 controlled by a keyword.
4020 (4) It is now possible to specify the separator character for use when
4021 multiple records are returned.
4023 38. The dnslists ACL condition has been extended: it is now possible to supply
4024 a list of IP addresses and/or domains to be looked up in a particular DNS
4027 39. Added log_selector=+queue_time_overall.
4029 40. When running the queue in the test harness, wait just a tad after forking a
4030 delivery process, to get repeatability of debugging output.
4032 41. Include certificate and key file names in error message when GnuTLS fails
4033 to set them up, because the GnuTLS error message doesn't include the name
4034 of the failing file when there is a problem reading it.
4036 42. Allow both -bf and -bF in the same test run.
4038 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
4040 44. The "Exiscan patch" is now merged into the mainline Exim source.
4042 45. Sometimes the final signoff response after QUIT could fail to get
4043 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4044 < 0 before doing a fflush(). This bug looks as though it goes back to the
4045 introduction of TLS in release 3.20, but "sometimes" must have been rare
4046 because the tests only now provoked it.
4048 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
4049 (this can affect the format of dates).
4051 47. exim_tidydb, when checking for the continued existence of a message for
4052 which it has found a message-specific retry record, was not finding
4053 messages that were in split spool directories. Consequently, it was
4054 deleting retry records that should have stayed in existence.
4056 48. Steve fixed some bugs in eximstats.
4058 49. The SPA authentication driver was not abandoning authentication and moving
4059 on to the next authenticator when an expansion was forced to fail,
4060 contradicting the general specification for all authenticators. Instead it
4061 was generating a temporary error. It now behaves as specified.
4063 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
4064 (the order specifies the preference for clients). The order is now AES256,
4065 AES128, 3DES, ARCFOUR128.
4067 51. Small patch to Sieve code - explicitly set From: when generating an
4070 52. Exim crashed if a remote delivery caused a very long error message to be
4071 recorded - for instance if somebody sent an entire SpamAssassin report back
4072 as a large number of 550 error lines. This bug was coincidentally fixed by
4073 increasing the size of one of Exim's internal buffers (big_buffer) that
4074 happened as part of the Exiscan merge. However, to be on the safe side, I
4075 have made the code more robust (and fixed the comments that describe what
4078 53. Now that there can be additional text after "Completed" in log lines (if
4079 the queue_time_overall log selector is set), a one-byte patch to exigrep
4080 was needed to allow it to recognize "Completed" as not the last thing in
4083 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
4084 patch that reportedly fixes this has been added. I am not expert enough to
4085 create a test for it. This is what the patch creator wrote:
4087 "I found a little strange behaviour of ldap code when working with
4088 Windows 2003 AD Domain, where users was placed in more than one
4089 Organization Units. When I tried to give exim partial DN, the exit code
4090 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
4091 But simultaneously result of request was absolutely normal ldap result,
4092 so I produce this patch..."
4094 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
4095 REFERENCE, so I have modified the code to exclude the patch when that macro
4098 55. Some experimental protocols are using DNS PTR records for new purposes. The
4099 keys for these records are domain names, not reversed IP addresses. The
4100 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4101 leaves it alone. Component reversal etc. now happens only for IP addresses.
4104 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
4106 57. Double the size of the debug message buffer (to 2048) so that more of very
4107 long debug lines gets shown.
4109 58. The exicyclog utility now does better if the number of log files to keep
4110 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
4112 59. Two changes related to the smtp_active_hostname option:
4114 (1) $smtp_active_hostname is now available as a variable.
4115 (2) The default for smtp_banner uses $smtp_active_hostname instead
4116 of $primary_hostname.
4118 60. The host_aton() function is supposed to be passed a string that is known
4119 to be a valid IP address. However, in the case of IPv6 addresses, it was
4120 not checking this. This is a hostage to fortune. Exim now panics and dies
4121 if the condition is not met. A case was found where this could be provoked
4122 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4123 components; fortuitously, this particular loophole had already been fixed
4124 by change 4.50/55 above.
4126 If there are any other similar loopholes, the new check in host_aton()
4127 itself should stop them being exploited. The report I received stated that
4128 data on the command line could provoke the exploit when Exim was running as
4129 exim, but did not say which command line option was involved. All I could
4130 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4131 running as the user.
4134 61. There was a buffer overflow vulnerability in the SPA authentication code
4135 (which came originally from the Samba project). I have added a test to the
4136 spa_base64_to_bits() function which I hope fixes it.
4139 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
4140 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
4142 63. The daemon start-up calls getloadavg() while still root for those OS that
4143 need the first call to be done as root, but it missed one case: when
4144 deliver_queue_load_max is set with deliver_drop_privilege. This is
4145 necessary for the benefit of the queue runner, because there is no re-exec
4146 when deliver_drop_privilege is set.
4148 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
4149 This has been fixed.
4151 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4152 was in use, was not putting the data itself into the right store pool;
4153 consequently, it could be overwritten for a subsequent message in the same
4154 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4157 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
4159 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
4160 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
4161 the function were treating the return as a boolean value, which happened to
4162 work because 0=false and not-0=true, but is not correct code.
4164 68. The host_aton() function was not handling scoped IPv6 addresses (those
4165 with, for example, "%eth0" on the end) correctly.
4167 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
4168 negated items (that is, ~something) in unsigned ints. Some compilers
4169 apparently mutter when there is no cast.
4171 70. If an address verification called from an ACL failed, and did not produce a
4172 user-specific message (i.e. there was only a "system" message), nothing was
4173 put in $acl_verify_message. In this situation, it now puts the system
4176 71. Change 4.23/11 added synchronization checking at the start of an SMTP
4177 session; change 4.31/43 added the unwanted input to the log line - except
4178 that it did not do this in the start of session case. It now does.
4180 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
4181 This is wrong and can cause the other end to generate a synchronization
4182 error if it is another Exim or anything else that does the synchronization
4183 check. A QUIT command is no longer sent after a timeout.
4185 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
4186 during host lookups.
4188 74. The defer_ok option of callout verification was not working if it was used
4189 when verifying addresses in header lines, that is, for this case:
4191 verify = header_sender/callout=defer_ok
4193 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
4194 those file descriptors could be used for SMTP connections. If anything
4195 wrote to stderr (the example that came up was "warn" in embedded Perl), it
4196 could be sent to the SMTP client, causing chaos. The daemon now opens
4197 stdin, stdout, and stderr to /dev/null when it puts itself into the
4200 76. Arrange for output from Perl's "warn" command to be written to Exim's main
4201 log by default. The user can override this with suitable Perl magic.
4203 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
4204 the log message when discard triggers, was not working for the DATA ACL or
4205 for the non-SMTP ACL.
4207 78. Error message wording change in sieve.c.
4209 79. If smtp_accept_max_per_host was set, the number of connections could be
4210 restricted to fewer than expected, because the daemon was trying to set up
4211 a new connection before checking whether the processes handling previous
4212 connections had finished. The check for completed processes is now done
4213 earlier. On busy systems, this bug wouldn't be noticed because something
4214 else would have woken the daemon, and it would have reaped the completed
4217 80. If a message was submitted locally by a user whose login name contained one
4218 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
4219 It caused a spool format error. I have fixed the spool reading code. A
4220 related problem was that the "from" clause in the Received: line became
4221 illegal because of the space(s). It is now covered by ${quote_local_part.
4223 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
4226 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
4228 83. Patch from Sieve maintainer to fix unterminated string problem in
4229 "vacation" handling.
4231 84. Some minor changes to the Linux configuration files to help with other
4232 OS variants using glibc.
4234 85. One more patch for Sieve to update vacation handling to latest spec.
4237 ----------------------------------------------------
4238 See the note above about the 4.44 and 4.50 releases.
4239 ----------------------------------------------------
4245 1. Change 4.43/35 introduced a bug that caused file counts to be
4246 incorrectly computed when quota_filecount was set in an appendfile
4249 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4250 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4253 3. Give more explanation in the error message when the command for a transport
4254 filter fails to execute.
4256 4. There are several places where Exim runs a non-Exim command in a
4257 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4258 was being done only for the command run by the queryprogram router. It is
4259 now done for all such subprocesses. The other cases are: ${run, transport
4260 filters, and the commands run by the lmtp and pipe transports.
4262 5. Some older OS have a limit of 256 on the maximum number of file
4263 descriptors. Exim was using setrlimit() to set 1000 as a large value
4264 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4265 systems. I've change it so that if it can't get 1000, it tries for 256.
4267 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4268 was an oversight, and furthermore, ever since the addition of extra
4269 controls (e.g. 4.43/32), the checks on when to allow different forms of
4270 "control" were broken. There should now be diagnostics for all cases when a
4271 control that does not make sense is encountered.
4273 7. $recipients is now available in the predata ACL (oversight).
4275 8. Tidy the search cache before the fork to do a delivery from a message
4276 received from the command line. Otherwise the child will trigger a lookup
4277 failure and thereby defer the delivery if it tries to use (for example) a
4278 cached ldap connection that the parent has called unbind on.
4280 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4281 of $address_data from the recipient verification was clobbered by the
4282 sender verification.
4284 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4285 was its contents. (It was OK if the option was not defined at all.)
4287 11. A "Completed" log line is now written for messages that are removed from
4288 the spool by the -Mrm option.
4290 12. $host_address is now set to the target address during the checking of
4291 ignore_target_hosts.
4293 13. When checking ignore_target_hosts for an ipliteral router, no host name was
4294 being passed; this would have caused $sender_host_name to have been used if
4295 matching the list had actually called for a host name (not very likely,
4296 since this list is usually IP addresses). A host name is now passed as
4299 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4300 code that specifies a non-restarting handler (typically sigaction() in
4301 modern systems) in an attempt to fix a rare and obscure crash bug.
4303 15. Narrowed the window for a race in the daemon that could cause it to ignore
4304 SIGCHLD signals. This is not a major problem, because they are used only to
4305 wake it up if nothing else does.
4307 16. A malformed maildirsize file could cause Exim to calculate negative values
4308 for the mailbox size or file count. Odd effects could occur as a result.
4309 The maildirsize information is now recalculated if the size or filecount
4312 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4313 support for a long time. Removed HAVE_SYS_VFS_H.
4315 18. Updated exipick to current release from John Jetmore.
4317 19. Allow an empty sender to be matched against a lookup in an address list.
4318 Previously the only cases considered were a regular expression, or an
4321 20. Exim went into a mad DNS lookup loop when doing a callout where the
4322 host was specified on the transport, if the DNS lookup yielded more than
4325 21. The RFC2047 encoding function was originally intended for short strings
4326 such as real names; it was not keeping to the 75-character limit for
4327 encoded words that the RFC imposes. It now respects the limit, and
4328 generates multiple encoded words if necessary. To be on the safe side, I
4329 have increased the buffer size for the ${rfc2047: expansion operator from
4332 22. Failure to deliver a bounce message always caused it to be frozen, even if
4333 there was an errors_to setting on the router. The errors_to setting is now
4336 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4337 canonical form (fully expanded) before being placed in
4338 $sender_host_address.
4340 24. Updated eximstats to version 1.33
4342 25. Include certificate and key file names in error message when GnuTLS fails
4343 to set them up, because the GnuTLS error message doesn't include the name
4344 of the failing file when there is a problem reading it.
4346 26. Expand error message when OpenSSL has problems setting up cert/key files.
4349 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
4350 (this can affect the format of dates).
4352 28. exim_tidydb, when checking for the continued existence of a message for
4353 which it has found a message-specific retry record, was not finding
4354 messages that were in split spool directories. Consequently, it was
4355 deleting retry records that should have stayed in existence.
4357 29. eximstats updated to version 1.35
4358 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
4359 1.35 - bugfix such that pie charts by volume are generated correctly
4361 30. The SPA authentication driver was not abandoning authentication and moving
4362 on to the next authenticator when an expansion was forced to fail,
4363 contradicting the general specification for all authenticators. Instead it
4364 was generating a temporary error. It now behaves as specified.
4366 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
4367 (the order specifies the preference for clients). The order is now AES256,
4368 AES128, 3DES, ARCFOUR128.
4370 31. Small patch to Sieve code - explicitly set From: when generating an
4373 32. Exim crashed if a remote delivery caused a very long error message to be
4374 recorded - for instance if somebody sent an entire SpamAssassin report back
4375 as a large number of 550 error lines. This bug was coincidentally fixed by
4376 increasing the size of one of Exim's internal buffers (big_buffer) that
4377 happened as part of the Exiscan merge. However, to be on the safe side, I
4378 have made the code more robust (and fixed the comments that describe what
4381 33. Some experimental protocols are using DNS PTR records for new purposes. The
4382 keys for these records are domain names, not reversed IP addresses. The
4383 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4384 leaves it alone. Component reversal etc. now happens only for IP addresses.
4387 34. The host_aton() function is supposed to be passed a string that is known
4388 to be a valid IP address. However, in the case of IPv6 addresses, it was
4389 not checking this. This is a hostage to fortune. Exim now panics and dies
4390 if the condition is not met. A case was found where this could be provoked
4391 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4392 components; fortuitously, this particular loophole had already been fixed
4393 by change 4.50/55 or 4.44/33 above.
4395 If there are any other similar loopholes, the new check in host_aton()
4396 itself should stop them being exploited. The report I received stated that
4397 data on the command line could provoke the exploit when Exim was running as
4398 exim, but did not say which command line option was involved. All I could
4399 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4400 running as the user.
4403 35. There was a buffer overflow vulnerability in the SPA authentication code
4404 (which came originally from the Samba project). I have added a test to the
4405 spa_base64_to_bits() function which I hope fixes it.
4408 36. The daemon start-up calls getloadavg() while still root for those OS that
4409 need the first call to be done as root, but it missed one case: when
4410 deliver_queue_load_max is set with deliver_drop_privilege. This is
4411 necessary for the benefit of the queue runner, because there is no re-exec
4412 when deliver_drop_privilege is set.
4414 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4415 was in use, was not putting the data itself into the right store pool;
4416 consequently, it could be overwritten for a subsequent message in the same
4417 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4420 38. Sometimes the final signoff response after QUIT could fail to get
4421 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4422 < 0 before doing a fflush(). This bug looks as though it goes back to the
4423 introduction of TLS in release 3.20, but "sometimes" must have been rare
4424 because the tests only now provoked it.
4430 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
4431 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
4432 Now it returns an integer. A number of calls were still expecting a T/F
4433 return. Fortuitously, in all cases, the tests worked in OK situations,
4434 which is the norm. However, things would have gone wrong on any write
4435 failures on the smtp file descriptor. This function is used when sending
4436 messages over SMTP and also when doing verify callouts.
4438 2. When Exim is called to do synchronous delivery of a locally submitted
4439 message (the -odf or -odi options), it no longer closes stderr before doing
4442 3. Implemented the mua_wrapper option.
4444 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
4446 5. Implemented the functions header_remove(), header_testname(),
4447 header_add_at_position(), and receive_remove_recipient(), and exported them
4450 6. If an ACL "warn" statement specified the addition of headers, Exim already
4451 inserted X-ACL-Warn: at the start if there was no header name. However, it
4452 was not making this test for the second and subsequent header lines if
4453 there were newlines in the string. This meant that an invalid header could
4454 be inserted if Exim was badly configured.
4456 7. Allow an ACL "warn" statement to add header lines at the start or after all
4457 the Received: headers, as well as at the end.
4459 8. Added the rcpt_4xx retry error code.
4461 9. Added postmaster_mailfrom=xxx to callout verification option.
4463 10. Added mailfrom=xxxx to the callout verification option, for verify=
4466 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
4467 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
4469 12. Inserted some casts to stop certain compilers warning when using pointer
4470 differences as field lengths or precisions in printf-type calls (mostly
4471 affecting debugging statements).
4473 13. Added optional readline() support for -be (dynamically loaded).
4475 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
4476 same clock tick as a message's arrival, so that its received time was the
4477 same as the "first fail" time on the retry record, and that message
4478 remained on the queue past the ultimate address timeout, every queue runner
4479 would try a delivery (because it was past the ultimate address timeout) but
4480 after another failure, the ultimate address timeout, which should have then
4481 bounced the address, did not kick in. This was a "< instead of <=" error;
4482 in most cases the first failure would have been in the next clock tick
4483 after the received time, and all would be well.
4485 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
4486 being recognized when the domain list was tested by the match_domain
4487 condition in an expansion string.
4489 16. Added the ${str2b64: operator.
4491 17. Exim was always calling setrlimit() to set a large limit for the number of
4492 processes, without checking whether the existing limit was already
4493 adequate. (It did check for the limit on file descriptors.) Furthermore,
4494 errors from getrlimit() and setrlimit() were being ignored. Now they are
4495 logged to the main and panic logs, but Exim does carry on, to try to do its
4496 job under whatever limits there are.
4498 18. Imported PCRE 5.0.
4500 19. Trivial typo in log message " temporarily refused connection" (the leading
4503 20. If the log selector return_path_on_delivery was set and an address was
4504 redirected to /dev/null, the delivery process crashed because it assumed
4505 that a return path would always be set for a "successful" delivery. In this
4506 case, the whole delivery is bypassed as an optimization, and therefore no
4509 21. Internal re-arrangement: the function for sending a challenge and reading
4510 a response while authentication was assuming a zero-terminated challenge
4511 string. It's now changed to take a pointer and a length, to allow for
4512 binary data in such strings.
4514 22. Added the cyrus_sasl authenticator (code supplied by MBM).
4516 23. Exim was not respecting finduser_retries when seeking the login of the
4517 uid under which it was called; it was always trying 10 times. (The default
4518 setting of finduser_retries is zero.) Also, it was sleeping after the final
4519 failure, which is pointless.
4521 24. Implemented tls_on_connect_ports.
4523 25. Implemented acl_smtp_predata.
4525 26. If the domain in control=submission is set empty, Exim assumes that the
4526 authenticated id is a complete email address when it generates From: or
4527 Sender: header lines.
4529 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
4530 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
4531 chown and chgrp in /bin and hostname in /usr/bin.
4533 28. Exim was keeping the "process log" file open after each use, just as it
4534 does for the main log. This opens the possibility of it remaining open for
4535 long periods when the USR1 signal hits a daemon. Occasional processlog
4536 errors were reported, that could have been caused by this. Anyway, it seems
4537 much more sensible not to leave this file open at all, so that is what now
4540 29. The long-running daemon process does not normally write to the log once it
4541 has entered its main loop, and it closes the log before doing so. This is
4542 so that log files can straightforwardly be renamed and moved. However,
4543 there are a couple of unusual error situations where the daemon does write
4544 log entries, and I had neglected to close the log afterwards.
4546 30. The text of an SMTP error response that was received during a remote
4547 delivery was being truncated at 512 bytes. This is too short for some of
4548 the long messages that one sometimes sees. I've increased the limit to
4551 31. It is now possible to make retry rules that apply only when a message has a
4552 specific sender, in particular, an empty sender.
4554 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
4555 it possible to be selective about when SMTP synchronization is enforced.
4557 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
4559 32. Implemented hosts_connection_nolog.
4561 33. Added an ACL for QUIT.
4563 34. Setting "delay_warning=" to disable warnings was not working; it gave a
4566 35. Added mailbox_size and mailbox_filecount to appendfile.
4568 36. Added control = no_multiline_responses to ACLs.
4570 37. There was a bug in the logic of the code that waits for the clock to tick
4571 in the case where the clock went backwards by a substantial amount such
4572 that the microsecond fraction of "now" was more than the microsecond
4573 fraction of "then" (but the whole seconds number was less).
4575 38. Added support for the libradius Radius client library this is found on
4576 FreeBSD (previously only the radiusclient library was supported).
4582 1. When certain lookups returned multiple values in the form name=value, the
4583 quoting of the values was not always being done properly. Specifically:
4584 (a) If the value started with a double quote, but contained no whitespace,
4586 (b) If the value contained whitespace other than a space character (i.e.
4587 tabs or newlines or carriage returns) it was not quoted.
4588 This fix has been applied to the mysql and pgsql lookups by writing a
4589 separate quoting function and calling it from the lookup code. The fix
4590 should probably also be applied to nisplus, ibase and oracle lookups, but
4591 since I cannot test any of those, I have not disturbed their existing code.
4593 2. A hit in the callout cache for a specific address caused a log line with no
4594 reason for rejecting RCPT. Now it says "Previous (cached) callout
4595 verification failure".
4597 3. There was an off-by-one bug in the queryprogram router. An over-long
4598 return line was truncated at 256 instead of 255 characters, thereby
4599 overflowing its buffer with the terminating zero. As well as fixing this, I
4600 have increased the buffer size to 1024 (and made a note to document this).
4602 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
4603 when Exim is waiting for an SMTP response from a remote server, Exim
4604 restarts its select() call on the socket, thereby resetting its timeout.
4605 This is not a problem when such interrupts are rare. Somebody set up a cron
4606 job to run exiwhat every 2 minutes, which is less than the normal select()
4607 timeout (5 or 10 minutes). This meant that the select() timeout never
4608 kicked in because it was always reset. I have fixed this by comparing the
4609 time when an interrupt arrives with the time at the start of the first call
4610 to select(). If more time than the timeout has elapsed, the interrupt is
4611 treated as a timeout.
4613 5. Some internal re-factoring in preparation for the addition of Sieve
4614 extensions (by MH). In particular, the "personal" test is moved to a
4615 separate function, and given an option for scanning Cc: and Bcc: (which is
4616 not set for Exim filters).
4618 6. When Exim created an email address using the login of the caller as the
4619 local part (e.g. when creating a From: or Sender: header line), it was not
4620 quoting the local part when it contained special characters such as @.
4622 7. Installed new OpenBSD configuration files.
4624 8. Reworded some messages for syntax errors in "and" and "or" conditions to
4625 try to make them clearer.
4627 9. Callout options, other than the timeout value, were being ignored when
4628 verifying sender addresses in header lines. For example, when using
4630 verify = header_sender/callout=no_cache
4632 the cache was (incorrectly) being used.
4634 10. Added a missing instance of ${EXE} to the exim_install script; this affects
4635 only the Cygwin environment.
4637 11. When return_path_on_delivery was set as a log selector, if different remote
4638 addresses in the same message used different return paths and parallel
4639 remote delivery occurred, the wrong values would sometimes be logged.
4640 (Whenever a remote delivery process finished, the return path value from
4641 the most recently started remote delivery process was logged.)
4643 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
4644 lines when AUTH and/or TLS are in use. This is the "received protocol"
4645 field. Exim used to use "asmtp" for authenticated SMTP, without any
4646 indication (in the protocol name) for TLS use. Now it follows the RFC and
4647 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
4648 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
4649 names appear in log lines as well as in Received: header lines.
4651 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
4652 extensions, and comparison tests, and to fix some bugs.
4654 14. Changes to the "personal" filter test:
4656 (1) The test was buggy in that it was just doing the equivalent of
4657 "contains" tests on header lines. For example, if a user's address was
4658 anne@some.where, the "personal" test would incorrectly be true for
4660 To: susanne@some.where
4662 This test is now done by extracting each address from the header in turn,
4663 and checking the entire address. Other tests that are part of "personal"
4664 are now done using regular expressions (for example, to check local parts
4665 of addresses in From: header lines).
4667 (2) The list of non-personal local parts in From: addresses has been
4668 extended to include "listserv", "majordomo", "*-request", and "owner-*",
4669 taken from the Sieve specification recommendations.
4671 (3) If the message contains any header line starting with "List-" it is
4672 treated as non-personal.
4674 (4) The test for "circular" in the Subject: header line has been removed
4675 because it now seems ill-conceived.
4677 15. Minor typos in src/EDITME comments corrected.
4679 16. Installed latest exipick from John Jetmore.
4681 17. If headers_add on a router specified a text string that was too long for
4682 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
4683 of string_sprintf() is now avoided.
4685 18. $message_body_size was not set (it was always zero) when running the DATA
4686 ACL and the local_scan() function.
4688 19. For the "mail" command in an Exim filter, no default was being set for
4689 the once_repeat time, causing a random time value to be used if "once" was
4690 specified. (If the value happened to be <= 0, no repeat happened.) The
4691 default is now 0s, meaning "never repeat". The "vacation" command was OK
4692 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
4693 (I found it when inspecting the code).
4695 20. There is now an overall timeout for performing a callout verification. It
4696 defaults to 4 times the callout timeout, which applies to individual SMTP
4697 commands during the callout. The overall timeout applies when there is more
4698 than one host that can be tried. The timeout is checked before trying the
4699 next host. This prevents very long delays if there are a large number of
4700 hosts and all are timing out (e.g. when the network connections are timing
4701 out). The value of the overall timeout can be changed by specifying an
4702 additional sub-option for "callout", called "maxwait". For example:
4704 verify = sender/callout=5s,maxwait=20s
4706 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
4707 to the end before writing, but this should make it even safer).
4709 22. Exim was forgetting that it had advertised PIPELINING for the second and
4710 subsequent messages on an SMTP connection. It was also not resetting its
4711 memory on STARTTLS and an internal HELO.
4713 23. When Exim logs an SMTP synchronization error within a session, it now
4714 records whether PIPELINING has been advertised or not.
4716 24. Added 3 instances of "(long int)" casts to time_t variables that were being
4717 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
4718 rather than long int.
4720 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
4722 26. Added the never_mail option to autoreply.
4728 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
4729 crash if the getsockname() call failed; this can happen if a connection is
4730 closed very soon after it is established. The problem was simply in the
4731 order in which certain operations were done, causing Exim to try to write
4732 to the SMTP stream before it had set up the file descriptor. The bug has
4733 been fixed by making things happen in the correct order.
4739 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
4740 before the connection was closed, thus losing the rejection response.
4742 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
4743 some early Solaris releases, but causes trouble in current releases where
4744 socklen_t is defined.
4746 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
4749 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
4752 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
4753 that releasing the top of it at the end releases what was used for sub-
4754 expansions (unless the block got too big). However, discard this block if
4755 the first thing is a variable or header, so that we can use its block when
4756 it is dynamic (useful for very large $message_headers, for example).
4758 6. Lookups now cache *every* query, not just the most recent. A new, separate
4759 store pool is used for this. It can be recovered when all lookup caches are
4760 flushed. Lookups now release memory at the end of their result strings.
4761 This has involved some general refactoring of the lookup sources.
4763 7. Some code has been added to the store_xxx() functions to reduce the amount
4764 of flapping under certain conditions.
4766 8. log_incoming_interface used to affect only the <= reception log lines. Now
4767 it causes the local interface and port to be added to several more SMTP log
4768 lines, for example "SMTP connection from", and rejection lines.
4770 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
4772 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
4774 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
4775 could be overwritten at the end of the current message (or the start of a
4776 new message if it was set in a HELO ACL). The value is now preserved for
4777 the duration of the SMTP connection.
4779 12. If a transport had a headers_rewrite setting, and a matching header line
4780 contained an unqualified address, that address was qualified, even if it
4781 did not match any rewriting rules. The underlying bug was that the values
4782 of the flags that permit the existence of unqualified sender and recipient
4783 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
4784 non-local messages, and by -bnq for local messages) were not being
4785 preserved with the message after it was received.
4787 13. When Exim was logging an SMTP synchronization error, it could sometimes log
4788 "next input=" as part of the text comprising the host identity instead of
4789 the correct text. The code was using the same buffer for two different
4790 strings. However, depending on which order the printing function evaluated
4791 its arguments, the bug did not always show up. Under Linux, for example, my
4792 test suite worked just fine.
4794 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
4795 doesn't work with some older versions of Perl. It has been changed to "my",
4796 which in any case is probably the better facility to use.
4798 15. A really picky compiler found some instances of statements for creating
4799 error messages that either had too many or two few arguments for the format
4802 16. The size of the buffer for calls to the DNS resolver has been increased
4803 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
4804 for addresses that have a lot of PTR records. This alleviates a problem; it
4805 does not fully solve it.
4807 17. A dnsdb lookup for PTR records that receives more data than will fit in the
4808 buffer now truncates the list and logs the incident, which is the same
4809 action as happens when Exim is looking up a host name and its aliases.
4810 Previously in this situation something unpredictable would happen;
4811 sometimes it was "internal error: store_reset failed".
4813 18. If a server dropped the connection unexpectedly when an Exim client was
4814 using GnuTLS and trying to read a response, the client delivery process
4815 crashed while trying to generate an error log message.
4817 19. If a "warn" verb in an ACL added multiple headers to a message in a single
4818 string, for example:
4820 warn message = H1: something\nH2: something
4822 the text was added as a single header line from Exim's point of view
4823 though it ended up OK in the delivered message. However, searching for the
4824 second and subsequent header lines using $h_h2: did not work. This has been
4825 fixed. Similarly, if a system filter added multiple headers in this way,
4826 the routers could not see them.
4828 20. Expanded the error message when iplsearch is called with an invalid key to
4829 suggest using net-iplsearch in a host list.
4831 21. When running tests using -bh, any delays imposed by "delay" modifiers in
4832 ACLs are no longer actually imposed (and a message to that effect is
4835 22. If a "gecos" field in a passwd entry contained escaped characters, in
4836 particular, if it contained a \" sequence, Exim got it wrong when building
4837 a From: or a Sender: header from that name. A second bug also caused
4838 incorrect handling when an unquoted " was present following a character
4839 that needed quoting.
4841 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
4842 was not being matched caselessly.
4844 24. Arranged for all hyphens in the exim.8 source to be escaped with
4847 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
4848 properly. Recipient callout cache records were still being keyed to include
4849 the sender, even when use_sender was set false. This led to far more
4850 callouts that were necessary. The sender is no longer included in the key
4851 when use_sender is false.
4853 26. Added "control = submission" modifier to ACLs.
4855 27. Added the ${base62d: operator to decode base 62 numbers.
4857 28. dnsdb lookups can now access SRV records.
4859 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
4860 the configuration file.
4862 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
4863 (-v) mode. This makes the output for a verbose queue run more intelligible.
4865 31. Added a use_postmaster feature to recipient callouts.
4867 32. Added the $body_zerocount variable, containing the number of binary zero
4868 bytes in the message body.
4870 33. The time of last modification of the "new" subdirectory is now used as the
4871 "mailbox time last read" when there is a quota error for a maildir
4874 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
4876 35. Added +ignore_unknown as a special item in host lists.
4878 36. Code for decoding IPv6 addresses in host lists is now included, even if
4879 IPv6 support is not being compiled. This fixes a bug in which an IPv6
4880 address was recognized as an IP address, but was then not correctly decoded
4881 into binary, causing unexpected and incorrect effects when compared with
4888 1. Very minor rewording of debugging text in manualroute to say "list of
4889 hosts" instead of "hostlist".
4891 2. If verify=header_syntax was set, and a header line with an unqualified
4892 address (no domain) and a large number of spaces between the end of the
4893 name and the colon was received, the reception process suffered a buffer
4894 overflow, and (when I tested it) crashed. This was caused by some obsolete
4895 code that should have been removed. The fix is to remove it!
4897 3. When running in the test harness, delay a bit after writing a bounce
4898 message to get a bit more predictability in the log output.
4900 4. Added a call to search_tidyup() just before forking a reception process. In
4901 theory, someone could use a lookup in the expansion of smtp_accept_max_
4902 per_host which, without the tidyup, could leave open a database connection.
4904 5. Added the variables $recipient_data and $sender_data which get set from a
4905 lookup success in an ACL "recipients" or "senders" condition, or a router
4906 "senders" option, similar to $domain_data and $local_part_data.
4908 6. Moved the writing of debug_print from before to after the "senders" test
4911 7. Change 4.31/66 (moving the time when the Received: is generated) caused
4912 problems for message scanning, either using a data ACL, or using
4913 local_scan() because the Received: header was not generated till after they
4914 were called (in order to set the time as the time of reception completion).
4915 I have revised the way this works. The header is now generated after the
4916 body is received, but before the ACL or local_scan() are called. After they
4917 are run, the timestamp in the header is updated.
4923 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
4924 before starting a queue runner without re-exec. This happened only when
4925 deliver_drop_privilege was set or when the Exim user was set to root. The
4926 effect of the bug was that timeouts during subsequent deliveries caused
4927 crashes instead of being properly handled. The handler is now left at its
4928 default (and expected) setting.
4930 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
4931 message, again when deliver_drop_privilege is set or Exim is run as root.
4932 The bug described in (1) was not present in this case, but the tidying up
4933 of the other signals was missing. I have made the two cases consistent.
4935 3. The ignore_target_hosts setting on a manualroute router was being ignored
4936 for hosts that were looked up using the /MX notation.
4938 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
4941 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
4942 operated on the sender address. After changing the $sender_address to <>
4943 for the sender address verify, Exim was re-instated it as the original
4944 (before rewriting) address, but remembering that it had rewritten it, so it
4945 wasn't rewriting it again. This bug also had the effect of breaking the
4946 sender address verification caching when the sender address was rewritten.
4948 6. The ignore_target_hosts option was being ignored by the ipliteral router.
4949 This has been changed so that if the ip literal address matches
4950 ignore_target_hosts, the router declines.
4952 7. Added expansion conditions match_domain, match_address, and match_local_
4953 part (NOT match_host).
4955 8. The placeholder for the Received: header didn't have a length field set.
4957 9. Added code to Exim itself and to exim_lock to test for a specific race
4958 condition that could lead to file corruption when using MBX delivery. The
4959 issue is with the lockfile that is created in /tmp. If this file is removed
4960 after a process has opened it but before that process has acquired a lock,
4961 there is the potential for a second process to recreate the file and also
4962 acquire a lock. This could lead to two Exim processes writing to the file
4963 at the same time. The added code performs the same test as UW imapd; it
4964 checks after acquiring the lock that its file descriptor still refers to
4965 the same named file.
4967 10. The buffer for building added header lines was of fixed size, 8192 bytes.
4968 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
4971 11. Added the smtp_active_hostname option. If used, this will typically be made
4972 to depend on the incoming interface address. Because $interface_address is
4973 not set up until the daemon has forked a reception process, error responses
4974 that can happen earlier (such as "too many connections") no longer contain
4977 12. If an expansion in a condition on a "warn" statement fails because a lookup
4978 defers, the "warn" statement is abandoned, and the next ACL statement is
4979 processed. Previously this caused the whole ACL to be aborted.
4981 13. Added the iplsearch lookup type.
4983 14. Added ident_timeout as a log selector.
4985 15. Added tls_certificate_verified as a log selector.
4987 16. Added a global option tls_require_ciphers (compare the smtp transport
4988 option of the same name). This controls incoming TLS connections.
4990 17. I finally figured out how to make tls_require_ciphers do a similar thing
4991 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
4992 before starting the TLS session.
4994 18. Tabs are now shown as \t in -bP output.
4996 19. If the log selector return_path_on_delivery was set, Exim crashed when
4997 bouncing a message because it had too many Received: header lines.
4999 20. If two routers both had headers_remove settings, and the first one included
5000 a superfluous trailing colon, the final name in the first list and the
5001 first name in the second list were incorrectly joined into one item (with a
5002 colon in the middle).
5008 1. Added -C and -D options to the exinext utility, mainly to make it easier
5009 to include in the automated testing, but these could be helpful when
5010 multiple configurations are in use.
5012 2. The exinext utility was not formatting the output nicely when there was
5013 an alternate port involved in the retry record key, nor when there was a
5014 message id as well (for retries that were specific to a specific message
5015 and a specific host). It was also confused by IPv6 addresses, because of
5016 the additional colons they contain. I have fixed the IPv4 problem, and
5017 patched it up to do a reasonable job for IPv6.
5019 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
5020 delivery, the log line now contains "pipelined" if PIPELINING was used.
5022 4. An SMTP transport process used to panic and die if the bind() call to set
5023 an explicit outgoing interface failed. This has been changed; it is now
5024 treated in the same way as a connect() failure.
5026 5. A reference to $sender_host_name in the part of a conditional expansion
5027 that was being skipped was still causing a DNS lookup. This no longer
5030 6. The def: expansion condition was not recognizing references to header lines
5031 that used bh_ and bheader_.
5033 7. Added the _cache feature to named lists.
5035 8. The code for checking quota_filecount in the appendfile transport was
5036 allowing one more file than it should have been.
5038 9. For compatibility with Sendmail, the command line option
5046 and sets the incoming protocol and host name (for trusted callers). The
5047 host name and its colon can be omitted when only the protocol is to be set.
5048 Note the Exim already has two private options, -pd and -ps, that refer to
5049 embedded Perl. It is therefore impossible to set a protocol value of "d" or
5050 "s", but I don't think that's a major issue.
5052 10. A number of refactoring changes to the code, none of which should affect
5055 (a) The number of logging options was getting close to filling up the
5056 32-bit word that was used as a bit map. I have split them into two classes:
5057 those that are passed in the argument to log_write(), and those that are
5058 only ever tested independently outside of that function. These are now in
5059 separate 32-bit words, so there is plenty of room for expansion again.
5060 There is no change in the user interface or the logging behaviour.
5062 (b) When building, for example, log lines, the code previously used a
5063 macro that called string_cat() twice, in order to add two strings. This is
5064 not really sufficiently general. Furthermore, there was one instance where
5065 it was actually wrong because one of the argument was used twice, and in
5066 one call a function was used. (As it happened, calling the function twice
5067 did not affect the overall behaviour.) The macro has been replaced by a
5068 function that can join an arbitrary number of extra strings onto a growing
5071 (c) The code for expansion conditions now uses a table and a binary chop
5072 instead of a serial search (which was left over from when there were very
5073 few conditions). Also, it now recognizes conditions like "pam" even when
5074 the relevant support is not compiled in: a suitably worded error message is
5075 given if an attempt is made to use such a condition.
5077 11. Added ${time_interval:xxxxx}.
5079 12. A bug was causing one of the ddress fields not to be passed back correctly
5080 from remote delivery subprocesses. The field in question was not being
5081 subsequently used, so this caused to problems in practice.
5083 13. Added new log selectors queue_time and deliver_time.
5085 14. Might have fixed a bug in maildirsizefile handling that threw up
5086 "unexpected character" debug warnings, and recalculated the data
5087 unnecessarily. In any case, I expanded the warning message to give more
5090 15. Added the message "Restricted characters in address" to the statements in
5091 the default ACL that block characters like @ and % in local parts.
5093 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
5094 Three changes have been made:
5096 (a) There was a serious bug; a negative response to MAIL caused the whole
5097 recipient domain to be cached as invalid, thereby blocking all messages
5098 to all local parts at the same domain, from all senders. This bug has
5099 been fixed. The domain is no longer cached after a negative response to
5100 MAIL if the sender used is not empty.
5102 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
5105 (c) A new callout option, "use_sender" has been added for people who want
5106 the modified behaviour.
5112 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
5115 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
5116 indeed breaks things for older releases.
5118 3. Added additional logging to the case where there is a problem reading data
5119 from a filter that is running in a subprocess using a pipe, in order to
5120 try to track down a specific problem.
5122 4. Testing facility fudge: when running in the test harness and attempting
5123 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
5124 getting "No route to host". Convert this to a timeout.
5126 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
5129 6. Some OS don't have socklen_t but use size_t instead. This affects the
5130 fifth argument of getsockopt() amongst other things. This is now
5131 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
5132 can be set for individual OS. I have set it for SunOS5, OSF1, and
5133 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
5134 some earlier ones do not.
5136 7. Change 4.30/15 was not doing the test caselessly.
5138 8. The standard form for an IPv6 address literal was being rejected by address
5139 parsing in, for example, MAIL and RCPT commands. An example of this kind of
5140 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5141 this, as well as the form without the "IPv6" on the front (but only when
5142 address literals are enabled, of course).
5144 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
5146 10. Exim crashed if a message with an empty sender address specified by -f
5147 encountered a router with an errors_to setting. This could be provoked only
5148 by a command such as
5152 where an empty string was supplied; "<>" did not hit this bug.
5154 11. Installed PCRE release 4.5.
5156 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
5157 remained set. It is now erased.
5159 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
5160 times from message ids (which are base 36 rather than the normal 62).
5162 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
5163 were being counted as actual protocol errors, and logged if the log
5164 selector +smtp_protocol_error was set. One cannot be perfect in this test,
5165 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
5166 and DATA following a set of rejected RCPTs do not count as protocol errors.
5167 In other words, Exim assumes they were pipelined, though this may not
5168 actually be the case. Of course, in all cases the client gets an
5169 appropriate error code.
5171 15. If a lookup fails in an ACL condition, a message about the failure may
5172 be available; it is used if testing the ACL cannot continue, because most
5173 such messages specify what the cause of the deferral is. However, some
5174 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
5175 that caused an old message to be retained and used if a later statement
5176 caused a defer, replacing the real cause of the deferral.
5178 16. If an IP address had so many PTR records that the DNS lookup buffer
5179 was not large enough to hold them, Exim could crash while trying to process
5180 the truncated data. It now detects and logs this case.
5182 17. Further to 4.21/58, another change has been made: if (and only if) the
5183 first line of a message (the first header line) ends with CRLF, a bare LF
5184 in a subsequent header line has a space inserted after it, so as not to
5185 terminate the header.
5187 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
5188 unnecessarily, used atoi() instead of strtol(), and didn't check the
5189 termination when getting file sizes from file names by regex.
5191 19. Completely re-implemented the support for maildirsize files, in the light
5192 of a number of problems with the previous contributed implementation
5193 (4.30/29). In particular:
5195 . If the quota is zero, the maildirsize file is maintained, but no quota is
5198 . If the maildir directory does not exist, it is created before any attempt
5199 to write a maildirsize file.
5201 . The quota value in the file is just a cache; if the quota is changed in
5202 the transport, the new value overrides.
5204 . A regular expression is available for excluding directories from the
5207 20. The autoreply transport checks the characters in options that define the
5208 message's headers; it allows continued headers, but it was checking with
5209 isspace() after an embedded newline instead of explicitly looking for a
5212 21. If all the "regular" hosts to which an address was routed had passed their
5213 expiry times, and had not reached their retry times, the address was
5214 bounced, even if fallback hosts were defined. Now Exim should go on to try
5217 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
5218 equivalent code in the SMTP transport. Some hosts send humungous responses
5219 to HELO/EHLO, more than 1024 it seems.
5221 23. Refactoring: code in filter.c used (void *) for "any old type" but this
5222 gives compiler warnings in some environments. I've now done it "properly",
5225 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
5226 (because of problems with the built-in one) was declared to return uschar *
5227 instead of char *, causing compiler failure.
5229 25. Fixed a file descriptor leak when processing alias/forward files.
5231 26. Fixed a minor format string issue in dbfn.c.
5233 27. Typo in exim.c: ("dmbnz" for "dbmnz").
5235 28. If a filter file refered to $h_xxx or $message_headers, and the headers
5236 contained RFC 2047 "words", Exim's memory could, under certain conditions,
5239 29. When a sender address is verified, it is cached, to save repeating the test
5240 when there is more than one recipient in a message. However, when the
5241 verification involves a callout, it is possible for different callout
5242 options to be set for different recipients. It is too complicated to keep
5243 track of this in the cache, so now Exim always runs a verification when a
5244 callout is required, relying on the callout cache for the optimization.
5245 The overhead is duplication of the address routing, but this should not be
5248 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
5249 to be verified with callout=postmaster, and the main callout worked but the
5250 postmaster check failed, the verification correctly failed. However, if a
5251 subsequent RCPT command asked for sender verification *without* the
5252 postmaster check, incorrect caching caused this verification also to fail,
5255 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
5256 it was not caching the DNS options (qualify_single, search_parents) that
5257 were used when the lookup failed. A subsequent lookup with different
5258 options therefore always gave the same answer, though there were cases
5259 where it should not have. (Example: a "domains = !$mx_any" option on a
5260 dnslookup router: the "domains" option is always processed without any
5261 widening, but the router might have qualify_single set.) Now Exim uses the
5262 cached value only when the same options are set.
5264 32. Added John Jetmore's "exipick" utility to the distribution.
5266 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
5267 than a timeout (e.g. a certificate is required, and is not provided), an
5268 Exim server now closes the connection immediately. Previously it waited for
5269 the client to close - but if the client is SSL, it seems that they each
5270 wait for each other, leading to a delay before one of them times out.
5272 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
5273 maintained 0.8.x compatibility because I don't think many are using it, and
5274 it is clearly obsolete.
5276 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
5279 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
5280 client certificate was expired. A simple patch fixes this, though I don't
5281 understand the full logic of why the verify callback is called multiple
5284 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
5285 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
5286 which causes problems with some clients (such as the Certicom SSL Plus
5287 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
5288 disables the coutermeasure allowing Eudora to connect."
5290 38. Exim was not checking that a write() to a log file succeeded. This could
5291 lead to Bad Things if a log got too big, in particular if it hit a file
5292 size limit. Exim now panics and dies if it cannot write to a log file, just
5293 as it does if it cannot open a log file.
5295 39. Modified OS/Makefile-Linux so that it now contains
5297 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
5299 The two -D definitions ensure that Exim is compiled with large file
5300 support, which makes it possible to handle log files that are bigger than
5303 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
5304 instance) a domain was checked against a named list that involved a lookup,
5305 causing $domain_data to be set, then another domain was checked against the
5306 same list, then the first domain was re-checked, the value of $domain_data
5307 after the final check could be wrong. In particular, if the second check
5308 failed, it could be set empty. This bug probably also applied to
5311 41. The strip_trailing_dot option was not being applied to the address given
5312 with the -f command-line option.
5314 42. The code for reading a message's header from the spool was incrementing
5315 $received_count, but never initializing it. This meant that the value was
5316 incorrect (doubled) while delivering a message in the same process in which
5317 it was received. In the most common configuration of Exim, this never
5318 happens - a fresh exec is done - but it can happen when
5319 deliver_drop_privilege is set.
5321 43. When Exim logs an SMTP synchronization error - client data sent too soon -
5322 it now includes up to 150 characters of the unexpected data in the log
5325 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
5326 and building data strings. The size of both of these buffers was 10 000
5327 bytes - far larger than anybody would *ever* want, thought I. Needless to
5328 say, somebody hit the limit. I have increased the maximum line length to
5329 20 000 and the maximum data length of concatenated lines to 100 000. I have
5330 also fixed two bugs, because there was no checking on these buffers. Tsk,
5331 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
5332 buffer is too small.
5334 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
5335 lsearch lookups. Now it does.
5337 46. When parsing a route_list item in a manualroute router, a fixed-length
5338 buffer was used for the list of hosts. I made this 1024 bytes long,
5339 thinking that nobody would ever have a list of hosts that long. Wrong.
5340 Somebody had a whole pile of complicated expansion conditions, and the
5341 string was silently truncated, leading to an expansion error. It turns out
5342 that it is easier to change to an unlimited length (owing to other changes
5343 that have happened since this code was originally written) than to build
5344 structure for giving a limitation error. The length of the item that
5345 expands into the list of hosts is now unlimited.
5347 47. The lsearch lookup could not handle data where the length of text line was
5348 more than 4095 characters. Such lines were truncated, leading to shortened
5349 data being returned. It should now handle lines of any length.
5351 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
5352 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
5355 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
5356 replacing strings such as BIN_DIRECTORY in a source file: the replacement
5357 no longer happens in comment lines. A list of replacements is now placed
5358 at the head of all of the source files, except those whose only change is
5359 to replace PERL_COMMAND in the very first #! line.
5361 50. Replaced the slow insertion sort in queue.c, for sorting the list of
5362 messages on the queue, with a bottom-up merge sort, using code contributed
5363 by Michael Haardt. This should make operations like -bp somewhat faster on
5364 large queues. It won't affect queue runners, except when queue_run_in_order
5367 51. Installed eximstats 1.31 in the distribution.
5369 52. Added support for SRV lookups to the dnslookup router.
5371 53. If an ACL referred to $message_body or $message_body_end, the value was not
5372 reset for any messages that followed in the same SMTP session.
5374 54. The store-handling optimization for building very long strings was not
5375 differentiating between the different store pools. I don't think this
5376 actually made any difference in practice, but I've tidied it.
5378 55. While running the routers to verify a sender address, $sender_address
5379 was still set to the sender address. This is wrong, because when routing to
5380 send a bounce to the sender, it would be empty. Therefore, I have changed
5381 it so that, while verifying a sender address, $sender_address is set to <>.
5382 (There is no change to what happens when verifying a recipient address.)
5384 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
5385 target A or AAAA records (if not already returned) without resetting the
5386 qualify_single or search_parents options of the DNS resolver. These are
5387 inappropriate in this case because the targets of MX and SRV records must
5388 be FQDNs. A broken DNS record could cause trouble if it happened to have a
5389 target that, when qualified, matched something in the local domain. These
5390 two options are now turned off when doing these lookups.
5392 57. It seems that at least some releases of Reiserfs (which does not have the
5393 concept of a fixed number of inodes) returns zero and not -1 for the
5394 number of available inodes. This interacted badly with check_spool_inodes,
5395 which assumed that -1 was the "no such thing" setting. What I have done is
5396 to check that the total number of inodes is greater than zero before doing
5397 the test of how many are available.
5399 58. When a "warn" ACL statement has a log_message modifier, the message is
5400 remembered, and not repeated. This is to avoid a lot of repetition when a
5401 message has many recipients that cause the same warning to be written.
5402 Howewer, Exim was preserving the list of already written lines for an
5403 entire SMTP session, which doesn't seem right. The memory is now reset if a
5404 new message is started.
5406 59. The "rewrite" debugging flag was not showing the result of rewriting in the
5407 debugging output unless log_rewrite was also set.
5409 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
5410 of (int)(handle) when we know that handle contains (void *)(-1).
5412 61. The Exim daemon panic-logs an error return when it closes the incoming
5413 connection. However "connection reset by peer" seems to be common, and
5414 isn't really an error worthy of noting specially, so that particular error
5417 62. When Exim is trying to find all the local interfaces, it used to panic and
5418 die if the ioctl to get the interface flags failed. However, it seems that
5419 on at least one OS (Solaris 9) it is possible to have an interface that is
5420 included in the list of interfaces, but for which you get a failure error
5421 for this call. This happens when the interface is not "plumbed" into a
5422 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
5423 failure of the "get flags" call assumes that the interface is down.
5425 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
5426 makes life easier for people who are doing arithmetic on fields extracted
5427 from dates, where you often get leading zeros that should not be
5428 interpreted as octal.
5430 64. Added qualify_domain to the redirect router, to override the global
5433 65. If a pathologically long header line contained very many addresses (the
5434 report of this problem mentioned 10 000) and each of them was rewritten,
5435 Exim could use up a very large amount of memory. (It kept on making new
5436 copies of the header line as it rewrote, and never released the old ones.)
5437 At the expense of a bit more processing, the header rewriting function has
5438 been changed so that it no longer eats memory in this way.
5440 66. The generation of the Received: header has been moved from the time that a
5441 message starts to be received, to the time that it finishes. The timestamp
5442 in the Received: header should now be very close to that of the <= log
5443 line. There are two side-effects of this change:
5445 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
5446 logged header lines no longer include the local Received: line, because
5447 it has not yet been created. The same applies to a copy of the message
5448 that is returned to a non-SMTP sender when a message is rejected.
5450 (b) When a filter file is tested using -bf, no additional Received: header
5451 is added to the test message. After some thought, I decided that this
5454 This change does not affect the value of $received_for. It is still set
5455 after address rewriting, but before local_scan() is called.
5457 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
5459 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
5460 gave an unhelpful panic error message, and a defer error. I have managed to
5461 change this behaviour so that it now rejects any supplied certificate,
5462 which seems right, as the list of acceptable certificates is empty.
5464 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
5465 gave an unhelpful defer error. I have not managed to make this reject any
5466 supplied certificates, but the error message it gives is "no certificate
5467 supplied", which is not helpful.
5469 70. exigrep's output now also includes lines that are not associated with any
5470 message, but which match the given pattern. Implemented by a patch from
5471 Martin Sluka, which also tidied up the Perl a bit.
5473 71. Recipient callout verification, like sender verification, was using <> in
5474 the MAIL FROM command. This isn't really the right thing, since the actual
5475 sender may affect whether the remote host accepts the recipient or not. I
5476 have changed it to use the actual sender in the callout; this means that
5477 the cache record is now keyed on a recipient/sender pair, not just the
5478 recipient address. There doesn't seem to be a real danger of callout loops,
5479 since a callout by the remote host to check the sender would use <>.
5480 [SEE ABOVE: changed after hitting problems.]
5482 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
5483 temporary errors. However, in the case of such a code being given after
5484 the end of a data transmission (i.e. after ".") Exim was failing to write
5485 a retry record for the message. (Yes, there was some broken host that was
5486 actually sending 8xx at this point.)
5488 73. An unknown lookup type in a host list could cause Exim to panic-die when
5489 the list was checked. (An example that provoked this was putting <; in the
5490 middle of a list instead of at the start.) If this happened during a DATA
5491 ACL check, a -D file could be left lying around. This kind of configuration
5492 error no longer causes Exim to die; instead it causes a defer errror. The
5493 incident is still logged to the main and panic logs.
5495 74. Buglet left over from Exim 3 conversion. The message "too many messages
5496 in one connection" was written to the rejectlog but not the mainlog, except
5497 when address rewriting (yes!) was being logged.
5499 75. Added write_rejectlog option.
5501 76. When a system filter was run not as root (that is, when system_filter_user
5502 was set), the values of the $n variables were not being returned to the
5503 main process; thus, they were not subsequently available in the $sn
5506 77. Added +return_path_on_delivery log selector.
5508 78. A connection timeout was being treated differently from recipients deferred
5509 when testing hosts_max_try with a message that was older than the host's
5510 retry timeout. (The host should not be counted, thus allowing all hosts to
5511 be tried at least once before bouncing.) This may have been the cause of an
5512 occasionally reported bug whereby a message would remain on the queue
5513 longer than the retry timeout, but would be bounced if a delivery was
5514 forced. I say "may" because I never totally pinned down the problem;
5515 setting up timeout/retry tests is difficult. See also the next item.
5517 79. The ultimate address timeout was not being applied to errors that involved
5518 a combination of host plus message (for example, a timeout on a MAIL
5519 command). When an address resolved to a number of possible hosts, and they
5520 were not all tried for each delivery (e.g. because of hosts_max_try), a
5521 message could remain on the queue longer than the retry timeout.
5523 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
5526 81. Fixed an obscure SMTP outgoing bug which required at least the following
5527 conditions: (a) there was another message waiting for the same server;
5528 (b) the server returned 5xx to all RCPT commands in the first message so
5529 that the message was not completed; (c) the server dropped the connection
5530 or gave a negative response to the RSET that Exim sends to abort the
5531 transaction. The observed case was a dropped connection after DATA that had
5532 been sent in pipelining mode. That is, the server had advertised PIPELINING
5533 but was not implementing it correctly. The effect of the bug was incorrect
5534 behaviour, such as trying another host, and this could lead to a crash.
5540 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
5541 and daemon.c were passed as pointers to ints; they should have been
5542 pointers to socklen_t variables (which are typically unsigned ints).
5544 2. Some signed/unsigned type warnings in the os.c file for Linux have been
5547 3. Fixed a really odd bug that affected only the testing scheme; patching a
5548 certain fixed string in the binary changed the value of another string that
5549 happened to be identical to the end of the original first string.
5551 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
5552 name", it returns that address as the IP address. On some operating
5553 systems (e.g. Solaris), it also passes back the IP address string as the
5554 "host name". However, on others (e.g. Linux), it passes back an empty
5555 string. Exim wasn't checking for this, and was changing the host name to an
5556 empty string, assuming it had been canonicized.
5558 5. Although rare, it is permitted to have more than one PTR record for a given
5559 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
5560 all the names associated with an address, because they do in Solaris.
5561 However, it seems that they do not in Linux for data that comes from the
5562 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
5563 I found this out when I moved to a new Linux workstation and tried to run
5564 the Exim test suite.
5566 To get round this problem I have changed the code so that it now does its
5567 own call to the DNS to look up PTR records when searching for a host name.
5568 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
5569 addresses that are only in /etc/hosts are still found.
5571 This behaviour is, however, controlled by an option called host_lookup_
5572 order, which defaults to "bydns:byaddr". If people want to use the other
5573 order, or indeed, just use one or the other means of lookup, they can
5574 specify it in this variable.
5576 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
5577 some operating systems, this comes back from gethostbyaddr() as an empty
5578 string, and this is what Exim used to test for. However, it seems that in
5579 other systems, "." is yielded. Exim now tests for this case too.
5581 7. The values of check_spool_space and check_log_space are now held internally
5582 as a number of kilobytes instead of an absolute number of bytes. If a
5583 numbers is specified without 'K' or 'M', it is rounded up to the nearest
5584 kilobyte. This means that much larger values can be stored.
5586 8. Exim monitor: an attempt to get the action menu when not actually pointing
5587 at a message produces an empty menu entitled "No message selected". This
5588 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
5589 no entries in it ("Shell widget menu has zero width and/or height"). So I
5590 have added a single, blank menu entry in this case.
5592 9. Added ${quote_local_part.
5594 10. MIME decoding is now applied to the contents of Subject: header lines when
5597 11. Now that a reference to $sender_host_address automatically causes a reverse
5598 lookup to occur if necessary (4.13/18), there is no need to arrange for a
5599 host lookup before query-style lookups in lists that might use this
5600 variable. This has therefore been abolished, and the "net-" prefix is no
5601 longer necessary for query-style lookups.
5603 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
5604 have been a typo for LFLAGS, so it has been changed.
5606 13. The install script calls Exim with "-C /dev/null" in order to find the
5607 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
5608 to be output. Howeve, since Exim outputs its version number before the
5609 error, it didn't break the script. It just looked ugly. I fixed this by
5610 always allowing "-C /dev/null" if the caller is root.
5612 14. Ignore overlarge ACL variable number when reading spool file - insurance
5613 against a later release with more variables having written the file.
5615 15. The standard form for an IPv6 address literal was being rejected by EHLO.
5616 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5617 this, as well as the form without the "IPv6" on the front.
5619 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
5620 OS/Makefile-Darwin file.
5622 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
5623 only to LDAP libraries that do not have LDAP_OPT_DEREF.
5625 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
5626 variable. However, some OS use ints for inodes. I've added cast to long int
5627 to get rid of the compiler warning.
5629 19. I had forgotten to lock out "/../" in configuration file names when
5630 ALT_CONFIG_PREFIX was set.
5632 20. Routers used for verification do not need to specify transports. However,
5633 if such a router generated a host list, and callout was configured, Exim
5634 crashed, because it could not find a port number from the (non-existent)
5635 transport. It now assumes port 25 in this circumstance.
5637 21. Added the -t option to exigrep.
5639 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
5640 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
5641 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
5642 them is set, LOOKUP_LSEARCH is forced.
5644 23. "exim -bV" now outputs a list of lookups that are included in the binary.
5646 24. Added sender and host information to the "rejected by local_scan()" log
5647 line; previously there was no indication of these.
5649 25. Added .include_if_exists.
5651 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
5652 Linux. It turns out that not all file systems support this. Apparently some
5653 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
5654 do it.) To cope with this, the error EINVAL, which means that sync-ing is
5655 not supported on the file descriptor, is now ignored when Exim is trying to
5656 sync a directory. This applies only to Linux.
5658 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
5660 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
5661 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
5664 29. Added support for maildirsize files from supplied patch (modified a bit).
5666 30. The use of :fail: followed by an empty string could lead Exim to respond to
5667 sender verification failures with (e.g.):
5669 550 Verification failed for <xxx>
5670 550 Sender verify failed
5672 where the first response line was missing the '-' that indicates it is not
5673 the final line of the response.
5675 31. The loop for finding the name of the user that called Exim had a hardwired
5676 limit of 10; it now uses the value of finduser_retries, which is used for
5677 all other user lookups.
5679 32. Added $received_count variable, available in data and not_smtp ACLs, and at
5682 33. Exim was neglecting to zero errno before one call of strtol() when
5683 expanding a string and expecting an integer value. On some systems this
5684 resulted in spurious "integer overflow" errors. Also, it was casting the
5685 result into an int without checking.
5687 34. Testing for a connection timeout using "timeout_connect" in the retry rules
5688 did not work. The code looks as if it has *never* worked, though it appears
5689 to have been documented since at least releast 1.62. I have made it work.
5691 35. The "timeout_DNS" error in retry rules, also documented since at least
5692 1.62, also never worked. As it isn't clear exactly what this means, and
5693 clearly it isn't a major issue, I have abolished the feature by treating it
5694 as "timeout", and writing a warning to the main and panic logs.
5696 36. The display of retry rules for -brt wasn't always showing the error code
5699 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
5700 timeout_connect_A, timeout_connect_MX.
5702 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
5703 to the empty sender.
5705 39. The daemon was not analysing the content of -oX till after it had closed
5706 stderr and disconnected from the controlling terminal. This meant that any
5707 syntax errors were only noted on the panic log, and the return code from
5708 the command was 0. By re-arranging the code a little, I've made the
5709 decoding happen first, so such errors now appear on stderr, and the return
5710 code is 1. However, the actual setting up of the sockets still happens in
5711 the disconnected process, so errors there are still only recorded on the
5714 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
5715 connections (as happens on some IP stacks) was logged at start up time as
5716 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
5717 it from "IPv6 and IPv4", which means that two separate sockets are being
5720 41. The debug output for gethostbyname2() or getipnodebyname() failures now
5721 says whether AF_INET or AF_INET6 was passed as an argument.
5723 42. Exiwhat output was messed up when time zones were included in log
5726 43. Exiwhat now gives more information about the daemon's listening ports,
5727 and whether -tls-on-connect was used.
5729 44. The "port" option of the smtp transport is now expanded.
5731 45. A "message" modifier in a "warn" statement in a non-message ACL was being
5732 silently ignored. Now an error message is written to the main and panic
5735 46. There's a new ACL modifier called "logwrite" which writes to a log file
5736 as soon as it is encountered.
5738 47. Added $local_user_uid and $local_user_gid at routing time.
5740 48. Exim crashed when trying to verify a sender address that was being
5743 49. Exim was recognizing only a space character after ".include". It now also
5744 recognizes a tab character.
5746 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
5747 extracting the relevant information from the specification. The man page no
5748 longer contains scrambled data for the -d option, and I've added a section
5749 at the front about calling Exim under different names.
5751 51. Added "extra_headers" argument to the "mail" command in filter files.
5753 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
5756 53. Installed eximstats 1.29.
5758 54. Added transport_filter_timeout as a generic transport option.
5760 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
5761 Cc: header lines. This was required by RFC 822, but it not required by RFC
5764 56. Exim used to add From:, Date:, and Message-Id: header lines to any
5765 incoming messages that did not have them. Now it does so only if the
5766 message originates locally, that is, if there is no associated remote host
5767 address. When Resent- header lines are present, this applies to the Resent-
5768 lines rather than the non-Resent- lines.
5770 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
5771 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
5773 58. Messages for configuration errors now include the name of the main
5774 configuration file - useful now that there may be more than one file in a
5775 list (.included file names were always shown).
5777 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
5778 for those rare installations that do not start the daemon as root or run it
5779 setuid root. I've cut out the call to initgroups() if the daemon is not
5782 60. The Exim user and group can now be bound into the binary as text strings
5783 that are looked up at the start of Exim's processing.
5785 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
5787 62. Added $mailstore_basename variable.
5789 63. Installed patch to sieve.c from Michael Haardt.
5791 64. When Exim failed to open the panic log after failing to open the main log,
5792 the original message it was trying to log was written to stderr and debug
5793 output, but if they were not available (the usual case in production), it
5794 was lost. Now it is written to syslog before the two lines that record the
5795 failures to open the logs.
5797 65. Users' Exim filters run in subprocesses under the user's uid. It is
5798 possible for a "deliver" command or an alias in a "personal" command to
5799 provoke an address rewrite. If logging of address rewriting is configured,
5800 this fails because the process is not running as root or exim. There may be
5801 a better way of dealing with this, but for the moment (because 4.30 needs
5802 to be released), I have disabled address rewrite logging when running a
5803 filter in a non-root, non-exim process.
5809 1. The buildconfig auxiliary program wasn't quoting the value set for
5810 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
5811 not defined. This bug was masked in 4.22 by the effect that was fixed in
5814 2. Some messages that were rejected after a message id was allocated were
5815 shown as "incomplete" by exigrep. It no longer does this for messages that
5816 are rejected by local_scan() or the DATA or non-SMTP ACLs.
5818 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
5819 have allow_domain_literals set, the ID did not get logged in the <= line.
5820 Domain literals are now always recognized in Message-ID: header lines.
5822 4. The first argument for a ${extract expansion item is the key name or field
5823 number. Leading and trailing spaces in this item were not being ignored,
5824 causing some misleading effects.
5826 5. When deliver_drop_privilege was set, single queue runner processes started
5827 manually (i.e. by the command "exim -q") or by the daemon (which uses the
5828 same command in the process it spins off) were not dropping privilege.
5830 6. When the daemon running as "exim" started a queue runner, it always
5831 re-executed Exim in the spun-off process. This is a waste of effort when
5832 deliver_drop_privilege is set. The new process now just calls the
5833 queue-runner function directly.
5839 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
5842 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
5843 ignored. Though the use of -oP was forcing the writing of a pid file, it
5844 was always written to the default place.
5846 3. If the message "no IP address found for host xxxx" is generated during
5847 incoming verification, it is now followed by identification of the incoming
5848 connection (so you can more easily find what provoked it).
5850 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
5852 5. Added some features to "harden" Exim a bit more against certain attacks:
5854 (a) There is now a build-time option called FIXED_NEVER_USERS that can
5855 be put in Local/Makefile. This is like the never_users runtime option,
5856 but it cannot be overridden. The default setting is "root".
5858 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
5859 prefix string with which any file named in a -C command line option
5862 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
5863 is retained for -C and -D only if the caller of Exim is root. Without
5864 it, the exim user may also use -C and -D and retain privilege.
5866 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
5867 command line option is disabled.
5869 6. Macro names set by the -D option must start with an upper case letter, just
5870 like macro names defined in the configuration file.
5872 7. Added "dereference=" facility to LDAP.
5874 8. Two instances of the typo "uknown" in the source files are fixed.
5876 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
5877 the Configure-Makefile script screwed up while processing it.
5879 10. Incorporated PCRE 4.4.
5881 11. The SMTP synchronization check was not operating right at the start of an
5882 SMTP session. For example, it could not catch a HELO sent before the client
5883 waited for the greeting. There is now a check for outstanding input at the
5884 point when the greeting is written. Because of the duplex, asynchronous
5885 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
5886 way, but not yet received, when the check is performed.
5888 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
5889 on TCP/IP sockets, because this apparently causes some broken clients to
5892 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
5893 unchanged) from the Cygwin maintainer.
5895 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
5896 was supported instead of testing for mbx. Effectively a typo.
5898 15. The spa authenticator server code was not checking that the input it
5899 received was valid base64.
5901 16. The debug output line for the "set" modifier in ACLs was not showing the
5902 name of the variable that was being set.
5904 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
5906 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
5907 lookup on its own. Something else was needed to trigger the lookup. For
5908 example, a match in host_lookup or the need for a host name in a host list.
5909 Now, if $sender_host_name is referenced and the host name has not yet been
5910 looked up, a lookup is performed. If the lookup fails, the variable remains
5911 empty, and $host_lookup_failed is set to "1".
5913 19. Added "eqi" as a case-independent comparison operator.
5915 20. The saslauthd authentication condition could segfault if neither service
5916 nor realm was specified.
5918 21. If an overflowing value such as "2048M" was set for message_size_limit, the
5919 error message that was logged was misleading, and incoming SMTP
5920 connections were dropped. The message is now more accurate, and temporary
5921 errors are given to SMTP connections.
5923 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
5924 (except RSET) with a 421 error, until QUIT is received. However, it was
5925 failing to send a response to QUIT.
5927 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
5928 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
5929 ACL is now run after the helo_try_verify_hosts code.
5931 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
5932 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
5933 case-independent, so other case variants are also recognized). Apparently
5934 some systems use these upper case variants.
5936 25. If more than two messages were waiting for the same host, and a transport
5937 filter was specified for the transport, Exim sent two messages over the
5938 same TCP/IP connection, and then failed with "socket operation on non-
5939 socket" when it tried to send the third.
5941 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
5943 27. The extern definition of crypt16() in expand.c was not being excluded when
5944 the OS had its own crypt16() function.
5946 28. Added bounce_return_body as a new option, and bounce_return_size_limit
5947 as a preferred synonym for return_size_limit, both as an option and as an
5950 29. Added LIBS=-liconv to OS/Makefile-OSF1.
5952 30. Changed the default configuration ACL to relax the local part checking rule
5953 for addresses that are not in any local domains. For these addresses,
5954 slashes and pipe symbols are allowed within local parts, but the sequence
5955 /../ is explicitly forbidden.
5957 31. SPA server authentication was not clearing the challenge buffer before
5960 32. log_message in a "warn" ACL statement was writing to the reject log as
5961 well as to the main log, which contradicts the documentation and doesn't
5962 seem right (because no rejection is happening). So I have stopped it.
5964 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
5965 However, I am unable to do any testing of this.
5967 34. Fixed an infelicity in the appendfile transport. When checking directories
5968 for a mailbox, to see if any needed to be created, it was accidentally
5969 using path names with one or more superfluous leading slashes; tracing
5970 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
5972 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
5973 contents are added to the log line that is written for every discarded
5974 recipient. (Previously a log_message setting was ignored.)
5976 36. The ${quote: operator now quotes the string if it is empty.
5978 37. The install script runs exim in order to find its version number. If for
5979 some reason other than non-existence or emptiness, which it checks, it
5980 could not run './exim', it was installing it with an empty version number,
5981 i.e. as "exim-". This error state is now caught, and the installation is
5984 38. An argument was missing from the function that creates an error message
5985 when Exim fails to connect to the socket for saslauthd authentication.
5986 This could cause Exim to crash, or give a corrupted message.
5988 39. Added isip, isip4, and isip6 to ${if conditions.
5990 40. The ACL variables $acl_xx are now saved with the message, and can be
5991 accessed later in routers, transports, and filters.
5993 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
5994 strings in the file are not string-expanded.
5996 42. If a MAIL command specified a SIZE value that was too large to fit into an
5997 int variable, the check against message_size_limit failed. Such values are
5998 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
5999 day this will have to be increased, but I don't think I want to be around
6000 when emails are that large.
6007 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
6008 iconv() is not standard in FreeBSD.
6010 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
6011 IPv6 enabled. The observed symptom was a segmentation fault on return from
6012 the function os_common_find_running_interfaces() in src/os.c.
6014 3. In the check_special_case() function in daemon.c I had used "errno" as an
6015 argument name, which causes warnings on some systems. This was basically a
6016 typo, since it was named "eno" in the comments!
6018 4. The code that waits for the clock to tick (at a resolution of some fraction
6019 of a second) so as to ensure message-id uniqueness was always waiting for
6020 at least one whole tick, when it could have waited for less. [This is
6021 almost certainly not relevant at current processor speeds, where it is
6022 unlikely to ever wait at all. But we try to future-proof.]
6024 5. The function that sleeps for a time interval that includes fractions of a
6025 second contained a race. It did not block SIGALRM between setting the
6026 timer, and suspending (a couple of lines later). If the interval was short
6027 and the sigsuspend() was delayed until after it had expired, the suspension
6028 never ended. On busy systems this could lead to processes getting stuck for
6031 6. Some uncommon configurations may cause a lookup to happen in a queue runner
6032 process, before it forks any delivery processes. The open lookup caching
6033 mechanism meant that the open file or database connection was passed into
6034 the delivery process. The problem was that delivery processes always tidy
6035 up cached lookup data. This could cause a problem for the next delivery
6036 process started by the queue runner, because the external queue runner
6037 process does not know about the closure. So the next delivery process
6038 still has data in the lookup cache. In the case of a file lookup, there was
6039 no problem because closing a file descriptor in a subprocess doesn't affect
6040 the parent. However, if the lookup was caching a connection to a database,
6041 the connection was closed, and the second delivery process was likely to
6042 see errors such as "PGSQL: query failed: server closed the connection
6043 unexpectedly". The problem has been fixed by closing all cached lookups
6044 in a queue runner before running a delivery process.
6046 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
6047 seem to have the "const" qualifier which it has on other OS. I've
6050 8. Change 4.21/2 was too strict. It is only if there are two authenticators
6051 *of the same type* (client or server) with the same public name that an
6052 error should be diagnosed.
6054 9. When Exim looked up a host name for an IP address, but failed to find the
6055 original IP address when looking up the host name (a safety check), it
6056 output the message "<ip address> does not match any IP for NULL", which was
6057 confusing, to say the least. The bug was that the host name should have
6058 appeared instead of "NULL".
6060 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
6061 user that is built into the binary, and the -C or -D options is used, root
6062 privilege is dropped before the configuration file is read. In addition,
6063 logging is switched to stderr instead of the normal log files. If the
6064 configuration then re-defines the Exim user, the unprivileged environment
6065 is probably not what is expected, so Exim logs a panic warning message (but
6068 However, if deliver_drop_privilege is set, the unprivileged state may well
6069 be exactly what is intended, so the warning has been cut out in that case,
6070 and Exim is allowed to try to write to its normal log files.
6076 1. smtp_return_error_details was not giving details for temporary sender
6077 or receiver verification errors.
6079 2. Diagnose a configuration error if two authenticators have the same public
6082 3. Exim used not to create the message log file for a message until the first
6083 delivery attempt. This could be confusing when incoming messages were held
6084 for policy or load reasons. The message log file is now created at the time
6085 the message is received, and an initial "Received" line is written to it.
6087 4. The automatically generated man page for command line options had a minor
6088 bug that caused no ill effects; however, a more serious problem was that
6089 the procedure for building the man page automatically didn't always
6090 operate. Consequently, release 4.20 contains an out-of-date version. This
6091 shouldn't happen again.
6093 5. When building Exim with embedded Perl support, the script that builds the
6094 Makefile was calling 'perl' to find its compile-time parameters, ignoring
6095 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
6097 6. The freeze_tell option was not being used for messages that were frozen on
6098 arrival, either by an ACL or by local_scan().
6100 7. Added the smtp_incomplete_transaction log selector.
6102 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
6103 was accepting AUTH without a new EHLO.
6105 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
6106 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
6109 10. Logging of TCP/IP connections (when configured) now happens in the main
6110 daemon process instead of the child process, so that the TCP/IP connection
6111 count is more accurate (but it can never be perfect).
6113 11. The use of "drop" in a nested ACL was not being handled correctly in the
6114 outer ACL. Now, if condition failure induced by the nested "drop" causes
6115 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
6116 or "require"), the connection is dropped.
6118 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
6119 that yield "discard" can now be used with an "accept" or a "discard" verb,
6120 but an error is generated for any others (because I can't see a useful way
6121 to define what should happen).
6123 13. When an ACL is read dynamically from a file (or anywhere else), the lines
6124 are now processed in the same way as lines in the Exim configuration file.
6125 In particular, continuation lines are supported.
6127 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
6129 15. Added -ti meaning -t -i.
6131 16. Check for letters, digits, hyphens, and dots in the names of dnslist
6132 domains, and warn by logging if others are found.
6134 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
6135 returned from GIFCONF when Exim is trying to find the list of interfaces on
6136 a host. The code in os.c has been modified to copy each ifreq to an aligned
6137 structure in all cases.
6139 Also, in some cases, the returned ifreq's were being copied to a 'struct
6140 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
6141 means the last couple of bytes of an IPv6 address could be chopped if the
6142 ifreq contained only a normal sockaddr (14 bytes storage).
6144 18. Named domain lists were not supported in the hosts_treat_as_local option.
6145 An entry such as +xxxx was not recognized, and was treated as a literal
6148 19. Ensure that header lines added by a DATA ACL are included in the reject log
6149 if the ACL subsequently rejects the message.
6151 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
6152 MD5 (which is deprecated).
6154 21. When testing a filter file using -bf, Exim was writing a message when it
6155 took the sender from a "From " line in the message, but it was not doing so
6156 when it took $return_path from a Return-Path: header line. It now does.
6158 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
6159 with a valid header line field name (a series of printing characters
6160 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
6162 23. Changed "disc" in the source to "disk" to conform to the documentation and
6163 the book and for uniformity.
6165 24. Ignore Sendmail's -Ooption=value command line item.
6167 25. When execve() failed while trying to run a command in a pipe transport,
6168 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
6169 could be confused with a return value of 69 from the command itself. This
6170 has been changed to 127, the value the shell returns if it is asked to run
6171 a non-existent command. The wording for the related log line suggests a
6172 non-existent command as the problem.
6174 26. If received_header_text expands to an empty string, do not add a Received:
6175 header line to the message. (Well, it adds a token one on the spool, but
6176 marks it "old" so that it doesn't get used or transmitted.)
6178 27. Installed eximstats 1.28 (addition of -nt option).
6180 28. There was no check for failure on the call to getsockname() in the daemon
6181 code. This can fail if there is a shortage of resources on the system, with
6182 ENOMEM, for example. A temporary error is now given on failure.
6184 29. Contrary to the C standard, it seems that in some environments, the
6185 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
6186 program. Exim now does this explicitly; it affects the formatting of
6187 timestamps using strftime().
6189 30. If exiqsumm was given junk data, it threw up some uninitialized variable
6190 complaints. I've now initialized all the variables, to avoid this.
6192 32. Header lines added by a system filter were not being "seen" during
6193 transport-time rewrites.
6195 33. The info_callback() function passed to OpenSSL is set up with type void
6196 (*)(SSL *, int, int), as described somewhere. However, when calling the
6197 function (actually a macro) that sets it up, the type void(*)() is
6198 expected. I've put in a cast to prevent warnings from picky compilers.
6200 34. If a DNS black list lookup found a CNAME record, but there were no A
6201 records associated with the domain it pointed at, Exim crashed.
6203 35. If a DNS black list lookup returned more than one A record, Exim ignored
6204 all but the first. It now scans all returned addresses if a particular IP
6205 value is being sought. In this situation, the contents of the
6206 $dnslist_value variable are a list of all the addresses, separated by a
6209 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
6210 to accept a host name and all its aliases if the forward lookup for any of
6211 them yielded the IP address of the incoming connection. Now it accepts only
6212 those names whose forward lookup yields the correct IP address. Any other
6213 names are discarded. This closes a loophole whereby a rogue DNS
6214 administrator could create reverse DNS records to break through a
6215 wildcarded host restriction in an ACL.
6217 37. If a user filter or a system filter that ran in a subprocess used any of
6218 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
6219 the wrong values were passed to the pipe command ($thisaddress had the
6220 value of $0, $0 had the value of $1, etc). This bug was introduced by
6221 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
6223 38. Improved the line breaking for long SMTP error messages from ACLs.
6224 Previously, if there was no break point between 40 and 75 characters, Exim
6225 left the rest of the message alone. Two changes have been made: (a) I've
6226 reduced the minimum length to 35 characters; (b) if it can't find a break
6227 point between 35 and 75 characters, it looks ahead and uses the first one
6228 that it finds. This may give the occasional overlong line, but at least the
6229 remaining text gets split now.
6231 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
6232 file descriptors might be low, and that setting 1000 would always raise it.
6233 It turns out that in some environments, the limit is already over 1000 and
6234 that lowering it causes trouble. So now Exim takes care not to decrease it.
6236 40. When delivering a message, the value of $return_path is set to $sender_
6237 address at the start of routing (routers may change the value). By an
6238 oversight, this default was not being set up when an address was tested by
6239 -bt or -bv, which affected the outcome if any router or filter referred to
6242 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
6243 log only when "message" or "log_message" are set. However, if one of the
6244 conditions was an address verification, or a call to a nested ACL, the
6245 messages generated by the underlying test were being passed through. This
6246 no longer happens. The underlying message is available in $acl_verify_
6247 message for both "message" and "log_message" expansions, so it can be
6248 passed through if needed.
6250 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
6251 new expansion $bh_ to give the encoded byte string without charset
6252 translation. Translation happens only if iconv() is available; HAVE_ICONV
6253 indicates this at build time. HEADERS_CHARSET gives the charset to
6254 translate to; headers_charset can change it in the configuration, and
6255 "headers charset" can change it in an individual filter file.
6257 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
6258 that creates RFC 2047 encoded "words" labels them as that charset instead
6259 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
6260 expansion operator; (ii) when Exim creates a From: line for a local
6261 message; (iii) when a header line is rewritten to include a "phrase" part.
6263 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
6264 buggy, causing it to skip the first lines of messages whose message ID
6265 ended in 'D'. This would not have bitten before Exim release 4.14, because
6266 message IDs were unlikely to end in 'D' before then. The effect was to have
6267 incorrect size information for certain domains.
6269 45. #include "config.h" was missing at the start of the crypt16.c module. This
6270 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
6273 46. If there was a timeout during a "random" callout check, Exim treated it as
6274 a failure of the random address, and carried on sending RSET and the real
6275 address. If the delay was just some slowness somewhere, the response to the
6276 original RCPT would be taken as a response to RSET and so on, causing
6277 mayhem of various kinds.
6279 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
6280 when I implemented it. It didn't allow for the fact that some option values
6281 may legitimatetly be negative (e.g. size_addition), and it didn't even do
6282 the right test for positive values.
6284 48. Domain names in DNS records are case-independent. Exim always looks them up
6285 in lower case. Some resolvers return domain names in exactly the case they
6286 appear in the zone file, that is, they may contain uppercase letters. Not
6287 all resolvers do this - some return always lower case. Exim was treating a
6288 change of case by a resolver as a change of domain, similar to a widening
6289 of a domain abbreviation. This triggered its re-routing code and so it was
6290 trying to route what was effectively the same domain again. This normally
6291 caused routing to fail (because the router wouldn't handle the domain
6292 twice). Now Exim checks for this case specially, and just changes the
6293 casing of the domain that it ultimately uses when it transmits the message
6296 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
6299 50. If a filter generated a file delivery with a non-absolute name (possible if
6300 no home directory exists for the router), the forbid_file option was not
6303 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
6304 the existing equality matching.
6306 52. Exim was using ints instead of ino_t variables in some places where it was
6307 dealing with inode numbers.
6309 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
6310 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
6311 called TMPDIR, and if it finds it is different, it changes its value.
6313 54. The smtp_printf() function is now made available to local_scan() so
6314 additional output lines can be written before returning. There is also an
6315 smtp_fflush() function to enable the detection of a dropped connection.
6316 The variables smtp_input and smtp_batched_input are exported to
6319 55. Changed the default runtime configuration: the message "Unknown user"
6320 has been removed from the ACL, and instead placed on the localuser router,
6321 using the cannot_route_message feature. This means that any verification
6322 failures that generate their own messages won't get overridden. Similarly,
6323 the "Unrouteable address" message that was in the ACL for unverifiable
6324 relay addresses has also been removed.
6326 56. Added hosts_avoid_esmtp to the smtp transport.
6328 57. The exicyclog script was not checking for the esoteric option
6329 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
6330 will work only if exicyclog is run under the appropriate euid.
6332 58. Following a discussion on the list, the rules by which Exim recognises line
6333 endings on incoming messages have been changed. The -dropcr and drop_cr
6334 options are now no-ops, retained only for backwards compatibility. The
6335 following line terminators are recognized: LF CRLF CR. However, special
6336 processing applies to CR:
6338 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
6339 nor a local message in the state where . is a terminator.
6341 (ii) If a bare CR is encountered in a header line, an extra space is added
6342 after the line terminator so as not to end the header. The reasoning
6343 behind this is that bare CRs in header lines are most likely either
6344 to be mistakes, or people trying to play silly games.
6346 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
6347 was being incorrectly given as 18 bytes larger than it should have been.
6348 This is a VOB (very old bug).
6350 60. This may never have affected anything current, but just in case it has:
6351 When the local host is found other than at the start of a list of hosts,
6352 the local host, those with the same MX, and any that follow, are discarded.
6353 When the list in question was part of a longer list of hosts, the following
6354 hosts (not currently being processed) were also being discarded. This no
6355 longer happens. I'm not sure if this situation could ever has previously
6358 61. Added the "/MX" feature to lists of hosts in the manualroute and query
6361 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
6362 header. This is something that is recommended in a new Internet Draft, and
6363 is something that is documented as being done by Sendmail. There are two
6364 possible values. For messages generated by the autoreply transport, Exim
6367 Auto-Submitted: auto-replied
6369 whereas for all other generated messages (e.g. bounces) it adds
6371 Auto-Submitted: auto-generated
6373 63. The "personal" condition in filters now includes a test for the
6374 Auto-Submitted: header. If it contains the string "auto-" the message it
6375 not considered personal.
6377 64. Added rcpt_include_affixes as a generic transport option.
6379 65. Added queue_only_override (default true).
6381 66. Added the syslog_duplication option.
6383 67. If what should have been the first header line of a message consisted of
6384 a space followed by a colon, Exim was mis-interpreting it as a header line.
6385 It isn't of course - it is syntactically invalid and should therefore be
6386 treated as the start of the message body. The misbehaviour could have
6387 caused a number of strange effects, including loss of data in subsequent
6388 header lines, and spool format errors.
6390 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
6391 client host had authenticated. This control can now be exercised by an ACL
6392 for more flexibility.
6394 69. By default, callouts do not happen when testing with -bh. There is now a
6395 variant, -bhc, which does actually run the callout code, including
6396 consulting and updating the callout cache.
6398 70. Added support for saslauthd authentication, courtesy of Alexander
6401 71. If statvfs() failed on the spool or log directories while checking their
6402 size for availability, Exim confusingly gave the error "space shortage".
6403 Furthermore, in debugging mode it crashed with a floating point exception.
6404 These checks are done if check_{spool,log}_{space,inodes} are set, and when
6405 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
6406 serious problem, Exim now writes to the main and panic logs when this
6407 happens, with details of the failure. It then refuses to accept the
6408 incoming message, giving the message "spool directory problem" or "log
6409 directory problem" with a 421 code for SMTP messages.
6411 72. When Exim is about to re-exec itself, it ensures that the file descriptors
6412 0, 1, and 2 exist, because some OS complain for execs without them (see
6413 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
6414 descriptors. However, the code omitted to check that the open succeeded,
6415 causing mysterious errors if for some reason the permissions on /dev/null
6416 got screwed. Now Exim writes a message to the main and panic logs, and
6417 bombs out if it can't open /dev/null.
6419 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
6420 interact so that it is all more flexible. It is supposed to remain
6421 backwards compatible. Also added extra_local_interfaces.
6423 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
6424 to bomb out with an assertion failure - to the client this appears as a
6425 connection drop. This problem occurs in the part of the code that was taken
6426 from the Samba project. Fortunately, the assertion is in a very simple
6427 function, so I have fixed this by reproducing the function inline in the
6428 one place where it is called, and arranging for authentication to fail
6429 instead of killing the process with assert().
6431 75. The SPA client code was not working when the server requested OEM rather
6432 than Unicode encoding.
6434 76. Added code to make require_files with a specific uid setting more usable in
6435 the case where statting the file as root fails - usually a non-root-mounted
6436 NFS file system. When this happens and the failure is EACCES, Exim now
6437 forks a subprocess and does the per-uid checking as the relevant uid.
6439 77. Added process_log_path.
6441 78. If log_file_path was not explicitly set, a setting of check_log_space or
6442 check_log_inodes was ignored.
6444 79. If a space check for the spool or log partitions fails, the incident is now
6445 logged. Of course, in the latter case the data may get lost...
6447 80. Added the %p formatting code to string_format() so that it can be used to
6448 print addresses in debug_print(). Adjusted all the address printing in the
6449 debugging in store.c to use %p rather than %d.
6451 81. There was a concern that a line of code in smtp_in.c could overflow a
6452 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
6453 initially expressed, the concern was not well-founded, because trailing
6454 spaces are removed early. However, if the trailing spaces were followed by
6455 a NULL, they did not get removed, so the overflow was possible. Two fixes
6458 (a) I re-wrote the offending code in a cleaner fashion.
6459 (b) If an incoming SMTP command contains a NULL character, it is rejected
6462 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
6463 runs initgroups(), so that if the Exim user is in any additional groups,
6464 they will be used during message reception.
6470 The change log for 4.20 and earlier releases has been archived.
projects / exim.git / blob