1 Change log file for Exim from version 4.21
2 -------------------------------------------
7 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
8 and 3.4.4 - once the server is enabled to respond to an OCSP request
9 it does even when not requested, resulting in a stapling non-aware
10 client dropping the TLS connection.
12 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
13 support variable-length bit vectors. No functional change.
15 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
16 The I= interface field on outgoing lines is now after the H= remote
17 host field, same as incoming lines. There is a separate
18 outgoing_interface log selector which allows you to disable the
21 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
22 If not running log_selector +smtp_connection the mainlog would be held
23 open indefinitely after a "too many connections" event, including to a
24 deleted file after a log rotate. Leave the per net connection logging
25 leaving it open for efficiency as that will be quickly detected by the
26 check on the next write.
28 HS/01 Bug 1671: Fix post transport crash.
29 Processing the wait-<transport> messages could crash the delivery
30 process if the message IDs didn't exist for some reason. When
31 using 'split_spool_directory=yes' the construction of the spool
32 file name failed already, exposing the same netto behaviour.
34 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
35 mime_regex ACL conditions.
37 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
38 to DSN fail messages (bounces): remote IP, remote greeting, remote response
39 to HELO, local diagnostic string.
41 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
42 log line to debug. Even when configured with a tls authenticator many
43 client connections are expected to not authenticate in this way, so
44 an authenticate fail is not an error.
46 HS/02 Add the Exim version string to the process info. This way exiwhat
47 gives some more detail about the running daemon.
49 JH/06 Bug 1395: time-limit cacheing of DNS lookups, to the TTL value. This may
50 matter for fast-change records such as DNSBLs.
52 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
53 retry record, even if constant. There may be multiple transports with
54 different interface settings and the retry behaviour needs to be kept
57 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
59 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
61 JH/10 Bug 840: fix log_defer_output option of pipe transport
63 JH/11 Bug 830: use same host for all RCPTS of a message, even under
64 hosts_randomize. This matters a lot when combined with mua_wrapper.
66 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
67 ${quote_pgsql:<string>} operator.
69 JH/13 Bug 1708: avoid misaligned access in cached lookup.
71 JH/14 Change header file name for freeradius-client. Relevant if compiling
72 with Radius support; from the Gentoo tree and checked under Fedora.
74 JH/15 Bug 1712: Introduce $prdr_requested flag variable
76 JH/16 Bug 1714: Permit an empty string as expansion result for transport
77 option transport_filter, meaning no filtering.
79 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
81 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
82 defaults to "*" (all hosts). The variable is now available when not built
83 with TLS, default unset, mainly to enable keeping the testuite sane.
84 If a server certificate is not supplied (via tls_certificate) an error is
85 logged, and clients will find TLS connections fail on startup. Presumably
86 they will retry in-clear.
87 Packagers of Exim are strongly encouraged to create a server certificate
90 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
91 with the $config_file variable.
93 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
94 in transport context, after the attempt, and per-recipient. The latter type
95 is per host attempted. The event data is the error message, and the errno
96 information encodes the lookup type (A vs. MX) used for the (first) host,
97 and the trailing two digits of the smtp 4xx reponse.
99 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
100 to write to mainlog (or rejectlog, paniclog) in the window between file
101 creation and permissions/ownership being changed. Particularly affects
102 installations where exicyclog is run as root, rather than exim user;
103 result is that the running daemon panics and dies.
105 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
107 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
108 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
109 "pri" and "weight". Note that the previous implicit priority given by the
110 list order is no longer honoured.
112 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalisation
115 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
116 by defining SUPPORT_SOCKS.
118 JH/26 Move PROXY support from Experimental to mainline, enabled for a build
119 by defining SUPPORT_PROXY. Note that the proxy_required_hosts option
120 is renamed to hosts_proxy, and the proxy_{host,target}_{address,port}.
121 variables are renamed to proxy_{local,external}_{address,port}.
123 JH/27 Move Internationalisation support from Experimental to mainline, enabled
124 for a build by defining SUPPORT_I18N
126 JH/28 Bug 1745: Fix redis lookups to handle (quoted) spaces embedded in parts
127 of the query string, and make ${quote_redis:} do that quoting.
129 JH/29 Move Events support from Experimental to mainline, enabled by default
130 and removable for a build by defining DISABLE_EVENT.
132 JH/30 Updated DANE implementation code to current from Viktor Dukhovni.
134 JH/31 Fix bug with hosts_connection_nolog and named-lists which were wrongly
135 cached by the daemon.
137 JH/32 Move Redis support from Experimental to mainline, enabled for a build
138 by defining LOOKUP_REDIS. The libhiredis library is required.
140 JH/33 Bug 1748: Permit ACL dnslists= condition in non-smtp ACLs if explicit
141 keys are given for lookup.
143 JH/34 Bug 1192: replace the embedded copy of PolarSSL RSA routines in the DKIM
144 support, by using OpenSSL or GnuTLS library ones. This means DKIM is
145 only supported when built with TLS support. The PolarSSL SHA routines
146 are still used when the TLS library is too old for convenient support.
148 JH/35 Require SINGLE_DH_USE by default in OpenSSL (main config option
149 openssl_options), for security. OpenSSL forces this from version 1.1.0
150 server-side so match that on older versions.
152 JH/36 Bug 1778: longstanding bug in memory use by the ${run } expansion: A fresh
153 allocation for $value could be released as the expansion processing
154 concluded, but leaving the global pointer active for it.
156 JH/37 Bug 1769: Permit a VRFY ACL to override the default 252 response,
157 and to use the domains and local_parts ACL conditions.
159 JH/38 Fix cutthrough bug with body lines having a single dot. The dot was
160 incorrectly not doubled on cutthrough transmission, hence seen as a
161 body-termination at the receiving system - resulting in truncated mails.
162 Commonly the sender saw a TCP-level error, and retransmitted the message
163 via the normal store-and-forward channel. This could result in duplicates
164 received - but deduplicating mailstores were liable to retain only the
165 initial truncated version.
167 JH/39 Bug 1781: Fix use of DKIM private-keys having trailing '=' in the base-64.
169 JH/40 Fix crash in queryprogram router when compiled with EXPERIMENTAL_SRS.
171 JH/41 Bug 1792: Fix selection of headers to sign for DKIM: bottom-up. While
172 we're in there, support oversigning also; bug 1309.
174 JH/42 Bug 1796: Fix error logged on a malware scanner connection failure.
176 HS/04 Add support for keep_environment and add_environment options.
178 JH/43 Tidy coding issues detected by gcc --fsanitize=undefined. Some remain;
179 either intentional arithmetic overflow during PRNG, or testing config-
182 JH/44 Bug 1800: The combination of a -bhc commandline option and cutthrough
183 delivery resulted in actual delivery. Cancel cutthrough before DATA
186 JH/45 Fix cutthrough, when connection not opened by verify and target hard-
187 rejects a recipient: pass the reject to the originator.
189 JH/46 Multiple issues raised by Coverity. Some were obvious or plausible bugs.
190 Many were false-positives and ignorable, but it's worth fixing the
197 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
200 JH/02 The smtp transport option "multi_domain" is now expanded.
202 JH/03 The smtp transport now requests PRDR by default, if the server offers
205 JH/04 Certificate name checking on server certificates, when exim is a client,
206 is now done by default. The transport option tls_verify_cert_hostnames
207 can be used to disable this per-host. The build option
208 EXPERIMENTAL_CERTNAMES is withdrawn.
210 JH/05 The value of the tls_verify_certificates smtp transport and main options
211 default to the word "system" to access the system default CA bundle.
212 For GnuTLS, only version 3.0.20 or later.
214 JH/06 Verification of the server certificate for a TLS connection is now tried
215 (but not required) by default. The verification status is now logged by
216 default, for both outbound TLS and client-certificate supplying inbound
219 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
222 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
223 Status Notification (bounce) messages are now MIME format per RFC 3464.
224 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
225 under the control of the dsn_advertise_hosts option, and routers may
226 have a dsn_lasthop option.
228 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
229 default, modifiable by a malware= option. The list separator for
230 the options can now be changed in the usual way. Bug 68.
232 JH/10 The smtp_receive_timeout main option is now expanded before use.
234 JH/11 The incoming_interface log option now also enables logging of the
235 local interface on delivery outgoing connections.
237 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
238 if the interface and destination host and port all match.
240 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
243 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
244 Patch from Andrew Lewis.
246 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
247 now supports optional time-restrictions, weighting, and priority
248 modifiers per server. Patch originally by <rommer@active.by>.
250 JH/16 The spamd_address main option now supports a mixed list of local
251 and remote servers. Remote servers can be IPv6 addresses, and
252 specify a port-range.
254 JH/17 Bug 68: The spamd_address main option now supports an optional
255 timeout value per server.
257 JH/18 Bug 1581: Router and transport options headers_add/remove can
258 now have the list separator specified.
260 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
263 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
266 JH/21 Support for the A6 type of dns record is withdrawn.
268 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
269 rather than the verbs used.
271 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
272 from 255 to 1024 chars.
274 JH/24 Verification callouts now attempt to use TLS by default.
276 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
277 are generic router options now. The defaults didn't change.
279 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
280 Original patch from Alexander Shikoff, worked over by JH.
282 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
283 files if ZCAT_COMMAND is not executable.
285 JH/26 Bug 1539: Add timout/retry options on dnsdb lookups.
287 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
289 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
290 Normally benign, it bites when the pair was led to by a CNAME;
291 modern usage is to not canoicalize the domain to a CNAME target
292 (and we were inconsistent anyway for A-only vs AAAA+A).
294 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
296 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
297 when evaluating $sender_host_dnssec.
299 JH/31 Check the HELO verification lookup for DNSSEC, adding new
300 $sender_helo_dnssec variable.
302 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
304 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
306 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
308 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
309 documented as working, but never had. Support all but $spam_report.
311 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
312 added for tls authenticator.
317 TL/01 When running the test suite, the README says that variables such as
318 no_msglog_check are global and can be placed anywhere in a specific
319 test's script, however it was observed that placement needed to be near
320 the beginning for it to behave that way. Changed the runtest perl
321 script to read through the entire script once to detect and set these
322 variables, reset to the beginning of the script, and then run through
323 the script parsing/test process like normal.
325 TL/02 The BSD's have an arc4random API. One of the functions to induce
326 adding randomness was arc4random_stir(), but it has been removed in
327 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
328 function when detected.
330 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
331 cause callback expansion.
333 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
334 syntax errors in an expansion can be treated as a string instead of
335 logging or causing an error, due to the internal use of bool_lax
336 instead of bool when processing it.
338 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
339 server certificates when making smtp deliveries.
341 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
343 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
345 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
347 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
348 Merged patch from Sebastian Wiedenroth.
350 JH/05 Fix results-pipe from transport process. Several recipients, combined
351 with certificate use, exposed issues where response data items split
352 over buffer boundaries were not parsed properly. This eventually
353 resulted in duplicates being sent. This issue only became common enough
354 to notice due to the introduction of conection certificate information,
355 the item size being so much larger. Found and fixed by Wolfgang Breyha.
357 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
358 size buffer was used, resulting in syntax errors when an expansion
361 JH/07 Add support for directories of certificates when compiled with a GnuTLS
362 version 3.3.6 or later.
364 JH/08 Rename the TPDA expermimental facility to Event Actions. The #ifdef
365 is EXPERIMENTAL_EVENT, the main-configuration and transport options
366 both become "event_action", the variables become $event_name, $event_data
367 and $event_defer_errno. There is a new variable $verify_mode, usable in
368 routers, transports and related events. The tls:cert event is now also
369 raised for inbound connections, if the main configuration event_action
372 TL/06 In test suite, disable OCSP for old versions of openssl which contained
373 early OCSP support, but no stapling (appears to be less than 1.0.0).
375 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
376 server certificate names available under the smtp transport option
377 "tls_verify_cert_hostname" now do not permit multi-component wildcard
380 JH/10 Time-related extraction expansions from certificates now use the main
381 option "timezone" setting for output formatting, and are consistent
382 between OpenSSL and GnuTLS compilations. Bug 1541.
384 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
385 encoded parameter in the incoming message. Bug 1558.
387 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
388 include certificate info, eximon was claiming there were spoolfile
391 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
393 JH/14 Log delivery-related information more consistently, using the sequence
394 "H=<name> [<ip>]" wherever possible.
396 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
397 are problematic for Debian distribution, omit them from the release
400 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
402 JH/16 Fix string representation of time values on 64bit time_t anchitectures.
405 JH/17 Fix a null-indirection in certextract expansions when a nondefault
406 output list separator was used.
411 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
412 checkers that were complaining about end of non-void function with no
415 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
416 This was a regression intruduced in 4.83 by another bugfix.
418 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
420 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
421 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
427 TF/01 Correctly close the server side of TLS when forking for delivery.
429 When a message was received over SMTP with TLS, Exim failed to clear up
430 the incoming connection properly after forking off the child process to
431 deliver the message. In some situations the subsequent outgoing
432 delivery connection happened to have the same fd number as the incoming
433 connection previously had. Exim would try to use TLS and fail, logging
434 a "Bad file descriptor" error.
436 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
437 utilities have not been installed.
439 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
440 temporary space as the ACL may create new global variables.
442 TL/01 LDAP support uses per connection or global context settings, depending
443 upon the detected version of the libraries at build time.
445 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
446 to extract and use the src ip:port in logging and expansions as if it
447 were a direct connection from the outside internet. PPv2 support was
448 updated based on HAProxy spec change in May 2014.
450 JH/02 Add ${listextract {number}{list}{success}{fail}}.
452 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
453 Properly escape header and check for NULL return.
455 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
458 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
460 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
461 characters in header names, implemented as a verify condition.
462 Contributed by Michael Fischer v. Mollard.
464 TL/05 Rename SPF condition results err_perm and err_temp to standardized
465 results permerror and temperror. Previous values are deprecated but
466 still accepted. In a future release, err_perm and err_temp will be
467 completely removed, which will be a backward incompatibility if the
468 ACL tests for either of these two old results. Patch contributed by
469 user bes-internal on the mailing list.
471 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
473 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
474 selectors, in both main and reject logs.
476 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
479 JH/07 Add malware type "sock" for talking to simple daemon.
481 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
483 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
484 routers/transports under cutthrough routing.
486 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
487 numbers. Touch up "bool" conditional to keep the same definition.
489 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
491 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
493 JH/12 Expand items in router/transport headers_add or headers_remove lists
494 individually rather than the list as a whole. Bug 1452.
496 Required for reasonable handling of multiple headers_ options when
497 they may be empty; requires that headers_remove items with embedded
498 colons must have them doubled (or the list-separator changed).
500 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
501 view the policy declared in the DMARC record. Currently, $dmarc_status
502 is a combined value of both the record presence and the result of the
505 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
507 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
508 dnslookup router and the smtp transport (applying to the forward
511 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
512 of ldap servers used for a specific lookup. Patch provided by Heiko
515 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
516 New variable $lookup_dnssec_authenticated for observability.
518 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
519 Patch submitted by Lars Timman.
521 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
523 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
524 Requires trusted mode and valid format message id, aborts otherwise.
525 Patch contributed by Heiko Schlichting.
527 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
528 certextract with support for various fields. Bug 1358.
530 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
531 is requested by default, modifiable by smtp transport option
534 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
535 operate on certificate variables to give certificate fingerprints
536 Also new ${sha256:cert_variable}.
538 JH/23 The PRDR feature is moved from being Experimental into the mainline.
540 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
541 Christian Aistleitner.
543 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
545 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
546 file. Patch from Wolfgang Breyha.
548 JH/25 Expand the coverage of the delivery $host and $host_address to
549 client authenticators run in verify callout. Bug 1476.
551 JH/26 Port service names are now accepted for tls_on_connect_ports, to
552 align with daemon_smtp_ports. Bug 72.
554 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
555 support and error reporting did not work properly.
557 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
558 and is readable. Patch from Andrew Colin Kissa.
560 TL/14 Enhance documentation of ${run expansion and how it parses the
561 commandline after expansion, particularly in the case when an
562 unquoted variable expansion results in an empty value.
564 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
566 PP/02 Fix internal collision of T_APL on systems which support RFC3123
567 by renaming away from it. Addresses GH issue 15, reported by
570 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
572 TL/15 SECURITY: prevent double expansion in math comparison functions
573 (can expand unsanitized data). Not remotely exploitable.
580 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
582 PP/02 Make -n do something, by making it not do something.
583 When combined with -bP, the name of an option is not output.
585 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
588 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
589 $sender_host_name and config options to manage this, and basic check
592 PP/05 DSCP support for outbound connections and control modifier for inbound.
594 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
595 (Only plugin which currently uses this is kerberos4, which nobody should
596 be using, but we should make it available and other future plugins might
597 conceivably use it, even though it would break NAT; stuff *should* be
598 using channel bindings instead).
600 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
601 name; added for Sendmail compatibility; requires admin caller.
602 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
603 just ignore it); requires trusted caller.
604 Also parse but ignore: -Ac -Am -X<logfile>
607 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
609 TL/02 Add +smtp_confirmation as a default logging option.
611 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
612 Patch by Magnus Holmgren from 2007-02-20.
614 TL/04 Bugzilla 1281 - Spec typo.
615 Bugzilla 1283 - Spec typo.
616 Bugzilla 1290 - Spec grammar fixes.
618 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
620 TL/06 Add Experimental DMARC support using libopendmarc libraries.
622 TL/07 Fix an out of order global option causing a segfault. Reported to dev
623 mailing list by by Dmitry Isaikin.
625 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
627 JH/02 Support "G" suffix to numbers in ${if comparisons.
629 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
631 NM/01 Bugzilla 1197 - Spec typo
632 Bugzilla 1196 - Spec examples corrections
634 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
636 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
637 gnutls_enable_pkcs11, but renamed to more accurately indicate its
640 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
641 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
643 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
644 "acl {{name}{arg}...}", and optional args on acl condition
647 JH/05 Permit multiple router/transport headers_add/remove lines.
649 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
651 JH/07 Avoid using a waiting database for a single-message-only transport.
652 Performance patch from Paul Fisher. Bugzilla 1262.
654 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
657 JH/09 Add $headers_added variable, with content from use of ACL modifier
658 add_header (but not yet added to the message). Bugzilla 199.
660 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
661 Pulled from Bugzilla 817 by Wolfgang Breyha.
663 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
665 (nb: this is the same fix as in Exim 4.80.1)
667 JH/11 Add A= logging on delivery lines, and a client_set_id option on
670 JH/12 Add optional authenticated_sender logging to A= and a log_selector
673 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
675 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
676 advertise SMTP AUTH mechanism to us, instead of a generic
677 protocol violation error. Also, make Exim more robust to bad
678 data from the Dovecot auth socket.
680 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
682 When a queue runner is handling a message, Exim first routes the
683 recipient addresses, during which it prunes them based on the retry
684 hints database. After that it attempts to deliver the message to
685 any remaining recipients. It then updates the hints database using
688 So if a recipient address works intermittently, it can get repeatedly
689 deferred at routing time. The retry hints record remains fresh so the
690 address never reaches the final cutoff time.
692 This is a fairly common occurrence when a user is bumping up against
693 their storage quota. Exim had some logic in its local delivery code
694 to deal with this. However it did not apply to per-recipient defers
695 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
697 This change adds a proper retry rule check during routing so that the
698 final cutoff time is checked against the message's age. We only do
699 this check if there is an address retry record and there is not a
700 domain retry record; this implies that previous attempts to handle
701 the address had the retry_use_local_parts option turned on. We use
702 this as an approximation for the destination being like a local
703 delivery, as in LMTP.
705 I suspect this new check makes the old local delivery cutoff check
706 redundant, but I have not verified this so I left the code in place.
708 TF/02 Correct gecos expansion when From: is a prefix of the username.
710 Test 0254 submits a message to Exim with the header
714 When I ran the test suite under the user fanf2, Exim expanded
715 the header to contain my full name, whereas it should have added
716 a Resent-Sender: header. It erroneously treats any prefix of the
717 username as equal to the username.
719 This change corrects that bug.
721 GF/01 DCC debug and logging tidyup
722 Error conditions log to paniclog rather than rejectlog.
723 Debug lines prefixed by "DCC: " to remove any ambiguity.
725 TF/03 Avoid unnecessary rebuilds of lookup-related code.
727 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
728 Bug spotted by Jeremy Harris; was flawed since initial commit.
729 Would have resulted in OCSP responses post-SNI triggering an Exim
730 NULL dereference and crash.
732 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
734 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
735 Bug detection, analysis and fix by Samuel Thibault.
736 Bugzilla 1331, Debian bug #698092.
738 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
740 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
741 Server implementation by Todd Lyons, client by JH.
742 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
743 config variable "prdr_enable" controls whether the server
744 advertises the facility. If the client requests PRDR a new
745 acl_data_smtp_prdr ACL is called once for each recipient, after
746 the body content is received and before the acl_smtp_data ACL.
747 The client is controlled by bolth of: a hosts_try_prdr option
748 on the smtp transport, and the server advertisement.
749 Default client logging of deliveries and rejections involving
750 PRDR are flagged with the string "PRDR".
752 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
753 fclose(). Diagnosis by Todd Lyons.
755 PP/17 Update configure.default to handle IPv6 localhost better.
756 Patch by Alain Williams (plus minor tweaks).
759 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
760 This is now consistent with GnuTLS, and is now documented: the
761 previous undocumented portable approach to treating the option as
762 unset was to force an expansion failure. That still works, and
763 an empty string is now equivalent.
765 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
766 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
767 not performing validation itself.
769 PP/20 Added force_command boolean option to pipe transport.
770 Patch from Nick Koston, of cPanel Inc.
772 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
775 TF/04 Added udpsend ACL modifer and hexquote expansion operator
777 PP/21 Fix eximon continuous updating with timestamped log-files.
778 Broken in a format-string cleanup in 4.80, missed when I repaired the
779 other false fix of the same issue.
780 Report and fix from Heiko Schlichting.
783 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
784 Report from Prashanth Katuri.
786 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
787 It's SecureTransport, so affects any MacOS clients which use the
788 system-integrated TLS libraries, including email clients.
790 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
791 using a MIME ACL for non-SMTP local injection.
792 Report and assistance in diagnosis by Warren Baker.
794 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
796 JH/16 Fix comparisons for 64b. Bugzilla 1385.
798 TL/09 Add expansion variable $authenticated_fail_id to keep track of
799 last id that failed so it may be referenced in subsequent ACL's.
801 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
804 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
805 ldap library initialization, allowing self-signed CA's to be
806 used. Also properly sets require_cert option later in code by
807 using NULL (global ldap config) instead of ldap handle (per
808 session). Bug diagnosis and testing by alxgomz.
810 TL/12 Enhanced documentation in the ratelimit.pl script provided in
811 the src/util/ subdirectory.
813 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
814 renamed to Transport Post Delivery Action by Jeremy Harris, as
817 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
818 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
819 redis_servers = needs to be configured which will be used by the redis
820 lookup. Patch from Warren Baker, of The Packet Hub.
822 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
824 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
825 hostname or reverse DNS when processing a host list. Used suggestions
826 from multiple comments on this bug.
828 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
830 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
831 Missed a few lines, added it to make the runtest require no keyboard
834 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
835 contains upper case chars. Make router use caseful_local_part.
837 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
838 support when GnuTLS has been built with p11-kit.
844 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
846 This, or similar/improved, will also be change PP/11 of 4.82.
852 PP/01 Handle short writes when writing local log-files.
853 In practice, only affects FreeBSD (8 onwards).
854 Bugzilla 1053, with thanks to Dmitry Isaikin.
856 NM/01 Bugzilla 949 - Documentation tweak
858 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
861 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
863 PP/02 Implemented gsasl authenticator.
865 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
867 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
868 `pkg-config foo` for cflags/libs.
870 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
871 with rest of GSASL and with heimdal_gssapi.
873 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
874 `pkg-config foo` for cflags/libs for the TLS implementation.
876 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
877 properties get this fed in as external SSF. A number of robustness
878 and debugging improvements to the cyrus_sasl authenticator.
880 PP/08 cyrus_sasl server now expands the server_realm option.
882 PP/09 Bugzilla 1214 - Log authentication information in reject log.
883 Patch by Jeremy Harris.
885 PP/10 Added dbmjz lookup type.
887 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
889 PP/12 MAIL args handles TAB as well as SP, for better interop with
890 non-compliant senders.
891 Analysis and variant patch by Todd Lyons.
893 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
894 Bug report from Lars Müller <lars@samba.org> (via SUSE),
895 Patch from Dirk Mueller <dmueller@suse.com>
897 PP/13 tls_peerdn now print-escaped for spool files.
898 Observed some $tls_peerdn in wild which contained \n, which resulted
899 in spool file corruption.
901 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
902 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
903 or write after TLS renegotiation, which otherwise led to messages
906 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
907 as a tracking header (ie: a signed header comes before the signature).
908 Patch from Wolfgang Breyha.
910 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
911 comma-sep list; embedded commas doubled.
913 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
915 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
917 Report and patch from Dmitry Banschikov.
919 PP/16 Removed "dont_insert_empty_fragments" fron "openssl_options".
920 Removed SSL_clear() after SSL_new() which led to protocol negotiation
921 failures. We appear to now support TLS1.1+ with Exim.
923 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
924 lets Exim select keys and certificates based upon TLS SNI from client.
925 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
926 before an outbound SMTP session. New log_selector, +tls_sni.
928 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
929 NULL dereference. Report and patch from Alun Jones.
931 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
932 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
933 Not seeing resolver debug output on NetBSD, but suspect this is a
934 resolver implementation change.
936 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
937 Left warnings. Added "eximon gdb" invocation mode.
939 PP/21 Defaulting "accept_8bitmime" to true, not false.
941 PP/22 Added -bw for inetd wait mode support.
943 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
944 locate the relevant includes and libraries. Made this the default.
946 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
947 Bugzilla 1246, report and most of solution from Tomasz Kusy.
949 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
950 This may cause build issues on older platforms.
952 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
953 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
954 gnutls_require_mac & gnutls_require_protocols (no longer supported).
955 Added SNI support via GnuTLS too.
956 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
958 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
960 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
961 Applied second patch from Janne, implementing suggestion to default
962 multiple-strings-in-record handling to match SPF spec.
964 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
966 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
967 read-only, out of scope).
968 Patch from Wolfgang Breyha, report from Stuart Northfield.
970 PP/29 Fix three issues highlighted by clang analyser static analysis.
971 Only crash-plausible issue would require the Cambridge-specific
972 iplookup router and a misconfiguration.
973 Report from Marcin Mirosław.
975 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
977 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
978 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
979 As part of this, removing so much warning spew let me fix some minor
980 real issues in debug logging.
982 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
983 assignment on my part. Fixed.
985 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
986 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
987 Janne Snabb (who went above and beyond: thank you).
989 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
990 string otherwise requires a connection and a bunch more work and it's
991 relatively easy to get wrong. Should also expose TLS library linkage
994 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
995 64-bit ${eval} (JH/03).
997 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
998 GNU libc to support some of the 64-bit stuff, should not lead to
999 conflicts. Defined before os.h is pulled in, so if a given platform
1000 needs to override this, it can.
1002 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
1003 protection layer was required, which is not implemented.
1004 Bugzilla 1254, patch from Wolfgang Breyha.
1006 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
1007 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
1008 tls_dhparam take prime identifiers. Also unbreak combination of
1009 OpenSSL+DH_params+TLSSNI.
1011 PP/39 Disable SSLv2 by default in OpenSSL support.
1017 PP/01 Solaris build fix for Oracle's LDAP libraries.
1018 Bugzilla 1109, patch from Stephen Usher.
1020 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
1022 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
1025 TF/02 Fix a couple more cases where we did not log the error message
1026 when unlink() failed. See also change 4.74-TF/03.
1028 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
1029 lock up or crash if it happened to be inside a call to libc when it
1030 got a SIGUSR1 from exiwhat.
1032 The SIGUSR1 handler appends the current process status to the process
1033 log which is later printed by exiwhat. It used to use the general
1034 purpose logging code to do this, but several functions it calls are
1035 not safe for signals.
1037 The new output code in the SIGUSR1 handler is specific to the process
1038 log, and simple enough that it's easy to inspect for signal safety.
1039 Removing some special cases also simplifies the general logging code.
1040 Removing the spurious timestamps from the process log simplifies
1043 TF/04 Improved ratelimit ACL condition.
1045 The /noupdate option has been deprecated in favour of /readonly which
1046 has clearer semantics. The /leaky, /strict, and /readonly update modes
1047 are mutually exclusive. The update mode is no longer included in the
1048 database key; it just determines when the database is updated. (This
1049 means that when you upgrde Exim will forget old rate measurements.)
1051 Exim now checks that the per_* options are used with an update mode that
1052 makes sense for the current ACL. For example, when Exim is processing a
1053 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
1054 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
1055 must specify per_mail/readonly. If you omit the update mode it defaults to
1056 /leaky where that makes sense (as before) or /readonly where required.
1058 The /noupdate option is now undocumented but still supported for
1059 backwards compatibility. It is equivalent to /readonly except that in
1060 ACLs where /readonly is required you may specify /leaky/noupdate or
1061 /strict/noupdate which are treated the same as /readonly.
1063 A useful new feature is the /count= option. This is a generalization
1064 of the per_byte option, so that you can measure the throughput of other
1065 aggregate values. For example, the per_byte option is now equivalent
1066 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
1068 The per_rcpt option has been generalized using the /count= mechanism
1069 (though it's more complicated than the per_byte equivalence). When it is
1070 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
1071 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
1072 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
1073 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
1074 Note that using per_rcpt with a non-readonly update mode in more than
1075 one ACL will cause the recipients to be double-counted. (The per_mail
1076 and per_byte options don't have this problem.)
1078 The handling of very low rates has changed slightly. If the computed rate
1079 is less than the event's count (usually one) then this event is the first
1080 after a long gap. In this case the rate is set to the same as this event's
1081 count, so that the first message of a spam run is counted properly.
1083 The major new feature is a mechanism for counting the rate of unique
1084 events. The new per_addr option counts the number of different
1085 recipients that someone has sent messages to in the last time period. It
1086 behaves like per_rcpt if all the recipient addresses are different, but
1087 duplicate recipient addresses do not increase the measured rate. Like
1088 the /count= option this is a general mechanism, so the per_addr option
1089 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
1090 example, measure the rate that a client uses different sender addresses
1091 with the options per_mail/unique=$sender_address. There are further
1092 details in the main documentation.
1094 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
1096 TF/06 Removed a few PCRE remnants.
1098 TF/07 Automatically extract Exim's version number from tags in the git
1099 repository when doing development or release builds.
1101 PP/02 Raise smtp_cmd_buffer_size to 16kB.
1102 Bugzilla 879. Patch from Paul Fisher.
1104 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
1105 Heavily based on revision 40f9a89a from Simon Arlott's tree.
1108 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
1110 PP/05 Variable $av_failed, true if the AV scanner deferred.
1111 Bugzilla 1078. Patch from John Horne.
1113 PP/06 Stop make process more reliably on build failure.
1114 Bugzilla 1087. Patch from Heiko Schlittermann.
1116 PP/07 Make maildir_use_size_file an _expandable_ boolean.
1117 Bugzilla 1089. Patch from Heiko Schlittermann.
1119 PP/08 Handle ${run} returning more data than OS pipe buffer size.
1120 Bugzilla 1131. Patch from Holger Weiß.
1122 PP/09 Handle IPv6 addresses with SPF.
1123 Bugzilla 860. Patch from Wolfgang Breyha.
1125 PP/10 GnuTLS: support TLS 1.2 & 1.1.
1127 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
1130 PP/11 match_* no longer expand right-hand-side by default.
1131 New compile-time build option, EXPAND_LISTMATCH_RHS.
1132 New expansion conditions, "inlist", "inlisti".
1134 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
1136 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
1138 PP/14 fix log_write() format string regression from TF/03.
1139 Bugzilla 1152. Patch from Dmitry Isaikin.
1145 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
1147 PP/02 Harmonised TLS library version reporting; only show if debugging.
1148 Layout now matches that introduced for other libraries in 4.74 PP/03.
1150 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
1152 PP/04 New "dns_use_edns0" global option.
1154 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
1157 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
1158 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
1160 TK/01 Updated PolarSSL code to 0.14.2.
1161 Bugzilla 1097. Patch from Andreas Metzler.
1163 PP/07 Catch divide-by-zero in ${eval:...}.
1164 Fixes bugzilla 1102.
1166 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
1169 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
1170 format-string attack -- SECURITY: remote arbitrary code execution.
1172 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
1173 time unintentionally subject to list matching rules, letting the header
1174 cause arbitrary Exim lookups (of items which can occur in lists, *not*
1175 arbitrary string expansion). This allowed for information disclosure.
1177 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
1178 INT_MIN/-1 -- value coerced to INT_MAX.
1184 NM/01 Workround for PCRE version dependency in version reporting
1187 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
1188 This fixes portability to compilers other than gcc, notably
1189 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
1191 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
1192 makefiles for portability to HP-UX and POSIX correctness.
1194 PP/01 Permit LOOKUP_foo enabling on the make command-line.
1195 Also via indented variable definition in the Makefile.
1196 (Debugging by Oliver Heesakkers).
1198 PP/02 Restore caching of spamd results with expanded spamd_address.
1199 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
1201 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
1202 Improves build reliability. Fix from: Frank Elsner
1204 NM/02 Fix wide character breakage in the rfc2047 coding
1205 Fixes bug 1064. Patch from Andrey N. Oktyabrski
1207 NM/03 Allow underscore in dnslist lookups
1208 Fixes bug 1026. Patch from Graeme Fowler
1210 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
1211 Code patches from Adam Ciarcinski of NetBSD.
1213 NM/04 Fixed exiqgrep to cope with mailq missing size issue
1216 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
1217 is logged, to avoid truncation. Patch from John Horne.
1219 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
1220 Patch from Jakob Hirsch.
1222 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
1223 SQL string expansion failure details.
1224 Patch from Andrey Oktyabrski.
1226 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
1227 Patch from Simon Arlott.
1229 PP/09 New lookups functionality failed to compile on old gcc which rejects
1230 extern declarations in function scope.
1231 Patch from Oliver Fleischmann
1233 PP/10 Use sig_atomic_t for flags set from signal handlers.
1234 Check getgroups() return and improve debugging.
1235 Fixed developed for diagnosis in bug 927 (which turned out to be
1238 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
1239 Patch from Mark Zealey.
1241 PP/12 Bugzilla 1056: Improved spamd server selection.
1242 Patch from Mark Zealey.
1244 PP/13 Bugzilla 1086: Deal with maildir quota file races.
1245 Based on patch from Heiko Schlittermann.
1247 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
1248 Patch from Uwe Doering, sign-off by Michael Haardt.
1250 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
1251 variable declaration deep within a block. Bug and patch from
1254 PP/15 lookups-Makefile IRIX compatibilty coercion.
1256 PP/16 Make DISABLE_DKIM build knob functional.
1258 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
1259 Patch by Simon Arlott
1261 TF/03 Fix valgrind.h portability to C89 compilers that do not support
1262 variable argument macros. Our copy now differs from upstream.
1268 TF/01 Failure to get a lock on a hints database can have serious
1269 consequences so log it to the panic log.
1271 TF/02 Log LMTP confirmation messages in the same way as SMTP,
1272 controlled using the smtp_confirmation log selector.
1274 TF/03 Include the error message when we fail to unlink a spool file.
1276 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
1277 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
1278 for maintaining out-of-tree patches for some time.
1280 PP/01 Bugzilla 139: Documentation and portability issues.
1281 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
1282 Handle per-OS dynamic-module compilation flags.
1284 PP/02 Let /dev/null have normal permissions.
1285 The 4.73 fixes were a little too stringent and complained about the
1286 permissions on /dev/null. Exempt it from some checks.
1287 Reported by Andreas M. Kirchwitz.
1289 PP/03 Report version information for many libraries, including
1290 Exim version information for dynamically loaded libraries. Created
1291 version.h, now support a version extension string for distributors
1292 who patch heavily. Dynamic module ABI change.
1294 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
1295 privilege escalation vulnerability whereby the Exim run-time user
1296 can cause root to append content of the attacker's choosing to
1299 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
1302 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
1303 If dropping privileges for untrusted macros, we disabled normal logging
1304 on the basis that it would fail; for the Exim run-time user, this is not
1305 the case, and it resulted in successful deliveries going unlogged.
1306 Fixed. Reported by Andreas Metzler.
1312 PP/01 Date: & Message-Id: revert to normally being appended to a message,
1313 only prepend for the Resent-* case. Fixes regression introduced in
1314 Exim 4.70 by NM/22 for Bugzilla 607.
1316 PP/02 Include check_rfc2047_length in configure.default because we're seeing
1317 increasing numbers of administrators be bitten by this.
1319 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
1321 PP/03 Bugzilla 994: added openssl_options main configuration option.
1323 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
1325 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
1327 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
1329 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
1330 without a peer certificate, leading to a segfault because of an
1331 assumption that peers always have certificates. Be a little more
1332 paranoid. Problem reported by Martin Tscholak.
1334 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
1335 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
1336 NB: ClamAV planning to remove STREAM in "middle of 2010".
1337 CL also introduces -bmalware, various -d+acl logging additions and
1338 more caution in buffer sizes.
1340 PP/09 Implemented reverse_ip expansion operator.
1342 PP/10 Bugzilla 937: provide a "debug" ACL control.
1344 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
1346 PP/12 Bugzilla 973: Implement --version.
1348 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
1350 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
1352 PP/15 Bugzilla 816: support multiple condition rules on Routers.
1354 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
1355 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
1356 ignore trailing whitespace.
1358 JJ/02 prevent non-panic DKIM error from being sent to paniclog
1360 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
1363 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
1364 Notification from Dr Andrew Aitchison.
1366 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
1367 ExtendedDetectionInfo response format.
1368 Notification from John Horne.
1370 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
1373 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
1374 XSL and documented dependency on system catalogs, with examples of how
1377 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
1380 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
1381 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
1382 configuration file which is writeable by the Exim user or group.
1384 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
1385 of configuration files to cover files specified with the -C option if
1386 they are going to be used with root privileges, not just the default
1389 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
1390 option (effectively making it always true).
1392 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
1393 files to be used while preserving root privileges.
1395 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
1396 that rogue child processes cannot use them.
1398 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
1399 run-time user, instead of root.
1401 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
1402 Exim run-time user without dropping privileges.
1404 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
1405 result string, instead of calling string_vformat() twice with the same
1408 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
1409 for other users. Others should always drop root privileges if they use
1410 -C on the command line, even for a whitelisted configure file.
1412 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
1414 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
1420 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
1421 $data_path, and $header_path variables; fixed documentation bugs and
1424 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
1425 exipick to access non-standard spools, including the "frozen" queue
1428 NM/01 Bugzilla 965: Support mysql stored procedures.
1429 Patch from Alain Williams
1431 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
1433 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
1434 Patch from Andreas Metzler
1436 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
1437 Patch from Kirill Miazine
1439 NM/05 Bugzilla 671: Added umask to procmail example.
1441 JJ/03 installed exipick 20100323.0, fixing doc bug
1443 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
1444 directory. Notification and patch from Dan Rosenberg.
1446 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
1448 TK/02 Improve log output when DKIM signing operation fails.
1450 MH/01 Treat the transport option dkim_domain as a colon separated
1451 list, not as a single string, and sign the message with each element,
1452 omitting multiple occurences of the same signer.
1454 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
1455 Bugzilla 985, 986. Patch by Simon Arlott
1457 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
1458 Patch by Simon Arlott
1460 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
1461 MBX locking. Notification from Dan Rosenberg.
1467 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
1469 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
1471 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
1473 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
1475 NM/04 Bugzilla 915: Flag broken perl installation during build.
1481 TK/01 Added patch by Johannes Berg that expands the main option
1482 "spamd_address" if it starts with a dollar sign.
1484 TK/02 Write list of recipients to X-Envelope-Sender header when building
1485 the mbox-format spool file for content scanning (suggested by Jakob
1488 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
1489 (http://www.dcc-servers.net/) support via dccifd. Activated by
1490 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
1492 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
1493 by Mark Daniel Reidel <mr@df.eu>.
1495 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
1496 When building exim an external PCRE library is now needed -
1497 PCRE is a system library on the majority of modern systems.
1498 See entry on PCRE_LIBS in EDITME file.
1500 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
1501 conversation. Added nologin parameter to request.
1502 Patch contributed by Kirill Miazine.
1504 TF/01 Do not log submission mode rewrites if they do not change the address.
1506 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
1508 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
1509 log files in place. Contributed by Roberto Lima.
1511 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
1513 TF/03 Bugzilla 615: When checking the local_parts router precondition
1514 after a local_part_suffix or local_part_prefix option, Exim now
1515 does not use the address's named list lookup cache, since this
1516 contains cached lookups for the whole local part.
1518 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
1519 Robert Millan. Documentation is in experimental-spec.txt.
1521 TF/04 Bugzilla 668: Fix parallel build (make -j).
1523 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
1525 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
1526 Patch provided by Jan Srzednicki.
1528 TF/05 Leading white space used to be stripped from $spam_report which
1529 wrecked the formatting. Now it is preserved.
1531 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
1532 that they are available at delivery time.
1534 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
1536 TF/08 TLS error reporting now respects the incoming_interface and
1537 incoming_port log selectors.
1539 TF/09 Produce a more useful error message if an SMTP transport's hosts
1540 setting expands to an empty string.
1542 NM/06 Bugzilla 744: EXPN did not work under TLS.
1543 Patch provided by Phil Pennock.
1545 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
1546 Patch provided by Richard Godbee.
1548 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
1549 acl_smtp_notquit, added index entry.
1551 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
1552 Patch provided by Eugene Bujak.
1554 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
1555 accept(). Patch provided by Maxim Dounin.
1557 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
1558 Patch provided by Phil Pennock.
1560 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
1562 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
1563 Patch provided by Brad "anomie" Jorsch.
1565 NM/14 Bugzilla 622: Added timeout setting to transport filter.
1566 Patch provided by Dean Brooks.
1568 TK/05 Add native DKIM support (does not depend on external libraries).
1570 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
1571 Patch provided by Graeme Fowler.
1573 NM/16 Bugzilla 851: Documentation example syntax fix.
1575 NM/17 Changed NOTICE file to remove references to embedded PCRE.
1577 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
1580 NM/19 Bugzilla 745: TLS version reporting.
1581 Patch provided by Phil Pennock.
1583 NM/20 Bugzilla 167: bool: condition support.
1584 Patch provided by Phil Pennock.
1586 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
1587 clients. Patch provided by Phil Pennock.
1589 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
1590 Patch provided by Brad "anomie" Jorsch.
1592 NM/23 Bugzilla 687: Fix misparses in eximstats.
1593 Patch provided by Heiko Schlittermann.
1595 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
1596 Patch provided by Heiko Schlittermann.
1598 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
1599 plus update to original patch.
1601 NM/26 Bugzilla 799: Documentation correction for ratelimit.
1603 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
1604 Patch provided by David Brownlee.
1606 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
1608 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
1610 NM/30 Bugzilla 888: TLS documentation bugfixes.
1612 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
1614 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
1615 Unlike the original bugzilla I have changed all shell scripts in src tree.
1617 NM/33 Bugzilla 898: Transport filter timeout fix.
1618 Patch by Todd Rinaldo.
1620 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
1621 Patch by Serge Demonchaux.
1623 NM/35 Bugzilla 39: Base64 decode bug fixes.
1624 Patch by Jakob Hirsch.
1626 NM/36 Bugzilla 909: Correct connect() call in dcc code.
1628 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
1630 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
1632 NM/39 Bugzilla 911: Fixed MakeLinks build script.
1638 TK/01 Add preliminary DKIM support. Currently requires a forked version of
1639 ALT-N's libdkim that I have put here:
1640 http://duncanthrax.net/exim-experimental/
1642 Note to Michael Haardt: I had to rename some vars in sieve.c. They
1643 were called 'true' and it seems that C99 defines that as a reserved
1644 keyword to be used with 'bool' variable types. That means you could
1645 not include C99-style headers which use bools without triggering
1646 build errors in sieve.c.
1648 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
1649 as mailq or other aliases. Changed the --help handling significantly
1650 to do whats expected. exim_usage() emits usage/help information.
1652 SC/01 Added the -bylocaldomain option to eximstats.
1654 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
1656 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
1658 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
1664 PH/01 Another patch from the Sieve maintainer.
1666 PH/02 When an IPv6 address is converted to a string for single-key lookup
1667 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
1668 dots are used instead of colons so that keys in lsearch files need not
1669 contain colons. This was done some time before quoting was made available
1670 in lsearch files. However, iplsearch files do require colons in IPv6 keys
1671 (notated using the quote facility) so as to distinguish them from IPv4
1672 keys. This meant that lookups for IP addresses in host lists did not work
1673 for iplsearch lookups.
1675 This has been fixed by arranging for IPv6 addresses to be expressed with
1676 colons if the lookup type is iplsearch. This is not incompatible, because
1677 previously such lookups could never work.
1679 The situation is now rather anomolous, since one *can* have colons in
1680 ordinary lsearch keys. However, making the change in all cases is
1681 incompatible and would probably break a number of configurations.
1683 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
1686 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
1687 conversion specification without a maximum field width, thereby enabling
1688 a rogue spamd server to cause a buffer overflow. While nobody in their
1689 right mind would setup Exim to query an untrusted spamd server, an
1690 attacker that gains access to a server running spamd could potentially
1691 exploit this vulnerability to run arbitrary code as the Exim user.
1693 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
1694 $primary_hostname instead of what libspf2 thinks the hosts name is.
1696 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
1697 a directory entry by the name of the lookup key. Previously, if a
1698 symlink pointed to a non-existing file or a file in a directory that
1699 Exim lacked permissions to read, a lookup for a key matching that
1700 symlink would fail. Now it is enough that a matching directory entry
1701 exists, symlink or not. (Bugzilla 503.)
1703 PH/03 The body_linecount and body_zerocount variables are now exported in the
1706 PH/04 Added the $dnslist_matched variable.
1708 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
1709 This means they are set thereafter only if the connection becomes
1712 PH/06 Added the client_condition to authenticators so that some can be skipped
1713 by clients under certain conditions.
1715 PH/07 The error message for a badly-placed control=no_multiline_responses left
1716 "_responses" off the end of the name.
1718 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
1720 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
1721 (without spaces) instead of just copying the configuration text.
1723 PH/10 Added the /noupdate option to the ratelimit ACL condition.
1725 PH/11 Added $max_received_linelength.
1727 PH/12 Added +ignore_defer and +include_defer to host lists.
1729 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
1730 way in which PCRE > 7.0 is built.
1732 PH/14 Implemented queue_only_load_latch.
1734 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
1735 MAIL command. The effect was to mangle the value on 64-bit systems.
1737 PH/16 Another patch from the Sieve maintainer.
1739 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
1741 PH/18 If a system quota error occurred while trying to create the file for
1742 a maildir delivery, the message "Mailbox is full" was not appended to the
1743 bounce if the delivery eventually timed out. Change 4.67/27 below applied
1744 only to a quota excession during the actual writing of the file.
1746 PH/19 It seems that peer DN values may contain newlines (and other non-printing
1747 characters?) which causes problems in log lines. The DN values are now
1748 passed through string_printing() before being added to log lines.
1750 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
1751 and InterBase are left for another time.)
1753 PH/21 Added message_body_newlines option.
1755 PH/22 Guard against possible overflow in moan_check_errorcopy().
1757 PH/23 POSIX allows open() to be a macro; guard against that.
1759 PH/24 If the recipient of an error message contained an @ in the local part
1760 (suitably quoted, of course), incorrect values were put in $domain and
1761 $local_part during the evaluation of errors_copy.
1767 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
1768 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
1771 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
1772 issue a MAIL command.
1774 PH/02 In an ACL statement such as
1776 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
1778 if a client was not listed at all, or was listed with a value other than
1779 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
1780 the condition was not true (as it should be), so access was not denied.
1781 The bug was that the ! inversion was incorrectly passed on to the second
1782 item. This has been fixed.
1784 PH/03 Added additional dnslists conditions == and =& which are different from
1785 = and & when the dns lookup returns more than one IP address.
1787 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
1788 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
1790 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
1791 FSYNC, which compiles an option called disable_fsync that allows for
1792 bypassing fsync(). The documentation is heavily laced with warnings.
1794 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
1796 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
1797 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
1798 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
1799 including adding "make clean"; (3) Added -fPIC when compiling the test
1800 dynamically loaded module, to get rid of a warning.
1802 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
1803 message fails, move_frozen_messages = true and ignore_bounce_errors_after
1804 = 0s. The bug is otherwise harmless.
1806 PH/07 There was a bug in the dovecot authenticator such that the value of
1807 $auth1 could be overwritten, and so not correctly preserved, after a
1808 successful authentication. This usually meant that the value preserved by
1809 the server_setid option was incorrect.
1811 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
1813 PH/09 Installed PCRE release 7.0.
1815 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
1816 run for batched SMTP input. It is now run at the start of every message
1817 in the batch. While fixing this I discovered that the process information
1818 (output by running exiwhat) was not always getting set for -bs and -bS
1819 input. This is fixed, and it now also says "batched" for BSMTP.
1821 PH/11 Added control=no_pipelining.
1823 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
1824 patch, slightly modified), and move the expansion of helo_data till after
1825 the connection is made in the smtp transport (so it can use these
1828 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
1830 PH/14 Added log_selector = +pid.
1832 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
1834 PH/16 Add ${if forany and ${if forall.
1836 PH/17 Added dsn_from option to vary the From: line in DSNs.
1838 PH/18 Flush SMTP output before performing a callout, unless control =
1839 no_callout_flush is set.
1841 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
1842 was true (the default) a successful delivery failed to delete the retry
1843 item, thus causing premature timeout of the address. The bug is now
1846 PH/20 Added hosts_avoid_pipelining to the smtp transport.
1848 PH/21 Long custom messages for fakedefer and fakereject are now split up
1849 into multiline reponses in the same way that messages for "deny" and
1850 other ACL rejections are.
1852 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
1853 with slight modification.
1855 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
1856 draft, changing the syntax and factoring some duplicate code".
1858 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
1859 for deliveries of the second and subsequent messages over the same SMTP
1862 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
1863 ${reduce, with only minor "tidies".
1865 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
1867 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
1868 expansion side effects.
1870 PH/27 When a message times out after an over-quota error from an Exim-imposed
1871 quota, the bounce message says "mailbox is full". This message was not
1872 being given when it was a system quota that was exceeded. It now should
1875 MH/03 Made $recipients available in local_scan(). local_scan() already has
1876 better access to the recipient list through recipients_list[], but
1877 $recipients can be useful in postmaster-provided expansion strings.
1879 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
1880 in the case of a MAIL command with additional options following the
1881 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
1882 were accidentally chopped off.
1884 PH/29 SMTP synchronization checks are implemented when a command is read -
1885 there is a check that no more input is waiting when there shouldn't be
1886 any. However, for some commands, a delay in an ACL can mean that it is
1887 some time before the response is written. In this time, more input might
1888 arrive, invalidly. So now there are extra checks after an ACL has run for
1889 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
1890 pipelining has not been advertised.
1892 PH/30 MH's patch to allow iscntrl() characters to be list separators.
1894 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
1895 returned in the SMTP response when smtp_return_error_details was false.
1896 This has been fixed.
1898 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
1899 instead of the C I/O that was originally supplied, because problems were
1900 reported on Solaris.
1902 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
1903 Exim which did not show up earlier: it was assuming that a call to
1904 SSL_CTX_set_info_callback() might give an error value. In fact, there is
1905 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
1906 was a macro that became an assignment, so it seemed to work. This has
1907 changed to a proper function call with a void return, hence the compile
1908 error. Exim's code has been fixed.
1910 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
1913 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
1915 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
1917 PH/37 If a message is not accepted after it has had an id assigned (e.g.
1918 because it turns out to be too big or there is a timeout) there is no
1919 "Completed" line in the log. When some messages of this type were
1920 selected by exigrep, they were listed as "not completed". Others were
1921 picked up by some special patterns. I have improved the selection
1922 criteria to be more general.
1924 PH/38 The host_find_failed option in the manualroute router can now be set
1925 to "ignore", to completely ignore a host whose IP address cannot be
1926 found. If all hosts are ignored, the behaviour is controlled by the new
1927 host_all_ignored option.
1929 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
1930 homing or because of multiple MX records with /mx) generated more than
1931 one IP address, and the following item turned out to be the local host,
1932 all the secondary addresses of the first item were incorrectly removed
1933 from the list, along with the local host and any following hosts (which
1934 is what is supposed to happen).
1936 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
1937 whoever called Exim into the -H file. In the case of the daemon it was
1938 behaving confusingly. When first started, it used values for whoever
1939 started the daemon, but after a SIGHUP it used the Exim user (because it
1940 calls itself on a restart). I have changed the code so that it now always
1943 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
1944 message are rejected with the same error (e.g. no authentication or bad
1945 sender address), and a DATA command is nevertheless sent (as can happen
1946 with PIPELINING or a stupid MUA), the error message that was given to the
1947 RCPT commands is included in the rejection of the DATA command. This is
1948 intended to be helpful for MUAs that show only the final error to their
1951 PH/42 Another patch from the Sieve maintainer.
1953 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
1954 Eximstats - Fixed some broken HTML links and added missing column headers
1956 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
1957 columns for Rejects, Temp Rejects, Ham, and Spam rows.
1959 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
1961 PH/43 Yet another patch from the Sieve maintainer.
1963 PH/44 I found a way to check for a TCP/IP connection going away before sending
1964 the response to the final '.' that terminates a message, but only in the
1965 case where the client has not sent further data following the '.'
1966 (unfortunately, this is allowed). However, in many cases there won't be
1967 any further data because there won't be any more messages to send. A call
1968 to select() can be used: if it shows that the input is "ready", there is
1969 either input waiting, or the socket has been closed. An attempt to read
1970 the next input character can distinguish the two cases. Previously, Exim
1971 would have sent an OK response which the client would never have see.
1972 This could lead to message repetition. This fix should cure that, at
1973 least in a lot of common cases.
1975 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
1976 advertised in response to EHLO.
1982 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
1983 fixed by 4.65/MH/01 (is this a record?) are fixed:
1985 (i) An empty string was always treated as zero by the numeric comparison
1986 operators. This behaviour has been restored.
1988 (ii) It is documented that the numeric comparison operators always treat
1989 their arguments as decimal numbers. This was broken in that numbers
1990 starting with 0 were being interpreted as octal.
1992 While fixing these problems I realized that there was another issue that
1993 hadn't been noticed. Values of message_size_limit (both the global option
1994 and the transport option) were treated as octal if they started with 0.
1995 The documentation was vague. These values are now always treated as
1996 decimal, and I will make that clear in the documentation.
2002 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
2003 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
2006 MH/01 Don't check that the operands of numeric comparison operators are
2007 integers when their expansion is in "skipping" mode (fixes bug
2008 introduced by 4.64-PH/07).
2010 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
2011 child addresses, Exim now panics and dies. Previously, because the count
2012 is held in a short int, deliveries were likely to be lost. As such a
2013 large number of recipients for a single message is ridiculous
2014 (performance will be very, very poor), I have chosen to impose a limit
2015 rather than extend the field.
2021 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
2022 leftover -K file (the existence of which was triggered by #402).
2023 While we were at it, introduced process PID as part of the -K
2024 filename. This should rule out race conditions when creating
2027 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
2028 processing considerably. Previous code took too long for large mails,
2029 triggering a timeout which in turn triggers #401.
2031 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
2032 in the DK code in transports.c. sendfile() is not really portable,
2033 hence the _LINUX specificness.
2035 TF/01 In the add_headers option to the mail command in an Exim filter,
2036 there was a bug that Exim would claim a syntax error in any
2037 header after the first one which had an odd number of characters
2040 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
2041 callout verification, Exim cached a "reject" for the entire domain. This
2042 is correct for most verifications, but it is not correct for a recipient
2043 verification with use_sender or use_postmaster set, because in that case
2044 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
2045 case of MAIL FROM:<> rejection from other early rejections (e.g.
2046 rejection of HELO). When verifying a recipient using a non-null MAIL
2047 address, the cache is ignored if it shows MAIL FROM:<> rejection.
2048 Whatever the result of the callout, the value of the domain cache is
2049 left unchanged (for any other kind of callout, getting as far as trying
2050 RCPT means that the domain itself is ok).
2052 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
2055 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
2056 manifest itself as EPIPE rather than ECONNECT. When tidying away a
2057 session, the daemon ignores ECONNECT errors and logs others; it now
2058 ignores EPIPE as well.
2060 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
2061 (quoted-printable decoding).
2063 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
2064 later the small subsequent patch to fix an introduced bug.
2066 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
2068 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
2070 PH/08 An error is now given if message_size_limit is specified negative.
2072 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
2073 to be given (somewhat) arbitrary names.
2075 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
2078 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
2079 miscellaneous code fixes
2081 PH/10 Added the log_reject_target ACL modifier to specify where to log
2084 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
2085 hostname. This is wrong, because it relates to the incoming message (and
2086 probably the interface on which it is arriving) and not to the outgoing
2087 callout (which could be using a different interface). This has been
2088 changed to use the value of the helo_data option from the smtp transport
2089 instead - this is what is used when a message is actually being sent. If
2090 there is no remote transport (possible with a router that sets up host
2091 addresses), $smtp_active_hostname is used.
2093 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
2094 tweaks were necessary in order to get it to work (see also 21 below):
2095 (a) The code assumed that strncpy() returns a negative number on buffer
2096 overflow, which isn't the case. Replaced with Exim's string_format()
2098 (b) There were several signed/unsigned issues. I just did the minimum
2099 hacking in of casts. There is scope for a larger refactoring.
2100 (c) The code used strcasecmp() which is not a standard C function.
2101 Replaced with Exim's strcmpic() function.
2102 (d) The code set only $1; it now sets $auth1 as well.
2103 (e) A simple test gave the error "authentication client didn't specify
2104 service in request". It would seem that Dovecot has changed its
2105 interface. Fortunately there's a specification; I followed it and
2106 changed what the client sends and it appears to be working now.
2108 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
2111 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
2112 address A is aliased to B and C, where B exists and C does not. Without
2113 -v the output is "A verified" because verification stops after a
2114 successful redirection if more than one address is generated. However,
2115 with -v the child addresses are also verified. Exim was outputting "A
2116 failed to verify" and then showing the successful verification for C,
2117 with its parentage. It now outputs "B failed to verify", showing B's
2118 parentage before showing the successful verification of C.
2120 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
2121 look up a TXT record in a specific list after matching in a combined
2124 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
2125 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
2126 they consult the DNS. I had assumed they would set it the way they
2127 wanted; and indeed my experiments on Linux seem to show that in some
2128 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
2129 To be on the safe side, however, I have now made the interface to
2130 host_find_byname() similar to host_find_bydns(), with an argument
2131 containing the DNS resolver options. The host_find_byname() function now
2132 sets these options at its start, just as host_find_bydns() does. The smtp
2133 transport options dns_qualify_single and dns_search_parents are passed to
2134 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
2135 of host_find_byname() use the default settings of RES_DEFNAMES
2136 (qualify_single) but not RES_DNSRCH (search_parents).
2138 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
2139 spool_read_header() do less string testing, by means of a preliminary
2140 switch on the second character of optional "-foo" lines. (This is
2141 overdue, caused by the large number of possibilities that now exist.
2142 Originally there were few.) While I was there, I also converted the
2143 str(n)cmp tests so they don't re-test the leading "-" and the first
2144 character, in the hope this might squeeze out yet more improvement.
2146 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
2147 flag allowing group syntax was set by the header_syntax check but not
2148 turned off, possible causing trouble later; (2) The flag was not being
2149 set at all for the header_verify test, causing "group"-style headers to
2150 be rejected. I have now set it in this case, and also caused header_
2151 verify to ignore an empty address taken from a group. While doing this, I
2152 came across some other cases where the code for allowing group syntax
2153 while scanning a header line wasn't quite right (mostly, not resetting
2154 the flag correctly in the right place). These bugs could have caused
2155 trouble for malformed header lines. I hope it is now all correct.
2157 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
2158 with the "reply" argument non-NULL. The code, however (which originally
2159 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
2160 but it didn't always do it. This confused somebody who was copying the
2161 code for some other use. I have removed all the tests.
2163 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
2164 feature that was used to support insecure browsers during the U.S. crypto
2165 embargo. It requires special client support, and Exim is probably the
2166 only MTA that supported it -- and would never use it because real RSA is
2167 always available. This code has been removed, because it had the bad
2168 effect of slowing Exim down by computing (never used) parameters for the
2169 RSA_EXPORT functionality.
2171 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
2172 authenticator to fail if there's a tab character in the incoming data
2173 (there should never be unless someone is messing about, as it's supposed
2174 to be base64-encoded). Also added, on Timo's advice, the "secured" option
2175 if the connection is using TLS or if the remote IP is the same as the
2176 local IP, and the "valid-client-cert option" if a client certificate has
2179 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
2180 authenticators. This can be used for authorization after authentication
2181 succeeds. (In the case of plaintext, it servers for both authentication
2184 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
2185 if any retry times were supplied.
2187 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
2188 connection, where there is no client IP address to check. In this
2189 situation, the verify now always succeeds.
2191 PH/25 Applied John Jetmore's -Mset patch.
2193 PH/26 Added -bem to be like -Mset, but loading a message from a file.
2195 PH/27 In a string expansion for a processed (not raw) header when multiple
2196 headers of the same name were present, leading whitespace was being
2197 removed from all of them, but trailing whitespace was being removed only
2198 from the last one. Now trailing whitespace is removed from each header
2199 before concatenation. Completely empty headers in a concatenation (as
2200 before) are ignored.
2202 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
2203 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
2205 PH/29 [Removed. This was a change that I later backed out, and forgot to
2206 correct the ChangeLog entry (that I had efficiently created) before
2207 committing the later change.]
2209 PH/30 Exim was sometimes attempting to deliver messages that had suffered
2210 address errors (4xx response to RCPT) over the same connection as other
2211 messages routed to the same hosts. Such deliveries are always "forced",
2212 so retry times are not inspected. This resulted in far too many retries
2213 for the affected addresses. The effect occurred only when there were more
2214 hosts than the hosts_max_try setting in the smtp transport when it had
2215 the 4xx errors. Those hosts that it had tried were not added to the list
2216 of hosts for which the message was waiting, so if all were tried, there
2217 was no problem. Two fixes have been applied:
2219 (i) If there are any address or message errors in an SMTP delivery, none
2220 of the hosts (tried or untried) are now added to the list of hosts
2221 for which the message is waiting, so the message should not be a
2222 candidate for sending over the same connection that was used for a
2223 successful delivery of some other message. This seems entirely
2224 reasonable: after all the message is NOT "waiting for some host".
2225 This is so "obvious" that I'm not sure why it wasn't done
2226 previously. Hope I haven't missed anything, but it can't do any
2227 harm, as the worst effect is to miss an optimization.
2229 (ii) If, despite (i), such a delivery is accidentally attempted, the
2230 routing retry time is respected, so at least it doesn't keep
2231 hammering the server.
2233 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
2234 in ${readsocket because some servers need this prod.
2236 PH/32 Added some extra debug output when updating a wait-xxx database.
2238 PH/33 The hint "could be header name not terminated by colon", which has been
2239 given for certain expansion errors for a long time, was not being given
2240 for the ${if def:h_colon_omitted{... case.
2242 PH/34 The spec says: "With one important exception, whenever a domain list is
2243 being scanned, $domain contains the subject domain." There was at least
2244 one case where this was not true.
2246 PH/35 The error "getsockname() failed: connection reset by peer" was being
2247 written to the panic log as well as the main log, but it isn't really
2248 panic-worthy as it just means the connection died rather early on. I have
2249 removed the panic log writing for the ECONNRESET error when getsockname()
2252 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
2253 runs only) independently of the message's sender address. This meant
2254 that, if the 4xx error was in fact related to the sender, a different
2255 message to the same recipient with a different sender could confuse
2256 things. In particualar, this can happen when sending to a greylisting
2257 server, but other circumstances could also provoke similar problems.
2258 I have changed the default so that the retry time for these errors is now
2259 based a combination of the sender and recipient addresses. This change
2260 can be overridden by setting address_retry_include_sender=false in the
2263 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
2264 remote server are returned as part of bounce messages. This was not
2265 happening for LMTP over a pipe (the lmtp transport), but now it is the
2266 same for both kinds of LMTP.
2268 PH/38 Despite being documented as not happening, Exim was rewriting addresses
2269 in header lines that were in fact CNAMEs. This is no longer the case.
2271 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
2272 and queue runs started by the daemon processed all messages. This has
2273 been fixed so that -R and -S can now usefully be given with -q<time>.
2275 PH/40 Import PCRE release 6.7 (fixes some bugs).
2277 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
2279 PH/42 Give an error if -q is specified more than once.
2281 PH/43 Renamed the variables $interface_address and $interface_port as
2282 $received_ip_address and $received_port, to make it clear that these
2283 values apply to message reception, and not to the outgoing interface when
2284 a message is delivered. (The old names remain recognized, of course.)
2286 PH/44 There was no timeout on the connect() call when using a Unix domain
2287 socket in the ${readsocket expansion. There now is.
2289 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
2290 be meaningful with "accept".
2292 SC/01 Eximstats V1.43
2293 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
2295 SC/02 Eximstats V1.44
2296 Use a glob alias rather than an array ref in the generated
2297 parser. This improves both readability and performance.
2299 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
2300 Collect SpamAssassin and rejection statistics.
2301 Don't display local sender or destination tables unless
2302 there is data to show.
2303 Added average volumes into the top table text output.
2305 SC/04 Eximstats V1.46
2306 Collect data on the number of addresses (recipients)
2307 as well as the number of messages.
2309 SC/05 Eximstats V1.47
2310 Added 'Message too big' to the list of mail rejection
2311 reasons (thanks to Marco Gaiarin).
2313 SC/06 Eximstats V1.48
2314 Mainlog lines which have GMT offsets and are too short to
2315 have a flag are now skipped.
2317 SC/07 Eximstats V1.49 (Alain Williams)
2318 Added the -emptyok flag.
2320 SC/08 Eximstats V1.50
2321 Fixes for obtaining the IP address from reject messages.
2323 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
2324 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
2325 whitesspace changes from 4.64-PH/27
2327 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
2330 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
2331 are found, allow negative numbers in numeric criteria)
2333 JJ/06 exipick.20061117.2, added new $message_body_missing variable
2335 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
2336 to match changes made in 4.64-PH/43
2338 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
2340 PH/47 Put in an explicit test for a DNS lookup of an address record where the
2341 "domain" is actually an IP address, and force a failure. This locks out
2342 those revolvers/nameservers that support "A-for-A" lookups, in
2343 contravention of the specifications.
2345 PH/48 When a host name was looked up from an IP address, and the subsequent
2346 forward lookup of the name timed out, the host name was left in
2347 $sender_host_name, contrary to the specification.
2349 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
2350 restricted to single-key lookups, Exim was not diagnosing an error if
2351 * or *@ was used with a query-style lookup.
2353 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
2355 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
2356 long ago, but noone interested enough thought of it. Let's just say that
2357 the "1.1" means that there are some new functions that weren't there at
2358 some point in the past.
2360 PH/51 Error processing for expansion failure of helo_data from an smtp
2361 transport during callout processing was broken.
2363 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
2364 tested/used via the -bh/-bhc/-bs options.
2366 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
2367 bug, fixed in subsequent PCRE releases).
2369 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
2370 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
2372 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
2378 SC/01 Use a glob alias rather than an array ref in eximstats generated
2379 parser. This improves both readability and performance.
2381 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
2382 Don't display local sender or destination tables in eximstats unless
2383 there is data to show.
2384 Added average volumes into the eximstats top table text output.
2386 SC/03 Collect data on the number of addresses (recipients) as well
2387 as the number of messages in eximstats.
2389 TF/01 Correct an error in the documentation for the redirect router. Exim
2390 does (usually) call initgroups() when daemonizing.
2392 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
2393 with consistent privilege compared to when running as a daemon.
2395 TF/03 Note in the spec that $authenticated_id is not set for local
2396 submissions from trusted users.
2398 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
2399 Thanks to Dean Brooks <dean@iglou.com> for the patch.
2401 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
2402 by adding some example configuration directives to the default
2403 configuration file. A little bit of work is required to uncomment the
2404 directives and define how usernames and passwords are checked, but
2405 there is now a framework to start from.
2407 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
2408 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
2409 without this. I don't know how relevant this is to other LDAP libraries.
2411 PH/02 Add the verb name to the "unknown ACL verb" error.
2413 PH/03 Magnus Holmgren's patch for filter_prepend_home.
2415 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
2417 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
2418 directory not expanded when it should be if an expanded home directory
2419 was set for the address (which is overridden by the transport).
2421 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
2424 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
2425 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
2426 because it is too late at that time, and has no effect.
2428 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
2429 security issue with \' (bugzilla #107). I could not use the
2430 PQescapeStringConn() function, because it needs a PGconn value as one of
2433 PH/08 When testing addresses using -bt, indicate those final addresses that
2434 are duplicates that would not cause an additional delivery. At least one
2435 person was confused, thinking that -bt output corresponded to deliveries.
2436 (Suppressing duplicates isn't a good idea as you lose the information
2437 about possibly different redirections that led to the duplicates.)
2439 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
2440 systems where poll() doesn't work, in particular OS X.
2442 PH/10 Added more information to debugging output for retry time not reached.
2444 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
2445 operations in malware.c.
2447 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
2450 PH/13 If write_rejectlog was set false when logging was sent to syslog with
2451 syslog_duplication set false, log lines that would normally be written
2452 both the the main log and to the reject log were not written to syslog at
2455 PH/14 In the default configuration, change the use of "message" in ACL warn
2456 statements to "add_header".
2458 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
2459 not followed by a command (e.g. "seen endif").
2461 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
2462 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
2465 PH/17 Added extra conditions to the default value of delay_warning_condition
2469 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
2470 { match{$h_precedence:}{(?i)bulk|list|junk} } \
2471 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
2474 The Auto-Submitted: and various List- headers are standardised, whereas I
2475 don't think Precedence: ever was.
2477 PH/18 Refactored debugging code in route_finduser() to show more information,
2478 in particular, the error code if getpwnam() issues one.
2480 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
2481 This is apparently needed in addition to the PH/07 change above to avoid
2482 any possible encoding problems.
2484 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
2485 but not after initializing Perl.
2487 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
2488 output them only if debugging. By default they are written stderr,
2489 apparently, which is not desirable.
2491 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
2494 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
2497 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
2499 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
2500 authenticated or an ident call has been made. Suppress the default
2501 values for $authenticated_id and $authenticated_sender (but permit -oMai
2502 and -oMas) when testing with -bh.
2504 PH/24 Re-jigged the order of the tests in the default configuration so that the
2505 tests for valid domains and recipients precede the DNS black list and CSA
2506 tests, on the grounds that those ones are more expensive.
2508 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
2509 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
2510 This bug exists in every version of Exim that I still have, right back to
2513 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
2514 However, an attempt to turn on case-sensitivity in a regex key by
2515 including (?-i) didn't work because the subject string was already
2516 lowercased, and the effects were non-intuitive. It turns out that a
2517 one-line patch can be used to allow (?-i) to work as expected.
2523 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
2524 other effects) broke the use of negated acl sub-conditions.
2526 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
2529 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
2530 "Deny" causes Exim to reject the incoming connection with a 554 error.
2531 Unfortunately, if there is a major crisis, such as a disk failure,
2532 tcp-wrappers gives "deny", whereas what one would like would be some
2533 kind of temporary error. A kludge has been added to help with this.
2534 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
2535 554 error is used if errno is still zero or contains ENOENT (which occurs
2536 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
2539 PH/03 Add -lutil to the default FreeBSD LIBS setting.
2541 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
2542 errors. Otherwise a message that provokes a temporary error (when other
2543 messages do not) can cause a whole host to time out.
2545 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
2546 addresses were routed directly to files or pipes from a redirect router.
2547 File deliveries just didn't batch; pipe deliveries might have suffered
2550 PH/06 A failure to get a lock for a hints database would erroneously always say
2551 "Failed to get write lock", even when it was really a read lock.
2553 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
2554 of 0600. This has been changed to use the value of the lockfile_mode
2555 option (which defaults to 0600).
2557 PH/08 Applied small patch from the Sieve maintainer.
2559 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
2560 folder from quota calculations, a direct delivery into this folder messed
2561 up the contents of the maildirsize file. This was because the regex was
2562 used only to exclude .Trash (or whatever) when the size of the mailbox
2563 was calculated. There was no check that a delivery was happening into an
2564 excluded directory. This bug has been fixed by ignoring all quota
2565 processing for deliveries into excluded directories.
2567 PH/10 Added the maildirfolder_create_regex option to appendfile.
2573 PH/01 The code for finding all the local interface addresses on a FreeBSD
2574 system running IPv6 was broken. This may well have applied to all BSD
2575 systems, as well as to others that have similar system calls. The broken
2576 code found IPv4 interfaces correctly, but gave incorrect values for the
2577 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
2578 that it would not match correctly against @[] and not recognize the IPv6
2581 PH/02 The ipliteral router was not recognizing addresses of the form user@
2582 [ipv6:....] because it didn't know about the "ipv6:" prefix.
2584 PH/03 Added disable_ipv6.
2586 PH/04 Changed $reply_address to use the raw form of the headers instead of the
2587 decoded form, because it is most often used to construct To: headers
2588 lines in autoreplies, and the decoded form may well be syntactically
2589 invalid. However, $reply_address has leading white space removed, and all
2590 newlines turned into spaces so that the autoreply transport does not
2593 PH/05 If group was specified without a user on a router, and no group or user
2594 was specified on a transport, the group from the router was ignored.
2596 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
2597 for visible compile-time settings that can be used to change these
2598 numbers, for those that want even more. Backwards compatibility with old
2599 spool files has been maintained. However, going back to a previous Exim
2600 release will lost any variables that are in spool files.
2602 PH/07 Two small changes when running in the test harness: increase delay when
2603 passing a TCP/IP connection to a new process, in case the original
2604 process has to generate a bounce, and remove special handling of
2605 127.0.0.2 (sic), which is no longer necessary.
2607 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
2608 be the same on different OS.
2610 PH/09 Moved a debug statement in filter processing to avoid a race problem when
2613 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
2614 whether --show-vars was specified or not
2616 JJ/02 exipick: Added support for new ACL variable spool format introduced
2619 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
2620 syntactically invalid From: or Reply-to: line, and a filter used this to
2621 generate an autoreply, and therefore failed to obtain an address for the
2622 autoreply, Exim could try to deliver to a non-existent relative file
2623 name, causing unrelated and misleading errors. What now happens is that
2624 it logs this as a hard delivery error, but does not attempt to create a
2627 PH/11 The exinext utility has a -C option for testing purposes, but although
2628 the given file was scanned by exinext itself; it wasn't being passed on
2629 when Exim was called.
2631 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
2632 an end-of-file indication when reading a command response.
2634 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
2635 compiled. In many other places in Exim, IPv6 addresses are always
2636 recognized, so I have changed this. It also means that IPv4 domain
2637 literals of the form [IPV4:n.n.n.n] are now always recognized.
2639 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
2640 used if the router is not running as root, for example, when verifying at
2641 ACL time, or when using -bh. The debugging output from this situation was
2642 non-existent - all you got was a failure to exec. I have made two
2645 (a) Failures to set uid/gid, the current directory, or a process leader
2646 in a subprocess such as that created by queryprogram now generate
2647 suitable debugging ouput when -d is set.
2649 (b) The queryprogram router detects when it is not running as root,
2650 outputs suitable debugging information if -d is set, and then runs
2651 the subprocess without attempting to change uid/gid.
2653 PH/15 Minor change to Makefile for building test_host (undocumented testing
2656 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
2657 additional section of a DNS packet that returns MX or SRV records.
2658 Instead, it always explicitly searches for A/AAAA records. This avoids
2659 major problems that occur when a DNS server includes only records of one
2660 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
2661 fixed another bug: if SRV records were looked up and the corresponding
2662 address records were *not* found in the additional section, the port
2663 values from the SRV records were lost.
2665 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
2666 using the correct key (the original address) when searching the retry
2667 rules in order to find which one to use for generating the retry hint.
2669 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
2670 adding the default one. Similarly, if it contains a Reply-To: header, the
2671 errors_reply_to option, if set, is not used.
2673 PH/19 When calculating a retry time, Exim used to measure the "time since
2674 failure" by looking at the "first failed" field in the retry record. Now
2675 it does not use this if it is later than than the arrival time of the
2676 message. Instead it uses the arrival time. This makes for better
2677 behaviour in cases where some deliveries succeed, thus re-setting the
2678 "first failed" field. An example is a quota failure for a huge message
2679 when small messages continue to be delivered. Without this change, the
2680 "time since failure" will always be short, possible causing more frequent
2681 delivery attempts for the huge message than are intended.
2682 [Note: This change was subsequently modified - see PH/04 for 4.62.]
2684 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
2685 $1, $2, $3) because the numerical variables can be reset during some
2686 expansion items (e.g. "match"), thereby losing the authentication data.
2688 PH/21 Make -bV show the size of off_t variables so that the test suite can
2689 decide whether to run tests for quotas > 2G.
2691 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
2692 mailbox_size, and mailbox_filecount in the appendfile transport. If a
2693 filecount value is greater than 2G or if a quota value is greater than 2G
2694 on a system where the size of off_t is not greater than 4, a panic error
2697 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
2698 never match. The debug and -bh output now contains an explicit error
2699 message indicating a malformed IPv4 address or mask.
2701 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
2702 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
2703 PH/23 above applies.
2705 PH/25 Do not write to syslog when running in the test harness. The only
2706 occasion when this arises is a failure to open the main or panic logs
2707 (for which there is an explicit test).
2709 PH/26 Added the /no_tell option to "control=freeze".
2711 PH/27 If a host name lookup failed very early in a connection, for example, if
2712 the IP address matched host_lookup and the reverse lookup yielded a name
2713 that did not have a forward lookup, an error message of the form "no IP
2714 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
2715 could be logged. Now it outputs the IP address instead of "NULL".
2717 PH/28 An enabling patch from MH: add new function child_open_exim2() which
2718 allows the sender and the authenticated sender to be set when
2719 submitting a message from within Exim. Since child_open_exim() is
2720 documented for local_scan(), the new function should be too.
2722 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
2723 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
2724 results in an empty string is now treated as unset.
2726 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
2728 PH/31 Added sender_verify_fail logging option.
2730 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
2731 needed by RFC 822 but not by RFC 2822 was commented out. I have now
2732 tidied the source and removed it altogether.
2734 PH/33 When a queue run was abandoned because the load average was too high, a
2735 log line was always written; now it is written only if the queue_run log
2736 selector is set. In addition, the log line for abandonment now contains
2737 information about the queue run such as the pid. This is always present
2738 in "start" and "stop" lines but was omitted from the "abandon" line.
2740 PH/34 Omit spaces between a header name and the colon in the error message that
2741 is given when verify = headers_syntax fails (if there are lots of them,
2742 the message gets confusing).
2744 PH/35 Change the default for dns_check_names_pattern to allow slashes within
2745 names, as there are now some PTR records that contain slashes. This check
2746 is only to protect against broken name servers that fall over on strange
2747 characters, so the fact that it applies to all lookups doesn't matter.
2749 PH/36 Now that the new test suite is complete, we can remove some of the
2750 special code in Exim that was needed for the old test suite. For example,
2751 sorting DNS records because real resolvers return them in an arbitrary
2752 order. The new test suite's fake resolver always returns records in the
2755 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
2756 bounces) except when queue_only is set, to avoid logging races between
2757 the different processes.
2759 PH/38 Panic-die if .include specifies a non-absolute path.
2761 PH/39 A tweak to the "H" retry rule from its user.
2763 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
2764 a label. They prevented compilation on older perls.
2766 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
2767 a warning to be raised on newish perls.
2769 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
2770 on queue. Changes to match documented behaviour of showing count of
2771 messages matching specified criteria.
2773 PH/40 Changed the default ident timeout from 30s to 5s.
2775 PH/41 Added support for the use of login_cap features, on those BSD systems
2776 that have them, for controlling the resources used by pipe deliveries.
2778 PH/42 The content-scanning code uses fopen() to create files in which to put
2779 message data. Previously it was not paying any attention to the mode of
2780 the files. Exim runs with umask(0) because the rest of the code creates
2781 files with open(), and sets the required mode explicitly. Thus, these
2782 files were ending up world-writeable. This was not a big issue, because,
2783 being within the spool directory, they were not world-accessible. I have
2784 created a function called modefopen, which takes an additional mode
2785 argument. It sets umask(777), creates the file, chmods it to the required
2786 mode, then resets the umask. All the relevant calls to fopen() in the
2787 content scanning code have been changed to use this function.
2789 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
2790 to 24 hours. This avoids potential overflow problems when processing G
2791 and H retry rules. I suspect nobody ever tinkers with this value.
2793 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
2795 PH/45 When the plaintext authenticator is running as a client, the server's
2796 challenges are checked to ensure they are valid base64 strings. By
2797 default, the authentication attempt is cancelled if an invalid string is
2798 received. Setting client_ignore_invalid_base64 true ignores these errors.
2799 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
2800 they are received. Thus, the responses can be made to depend on the
2801 challenges. If an invalid string is ignored, an empty string is placed in
2804 PH/46 Messages that are created by the autoreply transport now contains a
2805 References: header, in accordance with RFCs 2822 and 3834.
2807 PH/47 Added authenticated_sender_force to the smtp transport.
2809 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
2811 PH/49 Installed latest patch from the Sieve maintainer.
2813 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
2814 was also set, the appendfile transport was unnecessarily scanning a
2815 directory of message files (e.g. for maildir delivery) to find the count
2816 of files (along with the size), even though it did not need this
2817 information. It now does the scan only if it needs to find either the
2818 size of the count of files.
2820 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
2822 PH/52 Two bugs concerned with error handling when the smtp transport is
2825 (i) Exim was not creating retry information for temporary errors given
2826 for individual recipients after the DATA command when the smtp transport
2827 was used in LMTP mode. This meant that they could be retried too
2828 frequently, and not timed out correctly.
2830 (ii) Exim was setting the flag that allows error details to be returned
2831 for LMTP errors on RCPT commands, but not for LMTP errors for individual
2832 recipients that were returned after the DATA command.
2834 PH/53 This is related to PH/52, but is more general: for any failing address,
2835 when detailed error information was permitted to be returned to the
2836 sender, but the error was temporary, then after the final timeout, only
2837 "retry timeout exceeded" was returned. Now it returns the full error as
2838 well as "retry timeout exceeded".
2840 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
2841 do this, and (what is worse) MTAs that accept it.
2843 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
2844 will now be deprecated.
2846 PH/56 New os.c-cygwin from the Cygwin maintainer.
2848 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
2849 formats (previously only available in exim formats via -bpr, -bpru,
2850 and -bpra. Now also available in native and exiqgrep formats)
2852 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
2853 with very large, slow to parse queues
2855 JJ/08 exipick: added ! as generic prefix to negate any criteria format
2857 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
2859 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
2860 responses to authentication challenges, though it was showing the
2861 challenges; (ii) I've removed the CR characters from the debug output for
2864 PH/58 Allow for the insertion of a newline as well as a space when a string
2865 is turned into more than one encoded-word during RFC 2047 encoding. The
2866 Sieve code now uses this.
2868 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
2869 data_4xx, lost_connection, tls_required.
2871 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
2872 message was being sent as an SMTP response.
2874 PH/61 Add -l and -k options to exicyclog.
2876 PH/62 When verifying, if an address was redirected to one new address, so that
2877 verification continued, and the new address failed or deferred after
2878 having set something in $address_data, the value of $address_data was not
2879 passed back to the ACL. This was different to the case when no
2880 redirection occurred. The value is now passed back in both cases.
2882 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
2883 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
2884 use login_cap.h, so on its own it isn't the distinguishing feature. The
2885 new name refers directly to the setclassresources() function.
2887 PH/65 Added configuration files for NetBSD3.
2889 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
2891 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
2892 is preferred over IPv4.
2894 PH/68 The bounce_return_message and bounce_return_body options were not being
2895 honoured for bounces generated during the reception of non-SMTP messages.
2896 In particular, this applied to messages rejected by the ACL. This bug has
2897 been fixed. However, if bounce_return_message is true and bounce_return_
2898 body is false, the headers that are returned for a non-SMTP message
2899 include only those that have been read before the error was detected.
2900 (In the case of an ACL rejection, they have all been read.)
2902 PH/69 The HTML version of the specification is now built in a directory called
2903 spec_html instead of spec.html, because the latter looks like a path with
2904 a MIME-type, and this confuses some software.
2906 PH/70 Catch two compiler warnings in sieve.c.
2908 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
2909 function verify_get_ident() calls ip_connect() to connect a socket, but
2910 if the "connect()" function timed out, ip_connect() used to close the
2911 socket. However, verify_get_ident() also closes the socket later, and in
2912 between Exim writes to the log, which may get opened at this point. When
2913 the socket was closed in ip_connect(), the log could get the same file
2914 descriptor number as the socket. This naturally causes chaos. The fix is
2915 not to close the socket in ip_connect(); the socket should be closed by
2916 the function that creates it. There was only one place in the code where
2917 this was missing, in the iplookup router, which I don't think anybody now
2918 uses, but I've fixed it anyway.
2920 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
2921 well as to direct DNS lookups. Otherwise the handling of names in host
2922 lists is inconsistent and therefore confusing.
2928 PH/01 Two changes to the default runtime configuration:
2930 (1) Move the checks for relay_from_hosts and authenticated clients from
2931 after to before the (commented out) DNS black list checks.
2933 (2) Add control=submission to the relay_from_hosts and authenticated
2934 clients checks, on the grounds that messages accepted by these
2935 statements are most likely to be submissions.
2937 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
2939 (1) Generate an error if the third argument for the ${prvs expansion is
2942 (2) Treat a missing third argument of ${prvscheck as if it were an empty
2945 (3) Reset the variables that are obtained from the first argument of
2946 ${prvscheck and used in the second argument before leaving the code,
2947 because their memory is reclaimed, so using them afterwards may do
2950 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
2951 one (it's much easier than Tom thought :-).
2953 (5) Because of (4), we can now allow for the use of $prvscheck_result
2954 inside the third argument.
2956 PH/03 For some reason, the default setting of PATH when running a command from
2957 a pipe transport was just "/usr/bin". I have changed it to
2960 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
2961 anything to be listed in the output from -bV.
2963 PH/05 When a filter generated an autoreply, the entire To: header line was
2964 quoted in the delivery log line, like this:
2966 => >A.N.Other <ano@some.domain> <original@ddress> ...
2968 This has been changed so that it extracts the operative address. There
2969 may be more than one such address. If so, they are comma-separated, like
2972 => >ano@some.domain,ona@other.domain <original@ddress> ...
2974 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
2975 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
2976 not being looked up in the rDNS to get a host name, Exim was showing the
2977 IP address twice in Received: lines, even though the IP addresses were
2978 identical. For example:
2980 Received: from [1.2.3.4] (helo=[1.2.3.4])
2982 However, if the real host name was known, it was omitting the HELO data
2983 if it matched the actual IP address. This has been tidied up so that it
2984 doesn't show the same IP address twice.
2986 PH/07 When both +timestamp and +memory debugging was on, the value given by
2987 $tod_xxx expansions could be wrong, because the tod_stamp() function was
2988 called by the debug printing, thereby overwriting the timestamp buffer.
2989 Debugging no longer uses the tod_stamp() function when +timestamp is set.
2991 PH/08 When the original message was included in an autoreply transport, it
2992 always said "this is a copy of the message, including all the headers",
2993 even if body_only or headers_only was set. It now gives an appropriate
2996 PH/09 Applied a patch from the Sieve maintainer which:
2998 o fixes some comments
2999 o adds the (disabled) notify extension core
3000 o adds some debug output for the result of if/elsif tests
3001 o points to the current vacation draft in the documentation
3002 and documents the missing references header update
3006 o fixes a bug in processing the envelope test (when testing
3007 multiple envelope elements, the last element determinted the
3010 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
3011 Electronic Mail") by including:
3013 Auto-submitted: auto-generated
3015 in the messages that it generates (bounce messages and others, such as
3016 warnings). In the case of bounce messages for non-SMTP mesages, there was
3017 also a typo: it was using "Auto_submitted" (underscore instead of
3018 hyphen). Since every message generated by Exim is necessarily in response
3019 to another message, thes have all been changed to:
3021 Auto-Submitted: auto-replied
3023 in accordance with these statements in the RFC:
3025 The auto-replied keyword:
3027 - SHOULD be used on messages sent in direct response to another
3028 message by an automatic process,
3030 - MUST NOT be used on manually-generated messages,
3032 - MAY be used on Delivery Status Notifications (DSNs) and Message
3033 Disposition Notifications (MDNs),
3035 - MUST NOT be used on messages generated by automatic or periodic
3036 processes, except for messages which are automatic responses to
3039 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
3040 to the default Received: header definition.
3042 PH/12 Added log selector acl_warn_skipped (default on).
3044 PH/13 After a successful wildlsearch lookup, discard the values of numeric
3045 variables because (a) they are in the wrong storage pool and (b) even if
3046 they were copied, it wouldn't work properly because of the caching.
3048 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
3049 checking when decoding. Apparently there are clients that generate
3050 overlong encoded strings. Why am I not surprised?
3052 PH/15 If the first argument of "${if match_address" was not empty, but did not
3053 contain an "@" character, Exim crashed. Now it writes a panic log message
3054 and treats the condition as false.
3056 PH/16 In autoreply, treat an empty string for "once" the same as unset.
3058 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
3059 extension "envelope-auth". The code is finished and in agreement with
3060 other implementations, but there is no documentation so far and in fact,
3061 nobody wrote the draft yet. This extension is currently #undef'ed, thus
3062 not changing the active code.
3064 Print executed "if" and "elsif" statements when debugging is used. This
3065 helps a great deal to understand what a filter does.
3067 Document more things not specified clearly in RFC3028. I had all this
3068 sorted out, when out of a sudden new issues came to my mind. Oops."
3070 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
3073 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
3074 canonical form (as documented). However, after a host name lookup from
3075 the IP address, check_host() was doing a simple string comparison with
3076 addresses acquired from the DNS when checking that the found name did
3077 have the original IP as one of its addresses. Since any found IPv6
3078 addresses are likely to be in abbreviated form, the comparison could
3079 fail. Luckily, there already exists a function for doing the comparison
3080 by converting both addresses to binary, so now that is used instead of
3081 the text comparison.
3083 PH/20 There was another similar case to PH/19, when a complete host name was
3084 given in a host list; looking up its IP address could give an abbreviated
3085 form, whereas the current host's name might or might not be abbreviated.
3086 The same fix has been applied.
3092 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
3093 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
3096 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
3097 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
3099 PH/03 Typo: missing ".o" in src/pcre/Makefile.
3101 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
3102 header line, restrict the check to what is listed in RFCs 2369 and 2929.
3103 Also, for "Auto-Submitted", treat anything other than "no" as
3104 non-personal, in accordance with RFC 3834. (Previously it treated
3105 anything starting "auto-" as non-personal.)
3107 TF/01 The control=submission/name=... option had a problem with syntax
3108 errors if the name included a slash character. The /name= option
3109 now slurps the rest of the string, so it can include any characters
3110 but it must come last in the list of options (after /sender_retain
3113 PH/05 Some modifications to the interface to the fake nameserver for the new
3121 TK/01 Added the "success_on_redirect" address verification option. See
3122 NewStuff for rationale and an example.
3124 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
3126 PH/02 Patch to exigrep to allow it to work on syslog lines.
3128 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
3129 fread() to skip over the body file's header line, because in Cygwin the
3130 header line is locked and is inaccessible.
3132 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
3133 co-exist for some time) to make it clear that it is the Exim ID that is
3134 referenced, not the Message-ID: header line.
3136 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
3137 string_format() function, because snprintf() does not exist on all
3140 PH/06 The use of forbid_filter_existstest now also locks out the use of the
3141 ${stat: expansion item.
3143 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
3144 protocol synchronization error", to keep the pedants happy.
3146 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
3147 well as for IRIX systems, when gcc is being used. See the host.c source
3150 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
3152 PH/10 Named domain lists were not working if used in a queue_smtp_domains
3155 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
3156 transport and to the smtp transport in LMTP mode.
3158 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
3160 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
3161 run a filter in a subprocess. This could lead to confusion in subsequent
3162 lookups in the parent process. There should also be a search_tidyup() at
3163 the end of the subprocess.
3165 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
3166 only if the host matched helo_try_verify_hosts, which caused the
3167 verification to occur when the EHLO/HELO command was issued. The ACL just
3168 tested the remembered result. Now, if a previous verification attempt has
3169 not happened, "verify = helo" does it there and then.
3171 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
3173 TK/03 Fix log output including CR from clamd.
3175 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
3176 exist provoked a memory error which could cause a segfault.
3178 PH/15 Installed PCRE 6.2
3180 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
3182 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
3183 of the problem. Specifically, suggested +O2 rather than +O1 for the
3186 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
3188 PH/20 If a delivery was routed to a non-standard port by means of an SRV
3189 record, the port was not correctly logged when the outgoing_port log
3190 selector was set (it logged the transort's default port).
3192 PH/21 Added support for host-specific ports to manualroute, queryprogram,
3193 fallback_hosts, and "hosts" in the smtp transport.
3195 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
3196 host errors such as "Connection refused".
3198 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
3199 authentication with radiusclient 0.4.9:
3201 - Error returned from rc_read_config was caught wrongly
3202 - Username/password not passed on to radius server due to wrong length.
3204 The presumption is that some radiusclient API changes for 4.51/PH/17
3205 were not taken care of correctly. The code is still untested by me (my
3206 Linux distribution still has 0.3.2 of radiusclient), but it was
3207 contributed by a Radius user.
3209 PH/24 When doing a callout, the value of $domain wasn't set correctly when
3210 expanding the "port" option of the smtp transport.
3212 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
3213 while reading a MIME header. Thanks to Tom Hughes for a patch.
3215 PH/24 Include config.h inside local_scan.h so that configuration settings are
3218 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
3219 that in an ACL for RCPT (for example), you can examine exactly what was
3222 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
3223 commands, but it was not correctly comparing the address with the actual
3224 client host address. Thus, it would show the EHLO address in Received:
3225 header lines when this was not necessary.
3227 PH/27 Added the % operator to ${eval:}.
3229 PH/28 Exim tries to create and chdir to its spool directory when it starts;
3230 it should be ignoring failures (because with -C, for example, it has lost
3231 privilege). It wasn't ignoring creation failures other than "already
3234 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
3237 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
3238 because an input line was too long, either on its own, or by virtue of
3239 too many continuations, the temporary file was not being removed, and the
3240 return code was incorrect.
3242 PH/31 Missing "BOOL" in function definition in filtertest.c.
3244 PH/32 Applied Sieve patches from the maintainer.
3246 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
3248 PH/33 Added "verify = not_blind".
3250 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
3251 Local/Makefile (with some defaults set). These are used in built scripts
3252 such as exicyclog, but they have never been used in the exim_install
3253 script (though there are many overriding facilities there). I have
3254 arranged that the exim_install script now takes note of these two
3257 PH/35 Installed configuration files for Dragonfly.
3259 PH/36 When a locally submitted message by a trusted user did not contain a
3260 From: header, and the sender address was obtained from -f or from an SMTP
3261 MAIL command, and the trusted user did not use -F to supply a sender
3262 name, $originator_name was incorrectly used when constructing a From:
3263 header. Furthermore, $originator_name was used for submission mode
3264 messages from external hosts without From: headers in a similar way,
3265 which is clearly wrong.
3267 PH/37 Added control=suppress_local_fixups.
3269 PH/38 When log_selector = +received_sender was set, and the addition of the
3270 sender made the log line's construction buffer exactly full, or one byte
3271 less than full, an overflow happened when the terminating "\n" was
3274 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
3275 when the result of a list match is failure because a DNS lookup failed.
3277 PH/40 RM_COMMAND is now used in the building process.
3279 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
3280 the "build-* directories that it finds.
3282 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
3283 address in a domain literal was a prefix of an interface address.
3285 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
3286 when verifying a sender address, unless rewrite_headers is false.
3288 PH/44 Wrote a long comment about why errors_to addresses are verified as
3289 recipients, not senders.
3291 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
3292 the ratelimit ACL was added.
3294 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
3296 PH/46 Added extra information about PostgreSQL errors to the error string.
3298 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
3299 avoiding the need to install special zones in a real server. This is
3300 backwards compatible; if it can't find the fake resolver, it drops back.
3301 Thus, both old and new test suites can be run.
3303 TF/02 Added util/ratelimit.pl
3305 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
3306 clock is set back in time.
3308 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
3309 Candler <B.Candler@pobox.com>.
3311 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
3312 OK for addresses that are the result of redirections.
3314 PH/48 A number of further additions for the benefit of the new test suite,
3315 including a fake gethostbyname() that interfaces to the fake DNS resolver
3318 TF/06 The fix for widen_domains has also been applied to qualify_single and
3319 search_parents which are the other dnslookup options that can cause
3322 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
3325 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
3327 TF/07 Exim produced the error message "an SRV record indicated no SMTP
3328 service" if it encountered an MX record with an empty target hostname.
3329 The message is now "an MX or SRV record indicated no SMTP service".
3331 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
3332 if the DNS of the sending site is misconfigured. This is quite a
3333 common situation. This change restores the behaviour of treating a
3334 helo verification defer as a failure.
3336 PH/51 If self=fail was set on a router, the bounce message did not include the
3337 actual error message.
3343 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
3345 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
3346 command itself ended in error, the underlying message about the transport
3347 filter timeout was being overwritten with the pipe command error. Now the
3348 underlying error message should be appended to the second error message.
3350 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
3352 PH/02 Reduce the amount of output that "make" produces by default. Full output
3353 can still be requested.
3355 PH/03 The warning log line about a condition test deferring for a "warn" verb
3356 was being output only once per connection, rather than after each
3357 occurrence (because it was using the same function as for successful
3358 "warn" verbs). This seems wrong, so I have changed it.
3360 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
3361 it should not have, which might have caused a crash in the right
3362 circumstances, but probably never did.
3364 PH/04 Installed a modified version of Tony Finch's patch to make submission
3365 mode fix the return path as well as the Sender: header line, and to
3366 add a /name= option so that you can make the user's friendly name appear
3369 TF/03 Added the control = fakedefer ACL modifier.
3371 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
3372 Mark Lowes for thorough testing.
3374 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
3376 TK/03 Merged latest SRS patch from Miles Wilton.
3378 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
3379 with the definition in sysexits.h (which is #included earlier).
3380 Fortunately, Exim does not actually use EX_OK. The code used to try to
3381 preserve the sysexits.h value, by assumimg that macro definitions were
3382 scanned for macro replacements. I have been disabused of this notion,
3383 so now the code just undefines EX_OK before #including unistd.h.
3385 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
3386 in the smtp transport. When a block could not be written in a single
3387 write() function, the timeout was being re-applied to each part-write.
3388 This seems wrong - if the receiver was accepting one byte at a time it
3389 would take for ever. The timeout is now adjusted when this happens. It
3390 doesn't have to be particularly precise.
3392 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
3393 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
3395 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
3396 without a domain if the check to <postmaster@domain> fails.
3398 SC/01 Eximstats: added -xls and the ability to specify output files
3399 (patch written by Frank Heydlauf).
3401 SC/02 Eximstats: use FileHandles for outputing results.
3403 SC/03 Eximstats: allow any combination of xls, txt, and html output.
3405 SC/04 Eximstats: fixed display of large numbers with -nvr option
3407 SC/05 Eximstats: fixed merging of reports with empty tables.
3409 SC/06 Eximstats: added the -include_original_destination flag
3411 SC/07 Eximstats: removed tabs and trailing whitespace.
3413 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
3415 TK/06 MBOX spool code: Add real "From " MBOX separator line
3416 so the .eml file is really in mbox format (even though
3417 most programs do not really care). Patch from Alex Miller.
3419 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
3420 The latter is generated from $received_to and is only set if the
3421 message has one envelope recipient. SA can use these headers,
3422 obviously out-of-the-box. Patch from Alex Miller.
3424 PH/08 The ${def test on a variable was returning false if the variable's
3425 value was "0", contrary to what the specification has always said!
3426 The result should be true unless the variable is empty.
3428 PH/09 The syntax error of a character other than { following "${if
3429 def:variable_name" (after optional whitespace) was not being diagnosed.
3430 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
3431 accidental colon was present, for example, could give incorrect results.
3433 PH/10 Tidied the code in a number of places where the st_size field of a stat()
3434 result is used (not including appendfile, where other changes are about
3437 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
3438 This involved changing a lot of size variables from int to off_t. It
3439 should work with maildirs and everything.
3441 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
3442 spamd dying while we are connected to it.
3444 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
3447 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
3448 function for BDB 4.3.
3450 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
3452 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
3453 the sources for PCRE have been re-arranged and more files are now
3456 PH/15 The code I had for printing potentially long long variables in PH/11
3457 above was not the best (it lost precision). The length of off_t variables
3458 is now inspected at build time, and an appropriate printing format (%ld
3459 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
3460 to be "long long int" or "long int". This is needed for the internal
3461 formatting function string_vformat().
3463 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
3464 the configuration file to be ":syslog", then the script "guesses" where
3465 the logs files are, rather than using the compiled in default. In our
3466 case the guess is not the same as the compiled default, so the script
3467 suddenly stopped working when I started to use syslog. The patch checks
3468 to see if log_file_path is "". If so, it attempts to read it from exim
3469 with no configuration file to get the compiled in version, before it
3470 falls back to the previous guessing code."
3472 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
3473 implementing BATV in an Exim configuration. See NewStuff for the gory
3476 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
3477 Makefile that are specific to HP-UX.
3479 PH/18 If the "use_postmaster" option was set for a recipient callout together
3480 with the "random" option, the postmaster address was used as the MAIL
3481 FROM address for the random test, but not for the subsequent recipient
3482 test. It is now used for both.
3484 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
3485 patch removes a few documentation additions to RFC 3028, because the
3486 latest draft now contains them. It adds the new en;ascii-case comparator
3487 and a new error check for 8bit text in MIME parts. Comparator and
3488 require names are now matched exactly. I enabled the subaddress
3489 extension, but it is not well tested yet (read: it works for me)."
3491 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
3492 rework some of the code of TK/09 above to avoid the hardwired use of
3493 "%lld" and "long long". Replaced the call to snprintf() with a call to
3496 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
3497 records point to non-existent hosts", "retry timeout exceeded", and
3498 "retry time not reached for any host after a long failure period".
3500 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
3501 experimental DomainKeys support:
3503 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
3504 (2) On an error such as an illegally used "control", the wrong name for
3505 the control was given.
3507 These problems did NOT occur unless DomainKeys support was compiled.
3509 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
3511 PH/24 Added ${if match_ip condition.
3513 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
3514 databases so that it will be absolutely obvious if a crash occurs in the
3515 DB library. This is a regular occurrence (often caused by mis-matched
3518 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
3519 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
3520 picked up on a user's system that detects such things. There doesn't seem
3521 to be a gcc warning option for this - only an attribute that has to be
3522 put on the function's prototype. It seems that in Fedora Core 4 they have
3523 set this on a number of new functions. No doubt there will be more in due
3526 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
3527 specify a transport. However, if an address that was verified by such a
3528 router was the subject of a callout, Exim crashed because it tried to
3529 read the rcpt_include_affixes from the non-existent transport. Now it
3530 just assumes that the setting of that option is false. This bug was
3531 introduced by 4.51/PH/31.
3533 PH/28 Changed -d+all to exclude +memory, because that information is very
3534 rarely of interest, but it makes the output a lot bigger. People tend to
3535 do -d+all out of habit.
3537 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
3538 code in os-type was giving problems when libc.so lives in lib64, like on
3541 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
3542 aren't the modern standard, and it seems that some systems' include files
3543 don't always have them. Exim was already checking for some of the newer
3544 ones like T_AAAA, and defining it itself. I've added checks for all the
3545 record types that Exim uses.
3547 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
3548 not automatically generating a new one, as it is supposed to. This
3549 prevented TLS from working. If the file did exist, but contained invalid
3550 data, a new version was generated, as expected. It was only the case of a
3551 non-existent file that was broken.
3553 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
3554 with a change in libdomainkeys > 0.64.
3556 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
3557 from DNS. If the selector record carries the flag, it now has
3558 precedence over the domain-wide flag.
3560 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
3562 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
3563 the use of a port name that isn't defined in /etc/services) occurred, the
3564 message was deferred as in a normal delivery, and thus remained on the
3565 spool, instead of being failed because of the mua_wrapper setting. This
3566 is now fixed, and I tidied up some of the mua_wrapper messages at the
3569 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
3570 the messages in a hash of arrays rather than using individual hashes.
3571 This is a bit cleaner and results in dramatic memory savings, albeit
3572 at a slight CPU cost.
3574 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
3575 as requested by Marc Sherman.
3577 SC/10 Eximstats: added histograms for user specified patterns as requested
3580 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
3582 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
3583 fopen() in the content-scanning modules that did not already have it.
3589 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
3590 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
3592 TK/02 Fix ACL "control" statement not being available in MIME ACL.
3594 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
3596 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
3597 to test Sieve filters that use "vacation".
3599 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
3600 that changes the way the GnuTLS parameters are stored in the cache file.
3601 The new format can be generated externally. For backward compatibility,
3602 if the data in the cache doesn't make sense, Exim assumes it has read an
3603 old-format file, and it generates new data and writes a new file. This
3604 means that you can't go back to an older release without removing the
3607 PH/03 A redirect router that has both "unseen" and "one_time" set does not
3608 work if there are any delivery delays because "one_time" forces the
3609 parent to be marked "delivered", so its unseen clone is never tried
3610 again. For this reason, Exim now forbids the simultaneous setting of
3613 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
3614 redirected to themselves ("homonym" addresses). Read the long ChangeLog
3615 entry if you want to know the details. The fix, however, neglected to
3616 consider the case when local delivery batching is involved. The test for
3617 "previously delivered" was not happening when checking to see if an
3618 address could be batched with a previous (undelivered) one; under
3619 certain circumstances this could lead to multiple deliveries to the same
3622 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
3623 in its include files, and this causes problems building Exim.
3625 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
3626 header_syntax) but Exim was just ignoring anything given after a slash.
3627 In particular, this caused confusion with an attempt to use "verify =
3628 reverse_host_lookup/defer_ok". An error is now given when options are
3629 supplied for verify items that do not have them. (Maybe reverse_host_
3630 lookup should have a defer_ok option, but that's a different point.)
3632 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
3633 defined by RFC 821) to 2048, because there were problems with some AUTH
3634 commands, and RFC 1869 says the size should be increased for extended
3635 SMTP commands that take arguments.
3637 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
3640 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
3641 "unknown" error; now it says that the functionality isn't in the binary.
3643 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
3644 an address' error message when a string expansion fails (syntax or
3645 whatever). Otherwise the password may appear in the log. Following change
3646 PH/42 below, there is no longer a chance of it appearing in a bounce
3649 PH/11 Installed exipick version 20050225.0 from John Jetmore.
3651 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
3652 of its addresses was ever tried. (Bugzilla bug #2.)
3654 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
3655 the result incorrectly in the debug output. (It correctly added a newline
3656 to what was transported.)
3658 TF/01 Added $received_time.
3660 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
3661 commented out examples of how to interface to a virus scanner and to
3662 SpamAssassin. Also added commented examples of av_scanner and
3663 spamd_address settings.
3665 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
3666 and controls are allowed in which ACLs. There were a couple of minor
3667 errors. Some of the entries in the conditions table (which is a table of
3668 where they are NOT allowed) were getting very unwieldy; rewrote them as a
3669 negation of where the condition IS allowed.
3671 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
3673 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
3674 header file does not have a version number, so I've had to invent a new
3675 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
3676 API. The code is untested by me (my Linux distribution still has 0.3.2 of
3677 radiusclient), but it was contributed by a Radius user.
3679 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
3680 files or directories, for OpenSSL.
3682 PH/19 When an Exim process that is running as root has to create an Exim log
3683 file, it does so in a subprocess that runs as exim:exim so as to get the
3684 ownership right at creation (otherwise, other Exim processes might see
3685 the file with the wrong ownership). There was no test for failure of this
3686 fork() call, which would lead to the process getting stuck as it waited
3687 for a non-existent subprocess. Forks do occasionally fail when resources
3688 run out. I reviewed all the other calls to fork(); they all seem to check
3691 PH/20 When checking for unexpected SMTP input at connect time (before writing
3692 the banner), Exim was not dealing correctly with a non-positive return
3693 from the read() function. If the client had disconnected by this time,
3694 the result was a log entry for a synchronization error with an empty
3695 string after "input=" when read() returned zero. If read() returned -1
3696 (an event I could not check), uninitialized data bytes were printed.
3697 There were reports of junk text (parts of files, etc) appearing after
3700 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
3702 PH/22 Added support for macro redefinition, and (re)definition in between
3703 driver and ACL definitions.
3705 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
3706 forgetting to use the resulting value; it was using the unexpanded value.
3708 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
3709 hadn't been configured. The fix is from Juergen Kreileder, who
3710 understands it better than I do:
3712 "Here's what I see happening with three configured cyrus_sasl
3713 authenticators configured (plain, login, cram-md5):
3715 On startup auth_cyrus_sasl_init() gets called for each of these.
3716 This means three calls to sasl_listmech() without a specified mech_list.
3717 => SASL tests which mechs of all available mechs actually work
3718 => three warnings about OTP not working
3719 => the returned list contains: plain, login, cram-md5, digest-md5, ...
3721 With the patch, sasl_listmech() also gets called three times. But now
3722 SASL's mech_list option is set to the server_mech specified in the the
3723 authenticator. Or in other words, the answer from sasl_listmech()
3724 gets limited to just the mech you're testing for (which is different
3726 => the return list contains just 'plain' or 'login', 'cram-md5' or
3727 nothing depending on the value of ob->server_mech.
3729 I've just tested the patch: Authentication still works fine,
3730 unavailable mechs specified in the exim configuration are still
3731 caught, and the auth.log warnings about OTP are gone."
3733 PH/25 When debugging is enabled, the contents of the command line are added
3734 to the debugging output, even when log_selector=+arguments is not
3737 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
3738 answer is "GNU", and only if the return is "GNU/something" is the answer
3741 PH/27 $acl_verify_message is now set immediately after the failure of a
3742 verification in an ACL, and so is available in subsequent modifiers. In
3743 particular, the message can be preserved by coding like this:
3745 warn !verify = sender
3746 set acl_m0 = $acl_verify_message
3748 Previously, $acl_verify_message was set only while expanding "message"
3749 and "log_message" when a very denied access.
3751 PH/28 Modified OS/os.c-Linux with
3753 -#ifndef OS_LOAD_AVERAGE
3754 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
3756 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
3757 nomenclature these days.)
3759 PH/29 Installed patch from the Sieve maintainer that adds the options
3760 sieve_useraddress and sieve_subaddress to the redirect router.
3762 PH/30 In these circumstances:
3763 . Two addresses routed to the same list of hosts;
3764 . First host does not offer TLS;
3765 . First host accepts first address;
3766 . First host gives temporary error to second address;
3767 . Second host offers TLS and a TLS session is established;
3768 . Second host accepts second address.
3769 Exim incorrectly logged both deliveries with the TLS parameters (cipher
3770 and peerdn, if requested) that were in fact used only for the second
3773 PH/31 When doing a callout as part of verifying an address, Exim was not paying
3774 attention to any local part prefix or suffix that was matched by the
3775 router that accepted the address. It now behaves in the same way as it
3776 does for delivery: the affixes are removed from the local part unless
3777 rcpt_include_affixes is set on the transport.
3779 PH/32 Add the sender address, as F=<...>, to the log line when logging a
3780 timeout during the DATA phase of an incoming message.
3782 PH/33 Sieve envelope tests were broken for match types other than :is. I have
3783 applied a patch sanctioned by the Sieve maintainer.
3785 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
3786 the uid or gid is negative. A case of a negative gid caused this to be
3787 noticed. The fix allows for either to be negative.
3789 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
3790 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
3793 PH/36 The change PH/12 above was broken. Fixed it.
3795 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
3796 the grounds that routing the same address twice would always produce the
3797 same answer. This might have been true once, but it is certainly no
3798 longer true now. Routing a child address may depend on the previous
3799 routing that produced that child. Some complicated redirection strategies
3800 went wrong when messages had multiple recipients, and made Exim's
3801 behaviour dependent on the order in which the addresses were given.
3803 I have moved the duplicate checking until after the routing is complete.
3804 Exim scans the addresses that are assigned to local and remote
3805 transports, and removes any duplicates. This means that more work will be
3806 done, as duplicates will always all be routed, but duplicates are
3807 presumably rare, so I don't expect this is of any significance.
3809 For deliveries to pipes, files, and autoreplies, the duplicate checking
3810 still happens during the routing process, since they are not going to be
3813 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
3814 It corrects a timeout issue with spamd. This is Ian's comment: "The
3815 background is that sometimes spamd either never reads data from a
3816 connection it has accepted, or it never writes response data. The exiscan
3817 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
3818 blindly assumes that writes won't block so it may never time out."
3820 PH/39 Allow G after quota size as well as K and M.
3822 PH/40 The value set for $authenticated_id in an authenticator may not contain
3823 binary zeroes or newlines because the value is written to log lines and
3824 to spool files. There was no check on this. Now the value is run through
3825 the string_printing() function so that such characters are converted to
3826 printable escape sequences.
3828 PH/41 $message_linecount is a new variable that contains the total number of
3829 lines in the message. Compare $body_linecount, which is the count for the
3832 PH/42 Exim no longer gives details of delivery errors for specific addresses in
3833 bounce and delay warning messages, except in certain special cases, which
3836 (a) An SMTP error message from a remote host;
3837 (b) A message specified in a :fail: redirection;
3838 (c) A message specified in a "fail" command in a system filter;
3839 (d) A message specified in a FAIL return from the queryprogram router;
3840 (e) A message specified by the cannot_route_message router option.
3842 In these cases only, Exim does include the error details in bounce and
3843 warning messages. There are also a few cases where bland messages such
3844 as "unrouteable address" or "local delivery error" are given.
3846 PH/43 $value is now also set for the "else" part of a ${run expansion.
3848 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
3849 being worked on, but at least Exim now implements the latest version to
3852 PH/45 In a pipe transport, although a timeout while waiting for the pipe
3853 process to complete was treated as a delivery failure, a timeout while
3854 writing the message to the pipe was logged, but erroneously treated as a
3855 successful delivery. Such timeouts include transport filter timeouts. For
3856 consistency with the overall process timeout, these timeouts are now
3857 treated as errors, giving rise to delivery failures by default. However,
3858 there is now a new Boolean option for the pipe transport called
3859 timeout_defer, which, if set TRUE, converts the failures into defers for
3860 both kinds of timeout. A transport filter timeout is now identified in
3863 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
3864 systems where "make" and "gmake" are different, calling "gmake" at top
3865 level broke things. I've arranged for the value of $(MAKE) to be passed
3866 from the Makefile to this script so that it can call the same version of
3870 A note about Exim versions 4.44 and 4.50
3871 ----------------------------------------
3873 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
3874 changes of various kinds. As a consequence, a big documentation update was
3875 needed. This delayed the release for rather longer than seemed good, especially
3876 in the light of a couple of (minor) security issues. Therefore, the changes
3877 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
3878 release. So 4.44 and 4.50 are in effect two different branches that both start
3881 I have left the 4.50 change log unchanged; it contains all the changes since
3882 4.43. The change log for 4.44 is below; many of its items are identical to
3883 those for 4.50. This seems to be the most sensible way to preserve the
3884 historical information.
3890 1. Minor wording change to the doc/README.SIEVE file.
3892 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
3893 computation of the current number of files was incorrect.
3895 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
3896 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
3899 4. Give more explanation in the error message when the command for a transport
3900 filter fails to execute.
3902 5. There are several places where Exim runs a non-Exim command in a
3903 subprocess. The SIGUSR1 signal should be disabled for these processes. This
3904 was being done only for the command run by the queryprogram router. It is
3905 now done for all such subprocesses. The other cases are: ${run, transport
3906 filters, and the commands run by the lmtp and pipe transports.
3908 6. Added CONFIGURE_GROUP build-time option.
3910 7. Some older OS have a limit of 256 on the maximum number of file
3911 descriptors. Exim was using setrlimit() to set 1000 as a large value
3912 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
3913 systems. I've change it so that if it can't get 1000, it tries for 256.
3915 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
3916 was an oversight, and furthermore, ever since the addition of extra
3917 controls (e.g. 4.43/32), the checks on when to allow different forms of
3918 "control" were broken. There should now be diagnostics for all cases when a
3919 control that does not make sense is encountered.
3921 9. Added the /retain_sender option to "control=submission".
3923 10. $recipients is now available in the predata ACL (oversight).
3925 11. Tidy the search cache before the fork to do a delivery from a message
3926 received from the command line. Otherwise the child will trigger a lookup
3927 failure and thereby defer the delivery if it tries to use (for example) a
3928 cached ldap connection that the parent has called unbind on.
3930 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
3931 of $address_data from the recipient verification was clobbered by the
3932 sender verification.
3934 13. The value of address_data from a sender verification is now available in
3935 $sender_address_data in subsequent conditions in the ACL statement.
3937 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
3939 15. Added a new option "connect=<time>" to callout options, to set a different
3942 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
3943 was its contents. (It was OK if the option was not defined at all.)
3945 17. A "Completed" log line is now written for messages that are removed from
3946 the spool by the -Mrm option.
3948 18. New variables $sender_verify_failure and $recipient_verify_failure contain
3949 information about exactly what failed.
3951 19. Added -dd to debug only the daemon process.
3953 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
3954 handles timeouts, both on the server side and network timeouts. Renamed the
3955 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
3957 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
3958 It is now set to "smtps".
3960 22. $host_address is now set to the target address during the checking of
3961 ignore_target_hosts.
3963 23. When checking ignore_target_hosts for an ipliteral router, no host name was
3964 being passed; this would have caused $sender_host_name to have been used if
3965 matching the list had actually called for a host name (not very likely,
3966 since this list is usually IP addresses). A host name is now passed as
3969 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
3970 code that specifies a non-restarting handler (typically sigaction() in
3971 modern systems) in an attempt to fix a rare and obscure crash bug.
3973 25. Narrowed the window for a race in the daemon that could cause it to ignore
3974 SIGCHLD signals. This is not a major problem, because they are used only to
3975 wake it up if nothing else does.
3977 26. A malformed maildirsize file could cause Exim to calculate negative values
3978 for the mailbox size or file count. Odd effects could occur as a result.
3979 The maildirsize information is now recalculated if the size or filecount
3982 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
3983 support for a long time. Removed HAVE_SYS_VFS_H.
3985 28. Installed the latest version of exipick from John Jetmore.
3987 29. In an address list, if the pattern was not a regular expression, an empty
3988 subject address (from a bounce message) matched only if the pattern was an
3989 empty string. Non-empty patterns were not even tested. This was the wrong
3990 because it is perfectly reasonable to use an empty address as part of a
3991 database query. An empty address is now tested by patterns that are
3992 lookups. However, all the other forms of pattern expect the subject to
3993 contain a local part and a domain, and therefore, for them, an empty
3994 address still always fails if the pattern is not itself empty.
3996 30. Exim went into a mad DNS loop when attempting to do a callout where the
3997 host was specified on an smtp transport, and looking it up yielded more
3998 than one IP address.
4000 31. Re-factored the code for checking spool and log partition space into a
4001 function that finds that data and another that does the check. The former
4002 is then used to implement four new variables: $spool_space, $log_space,
4003 $spool_inodes, and $log_inodes.
4005 32. The RFC2047 encoding function was originally intended for short strings
4006 such as real names; it was not keeping to the 75-character limit for
4007 encoded words that the RFC imposes. It now respects the limit, and
4008 generates multiple encoded words if necessary. To be on the safe side, I
4009 have increased the buffer size for the ${rfc2047: expansion operator from
4012 33. It is now permitted to omit both strings after an "if" condition; if the
4013 condition is true, the result is "true". As before, when the second string
4014 is omitted, a false condition yields an empty string. This makes it less
4015 cumbersome to write custom ACL and router conditions.
4017 34. Failure to deliver a bounce message always caused it to be frozen, even if
4018 there was an errors_to setting on the router. The errors_to setting is now
4021 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4022 canonical form (fully expanded) before being placed in
4023 $sender_host_address.
4025 36. The table in the code that translates DNS record types into text (T_A to
4026 "A" for instance) was missing entries for NS and CNAME. It is just possible
4027 that this could have caused confusion if both these types were looked up
4028 for the same domain, because the text type is used as part of Exim's
4029 per-process caching. But the chance of anyone hitting this buglet seems
4032 37. The dnsdb lookup has been extended in a number of ways.
4034 (1) There is a new type, "zns", which walks up the domain tree until it
4035 finds some nameserver records. It should be used with care.
4037 (2) There is a new type, "mxh", which is like "mx" except that it returns
4038 just the host names, not the priorities.
4040 (3) It is now possible to give a list of domains (or IP addresses) to be
4041 looked up. The behaviour when one of the lookups defers can be
4042 controlled by a keyword.
4044 (4) It is now possible to specify the separator character for use when
4045 multiple records are returned.
4047 38. The dnslists ACL condition has been extended: it is now possible to supply
4048 a list of IP addresses and/or domains to be looked up in a particular DNS
4051 39. Added log_selector=+queue_time_overall.
4053 40. When running the queue in the test harness, wait just a tad after forking a
4054 delivery process, to get repeatability of debugging output.
4056 41. Include certificate and key file names in error message when GnuTLS fails
4057 to set them up, because the GnuTLS error message doesn't include the name
4058 of the failing file when there is a problem reading it.
4060 42. Allow both -bf and -bF in the same test run.
4062 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
4064 44. The "Exiscan patch" is now merged into the mainline Exim source.
4066 45. Sometimes the final signoff response after QUIT could fail to get
4067 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4068 < 0 before doing a fflush(). This bug looks as though it goes back to the
4069 introduction of TLS in release 3.20, but "sometimes" must have been rare
4070 because the tests only now provoked it.
4072 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
4073 (this can affect the format of dates).
4075 47. exim_tidydb, when checking for the continued existence of a message for
4076 which it has found a message-specific retry record, was not finding
4077 messages that were in split spool directories. Consequently, it was
4078 deleting retry records that should have stayed in existence.
4080 48. Steve fixed some bugs in eximstats.
4082 49. The SPA authentication driver was not abandoning authentication and moving
4083 on to the next authenticator when an expansion was forced to fail,
4084 contradicting the general specification for all authenticators. Instead it
4085 was generating a temporary error. It now behaves as specified.
4087 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
4088 (the order specifies the preference for clients). The order is now AES256,
4089 AES128, 3DES, ARCFOUR128.
4091 51. Small patch to Sieve code - explicitly set From: when generating an
4094 52. Exim crashed if a remote delivery caused a very long error message to be
4095 recorded - for instance if somebody sent an entire SpamAssassin report back
4096 as a large number of 550 error lines. This bug was coincidentally fixed by
4097 increasing the size of one of Exim's internal buffers (big_buffer) that
4098 happened as part of the Exiscan merge. However, to be on the safe side, I
4099 have made the code more robust (and fixed the comments that describe what
4102 53. Now that there can be additional text after "Completed" in log lines (if
4103 the queue_time_overall log selector is set), a one-byte patch to exigrep
4104 was needed to allow it to recognize "Completed" as not the last thing in
4107 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
4108 patch that reportedly fixes this has been added. I am not expert enough to
4109 create a test for it. This is what the patch creator wrote:
4111 "I found a little strange behaviour of ldap code when working with
4112 Windows 2003 AD Domain, where users was placed in more than one
4113 Organization Units. When I tried to give exim partial DN, the exit code
4114 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
4115 But simultaneously result of request was absolutely normal ldap result,
4116 so I produce this patch..."
4118 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
4119 REFERENCE, so I have modified the code to exclude the patch when that macro
4122 55. Some experimental protocols are using DNS PTR records for new purposes. The
4123 keys for these records are domain names, not reversed IP addresses. The
4124 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4125 leaves it alone. Component reversal etc. now happens only for IP addresses.
4128 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
4130 57. Double the size of the debug message buffer (to 2048) so that more of very
4131 long debug lines gets shown.
4133 58. The exicyclog utility now does better if the number of log files to keep
4134 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
4136 59. Two changes related to the smtp_active_hostname option:
4138 (1) $smtp_active_hostname is now available as a variable.
4139 (2) The default for smtp_banner uses $smtp_active_hostname instead
4140 of $primary_hostname.
4142 60. The host_aton() function is supposed to be passed a string that is known
4143 to be a valid IP address. However, in the case of IPv6 addresses, it was
4144 not checking this. This is a hostage to fortune. Exim now panics and dies
4145 if the condition is not met. A case was found where this could be provoked
4146 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4147 components; fortuitously, this particular loophole had already been fixed
4148 by change 4.50/55 above.
4150 If there are any other similar loopholes, the new check in host_aton()
4151 itself should stop them being exploited. The report I received stated that
4152 data on the command line could provoke the exploit when Exim was running as
4153 exim, but did not say which command line option was involved. All I could
4154 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4155 running as the user.
4158 61. There was a buffer overflow vulnerability in the SPA authentication code
4159 (which came originally from the Samba project). I have added a test to the
4160 spa_base64_to_bits() function which I hope fixes it.
4163 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
4164 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
4166 63. The daemon start-up calls getloadavg() while still root for those OS that
4167 need the first call to be done as root, but it missed one case: when
4168 deliver_queue_load_max is set with deliver_drop_privilege. This is
4169 necessary for the benefit of the queue runner, because there is no re-exec
4170 when deliver_drop_privilege is set.
4172 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
4173 This has been fixed.
4175 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4176 was in use, was not putting the data itself into the right store pool;
4177 consequently, it could be overwritten for a subsequent message in the same
4178 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4181 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
4183 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
4184 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
4185 the function were treating the return as a boolean value, which happened to
4186 work because 0=false and not-0=true, but is not correct code.
4188 68. The host_aton() function was not handling scoped IPv6 addresses (those
4189 with, for example, "%eth0" on the end) correctly.
4191 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
4192 negated items (that is, ~something) in unsigned ints. Some compilers
4193 apparently mutter when there is no cast.
4195 70. If an address verification called from an ACL failed, and did not produce a
4196 user-specific message (i.e. there was only a "system" message), nothing was
4197 put in $acl_verify_message. In this situation, it now puts the system
4200 71. Change 4.23/11 added synchronization checking at the start of an SMTP
4201 session; change 4.31/43 added the unwanted input to the log line - except
4202 that it did not do this in the start of session case. It now does.
4204 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
4205 This is wrong and can cause the other end to generate a synchronization
4206 error if it is another Exim or anything else that does the synchronization
4207 check. A QUIT command is no longer sent after a timeout.
4209 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
4210 during host lookups.
4212 74. The defer_ok option of callout verification was not working if it was used
4213 when verifying addresses in header lines, that is, for this case:
4215 verify = header_sender/callout=defer_ok
4217 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
4218 those file descriptors could be used for SMTP connections. If anything
4219 wrote to stderr (the example that came up was "warn" in embedded Perl), it
4220 could be sent to the SMTP client, causing chaos. The daemon now opens
4221 stdin, stdout, and stderr to /dev/null when it puts itself into the
4224 76. Arrange for output from Perl's "warn" command to be written to Exim's main
4225 log by default. The user can override this with suitable Perl magic.
4227 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
4228 the log message when discard triggers, was not working for the DATA ACL or
4229 for the non-SMTP ACL.
4231 78. Error message wording change in sieve.c.
4233 79. If smtp_accept_max_per_host was set, the number of connections could be
4234 restricted to fewer than expected, because the daemon was trying to set up
4235 a new connection before checking whether the processes handling previous
4236 connections had finished. The check for completed processes is now done
4237 earlier. On busy systems, this bug wouldn't be noticed because something
4238 else would have woken the daemon, and it would have reaped the completed
4241 80. If a message was submitted locally by a user whose login name contained one
4242 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
4243 It caused a spool format error. I have fixed the spool reading code. A
4244 related problem was that the "from" clause in the Received: line became
4245 illegal because of the space(s). It is now covered by ${quote_local_part.
4247 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
4250 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
4252 83. Patch from Sieve maintainer to fix unterminated string problem in
4253 "vacation" handling.
4255 84. Some minor changes to the Linux configuration files to help with other
4256 OS variants using glibc.
4258 85. One more patch for Sieve to update vacation handling to latest spec.
4261 ----------------------------------------------------
4262 See the note above about the 4.44 and 4.50 releases.
4263 ----------------------------------------------------
4269 1. Change 4.43/35 introduced a bug that caused file counts to be
4270 incorrectly computed when quota_filecount was set in an appendfile
4273 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4274 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4277 3. Give more explanation in the error message when the command for a transport
4278 filter fails to execute.
4280 4. There are several places where Exim runs a non-Exim command in a
4281 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4282 was being done only for the command run by the queryprogram router. It is
4283 now done for all such subprocesses. The other cases are: ${run, transport
4284 filters, and the commands run by the lmtp and pipe transports.
4286 5. Some older OS have a limit of 256 on the maximum number of file
4287 descriptors. Exim was using setrlimit() to set 1000 as a large value
4288 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4289 systems. I've change it so that if it can't get 1000, it tries for 256.
4291 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4292 was an oversight, and furthermore, ever since the addition of extra
4293 controls (e.g. 4.43/32), the checks on when to allow different forms of
4294 "control" were broken. There should now be diagnostics for all cases when a
4295 control that does not make sense is encountered.
4297 7. $recipients is now available in the predata ACL (oversight).
4299 8. Tidy the search cache before the fork to do a delivery from a message
4300 received from the command line. Otherwise the child will trigger a lookup
4301 failure and thereby defer the delivery if it tries to use (for example) a
4302 cached ldap connection that the parent has called unbind on.
4304 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4305 of $address_data from the recipient verification was clobbered by the
4306 sender verification.
4308 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4309 was its contents. (It was OK if the option was not defined at all.)
4311 11. A "Completed" log line is now written for messages that are removed from
4312 the spool by the -Mrm option.
4314 12. $host_address is now set to the target address during the checking of
4315 ignore_target_hosts.
4317 13. When checking ignore_target_hosts for an ipliteral router, no host name was
4318 being passed; this would have caused $sender_host_name to have been used if
4319 matching the list had actually called for a host name (not very likely,
4320 since this list is usually IP addresses). A host name is now passed as
4323 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4324 code that specifies a non-restarting handler (typically sigaction() in
4325 modern systems) in an attempt to fix a rare and obscure crash bug.
4327 15. Narrowed the window for a race in the daemon that could cause it to ignore
4328 SIGCHLD signals. This is not a major problem, because they are used only to
4329 wake it up if nothing else does.
4331 16. A malformed maildirsize file could cause Exim to calculate negative values
4332 for the mailbox size or file count. Odd effects could occur as a result.
4333 The maildirsize information is now recalculated if the size or filecount
4336 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4337 support for a long time. Removed HAVE_SYS_VFS_H.
4339 18. Updated exipick to current release from John Jetmore.
4341 19. Allow an empty sender to be matched against a lookup in an address list.
4342 Previously the only cases considered were a regular expression, or an
4345 20. Exim went into a mad DNS lookup loop when doing a callout where the
4346 host was specified on the transport, if the DNS lookup yielded more than
4349 21. The RFC2047 encoding function was originally intended for short strings
4350 such as real names; it was not keeping to the 75-character limit for
4351 encoded words that the RFC imposes. It now respects the limit, and
4352 generates multiple encoded words if necessary. To be on the safe side, I
4353 have increased the buffer size for the ${rfc2047: expansion operator from
4356 22. Failure to deliver a bounce message always caused it to be frozen, even if
4357 there was an errors_to setting on the router. The errors_to setting is now
4360 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4361 canonical form (fully expanded) before being placed in
4362 $sender_host_address.
4364 24. Updated eximstats to version 1.33
4366 25. Include certificate and key file names in error message when GnuTLS fails
4367 to set them up, because the GnuTLS error message doesn't include the name
4368 of the failing file when there is a problem reading it.
4370 26. Expand error message when OpenSSL has problems setting up cert/key files.
4373 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
4374 (this can affect the format of dates).
4376 28. exim_tidydb, when checking for the continued existence of a message for
4377 which it has found a message-specific retry record, was not finding
4378 messages that were in split spool directories. Consequently, it was
4379 deleting retry records that should have stayed in existence.
4381 29. eximstats updated to version 1.35
4382 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
4383 1.35 - bugfix such that pie charts by volume are generated correctly
4385 30. The SPA authentication driver was not abandoning authentication and moving
4386 on to the next authenticator when an expansion was forced to fail,
4387 contradicting the general specification for all authenticators. Instead it
4388 was generating a temporary error. It now behaves as specified.
4390 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
4391 (the order specifies the preference for clients). The order is now AES256,
4392 AES128, 3DES, ARCFOUR128.
4394 31. Small patch to Sieve code - explicitly set From: when generating an
4397 32. Exim crashed if a remote delivery caused a very long error message to be
4398 recorded - for instance if somebody sent an entire SpamAssassin report back
4399 as a large number of 550 error lines. This bug was coincidentally fixed by
4400 increasing the size of one of Exim's internal buffers (big_buffer) that
4401 happened as part of the Exiscan merge. However, to be on the safe side, I
4402 have made the code more robust (and fixed the comments that describe what
4405 33. Some experimental protocols are using DNS PTR records for new purposes. The
4406 keys for these records are domain names, not reversed IP addresses. The
4407 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4408 leaves it alone. Component reversal etc. now happens only for IP addresses.
4411 34. The host_aton() function is supposed to be passed a string that is known
4412 to be a valid IP address. However, in the case of IPv6 addresses, it was
4413 not checking this. This is a hostage to fortune. Exim now panics and dies
4414 if the condition is not met. A case was found where this could be provoked
4415 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4416 components; fortuitously, this particular loophole had already been fixed
4417 by change 4.50/55 or 4.44/33 above.
4419 If there are any other similar loopholes, the new check in host_aton()
4420 itself should stop them being exploited. The report I received stated that
4421 data on the command line could provoke the exploit when Exim was running as
4422 exim, but did not say which command line option was involved. All I could
4423 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4424 running as the user.
4427 35. There was a buffer overflow vulnerability in the SPA authentication code
4428 (which came originally from the Samba project). I have added a test to the
4429 spa_base64_to_bits() function which I hope fixes it.
4432 36. The daemon start-up calls getloadavg() while still root for those OS that
4433 need the first call to be done as root, but it missed one case: when
4434 deliver_queue_load_max is set with deliver_drop_privilege. This is
4435 necessary for the benefit of the queue runner, because there is no re-exec
4436 when deliver_drop_privilege is set.
4438 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4439 was in use, was not putting the data itself into the right store pool;
4440 consequently, it could be overwritten for a subsequent message in the same
4441 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4444 38. Sometimes the final signoff response after QUIT could fail to get
4445 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4446 < 0 before doing a fflush(). This bug looks as though it goes back to the
4447 introduction of TLS in release 3.20, but "sometimes" must have been rare
4448 because the tests only now provoked it.
4454 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
4455 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
4456 Now it returns an integer. A number of calls were still expecting a T/F
4457 return. Fortuitously, in all cases, the tests worked in OK situations,
4458 which is the norm. However, things would have gone wrong on any write
4459 failures on the smtp file descriptor. This function is used when sending
4460 messages over SMTP and also when doing verify callouts.
4462 2. When Exim is called to do synchronous delivery of a locally submitted
4463 message (the -odf or -odi options), it no longer closes stderr before doing
4466 3. Implemented the mua_wrapper option.
4468 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
4470 5. Implemented the functions header_remove(), header_testname(),
4471 header_add_at_position(), and receive_remove_recipient(), and exported them
4474 6. If an ACL "warn" statement specified the addition of headers, Exim already
4475 inserted X-ACL-Warn: at the start if there was no header name. However, it
4476 was not making this test for the second and subsequent header lines if
4477 there were newlines in the string. This meant that an invalid header could
4478 be inserted if Exim was badly configured.
4480 7. Allow an ACL "warn" statement to add header lines at the start or after all
4481 the Received: headers, as well as at the end.
4483 8. Added the rcpt_4xx retry error code.
4485 9. Added postmaster_mailfrom=xxx to callout verification option.
4487 10. Added mailfrom=xxxx to the callout verification option, for verify=
4490 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
4491 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
4493 12. Inserted some casts to stop certain compilers warning when using pointer
4494 differences as field lengths or precisions in printf-type calls (mostly
4495 affecting debugging statements).
4497 13. Added optional readline() support for -be (dynamically loaded).
4499 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
4500 same clock tick as a message's arrival, so that its received time was the
4501 same as the "first fail" time on the retry record, and that message
4502 remained on the queue past the ultimate address timeout, every queue runner
4503 would try a delivery (because it was past the ultimate address timeout) but
4504 after another failure, the ultimate address timeout, which should have then
4505 bounced the address, did not kick in. This was a "< instead of <=" error;
4506 in most cases the first failure would have been in the next clock tick
4507 after the received time, and all would be well.
4509 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
4510 being recognized when the domain list was tested by the match_domain
4511 condition in an expansion string.
4513 16. Added the ${str2b64: operator.
4515 17. Exim was always calling setrlimit() to set a large limit for the number of
4516 processes, without checking whether the existing limit was already
4517 adequate. (It did check for the limit on file descriptors.) Furthermore,
4518 errors from getrlimit() and setrlimit() were being ignored. Now they are
4519 logged to the main and panic logs, but Exim does carry on, to try to do its
4520 job under whatever limits there are.
4522 18. Imported PCRE 5.0.
4524 19. Trivial typo in log message " temporarily refused connection" (the leading
4527 20. If the log selector return_path_on_delivery was set and an address was
4528 redirected to /dev/null, the delivery process crashed because it assumed
4529 that a return path would always be set for a "successful" delivery. In this
4530 case, the whole delivery is bypassed as an optimization, and therefore no
4533 21. Internal re-arrangement: the function for sending a challenge and reading
4534 a response while authentication was assuming a zero-terminated challenge
4535 string. It's now changed to take a pointer and a length, to allow for
4536 binary data in such strings.
4538 22. Added the cyrus_sasl authenticator (code supplied by MBM).
4540 23. Exim was not respecting finduser_retries when seeking the login of the
4541 uid under which it was called; it was always trying 10 times. (The default
4542 setting of finduser_retries is zero.) Also, it was sleeping after the final
4543 failure, which is pointless.
4545 24. Implemented tls_on_connect_ports.
4547 25. Implemented acl_smtp_predata.
4549 26. If the domain in control=submission is set empty, Exim assumes that the
4550 authenticated id is a complete email address when it generates From: or
4551 Sender: header lines.
4553 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
4554 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
4555 chown and chgrp in /bin and hostname in /usr/bin.
4557 28. Exim was keeping the "process log" file open after each use, just as it
4558 does for the main log. This opens the possibility of it remaining open for
4559 long periods when the USR1 signal hits a daemon. Occasional processlog
4560 errors were reported, that could have been caused by this. Anyway, it seems
4561 much more sensible not to leave this file open at all, so that is what now
4564 29. The long-running daemon process does not normally write to the log once it
4565 has entered its main loop, and it closes the log before doing so. This is
4566 so that log files can straightforwardly be renamed and moved. However,
4567 there are a couple of unusual error situations where the daemon does write
4568 log entries, and I had neglected to close the log afterwards.
4570 30. The text of an SMTP error response that was received during a remote
4571 delivery was being truncated at 512 bytes. This is too short for some of
4572 the long messages that one sometimes sees. I've increased the limit to
4575 31. It is now possible to make retry rules that apply only when a message has a
4576 specific sender, in particular, an empty sender.
4578 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
4579 it possible to be selective about when SMTP synchronization is enforced.
4581 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
4583 32. Implemented hosts_connection_nolog.
4585 33. Added an ACL for QUIT.
4587 34. Setting "delay_warning=" to disable warnings was not working; it gave a
4590 35. Added mailbox_size and mailbox_filecount to appendfile.
4592 36. Added control = no_multiline_responses to ACLs.
4594 37. There was a bug in the logic of the code that waits for the clock to tick
4595 in the case where the clock went backwards by a substantial amount such
4596 that the microsecond fraction of "now" was more than the microsecond
4597 fraction of "then" (but the whole seconds number was less).
4599 38. Added support for the libradius Radius client library this is found on
4600 FreeBSD (previously only the radiusclient library was supported).
4606 1. When certain lookups returned multiple values in the form name=value, the
4607 quoting of the values was not always being done properly. Specifically:
4608 (a) If the value started with a double quote, but contained no whitespace,
4610 (b) If the value contained whitespace other than a space character (i.e.
4611 tabs or newlines or carriage returns) it was not quoted.
4612 This fix has been applied to the mysql and pgsql lookups by writing a
4613 separate quoting function and calling it from the lookup code. The fix
4614 should probably also be applied to nisplus, ibase and oracle lookups, but
4615 since I cannot test any of those, I have not disturbed their existing code.
4617 2. A hit in the callout cache for a specific address caused a log line with no
4618 reason for rejecting RCPT. Now it says "Previous (cached) callout
4619 verification failure".
4621 3. There was an off-by-one bug in the queryprogram router. An over-long
4622 return line was truncated at 256 instead of 255 characters, thereby
4623 overflowing its buffer with the terminating zero. As well as fixing this, I
4624 have increased the buffer size to 1024 (and made a note to document this).
4626 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
4627 when Exim is waiting for an SMTP response from a remote server, Exim
4628 restarts its select() call on the socket, thereby resetting its timeout.
4629 This is not a problem when such interrupts are rare. Somebody set up a cron
4630 job to run exiwhat every 2 minutes, which is less than the normal select()
4631 timeout (5 or 10 minutes). This meant that the select() timeout never
4632 kicked in because it was always reset. I have fixed this by comparing the
4633 time when an interrupt arrives with the time at the start of the first call
4634 to select(). If more time than the timeout has elapsed, the interrupt is
4635 treated as a timeout.
4637 5. Some internal re-factoring in preparation for the addition of Sieve
4638 extensions (by MH). In particular, the "personal" test is moved to a
4639 separate function, and given an option for scanning Cc: and Bcc: (which is
4640 not set for Exim filters).
4642 6. When Exim created an email address using the login of the caller as the
4643 local part (e.g. when creating a From: or Sender: header line), it was not
4644 quoting the local part when it contained special characters such as @.
4646 7. Installed new OpenBSD configuration files.
4648 8. Reworded some messages for syntax errors in "and" and "or" conditions to
4649 try to make them clearer.
4651 9. Callout options, other than the timeout value, were being ignored when
4652 verifying sender addresses in header lines. For example, when using
4654 verify = header_sender/callout=no_cache
4656 the cache was (incorrectly) being used.
4658 10. Added a missing instance of ${EXE} to the exim_install script; this affects
4659 only the Cygwin environment.
4661 11. When return_path_on_delivery was set as a log selector, if different remote
4662 addresses in the same message used different return paths and parallel
4663 remote delivery occurred, the wrong values would sometimes be logged.
4664 (Whenever a remote delivery process finished, the return path value from
4665 the most recently started remote delivery process was logged.)
4667 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
4668 lines when AUTH and/or TLS are in use. This is the "received protocol"
4669 field. Exim used to use "asmtp" for authenticated SMTP, without any
4670 indication (in the protocol name) for TLS use. Now it follows the RFC and
4671 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
4672 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
4673 names appear in log lines as well as in Received: header lines.
4675 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
4676 extensions, and comparison tests, and to fix some bugs.
4678 14. Changes to the "personal" filter test:
4680 (1) The test was buggy in that it was just doing the equivalent of
4681 "contains" tests on header lines. For example, if a user's address was
4682 anne@some.where, the "personal" test would incorrectly be true for
4684 To: susanne@some.where
4686 This test is now done by extracting each address from the header in turn,
4687 and checking the entire address. Other tests that are part of "personal"
4688 are now done using regular expressions (for example, to check local parts
4689 of addresses in From: header lines).
4691 (2) The list of non-personal local parts in From: addresses has been
4692 extended to include "listserv", "majordomo", "*-request", and "owner-*",
4693 taken from the Sieve specification recommendations.
4695 (3) If the message contains any header line starting with "List-" it is
4696 treated as non-personal.
4698 (4) The test for "circular" in the Subject: header line has been removed
4699 because it now seems ill-conceived.
4701 15. Minor typos in src/EDITME comments corrected.
4703 16. Installed latest exipick from John Jetmore.
4705 17. If headers_add on a router specified a text string that was too long for
4706 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
4707 of string_sprintf() is now avoided.
4709 18. $message_body_size was not set (it was always zero) when running the DATA
4710 ACL and the local_scan() function.
4712 19. For the "mail" command in an Exim filter, no default was being set for
4713 the once_repeat time, causing a random time value to be used if "once" was
4714 specified. (If the value happened to be <= 0, no repeat happened.) The
4715 default is now 0s, meaning "never repeat". The "vacation" command was OK
4716 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
4717 (I found it when inspecting the code).
4719 20. There is now an overall timeout for performing a callout verification. It
4720 defaults to 4 times the callout timeout, which applies to individual SMTP
4721 commands during the callout. The overall timeout applies when there is more
4722 than one host that can be tried. The timeout is checked before trying the
4723 next host. This prevents very long delays if there are a large number of
4724 hosts and all are timing out (e.g. when the network connections are timing
4725 out). The value of the overall timeout can be changed by specifying an
4726 additional sub-option for "callout", called "maxwait". For example:
4728 verify = sender/callout=5s,maxwait=20s
4730 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
4731 to the end before writing, but this should make it even safer).
4733 22. Exim was forgetting that it had advertised PIPELINING for the second and
4734 subsequent messages on an SMTP connection. It was also not resetting its
4735 memory on STARTTLS and an internal HELO.
4737 23. When Exim logs an SMTP synchronization error within a session, it now
4738 records whether PIPELINING has been advertised or not.
4740 24. Added 3 instances of "(long int)" casts to time_t variables that were being
4741 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
4742 rather than long int.
4744 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
4746 26. Added the never_mail option to autoreply.
4752 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
4753 crash if the getsockname() call failed; this can happen if a connection is
4754 closed very soon after it is established. The problem was simply in the
4755 order in which certain operations were done, causing Exim to try to write
4756 to the SMTP stream before it had set up the file descriptor. The bug has
4757 been fixed by making things happen in the correct order.
4763 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
4764 before the connection was closed, thus losing the rejection response.
4766 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
4767 some early Solaris releases, but causes trouble in current releases where
4768 socklen_t is defined.
4770 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
4773 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
4776 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
4777 that releasing the top of it at the end releases what was used for sub-
4778 expansions (unless the block got too big). However, discard this block if
4779 the first thing is a variable or header, so that we can use its block when
4780 it is dynamic (useful for very large $message_headers, for example).
4782 6. Lookups now cache *every* query, not just the most recent. A new, separate
4783 store pool is used for this. It can be recovered when all lookup caches are
4784 flushed. Lookups now release memory at the end of their result strings.
4785 This has involved some general refactoring of the lookup sources.
4787 7. Some code has been added to the store_xxx() functions to reduce the amount
4788 of flapping under certain conditions.
4790 8. log_incoming_interface used to affect only the <= reception log lines. Now
4791 it causes the local interface and port to be added to several more SMTP log
4792 lines, for example "SMTP connection from", and rejection lines.
4794 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
4796 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
4798 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
4799 could be overwritten at the end of the current message (or the start of a
4800 new message if it was set in a HELO ACL). The value is now preserved for
4801 the duration of the SMTP connection.
4803 12. If a transport had a headers_rewrite setting, and a matching header line
4804 contained an unqualified address, that address was qualified, even if it
4805 did not match any rewriting rules. The underlying bug was that the values
4806 of the flags that permit the existence of unqualified sender and recipient
4807 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
4808 non-local messages, and by -bnq for local messages) were not being
4809 preserved with the message after it was received.
4811 13. When Exim was logging an SMTP synchronization error, it could sometimes log
4812 "next input=" as part of the text comprising the host identity instead of
4813 the correct text. The code was using the same buffer for two different
4814 strings. However, depending on which order the printing function evaluated
4815 its arguments, the bug did not always show up. Under Linux, for example, my
4816 test suite worked just fine.
4818 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
4819 doesn't work with some older versions of Perl. It has been changed to "my",
4820 which in any case is probably the better facility to use.
4822 15. A really picky compiler found some instances of statements for creating
4823 error messages that either had too many or two few arguments for the format
4826 16. The size of the buffer for calls to the DNS resolver has been increased
4827 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
4828 for addresses that have a lot of PTR records. This alleviates a problem; it
4829 does not fully solve it.
4831 17. A dnsdb lookup for PTR records that receives more data than will fit in the
4832 buffer now truncates the list and logs the incident, which is the same
4833 action as happens when Exim is looking up a host name and its aliases.
4834 Previously in this situation something unpredictable would happen;
4835 sometimes it was "internal error: store_reset failed".
4837 18. If a server dropped the connection unexpectedly when an Exim client was
4838 using GnuTLS and trying to read a response, the client delivery process
4839 crashed while trying to generate an error log message.
4841 19. If a "warn" verb in an ACL added multiple headers to a message in a single
4842 string, for example:
4844 warn message = H1: something\nH2: something
4846 the text was added as a single header line from Exim's point of view
4847 though it ended up OK in the delivered message. However, searching for the
4848 second and subsequent header lines using $h_h2: did not work. This has been
4849 fixed. Similarly, if a system filter added multiple headers in this way,
4850 the routers could not see them.
4852 20. Expanded the error message when iplsearch is called with an invalid key to
4853 suggest using net-iplsearch in a host list.
4855 21. When running tests using -bh, any delays imposed by "delay" modifiers in
4856 ACLs are no longer actually imposed (and a message to that effect is
4859 22. If a "gecos" field in a passwd entry contained escaped characters, in
4860 particular, if it contained a \" sequence, Exim got it wrong when building
4861 a From: or a Sender: header from that name. A second bug also caused
4862 incorrect handling when an unquoted " was present following a character
4863 that needed quoting.
4865 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
4866 was not being matched caselessly.
4868 24. Arranged for all hyphens in the exim.8 source to be escaped with
4871 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
4872 properly. Recipient callout cache records were still being keyed to include
4873 the sender, even when use_sender was set false. This led to far more
4874 callouts that were necessary. The sender is no longer included in the key
4875 when use_sender is false.
4877 26. Added "control = submission" modifier to ACLs.
4879 27. Added the ${base62d: operator to decode base 62 numbers.
4881 28. dnsdb lookups can now access SRV records.
4883 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
4884 the configuration file.
4886 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
4887 (-v) mode. This makes the output for a verbose queue run more intelligible.
4889 31. Added a use_postmaster feature to recipient callouts.
4891 32. Added the $body_zerocount variable, containing the number of binary zero
4892 bytes in the message body.
4894 33. The time of last modification of the "new" subdirectory is now used as the
4895 "mailbox time last read" when there is a quota error for a maildir
4898 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
4900 35. Added +ignore_unknown as a special item in host lists.
4902 36. Code for decoding IPv6 addresses in host lists is now included, even if
4903 IPv6 support is not being compiled. This fixes a bug in which an IPv6
4904 address was recognized as an IP address, but was then not correctly decoded
4905 into binary, causing unexpected and incorrect effects when compared with
4912 1. Very minor rewording of debugging text in manualroute to say "list of
4913 hosts" instead of "hostlist".
4915 2. If verify=header_syntax was set, and a header line with an unqualified
4916 address (no domain) and a large number of spaces between the end of the
4917 name and the colon was received, the reception process suffered a buffer
4918 overflow, and (when I tested it) crashed. This was caused by some obsolete
4919 code that should have been removed. The fix is to remove it!
4921 3. When running in the test harness, delay a bit after writing a bounce
4922 message to get a bit more predictability in the log output.
4924 4. Added a call to search_tidyup() just before forking a reception process. In
4925 theory, someone could use a lookup in the expansion of smtp_accept_max_
4926 per_host which, without the tidyup, could leave open a database connection.
4928 5. Added the variables $recipient_data and $sender_data which get set from a
4929 lookup success in an ACL "recipients" or "senders" condition, or a router
4930 "senders" option, similar to $domain_data and $local_part_data.
4932 6. Moved the writing of debug_print from before to after the "senders" test
4935 7. Change 4.31/66 (moving the time when the Received: is generated) caused
4936 problems for message scanning, either using a data ACL, or using
4937 local_scan() because the Received: header was not generated till after they
4938 were called (in order to set the time as the time of reception completion).
4939 I have revised the way this works. The header is now generated after the
4940 body is received, but before the ACL or local_scan() are called. After they
4941 are run, the timestamp in the header is updated.
4947 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
4948 before starting a queue runner without re-exec. This happened only when
4949 deliver_drop_privilege was set or when the Exim user was set to root. The
4950 effect of the bug was that timeouts during subsequent deliveries caused
4951 crashes instead of being properly handled. The handler is now left at its
4952 default (and expected) setting.
4954 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
4955 message, again when deliver_drop_privilege is set or Exim is run as root.
4956 The bug described in (1) was not present in this case, but the tidying up
4957 of the other signals was missing. I have made the two cases consistent.
4959 3. The ignore_target_hosts setting on a manualroute router was being ignored
4960 for hosts that were looked up using the /MX notation.
4962 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
4965 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
4966 operated on the sender address. After changing the $sender_address to <>
4967 for the sender address verify, Exim was re-instated it as the original
4968 (before rewriting) address, but remembering that it had rewritten it, so it
4969 wasn't rewriting it again. This bug also had the effect of breaking the
4970 sender address verification caching when the sender address was rewritten.
4972 6. The ignore_target_hosts option was being ignored by the ipliteral router.
4973 This has been changed so that if the ip literal address matches
4974 ignore_target_hosts, the router declines.
4976 7. Added expansion conditions match_domain, match_address, and match_local_
4977 part (NOT match_host).
4979 8. The placeholder for the Received: header didn't have a length field set.
4981 9. Added code to Exim itself and to exim_lock to test for a specific race
4982 condition that could lead to file corruption when using MBX delivery. The
4983 issue is with the lockfile that is created in /tmp. If this file is removed
4984 after a process has opened it but before that process has acquired a lock,
4985 there is the potential for a second process to recreate the file and also
4986 acquire a lock. This could lead to two Exim processes writing to the file
4987 at the same time. The added code performs the same test as UW imapd; it
4988 checks after acquiring the lock that its file descriptor still refers to
4989 the same named file.
4991 10. The buffer for building added header lines was of fixed size, 8192 bytes.
4992 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
4995 11. Added the smtp_active_hostname option. If used, this will typically be made
4996 to depend on the incoming interface address. Because $interface_address is
4997 not set up until the daemon has forked a reception process, error responses
4998 that can happen earlier (such as "too many connections") no longer contain
5001 12. If an expansion in a condition on a "warn" statement fails because a lookup
5002 defers, the "warn" statement is abandoned, and the next ACL statement is
5003 processed. Previously this caused the whole ACL to be aborted.
5005 13. Added the iplsearch lookup type.
5007 14. Added ident_timeout as a log selector.
5009 15. Added tls_certificate_verified as a log selector.
5011 16. Added a global option tls_require_ciphers (compare the smtp transport
5012 option of the same name). This controls incoming TLS connections.
5014 17. I finally figured out how to make tls_require_ciphers do a similar thing
5015 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
5016 before starting the TLS session.
5018 18. Tabs are now shown as \t in -bP output.
5020 19. If the log selector return_path_on_delivery was set, Exim crashed when
5021 bouncing a message because it had too many Received: header lines.
5023 20. If two routers both had headers_remove settings, and the first one included
5024 a superfluous trailing colon, the final name in the first list and the
5025 first name in the second list were incorrectly joined into one item (with a
5026 colon in the middle).
5032 1. Added -C and -D options to the exinext utility, mainly to make it easier
5033 to include in the automated testing, but these could be helpful when
5034 multiple configurations are in use.
5036 2. The exinext utility was not formatting the output nicely when there was
5037 an alternate port involved in the retry record key, nor when there was a
5038 message id as well (for retries that were specific to a specific message
5039 and a specific host). It was also confused by IPv6 addresses, because of
5040 the additional colons they contain. I have fixed the IPv4 problem, and
5041 patched it up to do a reasonable job for IPv6.
5043 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
5044 delivery, the log line now contains "pipelined" if PIPELINING was used.
5046 4. An SMTP transport process used to panic and die if the bind() call to set
5047 an explicit outgoing interface failed. This has been changed; it is now
5048 treated in the same way as a connect() failure.
5050 5. A reference to $sender_host_name in the part of a conditional expansion
5051 that was being skipped was still causing a DNS lookup. This no longer
5054 6. The def: expansion condition was not recognizing references to header lines
5055 that used bh_ and bheader_.
5057 7. Added the _cache feature to named lists.
5059 8. The code for checking quota_filecount in the appendfile transport was
5060 allowing one more file than it should have been.
5062 9. For compatibility with Sendmail, the command line option
5070 and sets the incoming protocol and host name (for trusted callers). The
5071 host name and its colon can be omitted when only the protocol is to be set.
5072 Note the Exim already has two private options, -pd and -ps, that refer to
5073 embedded Perl. It is therefore impossible to set a protocol value of "d" or
5074 "s", but I don't think that's a major issue.
5076 10. A number of refactoring changes to the code, none of which should affect
5079 (a) The number of logging options was getting close to filling up the
5080 32-bit word that was used as a bit map. I have split them into two classes:
5081 those that are passed in the argument to log_write(), and those that are
5082 only ever tested independently outside of that function. These are now in
5083 separate 32-bit words, so there is plenty of room for expansion again.
5084 There is no change in the user interface or the logging behaviour.
5086 (b) When building, for example, log lines, the code previously used a
5087 macro that called string_cat() twice, in order to add two strings. This is
5088 not really sufficiently general. Furthermore, there was one instance where
5089 it was actually wrong because one of the argument was used twice, and in
5090 one call a function was used. (As it happened, calling the function twice
5091 did not affect the overall behaviour.) The macro has been replaced by a
5092 function that can join an arbitrary number of extra strings onto a growing
5095 (c) The code for expansion conditions now uses a table and a binary chop
5096 instead of a serial search (which was left over from when there were very
5097 few conditions). Also, it now recognizes conditions like "pam" even when
5098 the relevant support is not compiled in: a suitably worded error message is
5099 given if an attempt is made to use such a condition.
5101 11. Added ${time_interval:xxxxx}.
5103 12. A bug was causing one of the ddress fields not to be passed back correctly
5104 from remote delivery subprocesses. The field in question was not being
5105 subsequently used, so this caused to problems in practice.
5107 13. Added new log selectors queue_time and deliver_time.
5109 14. Might have fixed a bug in maildirsizefile handling that threw up
5110 "unexpected character" debug warnings, and recalculated the data
5111 unnecessarily. In any case, I expanded the warning message to give more
5114 15. Added the message "Restricted characters in address" to the statements in
5115 the default ACL that block characters like @ and % in local parts.
5117 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
5118 Three changes have been made:
5120 (a) There was a serious bug; a negative response to MAIL caused the whole
5121 recipient domain to be cached as invalid, thereby blocking all messages
5122 to all local parts at the same domain, from all senders. This bug has
5123 been fixed. The domain is no longer cached after a negative response to
5124 MAIL if the sender used is not empty.
5126 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
5129 (c) A new callout option, "use_sender" has been added for people who want
5130 the modified behaviour.
5136 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
5139 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
5140 indeed breaks things for older releases.
5142 3. Added additional logging to the case where there is a problem reading data
5143 from a filter that is running in a subprocess using a pipe, in order to
5144 try to track down a specific problem.
5146 4. Testing facility fudge: when running in the test harness and attempting
5147 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
5148 getting "No route to host". Convert this to a timeout.
5150 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
5153 6. Some OS don't have socklen_t but use size_t instead. This affects the
5154 fifth argument of getsockopt() amongst other things. This is now
5155 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
5156 can be set for individual OS. I have set it for SunOS5, OSF1, and
5157 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
5158 some earlier ones do not.
5160 7. Change 4.30/15 was not doing the test caselessly.
5162 8. The standard form for an IPv6 address literal was being rejected by address
5163 parsing in, for example, MAIL and RCPT commands. An example of this kind of
5164 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5165 this, as well as the form without the "IPv6" on the front (but only when
5166 address literals are enabled, of course).
5168 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
5170 10. Exim crashed if a message with an empty sender address specified by -f
5171 encountered a router with an errors_to setting. This could be provoked only
5172 by a command such as
5176 where an empty string was supplied; "<>" did not hit this bug.
5178 11. Installed PCRE release 4.5.
5180 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
5181 remained set. It is now erased.
5183 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
5184 times from message ids (which are base 36 rather than the normal 62).
5186 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
5187 were being counted as actual protocol errors, and logged if the log
5188 selector +smtp_protocol_error was set. One cannot be perfect in this test,
5189 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
5190 and DATA following a set of rejected RCPTs do not count as protocol errors.
5191 In other words, Exim assumes they were pipelined, though this may not
5192 actually be the case. Of course, in all cases the client gets an
5193 appropriate error code.
5195 15. If a lookup fails in an ACL condition, a message about the failure may
5196 be available; it is used if testing the ACL cannot continue, because most
5197 such messages specify what the cause of the deferral is. However, some
5198 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
5199 that caused an old message to be retained and used if a later statement
5200 caused a defer, replacing the real cause of the deferral.
5202 16. If an IP address had so many PTR records that the DNS lookup buffer
5203 was not large enough to hold them, Exim could crash while trying to process
5204 the truncated data. It now detects and logs this case.
5206 17. Further to 4.21/58, another change has been made: if (and only if) the
5207 first line of a message (the first header line) ends with CRLF, a bare LF
5208 in a subsequent header line has a space inserted after it, so as not to
5209 terminate the header.
5211 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
5212 unnecessarily, used atoi() instead of strtol(), and didn't check the
5213 termination when getting file sizes from file names by regex.
5215 19. Completely re-implemented the support for maildirsize files, in the light
5216 of a number of problems with the previous contributed implementation
5217 (4.30/29). In particular:
5219 . If the quota is zero, the maildirsize file is maintained, but no quota is
5222 . If the maildir directory does not exist, it is created before any attempt
5223 to write a maildirsize file.
5225 . The quota value in the file is just a cache; if the quota is changed in
5226 the transport, the new value overrides.
5228 . A regular expression is available for excluding directories from the
5231 20. The autoreply transport checks the characters in options that define the
5232 message's headers; it allows continued headers, but it was checking with
5233 isspace() after an embedded newline instead of explicitly looking for a
5236 21. If all the "regular" hosts to which an address was routed had passed their
5237 expiry times, and had not reached their retry times, the address was
5238 bounced, even if fallback hosts were defined. Now Exim should go on to try
5241 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
5242 equivalent code in the SMTP transport. Some hosts send humungous responses
5243 to HELO/EHLO, more than 1024 it seems.
5245 23. Refactoring: code in filter.c used (void *) for "any old type" but this
5246 gives compiler warnings in some environments. I've now done it "properly",
5249 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
5250 (because of problems with the built-in one) was declared to return uschar *
5251 instead of char *, causing compiler failure.
5253 25. Fixed a file descriptor leak when processing alias/forward files.
5255 26. Fixed a minor format string issue in dbfn.c.
5257 27. Typo in exim.c: ("dmbnz" for "dbmnz").
5259 28. If a filter file refered to $h_xxx or $message_headers, and the headers
5260 contained RFC 2047 "words", Exim's memory could, under certain conditions,
5263 29. When a sender address is verified, it is cached, to save repeating the test
5264 when there is more than one recipient in a message. However, when the
5265 verification involves a callout, it is possible for different callout
5266 options to be set for different recipients. It is too complicated to keep
5267 track of this in the cache, so now Exim always runs a verification when a
5268 callout is required, relying on the callout cache for the optimization.
5269 The overhead is duplication of the address routing, but this should not be
5272 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
5273 to be verified with callout=postmaster, and the main callout worked but the
5274 postmaster check failed, the verification correctly failed. However, if a
5275 subsequent RCPT command asked for sender verification *without* the
5276 postmaster check, incorrect caching caused this verification also to fail,
5279 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
5280 it was not caching the DNS options (qualify_single, search_parents) that
5281 were used when the lookup failed. A subsequent lookup with different
5282 options therefore always gave the same answer, though there were cases
5283 where it should not have. (Example: a "domains = !$mx_any" option on a
5284 dnslookup router: the "domains" option is always processed without any
5285 widening, but the router might have qualify_single set.) Now Exim uses the
5286 cached value only when the same options are set.
5288 32. Added John Jetmore's "exipick" utility to the distribution.
5290 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
5291 than a timeout (e.g. a certificate is required, and is not provided), an
5292 Exim server now closes the connection immediately. Previously it waited for
5293 the client to close - but if the client is SSL, it seems that they each
5294 wait for each other, leading to a delay before one of them times out.
5296 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
5297 maintained 0.8.x compatibility because I don't think many are using it, and
5298 it is clearly obsolete.
5300 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
5303 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
5304 client certificate was expired. A simple patch fixes this, though I don't
5305 understand the full logic of why the verify callback is called multiple
5308 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
5309 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
5310 which causes problems with some clients (such as the Certicom SSL Plus
5311 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
5312 disables the coutermeasure allowing Eudora to connect."
5314 38. Exim was not checking that a write() to a log file succeeded. This could
5315 lead to Bad Things if a log got too big, in particular if it hit a file
5316 size limit. Exim now panics and dies if it cannot write to a log file, just
5317 as it does if it cannot open a log file.
5319 39. Modified OS/Makefile-Linux so that it now contains
5321 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
5323 The two -D definitions ensure that Exim is compiled with large file
5324 support, which makes it possible to handle log files that are bigger than
5327 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
5328 instance) a domain was checked against a named list that involved a lookup,
5329 causing $domain_data to be set, then another domain was checked against the
5330 same list, then the first domain was re-checked, the value of $domain_data
5331 after the final check could be wrong. In particular, if the second check
5332 failed, it could be set empty. This bug probably also applied to
5335 41. The strip_trailing_dot option was not being applied to the address given
5336 with the -f command-line option.
5338 42. The code for reading a message's header from the spool was incrementing
5339 $received_count, but never initializing it. This meant that the value was
5340 incorrect (doubled) while delivering a message in the same process in which
5341 it was received. In the most common configuration of Exim, this never
5342 happens - a fresh exec is done - but it can happen when
5343 deliver_drop_privilege is set.
5345 43. When Exim logs an SMTP synchronization error - client data sent too soon -
5346 it now includes up to 150 characters of the unexpected data in the log
5349 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
5350 and building data strings. The size of both of these buffers was 10 000
5351 bytes - far larger than anybody would *ever* want, thought I. Needless to
5352 say, somebody hit the limit. I have increased the maximum line length to
5353 20 000 and the maximum data length of concatenated lines to 100 000. I have
5354 also fixed two bugs, because there was no checking on these buffers. Tsk,
5355 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
5356 buffer is too small.
5358 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
5359 lsearch lookups. Now it does.
5361 46. When parsing a route_list item in a manualroute router, a fixed-length
5362 buffer was used for the list of hosts. I made this 1024 bytes long,
5363 thinking that nobody would ever have a list of hosts that long. Wrong.
5364 Somebody had a whole pile of complicated expansion conditions, and the
5365 string was silently truncated, leading to an expansion error. It turns out
5366 that it is easier to change to an unlimited length (owing to other changes
5367 that have happened since this code was originally written) than to build
5368 structure for giving a limitation error. The length of the item that
5369 expands into the list of hosts is now unlimited.
5371 47. The lsearch lookup could not handle data where the length of text line was
5372 more than 4095 characters. Such lines were truncated, leading to shortened
5373 data being returned. It should now handle lines of any length.
5375 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
5376 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
5379 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
5380 replacing strings such as BIN_DIRECTORY in a source file: the replacement
5381 no longer happens in comment lines. A list of replacements is now placed
5382 at the head of all of the source files, except those whose only change is
5383 to replace PERL_COMMAND in the very first #! line.
5385 50. Replaced the slow insertion sort in queue.c, for sorting the list of
5386 messages on the queue, with a bottom-up merge sort, using code contributed
5387 by Michael Haardt. This should make operations like -bp somewhat faster on
5388 large queues. It won't affect queue runners, except when queue_run_in_order
5391 51. Installed eximstats 1.31 in the distribution.
5393 52. Added support for SRV lookups to the dnslookup router.
5395 53. If an ACL referred to $message_body or $message_body_end, the value was not
5396 reset for any messages that followed in the same SMTP session.
5398 54. The store-handling optimization for building very long strings was not
5399 differentiating between the different store pools. I don't think this
5400 actually made any difference in practice, but I've tidied it.
5402 55. While running the routers to verify a sender address, $sender_address
5403 was still set to the sender address. This is wrong, because when routing to
5404 send a bounce to the sender, it would be empty. Therefore, I have changed
5405 it so that, while verifying a sender address, $sender_address is set to <>.
5406 (There is no change to what happens when verifying a recipient address.)
5408 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
5409 target A or AAAA records (if not already returned) without resetting the
5410 qualify_single or search_parents options of the DNS resolver. These are
5411 inappropriate in this case because the targets of MX and SRV records must
5412 be FQDNs. A broken DNS record could cause trouble if it happened to have a
5413 target that, when qualified, matched something in the local domain. These
5414 two options are now turned off when doing these lookups.
5416 57. It seems that at least some releases of Reiserfs (which does not have the
5417 concept of a fixed number of inodes) returns zero and not -1 for the
5418 number of available inodes. This interacted badly with check_spool_inodes,
5419 which assumed that -1 was the "no such thing" setting. What I have done is
5420 to check that the total number of inodes is greater than zero before doing
5421 the test of how many are available.
5423 58. When a "warn" ACL statement has a log_message modifier, the message is
5424 remembered, and not repeated. This is to avoid a lot of repetition when a
5425 message has many recipients that cause the same warning to be written.
5426 Howewer, Exim was preserving the list of already written lines for an
5427 entire SMTP session, which doesn't seem right. The memory is now reset if a
5428 new message is started.
5430 59. The "rewrite" debugging flag was not showing the result of rewriting in the
5431 debugging output unless log_rewrite was also set.
5433 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
5434 of (int)(handle) when we know that handle contains (void *)(-1).
5436 61. The Exim daemon panic-logs an error return when it closes the incoming
5437 connection. However "connection reset by peer" seems to be common, and
5438 isn't really an error worthy of noting specially, so that particular error
5441 62. When Exim is trying to find all the local interfaces, it used to panic and
5442 die if the ioctl to get the interface flags failed. However, it seems that
5443 on at least one OS (Solaris 9) it is possible to have an interface that is
5444 included in the list of interfaces, but for which you get a failure error
5445 for this call. This happens when the interface is not "plumbed" into a
5446 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
5447 failure of the "get flags" call assumes that the interface is down.
5449 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
5450 makes life easier for people who are doing arithmetic on fields extracted
5451 from dates, where you often get leading zeros that should not be
5452 interpreted as octal.
5454 64. Added qualify_domain to the redirect router, to override the global
5457 65. If a pathologically long header line contained very many addresses (the
5458 report of this problem mentioned 10 000) and each of them was rewritten,
5459 Exim could use up a very large amount of memory. (It kept on making new
5460 copies of the header line as it rewrote, and never released the old ones.)
5461 At the expense of a bit more processing, the header rewriting function has
5462 been changed so that it no longer eats memory in this way.
5464 66. The generation of the Received: header has been moved from the time that a
5465 message starts to be received, to the time that it finishes. The timestamp
5466 in the Received: header should now be very close to that of the <= log
5467 line. There are two side-effects of this change:
5469 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
5470 logged header lines no longer include the local Received: line, because
5471 it has not yet been created. The same applies to a copy of the message
5472 that is returned to a non-SMTP sender when a message is rejected.
5474 (b) When a filter file is tested using -bf, no additional Received: header
5475 is added to the test message. After some thought, I decided that this
5478 This change does not affect the value of $received_for. It is still set
5479 after address rewriting, but before local_scan() is called.
5481 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
5483 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
5484 gave an unhelpful panic error message, and a defer error. I have managed to
5485 change this behaviour so that it now rejects any supplied certificate,
5486 which seems right, as the list of acceptable certificates is empty.
5488 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
5489 gave an unhelpful defer error. I have not managed to make this reject any
5490 supplied certificates, but the error message it gives is "no certificate
5491 supplied", which is not helpful.
5493 70. exigrep's output now also includes lines that are not associated with any
5494 message, but which match the given pattern. Implemented by a patch from
5495 Martin Sluka, which also tidied up the Perl a bit.
5497 71. Recipient callout verification, like sender verification, was using <> in
5498 the MAIL FROM command. This isn't really the right thing, since the actual
5499 sender may affect whether the remote host accepts the recipient or not. I
5500 have changed it to use the actual sender in the callout; this means that
5501 the cache record is now keyed on a recipient/sender pair, not just the
5502 recipient address. There doesn't seem to be a real danger of callout loops,
5503 since a callout by the remote host to check the sender would use <>.
5504 [SEE ABOVE: changed after hitting problems.]
5506 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
5507 temporary errors. However, in the case of such a code being given after
5508 the end of a data transmission (i.e. after ".") Exim was failing to write
5509 a retry record for the message. (Yes, there was some broken host that was
5510 actually sending 8xx at this point.)
5512 73. An unknown lookup type in a host list could cause Exim to panic-die when
5513 the list was checked. (An example that provoked this was putting <; in the
5514 middle of a list instead of at the start.) If this happened during a DATA
5515 ACL check, a -D file could be left lying around. This kind of configuration
5516 error no longer causes Exim to die; instead it causes a defer errror. The
5517 incident is still logged to the main and panic logs.
5519 74. Buglet left over from Exim 3 conversion. The message "too many messages
5520 in one connection" was written to the rejectlog but not the mainlog, except
5521 when address rewriting (yes!) was being logged.
5523 75. Added write_rejectlog option.
5525 76. When a system filter was run not as root (that is, when system_filter_user
5526 was set), the values of the $n variables were not being returned to the
5527 main process; thus, they were not subsequently available in the $sn
5530 77. Added +return_path_on_delivery log selector.
5532 78. A connection timeout was being treated differently from recipients deferred
5533 when testing hosts_max_try with a message that was older than the host's
5534 retry timeout. (The host should not be counted, thus allowing all hosts to
5535 be tried at least once before bouncing.) This may have been the cause of an
5536 occasionally reported bug whereby a message would remain on the queue
5537 longer than the retry timeout, but would be bounced if a delivery was
5538 forced. I say "may" because I never totally pinned down the problem;
5539 setting up timeout/retry tests is difficult. See also the next item.
5541 79. The ultimate address timeout was not being applied to errors that involved
5542 a combination of host plus message (for example, a timeout on a MAIL
5543 command). When an address resolved to a number of possible hosts, and they
5544 were not all tried for each delivery (e.g. because of hosts_max_try), a
5545 message could remain on the queue longer than the retry timeout.
5547 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
5550 81. Fixed an obscure SMTP outgoing bug which required at least the following
5551 conditions: (a) there was another message waiting for the same server;
5552 (b) the server returned 5xx to all RCPT commands in the first message so
5553 that the message was not completed; (c) the server dropped the connection
5554 or gave a negative response to the RSET that Exim sends to abort the
5555 transaction. The observed case was a dropped connection after DATA that had
5556 been sent in pipelining mode. That is, the server had advertised PIPELINING
5557 but was not implementing it correctly. The effect of the bug was incorrect
5558 behaviour, such as trying another host, and this could lead to a crash.
5564 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
5565 and daemon.c were passed as pointers to ints; they should have been
5566 pointers to socklen_t variables (which are typically unsigned ints).
5568 2. Some signed/unsigned type warnings in the os.c file for Linux have been
5571 3. Fixed a really odd bug that affected only the testing scheme; patching a
5572 certain fixed string in the binary changed the value of another string that
5573 happened to be identical to the end of the original first string.
5575 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
5576 name", it returns that address as the IP address. On some operating
5577 systems (e.g. Solaris), it also passes back the IP address string as the
5578 "host name". However, on others (e.g. Linux), it passes back an empty
5579 string. Exim wasn't checking for this, and was changing the host name to an
5580 empty string, assuming it had been canonicized.
5582 5. Although rare, it is permitted to have more than one PTR record for a given
5583 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
5584 all the names associated with an address, because they do in Solaris.
5585 However, it seems that they do not in Linux for data that comes from the
5586 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
5587 I found this out when I moved to a new Linux workstation and tried to run
5588 the Exim test suite.
5590 To get round this problem I have changed the code so that it now does its
5591 own call to the DNS to look up PTR records when searching for a host name.
5592 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
5593 addresses that are only in /etc/hosts are still found.
5595 This behaviour is, however, controlled by an option called host_lookup_
5596 order, which defaults to "bydns:byaddr". If people want to use the other
5597 order, or indeed, just use one or the other means of lookup, they can
5598 specify it in this variable.
5600 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
5601 some operating systems, this comes back from gethostbyaddr() as an empty
5602 string, and this is what Exim used to test for. However, it seems that in
5603 other systems, "." is yielded. Exim now tests for this case too.
5605 7. The values of check_spool_space and check_log_space are now held internally
5606 as a number of kilobytes instead of an absolute number of bytes. If a
5607 numbers is specified without 'K' or 'M', it is rounded up to the nearest
5608 kilobyte. This means that much larger values can be stored.
5610 8. Exim monitor: an attempt to get the action menu when not actually pointing
5611 at a message produces an empty menu entitled "No message selected". This
5612 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
5613 no entries in it ("Shell widget menu has zero width and/or height"). So I
5614 have added a single, blank menu entry in this case.
5616 9. Added ${quote_local_part.
5618 10. MIME decoding is now applied to the contents of Subject: header lines when
5621 11. Now that a reference to $sender_host_address automatically causes a reverse
5622 lookup to occur if necessary (4.13/18), there is no need to arrange for a
5623 host lookup before query-style lookups in lists that might use this
5624 variable. This has therefore been abolished, and the "net-" prefix is no
5625 longer necessary for query-style lookups.
5627 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
5628 have been a typo for LFLAGS, so it has been changed.
5630 13. The install script calls Exim with "-C /dev/null" in order to find the
5631 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
5632 to be output. Howeve, since Exim outputs its version number before the
5633 error, it didn't break the script. It just looked ugly. I fixed this by
5634 always allowing "-C /dev/null" if the caller is root.
5636 14. Ignore overlarge ACL variable number when reading spool file - insurance
5637 against a later release with more variables having written the file.
5639 15. The standard form for an IPv6 address literal was being rejected by EHLO.
5640 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5641 this, as well as the form without the "IPv6" on the front.
5643 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
5644 OS/Makefile-Darwin file.
5646 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
5647 only to LDAP libraries that do not have LDAP_OPT_DEREF.
5649 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
5650 variable. However, some OS use ints for inodes. I've added cast to long int
5651 to get rid of the compiler warning.
5653 19. I had forgotten to lock out "/../" in configuration file names when
5654 ALT_CONFIG_PREFIX was set.
5656 20. Routers used for verification do not need to specify transports. However,
5657 if such a router generated a host list, and callout was configured, Exim
5658 crashed, because it could not find a port number from the (non-existent)
5659 transport. It now assumes port 25 in this circumstance.
5661 21. Added the -t option to exigrep.
5663 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
5664 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
5665 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
5666 them is set, LOOKUP_LSEARCH is forced.
5668 23. "exim -bV" now outputs a list of lookups that are included in the binary.
5670 24. Added sender and host information to the "rejected by local_scan()" log
5671 line; previously there was no indication of these.
5673 25. Added .include_if_exists.
5675 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
5676 Linux. It turns out that not all file systems support this. Apparently some
5677 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
5678 do it.) To cope with this, the error EINVAL, which means that sync-ing is
5679 not supported on the file descriptor, is now ignored when Exim is trying to
5680 sync a directory. This applies only to Linux.
5682 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
5684 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
5685 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
5688 29. Added support for maildirsize files from supplied patch (modified a bit).
5690 30. The use of :fail: followed by an empty string could lead Exim to respond to
5691 sender verification failures with (e.g.):
5693 550 Verification failed for <xxx>
5694 550 Sender verify failed
5696 where the first response line was missing the '-' that indicates it is not
5697 the final line of the response.
5699 31. The loop for finding the name of the user that called Exim had a hardwired
5700 limit of 10; it now uses the value of finduser_retries, which is used for
5701 all other user lookups.
5703 32. Added $received_count variable, available in data and not_smtp ACLs, and at
5706 33. Exim was neglecting to zero errno before one call of strtol() when
5707 expanding a string and expecting an integer value. On some systems this
5708 resulted in spurious "integer overflow" errors. Also, it was casting the
5709 result into an int without checking.
5711 34. Testing for a connection timeout using "timeout_connect" in the retry rules
5712 did not work. The code looks as if it has *never* worked, though it appears
5713 to have been documented since at least releast 1.62. I have made it work.
5715 35. The "timeout_DNS" error in retry rules, also documented since at least
5716 1.62, also never worked. As it isn't clear exactly what this means, and
5717 clearly it isn't a major issue, I have abolished the feature by treating it
5718 as "timeout", and writing a warning to the main and panic logs.
5720 36. The display of retry rules for -brt wasn't always showing the error code
5723 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
5724 timeout_connect_A, timeout_connect_MX.
5726 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
5727 to the empty sender.
5729 39. The daemon was not analysing the content of -oX till after it had closed
5730 stderr and disconnected from the controlling terminal. This meant that any
5731 syntax errors were only noted on the panic log, and the return code from
5732 the command was 0. By re-arranging the code a little, I've made the
5733 decoding happen first, so such errors now appear on stderr, and the return
5734 code is 1. However, the actual setting up of the sockets still happens in
5735 the disconnected process, so errors there are still only recorded on the
5738 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
5739 connections (as happens on some IP stacks) was logged at start up time as
5740 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
5741 it from "IPv6 and IPv4", which means that two separate sockets are being
5744 41. The debug output for gethostbyname2() or getipnodebyname() failures now
5745 says whether AF_INET or AF_INET6 was passed as an argument.
5747 42. Exiwhat output was messed up when time zones were included in log
5750 43. Exiwhat now gives more information about the daemon's listening ports,
5751 and whether -tls-on-connect was used.
5753 44. The "port" option of the smtp transport is now expanded.
5755 45. A "message" modifier in a "warn" statement in a non-message ACL was being
5756 silently ignored. Now an error message is written to the main and panic
5759 46. There's a new ACL modifier called "logwrite" which writes to a log file
5760 as soon as it is encountered.
5762 47. Added $local_user_uid and $local_user_gid at routing time.
5764 48. Exim crashed when trying to verify a sender address that was being
5767 49. Exim was recognizing only a space character after ".include". It now also
5768 recognizes a tab character.
5770 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
5771 extracting the relevant information from the specification. The man page no
5772 longer contains scrambled data for the -d option, and I've added a section
5773 at the front about calling Exim under different names.
5775 51. Added "extra_headers" argument to the "mail" command in filter files.
5777 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
5780 53. Installed eximstats 1.29.
5782 54. Added transport_filter_timeout as a generic transport option.
5784 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
5785 Cc: header lines. This was required by RFC 822, but it not required by RFC
5788 56. Exim used to add From:, Date:, and Message-Id: header lines to any
5789 incoming messages that did not have them. Now it does so only if the
5790 message originates locally, that is, if there is no associated remote host
5791 address. When Resent- header lines are present, this applies to the Resent-
5792 lines rather than the non-Resent- lines.
5794 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
5795 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
5797 58. Messages for configuration errors now include the name of the main
5798 configuration file - useful now that there may be more than one file in a
5799 list (.included file names were always shown).
5801 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
5802 for those rare installations that do not start the daemon as root or run it
5803 setuid root. I've cut out the call to initgroups() if the daemon is not
5806 60. The Exim user and group can now be bound into the binary as text strings
5807 that are looked up at the start of Exim's processing.
5809 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
5811 62. Added $mailstore_basename variable.
5813 63. Installed patch to sieve.c from Michael Haardt.
5815 64. When Exim failed to open the panic log after failing to open the main log,
5816 the original message it was trying to log was written to stderr and debug
5817 output, but if they were not available (the usual case in production), it
5818 was lost. Now it is written to syslog before the two lines that record the
5819 failures to open the logs.
5821 65. Users' Exim filters run in subprocesses under the user's uid. It is
5822 possible for a "deliver" command or an alias in a "personal" command to
5823 provoke an address rewrite. If logging of address rewriting is configured,
5824 this fails because the process is not running as root or exim. There may be
5825 a better way of dealing with this, but for the moment (because 4.30 needs
5826 to be released), I have disabled address rewrite logging when running a
5827 filter in a non-root, non-exim process.
5833 1. The buildconfig auxiliary program wasn't quoting the value set for
5834 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
5835 not defined. This bug was masked in 4.22 by the effect that was fixed in
5838 2. Some messages that were rejected after a message id was allocated were
5839 shown as "incomplete" by exigrep. It no longer does this for messages that
5840 are rejected by local_scan() or the DATA or non-SMTP ACLs.
5842 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
5843 have allow_domain_literals set, the ID did not get logged in the <= line.
5844 Domain literals are now always recognized in Message-ID: header lines.
5846 4. The first argument for a ${extract expansion item is the key name or field
5847 number. Leading and trailing spaces in this item were not being ignored,
5848 causing some misleading effects.
5850 5. When deliver_drop_privilege was set, single queue runner processes started
5851 manually (i.e. by the command "exim -q") or by the daemon (which uses the
5852 same command in the process it spins off) were not dropping privilege.
5854 6. When the daemon running as "exim" started a queue runner, it always
5855 re-executed Exim in the spun-off process. This is a waste of effort when
5856 deliver_drop_privilege is set. The new process now just calls the
5857 queue-runner function directly.
5863 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
5866 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
5867 ignored. Though the use of -oP was forcing the writing of a pid file, it
5868 was always written to the default place.
5870 3. If the message "no IP address found for host xxxx" is generated during
5871 incoming verification, it is now followed by identification of the incoming
5872 connection (so you can more easily find what provoked it).
5874 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
5876 5. Added some features to "harden" Exim a bit more against certain attacks:
5878 (a) There is now a build-time option called FIXED_NEVER_USERS that can
5879 be put in Local/Makefile. This is like the never_users runtime option,
5880 but it cannot be overridden. The default setting is "root".
5882 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
5883 prefix string with which any file named in a -C command line option
5886 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
5887 is retained for -C and -D only if the caller of Exim is root. Without
5888 it, the exim user may also use -C and -D and retain privilege.
5890 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
5891 command line option is disabled.
5893 6. Macro names set by the -D option must start with an upper case letter, just
5894 like macro names defined in the configuration file.
5896 7. Added "dereference=" facility to LDAP.
5898 8. Two instances of the typo "uknown" in the source files are fixed.
5900 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
5901 the Configure-Makefile script screwed up while processing it.
5903 10. Incorporated PCRE 4.4.
5905 11. The SMTP synchronization check was not operating right at the start of an
5906 SMTP session. For example, it could not catch a HELO sent before the client
5907 waited for the greeting. There is now a check for outstanding input at the
5908 point when the greeting is written. Because of the duplex, asynchronous
5909 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
5910 way, but not yet received, when the check is performed.
5912 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
5913 on TCP/IP sockets, because this apparently causes some broken clients to
5916 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
5917 unchanged) from the Cygwin maintainer.
5919 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
5920 was supported instead of testing for mbx. Effectively a typo.
5922 15. The spa authenticator server code was not checking that the input it
5923 received was valid base64.
5925 16. The debug output line for the "set" modifier in ACLs was not showing the
5926 name of the variable that was being set.
5928 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
5930 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
5931 lookup on its own. Something else was needed to trigger the lookup. For
5932 example, a match in host_lookup or the need for a host name in a host list.
5933 Now, if $sender_host_name is referenced and the host name has not yet been
5934 looked up, a lookup is performed. If the lookup fails, the variable remains
5935 empty, and $host_lookup_failed is set to "1".
5937 19. Added "eqi" as a case-independent comparison operator.
5939 20. The saslauthd authentication condition could segfault if neither service
5940 nor realm was specified.
5942 21. If an overflowing value such as "2048M" was set for message_size_limit, the
5943 error message that was logged was misleading, and incoming SMTP
5944 connections were dropped. The message is now more accurate, and temporary
5945 errors are given to SMTP connections.
5947 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
5948 (except RSET) with a 421 error, until QUIT is received. However, it was
5949 failing to send a response to QUIT.
5951 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
5952 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
5953 ACL is now run after the helo_try_verify_hosts code.
5955 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
5956 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
5957 case-independent, so other case variants are also recognized). Apparently
5958 some systems use these upper case variants.
5960 25. If more than two messages were waiting for the same host, and a transport
5961 filter was specified for the transport, Exim sent two messages over the
5962 same TCP/IP connection, and then failed with "socket operation on non-
5963 socket" when it tried to send the third.
5965 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
5967 27. The extern definition of crypt16() in expand.c was not being excluded when
5968 the OS had its own crypt16() function.
5970 28. Added bounce_return_body as a new option, and bounce_return_size_limit
5971 as a preferred synonym for return_size_limit, both as an option and as an
5974 29. Added LIBS=-liconv to OS/Makefile-OSF1.
5976 30. Changed the default configuration ACL to relax the local part checking rule
5977 for addresses that are not in any local domains. For these addresses,
5978 slashes and pipe symbols are allowed within local parts, but the sequence
5979 /../ is explicitly forbidden.
5981 31. SPA server authentication was not clearing the challenge buffer before
5984 32. log_message in a "warn" ACL statement was writing to the reject log as
5985 well as to the main log, which contradicts the documentation and doesn't
5986 seem right (because no rejection is happening). So I have stopped it.
5988 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
5989 However, I am unable to do any testing of this.
5991 34. Fixed an infelicity in the appendfile transport. When checking directories
5992 for a mailbox, to see if any needed to be created, it was accidentally
5993 using path names with one or more superfluous leading slashes; tracing
5994 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
5996 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
5997 contents are added to the log line that is written for every discarded
5998 recipient. (Previously a log_message setting was ignored.)
6000 36. The ${quote: operator now quotes the string if it is empty.
6002 37. The install script runs exim in order to find its version number. If for
6003 some reason other than non-existence or emptiness, which it checks, it
6004 could not run './exim', it was installing it with an empty version number,
6005 i.e. as "exim-". This error state is now caught, and the installation is
6008 38. An argument was missing from the function that creates an error message
6009 when Exim fails to connect to the socket for saslauthd authentication.
6010 This could cause Exim to crash, or give a corrupted message.
6012 39. Added isip, isip4, and isip6 to ${if conditions.
6014 40. The ACL variables $acl_xx are now saved with the message, and can be
6015 accessed later in routers, transports, and filters.
6017 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
6018 strings in the file are not string-expanded.
6020 42. If a MAIL command specified a SIZE value that was too large to fit into an
6021 int variable, the check against message_size_limit failed. Such values are
6022 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
6023 day this will have to be increased, but I don't think I want to be around
6024 when emails are that large.
6031 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
6032 iconv() is not standard in FreeBSD.
6034 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
6035 IPv6 enabled. The observed symptom was a segmentation fault on return from
6036 the function os_common_find_running_interfaces() in src/os.c.
6038 3. In the check_special_case() function in daemon.c I had used "errno" as an
6039 argument name, which causes warnings on some systems. This was basically a
6040 typo, since it was named "eno" in the comments!
6042 4. The code that waits for the clock to tick (at a resolution of some fraction
6043 of a second) so as to ensure message-id uniqueness was always waiting for
6044 at least one whole tick, when it could have waited for less. [This is
6045 almost certainly not relevant at current processor speeds, where it is
6046 unlikely to ever wait at all. But we try to future-proof.]
6048 5. The function that sleeps for a time interval that includes fractions of a
6049 second contained a race. It did not block SIGALRM between setting the
6050 timer, and suspending (a couple of lines later). If the interval was short
6051 and the sigsuspend() was delayed until after it had expired, the suspension
6052 never ended. On busy systems this could lead to processes getting stuck for
6055 6. Some uncommon configurations may cause a lookup to happen in a queue runner
6056 process, before it forks any delivery processes. The open lookup caching
6057 mechanism meant that the open file or database connection was passed into
6058 the delivery process. The problem was that delivery processes always tidy
6059 up cached lookup data. This could cause a problem for the next delivery
6060 process started by the queue runner, because the external queue runner
6061 process does not know about the closure. So the next delivery process
6062 still has data in the lookup cache. In the case of a file lookup, there was
6063 no problem because closing a file descriptor in a subprocess doesn't affect
6064 the parent. However, if the lookup was caching a connection to a database,
6065 the connection was closed, and the second delivery process was likely to
6066 see errors such as "PGSQL: query failed: server closed the connection
6067 unexpectedly". The problem has been fixed by closing all cached lookups
6068 in a queue runner before running a delivery process.
6070 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
6071 seem to have the "const" qualifier which it has on other OS. I've
6074 8. Change 4.21/2 was too strict. It is only if there are two authenticators
6075 *of the same type* (client or server) with the same public name that an
6076 error should be diagnosed.
6078 9. When Exim looked up a host name for an IP address, but failed to find the
6079 original IP address when looking up the host name (a safety check), it
6080 output the message "<ip address> does not match any IP for NULL", which was
6081 confusing, to say the least. The bug was that the host name should have
6082 appeared instead of "NULL".
6084 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
6085 user that is built into the binary, and the -C or -D options is used, root
6086 privilege is dropped before the configuration file is read. In addition,
6087 logging is switched to stderr instead of the normal log files. If the
6088 configuration then re-defines the Exim user, the unprivileged environment
6089 is probably not what is expected, so Exim logs a panic warning message (but
6092 However, if deliver_drop_privilege is set, the unprivileged state may well
6093 be exactly what is intended, so the warning has been cut out in that case,
6094 and Exim is allowed to try to write to its normal log files.
6100 1. smtp_return_error_details was not giving details for temporary sender
6101 or receiver verification errors.
6103 2. Diagnose a configuration error if two authenticators have the same public
6106 3. Exim used not to create the message log file for a message until the first
6107 delivery attempt. This could be confusing when incoming messages were held
6108 for policy or load reasons. The message log file is now created at the time
6109 the message is received, and an initial "Received" line is written to it.
6111 4. The automatically generated man page for command line options had a minor
6112 bug that caused no ill effects; however, a more serious problem was that
6113 the procedure for building the man page automatically didn't always
6114 operate. Consequently, release 4.20 contains an out-of-date version. This
6115 shouldn't happen again.
6117 5. When building Exim with embedded Perl support, the script that builds the
6118 Makefile was calling 'perl' to find its compile-time parameters, ignoring
6119 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
6121 6. The freeze_tell option was not being used for messages that were frozen on
6122 arrival, either by an ACL or by local_scan().
6124 7. Added the smtp_incomplete_transaction log selector.
6126 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
6127 was accepting AUTH without a new EHLO.
6129 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
6130 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
6133 10. Logging of TCP/IP connections (when configured) now happens in the main
6134 daemon process instead of the child process, so that the TCP/IP connection
6135 count is more accurate (but it can never be perfect).
6137 11. The use of "drop" in a nested ACL was not being handled correctly in the
6138 outer ACL. Now, if condition failure induced by the nested "drop" causes
6139 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
6140 or "require"), the connection is dropped.
6142 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
6143 that yield "discard" can now be used with an "accept" or a "discard" verb,
6144 but an error is generated for any others (because I can't see a useful way
6145 to define what should happen).
6147 13. When an ACL is read dynamically from a file (or anywhere else), the lines
6148 are now processed in the same way as lines in the Exim configuration file.
6149 In particular, continuation lines are supported.
6151 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
6153 15. Added -ti meaning -t -i.
6155 16. Check for letters, digits, hyphens, and dots in the names of dnslist
6156 domains, and warn by logging if others are found.
6158 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
6159 returned from GIFCONF when Exim is trying to find the list of interfaces on
6160 a host. The code in os.c has been modified to copy each ifreq to an aligned
6161 structure in all cases.
6163 Also, in some cases, the returned ifreq's were being copied to a 'struct
6164 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
6165 means the last couple of bytes of an IPv6 address could be chopped if the
6166 ifreq contained only a normal sockaddr (14 bytes storage).
6168 18. Named domain lists were not supported in the hosts_treat_as_local option.
6169 An entry such as +xxxx was not recognized, and was treated as a literal
6172 19. Ensure that header lines added by a DATA ACL are included in the reject log
6173 if the ACL subsequently rejects the message.
6175 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
6176 MD5 (which is deprecated).
6178 21. When testing a filter file using -bf, Exim was writing a message when it
6179 took the sender from a "From " line in the message, but it was not doing so
6180 when it took $return_path from a Return-Path: header line. It now does.
6182 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
6183 with a valid header line field name (a series of printing characters
6184 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
6186 23. Changed "disc" in the source to "disk" to conform to the documentation and
6187 the book and for uniformity.
6189 24. Ignore Sendmail's -Ooption=value command line item.
6191 25. When execve() failed while trying to run a command in a pipe transport,
6192 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
6193 could be confused with a return value of 69 from the command itself. This
6194 has been changed to 127, the value the shell returns if it is asked to run
6195 a non-existent command. The wording for the related log line suggests a
6196 non-existent command as the problem.
6198 26. If received_header_text expands to an empty string, do not add a Received:
6199 header line to the message. (Well, it adds a token one on the spool, but
6200 marks it "old" so that it doesn't get used or transmitted.)
6202 27. Installed eximstats 1.28 (addition of -nt option).
6204 28. There was no check for failure on the call to getsockname() in the daemon
6205 code. This can fail if there is a shortage of resources on the system, with
6206 ENOMEM, for example. A temporary error is now given on failure.
6208 29. Contrary to the C standard, it seems that in some environments, the
6209 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
6210 program. Exim now does this explicitly; it affects the formatting of
6211 timestamps using strftime().
6213 30. If exiqsumm was given junk data, it threw up some uninitialized variable
6214 complaints. I've now initialized all the variables, to avoid this.
6216 32. Header lines added by a system filter were not being "seen" during
6217 transport-time rewrites.
6219 33. The info_callback() function passed to OpenSSL is set up with type void
6220 (*)(SSL *, int, int), as described somewhere. However, when calling the
6221 function (actually a macro) that sets it up, the type void(*)() is
6222 expected. I've put in a cast to prevent warnings from picky compilers.
6224 34. If a DNS black list lookup found a CNAME record, but there were no A
6225 records associated with the domain it pointed at, Exim crashed.
6227 35. If a DNS black list lookup returned more than one A record, Exim ignored
6228 all but the first. It now scans all returned addresses if a particular IP
6229 value is being sought. In this situation, the contents of the
6230 $dnslist_value variable are a list of all the addresses, separated by a
6233 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
6234 to accept a host name and all its aliases if the forward lookup for any of
6235 them yielded the IP address of the incoming connection. Now it accepts only
6236 those names whose forward lookup yields the correct IP address. Any other
6237 names are discarded. This closes a loophole whereby a rogue DNS
6238 administrator could create reverse DNS records to break through a
6239 wildcarded host restriction in an ACL.
6241 37. If a user filter or a system filter that ran in a subprocess used any of
6242 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
6243 the wrong values were passed to the pipe command ($thisaddress had the
6244 value of $0, $0 had the value of $1, etc). This bug was introduced by
6245 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
6247 38. Improved the line breaking for long SMTP error messages from ACLs.
6248 Previously, if there was no break point between 40 and 75 characters, Exim
6249 left the rest of the message alone. Two changes have been made: (a) I've
6250 reduced the minimum length to 35 characters; (b) if it can't find a break
6251 point between 35 and 75 characters, it looks ahead and uses the first one
6252 that it finds. This may give the occasional overlong line, but at least the
6253 remaining text gets split now.
6255 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
6256 file descriptors might be low, and that setting 1000 would always raise it.
6257 It turns out that in some environments, the limit is already over 1000 and
6258 that lowering it causes trouble. So now Exim takes care not to decrease it.
6260 40. When delivering a message, the value of $return_path is set to $sender_
6261 address at the start of routing (routers may change the value). By an
6262 oversight, this default was not being set up when an address was tested by
6263 -bt or -bv, which affected the outcome if any router or filter referred to
6266 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
6267 log only when "message" or "log_message" are set. However, if one of the
6268 conditions was an address verification, or a call to a nested ACL, the
6269 messages generated by the underlying test were being passed through. This
6270 no longer happens. The underlying message is available in $acl_verify_
6271 message for both "message" and "log_message" expansions, so it can be
6272 passed through if needed.
6274 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
6275 new expansion $bh_ to give the encoded byte string without charset
6276 translation. Translation happens only if iconv() is available; HAVE_ICONV
6277 indicates this at build time. HEADERS_CHARSET gives the charset to
6278 translate to; headers_charset can change it in the configuration, and
6279 "headers charset" can change it in an individual filter file.
6281 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
6282 that creates RFC 2047 encoded "words" labels them as that charset instead
6283 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
6284 expansion operator; (ii) when Exim creates a From: line for a local
6285 message; (iii) when a header line is rewritten to include a "phrase" part.
6287 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
6288 buggy, causing it to skip the first lines of messages whose message ID
6289 ended in 'D'. This would not have bitten before Exim release 4.14, because
6290 message IDs were unlikely to end in 'D' before then. The effect was to have
6291 incorrect size information for certain domains.
6293 45. #include "config.h" was missing at the start of the crypt16.c module. This
6294 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
6297 46. If there was a timeout during a "random" callout check, Exim treated it as
6298 a failure of the random address, and carried on sending RSET and the real
6299 address. If the delay was just some slowness somewhere, the response to the
6300 original RCPT would be taken as a response to RSET and so on, causing
6301 mayhem of various kinds.
6303 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
6304 when I implemented it. It didn't allow for the fact that some option values
6305 may legitimatetly be negative (e.g. size_addition), and it didn't even do
6306 the right test for positive values.
6308 48. Domain names in DNS records are case-independent. Exim always looks them up
6309 in lower case. Some resolvers return domain names in exactly the case they
6310 appear in the zone file, that is, they may contain uppercase letters. Not
6311 all resolvers do this - some return always lower case. Exim was treating a
6312 change of case by a resolver as a change of domain, similar to a widening
6313 of a domain abbreviation. This triggered its re-routing code and so it was
6314 trying to route what was effectively the same domain again. This normally
6315 caused routing to fail (because the router wouldn't handle the domain
6316 twice). Now Exim checks for this case specially, and just changes the
6317 casing of the domain that it ultimately uses when it transmits the message
6320 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
6323 50. If a filter generated a file delivery with a non-absolute name (possible if
6324 no home directory exists for the router), the forbid_file option was not
6327 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
6328 the existing equality matching.
6330 52. Exim was using ints instead of ino_t variables in some places where it was
6331 dealing with inode numbers.
6333 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
6334 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
6335 called TMPDIR, and if it finds it is different, it changes its value.
6337 54. The smtp_printf() function is now made available to local_scan() so
6338 additional output lines can be written before returning. There is also an
6339 smtp_fflush() function to enable the detection of a dropped connection.
6340 The variables smtp_input and smtp_batched_input are exported to
6343 55. Changed the default runtime configuration: the message "Unknown user"
6344 has been removed from the ACL, and instead placed on the localuser router,
6345 using the cannot_route_message feature. This means that any verification
6346 failures that generate their own messages won't get overridden. Similarly,
6347 the "Unrouteable address" message that was in the ACL for unverifiable
6348 relay addresses has also been removed.
6350 56. Added hosts_avoid_esmtp to the smtp transport.
6352 57. The exicyclog script was not checking for the esoteric option
6353 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
6354 will work only if exicyclog is run under the appropriate euid.
6356 58. Following a discussion on the list, the rules by which Exim recognises line
6357 endings on incoming messages have been changed. The -dropcr and drop_cr
6358 options are now no-ops, retained only for backwards compatibility. The
6359 following line terminators are recognized: LF CRLF CR. However, special
6360 processing applies to CR:
6362 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
6363 nor a local message in the state where . is a terminator.
6365 (ii) If a bare CR is encountered in a header line, an extra space is added
6366 after the line terminator so as not to end the header. The reasoning
6367 behind this is that bare CRs in header lines are most likely either
6368 to be mistakes, or people trying to play silly games.
6370 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
6371 was being incorrectly given as 18 bytes larger than it should have been.
6372 This is a VOB (very old bug).
6374 60. This may never have affected anything current, but just in case it has:
6375 When the local host is found other than at the start of a list of hosts,
6376 the local host, those with the same MX, and any that follow, are discarded.
6377 When the list in question was part of a longer list of hosts, the following
6378 hosts (not currently being processed) were also being discarded. This no
6379 longer happens. I'm not sure if this situation could ever has previously
6382 61. Added the "/MX" feature to lists of hosts in the manualroute and query
6385 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
6386 header. This is something that is recommended in a new Internet Draft, and
6387 is something that is documented as being done by Sendmail. There are two
6388 possible values. For messages generated by the autoreply transport, Exim
6391 Auto-Submitted: auto-replied
6393 whereas for all other generated messages (e.g. bounces) it adds
6395 Auto-Submitted: auto-generated
6397 63. The "personal" condition in filters now includes a test for the
6398 Auto-Submitted: header. If it contains the string "auto-" the message it
6399 not considered personal.
6401 64. Added rcpt_include_affixes as a generic transport option.
6403 65. Added queue_only_override (default true).
6405 66. Added the syslog_duplication option.
6407 67. If what should have been the first header line of a message consisted of
6408 a space followed by a colon, Exim was mis-interpreting it as a header line.
6409 It isn't of course - it is syntactically invalid and should therefore be
6410 treated as the start of the message body. The misbehaviour could have
6411 caused a number of strange effects, including loss of data in subsequent
6412 header lines, and spool format errors.
6414 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
6415 client host had authenticated. This control can now be exercised by an ACL
6416 for more flexibility.
6418 69. By default, callouts do not happen when testing with -bh. There is now a
6419 variant, -bhc, which does actually run the callout code, including
6420 consulting and updating the callout cache.
6422 70. Added support for saslauthd authentication, courtesy of Alexander
6425 71. If statvfs() failed on the spool or log directories while checking their
6426 size for availability, Exim confusingly gave the error "space shortage".
6427 Furthermore, in debugging mode it crashed with a floating point exception.
6428 These checks are done if check_{spool,log}_{space,inodes} are set, and when
6429 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
6430 serious problem, Exim now writes to the main and panic logs when this
6431 happens, with details of the failure. It then refuses to accept the
6432 incoming message, giving the message "spool directory problem" or "log
6433 directory problem" with a 421 code for SMTP messages.
6435 72. When Exim is about to re-exec itself, it ensures that the file descriptors
6436 0, 1, and 2 exist, because some OS complain for execs without them (see
6437 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
6438 descriptors. However, the code omitted to check that the open succeeded,
6439 causing mysterious errors if for some reason the permissions on /dev/null
6440 got screwed. Now Exim writes a message to the main and panic logs, and
6441 bombs out if it can't open /dev/null.
6443 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
6444 interact so that it is all more flexible. It is supposed to remain
6445 backwards compatible. Also added extra_local_interfaces.
6447 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
6448 to bomb out with an assertion failure - to the client this appears as a
6449 connection drop. This problem occurs in the part of the code that was taken
6450 from the Samba project. Fortunately, the assertion is in a very simple
6451 function, so I have fixed this by reproducing the function inline in the
6452 one place where it is called, and arranging for authentication to fail
6453 instead of killing the process with assert().
6455 75. The SPA client code was not working when the server requested OEM rather
6456 than Unicode encoding.
6458 76. Added code to make require_files with a specific uid setting more usable in
6459 the case where statting the file as root fails - usually a non-root-mounted
6460 NFS file system. When this happens and the failure is EACCES, Exim now
6461 forks a subprocess and does the per-uid checking as the relevant uid.
6463 77. Added process_log_path.
6465 78. If log_file_path was not explicitly set, a setting of check_log_space or
6466 check_log_inodes was ignored.
6468 79. If a space check for the spool or log partitions fails, the incident is now
6469 logged. Of course, in the latter case the data may get lost...
6471 80. Added the %p formatting code to string_format() so that it can be used to
6472 print addresses in debug_print(). Adjusted all the address printing in the
6473 debugging in store.c to use %p rather than %d.
6475 81. There was a concern that a line of code in smtp_in.c could overflow a
6476 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
6477 initially expressed, the concern was not well-founded, because trailing
6478 spaces are removed early. However, if the trailing spaces were followed by
6479 a NULL, they did not get removed, so the overflow was possible. Two fixes
6482 (a) I re-wrote the offending code in a cleaner fashion.
6483 (b) If an incoming SMTP command contains a NULL character, it is rejected
6486 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
6487 runs initgroups(), so that if the Exim user is in any additional groups,
6488 they will be used during message reception.
6494 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob