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
193 JH/47 Fix build on HP-UX and older Solaris, which need (un)setenv now also
194 for the new environment-manipulation done at startup. Move the routines
195 from being local to tls.c to being global via the os.c file.
201 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
204 JH/02 The smtp transport option "multi_domain" is now expanded.
206 JH/03 The smtp transport now requests PRDR by default, if the server offers
209 JH/04 Certificate name checking on server certificates, when exim is a client,
210 is now done by default. The transport option tls_verify_cert_hostnames
211 can be used to disable this per-host. The build option
212 EXPERIMENTAL_CERTNAMES is withdrawn.
214 JH/05 The value of the tls_verify_certificates smtp transport and main options
215 default to the word "system" to access the system default CA bundle.
216 For GnuTLS, only version 3.0.20 or later.
218 JH/06 Verification of the server certificate for a TLS connection is now tried
219 (but not required) by default. The verification status is now logged by
220 default, for both outbound TLS and client-certificate supplying inbound
223 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
226 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
227 Status Notification (bounce) messages are now MIME format per RFC 3464.
228 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
229 under the control of the dsn_advertise_hosts option, and routers may
230 have a dsn_lasthop option.
232 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
233 default, modifiable by a malware= option. The list separator for
234 the options can now be changed in the usual way. Bug 68.
236 JH/10 The smtp_receive_timeout main option is now expanded before use.
238 JH/11 The incoming_interface log option now also enables logging of the
239 local interface on delivery outgoing connections.
241 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
242 if the interface and destination host and port all match.
244 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
247 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
248 Patch from Andrew Lewis.
250 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
251 now supports optional time-restrictions, weighting, and priority
252 modifiers per server. Patch originally by <rommer@active.by>.
254 JH/16 The spamd_address main option now supports a mixed list of local
255 and remote servers. Remote servers can be IPv6 addresses, and
256 specify a port-range.
258 JH/17 Bug 68: The spamd_address main option now supports an optional
259 timeout value per server.
261 JH/18 Bug 1581: Router and transport options headers_add/remove can
262 now have the list separator specified.
264 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
267 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
270 JH/21 Support for the A6 type of dns record is withdrawn.
272 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
273 rather than the verbs used.
275 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
276 from 255 to 1024 chars.
278 JH/24 Verification callouts now attempt to use TLS by default.
280 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
281 are generic router options now. The defaults didn't change.
283 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
284 Original patch from Alexander Shikoff, worked over by JH.
286 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
287 files if ZCAT_COMMAND is not executable.
289 JH/26 Bug 1539: Add timout/retry options on dnsdb lookups.
291 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
293 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
294 Normally benign, it bites when the pair was led to by a CNAME;
295 modern usage is to not canoicalize the domain to a CNAME target
296 (and we were inconsistent anyway for A-only vs AAAA+A).
298 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
300 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
301 when evaluating $sender_host_dnssec.
303 JH/31 Check the HELO verification lookup for DNSSEC, adding new
304 $sender_helo_dnssec variable.
306 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
308 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
310 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
312 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
313 documented as working, but never had. Support all but $spam_report.
315 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
316 added for tls authenticator.
321 TL/01 When running the test suite, the README says that variables such as
322 no_msglog_check are global and can be placed anywhere in a specific
323 test's script, however it was observed that placement needed to be near
324 the beginning for it to behave that way. Changed the runtest perl
325 script to read through the entire script once to detect and set these
326 variables, reset to the beginning of the script, and then run through
327 the script parsing/test process like normal.
329 TL/02 The BSD's have an arc4random API. One of the functions to induce
330 adding randomness was arc4random_stir(), but it has been removed in
331 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
332 function when detected.
334 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
335 cause callback expansion.
337 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
338 syntax errors in an expansion can be treated as a string instead of
339 logging or causing an error, due to the internal use of bool_lax
340 instead of bool when processing it.
342 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
343 server certificates when making smtp deliveries.
345 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
347 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
349 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
351 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
352 Merged patch from Sebastian Wiedenroth.
354 JH/05 Fix results-pipe from transport process. Several recipients, combined
355 with certificate use, exposed issues where response data items split
356 over buffer boundaries were not parsed properly. This eventually
357 resulted in duplicates being sent. This issue only became common enough
358 to notice due to the introduction of conection certificate information,
359 the item size being so much larger. Found and fixed by Wolfgang Breyha.
361 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
362 size buffer was used, resulting in syntax errors when an expansion
365 JH/07 Add support for directories of certificates when compiled with a GnuTLS
366 version 3.3.6 or later.
368 JH/08 Rename the TPDA expermimental facility to Event Actions. The #ifdef
369 is EXPERIMENTAL_EVENT, the main-configuration and transport options
370 both become "event_action", the variables become $event_name, $event_data
371 and $event_defer_errno. There is a new variable $verify_mode, usable in
372 routers, transports and related events. The tls:cert event is now also
373 raised for inbound connections, if the main configuration event_action
376 TL/06 In test suite, disable OCSP for old versions of openssl which contained
377 early OCSP support, but no stapling (appears to be less than 1.0.0).
379 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
380 server certificate names available under the smtp transport option
381 "tls_verify_cert_hostname" now do not permit multi-component wildcard
384 JH/10 Time-related extraction expansions from certificates now use the main
385 option "timezone" setting for output formatting, and are consistent
386 between OpenSSL and GnuTLS compilations. Bug 1541.
388 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
389 encoded parameter in the incoming message. Bug 1558.
391 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
392 include certificate info, eximon was claiming there were spoolfile
395 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
397 JH/14 Log delivery-related information more consistently, using the sequence
398 "H=<name> [<ip>]" wherever possible.
400 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
401 are problematic for Debian distribution, omit them from the release
404 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
406 JH/16 Fix string representation of time values on 64bit time_t anchitectures.
409 JH/17 Fix a null-indirection in certextract expansions when a nondefault
410 output list separator was used.
415 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
416 checkers that were complaining about end of non-void function with no
419 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
420 This was a regression intruduced in 4.83 by another bugfix.
422 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
424 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
425 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
431 TF/01 Correctly close the server side of TLS when forking for delivery.
433 When a message was received over SMTP with TLS, Exim failed to clear up
434 the incoming connection properly after forking off the child process to
435 deliver the message. In some situations the subsequent outgoing
436 delivery connection happened to have the same fd number as the incoming
437 connection previously had. Exim would try to use TLS and fail, logging
438 a "Bad file descriptor" error.
440 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
441 utilities have not been installed.
443 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
444 temporary space as the ACL may create new global variables.
446 TL/01 LDAP support uses per connection or global context settings, depending
447 upon the detected version of the libraries at build time.
449 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
450 to extract and use the src ip:port in logging and expansions as if it
451 were a direct connection from the outside internet. PPv2 support was
452 updated based on HAProxy spec change in May 2014.
454 JH/02 Add ${listextract {number}{list}{success}{fail}}.
456 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
457 Properly escape header and check for NULL return.
459 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
462 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
464 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
465 characters in header names, implemented as a verify condition.
466 Contributed by Michael Fischer v. Mollard.
468 TL/05 Rename SPF condition results err_perm and err_temp to standardized
469 results permerror and temperror. Previous values are deprecated but
470 still accepted. In a future release, err_perm and err_temp will be
471 completely removed, which will be a backward incompatibility if the
472 ACL tests for either of these two old results. Patch contributed by
473 user bes-internal on the mailing list.
475 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
477 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
478 selectors, in both main and reject logs.
480 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
483 JH/07 Add malware type "sock" for talking to simple daemon.
485 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
487 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
488 routers/transports under cutthrough routing.
490 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
491 numbers. Touch up "bool" conditional to keep the same definition.
493 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
495 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
497 JH/12 Expand items in router/transport headers_add or headers_remove lists
498 individually rather than the list as a whole. Bug 1452.
500 Required for reasonable handling of multiple headers_ options when
501 they may be empty; requires that headers_remove items with embedded
502 colons must have them doubled (or the list-separator changed).
504 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
505 view the policy declared in the DMARC record. Currently, $dmarc_status
506 is a combined value of both the record presence and the result of the
509 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
511 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
512 dnslookup router and the smtp transport (applying to the forward
515 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
516 of ldap servers used for a specific lookup. Patch provided by Heiko
519 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
520 New variable $lookup_dnssec_authenticated for observability.
522 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
523 Patch submitted by Lars Timman.
525 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
527 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
528 Requires trusted mode and valid format message id, aborts otherwise.
529 Patch contributed by Heiko Schlichting.
531 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
532 certextract with support for various fields. Bug 1358.
534 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
535 is requested by default, modifiable by smtp transport option
538 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
539 operate on certificate variables to give certificate fingerprints
540 Also new ${sha256:cert_variable}.
542 JH/23 The PRDR feature is moved from being Experimental into the mainline.
544 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
545 Christian Aistleitner.
547 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
549 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
550 file. Patch from Wolfgang Breyha.
552 JH/25 Expand the coverage of the delivery $host and $host_address to
553 client authenticators run in verify callout. Bug 1476.
555 JH/26 Port service names are now accepted for tls_on_connect_ports, to
556 align with daemon_smtp_ports. Bug 72.
558 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
559 support and error reporting did not work properly.
561 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
562 and is readable. Patch from Andrew Colin Kissa.
564 TL/14 Enhance documentation of ${run expansion and how it parses the
565 commandline after expansion, particularly in the case when an
566 unquoted variable expansion results in an empty value.
568 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
570 PP/02 Fix internal collision of T_APL on systems which support RFC3123
571 by renaming away from it. Addresses GH issue 15, reported by
574 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
576 TL/15 SECURITY: prevent double expansion in math comparison functions
577 (can expand unsanitized data). Not remotely exploitable.
584 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
586 PP/02 Make -n do something, by making it not do something.
587 When combined with -bP, the name of an option is not output.
589 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
592 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
593 $sender_host_name and config options to manage this, and basic check
596 PP/05 DSCP support for outbound connections and control modifier for inbound.
598 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
599 (Only plugin which currently uses this is kerberos4, which nobody should
600 be using, but we should make it available and other future plugins might
601 conceivably use it, even though it would break NAT; stuff *should* be
602 using channel bindings instead).
604 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
605 name; added for Sendmail compatibility; requires admin caller.
606 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
607 just ignore it); requires trusted caller.
608 Also parse but ignore: -Ac -Am -X<logfile>
611 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
613 TL/02 Add +smtp_confirmation as a default logging option.
615 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
616 Patch by Magnus Holmgren from 2007-02-20.
618 TL/04 Bugzilla 1281 - Spec typo.
619 Bugzilla 1283 - Spec typo.
620 Bugzilla 1290 - Spec grammar fixes.
622 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
624 TL/06 Add Experimental DMARC support using libopendmarc libraries.
626 TL/07 Fix an out of order global option causing a segfault. Reported to dev
627 mailing list by by Dmitry Isaikin.
629 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
631 JH/02 Support "G" suffix to numbers in ${if comparisons.
633 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
635 NM/01 Bugzilla 1197 - Spec typo
636 Bugzilla 1196 - Spec examples corrections
638 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
640 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
641 gnutls_enable_pkcs11, but renamed to more accurately indicate its
644 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
645 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
647 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
648 "acl {{name}{arg}...}", and optional args on acl condition
651 JH/05 Permit multiple router/transport headers_add/remove lines.
653 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
655 JH/07 Avoid using a waiting database for a single-message-only transport.
656 Performance patch from Paul Fisher. Bugzilla 1262.
658 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
661 JH/09 Add $headers_added variable, with content from use of ACL modifier
662 add_header (but not yet added to the message). Bugzilla 199.
664 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
665 Pulled from Bugzilla 817 by Wolfgang Breyha.
667 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
669 (nb: this is the same fix as in Exim 4.80.1)
671 JH/11 Add A= logging on delivery lines, and a client_set_id option on
674 JH/12 Add optional authenticated_sender logging to A= and a log_selector
677 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
679 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
680 advertise SMTP AUTH mechanism to us, instead of a generic
681 protocol violation error. Also, make Exim more robust to bad
682 data from the Dovecot auth socket.
684 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
686 When a queue runner is handling a message, Exim first routes the
687 recipient addresses, during which it prunes them based on the retry
688 hints database. After that it attempts to deliver the message to
689 any remaining recipients. It then updates the hints database using
692 So if a recipient address works intermittently, it can get repeatedly
693 deferred at routing time. The retry hints record remains fresh so the
694 address never reaches the final cutoff time.
696 This is a fairly common occurrence when a user is bumping up against
697 their storage quota. Exim had some logic in its local delivery code
698 to deal with this. However it did not apply to per-recipient defers
699 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
701 This change adds a proper retry rule check during routing so that the
702 final cutoff time is checked against the message's age. We only do
703 this check if there is an address retry record and there is not a
704 domain retry record; this implies that previous attempts to handle
705 the address had the retry_use_local_parts option turned on. We use
706 this as an approximation for the destination being like a local
707 delivery, as in LMTP.
709 I suspect this new check makes the old local delivery cutoff check
710 redundant, but I have not verified this so I left the code in place.
712 TF/02 Correct gecos expansion when From: is a prefix of the username.
714 Test 0254 submits a message to Exim with the header
718 When I ran the test suite under the user fanf2, Exim expanded
719 the header to contain my full name, whereas it should have added
720 a Resent-Sender: header. It erroneously treats any prefix of the
721 username as equal to the username.
723 This change corrects that bug.
725 GF/01 DCC debug and logging tidyup
726 Error conditions log to paniclog rather than rejectlog.
727 Debug lines prefixed by "DCC: " to remove any ambiguity.
729 TF/03 Avoid unnecessary rebuilds of lookup-related code.
731 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
732 Bug spotted by Jeremy Harris; was flawed since initial commit.
733 Would have resulted in OCSP responses post-SNI triggering an Exim
734 NULL dereference and crash.
736 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
738 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
739 Bug detection, analysis and fix by Samuel Thibault.
740 Bugzilla 1331, Debian bug #698092.
742 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
744 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
745 Server implementation by Todd Lyons, client by JH.
746 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
747 config variable "prdr_enable" controls whether the server
748 advertises the facility. If the client requests PRDR a new
749 acl_data_smtp_prdr ACL is called once for each recipient, after
750 the body content is received and before the acl_smtp_data ACL.
751 The client is controlled by bolth of: a hosts_try_prdr option
752 on the smtp transport, and the server advertisement.
753 Default client logging of deliveries and rejections involving
754 PRDR are flagged with the string "PRDR".
756 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
757 fclose(). Diagnosis by Todd Lyons.
759 PP/17 Update configure.default to handle IPv6 localhost better.
760 Patch by Alain Williams (plus minor tweaks).
763 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
764 This is now consistent with GnuTLS, and is now documented: the
765 previous undocumented portable approach to treating the option as
766 unset was to force an expansion failure. That still works, and
767 an empty string is now equivalent.
769 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
770 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
771 not performing validation itself.
773 PP/20 Added force_command boolean option to pipe transport.
774 Patch from Nick Koston, of cPanel Inc.
776 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
779 TF/04 Added udpsend ACL modifer and hexquote expansion operator
781 PP/21 Fix eximon continuous updating with timestamped log-files.
782 Broken in a format-string cleanup in 4.80, missed when I repaired the
783 other false fix of the same issue.
784 Report and fix from Heiko Schlichting.
787 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
788 Report from Prashanth Katuri.
790 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
791 It's SecureTransport, so affects any MacOS clients which use the
792 system-integrated TLS libraries, including email clients.
794 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
795 using a MIME ACL for non-SMTP local injection.
796 Report and assistance in diagnosis by Warren Baker.
798 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
800 JH/16 Fix comparisons for 64b. Bugzilla 1385.
802 TL/09 Add expansion variable $authenticated_fail_id to keep track of
803 last id that failed so it may be referenced in subsequent ACL's.
805 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
808 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
809 ldap library initialization, allowing self-signed CA's to be
810 used. Also properly sets require_cert option later in code by
811 using NULL (global ldap config) instead of ldap handle (per
812 session). Bug diagnosis and testing by alxgomz.
814 TL/12 Enhanced documentation in the ratelimit.pl script provided in
815 the src/util/ subdirectory.
817 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
818 renamed to Transport Post Delivery Action by Jeremy Harris, as
821 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
822 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
823 redis_servers = needs to be configured which will be used by the redis
824 lookup. Patch from Warren Baker, of The Packet Hub.
826 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
828 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
829 hostname or reverse DNS when processing a host list. Used suggestions
830 from multiple comments on this bug.
832 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
834 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
835 Missed a few lines, added it to make the runtest require no keyboard
838 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
839 contains upper case chars. Make router use caseful_local_part.
841 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
842 support when GnuTLS has been built with p11-kit.
848 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
850 This, or similar/improved, will also be change PP/11 of 4.82.
856 PP/01 Handle short writes when writing local log-files.
857 In practice, only affects FreeBSD (8 onwards).
858 Bugzilla 1053, with thanks to Dmitry Isaikin.
860 NM/01 Bugzilla 949 - Documentation tweak
862 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
865 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
867 PP/02 Implemented gsasl authenticator.
869 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
871 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
872 `pkg-config foo` for cflags/libs.
874 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
875 with rest of GSASL and with heimdal_gssapi.
877 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
878 `pkg-config foo` for cflags/libs for the TLS implementation.
880 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
881 properties get this fed in as external SSF. A number of robustness
882 and debugging improvements to the cyrus_sasl authenticator.
884 PP/08 cyrus_sasl server now expands the server_realm option.
886 PP/09 Bugzilla 1214 - Log authentication information in reject log.
887 Patch by Jeremy Harris.
889 PP/10 Added dbmjz lookup type.
891 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
893 PP/12 MAIL args handles TAB as well as SP, for better interop with
894 non-compliant senders.
895 Analysis and variant patch by Todd Lyons.
897 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
898 Bug report from Lars Müller <lars@samba.org> (via SUSE),
899 Patch from Dirk Mueller <dmueller@suse.com>
901 PP/13 tls_peerdn now print-escaped for spool files.
902 Observed some $tls_peerdn in wild which contained \n, which resulted
903 in spool file corruption.
905 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
906 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
907 or write after TLS renegotiation, which otherwise led to messages
910 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
911 as a tracking header (ie: a signed header comes before the signature).
912 Patch from Wolfgang Breyha.
914 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
915 comma-sep list; embedded commas doubled.
917 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
919 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
921 Report and patch from Dmitry Banschikov.
923 PP/16 Removed "dont_insert_empty_fragments" fron "openssl_options".
924 Removed SSL_clear() after SSL_new() which led to protocol negotiation
925 failures. We appear to now support TLS1.1+ with Exim.
927 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
928 lets Exim select keys and certificates based upon TLS SNI from client.
929 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
930 before an outbound SMTP session. New log_selector, +tls_sni.
932 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
933 NULL dereference. Report and patch from Alun Jones.
935 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
936 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
937 Not seeing resolver debug output on NetBSD, but suspect this is a
938 resolver implementation change.
940 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
941 Left warnings. Added "eximon gdb" invocation mode.
943 PP/21 Defaulting "accept_8bitmime" to true, not false.
945 PP/22 Added -bw for inetd wait mode support.
947 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
948 locate the relevant includes and libraries. Made this the default.
950 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
951 Bugzilla 1246, report and most of solution from Tomasz Kusy.
953 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
954 This may cause build issues on older platforms.
956 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
957 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
958 gnutls_require_mac & gnutls_require_protocols (no longer supported).
959 Added SNI support via GnuTLS too.
960 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
962 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
964 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
965 Applied second patch from Janne, implementing suggestion to default
966 multiple-strings-in-record handling to match SPF spec.
968 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
970 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
971 read-only, out of scope).
972 Patch from Wolfgang Breyha, report from Stuart Northfield.
974 PP/29 Fix three issues highlighted by clang analyser static analysis.
975 Only crash-plausible issue would require the Cambridge-specific
976 iplookup router and a misconfiguration.
977 Report from Marcin Mirosław.
979 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
981 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
982 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
983 As part of this, removing so much warning spew let me fix some minor
984 real issues in debug logging.
986 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
987 assignment on my part. Fixed.
989 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
990 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
991 Janne Snabb (who went above and beyond: thank you).
993 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
994 string otherwise requires a connection and a bunch more work and it's
995 relatively easy to get wrong. Should also expose TLS library linkage
998 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
999 64-bit ${eval} (JH/03).
1001 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
1002 GNU libc to support some of the 64-bit stuff, should not lead to
1003 conflicts. Defined before os.h is pulled in, so if a given platform
1004 needs to override this, it can.
1006 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
1007 protection layer was required, which is not implemented.
1008 Bugzilla 1254, patch from Wolfgang Breyha.
1010 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
1011 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
1012 tls_dhparam take prime identifiers. Also unbreak combination of
1013 OpenSSL+DH_params+TLSSNI.
1015 PP/39 Disable SSLv2 by default in OpenSSL support.
1021 PP/01 Solaris build fix for Oracle's LDAP libraries.
1022 Bugzilla 1109, patch from Stephen Usher.
1024 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
1026 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
1029 TF/02 Fix a couple more cases where we did not log the error message
1030 when unlink() failed. See also change 4.74-TF/03.
1032 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
1033 lock up or crash if it happened to be inside a call to libc when it
1034 got a SIGUSR1 from exiwhat.
1036 The SIGUSR1 handler appends the current process status to the process
1037 log which is later printed by exiwhat. It used to use the general
1038 purpose logging code to do this, but several functions it calls are
1039 not safe for signals.
1041 The new output code in the SIGUSR1 handler is specific to the process
1042 log, and simple enough that it's easy to inspect for signal safety.
1043 Removing some special cases also simplifies the general logging code.
1044 Removing the spurious timestamps from the process log simplifies
1047 TF/04 Improved ratelimit ACL condition.
1049 The /noupdate option has been deprecated in favour of /readonly which
1050 has clearer semantics. The /leaky, /strict, and /readonly update modes
1051 are mutually exclusive. The update mode is no longer included in the
1052 database key; it just determines when the database is updated. (This
1053 means that when you upgrde Exim will forget old rate measurements.)
1055 Exim now checks that the per_* options are used with an update mode that
1056 makes sense for the current ACL. For example, when Exim is processing a
1057 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
1058 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
1059 must specify per_mail/readonly. If you omit the update mode it defaults to
1060 /leaky where that makes sense (as before) or /readonly where required.
1062 The /noupdate option is now undocumented but still supported for
1063 backwards compatibility. It is equivalent to /readonly except that in
1064 ACLs where /readonly is required you may specify /leaky/noupdate or
1065 /strict/noupdate which are treated the same as /readonly.
1067 A useful new feature is the /count= option. This is a generalization
1068 of the per_byte option, so that you can measure the throughput of other
1069 aggregate values. For example, the per_byte option is now equivalent
1070 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
1072 The per_rcpt option has been generalized using the /count= mechanism
1073 (though it's more complicated than the per_byte equivalence). When it is
1074 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
1075 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
1076 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
1077 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
1078 Note that using per_rcpt with a non-readonly update mode in more than
1079 one ACL will cause the recipients to be double-counted. (The per_mail
1080 and per_byte options don't have this problem.)
1082 The handling of very low rates has changed slightly. If the computed rate
1083 is less than the event's count (usually one) then this event is the first
1084 after a long gap. In this case the rate is set to the same as this event's
1085 count, so that the first message of a spam run is counted properly.
1087 The major new feature is a mechanism for counting the rate of unique
1088 events. The new per_addr option counts the number of different
1089 recipients that someone has sent messages to in the last time period. It
1090 behaves like per_rcpt if all the recipient addresses are different, but
1091 duplicate recipient addresses do not increase the measured rate. Like
1092 the /count= option this is a general mechanism, so the per_addr option
1093 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
1094 example, measure the rate that a client uses different sender addresses
1095 with the options per_mail/unique=$sender_address. There are further
1096 details in the main documentation.
1098 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
1100 TF/06 Removed a few PCRE remnants.
1102 TF/07 Automatically extract Exim's version number from tags in the git
1103 repository when doing development or release builds.
1105 PP/02 Raise smtp_cmd_buffer_size to 16kB.
1106 Bugzilla 879. Patch from Paul Fisher.
1108 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
1109 Heavily based on revision 40f9a89a from Simon Arlott's tree.
1112 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
1114 PP/05 Variable $av_failed, true if the AV scanner deferred.
1115 Bugzilla 1078. Patch from John Horne.
1117 PP/06 Stop make process more reliably on build failure.
1118 Bugzilla 1087. Patch from Heiko Schlittermann.
1120 PP/07 Make maildir_use_size_file an _expandable_ boolean.
1121 Bugzilla 1089. Patch from Heiko Schlittermann.
1123 PP/08 Handle ${run} returning more data than OS pipe buffer size.
1124 Bugzilla 1131. Patch from Holger Weiß.
1126 PP/09 Handle IPv6 addresses with SPF.
1127 Bugzilla 860. Patch from Wolfgang Breyha.
1129 PP/10 GnuTLS: support TLS 1.2 & 1.1.
1131 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
1134 PP/11 match_* no longer expand right-hand-side by default.
1135 New compile-time build option, EXPAND_LISTMATCH_RHS.
1136 New expansion conditions, "inlist", "inlisti".
1138 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
1140 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
1142 PP/14 fix log_write() format string regression from TF/03.
1143 Bugzilla 1152. Patch from Dmitry Isaikin.
1149 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
1151 PP/02 Harmonised TLS library version reporting; only show if debugging.
1152 Layout now matches that introduced for other libraries in 4.74 PP/03.
1154 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
1156 PP/04 New "dns_use_edns0" global option.
1158 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
1161 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
1162 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
1164 TK/01 Updated PolarSSL code to 0.14.2.
1165 Bugzilla 1097. Patch from Andreas Metzler.
1167 PP/07 Catch divide-by-zero in ${eval:...}.
1168 Fixes bugzilla 1102.
1170 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
1173 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
1174 format-string attack -- SECURITY: remote arbitrary code execution.
1176 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
1177 time unintentionally subject to list matching rules, letting the header
1178 cause arbitrary Exim lookups (of items which can occur in lists, *not*
1179 arbitrary string expansion). This allowed for information disclosure.
1181 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
1182 INT_MIN/-1 -- value coerced to INT_MAX.
1188 NM/01 Workround for PCRE version dependency in version reporting
1191 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
1192 This fixes portability to compilers other than gcc, notably
1193 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
1195 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
1196 makefiles for portability to HP-UX and POSIX correctness.
1198 PP/01 Permit LOOKUP_foo enabling on the make command-line.
1199 Also via indented variable definition in the Makefile.
1200 (Debugging by Oliver Heesakkers).
1202 PP/02 Restore caching of spamd results with expanded spamd_address.
1203 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
1205 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
1206 Improves build reliability. Fix from: Frank Elsner
1208 NM/02 Fix wide character breakage in the rfc2047 coding
1209 Fixes bug 1064. Patch from Andrey N. Oktyabrski
1211 NM/03 Allow underscore in dnslist lookups
1212 Fixes bug 1026. Patch from Graeme Fowler
1214 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
1215 Code patches from Adam Ciarcinski of NetBSD.
1217 NM/04 Fixed exiqgrep to cope with mailq missing size issue
1220 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
1221 is logged, to avoid truncation. Patch from John Horne.
1223 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
1224 Patch from Jakob Hirsch.
1226 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
1227 SQL string expansion failure details.
1228 Patch from Andrey Oktyabrski.
1230 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
1231 Patch from Simon Arlott.
1233 PP/09 New lookups functionality failed to compile on old gcc which rejects
1234 extern declarations in function scope.
1235 Patch from Oliver Fleischmann
1237 PP/10 Use sig_atomic_t for flags set from signal handlers.
1238 Check getgroups() return and improve debugging.
1239 Fixed developed for diagnosis in bug 927 (which turned out to be
1242 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
1243 Patch from Mark Zealey.
1245 PP/12 Bugzilla 1056: Improved spamd server selection.
1246 Patch from Mark Zealey.
1248 PP/13 Bugzilla 1086: Deal with maildir quota file races.
1249 Based on patch from Heiko Schlittermann.
1251 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
1252 Patch from Uwe Doering, sign-off by Michael Haardt.
1254 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
1255 variable declaration deep within a block. Bug and patch from
1258 PP/15 lookups-Makefile IRIX compatibilty coercion.
1260 PP/16 Make DISABLE_DKIM build knob functional.
1262 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
1263 Patch by Simon Arlott
1265 TF/03 Fix valgrind.h portability to C89 compilers that do not support
1266 variable argument macros. Our copy now differs from upstream.
1272 TF/01 Failure to get a lock on a hints database can have serious
1273 consequences so log it to the panic log.
1275 TF/02 Log LMTP confirmation messages in the same way as SMTP,
1276 controlled using the smtp_confirmation log selector.
1278 TF/03 Include the error message when we fail to unlink a spool file.
1280 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
1281 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
1282 for maintaining out-of-tree patches for some time.
1284 PP/01 Bugzilla 139: Documentation and portability issues.
1285 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
1286 Handle per-OS dynamic-module compilation flags.
1288 PP/02 Let /dev/null have normal permissions.
1289 The 4.73 fixes were a little too stringent and complained about the
1290 permissions on /dev/null. Exempt it from some checks.
1291 Reported by Andreas M. Kirchwitz.
1293 PP/03 Report version information for many libraries, including
1294 Exim version information for dynamically loaded libraries. Created
1295 version.h, now support a version extension string for distributors
1296 who patch heavily. Dynamic module ABI change.
1298 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
1299 privilege escalation vulnerability whereby the Exim run-time user
1300 can cause root to append content of the attacker's choosing to
1303 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
1306 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
1307 If dropping privileges for untrusted macros, we disabled normal logging
1308 on the basis that it would fail; for the Exim run-time user, this is not
1309 the case, and it resulted in successful deliveries going unlogged.
1310 Fixed. Reported by Andreas Metzler.
1316 PP/01 Date: & Message-Id: revert to normally being appended to a message,
1317 only prepend for the Resent-* case. Fixes regression introduced in
1318 Exim 4.70 by NM/22 for Bugzilla 607.
1320 PP/02 Include check_rfc2047_length in configure.default because we're seeing
1321 increasing numbers of administrators be bitten by this.
1323 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
1325 PP/03 Bugzilla 994: added openssl_options main configuration option.
1327 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
1329 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
1331 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
1333 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
1334 without a peer certificate, leading to a segfault because of an
1335 assumption that peers always have certificates. Be a little more
1336 paranoid. Problem reported by Martin Tscholak.
1338 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
1339 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
1340 NB: ClamAV planning to remove STREAM in "middle of 2010".
1341 CL also introduces -bmalware, various -d+acl logging additions and
1342 more caution in buffer sizes.
1344 PP/09 Implemented reverse_ip expansion operator.
1346 PP/10 Bugzilla 937: provide a "debug" ACL control.
1348 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
1350 PP/12 Bugzilla 973: Implement --version.
1352 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
1354 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
1356 PP/15 Bugzilla 816: support multiple condition rules on Routers.
1358 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
1359 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
1360 ignore trailing whitespace.
1362 JJ/02 prevent non-panic DKIM error from being sent to paniclog
1364 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
1367 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
1368 Notification from Dr Andrew Aitchison.
1370 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
1371 ExtendedDetectionInfo response format.
1372 Notification from John Horne.
1374 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
1377 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
1378 XSL and documented dependency on system catalogs, with examples of how
1381 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
1384 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
1385 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
1386 configuration file which is writeable by the Exim user or group.
1388 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
1389 of configuration files to cover files specified with the -C option if
1390 they are going to be used with root privileges, not just the default
1393 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
1394 option (effectively making it always true).
1396 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
1397 files to be used while preserving root privileges.
1399 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
1400 that rogue child processes cannot use them.
1402 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
1403 run-time user, instead of root.
1405 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
1406 Exim run-time user without dropping privileges.
1408 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
1409 result string, instead of calling string_vformat() twice with the same
1412 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
1413 for other users. Others should always drop root privileges if they use
1414 -C on the command line, even for a whitelisted configure file.
1416 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
1418 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
1424 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
1425 $data_path, and $header_path variables; fixed documentation bugs and
1428 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
1429 exipick to access non-standard spools, including the "frozen" queue
1432 NM/01 Bugzilla 965: Support mysql stored procedures.
1433 Patch from Alain Williams
1435 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
1437 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
1438 Patch from Andreas Metzler
1440 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
1441 Patch from Kirill Miazine
1443 NM/05 Bugzilla 671: Added umask to procmail example.
1445 JJ/03 installed exipick 20100323.0, fixing doc bug
1447 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
1448 directory. Notification and patch from Dan Rosenberg.
1450 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
1452 TK/02 Improve log output when DKIM signing operation fails.
1454 MH/01 Treat the transport option dkim_domain as a colon separated
1455 list, not as a single string, and sign the message with each element,
1456 omitting multiple occurences of the same signer.
1458 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
1459 Bugzilla 985, 986. Patch by Simon Arlott
1461 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
1462 Patch by Simon Arlott
1464 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
1465 MBX locking. Notification from Dan Rosenberg.
1471 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
1473 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
1475 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
1477 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
1479 NM/04 Bugzilla 915: Flag broken perl installation during build.
1485 TK/01 Added patch by Johannes Berg that expands the main option
1486 "spamd_address" if it starts with a dollar sign.
1488 TK/02 Write list of recipients to X-Envelope-Sender header when building
1489 the mbox-format spool file for content scanning (suggested by Jakob
1492 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
1493 (http://www.dcc-servers.net/) support via dccifd. Activated by
1494 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
1496 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
1497 by Mark Daniel Reidel <mr@df.eu>.
1499 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
1500 When building exim an external PCRE library is now needed -
1501 PCRE is a system library on the majority of modern systems.
1502 See entry on PCRE_LIBS in EDITME file.
1504 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
1505 conversation. Added nologin parameter to request.
1506 Patch contributed by Kirill Miazine.
1508 TF/01 Do not log submission mode rewrites if they do not change the address.
1510 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
1512 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
1513 log files in place. Contributed by Roberto Lima.
1515 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
1517 TF/03 Bugzilla 615: When checking the local_parts router precondition
1518 after a local_part_suffix or local_part_prefix option, Exim now
1519 does not use the address's named list lookup cache, since this
1520 contains cached lookups for the whole local part.
1522 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
1523 Robert Millan. Documentation is in experimental-spec.txt.
1525 TF/04 Bugzilla 668: Fix parallel build (make -j).
1527 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
1529 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
1530 Patch provided by Jan Srzednicki.
1532 TF/05 Leading white space used to be stripped from $spam_report which
1533 wrecked the formatting. Now it is preserved.
1535 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
1536 that they are available at delivery time.
1538 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
1540 TF/08 TLS error reporting now respects the incoming_interface and
1541 incoming_port log selectors.
1543 TF/09 Produce a more useful error message if an SMTP transport's hosts
1544 setting expands to an empty string.
1546 NM/06 Bugzilla 744: EXPN did not work under TLS.
1547 Patch provided by Phil Pennock.
1549 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
1550 Patch provided by Richard Godbee.
1552 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
1553 acl_smtp_notquit, added index entry.
1555 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
1556 Patch provided by Eugene Bujak.
1558 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
1559 accept(). Patch provided by Maxim Dounin.
1561 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
1562 Patch provided by Phil Pennock.
1564 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
1566 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
1567 Patch provided by Brad "anomie" Jorsch.
1569 NM/14 Bugzilla 622: Added timeout setting to transport filter.
1570 Patch provided by Dean Brooks.
1572 TK/05 Add native DKIM support (does not depend on external libraries).
1574 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
1575 Patch provided by Graeme Fowler.
1577 NM/16 Bugzilla 851: Documentation example syntax fix.
1579 NM/17 Changed NOTICE file to remove references to embedded PCRE.
1581 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
1584 NM/19 Bugzilla 745: TLS version reporting.
1585 Patch provided by Phil Pennock.
1587 NM/20 Bugzilla 167: bool: condition support.
1588 Patch provided by Phil Pennock.
1590 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
1591 clients. Patch provided by Phil Pennock.
1593 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
1594 Patch provided by Brad "anomie" Jorsch.
1596 NM/23 Bugzilla 687: Fix misparses in eximstats.
1597 Patch provided by Heiko Schlittermann.
1599 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
1600 Patch provided by Heiko Schlittermann.
1602 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
1603 plus update to original patch.
1605 NM/26 Bugzilla 799: Documentation correction for ratelimit.
1607 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
1608 Patch provided by David Brownlee.
1610 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
1612 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
1614 NM/30 Bugzilla 888: TLS documentation bugfixes.
1616 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
1618 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
1619 Unlike the original bugzilla I have changed all shell scripts in src tree.
1621 NM/33 Bugzilla 898: Transport filter timeout fix.
1622 Patch by Todd Rinaldo.
1624 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
1625 Patch by Serge Demonchaux.
1627 NM/35 Bugzilla 39: Base64 decode bug fixes.
1628 Patch by Jakob Hirsch.
1630 NM/36 Bugzilla 909: Correct connect() call in dcc code.
1632 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
1634 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
1636 NM/39 Bugzilla 911: Fixed MakeLinks build script.
1642 TK/01 Add preliminary DKIM support. Currently requires a forked version of
1643 ALT-N's libdkim that I have put here:
1644 http://duncanthrax.net/exim-experimental/
1646 Note to Michael Haardt: I had to rename some vars in sieve.c. They
1647 were called 'true' and it seems that C99 defines that as a reserved
1648 keyword to be used with 'bool' variable types. That means you could
1649 not include C99-style headers which use bools without triggering
1650 build errors in sieve.c.
1652 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
1653 as mailq or other aliases. Changed the --help handling significantly
1654 to do whats expected. exim_usage() emits usage/help information.
1656 SC/01 Added the -bylocaldomain option to eximstats.
1658 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
1660 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
1662 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
1668 PH/01 Another patch from the Sieve maintainer.
1670 PH/02 When an IPv6 address is converted to a string for single-key lookup
1671 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
1672 dots are used instead of colons so that keys in lsearch files need not
1673 contain colons. This was done some time before quoting was made available
1674 in lsearch files. However, iplsearch files do require colons in IPv6 keys
1675 (notated using the quote facility) so as to distinguish them from IPv4
1676 keys. This meant that lookups for IP addresses in host lists did not work
1677 for iplsearch lookups.
1679 This has been fixed by arranging for IPv6 addresses to be expressed with
1680 colons if the lookup type is iplsearch. This is not incompatible, because
1681 previously such lookups could never work.
1683 The situation is now rather anomolous, since one *can* have colons in
1684 ordinary lsearch keys. However, making the change in all cases is
1685 incompatible and would probably break a number of configurations.
1687 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
1690 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
1691 conversion specification without a maximum field width, thereby enabling
1692 a rogue spamd server to cause a buffer overflow. While nobody in their
1693 right mind would setup Exim to query an untrusted spamd server, an
1694 attacker that gains access to a server running spamd could potentially
1695 exploit this vulnerability to run arbitrary code as the Exim user.
1697 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
1698 $primary_hostname instead of what libspf2 thinks the hosts name is.
1700 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
1701 a directory entry by the name of the lookup key. Previously, if a
1702 symlink pointed to a non-existing file or a file in a directory that
1703 Exim lacked permissions to read, a lookup for a key matching that
1704 symlink would fail. Now it is enough that a matching directory entry
1705 exists, symlink or not. (Bugzilla 503.)
1707 PH/03 The body_linecount and body_zerocount variables are now exported in the
1710 PH/04 Added the $dnslist_matched variable.
1712 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
1713 This means they are set thereafter only if the connection becomes
1716 PH/06 Added the client_condition to authenticators so that some can be skipped
1717 by clients under certain conditions.
1719 PH/07 The error message for a badly-placed control=no_multiline_responses left
1720 "_responses" off the end of the name.
1722 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
1724 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
1725 (without spaces) instead of just copying the configuration text.
1727 PH/10 Added the /noupdate option to the ratelimit ACL condition.
1729 PH/11 Added $max_received_linelength.
1731 PH/12 Added +ignore_defer and +include_defer to host lists.
1733 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
1734 way in which PCRE > 7.0 is built.
1736 PH/14 Implemented queue_only_load_latch.
1738 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
1739 MAIL command. The effect was to mangle the value on 64-bit systems.
1741 PH/16 Another patch from the Sieve maintainer.
1743 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
1745 PH/18 If a system quota error occurred while trying to create the file for
1746 a maildir delivery, the message "Mailbox is full" was not appended to the
1747 bounce if the delivery eventually timed out. Change 4.67/27 below applied
1748 only to a quota excession during the actual writing of the file.
1750 PH/19 It seems that peer DN values may contain newlines (and other non-printing
1751 characters?) which causes problems in log lines. The DN values are now
1752 passed through string_printing() before being added to log lines.
1754 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
1755 and InterBase are left for another time.)
1757 PH/21 Added message_body_newlines option.
1759 PH/22 Guard against possible overflow in moan_check_errorcopy().
1761 PH/23 POSIX allows open() to be a macro; guard against that.
1763 PH/24 If the recipient of an error message contained an @ in the local part
1764 (suitably quoted, of course), incorrect values were put in $domain and
1765 $local_part during the evaluation of errors_copy.
1771 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
1772 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
1775 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
1776 issue a MAIL command.
1778 PH/02 In an ACL statement such as
1780 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
1782 if a client was not listed at all, or was listed with a value other than
1783 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
1784 the condition was not true (as it should be), so access was not denied.
1785 The bug was that the ! inversion was incorrectly passed on to the second
1786 item. This has been fixed.
1788 PH/03 Added additional dnslists conditions == and =& which are different from
1789 = and & when the dns lookup returns more than one IP address.
1791 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
1792 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
1794 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
1795 FSYNC, which compiles an option called disable_fsync that allows for
1796 bypassing fsync(). The documentation is heavily laced with warnings.
1798 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
1800 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
1801 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
1802 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
1803 including adding "make clean"; (3) Added -fPIC when compiling the test
1804 dynamically loaded module, to get rid of a warning.
1806 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
1807 message fails, move_frozen_messages = true and ignore_bounce_errors_after
1808 = 0s. The bug is otherwise harmless.
1810 PH/07 There was a bug in the dovecot authenticator such that the value of
1811 $auth1 could be overwritten, and so not correctly preserved, after a
1812 successful authentication. This usually meant that the value preserved by
1813 the server_setid option was incorrect.
1815 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
1817 PH/09 Installed PCRE release 7.0.
1819 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
1820 run for batched SMTP input. It is now run at the start of every message
1821 in the batch. While fixing this I discovered that the process information
1822 (output by running exiwhat) was not always getting set for -bs and -bS
1823 input. This is fixed, and it now also says "batched" for BSMTP.
1825 PH/11 Added control=no_pipelining.
1827 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
1828 patch, slightly modified), and move the expansion of helo_data till after
1829 the connection is made in the smtp transport (so it can use these
1832 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
1834 PH/14 Added log_selector = +pid.
1836 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
1838 PH/16 Add ${if forany and ${if forall.
1840 PH/17 Added dsn_from option to vary the From: line in DSNs.
1842 PH/18 Flush SMTP output before performing a callout, unless control =
1843 no_callout_flush is set.
1845 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
1846 was true (the default) a successful delivery failed to delete the retry
1847 item, thus causing premature timeout of the address. The bug is now
1850 PH/20 Added hosts_avoid_pipelining to the smtp transport.
1852 PH/21 Long custom messages for fakedefer and fakereject are now split up
1853 into multiline reponses in the same way that messages for "deny" and
1854 other ACL rejections are.
1856 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
1857 with slight modification.
1859 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
1860 draft, changing the syntax and factoring some duplicate code".
1862 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
1863 for deliveries of the second and subsequent messages over the same SMTP
1866 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
1867 ${reduce, with only minor "tidies".
1869 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
1871 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
1872 expansion side effects.
1874 PH/27 When a message times out after an over-quota error from an Exim-imposed
1875 quota, the bounce message says "mailbox is full". This message was not
1876 being given when it was a system quota that was exceeded. It now should
1879 MH/03 Made $recipients available in local_scan(). local_scan() already has
1880 better access to the recipient list through recipients_list[], but
1881 $recipients can be useful in postmaster-provided expansion strings.
1883 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
1884 in the case of a MAIL command with additional options following the
1885 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
1886 were accidentally chopped off.
1888 PH/29 SMTP synchronization checks are implemented when a command is read -
1889 there is a check that no more input is waiting when there shouldn't be
1890 any. However, for some commands, a delay in an ACL can mean that it is
1891 some time before the response is written. In this time, more input might
1892 arrive, invalidly. So now there are extra checks after an ACL has run for
1893 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
1894 pipelining has not been advertised.
1896 PH/30 MH's patch to allow iscntrl() characters to be list separators.
1898 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
1899 returned in the SMTP response when smtp_return_error_details was false.
1900 This has been fixed.
1902 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
1903 instead of the C I/O that was originally supplied, because problems were
1904 reported on Solaris.
1906 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
1907 Exim which did not show up earlier: it was assuming that a call to
1908 SSL_CTX_set_info_callback() might give an error value. In fact, there is
1909 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
1910 was a macro that became an assignment, so it seemed to work. This has
1911 changed to a proper function call with a void return, hence the compile
1912 error. Exim's code has been fixed.
1914 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
1917 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
1919 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
1921 PH/37 If a message is not accepted after it has had an id assigned (e.g.
1922 because it turns out to be too big or there is a timeout) there is no
1923 "Completed" line in the log. When some messages of this type were
1924 selected by exigrep, they were listed as "not completed". Others were
1925 picked up by some special patterns. I have improved the selection
1926 criteria to be more general.
1928 PH/38 The host_find_failed option in the manualroute router can now be set
1929 to "ignore", to completely ignore a host whose IP address cannot be
1930 found. If all hosts are ignored, the behaviour is controlled by the new
1931 host_all_ignored option.
1933 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
1934 homing or because of multiple MX records with /mx) generated more than
1935 one IP address, and the following item turned out to be the local host,
1936 all the secondary addresses of the first item were incorrectly removed
1937 from the list, along with the local host and any following hosts (which
1938 is what is supposed to happen).
1940 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
1941 whoever called Exim into the -H file. In the case of the daemon it was
1942 behaving confusingly. When first started, it used values for whoever
1943 started the daemon, but after a SIGHUP it used the Exim user (because it
1944 calls itself on a restart). I have changed the code so that it now always
1947 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
1948 message are rejected with the same error (e.g. no authentication or bad
1949 sender address), and a DATA command is nevertheless sent (as can happen
1950 with PIPELINING or a stupid MUA), the error message that was given to the
1951 RCPT commands is included in the rejection of the DATA command. This is
1952 intended to be helpful for MUAs that show only the final error to their
1955 PH/42 Another patch from the Sieve maintainer.
1957 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
1958 Eximstats - Fixed some broken HTML links and added missing column headers
1960 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
1961 columns for Rejects, Temp Rejects, Ham, and Spam rows.
1963 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
1965 PH/43 Yet another patch from the Sieve maintainer.
1967 PH/44 I found a way to check for a TCP/IP connection going away before sending
1968 the response to the final '.' that terminates a message, but only in the
1969 case where the client has not sent further data following the '.'
1970 (unfortunately, this is allowed). However, in many cases there won't be
1971 any further data because there won't be any more messages to send. A call
1972 to select() can be used: if it shows that the input is "ready", there is
1973 either input waiting, or the socket has been closed. An attempt to read
1974 the next input character can distinguish the two cases. Previously, Exim
1975 would have sent an OK response which the client would never have see.
1976 This could lead to message repetition. This fix should cure that, at
1977 least in a lot of common cases.
1979 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
1980 advertised in response to EHLO.
1986 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
1987 fixed by 4.65/MH/01 (is this a record?) are fixed:
1989 (i) An empty string was always treated as zero by the numeric comparison
1990 operators. This behaviour has been restored.
1992 (ii) It is documented that the numeric comparison operators always treat
1993 their arguments as decimal numbers. This was broken in that numbers
1994 starting with 0 were being interpreted as octal.
1996 While fixing these problems I realized that there was another issue that
1997 hadn't been noticed. Values of message_size_limit (both the global option
1998 and the transport option) were treated as octal if they started with 0.
1999 The documentation was vague. These values are now always treated as
2000 decimal, and I will make that clear in the documentation.
2006 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
2007 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
2010 MH/01 Don't check that the operands of numeric comparison operators are
2011 integers when their expansion is in "skipping" mode (fixes bug
2012 introduced by 4.64-PH/07).
2014 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
2015 child addresses, Exim now panics and dies. Previously, because the count
2016 is held in a short int, deliveries were likely to be lost. As such a
2017 large number of recipients for a single message is ridiculous
2018 (performance will be very, very poor), I have chosen to impose a limit
2019 rather than extend the field.
2025 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
2026 leftover -K file (the existence of which was triggered by #402).
2027 While we were at it, introduced process PID as part of the -K
2028 filename. This should rule out race conditions when creating
2031 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
2032 processing considerably. Previous code took too long for large mails,
2033 triggering a timeout which in turn triggers #401.
2035 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
2036 in the DK code in transports.c. sendfile() is not really portable,
2037 hence the _LINUX specificness.
2039 TF/01 In the add_headers option to the mail command in an Exim filter,
2040 there was a bug that Exim would claim a syntax error in any
2041 header after the first one which had an odd number of characters
2044 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
2045 callout verification, Exim cached a "reject" for the entire domain. This
2046 is correct for most verifications, but it is not correct for a recipient
2047 verification with use_sender or use_postmaster set, because in that case
2048 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
2049 case of MAIL FROM:<> rejection from other early rejections (e.g.
2050 rejection of HELO). When verifying a recipient using a non-null MAIL
2051 address, the cache is ignored if it shows MAIL FROM:<> rejection.
2052 Whatever the result of the callout, the value of the domain cache is
2053 left unchanged (for any other kind of callout, getting as far as trying
2054 RCPT means that the domain itself is ok).
2056 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
2059 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
2060 manifest itself as EPIPE rather than ECONNECT. When tidying away a
2061 session, the daemon ignores ECONNECT errors and logs others; it now
2062 ignores EPIPE as well.
2064 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
2065 (quoted-printable decoding).
2067 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
2068 later the small subsequent patch to fix an introduced bug.
2070 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
2072 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
2074 PH/08 An error is now given if message_size_limit is specified negative.
2076 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
2077 to be given (somewhat) arbitrary names.
2079 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
2082 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
2083 miscellaneous code fixes
2085 PH/10 Added the log_reject_target ACL modifier to specify where to log
2088 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
2089 hostname. This is wrong, because it relates to the incoming message (and
2090 probably the interface on which it is arriving) and not to the outgoing
2091 callout (which could be using a different interface). This has been
2092 changed to use the value of the helo_data option from the smtp transport
2093 instead - this is what is used when a message is actually being sent. If
2094 there is no remote transport (possible with a router that sets up host
2095 addresses), $smtp_active_hostname is used.
2097 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
2098 tweaks were necessary in order to get it to work (see also 21 below):
2099 (a) The code assumed that strncpy() returns a negative number on buffer
2100 overflow, which isn't the case. Replaced with Exim's string_format()
2102 (b) There were several signed/unsigned issues. I just did the minimum
2103 hacking in of casts. There is scope for a larger refactoring.
2104 (c) The code used strcasecmp() which is not a standard C function.
2105 Replaced with Exim's strcmpic() function.
2106 (d) The code set only $1; it now sets $auth1 as well.
2107 (e) A simple test gave the error "authentication client didn't specify
2108 service in request". It would seem that Dovecot has changed its
2109 interface. Fortunately there's a specification; I followed it and
2110 changed what the client sends and it appears to be working now.
2112 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
2115 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
2116 address A is aliased to B and C, where B exists and C does not. Without
2117 -v the output is "A verified" because verification stops after a
2118 successful redirection if more than one address is generated. However,
2119 with -v the child addresses are also verified. Exim was outputting "A
2120 failed to verify" and then showing the successful verification for C,
2121 with its parentage. It now outputs "B failed to verify", showing B's
2122 parentage before showing the successful verification of C.
2124 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
2125 look up a TXT record in a specific list after matching in a combined
2128 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
2129 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
2130 they consult the DNS. I had assumed they would set it the way they
2131 wanted; and indeed my experiments on Linux seem to show that in some
2132 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
2133 To be on the safe side, however, I have now made the interface to
2134 host_find_byname() similar to host_find_bydns(), with an argument
2135 containing the DNS resolver options. The host_find_byname() function now
2136 sets these options at its start, just as host_find_bydns() does. The smtp
2137 transport options dns_qualify_single and dns_search_parents are passed to
2138 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
2139 of host_find_byname() use the default settings of RES_DEFNAMES
2140 (qualify_single) but not RES_DNSRCH (search_parents).
2142 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
2143 spool_read_header() do less string testing, by means of a preliminary
2144 switch on the second character of optional "-foo" lines. (This is
2145 overdue, caused by the large number of possibilities that now exist.
2146 Originally there were few.) While I was there, I also converted the
2147 str(n)cmp tests so they don't re-test the leading "-" and the first
2148 character, in the hope this might squeeze out yet more improvement.
2150 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
2151 flag allowing group syntax was set by the header_syntax check but not
2152 turned off, possible causing trouble later; (2) The flag was not being
2153 set at all for the header_verify test, causing "group"-style headers to
2154 be rejected. I have now set it in this case, and also caused header_
2155 verify to ignore an empty address taken from a group. While doing this, I
2156 came across some other cases where the code for allowing group syntax
2157 while scanning a header line wasn't quite right (mostly, not resetting
2158 the flag correctly in the right place). These bugs could have caused
2159 trouble for malformed header lines. I hope it is now all correct.
2161 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
2162 with the "reply" argument non-NULL. The code, however (which originally
2163 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
2164 but it didn't always do it. This confused somebody who was copying the
2165 code for some other use. I have removed all the tests.
2167 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
2168 feature that was used to support insecure browsers during the U.S. crypto
2169 embargo. It requires special client support, and Exim is probably the
2170 only MTA that supported it -- and would never use it because real RSA is
2171 always available. This code has been removed, because it had the bad
2172 effect of slowing Exim down by computing (never used) parameters for the
2173 RSA_EXPORT functionality.
2175 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
2176 authenticator to fail if there's a tab character in the incoming data
2177 (there should never be unless someone is messing about, as it's supposed
2178 to be base64-encoded). Also added, on Timo's advice, the "secured" option
2179 if the connection is using TLS or if the remote IP is the same as the
2180 local IP, and the "valid-client-cert option" if a client certificate has
2183 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
2184 authenticators. This can be used for authorization after authentication
2185 succeeds. (In the case of plaintext, it servers for both authentication
2188 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
2189 if any retry times were supplied.
2191 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
2192 connection, where there is no client IP address to check. In this
2193 situation, the verify now always succeeds.
2195 PH/25 Applied John Jetmore's -Mset patch.
2197 PH/26 Added -bem to be like -Mset, but loading a message from a file.
2199 PH/27 In a string expansion for a processed (not raw) header when multiple
2200 headers of the same name were present, leading whitespace was being
2201 removed from all of them, but trailing whitespace was being removed only
2202 from the last one. Now trailing whitespace is removed from each header
2203 before concatenation. Completely empty headers in a concatenation (as
2204 before) are ignored.
2206 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
2207 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
2209 PH/29 [Removed. This was a change that I later backed out, and forgot to
2210 correct the ChangeLog entry (that I had efficiently created) before
2211 committing the later change.]
2213 PH/30 Exim was sometimes attempting to deliver messages that had suffered
2214 address errors (4xx response to RCPT) over the same connection as other
2215 messages routed to the same hosts. Such deliveries are always "forced",
2216 so retry times are not inspected. This resulted in far too many retries
2217 for the affected addresses. The effect occurred only when there were more
2218 hosts than the hosts_max_try setting in the smtp transport when it had
2219 the 4xx errors. Those hosts that it had tried were not added to the list
2220 of hosts for which the message was waiting, so if all were tried, there
2221 was no problem. Two fixes have been applied:
2223 (i) If there are any address or message errors in an SMTP delivery, none
2224 of the hosts (tried or untried) are now added to the list of hosts
2225 for which the message is waiting, so the message should not be a
2226 candidate for sending over the same connection that was used for a
2227 successful delivery of some other message. This seems entirely
2228 reasonable: after all the message is NOT "waiting for some host".
2229 This is so "obvious" that I'm not sure why it wasn't done
2230 previously. Hope I haven't missed anything, but it can't do any
2231 harm, as the worst effect is to miss an optimization.
2233 (ii) If, despite (i), such a delivery is accidentally attempted, the
2234 routing retry time is respected, so at least it doesn't keep
2235 hammering the server.
2237 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
2238 in ${readsocket because some servers need this prod.
2240 PH/32 Added some extra debug output when updating a wait-xxx database.
2242 PH/33 The hint "could be header name not terminated by colon", which has been
2243 given for certain expansion errors for a long time, was not being given
2244 for the ${if def:h_colon_omitted{... case.
2246 PH/34 The spec says: "With one important exception, whenever a domain list is
2247 being scanned, $domain contains the subject domain." There was at least
2248 one case where this was not true.
2250 PH/35 The error "getsockname() failed: connection reset by peer" was being
2251 written to the panic log as well as the main log, but it isn't really
2252 panic-worthy as it just means the connection died rather early on. I have
2253 removed the panic log writing for the ECONNRESET error when getsockname()
2256 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
2257 runs only) independently of the message's sender address. This meant
2258 that, if the 4xx error was in fact related to the sender, a different
2259 message to the same recipient with a different sender could confuse
2260 things. In particualar, this can happen when sending to a greylisting
2261 server, but other circumstances could also provoke similar problems.
2262 I have changed the default so that the retry time for these errors is now
2263 based a combination of the sender and recipient addresses. This change
2264 can be overridden by setting address_retry_include_sender=false in the
2267 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
2268 remote server are returned as part of bounce messages. This was not
2269 happening for LMTP over a pipe (the lmtp transport), but now it is the
2270 same for both kinds of LMTP.
2272 PH/38 Despite being documented as not happening, Exim was rewriting addresses
2273 in header lines that were in fact CNAMEs. This is no longer the case.
2275 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
2276 and queue runs started by the daemon processed all messages. This has
2277 been fixed so that -R and -S can now usefully be given with -q<time>.
2279 PH/40 Import PCRE release 6.7 (fixes some bugs).
2281 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
2283 PH/42 Give an error if -q is specified more than once.
2285 PH/43 Renamed the variables $interface_address and $interface_port as
2286 $received_ip_address and $received_port, to make it clear that these
2287 values apply to message reception, and not to the outgoing interface when
2288 a message is delivered. (The old names remain recognized, of course.)
2290 PH/44 There was no timeout on the connect() call when using a Unix domain
2291 socket in the ${readsocket expansion. There now is.
2293 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
2294 be meaningful with "accept".
2296 SC/01 Eximstats V1.43
2297 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
2299 SC/02 Eximstats V1.44
2300 Use a glob alias rather than an array ref in the generated
2301 parser. This improves both readability and performance.
2303 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
2304 Collect SpamAssassin and rejection statistics.
2305 Don't display local sender or destination tables unless
2306 there is data to show.
2307 Added average volumes into the top table text output.
2309 SC/04 Eximstats V1.46
2310 Collect data on the number of addresses (recipients)
2311 as well as the number of messages.
2313 SC/05 Eximstats V1.47
2314 Added 'Message too big' to the list of mail rejection
2315 reasons (thanks to Marco Gaiarin).
2317 SC/06 Eximstats V1.48
2318 Mainlog lines which have GMT offsets and are too short to
2319 have a flag are now skipped.
2321 SC/07 Eximstats V1.49 (Alain Williams)
2322 Added the -emptyok flag.
2324 SC/08 Eximstats V1.50
2325 Fixes for obtaining the IP address from reject messages.
2327 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
2328 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
2329 whitesspace changes from 4.64-PH/27
2331 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
2334 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
2335 are found, allow negative numbers in numeric criteria)
2337 JJ/06 exipick.20061117.2, added new $message_body_missing variable
2339 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
2340 to match changes made in 4.64-PH/43
2342 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
2344 PH/47 Put in an explicit test for a DNS lookup of an address record where the
2345 "domain" is actually an IP address, and force a failure. This locks out
2346 those revolvers/nameservers that support "A-for-A" lookups, in
2347 contravention of the specifications.
2349 PH/48 When a host name was looked up from an IP address, and the subsequent
2350 forward lookup of the name timed out, the host name was left in
2351 $sender_host_name, contrary to the specification.
2353 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
2354 restricted to single-key lookups, Exim was not diagnosing an error if
2355 * or *@ was used with a query-style lookup.
2357 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
2359 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
2360 long ago, but noone interested enough thought of it. Let's just say that
2361 the "1.1" means that there are some new functions that weren't there at
2362 some point in the past.
2364 PH/51 Error processing for expansion failure of helo_data from an smtp
2365 transport during callout processing was broken.
2367 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
2368 tested/used via the -bh/-bhc/-bs options.
2370 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
2371 bug, fixed in subsequent PCRE releases).
2373 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
2374 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
2376 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
2382 SC/01 Use a glob alias rather than an array ref in eximstats generated
2383 parser. This improves both readability and performance.
2385 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
2386 Don't display local sender or destination tables in eximstats unless
2387 there is data to show.
2388 Added average volumes into the eximstats top table text output.
2390 SC/03 Collect data on the number of addresses (recipients) as well
2391 as the number of messages in eximstats.
2393 TF/01 Correct an error in the documentation for the redirect router. Exim
2394 does (usually) call initgroups() when daemonizing.
2396 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
2397 with consistent privilege compared to when running as a daemon.
2399 TF/03 Note in the spec that $authenticated_id is not set for local
2400 submissions from trusted users.
2402 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
2403 Thanks to Dean Brooks <dean@iglou.com> for the patch.
2405 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
2406 by adding some example configuration directives to the default
2407 configuration file. A little bit of work is required to uncomment the
2408 directives and define how usernames and passwords are checked, but
2409 there is now a framework to start from.
2411 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
2412 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
2413 without this. I don't know how relevant this is to other LDAP libraries.
2415 PH/02 Add the verb name to the "unknown ACL verb" error.
2417 PH/03 Magnus Holmgren's patch for filter_prepend_home.
2419 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
2421 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
2422 directory not expanded when it should be if an expanded home directory
2423 was set for the address (which is overridden by the transport).
2425 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
2428 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
2429 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
2430 because it is too late at that time, and has no effect.
2432 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
2433 security issue with \' (bugzilla #107). I could not use the
2434 PQescapeStringConn() function, because it needs a PGconn value as one of
2437 PH/08 When testing addresses using -bt, indicate those final addresses that
2438 are duplicates that would not cause an additional delivery. At least one
2439 person was confused, thinking that -bt output corresponded to deliveries.
2440 (Suppressing duplicates isn't a good idea as you lose the information
2441 about possibly different redirections that led to the duplicates.)
2443 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
2444 systems where poll() doesn't work, in particular OS X.
2446 PH/10 Added more information to debugging output for retry time not reached.
2448 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
2449 operations in malware.c.
2451 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
2454 PH/13 If write_rejectlog was set false when logging was sent to syslog with
2455 syslog_duplication set false, log lines that would normally be written
2456 both the the main log and to the reject log were not written to syslog at
2459 PH/14 In the default configuration, change the use of "message" in ACL warn
2460 statements to "add_header".
2462 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
2463 not followed by a command (e.g. "seen endif").
2465 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
2466 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
2469 PH/17 Added extra conditions to the default value of delay_warning_condition
2473 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
2474 { match{$h_precedence:}{(?i)bulk|list|junk} } \
2475 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
2478 The Auto-Submitted: and various List- headers are standardised, whereas I
2479 don't think Precedence: ever was.
2481 PH/18 Refactored debugging code in route_finduser() to show more information,
2482 in particular, the error code if getpwnam() issues one.
2484 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
2485 This is apparently needed in addition to the PH/07 change above to avoid
2486 any possible encoding problems.
2488 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
2489 but not after initializing Perl.
2491 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
2492 output them only if debugging. By default they are written stderr,
2493 apparently, which is not desirable.
2495 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
2498 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
2501 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
2503 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
2504 authenticated or an ident call has been made. Suppress the default
2505 values for $authenticated_id and $authenticated_sender (but permit -oMai
2506 and -oMas) when testing with -bh.
2508 PH/24 Re-jigged the order of the tests in the default configuration so that the
2509 tests for valid domains and recipients precede the DNS black list and CSA
2510 tests, on the grounds that those ones are more expensive.
2512 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
2513 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
2514 This bug exists in every version of Exim that I still have, right back to
2517 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
2518 However, an attempt to turn on case-sensitivity in a regex key by
2519 including (?-i) didn't work because the subject string was already
2520 lowercased, and the effects were non-intuitive. It turns out that a
2521 one-line patch can be used to allow (?-i) to work as expected.
2527 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
2528 other effects) broke the use of negated acl sub-conditions.
2530 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
2533 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
2534 "Deny" causes Exim to reject the incoming connection with a 554 error.
2535 Unfortunately, if there is a major crisis, such as a disk failure,
2536 tcp-wrappers gives "deny", whereas what one would like would be some
2537 kind of temporary error. A kludge has been added to help with this.
2538 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
2539 554 error is used if errno is still zero or contains ENOENT (which occurs
2540 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
2543 PH/03 Add -lutil to the default FreeBSD LIBS setting.
2545 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
2546 errors. Otherwise a message that provokes a temporary error (when other
2547 messages do not) can cause a whole host to time out.
2549 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
2550 addresses were routed directly to files or pipes from a redirect router.
2551 File deliveries just didn't batch; pipe deliveries might have suffered
2554 PH/06 A failure to get a lock for a hints database would erroneously always say
2555 "Failed to get write lock", even when it was really a read lock.
2557 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
2558 of 0600. This has been changed to use the value of the lockfile_mode
2559 option (which defaults to 0600).
2561 PH/08 Applied small patch from the Sieve maintainer.
2563 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
2564 folder from quota calculations, a direct delivery into this folder messed
2565 up the contents of the maildirsize file. This was because the regex was
2566 used only to exclude .Trash (or whatever) when the size of the mailbox
2567 was calculated. There was no check that a delivery was happening into an
2568 excluded directory. This bug has been fixed by ignoring all quota
2569 processing for deliveries into excluded directories.
2571 PH/10 Added the maildirfolder_create_regex option to appendfile.
2577 PH/01 The code for finding all the local interface addresses on a FreeBSD
2578 system running IPv6 was broken. This may well have applied to all BSD
2579 systems, as well as to others that have similar system calls. The broken
2580 code found IPv4 interfaces correctly, but gave incorrect values for the
2581 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
2582 that it would not match correctly against @[] and not recognize the IPv6
2585 PH/02 The ipliteral router was not recognizing addresses of the form user@
2586 [ipv6:....] because it didn't know about the "ipv6:" prefix.
2588 PH/03 Added disable_ipv6.
2590 PH/04 Changed $reply_address to use the raw form of the headers instead of the
2591 decoded form, because it is most often used to construct To: headers
2592 lines in autoreplies, and the decoded form may well be syntactically
2593 invalid. However, $reply_address has leading white space removed, and all
2594 newlines turned into spaces so that the autoreply transport does not
2597 PH/05 If group was specified without a user on a router, and no group or user
2598 was specified on a transport, the group from the router was ignored.
2600 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
2601 for visible compile-time settings that can be used to change these
2602 numbers, for those that want even more. Backwards compatibility with old
2603 spool files has been maintained. However, going back to a previous Exim
2604 release will lost any variables that are in spool files.
2606 PH/07 Two small changes when running in the test harness: increase delay when
2607 passing a TCP/IP connection to a new process, in case the original
2608 process has to generate a bounce, and remove special handling of
2609 127.0.0.2 (sic), which is no longer necessary.
2611 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
2612 be the same on different OS.
2614 PH/09 Moved a debug statement in filter processing to avoid a race problem when
2617 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
2618 whether --show-vars was specified or not
2620 JJ/02 exipick: Added support for new ACL variable spool format introduced
2623 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
2624 syntactically invalid From: or Reply-to: line, and a filter used this to
2625 generate an autoreply, and therefore failed to obtain an address for the
2626 autoreply, Exim could try to deliver to a non-existent relative file
2627 name, causing unrelated and misleading errors. What now happens is that
2628 it logs this as a hard delivery error, but does not attempt to create a
2631 PH/11 The exinext utility has a -C option for testing purposes, but although
2632 the given file was scanned by exinext itself; it wasn't being passed on
2633 when Exim was called.
2635 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
2636 an end-of-file indication when reading a command response.
2638 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
2639 compiled. In many other places in Exim, IPv6 addresses are always
2640 recognized, so I have changed this. It also means that IPv4 domain
2641 literals of the form [IPV4:n.n.n.n] are now always recognized.
2643 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
2644 used if the router is not running as root, for example, when verifying at
2645 ACL time, or when using -bh. The debugging output from this situation was
2646 non-existent - all you got was a failure to exec. I have made two
2649 (a) Failures to set uid/gid, the current directory, or a process leader
2650 in a subprocess such as that created by queryprogram now generate
2651 suitable debugging ouput when -d is set.
2653 (b) The queryprogram router detects when it is not running as root,
2654 outputs suitable debugging information if -d is set, and then runs
2655 the subprocess without attempting to change uid/gid.
2657 PH/15 Minor change to Makefile for building test_host (undocumented testing
2660 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
2661 additional section of a DNS packet that returns MX or SRV records.
2662 Instead, it always explicitly searches for A/AAAA records. This avoids
2663 major problems that occur when a DNS server includes only records of one
2664 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
2665 fixed another bug: if SRV records were looked up and the corresponding
2666 address records were *not* found in the additional section, the port
2667 values from the SRV records were lost.
2669 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
2670 using the correct key (the original address) when searching the retry
2671 rules in order to find which one to use for generating the retry hint.
2673 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
2674 adding the default one. Similarly, if it contains a Reply-To: header, the
2675 errors_reply_to option, if set, is not used.
2677 PH/19 When calculating a retry time, Exim used to measure the "time since
2678 failure" by looking at the "first failed" field in the retry record. Now
2679 it does not use this if it is later than than the arrival time of the
2680 message. Instead it uses the arrival time. This makes for better
2681 behaviour in cases where some deliveries succeed, thus re-setting the
2682 "first failed" field. An example is a quota failure for a huge message
2683 when small messages continue to be delivered. Without this change, the
2684 "time since failure" will always be short, possible causing more frequent
2685 delivery attempts for the huge message than are intended.
2686 [Note: This change was subsequently modified - see PH/04 for 4.62.]
2688 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
2689 $1, $2, $3) because the numerical variables can be reset during some
2690 expansion items (e.g. "match"), thereby losing the authentication data.
2692 PH/21 Make -bV show the size of off_t variables so that the test suite can
2693 decide whether to run tests for quotas > 2G.
2695 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
2696 mailbox_size, and mailbox_filecount in the appendfile transport. If a
2697 filecount value is greater than 2G or if a quota value is greater than 2G
2698 on a system where the size of off_t is not greater than 4, a panic error
2701 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
2702 never match. The debug and -bh output now contains an explicit error
2703 message indicating a malformed IPv4 address or mask.
2705 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
2706 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
2707 PH/23 above applies.
2709 PH/25 Do not write to syslog when running in the test harness. The only
2710 occasion when this arises is a failure to open the main or panic logs
2711 (for which there is an explicit test).
2713 PH/26 Added the /no_tell option to "control=freeze".
2715 PH/27 If a host name lookup failed very early in a connection, for example, if
2716 the IP address matched host_lookup and the reverse lookup yielded a name
2717 that did not have a forward lookup, an error message of the form "no IP
2718 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
2719 could be logged. Now it outputs the IP address instead of "NULL".
2721 PH/28 An enabling patch from MH: add new function child_open_exim2() which
2722 allows the sender and the authenticated sender to be set when
2723 submitting a message from within Exim. Since child_open_exim() is
2724 documented for local_scan(), the new function should be too.
2726 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
2727 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
2728 results in an empty string is now treated as unset.
2730 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
2732 PH/31 Added sender_verify_fail logging option.
2734 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
2735 needed by RFC 822 but not by RFC 2822 was commented out. I have now
2736 tidied the source and removed it altogether.
2738 PH/33 When a queue run was abandoned because the load average was too high, a
2739 log line was always written; now it is written only if the queue_run log
2740 selector is set. In addition, the log line for abandonment now contains
2741 information about the queue run such as the pid. This is always present
2742 in "start" and "stop" lines but was omitted from the "abandon" line.
2744 PH/34 Omit spaces between a header name and the colon in the error message that
2745 is given when verify = headers_syntax fails (if there are lots of them,
2746 the message gets confusing).
2748 PH/35 Change the default for dns_check_names_pattern to allow slashes within
2749 names, as there are now some PTR records that contain slashes. This check
2750 is only to protect against broken name servers that fall over on strange
2751 characters, so the fact that it applies to all lookups doesn't matter.
2753 PH/36 Now that the new test suite is complete, we can remove some of the
2754 special code in Exim that was needed for the old test suite. For example,
2755 sorting DNS records because real resolvers return them in an arbitrary
2756 order. The new test suite's fake resolver always returns records in the
2759 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
2760 bounces) except when queue_only is set, to avoid logging races between
2761 the different processes.
2763 PH/38 Panic-die if .include specifies a non-absolute path.
2765 PH/39 A tweak to the "H" retry rule from its user.
2767 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
2768 a label. They prevented compilation on older perls.
2770 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
2771 a warning to be raised on newish perls.
2773 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
2774 on queue. Changes to match documented behaviour of showing count of
2775 messages matching specified criteria.
2777 PH/40 Changed the default ident timeout from 30s to 5s.
2779 PH/41 Added support for the use of login_cap features, on those BSD systems
2780 that have them, for controlling the resources used by pipe deliveries.
2782 PH/42 The content-scanning code uses fopen() to create files in which to put
2783 message data. Previously it was not paying any attention to the mode of
2784 the files. Exim runs with umask(0) because the rest of the code creates
2785 files with open(), and sets the required mode explicitly. Thus, these
2786 files were ending up world-writeable. This was not a big issue, because,
2787 being within the spool directory, they were not world-accessible. I have
2788 created a function called modefopen, which takes an additional mode
2789 argument. It sets umask(777), creates the file, chmods it to the required
2790 mode, then resets the umask. All the relevant calls to fopen() in the
2791 content scanning code have been changed to use this function.
2793 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
2794 to 24 hours. This avoids potential overflow problems when processing G
2795 and H retry rules. I suspect nobody ever tinkers with this value.
2797 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
2799 PH/45 When the plaintext authenticator is running as a client, the server's
2800 challenges are checked to ensure they are valid base64 strings. By
2801 default, the authentication attempt is cancelled if an invalid string is
2802 received. Setting client_ignore_invalid_base64 true ignores these errors.
2803 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
2804 they are received. Thus, the responses can be made to depend on the
2805 challenges. If an invalid string is ignored, an empty string is placed in
2808 PH/46 Messages that are created by the autoreply transport now contains a
2809 References: header, in accordance with RFCs 2822 and 3834.
2811 PH/47 Added authenticated_sender_force to the smtp transport.
2813 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
2815 PH/49 Installed latest patch from the Sieve maintainer.
2817 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
2818 was also set, the appendfile transport was unnecessarily scanning a
2819 directory of message files (e.g. for maildir delivery) to find the count
2820 of files (along with the size), even though it did not need this
2821 information. It now does the scan only if it needs to find either the
2822 size of the count of files.
2824 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
2826 PH/52 Two bugs concerned with error handling when the smtp transport is
2829 (i) Exim was not creating retry information for temporary errors given
2830 for individual recipients after the DATA command when the smtp transport
2831 was used in LMTP mode. This meant that they could be retried too
2832 frequently, and not timed out correctly.
2834 (ii) Exim was setting the flag that allows error details to be returned
2835 for LMTP errors on RCPT commands, but not for LMTP errors for individual
2836 recipients that were returned after the DATA command.
2838 PH/53 This is related to PH/52, but is more general: for any failing address,
2839 when detailed error information was permitted to be returned to the
2840 sender, but the error was temporary, then after the final timeout, only
2841 "retry timeout exceeded" was returned. Now it returns the full error as
2842 well as "retry timeout exceeded".
2844 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
2845 do this, and (what is worse) MTAs that accept it.
2847 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
2848 will now be deprecated.
2850 PH/56 New os.c-cygwin from the Cygwin maintainer.
2852 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
2853 formats (previously only available in exim formats via -bpr, -bpru,
2854 and -bpra. Now also available in native and exiqgrep formats)
2856 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
2857 with very large, slow to parse queues
2859 JJ/08 exipick: added ! as generic prefix to negate any criteria format
2861 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
2863 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
2864 responses to authentication challenges, though it was showing the
2865 challenges; (ii) I've removed the CR characters from the debug output for
2868 PH/58 Allow for the insertion of a newline as well as a space when a string
2869 is turned into more than one encoded-word during RFC 2047 encoding. The
2870 Sieve code now uses this.
2872 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
2873 data_4xx, lost_connection, tls_required.
2875 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
2876 message was being sent as an SMTP response.
2878 PH/61 Add -l and -k options to exicyclog.
2880 PH/62 When verifying, if an address was redirected to one new address, so that
2881 verification continued, and the new address failed or deferred after
2882 having set something in $address_data, the value of $address_data was not
2883 passed back to the ACL. This was different to the case when no
2884 redirection occurred. The value is now passed back in both cases.
2886 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
2887 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
2888 use login_cap.h, so on its own it isn't the distinguishing feature. The
2889 new name refers directly to the setclassresources() function.
2891 PH/65 Added configuration files for NetBSD3.
2893 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
2895 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
2896 is preferred over IPv4.
2898 PH/68 The bounce_return_message and bounce_return_body options were not being
2899 honoured for bounces generated during the reception of non-SMTP messages.
2900 In particular, this applied to messages rejected by the ACL. This bug has
2901 been fixed. However, if bounce_return_message is true and bounce_return_
2902 body is false, the headers that are returned for a non-SMTP message
2903 include only those that have been read before the error was detected.
2904 (In the case of an ACL rejection, they have all been read.)
2906 PH/69 The HTML version of the specification is now built in a directory called
2907 spec_html instead of spec.html, because the latter looks like a path with
2908 a MIME-type, and this confuses some software.
2910 PH/70 Catch two compiler warnings in sieve.c.
2912 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
2913 function verify_get_ident() calls ip_connect() to connect a socket, but
2914 if the "connect()" function timed out, ip_connect() used to close the
2915 socket. However, verify_get_ident() also closes the socket later, and in
2916 between Exim writes to the log, which may get opened at this point. When
2917 the socket was closed in ip_connect(), the log could get the same file
2918 descriptor number as the socket. This naturally causes chaos. The fix is
2919 not to close the socket in ip_connect(); the socket should be closed by
2920 the function that creates it. There was only one place in the code where
2921 this was missing, in the iplookup router, which I don't think anybody now
2922 uses, but I've fixed it anyway.
2924 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
2925 well as to direct DNS lookups. Otherwise the handling of names in host
2926 lists is inconsistent and therefore confusing.
2932 PH/01 Two changes to the default runtime configuration:
2934 (1) Move the checks for relay_from_hosts and authenticated clients from
2935 after to before the (commented out) DNS black list checks.
2937 (2) Add control=submission to the relay_from_hosts and authenticated
2938 clients checks, on the grounds that messages accepted by these
2939 statements are most likely to be submissions.
2941 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
2943 (1) Generate an error if the third argument for the ${prvs expansion is
2946 (2) Treat a missing third argument of ${prvscheck as if it were an empty
2949 (3) Reset the variables that are obtained from the first argument of
2950 ${prvscheck and used in the second argument before leaving the code,
2951 because their memory is reclaimed, so using them afterwards may do
2954 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
2955 one (it's much easier than Tom thought :-).
2957 (5) Because of (4), we can now allow for the use of $prvscheck_result
2958 inside the third argument.
2960 PH/03 For some reason, the default setting of PATH when running a command from
2961 a pipe transport was just "/usr/bin". I have changed it to
2964 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
2965 anything to be listed in the output from -bV.
2967 PH/05 When a filter generated an autoreply, the entire To: header line was
2968 quoted in the delivery log line, like this:
2970 => >A.N.Other <ano@some.domain> <original@ddress> ...
2972 This has been changed so that it extracts the operative address. There
2973 may be more than one such address. If so, they are comma-separated, like
2976 => >ano@some.domain,ona@other.domain <original@ddress> ...
2978 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
2979 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
2980 not being looked up in the rDNS to get a host name, Exim was showing the
2981 IP address twice in Received: lines, even though the IP addresses were
2982 identical. For example:
2984 Received: from [1.2.3.4] (helo=[1.2.3.4])
2986 However, if the real host name was known, it was omitting the HELO data
2987 if it matched the actual IP address. This has been tidied up so that it
2988 doesn't show the same IP address twice.
2990 PH/07 When both +timestamp and +memory debugging was on, the value given by
2991 $tod_xxx expansions could be wrong, because the tod_stamp() function was
2992 called by the debug printing, thereby overwriting the timestamp buffer.
2993 Debugging no longer uses the tod_stamp() function when +timestamp is set.
2995 PH/08 When the original message was included in an autoreply transport, it
2996 always said "this is a copy of the message, including all the headers",
2997 even if body_only or headers_only was set. It now gives an appropriate
3000 PH/09 Applied a patch from the Sieve maintainer which:
3002 o fixes some comments
3003 o adds the (disabled) notify extension core
3004 o adds some debug output for the result of if/elsif tests
3005 o points to the current vacation draft in the documentation
3006 and documents the missing references header update
3010 o fixes a bug in processing the envelope test (when testing
3011 multiple envelope elements, the last element determinted the
3014 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
3015 Electronic Mail") by including:
3017 Auto-submitted: auto-generated
3019 in the messages that it generates (bounce messages and others, such as
3020 warnings). In the case of bounce messages for non-SMTP mesages, there was
3021 also a typo: it was using "Auto_submitted" (underscore instead of
3022 hyphen). Since every message generated by Exim is necessarily in response
3023 to another message, thes have all been changed to:
3025 Auto-Submitted: auto-replied
3027 in accordance with these statements in the RFC:
3029 The auto-replied keyword:
3031 - SHOULD be used on messages sent in direct response to another
3032 message by an automatic process,
3034 - MUST NOT be used on manually-generated messages,
3036 - MAY be used on Delivery Status Notifications (DSNs) and Message
3037 Disposition Notifications (MDNs),
3039 - MUST NOT be used on messages generated by automatic or periodic
3040 processes, except for messages which are automatic responses to
3043 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
3044 to the default Received: header definition.
3046 PH/12 Added log selector acl_warn_skipped (default on).
3048 PH/13 After a successful wildlsearch lookup, discard the values of numeric
3049 variables because (a) they are in the wrong storage pool and (b) even if
3050 they were copied, it wouldn't work properly because of the caching.
3052 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
3053 checking when decoding. Apparently there are clients that generate
3054 overlong encoded strings. Why am I not surprised?
3056 PH/15 If the first argument of "${if match_address" was not empty, but did not
3057 contain an "@" character, Exim crashed. Now it writes a panic log message
3058 and treats the condition as false.
3060 PH/16 In autoreply, treat an empty string for "once" the same as unset.
3062 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
3063 extension "envelope-auth". The code is finished and in agreement with
3064 other implementations, but there is no documentation so far and in fact,
3065 nobody wrote the draft yet. This extension is currently #undef'ed, thus
3066 not changing the active code.
3068 Print executed "if" and "elsif" statements when debugging is used. This
3069 helps a great deal to understand what a filter does.
3071 Document more things not specified clearly in RFC3028. I had all this
3072 sorted out, when out of a sudden new issues came to my mind. Oops."
3074 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
3077 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
3078 canonical form (as documented). However, after a host name lookup from
3079 the IP address, check_host() was doing a simple string comparison with
3080 addresses acquired from the DNS when checking that the found name did
3081 have the original IP as one of its addresses. Since any found IPv6
3082 addresses are likely to be in abbreviated form, the comparison could
3083 fail. Luckily, there already exists a function for doing the comparison
3084 by converting both addresses to binary, so now that is used instead of
3085 the text comparison.
3087 PH/20 There was another similar case to PH/19, when a complete host name was
3088 given in a host list; looking up its IP address could give an abbreviated
3089 form, whereas the current host's name might or might not be abbreviated.
3090 The same fix has been applied.
3096 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
3097 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
3100 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
3101 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
3103 PH/03 Typo: missing ".o" in src/pcre/Makefile.
3105 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
3106 header line, restrict the check to what is listed in RFCs 2369 and 2929.
3107 Also, for "Auto-Submitted", treat anything other than "no" as
3108 non-personal, in accordance with RFC 3834. (Previously it treated
3109 anything starting "auto-" as non-personal.)
3111 TF/01 The control=submission/name=... option had a problem with syntax
3112 errors if the name included a slash character. The /name= option
3113 now slurps the rest of the string, so it can include any characters
3114 but it must come last in the list of options (after /sender_retain
3117 PH/05 Some modifications to the interface to the fake nameserver for the new
3125 TK/01 Added the "success_on_redirect" address verification option. See
3126 NewStuff for rationale and an example.
3128 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
3130 PH/02 Patch to exigrep to allow it to work on syslog lines.
3132 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
3133 fread() to skip over the body file's header line, because in Cygwin the
3134 header line is locked and is inaccessible.
3136 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
3137 co-exist for some time) to make it clear that it is the Exim ID that is
3138 referenced, not the Message-ID: header line.
3140 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
3141 string_format() function, because snprintf() does not exist on all
3144 PH/06 The use of forbid_filter_existstest now also locks out the use of the
3145 ${stat: expansion item.
3147 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
3148 protocol synchronization error", to keep the pedants happy.
3150 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
3151 well as for IRIX systems, when gcc is being used. See the host.c source
3154 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
3156 PH/10 Named domain lists were not working if used in a queue_smtp_domains
3159 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
3160 transport and to the smtp transport in LMTP mode.
3162 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
3164 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
3165 run a filter in a subprocess. This could lead to confusion in subsequent
3166 lookups in the parent process. There should also be a search_tidyup() at
3167 the end of the subprocess.
3169 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
3170 only if the host matched helo_try_verify_hosts, which caused the
3171 verification to occur when the EHLO/HELO command was issued. The ACL just
3172 tested the remembered result. Now, if a previous verification attempt has
3173 not happened, "verify = helo" does it there and then.
3175 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
3177 TK/03 Fix log output including CR from clamd.
3179 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
3180 exist provoked a memory error which could cause a segfault.
3182 PH/15 Installed PCRE 6.2
3184 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
3186 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
3187 of the problem. Specifically, suggested +O2 rather than +O1 for the
3190 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
3192 PH/20 If a delivery was routed to a non-standard port by means of an SRV
3193 record, the port was not correctly logged when the outgoing_port log
3194 selector was set (it logged the transort's default port).
3196 PH/21 Added support for host-specific ports to manualroute, queryprogram,
3197 fallback_hosts, and "hosts" in the smtp transport.
3199 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
3200 host errors such as "Connection refused".
3202 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
3203 authentication with radiusclient 0.4.9:
3205 - Error returned from rc_read_config was caught wrongly
3206 - Username/password not passed on to radius server due to wrong length.
3208 The presumption is that some radiusclient API changes for 4.51/PH/17
3209 were not taken care of correctly. The code is still untested by me (my
3210 Linux distribution still has 0.3.2 of radiusclient), but it was
3211 contributed by a Radius user.
3213 PH/24 When doing a callout, the value of $domain wasn't set correctly when
3214 expanding the "port" option of the smtp transport.
3216 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
3217 while reading a MIME header. Thanks to Tom Hughes for a patch.
3219 PH/24 Include config.h inside local_scan.h so that configuration settings are
3222 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
3223 that in an ACL for RCPT (for example), you can examine exactly what was
3226 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
3227 commands, but it was not correctly comparing the address with the actual
3228 client host address. Thus, it would show the EHLO address in Received:
3229 header lines when this was not necessary.
3231 PH/27 Added the % operator to ${eval:}.
3233 PH/28 Exim tries to create and chdir to its spool directory when it starts;
3234 it should be ignoring failures (because with -C, for example, it has lost
3235 privilege). It wasn't ignoring creation failures other than "already
3238 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
3241 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
3242 because an input line was too long, either on its own, or by virtue of
3243 too many continuations, the temporary file was not being removed, and the
3244 return code was incorrect.
3246 PH/31 Missing "BOOL" in function definition in filtertest.c.
3248 PH/32 Applied Sieve patches from the maintainer.
3250 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
3252 PH/33 Added "verify = not_blind".
3254 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
3255 Local/Makefile (with some defaults set). These are used in built scripts
3256 such as exicyclog, but they have never been used in the exim_install
3257 script (though there are many overriding facilities there). I have
3258 arranged that the exim_install script now takes note of these two
3261 PH/35 Installed configuration files for Dragonfly.
3263 PH/36 When a locally submitted message by a trusted user did not contain a
3264 From: header, and the sender address was obtained from -f or from an SMTP
3265 MAIL command, and the trusted user did not use -F to supply a sender
3266 name, $originator_name was incorrectly used when constructing a From:
3267 header. Furthermore, $originator_name was used for submission mode
3268 messages from external hosts without From: headers in a similar way,
3269 which is clearly wrong.
3271 PH/37 Added control=suppress_local_fixups.
3273 PH/38 When log_selector = +received_sender was set, and the addition of the
3274 sender made the log line's construction buffer exactly full, or one byte
3275 less than full, an overflow happened when the terminating "\n" was
3278 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
3279 when the result of a list match is failure because a DNS lookup failed.
3281 PH/40 RM_COMMAND is now used in the building process.
3283 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
3284 the "build-* directories that it finds.
3286 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
3287 address in a domain literal was a prefix of an interface address.
3289 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
3290 when verifying a sender address, unless rewrite_headers is false.
3292 PH/44 Wrote a long comment about why errors_to addresses are verified as
3293 recipients, not senders.
3295 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
3296 the ratelimit ACL was added.
3298 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
3300 PH/46 Added extra information about PostgreSQL errors to the error string.
3302 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
3303 avoiding the need to install special zones in a real server. This is
3304 backwards compatible; if it can't find the fake resolver, it drops back.
3305 Thus, both old and new test suites can be run.
3307 TF/02 Added util/ratelimit.pl
3309 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
3310 clock is set back in time.
3312 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
3313 Candler <B.Candler@pobox.com>.
3315 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
3316 OK for addresses that are the result of redirections.
3318 PH/48 A number of further additions for the benefit of the new test suite,
3319 including a fake gethostbyname() that interfaces to the fake DNS resolver
3322 TF/06 The fix for widen_domains has also been applied to qualify_single and
3323 search_parents which are the other dnslookup options that can cause
3326 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
3329 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
3331 TF/07 Exim produced the error message "an SRV record indicated no SMTP
3332 service" if it encountered an MX record with an empty target hostname.
3333 The message is now "an MX or SRV record indicated no SMTP service".
3335 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
3336 if the DNS of the sending site is misconfigured. This is quite a
3337 common situation. This change restores the behaviour of treating a
3338 helo verification defer as a failure.
3340 PH/51 If self=fail was set on a router, the bounce message did not include the
3341 actual error message.
3347 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
3349 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
3350 command itself ended in error, the underlying message about the transport
3351 filter timeout was being overwritten with the pipe command error. Now the
3352 underlying error message should be appended to the second error message.
3354 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
3356 PH/02 Reduce the amount of output that "make" produces by default. Full output
3357 can still be requested.
3359 PH/03 The warning log line about a condition test deferring for a "warn" verb
3360 was being output only once per connection, rather than after each
3361 occurrence (because it was using the same function as for successful
3362 "warn" verbs). This seems wrong, so I have changed it.
3364 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
3365 it should not have, which might have caused a crash in the right
3366 circumstances, but probably never did.
3368 PH/04 Installed a modified version of Tony Finch's patch to make submission
3369 mode fix the return path as well as the Sender: header line, and to
3370 add a /name= option so that you can make the user's friendly name appear
3373 TF/03 Added the control = fakedefer ACL modifier.
3375 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
3376 Mark Lowes for thorough testing.
3378 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
3380 TK/03 Merged latest SRS patch from Miles Wilton.
3382 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
3383 with the definition in sysexits.h (which is #included earlier).
3384 Fortunately, Exim does not actually use EX_OK. The code used to try to
3385 preserve the sysexits.h value, by assumimg that macro definitions were
3386 scanned for macro replacements. I have been disabused of this notion,
3387 so now the code just undefines EX_OK before #including unistd.h.
3389 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
3390 in the smtp transport. When a block could not be written in a single
3391 write() function, the timeout was being re-applied to each part-write.
3392 This seems wrong - if the receiver was accepting one byte at a time it
3393 would take for ever. The timeout is now adjusted when this happens. It
3394 doesn't have to be particularly precise.
3396 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
3397 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
3399 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
3400 without a domain if the check to <postmaster@domain> fails.
3402 SC/01 Eximstats: added -xls and the ability to specify output files
3403 (patch written by Frank Heydlauf).
3405 SC/02 Eximstats: use FileHandles for outputing results.
3407 SC/03 Eximstats: allow any combination of xls, txt, and html output.
3409 SC/04 Eximstats: fixed display of large numbers with -nvr option
3411 SC/05 Eximstats: fixed merging of reports with empty tables.
3413 SC/06 Eximstats: added the -include_original_destination flag
3415 SC/07 Eximstats: removed tabs and trailing whitespace.
3417 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
3419 TK/06 MBOX spool code: Add real "From " MBOX separator line
3420 so the .eml file is really in mbox format (even though
3421 most programs do not really care). Patch from Alex Miller.
3423 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
3424 The latter is generated from $received_to and is only set if the
3425 message has one envelope recipient. SA can use these headers,
3426 obviously out-of-the-box. Patch from Alex Miller.
3428 PH/08 The ${def test on a variable was returning false if the variable's
3429 value was "0", contrary to what the specification has always said!
3430 The result should be true unless the variable is empty.
3432 PH/09 The syntax error of a character other than { following "${if
3433 def:variable_name" (after optional whitespace) was not being diagnosed.
3434 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
3435 accidental colon was present, for example, could give incorrect results.
3437 PH/10 Tidied the code in a number of places where the st_size field of a stat()
3438 result is used (not including appendfile, where other changes are about
3441 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
3442 This involved changing a lot of size variables from int to off_t. It
3443 should work with maildirs and everything.
3445 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
3446 spamd dying while we are connected to it.
3448 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
3451 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
3452 function for BDB 4.3.
3454 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
3456 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
3457 the sources for PCRE have been re-arranged and more files are now
3460 PH/15 The code I had for printing potentially long long variables in PH/11
3461 above was not the best (it lost precision). The length of off_t variables
3462 is now inspected at build time, and an appropriate printing format (%ld
3463 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
3464 to be "long long int" or "long int". This is needed for the internal
3465 formatting function string_vformat().
3467 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
3468 the configuration file to be ":syslog", then the script "guesses" where
3469 the logs files are, rather than using the compiled in default. In our
3470 case the guess is not the same as the compiled default, so the script
3471 suddenly stopped working when I started to use syslog. The patch checks
3472 to see if log_file_path is "". If so, it attempts to read it from exim
3473 with no configuration file to get the compiled in version, before it
3474 falls back to the previous guessing code."
3476 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
3477 implementing BATV in an Exim configuration. See NewStuff for the gory
3480 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
3481 Makefile that are specific to HP-UX.
3483 PH/18 If the "use_postmaster" option was set for a recipient callout together
3484 with the "random" option, the postmaster address was used as the MAIL
3485 FROM address for the random test, but not for the subsequent recipient
3486 test. It is now used for both.
3488 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
3489 patch removes a few documentation additions to RFC 3028, because the
3490 latest draft now contains them. It adds the new en;ascii-case comparator
3491 and a new error check for 8bit text in MIME parts. Comparator and
3492 require names are now matched exactly. I enabled the subaddress
3493 extension, but it is not well tested yet (read: it works for me)."
3495 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
3496 rework some of the code of TK/09 above to avoid the hardwired use of
3497 "%lld" and "long long". Replaced the call to snprintf() with a call to
3500 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
3501 records point to non-existent hosts", "retry timeout exceeded", and
3502 "retry time not reached for any host after a long failure period".
3504 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
3505 experimental DomainKeys support:
3507 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
3508 (2) On an error such as an illegally used "control", the wrong name for
3509 the control was given.
3511 These problems did NOT occur unless DomainKeys support was compiled.
3513 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
3515 PH/24 Added ${if match_ip condition.
3517 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
3518 databases so that it will be absolutely obvious if a crash occurs in the
3519 DB library. This is a regular occurrence (often caused by mis-matched
3522 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
3523 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
3524 picked up on a user's system that detects such things. There doesn't seem
3525 to be a gcc warning option for this - only an attribute that has to be
3526 put on the function's prototype. It seems that in Fedora Core 4 they have
3527 set this on a number of new functions. No doubt there will be more in due
3530 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
3531 specify a transport. However, if an address that was verified by such a
3532 router was the subject of a callout, Exim crashed because it tried to
3533 read the rcpt_include_affixes from the non-existent transport. Now it
3534 just assumes that the setting of that option is false. This bug was
3535 introduced by 4.51/PH/31.
3537 PH/28 Changed -d+all to exclude +memory, because that information is very
3538 rarely of interest, but it makes the output a lot bigger. People tend to
3539 do -d+all out of habit.
3541 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
3542 code in os-type was giving problems when libc.so lives in lib64, like on
3545 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
3546 aren't the modern standard, and it seems that some systems' include files
3547 don't always have them. Exim was already checking for some of the newer
3548 ones like T_AAAA, and defining it itself. I've added checks for all the
3549 record types that Exim uses.
3551 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
3552 not automatically generating a new one, as it is supposed to. This
3553 prevented TLS from working. If the file did exist, but contained invalid
3554 data, a new version was generated, as expected. It was only the case of a
3555 non-existent file that was broken.
3557 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
3558 with a change in libdomainkeys > 0.64.
3560 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
3561 from DNS. If the selector record carries the flag, it now has
3562 precedence over the domain-wide flag.
3564 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
3566 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
3567 the use of a port name that isn't defined in /etc/services) occurred, the
3568 message was deferred as in a normal delivery, and thus remained on the
3569 spool, instead of being failed because of the mua_wrapper setting. This
3570 is now fixed, and I tidied up some of the mua_wrapper messages at the
3573 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
3574 the messages in a hash of arrays rather than using individual hashes.
3575 This is a bit cleaner and results in dramatic memory savings, albeit
3576 at a slight CPU cost.
3578 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
3579 as requested by Marc Sherman.
3581 SC/10 Eximstats: added histograms for user specified patterns as requested
3584 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
3586 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
3587 fopen() in the content-scanning modules that did not already have it.
3593 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
3594 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
3596 TK/02 Fix ACL "control" statement not being available in MIME ACL.
3598 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
3600 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
3601 to test Sieve filters that use "vacation".
3603 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
3604 that changes the way the GnuTLS parameters are stored in the cache file.
3605 The new format can be generated externally. For backward compatibility,
3606 if the data in the cache doesn't make sense, Exim assumes it has read an
3607 old-format file, and it generates new data and writes a new file. This
3608 means that you can't go back to an older release without removing the
3611 PH/03 A redirect router that has both "unseen" and "one_time" set does not
3612 work if there are any delivery delays because "one_time" forces the
3613 parent to be marked "delivered", so its unseen clone is never tried
3614 again. For this reason, Exim now forbids the simultaneous setting of
3617 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
3618 redirected to themselves ("homonym" addresses). Read the long ChangeLog
3619 entry if you want to know the details. The fix, however, neglected to
3620 consider the case when local delivery batching is involved. The test for
3621 "previously delivered" was not happening when checking to see if an
3622 address could be batched with a previous (undelivered) one; under
3623 certain circumstances this could lead to multiple deliveries to the same
3626 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
3627 in its include files, and this causes problems building Exim.
3629 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
3630 header_syntax) but Exim was just ignoring anything given after a slash.
3631 In particular, this caused confusion with an attempt to use "verify =
3632 reverse_host_lookup/defer_ok". An error is now given when options are
3633 supplied for verify items that do not have them. (Maybe reverse_host_
3634 lookup should have a defer_ok option, but that's a different point.)
3636 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
3637 defined by RFC 821) to 2048, because there were problems with some AUTH
3638 commands, and RFC 1869 says the size should be increased for extended
3639 SMTP commands that take arguments.
3641 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
3644 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
3645 "unknown" error; now it says that the functionality isn't in the binary.
3647 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
3648 an address' error message when a string expansion fails (syntax or
3649 whatever). Otherwise the password may appear in the log. Following change
3650 PH/42 below, there is no longer a chance of it appearing in a bounce
3653 PH/11 Installed exipick version 20050225.0 from John Jetmore.
3655 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
3656 of its addresses was ever tried. (Bugzilla bug #2.)
3658 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
3659 the result incorrectly in the debug output. (It correctly added a newline
3660 to what was transported.)
3662 TF/01 Added $received_time.
3664 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
3665 commented out examples of how to interface to a virus scanner and to
3666 SpamAssassin. Also added commented examples of av_scanner and
3667 spamd_address settings.
3669 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
3670 and controls are allowed in which ACLs. There were a couple of minor
3671 errors. Some of the entries in the conditions table (which is a table of
3672 where they are NOT allowed) were getting very unwieldy; rewrote them as a
3673 negation of where the condition IS allowed.
3675 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
3677 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
3678 header file does not have a version number, so I've had to invent a new
3679 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
3680 API. The code is untested by me (my Linux distribution still has 0.3.2 of
3681 radiusclient), but it was contributed by a Radius user.
3683 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
3684 files or directories, for OpenSSL.
3686 PH/19 When an Exim process that is running as root has to create an Exim log
3687 file, it does so in a subprocess that runs as exim:exim so as to get the
3688 ownership right at creation (otherwise, other Exim processes might see
3689 the file with the wrong ownership). There was no test for failure of this
3690 fork() call, which would lead to the process getting stuck as it waited
3691 for a non-existent subprocess. Forks do occasionally fail when resources
3692 run out. I reviewed all the other calls to fork(); they all seem to check
3695 PH/20 When checking for unexpected SMTP input at connect time (before writing
3696 the banner), Exim was not dealing correctly with a non-positive return
3697 from the read() function. If the client had disconnected by this time,
3698 the result was a log entry for a synchronization error with an empty
3699 string after "input=" when read() returned zero. If read() returned -1
3700 (an event I could not check), uninitialized data bytes were printed.
3701 There were reports of junk text (parts of files, etc) appearing after
3704 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
3706 PH/22 Added support for macro redefinition, and (re)definition in between
3707 driver and ACL definitions.
3709 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
3710 forgetting to use the resulting value; it was using the unexpanded value.
3712 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
3713 hadn't been configured. The fix is from Juergen Kreileder, who
3714 understands it better than I do:
3716 "Here's what I see happening with three configured cyrus_sasl
3717 authenticators configured (plain, login, cram-md5):
3719 On startup auth_cyrus_sasl_init() gets called for each of these.
3720 This means three calls to sasl_listmech() without a specified mech_list.
3721 => SASL tests which mechs of all available mechs actually work
3722 => three warnings about OTP not working
3723 => the returned list contains: plain, login, cram-md5, digest-md5, ...
3725 With the patch, sasl_listmech() also gets called three times. But now
3726 SASL's mech_list option is set to the server_mech specified in the the
3727 authenticator. Or in other words, the answer from sasl_listmech()
3728 gets limited to just the mech you're testing for (which is different
3730 => the return list contains just 'plain' or 'login', 'cram-md5' or
3731 nothing depending on the value of ob->server_mech.
3733 I've just tested the patch: Authentication still works fine,
3734 unavailable mechs specified in the exim configuration are still
3735 caught, and the auth.log warnings about OTP are gone."
3737 PH/25 When debugging is enabled, the contents of the command line are added
3738 to the debugging output, even when log_selector=+arguments is not
3741 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
3742 answer is "GNU", and only if the return is "GNU/something" is the answer
3745 PH/27 $acl_verify_message is now set immediately after the failure of a
3746 verification in an ACL, and so is available in subsequent modifiers. In
3747 particular, the message can be preserved by coding like this:
3749 warn !verify = sender
3750 set acl_m0 = $acl_verify_message
3752 Previously, $acl_verify_message was set only while expanding "message"
3753 and "log_message" when a very denied access.
3755 PH/28 Modified OS/os.c-Linux with
3757 -#ifndef OS_LOAD_AVERAGE
3758 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
3760 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
3761 nomenclature these days.)
3763 PH/29 Installed patch from the Sieve maintainer that adds the options
3764 sieve_useraddress and sieve_subaddress to the redirect router.
3766 PH/30 In these circumstances:
3767 . Two addresses routed to the same list of hosts;
3768 . First host does not offer TLS;
3769 . First host accepts first address;
3770 . First host gives temporary error to second address;
3771 . Second host offers TLS and a TLS session is established;
3772 . Second host accepts second address.
3773 Exim incorrectly logged both deliveries with the TLS parameters (cipher
3774 and peerdn, if requested) that were in fact used only for the second
3777 PH/31 When doing a callout as part of verifying an address, Exim was not paying
3778 attention to any local part prefix or suffix that was matched by the
3779 router that accepted the address. It now behaves in the same way as it
3780 does for delivery: the affixes are removed from the local part unless
3781 rcpt_include_affixes is set on the transport.
3783 PH/32 Add the sender address, as F=<...>, to the log line when logging a
3784 timeout during the DATA phase of an incoming message.
3786 PH/33 Sieve envelope tests were broken for match types other than :is. I have
3787 applied a patch sanctioned by the Sieve maintainer.
3789 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
3790 the uid or gid is negative. A case of a negative gid caused this to be
3791 noticed. The fix allows for either to be negative.
3793 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
3794 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
3797 PH/36 The change PH/12 above was broken. Fixed it.
3799 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
3800 the grounds that routing the same address twice would always produce the
3801 same answer. This might have been true once, but it is certainly no
3802 longer true now. Routing a child address may depend on the previous
3803 routing that produced that child. Some complicated redirection strategies
3804 went wrong when messages had multiple recipients, and made Exim's
3805 behaviour dependent on the order in which the addresses were given.
3807 I have moved the duplicate checking until after the routing is complete.
3808 Exim scans the addresses that are assigned to local and remote
3809 transports, and removes any duplicates. This means that more work will be
3810 done, as duplicates will always all be routed, but duplicates are
3811 presumably rare, so I don't expect this is of any significance.
3813 For deliveries to pipes, files, and autoreplies, the duplicate checking
3814 still happens during the routing process, since they are not going to be
3817 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
3818 It corrects a timeout issue with spamd. This is Ian's comment: "The
3819 background is that sometimes spamd either never reads data from a
3820 connection it has accepted, or it never writes response data. The exiscan
3821 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
3822 blindly assumes that writes won't block so it may never time out."
3824 PH/39 Allow G after quota size as well as K and M.
3826 PH/40 The value set for $authenticated_id in an authenticator may not contain
3827 binary zeroes or newlines because the value is written to log lines and
3828 to spool files. There was no check on this. Now the value is run through
3829 the string_printing() function so that such characters are converted to
3830 printable escape sequences.
3832 PH/41 $message_linecount is a new variable that contains the total number of
3833 lines in the message. Compare $body_linecount, which is the count for the
3836 PH/42 Exim no longer gives details of delivery errors for specific addresses in
3837 bounce and delay warning messages, except in certain special cases, which
3840 (a) An SMTP error message from a remote host;
3841 (b) A message specified in a :fail: redirection;
3842 (c) A message specified in a "fail" command in a system filter;
3843 (d) A message specified in a FAIL return from the queryprogram router;
3844 (e) A message specified by the cannot_route_message router option.
3846 In these cases only, Exim does include the error details in bounce and
3847 warning messages. There are also a few cases where bland messages such
3848 as "unrouteable address" or "local delivery error" are given.
3850 PH/43 $value is now also set for the "else" part of a ${run expansion.
3852 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
3853 being worked on, but at least Exim now implements the latest version to
3856 PH/45 In a pipe transport, although a timeout while waiting for the pipe
3857 process to complete was treated as a delivery failure, a timeout while
3858 writing the message to the pipe was logged, but erroneously treated as a
3859 successful delivery. Such timeouts include transport filter timeouts. For
3860 consistency with the overall process timeout, these timeouts are now
3861 treated as errors, giving rise to delivery failures by default. However,
3862 there is now a new Boolean option for the pipe transport called
3863 timeout_defer, which, if set TRUE, converts the failures into defers for
3864 both kinds of timeout. A transport filter timeout is now identified in
3867 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
3868 systems where "make" and "gmake" are different, calling "gmake" at top
3869 level broke things. I've arranged for the value of $(MAKE) to be passed
3870 from the Makefile to this script so that it can call the same version of
3874 A note about Exim versions 4.44 and 4.50
3875 ----------------------------------------
3877 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
3878 changes of various kinds. As a consequence, a big documentation update was
3879 needed. This delayed the release for rather longer than seemed good, especially
3880 in the light of a couple of (minor) security issues. Therefore, the changes
3881 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
3882 release. So 4.44 and 4.50 are in effect two different branches that both start
3885 I have left the 4.50 change log unchanged; it contains all the changes since
3886 4.43. The change log for 4.44 is below; many of its items are identical to
3887 those for 4.50. This seems to be the most sensible way to preserve the
3888 historical information.
3894 1. Minor wording change to the doc/README.SIEVE file.
3896 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
3897 computation of the current number of files was incorrect.
3899 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
3900 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
3903 4. Give more explanation in the error message when the command for a transport
3904 filter fails to execute.
3906 5. There are several places where Exim runs a non-Exim command in a
3907 subprocess. The SIGUSR1 signal should be disabled for these processes. This
3908 was being done only for the command run by the queryprogram router. It is
3909 now done for all such subprocesses. The other cases are: ${run, transport
3910 filters, and the commands run by the lmtp and pipe transports.
3912 6. Added CONFIGURE_GROUP build-time option.
3914 7. Some older OS have a limit of 256 on the maximum number of file
3915 descriptors. Exim was using setrlimit() to set 1000 as a large value
3916 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
3917 systems. I've change it so that if it can't get 1000, it tries for 256.
3919 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
3920 was an oversight, and furthermore, ever since the addition of extra
3921 controls (e.g. 4.43/32), the checks on when to allow different forms of
3922 "control" were broken. There should now be diagnostics for all cases when a
3923 control that does not make sense is encountered.
3925 9. Added the /retain_sender option to "control=submission".
3927 10. $recipients is now available in the predata ACL (oversight).
3929 11. Tidy the search cache before the fork to do a delivery from a message
3930 received from the command line. Otherwise the child will trigger a lookup
3931 failure and thereby defer the delivery if it tries to use (for example) a
3932 cached ldap connection that the parent has called unbind on.
3934 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
3935 of $address_data from the recipient verification was clobbered by the
3936 sender verification.
3938 13. The value of address_data from a sender verification is now available in
3939 $sender_address_data in subsequent conditions in the ACL statement.
3941 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
3943 15. Added a new option "connect=<time>" to callout options, to set a different
3946 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
3947 was its contents. (It was OK if the option was not defined at all.)
3949 17. A "Completed" log line is now written for messages that are removed from
3950 the spool by the -Mrm option.
3952 18. New variables $sender_verify_failure and $recipient_verify_failure contain
3953 information about exactly what failed.
3955 19. Added -dd to debug only the daemon process.
3957 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
3958 handles timeouts, both on the server side and network timeouts. Renamed the
3959 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
3961 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
3962 It is now set to "smtps".
3964 22. $host_address is now set to the target address during the checking of
3965 ignore_target_hosts.
3967 23. When checking ignore_target_hosts for an ipliteral router, no host name was
3968 being passed; this would have caused $sender_host_name to have been used if
3969 matching the list had actually called for a host name (not very likely,
3970 since this list is usually IP addresses). A host name is now passed as
3973 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
3974 code that specifies a non-restarting handler (typically sigaction() in
3975 modern systems) in an attempt to fix a rare and obscure crash bug.
3977 25. Narrowed the window for a race in the daemon that could cause it to ignore
3978 SIGCHLD signals. This is not a major problem, because they are used only to
3979 wake it up if nothing else does.
3981 26. A malformed maildirsize file could cause Exim to calculate negative values
3982 for the mailbox size or file count. Odd effects could occur as a result.
3983 The maildirsize information is now recalculated if the size or filecount
3986 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
3987 support for a long time. Removed HAVE_SYS_VFS_H.
3989 28. Installed the latest version of exipick from John Jetmore.
3991 29. In an address list, if the pattern was not a regular expression, an empty
3992 subject address (from a bounce message) matched only if the pattern was an
3993 empty string. Non-empty patterns were not even tested. This was the wrong
3994 because it is perfectly reasonable to use an empty address as part of a
3995 database query. An empty address is now tested by patterns that are
3996 lookups. However, all the other forms of pattern expect the subject to
3997 contain a local part and a domain, and therefore, for them, an empty
3998 address still always fails if the pattern is not itself empty.
4000 30. Exim went into a mad DNS loop when attempting to do a callout where the
4001 host was specified on an smtp transport, and looking it up yielded more
4002 than one IP address.
4004 31. Re-factored the code for checking spool and log partition space into a
4005 function that finds that data and another that does the check. The former
4006 is then used to implement four new variables: $spool_space, $log_space,
4007 $spool_inodes, and $log_inodes.
4009 32. The RFC2047 encoding function was originally intended for short strings
4010 such as real names; it was not keeping to the 75-character limit for
4011 encoded words that the RFC imposes. It now respects the limit, and
4012 generates multiple encoded words if necessary. To be on the safe side, I
4013 have increased the buffer size for the ${rfc2047: expansion operator from
4016 33. It is now permitted to omit both strings after an "if" condition; if the
4017 condition is true, the result is "true". As before, when the second string
4018 is omitted, a false condition yields an empty string. This makes it less
4019 cumbersome to write custom ACL and router conditions.
4021 34. Failure to deliver a bounce message always caused it to be frozen, even if
4022 there was an errors_to setting on the router. The errors_to setting is now
4025 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4026 canonical form (fully expanded) before being placed in
4027 $sender_host_address.
4029 36. The table in the code that translates DNS record types into text (T_A to
4030 "A" for instance) was missing entries for NS and CNAME. It is just possible
4031 that this could have caused confusion if both these types were looked up
4032 for the same domain, because the text type is used as part of Exim's
4033 per-process caching. But the chance of anyone hitting this buglet seems
4036 37. The dnsdb lookup has been extended in a number of ways.
4038 (1) There is a new type, "zns", which walks up the domain tree until it
4039 finds some nameserver records. It should be used with care.
4041 (2) There is a new type, "mxh", which is like "mx" except that it returns
4042 just the host names, not the priorities.
4044 (3) It is now possible to give a list of domains (or IP addresses) to be
4045 looked up. The behaviour when one of the lookups defers can be
4046 controlled by a keyword.
4048 (4) It is now possible to specify the separator character for use when
4049 multiple records are returned.
4051 38. The dnslists ACL condition has been extended: it is now possible to supply
4052 a list of IP addresses and/or domains to be looked up in a particular DNS
4055 39. Added log_selector=+queue_time_overall.
4057 40. When running the queue in the test harness, wait just a tad after forking a
4058 delivery process, to get repeatability of debugging output.
4060 41. Include certificate and key file names in error message when GnuTLS fails
4061 to set them up, because the GnuTLS error message doesn't include the name
4062 of the failing file when there is a problem reading it.
4064 42. Allow both -bf and -bF in the same test run.
4066 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
4068 44. The "Exiscan patch" is now merged into the mainline Exim source.
4070 45. Sometimes the final signoff response after QUIT could fail to get
4071 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4072 < 0 before doing a fflush(). This bug looks as though it goes back to the
4073 introduction of TLS in release 3.20, but "sometimes" must have been rare
4074 because the tests only now provoked it.
4076 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
4077 (this can affect the format of dates).
4079 47. exim_tidydb, when checking for the continued existence of a message for
4080 which it has found a message-specific retry record, was not finding
4081 messages that were in split spool directories. Consequently, it was
4082 deleting retry records that should have stayed in existence.
4084 48. Steve fixed some bugs in eximstats.
4086 49. The SPA authentication driver was not abandoning authentication and moving
4087 on to the next authenticator when an expansion was forced to fail,
4088 contradicting the general specification for all authenticators. Instead it
4089 was generating a temporary error. It now behaves as specified.
4091 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
4092 (the order specifies the preference for clients). The order is now AES256,
4093 AES128, 3DES, ARCFOUR128.
4095 51. Small patch to Sieve code - explicitly set From: when generating an
4098 52. Exim crashed if a remote delivery caused a very long error message to be
4099 recorded - for instance if somebody sent an entire SpamAssassin report back
4100 as a large number of 550 error lines. This bug was coincidentally fixed by
4101 increasing the size of one of Exim's internal buffers (big_buffer) that
4102 happened as part of the Exiscan merge. However, to be on the safe side, I
4103 have made the code more robust (and fixed the comments that describe what
4106 53. Now that there can be additional text after "Completed" in log lines (if
4107 the queue_time_overall log selector is set), a one-byte patch to exigrep
4108 was needed to allow it to recognize "Completed" as not the last thing in
4111 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
4112 patch that reportedly fixes this has been added. I am not expert enough to
4113 create a test for it. This is what the patch creator wrote:
4115 "I found a little strange behaviour of ldap code when working with
4116 Windows 2003 AD Domain, where users was placed in more than one
4117 Organization Units. When I tried to give exim partial DN, the exit code
4118 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
4119 But simultaneously result of request was absolutely normal ldap result,
4120 so I produce this patch..."
4122 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
4123 REFERENCE, so I have modified the code to exclude the patch when that macro
4126 55. Some experimental protocols are using DNS PTR records for new purposes. The
4127 keys for these records are domain names, not reversed IP addresses. The
4128 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4129 leaves it alone. Component reversal etc. now happens only for IP addresses.
4132 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
4134 57. Double the size of the debug message buffer (to 2048) so that more of very
4135 long debug lines gets shown.
4137 58. The exicyclog utility now does better if the number of log files to keep
4138 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
4140 59. Two changes related to the smtp_active_hostname option:
4142 (1) $smtp_active_hostname is now available as a variable.
4143 (2) The default for smtp_banner uses $smtp_active_hostname instead
4144 of $primary_hostname.
4146 60. The host_aton() function is supposed to be passed a string that is known
4147 to be a valid IP address. However, in the case of IPv6 addresses, it was
4148 not checking this. This is a hostage to fortune. Exim now panics and dies
4149 if the condition is not met. A case was found where this could be provoked
4150 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4151 components; fortuitously, this particular loophole had already been fixed
4152 by change 4.50/55 above.
4154 If there are any other similar loopholes, the new check in host_aton()
4155 itself should stop them being exploited. The report I received stated that
4156 data on the command line could provoke the exploit when Exim was running as
4157 exim, but did not say which command line option was involved. All I could
4158 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4159 running as the user.
4162 61. There was a buffer overflow vulnerability in the SPA authentication code
4163 (which came originally from the Samba project). I have added a test to the
4164 spa_base64_to_bits() function which I hope fixes it.
4167 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
4168 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
4170 63. The daemon start-up calls getloadavg() while still root for those OS that
4171 need the first call to be done as root, but it missed one case: when
4172 deliver_queue_load_max is set with deliver_drop_privilege. This is
4173 necessary for the benefit of the queue runner, because there is no re-exec
4174 when deliver_drop_privilege is set.
4176 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
4177 This has been fixed.
4179 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4180 was in use, was not putting the data itself into the right store pool;
4181 consequently, it could be overwritten for a subsequent message in the same
4182 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4185 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
4187 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
4188 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
4189 the function were treating the return as a boolean value, which happened to
4190 work because 0=false and not-0=true, but is not correct code.
4192 68. The host_aton() function was not handling scoped IPv6 addresses (those
4193 with, for example, "%eth0" on the end) correctly.
4195 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
4196 negated items (that is, ~something) in unsigned ints. Some compilers
4197 apparently mutter when there is no cast.
4199 70. If an address verification called from an ACL failed, and did not produce a
4200 user-specific message (i.e. there was only a "system" message), nothing was
4201 put in $acl_verify_message. In this situation, it now puts the system
4204 71. Change 4.23/11 added synchronization checking at the start of an SMTP
4205 session; change 4.31/43 added the unwanted input to the log line - except
4206 that it did not do this in the start of session case. It now does.
4208 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
4209 This is wrong and can cause the other end to generate a synchronization
4210 error if it is another Exim or anything else that does the synchronization
4211 check. A QUIT command is no longer sent after a timeout.
4213 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
4214 during host lookups.
4216 74. The defer_ok option of callout verification was not working if it was used
4217 when verifying addresses in header lines, that is, for this case:
4219 verify = header_sender/callout=defer_ok
4221 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
4222 those file descriptors could be used for SMTP connections. If anything
4223 wrote to stderr (the example that came up was "warn" in embedded Perl), it
4224 could be sent to the SMTP client, causing chaos. The daemon now opens
4225 stdin, stdout, and stderr to /dev/null when it puts itself into the
4228 76. Arrange for output from Perl's "warn" command to be written to Exim's main
4229 log by default. The user can override this with suitable Perl magic.
4231 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
4232 the log message when discard triggers, was not working for the DATA ACL or
4233 for the non-SMTP ACL.
4235 78. Error message wording change in sieve.c.
4237 79. If smtp_accept_max_per_host was set, the number of connections could be
4238 restricted to fewer than expected, because the daemon was trying to set up
4239 a new connection before checking whether the processes handling previous
4240 connections had finished. The check for completed processes is now done
4241 earlier. On busy systems, this bug wouldn't be noticed because something
4242 else would have woken the daemon, and it would have reaped the completed
4245 80. If a message was submitted locally by a user whose login name contained one
4246 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
4247 It caused a spool format error. I have fixed the spool reading code. A
4248 related problem was that the "from" clause in the Received: line became
4249 illegal because of the space(s). It is now covered by ${quote_local_part.
4251 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
4254 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
4256 83. Patch from Sieve maintainer to fix unterminated string problem in
4257 "vacation" handling.
4259 84. Some minor changes to the Linux configuration files to help with other
4260 OS variants using glibc.
4262 85. One more patch for Sieve to update vacation handling to latest spec.
4265 ----------------------------------------------------
4266 See the note above about the 4.44 and 4.50 releases.
4267 ----------------------------------------------------
4273 1. Change 4.43/35 introduced a bug that caused file counts to be
4274 incorrectly computed when quota_filecount was set in an appendfile
4277 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4278 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4281 3. Give more explanation in the error message when the command for a transport
4282 filter fails to execute.
4284 4. There are several places where Exim runs a non-Exim command in a
4285 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4286 was being done only for the command run by the queryprogram router. It is
4287 now done for all such subprocesses. The other cases are: ${run, transport
4288 filters, and the commands run by the lmtp and pipe transports.
4290 5. Some older OS have a limit of 256 on the maximum number of file
4291 descriptors. Exim was using setrlimit() to set 1000 as a large value
4292 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4293 systems. I've change it so that if it can't get 1000, it tries for 256.
4295 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4296 was an oversight, and furthermore, ever since the addition of extra
4297 controls (e.g. 4.43/32), the checks on when to allow different forms of
4298 "control" were broken. There should now be diagnostics for all cases when a
4299 control that does not make sense is encountered.
4301 7. $recipients is now available in the predata ACL (oversight).
4303 8. Tidy the search cache before the fork to do a delivery from a message
4304 received from the command line. Otherwise the child will trigger a lookup
4305 failure and thereby defer the delivery if it tries to use (for example) a
4306 cached ldap connection that the parent has called unbind on.
4308 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4309 of $address_data from the recipient verification was clobbered by the
4310 sender verification.
4312 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4313 was its contents. (It was OK if the option was not defined at all.)
4315 11. A "Completed" log line is now written for messages that are removed from
4316 the spool by the -Mrm option.
4318 12. $host_address is now set to the target address during the checking of
4319 ignore_target_hosts.
4321 13. When checking ignore_target_hosts for an ipliteral router, no host name was
4322 being passed; this would have caused $sender_host_name to have been used if
4323 matching the list had actually called for a host name (not very likely,
4324 since this list is usually IP addresses). A host name is now passed as
4327 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4328 code that specifies a non-restarting handler (typically sigaction() in
4329 modern systems) in an attempt to fix a rare and obscure crash bug.
4331 15. Narrowed the window for a race in the daemon that could cause it to ignore
4332 SIGCHLD signals. This is not a major problem, because they are used only to
4333 wake it up if nothing else does.
4335 16. A malformed maildirsize file could cause Exim to calculate negative values
4336 for the mailbox size or file count. Odd effects could occur as a result.
4337 The maildirsize information is now recalculated if the size or filecount
4340 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4341 support for a long time. Removed HAVE_SYS_VFS_H.
4343 18. Updated exipick to current release from John Jetmore.
4345 19. Allow an empty sender to be matched against a lookup in an address list.
4346 Previously the only cases considered were a regular expression, or an
4349 20. Exim went into a mad DNS lookup loop when doing a callout where the
4350 host was specified on the transport, if the DNS lookup yielded more than
4353 21. The RFC2047 encoding function was originally intended for short strings
4354 such as real names; it was not keeping to the 75-character limit for
4355 encoded words that the RFC imposes. It now respects the limit, and
4356 generates multiple encoded words if necessary. To be on the safe side, I
4357 have increased the buffer size for the ${rfc2047: expansion operator from
4360 22. Failure to deliver a bounce message always caused it to be frozen, even if
4361 there was an errors_to setting on the router. The errors_to setting is now
4364 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4365 canonical form (fully expanded) before being placed in
4366 $sender_host_address.
4368 24. Updated eximstats to version 1.33
4370 25. Include certificate and key file names in error message when GnuTLS fails
4371 to set them up, because the GnuTLS error message doesn't include the name
4372 of the failing file when there is a problem reading it.
4374 26. Expand error message when OpenSSL has problems setting up cert/key files.
4377 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
4378 (this can affect the format of dates).
4380 28. exim_tidydb, when checking for the continued existence of a message for
4381 which it has found a message-specific retry record, was not finding
4382 messages that were in split spool directories. Consequently, it was
4383 deleting retry records that should have stayed in existence.
4385 29. eximstats updated to version 1.35
4386 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
4387 1.35 - bugfix such that pie charts by volume are generated correctly
4389 30. The SPA authentication driver was not abandoning authentication and moving
4390 on to the next authenticator when an expansion was forced to fail,
4391 contradicting the general specification for all authenticators. Instead it
4392 was generating a temporary error. It now behaves as specified.
4394 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
4395 (the order specifies the preference for clients). The order is now AES256,
4396 AES128, 3DES, ARCFOUR128.
4398 31. Small patch to Sieve code - explicitly set From: when generating an
4401 32. Exim crashed if a remote delivery caused a very long error message to be
4402 recorded - for instance if somebody sent an entire SpamAssassin report back
4403 as a large number of 550 error lines. This bug was coincidentally fixed by
4404 increasing the size of one of Exim's internal buffers (big_buffer) that
4405 happened as part of the Exiscan merge. However, to be on the safe side, I
4406 have made the code more robust (and fixed the comments that describe what
4409 33. Some experimental protocols are using DNS PTR records for new purposes. The
4410 keys for these records are domain names, not reversed IP addresses. The
4411 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4412 leaves it alone. Component reversal etc. now happens only for IP addresses.
4415 34. The host_aton() function is supposed to be passed a string that is known
4416 to be a valid IP address. However, in the case of IPv6 addresses, it was
4417 not checking this. This is a hostage to fortune. Exim now panics and dies
4418 if the condition is not met. A case was found where this could be provoked
4419 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4420 components; fortuitously, this particular loophole had already been fixed
4421 by change 4.50/55 or 4.44/33 above.
4423 If there are any other similar loopholes, the new check in host_aton()
4424 itself should stop them being exploited. The report I received stated that
4425 data on the command line could provoke the exploit when Exim was running as
4426 exim, but did not say which command line option was involved. All I could
4427 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4428 running as the user.
4431 35. There was a buffer overflow vulnerability in the SPA authentication code
4432 (which came originally from the Samba project). I have added a test to the
4433 spa_base64_to_bits() function which I hope fixes it.
4436 36. The daemon start-up calls getloadavg() while still root for those OS that
4437 need the first call to be done as root, but it missed one case: when
4438 deliver_queue_load_max is set with deliver_drop_privilege. This is
4439 necessary for the benefit of the queue runner, because there is no re-exec
4440 when deliver_drop_privilege is set.
4442 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4443 was in use, was not putting the data itself into the right store pool;
4444 consequently, it could be overwritten for a subsequent message in the same
4445 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4448 38. Sometimes the final signoff response after QUIT could fail to get
4449 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4450 < 0 before doing a fflush(). This bug looks as though it goes back to the
4451 introduction of TLS in release 3.20, but "sometimes" must have been rare
4452 because the tests only now provoked it.
4458 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
4459 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
4460 Now it returns an integer. A number of calls were still expecting a T/F
4461 return. Fortuitously, in all cases, the tests worked in OK situations,
4462 which is the norm. However, things would have gone wrong on any write
4463 failures on the smtp file descriptor. This function is used when sending
4464 messages over SMTP and also when doing verify callouts.
4466 2. When Exim is called to do synchronous delivery of a locally submitted
4467 message (the -odf or -odi options), it no longer closes stderr before doing
4470 3. Implemented the mua_wrapper option.
4472 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
4474 5. Implemented the functions header_remove(), header_testname(),
4475 header_add_at_position(), and receive_remove_recipient(), and exported them
4478 6. If an ACL "warn" statement specified the addition of headers, Exim already
4479 inserted X-ACL-Warn: at the start if there was no header name. However, it
4480 was not making this test for the second and subsequent header lines if
4481 there were newlines in the string. This meant that an invalid header could
4482 be inserted if Exim was badly configured.
4484 7. Allow an ACL "warn" statement to add header lines at the start or after all
4485 the Received: headers, as well as at the end.
4487 8. Added the rcpt_4xx retry error code.
4489 9. Added postmaster_mailfrom=xxx to callout verification option.
4491 10. Added mailfrom=xxxx to the callout verification option, for verify=
4494 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
4495 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
4497 12. Inserted some casts to stop certain compilers warning when using pointer
4498 differences as field lengths or precisions in printf-type calls (mostly
4499 affecting debugging statements).
4501 13. Added optional readline() support for -be (dynamically loaded).
4503 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
4504 same clock tick as a message's arrival, so that its received time was the
4505 same as the "first fail" time on the retry record, and that message
4506 remained on the queue past the ultimate address timeout, every queue runner
4507 would try a delivery (because it was past the ultimate address timeout) but
4508 after another failure, the ultimate address timeout, which should have then
4509 bounced the address, did not kick in. This was a "< instead of <=" error;
4510 in most cases the first failure would have been in the next clock tick
4511 after the received time, and all would be well.
4513 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
4514 being recognized when the domain list was tested by the match_domain
4515 condition in an expansion string.
4517 16. Added the ${str2b64: operator.
4519 17. Exim was always calling setrlimit() to set a large limit for the number of
4520 processes, without checking whether the existing limit was already
4521 adequate. (It did check for the limit on file descriptors.) Furthermore,
4522 errors from getrlimit() and setrlimit() were being ignored. Now they are
4523 logged to the main and panic logs, but Exim does carry on, to try to do its
4524 job under whatever limits there are.
4526 18. Imported PCRE 5.0.
4528 19. Trivial typo in log message " temporarily refused connection" (the leading
4531 20. If the log selector return_path_on_delivery was set and an address was
4532 redirected to /dev/null, the delivery process crashed because it assumed
4533 that a return path would always be set for a "successful" delivery. In this
4534 case, the whole delivery is bypassed as an optimization, and therefore no
4537 21. Internal re-arrangement: the function for sending a challenge and reading
4538 a response while authentication was assuming a zero-terminated challenge
4539 string. It's now changed to take a pointer and a length, to allow for
4540 binary data in such strings.
4542 22. Added the cyrus_sasl authenticator (code supplied by MBM).
4544 23. Exim was not respecting finduser_retries when seeking the login of the
4545 uid under which it was called; it was always trying 10 times. (The default
4546 setting of finduser_retries is zero.) Also, it was sleeping after the final
4547 failure, which is pointless.
4549 24. Implemented tls_on_connect_ports.
4551 25. Implemented acl_smtp_predata.
4553 26. If the domain in control=submission is set empty, Exim assumes that the
4554 authenticated id is a complete email address when it generates From: or
4555 Sender: header lines.
4557 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
4558 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
4559 chown and chgrp in /bin and hostname in /usr/bin.
4561 28. Exim was keeping the "process log" file open after each use, just as it
4562 does for the main log. This opens the possibility of it remaining open for
4563 long periods when the USR1 signal hits a daemon. Occasional processlog
4564 errors were reported, that could have been caused by this. Anyway, it seems
4565 much more sensible not to leave this file open at all, so that is what now
4568 29. The long-running daemon process does not normally write to the log once it
4569 has entered its main loop, and it closes the log before doing so. This is
4570 so that log files can straightforwardly be renamed and moved. However,
4571 there are a couple of unusual error situations where the daemon does write
4572 log entries, and I had neglected to close the log afterwards.
4574 30. The text of an SMTP error response that was received during a remote
4575 delivery was being truncated at 512 bytes. This is too short for some of
4576 the long messages that one sometimes sees. I've increased the limit to
4579 31. It is now possible to make retry rules that apply only when a message has a
4580 specific sender, in particular, an empty sender.
4582 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
4583 it possible to be selective about when SMTP synchronization is enforced.
4585 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
4587 32. Implemented hosts_connection_nolog.
4589 33. Added an ACL for QUIT.
4591 34. Setting "delay_warning=" to disable warnings was not working; it gave a
4594 35. Added mailbox_size and mailbox_filecount to appendfile.
4596 36. Added control = no_multiline_responses to ACLs.
4598 37. There was a bug in the logic of the code that waits for the clock to tick
4599 in the case where the clock went backwards by a substantial amount such
4600 that the microsecond fraction of "now" was more than the microsecond
4601 fraction of "then" (but the whole seconds number was less).
4603 38. Added support for the libradius Radius client library this is found on
4604 FreeBSD (previously only the radiusclient library was supported).
4610 1. When certain lookups returned multiple values in the form name=value, the
4611 quoting of the values was not always being done properly. Specifically:
4612 (a) If the value started with a double quote, but contained no whitespace,
4614 (b) If the value contained whitespace other than a space character (i.e.
4615 tabs or newlines or carriage returns) it was not quoted.
4616 This fix has been applied to the mysql and pgsql lookups by writing a
4617 separate quoting function and calling it from the lookup code. The fix
4618 should probably also be applied to nisplus, ibase and oracle lookups, but
4619 since I cannot test any of those, I have not disturbed their existing code.
4621 2. A hit in the callout cache for a specific address caused a log line with no
4622 reason for rejecting RCPT. Now it says "Previous (cached) callout
4623 verification failure".
4625 3. There was an off-by-one bug in the queryprogram router. An over-long
4626 return line was truncated at 256 instead of 255 characters, thereby
4627 overflowing its buffer with the terminating zero. As well as fixing this, I
4628 have increased the buffer size to 1024 (and made a note to document this).
4630 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
4631 when Exim is waiting for an SMTP response from a remote server, Exim
4632 restarts its select() call on the socket, thereby resetting its timeout.
4633 This is not a problem when such interrupts are rare. Somebody set up a cron
4634 job to run exiwhat every 2 minutes, which is less than the normal select()
4635 timeout (5 or 10 minutes). This meant that the select() timeout never
4636 kicked in because it was always reset. I have fixed this by comparing the
4637 time when an interrupt arrives with the time at the start of the first call
4638 to select(). If more time than the timeout has elapsed, the interrupt is
4639 treated as a timeout.
4641 5. Some internal re-factoring in preparation for the addition of Sieve
4642 extensions (by MH). In particular, the "personal" test is moved to a
4643 separate function, and given an option for scanning Cc: and Bcc: (which is
4644 not set for Exim filters).
4646 6. When Exim created an email address using the login of the caller as the
4647 local part (e.g. when creating a From: or Sender: header line), it was not
4648 quoting the local part when it contained special characters such as @.
4650 7. Installed new OpenBSD configuration files.
4652 8. Reworded some messages for syntax errors in "and" and "or" conditions to
4653 try to make them clearer.
4655 9. Callout options, other than the timeout value, were being ignored when
4656 verifying sender addresses in header lines. For example, when using
4658 verify = header_sender/callout=no_cache
4660 the cache was (incorrectly) being used.
4662 10. Added a missing instance of ${EXE} to the exim_install script; this affects
4663 only the Cygwin environment.
4665 11. When return_path_on_delivery was set as a log selector, if different remote
4666 addresses in the same message used different return paths and parallel
4667 remote delivery occurred, the wrong values would sometimes be logged.
4668 (Whenever a remote delivery process finished, the return path value from
4669 the most recently started remote delivery process was logged.)
4671 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
4672 lines when AUTH and/or TLS are in use. This is the "received protocol"
4673 field. Exim used to use "asmtp" for authenticated SMTP, without any
4674 indication (in the protocol name) for TLS use. Now it follows the RFC and
4675 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
4676 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
4677 names appear in log lines as well as in Received: header lines.
4679 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
4680 extensions, and comparison tests, and to fix some bugs.
4682 14. Changes to the "personal" filter test:
4684 (1) The test was buggy in that it was just doing the equivalent of
4685 "contains" tests on header lines. For example, if a user's address was
4686 anne@some.where, the "personal" test would incorrectly be true for
4688 To: susanne@some.where
4690 This test is now done by extracting each address from the header in turn,
4691 and checking the entire address. Other tests that are part of "personal"
4692 are now done using regular expressions (for example, to check local parts
4693 of addresses in From: header lines).
4695 (2) The list of non-personal local parts in From: addresses has been
4696 extended to include "listserv", "majordomo", "*-request", and "owner-*",
4697 taken from the Sieve specification recommendations.
4699 (3) If the message contains any header line starting with "List-" it is
4700 treated as non-personal.
4702 (4) The test for "circular" in the Subject: header line has been removed
4703 because it now seems ill-conceived.
4705 15. Minor typos in src/EDITME comments corrected.
4707 16. Installed latest exipick from John Jetmore.
4709 17. If headers_add on a router specified a text string that was too long for
4710 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
4711 of string_sprintf() is now avoided.
4713 18. $message_body_size was not set (it was always zero) when running the DATA
4714 ACL and the local_scan() function.
4716 19. For the "mail" command in an Exim filter, no default was being set for
4717 the once_repeat time, causing a random time value to be used if "once" was
4718 specified. (If the value happened to be <= 0, no repeat happened.) The
4719 default is now 0s, meaning "never repeat". The "vacation" command was OK
4720 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
4721 (I found it when inspecting the code).
4723 20. There is now an overall timeout for performing a callout verification. It
4724 defaults to 4 times the callout timeout, which applies to individual SMTP
4725 commands during the callout. The overall timeout applies when there is more
4726 than one host that can be tried. The timeout is checked before trying the
4727 next host. This prevents very long delays if there are a large number of
4728 hosts and all are timing out (e.g. when the network connections are timing
4729 out). The value of the overall timeout can be changed by specifying an
4730 additional sub-option for "callout", called "maxwait". For example:
4732 verify = sender/callout=5s,maxwait=20s
4734 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
4735 to the end before writing, but this should make it even safer).
4737 22. Exim was forgetting that it had advertised PIPELINING for the second and
4738 subsequent messages on an SMTP connection. It was also not resetting its
4739 memory on STARTTLS and an internal HELO.
4741 23. When Exim logs an SMTP synchronization error within a session, it now
4742 records whether PIPELINING has been advertised or not.
4744 24. Added 3 instances of "(long int)" casts to time_t variables that were being
4745 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
4746 rather than long int.
4748 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
4750 26. Added the never_mail option to autoreply.
4756 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
4757 crash if the getsockname() call failed; this can happen if a connection is
4758 closed very soon after it is established. The problem was simply in the
4759 order in which certain operations were done, causing Exim to try to write
4760 to the SMTP stream before it had set up the file descriptor. The bug has
4761 been fixed by making things happen in the correct order.
4767 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
4768 before the connection was closed, thus losing the rejection response.
4770 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
4771 some early Solaris releases, but causes trouble in current releases where
4772 socklen_t is defined.
4774 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
4777 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
4780 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
4781 that releasing the top of it at the end releases what was used for sub-
4782 expansions (unless the block got too big). However, discard this block if
4783 the first thing is a variable or header, so that we can use its block when
4784 it is dynamic (useful for very large $message_headers, for example).
4786 6. Lookups now cache *every* query, not just the most recent. A new, separate
4787 store pool is used for this. It can be recovered when all lookup caches are
4788 flushed. Lookups now release memory at the end of their result strings.
4789 This has involved some general refactoring of the lookup sources.
4791 7. Some code has been added to the store_xxx() functions to reduce the amount
4792 of flapping under certain conditions.
4794 8. log_incoming_interface used to affect only the <= reception log lines. Now
4795 it causes the local interface and port to be added to several more SMTP log
4796 lines, for example "SMTP connection from", and rejection lines.
4798 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
4800 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
4802 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
4803 could be overwritten at the end of the current message (or the start of a
4804 new message if it was set in a HELO ACL). The value is now preserved for
4805 the duration of the SMTP connection.
4807 12. If a transport had a headers_rewrite setting, and a matching header line
4808 contained an unqualified address, that address was qualified, even if it
4809 did not match any rewriting rules. The underlying bug was that the values
4810 of the flags that permit the existence of unqualified sender and recipient
4811 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
4812 non-local messages, and by -bnq for local messages) were not being
4813 preserved with the message after it was received.
4815 13. When Exim was logging an SMTP synchronization error, it could sometimes log
4816 "next input=" as part of the text comprising the host identity instead of
4817 the correct text. The code was using the same buffer for two different
4818 strings. However, depending on which order the printing function evaluated
4819 its arguments, the bug did not always show up. Under Linux, for example, my
4820 test suite worked just fine.
4822 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
4823 doesn't work with some older versions of Perl. It has been changed to "my",
4824 which in any case is probably the better facility to use.
4826 15. A really picky compiler found some instances of statements for creating
4827 error messages that either had too many or two few arguments for the format
4830 16. The size of the buffer for calls to the DNS resolver has been increased
4831 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
4832 for addresses that have a lot of PTR records. This alleviates a problem; it
4833 does not fully solve it.
4835 17. A dnsdb lookup for PTR records that receives more data than will fit in the
4836 buffer now truncates the list and logs the incident, which is the same
4837 action as happens when Exim is looking up a host name and its aliases.
4838 Previously in this situation something unpredictable would happen;
4839 sometimes it was "internal error: store_reset failed".
4841 18. If a server dropped the connection unexpectedly when an Exim client was
4842 using GnuTLS and trying to read a response, the client delivery process
4843 crashed while trying to generate an error log message.
4845 19. If a "warn" verb in an ACL added multiple headers to a message in a single
4846 string, for example:
4848 warn message = H1: something\nH2: something
4850 the text was added as a single header line from Exim's point of view
4851 though it ended up OK in the delivered message. However, searching for the
4852 second and subsequent header lines using $h_h2: did not work. This has been
4853 fixed. Similarly, if a system filter added multiple headers in this way,
4854 the routers could not see them.
4856 20. Expanded the error message when iplsearch is called with an invalid key to
4857 suggest using net-iplsearch in a host list.
4859 21. When running tests using -bh, any delays imposed by "delay" modifiers in
4860 ACLs are no longer actually imposed (and a message to that effect is
4863 22. If a "gecos" field in a passwd entry contained escaped characters, in
4864 particular, if it contained a \" sequence, Exim got it wrong when building
4865 a From: or a Sender: header from that name. A second bug also caused
4866 incorrect handling when an unquoted " was present following a character
4867 that needed quoting.
4869 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
4870 was not being matched caselessly.
4872 24. Arranged for all hyphens in the exim.8 source to be escaped with
4875 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
4876 properly. Recipient callout cache records were still being keyed to include
4877 the sender, even when use_sender was set false. This led to far more
4878 callouts that were necessary. The sender is no longer included in the key
4879 when use_sender is false.
4881 26. Added "control = submission" modifier to ACLs.
4883 27. Added the ${base62d: operator to decode base 62 numbers.
4885 28. dnsdb lookups can now access SRV records.
4887 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
4888 the configuration file.
4890 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
4891 (-v) mode. This makes the output for a verbose queue run more intelligible.
4893 31. Added a use_postmaster feature to recipient callouts.
4895 32. Added the $body_zerocount variable, containing the number of binary zero
4896 bytes in the message body.
4898 33. The time of last modification of the "new" subdirectory is now used as the
4899 "mailbox time last read" when there is a quota error for a maildir
4902 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
4904 35. Added +ignore_unknown as a special item in host lists.
4906 36. Code for decoding IPv6 addresses in host lists is now included, even if
4907 IPv6 support is not being compiled. This fixes a bug in which an IPv6
4908 address was recognized as an IP address, but was then not correctly decoded
4909 into binary, causing unexpected and incorrect effects when compared with
4916 1. Very minor rewording of debugging text in manualroute to say "list of
4917 hosts" instead of "hostlist".
4919 2. If verify=header_syntax was set, and a header line with an unqualified
4920 address (no domain) and a large number of spaces between the end of the
4921 name and the colon was received, the reception process suffered a buffer
4922 overflow, and (when I tested it) crashed. This was caused by some obsolete
4923 code that should have been removed. The fix is to remove it!
4925 3. When running in the test harness, delay a bit after writing a bounce
4926 message to get a bit more predictability in the log output.
4928 4. Added a call to search_tidyup() just before forking a reception process. In
4929 theory, someone could use a lookup in the expansion of smtp_accept_max_
4930 per_host which, without the tidyup, could leave open a database connection.
4932 5. Added the variables $recipient_data and $sender_data which get set from a
4933 lookup success in an ACL "recipients" or "senders" condition, or a router
4934 "senders" option, similar to $domain_data and $local_part_data.
4936 6. Moved the writing of debug_print from before to after the "senders" test
4939 7. Change 4.31/66 (moving the time when the Received: is generated) caused
4940 problems for message scanning, either using a data ACL, or using
4941 local_scan() because the Received: header was not generated till after they
4942 were called (in order to set the time as the time of reception completion).
4943 I have revised the way this works. The header is now generated after the
4944 body is received, but before the ACL or local_scan() are called. After they
4945 are run, the timestamp in the header is updated.
4951 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
4952 before starting a queue runner without re-exec. This happened only when
4953 deliver_drop_privilege was set or when the Exim user was set to root. The
4954 effect of the bug was that timeouts during subsequent deliveries caused
4955 crashes instead of being properly handled. The handler is now left at its
4956 default (and expected) setting.
4958 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
4959 message, again when deliver_drop_privilege is set or Exim is run as root.
4960 The bug described in (1) was not present in this case, but the tidying up
4961 of the other signals was missing. I have made the two cases consistent.
4963 3. The ignore_target_hosts setting on a manualroute router was being ignored
4964 for hosts that were looked up using the /MX notation.
4966 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
4969 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
4970 operated on the sender address. After changing the $sender_address to <>
4971 for the sender address verify, Exim was re-instated it as the original
4972 (before rewriting) address, but remembering that it had rewritten it, so it
4973 wasn't rewriting it again. This bug also had the effect of breaking the
4974 sender address verification caching when the sender address was rewritten.
4976 6. The ignore_target_hosts option was being ignored by the ipliteral router.
4977 This has been changed so that if the ip literal address matches
4978 ignore_target_hosts, the router declines.
4980 7. Added expansion conditions match_domain, match_address, and match_local_
4981 part (NOT match_host).
4983 8. The placeholder for the Received: header didn't have a length field set.
4985 9. Added code to Exim itself and to exim_lock to test for a specific race
4986 condition that could lead to file corruption when using MBX delivery. The
4987 issue is with the lockfile that is created in /tmp. If this file is removed
4988 after a process has opened it but before that process has acquired a lock,
4989 there is the potential for a second process to recreate the file and also
4990 acquire a lock. This could lead to two Exim processes writing to the file
4991 at the same time. The added code performs the same test as UW imapd; it
4992 checks after acquiring the lock that its file descriptor still refers to
4993 the same named file.
4995 10. The buffer for building added header lines was of fixed size, 8192 bytes.
4996 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
4999 11. Added the smtp_active_hostname option. If used, this will typically be made
5000 to depend on the incoming interface address. Because $interface_address is
5001 not set up until the daemon has forked a reception process, error responses
5002 that can happen earlier (such as "too many connections") no longer contain
5005 12. If an expansion in a condition on a "warn" statement fails because a lookup
5006 defers, the "warn" statement is abandoned, and the next ACL statement is
5007 processed. Previously this caused the whole ACL to be aborted.
5009 13. Added the iplsearch lookup type.
5011 14. Added ident_timeout as a log selector.
5013 15. Added tls_certificate_verified as a log selector.
5015 16. Added a global option tls_require_ciphers (compare the smtp transport
5016 option of the same name). This controls incoming TLS connections.
5018 17. I finally figured out how to make tls_require_ciphers do a similar thing
5019 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
5020 before starting the TLS session.
5022 18. Tabs are now shown as \t in -bP output.
5024 19. If the log selector return_path_on_delivery was set, Exim crashed when
5025 bouncing a message because it had too many Received: header lines.
5027 20. If two routers both had headers_remove settings, and the first one included
5028 a superfluous trailing colon, the final name in the first list and the
5029 first name in the second list were incorrectly joined into one item (with a
5030 colon in the middle).
5036 1. Added -C and -D options to the exinext utility, mainly to make it easier
5037 to include in the automated testing, but these could be helpful when
5038 multiple configurations are in use.
5040 2. The exinext utility was not formatting the output nicely when there was
5041 an alternate port involved in the retry record key, nor when there was a
5042 message id as well (for retries that were specific to a specific message
5043 and a specific host). It was also confused by IPv6 addresses, because of
5044 the additional colons they contain. I have fixed the IPv4 problem, and
5045 patched it up to do a reasonable job for IPv6.
5047 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
5048 delivery, the log line now contains "pipelined" if PIPELINING was used.
5050 4. An SMTP transport process used to panic and die if the bind() call to set
5051 an explicit outgoing interface failed. This has been changed; it is now
5052 treated in the same way as a connect() failure.
5054 5. A reference to $sender_host_name in the part of a conditional expansion
5055 that was being skipped was still causing a DNS lookup. This no longer
5058 6. The def: expansion condition was not recognizing references to header lines
5059 that used bh_ and bheader_.
5061 7. Added the _cache feature to named lists.
5063 8. The code for checking quota_filecount in the appendfile transport was
5064 allowing one more file than it should have been.
5066 9. For compatibility with Sendmail, the command line option
5074 and sets the incoming protocol and host name (for trusted callers). The
5075 host name and its colon can be omitted when only the protocol is to be set.
5076 Note the Exim already has two private options, -pd and -ps, that refer to
5077 embedded Perl. It is therefore impossible to set a protocol value of "d" or
5078 "s", but I don't think that's a major issue.
5080 10. A number of refactoring changes to the code, none of which should affect
5083 (a) The number of logging options was getting close to filling up the
5084 32-bit word that was used as a bit map. I have split them into two classes:
5085 those that are passed in the argument to log_write(), and those that are
5086 only ever tested independently outside of that function. These are now in
5087 separate 32-bit words, so there is plenty of room for expansion again.
5088 There is no change in the user interface or the logging behaviour.
5090 (b) When building, for example, log lines, the code previously used a
5091 macro that called string_cat() twice, in order to add two strings. This is
5092 not really sufficiently general. Furthermore, there was one instance where
5093 it was actually wrong because one of the argument was used twice, and in
5094 one call a function was used. (As it happened, calling the function twice
5095 did not affect the overall behaviour.) The macro has been replaced by a
5096 function that can join an arbitrary number of extra strings onto a growing
5099 (c) The code for expansion conditions now uses a table and a binary chop
5100 instead of a serial search (which was left over from when there were very
5101 few conditions). Also, it now recognizes conditions like "pam" even when
5102 the relevant support is not compiled in: a suitably worded error message is
5103 given if an attempt is made to use such a condition.
5105 11. Added ${time_interval:xxxxx}.
5107 12. A bug was causing one of the ddress fields not to be passed back correctly
5108 from remote delivery subprocesses. The field in question was not being
5109 subsequently used, so this caused to problems in practice.
5111 13. Added new log selectors queue_time and deliver_time.
5113 14. Might have fixed a bug in maildirsizefile handling that threw up
5114 "unexpected character" debug warnings, and recalculated the data
5115 unnecessarily. In any case, I expanded the warning message to give more
5118 15. Added the message "Restricted characters in address" to the statements in
5119 the default ACL that block characters like @ and % in local parts.
5121 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
5122 Three changes have been made:
5124 (a) There was a serious bug; a negative response to MAIL caused the whole
5125 recipient domain to be cached as invalid, thereby blocking all messages
5126 to all local parts at the same domain, from all senders. This bug has
5127 been fixed. The domain is no longer cached after a negative response to
5128 MAIL if the sender used is not empty.
5130 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
5133 (c) A new callout option, "use_sender" has been added for people who want
5134 the modified behaviour.
5140 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
5143 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
5144 indeed breaks things for older releases.
5146 3. Added additional logging to the case where there is a problem reading data
5147 from a filter that is running in a subprocess using a pipe, in order to
5148 try to track down a specific problem.
5150 4. Testing facility fudge: when running in the test harness and attempting
5151 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
5152 getting "No route to host". Convert this to a timeout.
5154 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
5157 6. Some OS don't have socklen_t but use size_t instead. This affects the
5158 fifth argument of getsockopt() amongst other things. This is now
5159 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
5160 can be set for individual OS. I have set it for SunOS5, OSF1, and
5161 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
5162 some earlier ones do not.
5164 7. Change 4.30/15 was not doing the test caselessly.
5166 8. The standard form for an IPv6 address literal was being rejected by address
5167 parsing in, for example, MAIL and RCPT commands. An example of this kind of
5168 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5169 this, as well as the form without the "IPv6" on the front (but only when
5170 address literals are enabled, of course).
5172 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
5174 10. Exim crashed if a message with an empty sender address specified by -f
5175 encountered a router with an errors_to setting. This could be provoked only
5176 by a command such as
5180 where an empty string was supplied; "<>" did not hit this bug.
5182 11. Installed PCRE release 4.5.
5184 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
5185 remained set. It is now erased.
5187 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
5188 times from message ids (which are base 36 rather than the normal 62).
5190 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
5191 were being counted as actual protocol errors, and logged if the log
5192 selector +smtp_protocol_error was set. One cannot be perfect in this test,
5193 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
5194 and DATA following a set of rejected RCPTs do not count as protocol errors.
5195 In other words, Exim assumes they were pipelined, though this may not
5196 actually be the case. Of course, in all cases the client gets an
5197 appropriate error code.
5199 15. If a lookup fails in an ACL condition, a message about the failure may
5200 be available; it is used if testing the ACL cannot continue, because most
5201 such messages specify what the cause of the deferral is. However, some
5202 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
5203 that caused an old message to be retained and used if a later statement
5204 caused a defer, replacing the real cause of the deferral.
5206 16. If an IP address had so many PTR records that the DNS lookup buffer
5207 was not large enough to hold them, Exim could crash while trying to process
5208 the truncated data. It now detects and logs this case.
5210 17. Further to 4.21/58, another change has been made: if (and only if) the
5211 first line of a message (the first header line) ends with CRLF, a bare LF
5212 in a subsequent header line has a space inserted after it, so as not to
5213 terminate the header.
5215 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
5216 unnecessarily, used atoi() instead of strtol(), and didn't check the
5217 termination when getting file sizes from file names by regex.
5219 19. Completely re-implemented the support for maildirsize files, in the light
5220 of a number of problems with the previous contributed implementation
5221 (4.30/29). In particular:
5223 . If the quota is zero, the maildirsize file is maintained, but no quota is
5226 . If the maildir directory does not exist, it is created before any attempt
5227 to write a maildirsize file.
5229 . The quota value in the file is just a cache; if the quota is changed in
5230 the transport, the new value overrides.
5232 . A regular expression is available for excluding directories from the
5235 20. The autoreply transport checks the characters in options that define the
5236 message's headers; it allows continued headers, but it was checking with
5237 isspace() after an embedded newline instead of explicitly looking for a
5240 21. If all the "regular" hosts to which an address was routed had passed their
5241 expiry times, and had not reached their retry times, the address was
5242 bounced, even if fallback hosts were defined. Now Exim should go on to try
5245 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
5246 equivalent code in the SMTP transport. Some hosts send humungous responses
5247 to HELO/EHLO, more than 1024 it seems.
5249 23. Refactoring: code in filter.c used (void *) for "any old type" but this
5250 gives compiler warnings in some environments. I've now done it "properly",
5253 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
5254 (because of problems with the built-in one) was declared to return uschar *
5255 instead of char *, causing compiler failure.
5257 25. Fixed a file descriptor leak when processing alias/forward files.
5259 26. Fixed a minor format string issue in dbfn.c.
5261 27. Typo in exim.c: ("dmbnz" for "dbmnz").
5263 28. If a filter file refered to $h_xxx or $message_headers, and the headers
5264 contained RFC 2047 "words", Exim's memory could, under certain conditions,
5267 29. When a sender address is verified, it is cached, to save repeating the test
5268 when there is more than one recipient in a message. However, when the
5269 verification involves a callout, it is possible for different callout
5270 options to be set for different recipients. It is too complicated to keep
5271 track of this in the cache, so now Exim always runs a verification when a
5272 callout is required, relying on the callout cache for the optimization.
5273 The overhead is duplication of the address routing, but this should not be
5276 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
5277 to be verified with callout=postmaster, and the main callout worked but the
5278 postmaster check failed, the verification correctly failed. However, if a
5279 subsequent RCPT command asked for sender verification *without* the
5280 postmaster check, incorrect caching caused this verification also to fail,
5283 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
5284 it was not caching the DNS options (qualify_single, search_parents) that
5285 were used when the lookup failed. A subsequent lookup with different
5286 options therefore always gave the same answer, though there were cases
5287 where it should not have. (Example: a "domains = !$mx_any" option on a
5288 dnslookup router: the "domains" option is always processed without any
5289 widening, but the router might have qualify_single set.) Now Exim uses the
5290 cached value only when the same options are set.
5292 32. Added John Jetmore's "exipick" utility to the distribution.
5294 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
5295 than a timeout (e.g. a certificate is required, and is not provided), an
5296 Exim server now closes the connection immediately. Previously it waited for
5297 the client to close - but if the client is SSL, it seems that they each
5298 wait for each other, leading to a delay before one of them times out.
5300 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
5301 maintained 0.8.x compatibility because I don't think many are using it, and
5302 it is clearly obsolete.
5304 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
5307 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
5308 client certificate was expired. A simple patch fixes this, though I don't
5309 understand the full logic of why the verify callback is called multiple
5312 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
5313 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
5314 which causes problems with some clients (such as the Certicom SSL Plus
5315 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
5316 disables the coutermeasure allowing Eudora to connect."
5318 38. Exim was not checking that a write() to a log file succeeded. This could
5319 lead to Bad Things if a log got too big, in particular if it hit a file
5320 size limit. Exim now panics and dies if it cannot write to a log file, just
5321 as it does if it cannot open a log file.
5323 39. Modified OS/Makefile-Linux so that it now contains
5325 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
5327 The two -D definitions ensure that Exim is compiled with large file
5328 support, which makes it possible to handle log files that are bigger than
5331 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
5332 instance) a domain was checked against a named list that involved a lookup,
5333 causing $domain_data to be set, then another domain was checked against the
5334 same list, then the first domain was re-checked, the value of $domain_data
5335 after the final check could be wrong. In particular, if the second check
5336 failed, it could be set empty. This bug probably also applied to
5339 41. The strip_trailing_dot option was not being applied to the address given
5340 with the -f command-line option.
5342 42. The code for reading a message's header from the spool was incrementing
5343 $received_count, but never initializing it. This meant that the value was
5344 incorrect (doubled) while delivering a message in the same process in which
5345 it was received. In the most common configuration of Exim, this never
5346 happens - a fresh exec is done - but it can happen when
5347 deliver_drop_privilege is set.
5349 43. When Exim logs an SMTP synchronization error - client data sent too soon -
5350 it now includes up to 150 characters of the unexpected data in the log
5353 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
5354 and building data strings. The size of both of these buffers was 10 000
5355 bytes - far larger than anybody would *ever* want, thought I. Needless to
5356 say, somebody hit the limit. I have increased the maximum line length to
5357 20 000 and the maximum data length of concatenated lines to 100 000. I have
5358 also fixed two bugs, because there was no checking on these buffers. Tsk,
5359 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
5360 buffer is too small.
5362 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
5363 lsearch lookups. Now it does.
5365 46. When parsing a route_list item in a manualroute router, a fixed-length
5366 buffer was used for the list of hosts. I made this 1024 bytes long,
5367 thinking that nobody would ever have a list of hosts that long. Wrong.
5368 Somebody had a whole pile of complicated expansion conditions, and the
5369 string was silently truncated, leading to an expansion error. It turns out
5370 that it is easier to change to an unlimited length (owing to other changes
5371 that have happened since this code was originally written) than to build
5372 structure for giving a limitation error. The length of the item that
5373 expands into the list of hosts is now unlimited.
5375 47. The lsearch lookup could not handle data where the length of text line was
5376 more than 4095 characters. Such lines were truncated, leading to shortened
5377 data being returned. It should now handle lines of any length.
5379 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
5380 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
5383 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
5384 replacing strings such as BIN_DIRECTORY in a source file: the replacement
5385 no longer happens in comment lines. A list of replacements is now placed
5386 at the head of all of the source files, except those whose only change is
5387 to replace PERL_COMMAND in the very first #! line.
5389 50. Replaced the slow insertion sort in queue.c, for sorting the list of
5390 messages on the queue, with a bottom-up merge sort, using code contributed
5391 by Michael Haardt. This should make operations like -bp somewhat faster on
5392 large queues. It won't affect queue runners, except when queue_run_in_order
5395 51. Installed eximstats 1.31 in the distribution.
5397 52. Added support for SRV lookups to the dnslookup router.
5399 53. If an ACL referred to $message_body or $message_body_end, the value was not
5400 reset for any messages that followed in the same SMTP session.
5402 54. The store-handling optimization for building very long strings was not
5403 differentiating between the different store pools. I don't think this
5404 actually made any difference in practice, but I've tidied it.
5406 55. While running the routers to verify a sender address, $sender_address
5407 was still set to the sender address. This is wrong, because when routing to
5408 send a bounce to the sender, it would be empty. Therefore, I have changed
5409 it so that, while verifying a sender address, $sender_address is set to <>.
5410 (There is no change to what happens when verifying a recipient address.)
5412 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
5413 target A or AAAA records (if not already returned) without resetting the
5414 qualify_single or search_parents options of the DNS resolver. These are
5415 inappropriate in this case because the targets of MX and SRV records must
5416 be FQDNs. A broken DNS record could cause trouble if it happened to have a
5417 target that, when qualified, matched something in the local domain. These
5418 two options are now turned off when doing these lookups.
5420 57. It seems that at least some releases of Reiserfs (which does not have the
5421 concept of a fixed number of inodes) returns zero and not -1 for the
5422 number of available inodes. This interacted badly with check_spool_inodes,
5423 which assumed that -1 was the "no such thing" setting. What I have done is
5424 to check that the total number of inodes is greater than zero before doing
5425 the test of how many are available.
5427 58. When a "warn" ACL statement has a log_message modifier, the message is
5428 remembered, and not repeated. This is to avoid a lot of repetition when a
5429 message has many recipients that cause the same warning to be written.
5430 Howewer, Exim was preserving the list of already written lines for an
5431 entire SMTP session, which doesn't seem right. The memory is now reset if a
5432 new message is started.
5434 59. The "rewrite" debugging flag was not showing the result of rewriting in the
5435 debugging output unless log_rewrite was also set.
5437 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
5438 of (int)(handle) when we know that handle contains (void *)(-1).
5440 61. The Exim daemon panic-logs an error return when it closes the incoming
5441 connection. However "connection reset by peer" seems to be common, and
5442 isn't really an error worthy of noting specially, so that particular error
5445 62. When Exim is trying to find all the local interfaces, it used to panic and
5446 die if the ioctl to get the interface flags failed. However, it seems that
5447 on at least one OS (Solaris 9) it is possible to have an interface that is
5448 included in the list of interfaces, but for which you get a failure error
5449 for this call. This happens when the interface is not "plumbed" into a
5450 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
5451 failure of the "get flags" call assumes that the interface is down.
5453 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
5454 makes life easier for people who are doing arithmetic on fields extracted
5455 from dates, where you often get leading zeros that should not be
5456 interpreted as octal.
5458 64. Added qualify_domain to the redirect router, to override the global
5461 65. If a pathologically long header line contained very many addresses (the
5462 report of this problem mentioned 10 000) and each of them was rewritten,
5463 Exim could use up a very large amount of memory. (It kept on making new
5464 copies of the header line as it rewrote, and never released the old ones.)
5465 At the expense of a bit more processing, the header rewriting function has
5466 been changed so that it no longer eats memory in this way.
5468 66. The generation of the Received: header has been moved from the time that a
5469 message starts to be received, to the time that it finishes. The timestamp
5470 in the Received: header should now be very close to that of the <= log
5471 line. There are two side-effects of this change:
5473 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
5474 logged header lines no longer include the local Received: line, because
5475 it has not yet been created. The same applies to a copy of the message
5476 that is returned to a non-SMTP sender when a message is rejected.
5478 (b) When a filter file is tested using -bf, no additional Received: header
5479 is added to the test message. After some thought, I decided that this
5482 This change does not affect the value of $received_for. It is still set
5483 after address rewriting, but before local_scan() is called.
5485 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
5487 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
5488 gave an unhelpful panic error message, and a defer error. I have managed to
5489 change this behaviour so that it now rejects any supplied certificate,
5490 which seems right, as the list of acceptable certificates is empty.
5492 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
5493 gave an unhelpful defer error. I have not managed to make this reject any
5494 supplied certificates, but the error message it gives is "no certificate
5495 supplied", which is not helpful.
5497 70. exigrep's output now also includes lines that are not associated with any
5498 message, but which match the given pattern. Implemented by a patch from
5499 Martin Sluka, which also tidied up the Perl a bit.
5501 71. Recipient callout verification, like sender verification, was using <> in
5502 the MAIL FROM command. This isn't really the right thing, since the actual
5503 sender may affect whether the remote host accepts the recipient or not. I
5504 have changed it to use the actual sender in the callout; this means that
5505 the cache record is now keyed on a recipient/sender pair, not just the
5506 recipient address. There doesn't seem to be a real danger of callout loops,
5507 since a callout by the remote host to check the sender would use <>.
5508 [SEE ABOVE: changed after hitting problems.]
5510 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
5511 temporary errors. However, in the case of such a code being given after
5512 the end of a data transmission (i.e. after ".") Exim was failing to write
5513 a retry record for the message. (Yes, there was some broken host that was
5514 actually sending 8xx at this point.)
5516 73. An unknown lookup type in a host list could cause Exim to panic-die when
5517 the list was checked. (An example that provoked this was putting <; in the
5518 middle of a list instead of at the start.) If this happened during a DATA
5519 ACL check, a -D file could be left lying around. This kind of configuration
5520 error no longer causes Exim to die; instead it causes a defer errror. The
5521 incident is still logged to the main and panic logs.
5523 74. Buglet left over from Exim 3 conversion. The message "too many messages
5524 in one connection" was written to the rejectlog but not the mainlog, except
5525 when address rewriting (yes!) was being logged.
5527 75. Added write_rejectlog option.
5529 76. When a system filter was run not as root (that is, when system_filter_user
5530 was set), the values of the $n variables were not being returned to the
5531 main process; thus, they were not subsequently available in the $sn
5534 77. Added +return_path_on_delivery log selector.
5536 78. A connection timeout was being treated differently from recipients deferred
5537 when testing hosts_max_try with a message that was older than the host's
5538 retry timeout. (The host should not be counted, thus allowing all hosts to
5539 be tried at least once before bouncing.) This may have been the cause of an
5540 occasionally reported bug whereby a message would remain on the queue
5541 longer than the retry timeout, but would be bounced if a delivery was
5542 forced. I say "may" because I never totally pinned down the problem;
5543 setting up timeout/retry tests is difficult. See also the next item.
5545 79. The ultimate address timeout was not being applied to errors that involved
5546 a combination of host plus message (for example, a timeout on a MAIL
5547 command). When an address resolved to a number of possible hosts, and they
5548 were not all tried for each delivery (e.g. because of hosts_max_try), a
5549 message could remain on the queue longer than the retry timeout.
5551 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
5554 81. Fixed an obscure SMTP outgoing bug which required at least the following
5555 conditions: (a) there was another message waiting for the same server;
5556 (b) the server returned 5xx to all RCPT commands in the first message so
5557 that the message was not completed; (c) the server dropped the connection
5558 or gave a negative response to the RSET that Exim sends to abort the
5559 transaction. The observed case was a dropped connection after DATA that had
5560 been sent in pipelining mode. That is, the server had advertised PIPELINING
5561 but was not implementing it correctly. The effect of the bug was incorrect
5562 behaviour, such as trying another host, and this could lead to a crash.
5568 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
5569 and daemon.c were passed as pointers to ints; they should have been
5570 pointers to socklen_t variables (which are typically unsigned ints).
5572 2. Some signed/unsigned type warnings in the os.c file for Linux have been
5575 3. Fixed a really odd bug that affected only the testing scheme; patching a
5576 certain fixed string in the binary changed the value of another string that
5577 happened to be identical to the end of the original first string.
5579 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
5580 name", it returns that address as the IP address. On some operating
5581 systems (e.g. Solaris), it also passes back the IP address string as the
5582 "host name". However, on others (e.g. Linux), it passes back an empty
5583 string. Exim wasn't checking for this, and was changing the host name to an
5584 empty string, assuming it had been canonicized.
5586 5. Although rare, it is permitted to have more than one PTR record for a given
5587 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
5588 all the names associated with an address, because they do in Solaris.
5589 However, it seems that they do not in Linux for data that comes from the
5590 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
5591 I found this out when I moved to a new Linux workstation and tried to run
5592 the Exim test suite.
5594 To get round this problem I have changed the code so that it now does its
5595 own call to the DNS to look up PTR records when searching for a host name.
5596 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
5597 addresses that are only in /etc/hosts are still found.
5599 This behaviour is, however, controlled by an option called host_lookup_
5600 order, which defaults to "bydns:byaddr". If people want to use the other
5601 order, or indeed, just use one or the other means of lookup, they can
5602 specify it in this variable.
5604 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
5605 some operating systems, this comes back from gethostbyaddr() as an empty
5606 string, and this is what Exim used to test for. However, it seems that in
5607 other systems, "." is yielded. Exim now tests for this case too.
5609 7. The values of check_spool_space and check_log_space are now held internally
5610 as a number of kilobytes instead of an absolute number of bytes. If a
5611 numbers is specified without 'K' or 'M', it is rounded up to the nearest
5612 kilobyte. This means that much larger values can be stored.
5614 8. Exim monitor: an attempt to get the action menu when not actually pointing
5615 at a message produces an empty menu entitled "No message selected". This
5616 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
5617 no entries in it ("Shell widget menu has zero width and/or height"). So I
5618 have added a single, blank menu entry in this case.
5620 9. Added ${quote_local_part.
5622 10. MIME decoding is now applied to the contents of Subject: header lines when
5625 11. Now that a reference to $sender_host_address automatically causes a reverse
5626 lookup to occur if necessary (4.13/18), there is no need to arrange for a
5627 host lookup before query-style lookups in lists that might use this
5628 variable. This has therefore been abolished, and the "net-" prefix is no
5629 longer necessary for query-style lookups.
5631 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
5632 have been a typo for LFLAGS, so it has been changed.
5634 13. The install script calls Exim with "-C /dev/null" in order to find the
5635 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
5636 to be output. Howeve, since Exim outputs its version number before the
5637 error, it didn't break the script. It just looked ugly. I fixed this by
5638 always allowing "-C /dev/null" if the caller is root.
5640 14. Ignore overlarge ACL variable number when reading spool file - insurance
5641 against a later release with more variables having written the file.
5643 15. The standard form for an IPv6 address literal was being rejected by EHLO.
5644 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5645 this, as well as the form without the "IPv6" on the front.
5647 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
5648 OS/Makefile-Darwin file.
5650 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
5651 only to LDAP libraries that do not have LDAP_OPT_DEREF.
5653 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
5654 variable. However, some OS use ints for inodes. I've added cast to long int
5655 to get rid of the compiler warning.
5657 19. I had forgotten to lock out "/../" in configuration file names when
5658 ALT_CONFIG_PREFIX was set.
5660 20. Routers used for verification do not need to specify transports. However,
5661 if such a router generated a host list, and callout was configured, Exim
5662 crashed, because it could not find a port number from the (non-existent)
5663 transport. It now assumes port 25 in this circumstance.
5665 21. Added the -t option to exigrep.
5667 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
5668 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
5669 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
5670 them is set, LOOKUP_LSEARCH is forced.
5672 23. "exim -bV" now outputs a list of lookups that are included in the binary.
5674 24. Added sender and host information to the "rejected by local_scan()" log
5675 line; previously there was no indication of these.
5677 25. Added .include_if_exists.
5679 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
5680 Linux. It turns out that not all file systems support this. Apparently some
5681 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
5682 do it.) To cope with this, the error EINVAL, which means that sync-ing is
5683 not supported on the file descriptor, is now ignored when Exim is trying to
5684 sync a directory. This applies only to Linux.
5686 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
5688 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
5689 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
5692 29. Added support for maildirsize files from supplied patch (modified a bit).
5694 30. The use of :fail: followed by an empty string could lead Exim to respond to
5695 sender verification failures with (e.g.):
5697 550 Verification failed for <xxx>
5698 550 Sender verify failed
5700 where the first response line was missing the '-' that indicates it is not
5701 the final line of the response.
5703 31. The loop for finding the name of the user that called Exim had a hardwired
5704 limit of 10; it now uses the value of finduser_retries, which is used for
5705 all other user lookups.
5707 32. Added $received_count variable, available in data and not_smtp ACLs, and at
5710 33. Exim was neglecting to zero errno before one call of strtol() when
5711 expanding a string and expecting an integer value. On some systems this
5712 resulted in spurious "integer overflow" errors. Also, it was casting the
5713 result into an int without checking.
5715 34. Testing for a connection timeout using "timeout_connect" in the retry rules
5716 did not work. The code looks as if it has *never* worked, though it appears
5717 to have been documented since at least releast 1.62. I have made it work.
5719 35. The "timeout_DNS" error in retry rules, also documented since at least
5720 1.62, also never worked. As it isn't clear exactly what this means, and
5721 clearly it isn't a major issue, I have abolished the feature by treating it
5722 as "timeout", and writing a warning to the main and panic logs.
5724 36. The display of retry rules for -brt wasn't always showing the error code
5727 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
5728 timeout_connect_A, timeout_connect_MX.
5730 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
5731 to the empty sender.
5733 39. The daemon was not analysing the content of -oX till after it had closed
5734 stderr and disconnected from the controlling terminal. This meant that any
5735 syntax errors were only noted on the panic log, and the return code from
5736 the command was 0. By re-arranging the code a little, I've made the
5737 decoding happen first, so such errors now appear on stderr, and the return
5738 code is 1. However, the actual setting up of the sockets still happens in
5739 the disconnected process, so errors there are still only recorded on the
5742 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
5743 connections (as happens on some IP stacks) was logged at start up time as
5744 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
5745 it from "IPv6 and IPv4", which means that two separate sockets are being
5748 41. The debug output for gethostbyname2() or getipnodebyname() failures now
5749 says whether AF_INET or AF_INET6 was passed as an argument.
5751 42. Exiwhat output was messed up when time zones were included in log
5754 43. Exiwhat now gives more information about the daemon's listening ports,
5755 and whether -tls-on-connect was used.
5757 44. The "port" option of the smtp transport is now expanded.
5759 45. A "message" modifier in a "warn" statement in a non-message ACL was being
5760 silently ignored. Now an error message is written to the main and panic
5763 46. There's a new ACL modifier called "logwrite" which writes to a log file
5764 as soon as it is encountered.
5766 47. Added $local_user_uid and $local_user_gid at routing time.
5768 48. Exim crashed when trying to verify a sender address that was being
5771 49. Exim was recognizing only a space character after ".include". It now also
5772 recognizes a tab character.
5774 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
5775 extracting the relevant information from the specification. The man page no
5776 longer contains scrambled data for the -d option, and I've added a section
5777 at the front about calling Exim under different names.
5779 51. Added "extra_headers" argument to the "mail" command in filter files.
5781 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
5784 53. Installed eximstats 1.29.
5786 54. Added transport_filter_timeout as a generic transport option.
5788 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
5789 Cc: header lines. This was required by RFC 822, but it not required by RFC
5792 56. Exim used to add From:, Date:, and Message-Id: header lines to any
5793 incoming messages that did not have them. Now it does so only if the
5794 message originates locally, that is, if there is no associated remote host
5795 address. When Resent- header lines are present, this applies to the Resent-
5796 lines rather than the non-Resent- lines.
5798 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
5799 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
5801 58. Messages for configuration errors now include the name of the main
5802 configuration file - useful now that there may be more than one file in a
5803 list (.included file names were always shown).
5805 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
5806 for those rare installations that do not start the daemon as root or run it
5807 setuid root. I've cut out the call to initgroups() if the daemon is not
5810 60. The Exim user and group can now be bound into the binary as text strings
5811 that are looked up at the start of Exim's processing.
5813 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
5815 62. Added $mailstore_basename variable.
5817 63. Installed patch to sieve.c from Michael Haardt.
5819 64. When Exim failed to open the panic log after failing to open the main log,
5820 the original message it was trying to log was written to stderr and debug
5821 output, but if they were not available (the usual case in production), it
5822 was lost. Now it is written to syslog before the two lines that record the
5823 failures to open the logs.
5825 65. Users' Exim filters run in subprocesses under the user's uid. It is
5826 possible for a "deliver" command or an alias in a "personal" command to
5827 provoke an address rewrite. If logging of address rewriting is configured,
5828 this fails because the process is not running as root or exim. There may be
5829 a better way of dealing with this, but for the moment (because 4.30 needs
5830 to be released), I have disabled address rewrite logging when running a
5831 filter in a non-root, non-exim process.
5837 1. The buildconfig auxiliary program wasn't quoting the value set for
5838 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
5839 not defined. This bug was masked in 4.22 by the effect that was fixed in
5842 2. Some messages that were rejected after a message id was allocated were
5843 shown as "incomplete" by exigrep. It no longer does this for messages that
5844 are rejected by local_scan() or the DATA or non-SMTP ACLs.
5846 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
5847 have allow_domain_literals set, the ID did not get logged in the <= line.
5848 Domain literals are now always recognized in Message-ID: header lines.
5850 4. The first argument for a ${extract expansion item is the key name or field
5851 number. Leading and trailing spaces in this item were not being ignored,
5852 causing some misleading effects.
5854 5. When deliver_drop_privilege was set, single queue runner processes started
5855 manually (i.e. by the command "exim -q") or by the daemon (which uses the
5856 same command in the process it spins off) were not dropping privilege.
5858 6. When the daemon running as "exim" started a queue runner, it always
5859 re-executed Exim in the spun-off process. This is a waste of effort when
5860 deliver_drop_privilege is set. The new process now just calls the
5861 queue-runner function directly.
5867 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
5870 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
5871 ignored. Though the use of -oP was forcing the writing of a pid file, it
5872 was always written to the default place.
5874 3. If the message "no IP address found for host xxxx" is generated during
5875 incoming verification, it is now followed by identification of the incoming
5876 connection (so you can more easily find what provoked it).
5878 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
5880 5. Added some features to "harden" Exim a bit more against certain attacks:
5882 (a) There is now a build-time option called FIXED_NEVER_USERS that can
5883 be put in Local/Makefile. This is like the never_users runtime option,
5884 but it cannot be overridden. The default setting is "root".
5886 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
5887 prefix string with which any file named in a -C command line option
5890 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
5891 is retained for -C and -D only if the caller of Exim is root. Without
5892 it, the exim user may also use -C and -D and retain privilege.
5894 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
5895 command line option is disabled.
5897 6. Macro names set by the -D option must start with an upper case letter, just
5898 like macro names defined in the configuration file.
5900 7. Added "dereference=" facility to LDAP.
5902 8. Two instances of the typo "uknown" in the source files are fixed.
5904 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
5905 the Configure-Makefile script screwed up while processing it.
5907 10. Incorporated PCRE 4.4.
5909 11. The SMTP synchronization check was not operating right at the start of an
5910 SMTP session. For example, it could not catch a HELO sent before the client
5911 waited for the greeting. There is now a check for outstanding input at the
5912 point when the greeting is written. Because of the duplex, asynchronous
5913 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
5914 way, but not yet received, when the check is performed.
5916 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
5917 on TCP/IP sockets, because this apparently causes some broken clients to
5920 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
5921 unchanged) from the Cygwin maintainer.
5923 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
5924 was supported instead of testing for mbx. Effectively a typo.
5926 15. The spa authenticator server code was not checking that the input it
5927 received was valid base64.
5929 16. The debug output line for the "set" modifier in ACLs was not showing the
5930 name of the variable that was being set.
5932 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
5934 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
5935 lookup on its own. Something else was needed to trigger the lookup. For
5936 example, a match in host_lookup or the need for a host name in a host list.
5937 Now, if $sender_host_name is referenced and the host name has not yet been
5938 looked up, a lookup is performed. If the lookup fails, the variable remains
5939 empty, and $host_lookup_failed is set to "1".
5941 19. Added "eqi" as a case-independent comparison operator.
5943 20. The saslauthd authentication condition could segfault if neither service
5944 nor realm was specified.
5946 21. If an overflowing value such as "2048M" was set for message_size_limit, the
5947 error message that was logged was misleading, and incoming SMTP
5948 connections were dropped. The message is now more accurate, and temporary
5949 errors are given to SMTP connections.
5951 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
5952 (except RSET) with a 421 error, until QUIT is received. However, it was
5953 failing to send a response to QUIT.
5955 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
5956 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
5957 ACL is now run after the helo_try_verify_hosts code.
5959 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
5960 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
5961 case-independent, so other case variants are also recognized). Apparently
5962 some systems use these upper case variants.
5964 25. If more than two messages were waiting for the same host, and a transport
5965 filter was specified for the transport, Exim sent two messages over the
5966 same TCP/IP connection, and then failed with "socket operation on non-
5967 socket" when it tried to send the third.
5969 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
5971 27. The extern definition of crypt16() in expand.c was not being excluded when
5972 the OS had its own crypt16() function.
5974 28. Added bounce_return_body as a new option, and bounce_return_size_limit
5975 as a preferred synonym for return_size_limit, both as an option and as an
5978 29. Added LIBS=-liconv to OS/Makefile-OSF1.
5980 30. Changed the default configuration ACL to relax the local part checking rule
5981 for addresses that are not in any local domains. For these addresses,
5982 slashes and pipe symbols are allowed within local parts, but the sequence
5983 /../ is explicitly forbidden.
5985 31. SPA server authentication was not clearing the challenge buffer before
5988 32. log_message in a "warn" ACL statement was writing to the reject log as
5989 well as to the main log, which contradicts the documentation and doesn't
5990 seem right (because no rejection is happening). So I have stopped it.
5992 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
5993 However, I am unable to do any testing of this.
5995 34. Fixed an infelicity in the appendfile transport. When checking directories
5996 for a mailbox, to see if any needed to be created, it was accidentally
5997 using path names with one or more superfluous leading slashes; tracing
5998 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
6000 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
6001 contents are added to the log line that is written for every discarded
6002 recipient. (Previously a log_message setting was ignored.)
6004 36. The ${quote: operator now quotes the string if it is empty.
6006 37. The install script runs exim in order to find its version number. If for
6007 some reason other than non-existence or emptiness, which it checks, it
6008 could not run './exim', it was installing it with an empty version number,
6009 i.e. as "exim-". This error state is now caught, and the installation is
6012 38. An argument was missing from the function that creates an error message
6013 when Exim fails to connect to the socket for saslauthd authentication.
6014 This could cause Exim to crash, or give a corrupted message.
6016 39. Added isip, isip4, and isip6 to ${if conditions.
6018 40. The ACL variables $acl_xx are now saved with the message, and can be
6019 accessed later in routers, transports, and filters.
6021 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
6022 strings in the file are not string-expanded.
6024 42. If a MAIL command specified a SIZE value that was too large to fit into an
6025 int variable, the check against message_size_limit failed. Such values are
6026 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
6027 day this will have to be increased, but I don't think I want to be around
6028 when emails are that large.
6035 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
6036 iconv() is not standard in FreeBSD.
6038 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
6039 IPv6 enabled. The observed symptom was a segmentation fault on return from
6040 the function os_common_find_running_interfaces() in src/os.c.
6042 3. In the check_special_case() function in daemon.c I had used "errno" as an
6043 argument name, which causes warnings on some systems. This was basically a
6044 typo, since it was named "eno" in the comments!
6046 4. The code that waits for the clock to tick (at a resolution of some fraction
6047 of a second) so as to ensure message-id uniqueness was always waiting for
6048 at least one whole tick, when it could have waited for less. [This is
6049 almost certainly not relevant at current processor speeds, where it is
6050 unlikely to ever wait at all. But we try to future-proof.]
6052 5. The function that sleeps for a time interval that includes fractions of a
6053 second contained a race. It did not block SIGALRM between setting the
6054 timer, and suspending (a couple of lines later). If the interval was short
6055 and the sigsuspend() was delayed until after it had expired, the suspension
6056 never ended. On busy systems this could lead to processes getting stuck for
6059 6. Some uncommon configurations may cause a lookup to happen in a queue runner
6060 process, before it forks any delivery processes. The open lookup caching
6061 mechanism meant that the open file or database connection was passed into
6062 the delivery process. The problem was that delivery processes always tidy
6063 up cached lookup data. This could cause a problem for the next delivery
6064 process started by the queue runner, because the external queue runner
6065 process does not know about the closure. So the next delivery process
6066 still has data in the lookup cache. In the case of a file lookup, there was
6067 no problem because closing a file descriptor in a subprocess doesn't affect
6068 the parent. However, if the lookup was caching a connection to a database,
6069 the connection was closed, and the second delivery process was likely to
6070 see errors such as "PGSQL: query failed: server closed the connection
6071 unexpectedly". The problem has been fixed by closing all cached lookups
6072 in a queue runner before running a delivery process.
6074 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
6075 seem to have the "const" qualifier which it has on other OS. I've
6078 8. Change 4.21/2 was too strict. It is only if there are two authenticators
6079 *of the same type* (client or server) with the same public name that an
6080 error should be diagnosed.
6082 9. When Exim looked up a host name for an IP address, but failed to find the
6083 original IP address when looking up the host name (a safety check), it
6084 output the message "<ip address> does not match any IP for NULL", which was
6085 confusing, to say the least. The bug was that the host name should have
6086 appeared instead of "NULL".
6088 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
6089 user that is built into the binary, and the -C or -D options is used, root
6090 privilege is dropped before the configuration file is read. In addition,
6091 logging is switched to stderr instead of the normal log files. If the
6092 configuration then re-defines the Exim user, the unprivileged environment
6093 is probably not what is expected, so Exim logs a panic warning message (but
6096 However, if deliver_drop_privilege is set, the unprivileged state may well
6097 be exactly what is intended, so the warning has been cut out in that case,
6098 and Exim is allowed to try to write to its normal log files.
6104 1. smtp_return_error_details was not giving details for temporary sender
6105 or receiver verification errors.
6107 2. Diagnose a configuration error if two authenticators have the same public
6110 3. Exim used not to create the message log file for a message until the first
6111 delivery attempt. This could be confusing when incoming messages were held
6112 for policy or load reasons. The message log file is now created at the time
6113 the message is received, and an initial "Received" line is written to it.
6115 4. The automatically generated man page for command line options had a minor
6116 bug that caused no ill effects; however, a more serious problem was that
6117 the procedure for building the man page automatically didn't always
6118 operate. Consequently, release 4.20 contains an out-of-date version. This
6119 shouldn't happen again.
6121 5. When building Exim with embedded Perl support, the script that builds the
6122 Makefile was calling 'perl' to find its compile-time parameters, ignoring
6123 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
6125 6. The freeze_tell option was not being used for messages that were frozen on
6126 arrival, either by an ACL or by local_scan().
6128 7. Added the smtp_incomplete_transaction log selector.
6130 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
6131 was accepting AUTH without a new EHLO.
6133 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
6134 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
6137 10. Logging of TCP/IP connections (when configured) now happens in the main
6138 daemon process instead of the child process, so that the TCP/IP connection
6139 count is more accurate (but it can never be perfect).
6141 11. The use of "drop" in a nested ACL was not being handled correctly in the
6142 outer ACL. Now, if condition failure induced by the nested "drop" causes
6143 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
6144 or "require"), the connection is dropped.
6146 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
6147 that yield "discard" can now be used with an "accept" or a "discard" verb,
6148 but an error is generated for any others (because I can't see a useful way
6149 to define what should happen).
6151 13. When an ACL is read dynamically from a file (or anywhere else), the lines
6152 are now processed in the same way as lines in the Exim configuration file.
6153 In particular, continuation lines are supported.
6155 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
6157 15. Added -ti meaning -t -i.
6159 16. Check for letters, digits, hyphens, and dots in the names of dnslist
6160 domains, and warn by logging if others are found.
6162 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
6163 returned from GIFCONF when Exim is trying to find the list of interfaces on
6164 a host. The code in os.c has been modified to copy each ifreq to an aligned
6165 structure in all cases.
6167 Also, in some cases, the returned ifreq's were being copied to a 'struct
6168 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
6169 means the last couple of bytes of an IPv6 address could be chopped if the
6170 ifreq contained only a normal sockaddr (14 bytes storage).
6172 18. Named domain lists were not supported in the hosts_treat_as_local option.
6173 An entry such as +xxxx was not recognized, and was treated as a literal
6176 19. Ensure that header lines added by a DATA ACL are included in the reject log
6177 if the ACL subsequently rejects the message.
6179 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
6180 MD5 (which is deprecated).
6182 21. When testing a filter file using -bf, Exim was writing a message when it
6183 took the sender from a "From " line in the message, but it was not doing so
6184 when it took $return_path from a Return-Path: header line. It now does.
6186 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
6187 with a valid header line field name (a series of printing characters
6188 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
6190 23. Changed "disc" in the source to "disk" to conform to the documentation and
6191 the book and for uniformity.
6193 24. Ignore Sendmail's -Ooption=value command line item.
6195 25. When execve() failed while trying to run a command in a pipe transport,
6196 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
6197 could be confused with a return value of 69 from the command itself. This
6198 has been changed to 127, the value the shell returns if it is asked to run
6199 a non-existent command. The wording for the related log line suggests a
6200 non-existent command as the problem.
6202 26. If received_header_text expands to an empty string, do not add a Received:
6203 header line to the message. (Well, it adds a token one on the spool, but
6204 marks it "old" so that it doesn't get used or transmitted.)
6206 27. Installed eximstats 1.28 (addition of -nt option).
6208 28. There was no check for failure on the call to getsockname() in the daemon
6209 code. This can fail if there is a shortage of resources on the system, with
6210 ENOMEM, for example. A temporary error is now given on failure.
6212 29. Contrary to the C standard, it seems that in some environments, the
6213 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
6214 program. Exim now does this explicitly; it affects the formatting of
6215 timestamps using strftime().
6217 30. If exiqsumm was given junk data, it threw up some uninitialized variable
6218 complaints. I've now initialized all the variables, to avoid this.
6220 32. Header lines added by a system filter were not being "seen" during
6221 transport-time rewrites.
6223 33. The info_callback() function passed to OpenSSL is set up with type void
6224 (*)(SSL *, int, int), as described somewhere. However, when calling the
6225 function (actually a macro) that sets it up, the type void(*)() is
6226 expected. I've put in a cast to prevent warnings from picky compilers.
6228 34. If a DNS black list lookup found a CNAME record, but there were no A
6229 records associated with the domain it pointed at, Exim crashed.
6231 35. If a DNS black list lookup returned more than one A record, Exim ignored
6232 all but the first. It now scans all returned addresses if a particular IP
6233 value is being sought. In this situation, the contents of the
6234 $dnslist_value variable are a list of all the addresses, separated by a
6237 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
6238 to accept a host name and all its aliases if the forward lookup for any of
6239 them yielded the IP address of the incoming connection. Now it accepts only
6240 those names whose forward lookup yields the correct IP address. Any other
6241 names are discarded. This closes a loophole whereby a rogue DNS
6242 administrator could create reverse DNS records to break through a
6243 wildcarded host restriction in an ACL.
6245 37. If a user filter or a system filter that ran in a subprocess used any of
6246 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
6247 the wrong values were passed to the pipe command ($thisaddress had the
6248 value of $0, $0 had the value of $1, etc). This bug was introduced by
6249 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
6251 38. Improved the line breaking for long SMTP error messages from ACLs.
6252 Previously, if there was no break point between 40 and 75 characters, Exim
6253 left the rest of the message alone. Two changes have been made: (a) I've
6254 reduced the minimum length to 35 characters; (b) if it can't find a break
6255 point between 35 and 75 characters, it looks ahead and uses the first one
6256 that it finds. This may give the occasional overlong line, but at least the
6257 remaining text gets split now.
6259 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
6260 file descriptors might be low, and that setting 1000 would always raise it.
6261 It turns out that in some environments, the limit is already over 1000 and
6262 that lowering it causes trouble. So now Exim takes care not to decrease it.
6264 40. When delivering a message, the value of $return_path is set to $sender_
6265 address at the start of routing (routers may change the value). By an
6266 oversight, this default was not being set up when an address was tested by
6267 -bt or -bv, which affected the outcome if any router or filter referred to
6270 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
6271 log only when "message" or "log_message" are set. However, if one of the
6272 conditions was an address verification, or a call to a nested ACL, the
6273 messages generated by the underlying test were being passed through. This
6274 no longer happens. The underlying message is available in $acl_verify_
6275 message for both "message" and "log_message" expansions, so it can be
6276 passed through if needed.
6278 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
6279 new expansion $bh_ to give the encoded byte string without charset
6280 translation. Translation happens only if iconv() is available; HAVE_ICONV
6281 indicates this at build time. HEADERS_CHARSET gives the charset to
6282 translate to; headers_charset can change it in the configuration, and
6283 "headers charset" can change it in an individual filter file.
6285 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
6286 that creates RFC 2047 encoded "words" labels them as that charset instead
6287 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
6288 expansion operator; (ii) when Exim creates a From: line for a local
6289 message; (iii) when a header line is rewritten to include a "phrase" part.
6291 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
6292 buggy, causing it to skip the first lines of messages whose message ID
6293 ended in 'D'. This would not have bitten before Exim release 4.14, because
6294 message IDs were unlikely to end in 'D' before then. The effect was to have
6295 incorrect size information for certain domains.
6297 45. #include "config.h" was missing at the start of the crypt16.c module. This
6298 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
6301 46. If there was a timeout during a "random" callout check, Exim treated it as
6302 a failure of the random address, and carried on sending RSET and the real
6303 address. If the delay was just some slowness somewhere, the response to the
6304 original RCPT would be taken as a response to RSET and so on, causing
6305 mayhem of various kinds.
6307 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
6308 when I implemented it. It didn't allow for the fact that some option values
6309 may legitimatetly be negative (e.g. size_addition), and it didn't even do
6310 the right test for positive values.
6312 48. Domain names in DNS records are case-independent. Exim always looks them up
6313 in lower case. Some resolvers return domain names in exactly the case they
6314 appear in the zone file, that is, they may contain uppercase letters. Not
6315 all resolvers do this - some return always lower case. Exim was treating a
6316 change of case by a resolver as a change of domain, similar to a widening
6317 of a domain abbreviation. This triggered its re-routing code and so it was
6318 trying to route what was effectively the same domain again. This normally
6319 caused routing to fail (because the router wouldn't handle the domain
6320 twice). Now Exim checks for this case specially, and just changes the
6321 casing of the domain that it ultimately uses when it transmits the message
6324 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
6327 50. If a filter generated a file delivery with a non-absolute name (possible if
6328 no home directory exists for the router), the forbid_file option was not
6331 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
6332 the existing equality matching.
6334 52. Exim was using ints instead of ino_t variables in some places where it was
6335 dealing with inode numbers.
6337 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
6338 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
6339 called TMPDIR, and if it finds it is different, it changes its value.
6341 54. The smtp_printf() function is now made available to local_scan() so
6342 additional output lines can be written before returning. There is also an
6343 smtp_fflush() function to enable the detection of a dropped connection.
6344 The variables smtp_input and smtp_batched_input are exported to
6347 55. Changed the default runtime configuration: the message "Unknown user"
6348 has been removed from the ACL, and instead placed on the localuser router,
6349 using the cannot_route_message feature. This means that any verification
6350 failures that generate their own messages won't get overridden. Similarly,
6351 the "Unrouteable address" message that was in the ACL for unverifiable
6352 relay addresses has also been removed.
6354 56. Added hosts_avoid_esmtp to the smtp transport.
6356 57. The exicyclog script was not checking for the esoteric option
6357 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
6358 will work only if exicyclog is run under the appropriate euid.
6360 58. Following a discussion on the list, the rules by which Exim recognises line
6361 endings on incoming messages have been changed. The -dropcr and drop_cr
6362 options are now no-ops, retained only for backwards compatibility. The
6363 following line terminators are recognized: LF CRLF CR. However, special
6364 processing applies to CR:
6366 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
6367 nor a local message in the state where . is a terminator.
6369 (ii) If a bare CR is encountered in a header line, an extra space is added
6370 after the line terminator so as not to end the header. The reasoning
6371 behind this is that bare CRs in header lines are most likely either
6372 to be mistakes, or people trying to play silly games.
6374 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
6375 was being incorrectly given as 18 bytes larger than it should have been.
6376 This is a VOB (very old bug).
6378 60. This may never have affected anything current, but just in case it has:
6379 When the local host is found other than at the start of a list of hosts,
6380 the local host, those with the same MX, and any that follow, are discarded.
6381 When the list in question was part of a longer list of hosts, the following
6382 hosts (not currently being processed) were also being discarded. This no
6383 longer happens. I'm not sure if this situation could ever has previously
6386 61. Added the "/MX" feature to lists of hosts in the manualroute and query
6389 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
6390 header. This is something that is recommended in a new Internet Draft, and
6391 is something that is documented as being done by Sendmail. There are two
6392 possible values. For messages generated by the autoreply transport, Exim
6395 Auto-Submitted: auto-replied
6397 whereas for all other generated messages (e.g. bounces) it adds
6399 Auto-Submitted: auto-generated
6401 63. The "personal" condition in filters now includes a test for the
6402 Auto-Submitted: header. If it contains the string "auto-" the message it
6403 not considered personal.
6405 64. Added rcpt_include_affixes as a generic transport option.
6407 65. Added queue_only_override (default true).
6409 66. Added the syslog_duplication option.
6411 67. If what should have been the first header line of a message consisted of
6412 a space followed by a colon, Exim was mis-interpreting it as a header line.
6413 It isn't of course - it is syntactically invalid and should therefore be
6414 treated as the start of the message body. The misbehaviour could have
6415 caused a number of strange effects, including loss of data in subsequent
6416 header lines, and spool format errors.
6418 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
6419 client host had authenticated. This control can now be exercised by an ACL
6420 for more flexibility.
6422 69. By default, callouts do not happen when testing with -bh. There is now a
6423 variant, -bhc, which does actually run the callout code, including
6424 consulting and updating the callout cache.
6426 70. Added support for saslauthd authentication, courtesy of Alexander
6429 71. If statvfs() failed on the spool or log directories while checking their
6430 size for availability, Exim confusingly gave the error "space shortage".
6431 Furthermore, in debugging mode it crashed with a floating point exception.
6432 These checks are done if check_{spool,log}_{space,inodes} are set, and when
6433 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
6434 serious problem, Exim now writes to the main and panic logs when this
6435 happens, with details of the failure. It then refuses to accept the
6436 incoming message, giving the message "spool directory problem" or "log
6437 directory problem" with a 421 code for SMTP messages.
6439 72. When Exim is about to re-exec itself, it ensures that the file descriptors
6440 0, 1, and 2 exist, because some OS complain for execs without them (see
6441 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
6442 descriptors. However, the code omitted to check that the open succeeded,
6443 causing mysterious errors if for some reason the permissions on /dev/null
6444 got screwed. Now Exim writes a message to the main and panic logs, and
6445 bombs out if it can't open /dev/null.
6447 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
6448 interact so that it is all more flexible. It is supposed to remain
6449 backwards compatible. Also added extra_local_interfaces.
6451 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
6452 to bomb out with an assertion failure - to the client this appears as a
6453 connection drop. This problem occurs in the part of the code that was taken
6454 from the Samba project. Fortunately, the assertion is in a very simple
6455 function, so I have fixed this by reproducing the function inline in the
6456 one place where it is called, and arranging for authentication to fail
6457 instead of killing the process with assert().
6459 75. The SPA client code was not working when the server requested OEM rather
6460 than Unicode encoding.
6462 76. Added code to make require_files with a specific uid setting more usable in
6463 the case where statting the file as root fails - usually a non-root-mounted
6464 NFS file system. When this happens and the failure is EACCES, Exim now
6465 forks a subprocess and does the per-uid checking as the relevant uid.
6467 77. Added process_log_path.
6469 78. If log_file_path was not explicitly set, a setting of check_log_space or
6470 check_log_inodes was ignored.
6472 79. If a space check for the spool or log partitions fails, the incident is now
6473 logged. Of course, in the latter case the data may get lost...
6475 80. Added the %p formatting code to string_format() so that it can be used to
6476 print addresses in debug_print(). Adjusted all the address printing in the
6477 debugging in store.c to use %p rather than %d.
6479 81. There was a concern that a line of code in smtp_in.c could overflow a
6480 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
6481 initially expressed, the concern was not well-founded, because trailing
6482 spaces are removed early. However, if the trailing spaces were followed by
6483 a NULL, they did not get removed, so the overflow was possible. Two fixes
6486 (a) I re-wrote the offending code in a cleaner fashion.
6487 (b) If an incoming SMTP command contains a NULL character, it is rejected
6490 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
6491 runs initgroups(), so that if the Exim user is in any additional groups,
6492 they will be used during message reception.
6498 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob