1 Change log file for Exim from version 4.21
2 -------------------------------------------
6 JH/01 Use SIZE on MAIL FROM in a cutthrough connection, if the destination
7 supports it and a size is available (ie. the sending peer gave us one).
9 JH/02 The obsolete acl condition "demime" is removed (finally, after ten
10 years of being deprecated). The replacements are the ACLs
11 acl_smtp_mime and acl_not_smtp_mime.
13 JH/03 Upgrade security requirements imposed for hosts_try_dane: previously
14 a downgraded non-dane trust-anchor for the TLS connection (CA-style)
15 or even an in-clear connection were permitted. Now, if the host lookup
16 was dnssec and dane was requested then the host is only used if the
17 TLSA lookup succeeds and is dnssec. Further hosts (eg. lower priority
18 MXs) will be tried (for hosts_try_dane though not for hosts_require_dane)
19 if one fails this test.
20 This means that a poorly-configured remote DNS will make it incommunicado;
21 but it protects against a DNS-interception attack on it.
23 JH/04 Bug 1810: make continued-use of an open smtp transport connection
24 non-noisy when a race steals the message being considered.
26 JH/05 If main configuration option tls_certificate is unset, generate a
27 selfsigned certificate for inbound TLS connections.
29 JH/06 Bug 165: hide more cases of password exposure - this time in expansions
30 in rewrites and routers.
32 JH/07 Retire gnutls_require_mac et.al. These were nonfunctional since 4.80
33 and logged a warning sing 4.83; now they are a configuration file error.
38 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
39 and 3.4.4 - once the server is enabled to respond to an OCSP request
40 it does even when not requested, resulting in a stapling non-aware
41 client dropping the TLS connection.
43 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
44 support variable-length bit vectors. No functional change.
46 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
47 The I= interface field on outgoing lines is now after the H= remote
48 host field, same as incoming lines. There is a separate
49 outgoing_interface log selector which allows you to disable the
52 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
53 If not running log_selector +smtp_connection the mainlog would be held
54 open indefinitely after a "too many connections" event, including to a
55 deleted file after a log rotate. Leave the per net connection logging
56 leaving it open for efficiency as that will be quickly detected by the
57 check on the next write.
59 HS/01 Bug 1671: Fix post transport crash.
60 Processing the wait-<transport> messages could crash the delivery
61 process if the message IDs didn't exist for some reason. When
62 using 'split_spool_directory=yes' the construction of the spool
63 file name failed already, exposing the same netto behaviour.
65 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
66 mime_regex ACL conditions.
68 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
69 to DSN fail messages (bounces): remote IP, remote greeting, remote response
70 to HELO, local diagnostic string.
72 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
73 log line to debug. Even when configured with a tls authenticator many
74 client connections are expected to not authenticate in this way, so
75 an authenticate fail is not an error.
77 HS/02 Add the Exim version string to the process info. This way exiwhat
78 gives some more detail about the running daemon.
80 JH/06 Bug 1395: time-limit cacheing of DNS lookups, to the TTL value. This may
81 matter for fast-change records such as DNSBLs.
83 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
84 retry record, even if constant. There may be multiple transports with
85 different interface settings and the retry behaviour needs to be kept
88 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
90 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
92 JH/10 Bug 840: fix log_defer_output option of pipe transport
94 JH/11 Bug 830: use same host for all RCPTS of a message, even under
95 hosts_randomize. This matters a lot when combined with mua_wrapper.
97 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
98 ${quote_pgsql:<string>} operator.
100 JH/13 Bug 1708: avoid misaligned access in cached lookup.
102 JH/14 Change header file name for freeradius-client. Relevant if compiling
103 with Radius support; from the Gentoo tree and checked under Fedora.
105 JH/15 Bug 1712: Introduce $prdr_requested flag variable
107 JH/16 Bug 1714: Permit an empty string as expansion result for transport
108 option transport_filter, meaning no filtering.
110 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
112 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
113 defaults to "*" (all hosts). The variable is now available when not built
114 with TLS, default unset, mainly to enable keeping the testuite sane.
115 If a server certificate is not supplied (via tls_certificate) an error is
116 logged, and clients will find TLS connections fail on startup. Presumably
117 they will retry in-clear.
118 Packagers of Exim are strongly encouraged to create a server certificate
119 at installation time.
121 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
122 with the $config_file variable.
124 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
125 in transport context, after the attempt, and per-recipient. The latter type
126 is per host attempted. The event data is the error message, and the errno
127 information encodes the lookup type (A vs. MX) used for the (first) host,
128 and the trailing two digits of the smtp 4xx reponse.
130 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
131 to write to mainlog (or rejectlog, paniclog) in the window between file
132 creation and permissions/ownership being changed. Particularly affects
133 installations where exicyclog is run as root, rather than exim user;
134 result is that the running daemon panics and dies.
136 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
138 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
139 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
140 "pri" and "weight". Note that the previous implicit priority given by the
141 list order is no longer honoured.
143 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalisation
146 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
147 by defining SUPPORT_SOCKS.
149 JH/26 Move PROXY support from Experimental to mainline, enabled for a build
150 by defining SUPPORT_PROXY. Note that the proxy_required_hosts option
151 is renamed to hosts_proxy, and the proxy_{host,target}_{address,port}.
152 variables are renamed to proxy_{local,external}_{address,port}.
154 JH/27 Move Internationalisation support from Experimental to mainline, enabled
155 for a build by defining SUPPORT_I18N
157 JH/28 Bug 1745: Fix redis lookups to handle (quoted) spaces embedded in parts
158 of the query string, and make ${quote_redis:} do that quoting.
160 JH/29 Move Events support from Experimental to mainline, enabled by default
161 and removable for a build by defining DISABLE_EVENT.
163 JH/30 Updated DANE implementation code to current from Viktor Dukhovni.
165 JH/31 Fix bug with hosts_connection_nolog and named-lists which were wrongly
166 cached by the daemon.
168 JH/32 Move Redis support from Experimental to mainline, enabled for a build
169 by defining LOOKUP_REDIS. The libhiredis library is required.
171 JH/33 Bug 1748: Permit ACL dnslists= condition in non-smtp ACLs if explicit
172 keys are given for lookup.
174 JH/34 Bug 1192: replace the embedded copy of PolarSSL RSA routines in the DKIM
175 support, by using OpenSSL or GnuTLS library ones. This means DKIM is
176 only supported when built with TLS support. The PolarSSL SHA routines
177 are still used when the TLS library is too old for convenient support.
179 JH/35 Require SINGLE_DH_USE by default in OpenSSL (main config option
180 openssl_options), for security. OpenSSL forces this from version 1.1.0
181 server-side so match that on older versions.
183 JH/36 Bug 1778: longstanding bug in memory use by the ${run } expansion: A fresh
184 allocation for $value could be released as the expansion processing
185 concluded, but leaving the global pointer active for it.
187 JH/37 Bug 1769: Permit a VRFY ACL to override the default 252 response,
188 and to use the domains and local_parts ACL conditions.
190 JH/38 Fix cutthrough bug with body lines having a single dot. The dot was
191 incorrectly not doubled on cutthrough transmission, hence seen as a
192 body-termination at the receiving system - resulting in truncated mails.
193 Commonly the sender saw a TCP-level error, and retransmitted the message
194 via the normal store-and-forward channel. This could result in duplicates
195 received - but deduplicating mailstores were liable to retain only the
196 initial truncated version.
198 JH/39 Bug 1781: Fix use of DKIM private-keys having trailing '=' in the base-64.
200 JH/40 Fix crash in queryprogram router when compiled with EXPERIMENTAL_SRS.
202 JH/41 Bug 1792: Fix selection of headers to sign for DKIM: bottom-up. While
203 we're in there, support oversigning also; bug 1309.
205 JH/42 Bug 1796: Fix error logged on a malware scanner connection failure.
207 HS/04 Add support for keep_environment and add_environment options.
209 JH/43 Tidy coding issues detected by gcc --fsanitize=undefined. Some remain;
210 either intentional arithmetic overflow during PRNG, or testing config-
213 JH/44 Bug 1800: The combination of a -bhc commandline option and cutthrough
214 delivery resulted in actual delivery. Cancel cutthrough before DATA
217 JH/45 Fix cutthrough, when connection not opened by verify and target hard-
218 rejects a recipient: pass the reject to the originator.
220 JH/46 Multiple issues raised by Coverity. Some were obvious or plausible bugs.
221 Many were false-positives and ignorable, but it's worth fixing the
224 JH/47 Fix build on HP-UX and older Solaris, which need (un)setenv now also
225 for the new environment-manipulation done at startup. Move the routines
226 from being local to tls.c to being global via the os.c file.
228 JH/48 Bug 1807: Fix ${extract } for the numeric/3-string case. While preparsing
229 an extract embedded as result-arg for a map, the first arg for extract
230 is unavailable so we cannot tell if this is a numbered or keyed
231 extraction. Accept either.
237 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
240 JH/02 The smtp transport option "multi_domain" is now expanded.
242 JH/03 The smtp transport now requests PRDR by default, if the server offers
245 JH/04 Certificate name checking on server certificates, when exim is a client,
246 is now done by default. The transport option tls_verify_cert_hostnames
247 can be used to disable this per-host. The build option
248 EXPERIMENTAL_CERTNAMES is withdrawn.
250 JH/05 The value of the tls_verify_certificates smtp transport and main options
251 default to the word "system" to access the system default CA bundle.
252 For GnuTLS, only version 3.0.20 or later.
254 JH/06 Verification of the server certificate for a TLS connection is now tried
255 (but not required) by default. The verification status is now logged by
256 default, for both outbound TLS and client-certificate supplying inbound
259 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
262 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
263 Status Notification (bounce) messages are now MIME format per RFC 3464.
264 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
265 under the control of the dsn_advertise_hosts option, and routers may
266 have a dsn_lasthop option.
268 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
269 default, modifiable by a malware= option. The list separator for
270 the options can now be changed in the usual way. Bug 68.
272 JH/10 The smtp_receive_timeout main option is now expanded before use.
274 JH/11 The incoming_interface log option now also enables logging of the
275 local interface on delivery outgoing connections.
277 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
278 if the interface and destination host and port all match.
280 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
283 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
284 Patch from Andrew Lewis.
286 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
287 now supports optional time-restrictions, weighting, and priority
288 modifiers per server. Patch originally by <rommer@active.by>.
290 JH/16 The spamd_address main option now supports a mixed list of local
291 and remote servers. Remote servers can be IPv6 addresses, and
292 specify a port-range.
294 JH/17 Bug 68: The spamd_address main option now supports an optional
295 timeout value per server.
297 JH/18 Bug 1581: Router and transport options headers_add/remove can
298 now have the list separator specified.
300 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
303 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
306 JH/21 Support for the A6 type of dns record is withdrawn.
308 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
309 rather than the verbs used.
311 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
312 from 255 to 1024 chars.
314 JH/24 Verification callouts now attempt to use TLS by default.
316 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
317 are generic router options now. The defaults didn't change.
319 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
320 Original patch from Alexander Shikoff, worked over by JH.
322 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
323 files if ZCAT_COMMAND is not executable.
325 JH/26 Bug 1539: Add timout/retry options on dnsdb lookups.
327 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
329 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
330 Normally benign, it bites when the pair was led to by a CNAME;
331 modern usage is to not canoicalize the domain to a CNAME target
332 (and we were inconsistent anyway for A-only vs AAAA+A).
334 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
336 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
337 when evaluating $sender_host_dnssec.
339 JH/31 Check the HELO verification lookup for DNSSEC, adding new
340 $sender_helo_dnssec variable.
342 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
344 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
346 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
348 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
349 documented as working, but never had. Support all but $spam_report.
351 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
352 added for tls authenticator.
354 HS/03 Add perl_taintmode main config option
359 TL/01 When running the test suite, the README says that variables such as
360 no_msglog_check are global and can be placed anywhere in a specific
361 test's script, however it was observed that placement needed to be near
362 the beginning for it to behave that way. Changed the runtest perl
363 script to read through the entire script once to detect and set these
364 variables, reset to the beginning of the script, and then run through
365 the script parsing/test process like normal.
367 TL/02 The BSD's have an arc4random API. One of the functions to induce
368 adding randomness was arc4random_stir(), but it has been removed in
369 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
370 function when detected.
372 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
373 cause callback expansion.
375 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
376 syntax errors in an expansion can be treated as a string instead of
377 logging or causing an error, due to the internal use of bool_lax
378 instead of bool when processing it.
380 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
381 server certificates when making smtp deliveries.
383 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
385 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
387 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
389 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
390 Merged patch from Sebastian Wiedenroth.
392 JH/05 Fix results-pipe from transport process. Several recipients, combined
393 with certificate use, exposed issues where response data items split
394 over buffer boundaries were not parsed properly. This eventually
395 resulted in duplicates being sent. This issue only became common enough
396 to notice due to the introduction of conection certificate information,
397 the item size being so much larger. Found and fixed by Wolfgang Breyha.
399 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
400 size buffer was used, resulting in syntax errors when an expansion
403 JH/07 Add support for directories of certificates when compiled with a GnuTLS
404 version 3.3.6 or later.
406 JH/08 Rename the TPDA expermimental facility to Event Actions. The #ifdef
407 is EXPERIMENTAL_EVENT, the main-configuration and transport options
408 both become "event_action", the variables become $event_name, $event_data
409 and $event_defer_errno. There is a new variable $verify_mode, usable in
410 routers, transports and related events. The tls:cert event is now also
411 raised for inbound connections, if the main configuration event_action
414 TL/06 In test suite, disable OCSP for old versions of openssl which contained
415 early OCSP support, but no stapling (appears to be less than 1.0.0).
417 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
418 server certificate names available under the smtp transport option
419 "tls_verify_cert_hostname" now do not permit multi-component wildcard
422 JH/10 Time-related extraction expansions from certificates now use the main
423 option "timezone" setting for output formatting, and are consistent
424 between OpenSSL and GnuTLS compilations. Bug 1541.
426 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
427 encoded parameter in the incoming message. Bug 1558.
429 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
430 include certificate info, eximon was claiming there were spoolfile
433 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
435 JH/14 Log delivery-related information more consistently, using the sequence
436 "H=<name> [<ip>]" wherever possible.
438 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
439 are problematic for Debian distribution, omit them from the release
442 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
444 JH/16 Fix string representation of time values on 64bit time_t anchitectures.
447 JH/17 Fix a null-indirection in certextract expansions when a nondefault
448 output list separator was used.
453 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
454 checkers that were complaining about end of non-void function with no
457 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
458 This was a regression intruduced in 4.83 by another bugfix.
460 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
462 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
463 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
469 TF/01 Correctly close the server side of TLS when forking for delivery.
471 When a message was received over SMTP with TLS, Exim failed to clear up
472 the incoming connection properly after forking off the child process to
473 deliver the message. In some situations the subsequent outgoing
474 delivery connection happened to have the same fd number as the incoming
475 connection previously had. Exim would try to use TLS and fail, logging
476 a "Bad file descriptor" error.
478 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
479 utilities have not been installed.
481 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
482 temporary space as the ACL may create new global variables.
484 TL/01 LDAP support uses per connection or global context settings, depending
485 upon the detected version of the libraries at build time.
487 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
488 to extract and use the src ip:port in logging and expansions as if it
489 were a direct connection from the outside internet. PPv2 support was
490 updated based on HAProxy spec change in May 2014.
492 JH/02 Add ${listextract {number}{list}{success}{fail}}.
494 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
495 Properly escape header and check for NULL return.
497 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
500 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
502 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
503 characters in header names, implemented as a verify condition.
504 Contributed by Michael Fischer v. Mollard.
506 TL/05 Rename SPF condition results err_perm and err_temp to standardized
507 results permerror and temperror. Previous values are deprecated but
508 still accepted. In a future release, err_perm and err_temp will be
509 completely removed, which will be a backward incompatibility if the
510 ACL tests for either of these two old results. Patch contributed by
511 user bes-internal on the mailing list.
513 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
515 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
516 selectors, in both main and reject logs.
518 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
521 JH/07 Add malware type "sock" for talking to simple daemon.
523 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
525 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
526 routers/transports under cutthrough routing.
528 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
529 numbers. Touch up "bool" conditional to keep the same definition.
531 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
533 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
535 JH/12 Expand items in router/transport headers_add or headers_remove lists
536 individually rather than the list as a whole. Bug 1452.
538 Required for reasonable handling of multiple headers_ options when
539 they may be empty; requires that headers_remove items with embedded
540 colons must have them doubled (or the list-separator changed).
542 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
543 view the policy declared in the DMARC record. Currently, $dmarc_status
544 is a combined value of both the record presence and the result of the
547 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
549 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
550 dnslookup router and the smtp transport (applying to the forward
553 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
554 of ldap servers used for a specific lookup. Patch provided by Heiko
557 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
558 New variable $lookup_dnssec_authenticated for observability.
560 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
561 Patch submitted by Lars Timman.
563 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
565 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
566 Requires trusted mode and valid format message id, aborts otherwise.
567 Patch contributed by Heiko Schlichting.
569 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
570 certextract with support for various fields. Bug 1358.
572 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
573 is requested by default, modifiable by smtp transport option
576 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
577 operate on certificate variables to give certificate fingerprints
578 Also new ${sha256:cert_variable}.
580 JH/23 The PRDR feature is moved from being Experimental into the mainline.
582 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
583 Christian Aistleitner.
585 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
587 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
588 file. Patch from Wolfgang Breyha.
590 JH/25 Expand the coverage of the delivery $host and $host_address to
591 client authenticators run in verify callout. Bug 1476.
593 JH/26 Port service names are now accepted for tls_on_connect_ports, to
594 align with daemon_smtp_ports. Bug 72.
596 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
597 support and error reporting did not work properly.
599 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
600 and is readable. Patch from Andrew Colin Kissa.
602 TL/14 Enhance documentation of ${run expansion and how it parses the
603 commandline after expansion, particularly in the case when an
604 unquoted variable expansion results in an empty value.
606 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
608 PP/02 Fix internal collision of T_APL on systems which support RFC3123
609 by renaming away from it. Addresses GH issue 15, reported by
612 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
614 TL/15 SECURITY: prevent double expansion in math comparison functions
615 (can expand unsanitized data). Not remotely exploitable.
622 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
624 PP/02 Make -n do something, by making it not do something.
625 When combined with -bP, the name of an option is not output.
627 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
630 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
631 $sender_host_name and config options to manage this, and basic check
634 PP/05 DSCP support for outbound connections and control modifier for inbound.
636 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
637 (Only plugin which currently uses this is kerberos4, which nobody should
638 be using, but we should make it available and other future plugins might
639 conceivably use it, even though it would break NAT; stuff *should* be
640 using channel bindings instead).
642 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
643 name; added for Sendmail compatibility; requires admin caller.
644 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
645 just ignore it); requires trusted caller.
646 Also parse but ignore: -Ac -Am -X<logfile>
649 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
651 TL/02 Add +smtp_confirmation as a default logging option.
653 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
654 Patch by Magnus Holmgren from 2007-02-20.
656 TL/04 Bugzilla 1281 - Spec typo.
657 Bugzilla 1283 - Spec typo.
658 Bugzilla 1290 - Spec grammar fixes.
660 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
662 TL/06 Add Experimental DMARC support using libopendmarc libraries.
664 TL/07 Fix an out of order global option causing a segfault. Reported to dev
665 mailing list by by Dmitry Isaikin.
667 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
669 JH/02 Support "G" suffix to numbers in ${if comparisons.
671 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
673 NM/01 Bugzilla 1197 - Spec typo
674 Bugzilla 1196 - Spec examples corrections
676 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
678 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
679 gnutls_enable_pkcs11, but renamed to more accurately indicate its
682 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
683 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
685 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
686 "acl {{name}{arg}...}", and optional args on acl condition
689 JH/05 Permit multiple router/transport headers_add/remove lines.
691 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
693 JH/07 Avoid using a waiting database for a single-message-only transport.
694 Performance patch from Paul Fisher. Bugzilla 1262.
696 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
699 JH/09 Add $headers_added variable, with content from use of ACL modifier
700 add_header (but not yet added to the message). Bugzilla 199.
702 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
703 Pulled from Bugzilla 817 by Wolfgang Breyha.
705 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
707 (nb: this is the same fix as in Exim 4.80.1)
709 JH/11 Add A= logging on delivery lines, and a client_set_id option on
712 JH/12 Add optional authenticated_sender logging to A= and a log_selector
715 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
717 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
718 advertise SMTP AUTH mechanism to us, instead of a generic
719 protocol violation error. Also, make Exim more robust to bad
720 data from the Dovecot auth socket.
722 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
724 When a queue runner is handling a message, Exim first routes the
725 recipient addresses, during which it prunes them based on the retry
726 hints database. After that it attempts to deliver the message to
727 any remaining recipients. It then updates the hints database using
730 So if a recipient address works intermittently, it can get repeatedly
731 deferred at routing time. The retry hints record remains fresh so the
732 address never reaches the final cutoff time.
734 This is a fairly common occurrence when a user is bumping up against
735 their storage quota. Exim had some logic in its local delivery code
736 to deal with this. However it did not apply to per-recipient defers
737 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
739 This change adds a proper retry rule check during routing so that the
740 final cutoff time is checked against the message's age. We only do
741 this check if there is an address retry record and there is not a
742 domain retry record; this implies that previous attempts to handle
743 the address had the retry_use_local_parts option turned on. We use
744 this as an approximation for the destination being like a local
745 delivery, as in LMTP.
747 I suspect this new check makes the old local delivery cutoff check
748 redundant, but I have not verified this so I left the code in place.
750 TF/02 Correct gecos expansion when From: is a prefix of the username.
752 Test 0254 submits a message to Exim with the header
756 When I ran the test suite under the user fanf2, Exim expanded
757 the header to contain my full name, whereas it should have added
758 a Resent-Sender: header. It erroneously treats any prefix of the
759 username as equal to the username.
761 This change corrects that bug.
763 GF/01 DCC debug and logging tidyup
764 Error conditions log to paniclog rather than rejectlog.
765 Debug lines prefixed by "DCC: " to remove any ambiguity.
767 TF/03 Avoid unnecessary rebuilds of lookup-related code.
769 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
770 Bug spotted by Jeremy Harris; was flawed since initial commit.
771 Would have resulted in OCSP responses post-SNI triggering an Exim
772 NULL dereference and crash.
774 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
776 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
777 Bug detection, analysis and fix by Samuel Thibault.
778 Bugzilla 1331, Debian bug #698092.
780 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
782 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
783 Server implementation by Todd Lyons, client by JH.
784 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
785 config variable "prdr_enable" controls whether the server
786 advertises the facility. If the client requests PRDR a new
787 acl_data_smtp_prdr ACL is called once for each recipient, after
788 the body content is received and before the acl_smtp_data ACL.
789 The client is controlled by bolth of: a hosts_try_prdr option
790 on the smtp transport, and the server advertisement.
791 Default client logging of deliveries and rejections involving
792 PRDR are flagged with the string "PRDR".
794 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
795 fclose(). Diagnosis by Todd Lyons.
797 PP/17 Update configure.default to handle IPv6 localhost better.
798 Patch by Alain Williams (plus minor tweaks).
801 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
802 This is now consistent with GnuTLS, and is now documented: the
803 previous undocumented portable approach to treating the option as
804 unset was to force an expansion failure. That still works, and
805 an empty string is now equivalent.
807 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
808 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
809 not performing validation itself.
811 PP/20 Added force_command boolean option to pipe transport.
812 Patch from Nick Koston, of cPanel Inc.
814 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
817 TF/04 Added udpsend ACL modifer and hexquote expansion operator
819 PP/21 Fix eximon continuous updating with timestamped log-files.
820 Broken in a format-string cleanup in 4.80, missed when I repaired the
821 other false fix of the same issue.
822 Report and fix from Heiko Schlichting.
825 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
826 Report from Prashanth Katuri.
828 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
829 It's SecureTransport, so affects any MacOS clients which use the
830 system-integrated TLS libraries, including email clients.
832 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
833 using a MIME ACL for non-SMTP local injection.
834 Report and assistance in diagnosis by Warren Baker.
836 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
838 JH/16 Fix comparisons for 64b. Bugzilla 1385.
840 TL/09 Add expansion variable $authenticated_fail_id to keep track of
841 last id that failed so it may be referenced in subsequent ACL's.
843 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
846 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
847 ldap library initialization, allowing self-signed CA's to be
848 used. Also properly sets require_cert option later in code by
849 using NULL (global ldap config) instead of ldap handle (per
850 session). Bug diagnosis and testing by alxgomz.
852 TL/12 Enhanced documentation in the ratelimit.pl script provided in
853 the src/util/ subdirectory.
855 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
856 renamed to Transport Post Delivery Action by Jeremy Harris, as
859 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
860 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
861 redis_servers = needs to be configured which will be used by the redis
862 lookup. Patch from Warren Baker, of The Packet Hub.
864 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
866 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
867 hostname or reverse DNS when processing a host list. Used suggestions
868 from multiple comments on this bug.
870 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
872 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
873 Missed a few lines, added it to make the runtest require no keyboard
876 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
877 contains upper case chars. Make router use caseful_local_part.
879 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
880 support when GnuTLS has been built with p11-kit.
886 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
888 This, or similar/improved, will also be change PP/11 of 4.82.
894 PP/01 Handle short writes when writing local log-files.
895 In practice, only affects FreeBSD (8 onwards).
896 Bugzilla 1053, with thanks to Dmitry Isaikin.
898 NM/01 Bugzilla 949 - Documentation tweak
900 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
903 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
905 PP/02 Implemented gsasl authenticator.
907 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
909 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
910 `pkg-config foo` for cflags/libs.
912 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
913 with rest of GSASL and with heimdal_gssapi.
915 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
916 `pkg-config foo` for cflags/libs for the TLS implementation.
918 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
919 properties get this fed in as external SSF. A number of robustness
920 and debugging improvements to the cyrus_sasl authenticator.
922 PP/08 cyrus_sasl server now expands the server_realm option.
924 PP/09 Bugzilla 1214 - Log authentication information in reject log.
925 Patch by Jeremy Harris.
927 PP/10 Added dbmjz lookup type.
929 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
931 PP/12 MAIL args handles TAB as well as SP, for better interop with
932 non-compliant senders.
933 Analysis and variant patch by Todd Lyons.
935 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
936 Bug report from Lars Müller <lars@samba.org> (via SUSE),
937 Patch from Dirk Mueller <dmueller@suse.com>
939 PP/13 tls_peerdn now print-escaped for spool files.
940 Observed some $tls_peerdn in wild which contained \n, which resulted
941 in spool file corruption.
943 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
944 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
945 or write after TLS renegotiation, which otherwise led to messages
948 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
949 as a tracking header (ie: a signed header comes before the signature).
950 Patch from Wolfgang Breyha.
952 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
953 comma-sep list; embedded commas doubled.
955 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
957 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
959 Report and patch from Dmitry Banschikov.
961 PP/16 Removed "dont_insert_empty_fragments" fron "openssl_options".
962 Removed SSL_clear() after SSL_new() which led to protocol negotiation
963 failures. We appear to now support TLS1.1+ with Exim.
965 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
966 lets Exim select keys and certificates based upon TLS SNI from client.
967 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
968 before an outbound SMTP session. New log_selector, +tls_sni.
970 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
971 NULL dereference. Report and patch from Alun Jones.
973 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
974 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
975 Not seeing resolver debug output on NetBSD, but suspect this is a
976 resolver implementation change.
978 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
979 Left warnings. Added "eximon gdb" invocation mode.
981 PP/21 Defaulting "accept_8bitmime" to true, not false.
983 PP/22 Added -bw for inetd wait mode support.
985 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
986 locate the relevant includes and libraries. Made this the default.
988 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
989 Bugzilla 1246, report and most of solution from Tomasz Kusy.
991 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
992 This may cause build issues on older platforms.
994 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
995 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
996 gnutls_require_mac & gnutls_require_protocols (no longer supported).
997 Added SNI support via GnuTLS too.
998 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
1000 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
1002 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
1003 Applied second patch from Janne, implementing suggestion to default
1004 multiple-strings-in-record handling to match SPF spec.
1006 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
1008 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
1009 read-only, out of scope).
1010 Patch from Wolfgang Breyha, report from Stuart Northfield.
1012 PP/29 Fix three issues highlighted by clang analyser static analysis.
1013 Only crash-plausible issue would require the Cambridge-specific
1014 iplookup router and a misconfiguration.
1015 Report from Marcin Mirosław.
1017 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
1019 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
1020 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
1021 As part of this, removing so much warning spew let me fix some minor
1022 real issues in debug logging.
1024 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
1025 assignment on my part. Fixed.
1027 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
1028 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
1029 Janne Snabb (who went above and beyond: thank you).
1031 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
1032 string otherwise requires a connection and a bunch more work and it's
1033 relatively easy to get wrong. Should also expose TLS library linkage
1036 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
1037 64-bit ${eval} (JH/03).
1039 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
1040 GNU libc to support some of the 64-bit stuff, should not lead to
1041 conflicts. Defined before os.h is pulled in, so if a given platform
1042 needs to override this, it can.
1044 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
1045 protection layer was required, which is not implemented.
1046 Bugzilla 1254, patch from Wolfgang Breyha.
1048 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
1049 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
1050 tls_dhparam take prime identifiers. Also unbreak combination of
1051 OpenSSL+DH_params+TLSSNI.
1053 PP/39 Disable SSLv2 by default in OpenSSL support.
1059 PP/01 Solaris build fix for Oracle's LDAP libraries.
1060 Bugzilla 1109, patch from Stephen Usher.
1062 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
1064 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
1067 TF/02 Fix a couple more cases where we did not log the error message
1068 when unlink() failed. See also change 4.74-TF/03.
1070 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
1071 lock up or crash if it happened to be inside a call to libc when it
1072 got a SIGUSR1 from exiwhat.
1074 The SIGUSR1 handler appends the current process status to the process
1075 log which is later printed by exiwhat. It used to use the general
1076 purpose logging code to do this, but several functions it calls are
1077 not safe for signals.
1079 The new output code in the SIGUSR1 handler is specific to the process
1080 log, and simple enough that it's easy to inspect for signal safety.
1081 Removing some special cases also simplifies the general logging code.
1082 Removing the spurious timestamps from the process log simplifies
1085 TF/04 Improved ratelimit ACL condition.
1087 The /noupdate option has been deprecated in favour of /readonly which
1088 has clearer semantics. The /leaky, /strict, and /readonly update modes
1089 are mutually exclusive. The update mode is no longer included in the
1090 database key; it just determines when the database is updated. (This
1091 means that when you upgrde Exim will forget old rate measurements.)
1093 Exim now checks that the per_* options are used with an update mode that
1094 makes sense for the current ACL. For example, when Exim is processing a
1095 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
1096 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
1097 must specify per_mail/readonly. If you omit the update mode it defaults to
1098 /leaky where that makes sense (as before) or /readonly where required.
1100 The /noupdate option is now undocumented but still supported for
1101 backwards compatibility. It is equivalent to /readonly except that in
1102 ACLs where /readonly is required you may specify /leaky/noupdate or
1103 /strict/noupdate which are treated the same as /readonly.
1105 A useful new feature is the /count= option. This is a generalization
1106 of the per_byte option, so that you can measure the throughput of other
1107 aggregate values. For example, the per_byte option is now equivalent
1108 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
1110 The per_rcpt option has been generalized using the /count= mechanism
1111 (though it's more complicated than the per_byte equivalence). When it is
1112 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
1113 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
1114 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
1115 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
1116 Note that using per_rcpt with a non-readonly update mode in more than
1117 one ACL will cause the recipients to be double-counted. (The per_mail
1118 and per_byte options don't have this problem.)
1120 The handling of very low rates has changed slightly. If the computed rate
1121 is less than the event's count (usually one) then this event is the first
1122 after a long gap. In this case the rate is set to the same as this event's
1123 count, so that the first message of a spam run is counted properly.
1125 The major new feature is a mechanism for counting the rate of unique
1126 events. The new per_addr option counts the number of different
1127 recipients that someone has sent messages to in the last time period. It
1128 behaves like per_rcpt if all the recipient addresses are different, but
1129 duplicate recipient addresses do not increase the measured rate. Like
1130 the /count= option this is a general mechanism, so the per_addr option
1131 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
1132 example, measure the rate that a client uses different sender addresses
1133 with the options per_mail/unique=$sender_address. There are further
1134 details in the main documentation.
1136 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
1138 TF/06 Removed a few PCRE remnants.
1140 TF/07 Automatically extract Exim's version number from tags in the git
1141 repository when doing development or release builds.
1143 PP/02 Raise smtp_cmd_buffer_size to 16kB.
1144 Bugzilla 879. Patch from Paul Fisher.
1146 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
1147 Heavily based on revision 40f9a89a from Simon Arlott's tree.
1150 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
1152 PP/05 Variable $av_failed, true if the AV scanner deferred.
1153 Bugzilla 1078. Patch from John Horne.
1155 PP/06 Stop make process more reliably on build failure.
1156 Bugzilla 1087. Patch from Heiko Schlittermann.
1158 PP/07 Make maildir_use_size_file an _expandable_ boolean.
1159 Bugzilla 1089. Patch from Heiko Schlittermann.
1161 PP/08 Handle ${run} returning more data than OS pipe buffer size.
1162 Bugzilla 1131. Patch from Holger Weiß.
1164 PP/09 Handle IPv6 addresses with SPF.
1165 Bugzilla 860. Patch from Wolfgang Breyha.
1167 PP/10 GnuTLS: support TLS 1.2 & 1.1.
1169 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
1172 PP/11 match_* no longer expand right-hand-side by default.
1173 New compile-time build option, EXPAND_LISTMATCH_RHS.
1174 New expansion conditions, "inlist", "inlisti".
1176 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
1178 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
1180 PP/14 fix log_write() format string regression from TF/03.
1181 Bugzilla 1152. Patch from Dmitry Isaikin.
1187 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
1189 PP/02 Harmonised TLS library version reporting; only show if debugging.
1190 Layout now matches that introduced for other libraries in 4.74 PP/03.
1192 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
1194 PP/04 New "dns_use_edns0" global option.
1196 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
1199 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
1200 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
1202 TK/01 Updated PolarSSL code to 0.14.2.
1203 Bugzilla 1097. Patch from Andreas Metzler.
1205 PP/07 Catch divide-by-zero in ${eval:...}.
1206 Fixes bugzilla 1102.
1208 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
1211 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
1212 format-string attack -- SECURITY: remote arbitrary code execution.
1214 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
1215 time unintentionally subject to list matching rules, letting the header
1216 cause arbitrary Exim lookups (of items which can occur in lists, *not*
1217 arbitrary string expansion). This allowed for information disclosure.
1219 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
1220 INT_MIN/-1 -- value coerced to INT_MAX.
1226 NM/01 Workround for PCRE version dependency in version reporting
1229 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
1230 This fixes portability to compilers other than gcc, notably
1231 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
1233 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
1234 makefiles for portability to HP-UX and POSIX correctness.
1236 PP/01 Permit LOOKUP_foo enabling on the make command-line.
1237 Also via indented variable definition in the Makefile.
1238 (Debugging by Oliver Heesakkers).
1240 PP/02 Restore caching of spamd results with expanded spamd_address.
1241 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
1243 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
1244 Improves build reliability. Fix from: Frank Elsner
1246 NM/02 Fix wide character breakage in the rfc2047 coding
1247 Fixes bug 1064. Patch from Andrey N. Oktyabrski
1249 NM/03 Allow underscore in dnslist lookups
1250 Fixes bug 1026. Patch from Graeme Fowler
1252 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
1253 Code patches from Adam Ciarcinski of NetBSD.
1255 NM/04 Fixed exiqgrep to cope with mailq missing size issue
1258 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
1259 is logged, to avoid truncation. Patch from John Horne.
1261 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
1262 Patch from Jakob Hirsch.
1264 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
1265 SQL string expansion failure details.
1266 Patch from Andrey Oktyabrski.
1268 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
1269 Patch from Simon Arlott.
1271 PP/09 New lookups functionality failed to compile on old gcc which rejects
1272 extern declarations in function scope.
1273 Patch from Oliver Fleischmann
1275 PP/10 Use sig_atomic_t for flags set from signal handlers.
1276 Check getgroups() return and improve debugging.
1277 Fixed developed for diagnosis in bug 927 (which turned out to be
1280 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
1281 Patch from Mark Zealey.
1283 PP/12 Bugzilla 1056: Improved spamd server selection.
1284 Patch from Mark Zealey.
1286 PP/13 Bugzilla 1086: Deal with maildir quota file races.
1287 Based on patch from Heiko Schlittermann.
1289 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
1290 Patch from Uwe Doering, sign-off by Michael Haardt.
1292 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
1293 variable declaration deep within a block. Bug and patch from
1296 PP/15 lookups-Makefile IRIX compatibilty coercion.
1298 PP/16 Make DISABLE_DKIM build knob functional.
1300 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
1301 Patch by Simon Arlott
1303 TF/03 Fix valgrind.h portability to C89 compilers that do not support
1304 variable argument macros. Our copy now differs from upstream.
1310 TF/01 Failure to get a lock on a hints database can have serious
1311 consequences so log it to the panic log.
1313 TF/02 Log LMTP confirmation messages in the same way as SMTP,
1314 controlled using the smtp_confirmation log selector.
1316 TF/03 Include the error message when we fail to unlink a spool file.
1318 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
1319 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
1320 for maintaining out-of-tree patches for some time.
1322 PP/01 Bugzilla 139: Documentation and portability issues.
1323 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
1324 Handle per-OS dynamic-module compilation flags.
1326 PP/02 Let /dev/null have normal permissions.
1327 The 4.73 fixes were a little too stringent and complained about the
1328 permissions on /dev/null. Exempt it from some checks.
1329 Reported by Andreas M. Kirchwitz.
1331 PP/03 Report version information for many libraries, including
1332 Exim version information for dynamically loaded libraries. Created
1333 version.h, now support a version extension string for distributors
1334 who patch heavily. Dynamic module ABI change.
1336 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
1337 privilege escalation vulnerability whereby the Exim run-time user
1338 can cause root to append content of the attacker's choosing to
1341 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
1344 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
1345 If dropping privileges for untrusted macros, we disabled normal logging
1346 on the basis that it would fail; for the Exim run-time user, this is not
1347 the case, and it resulted in successful deliveries going unlogged.
1348 Fixed. Reported by Andreas Metzler.
1354 PP/01 Date: & Message-Id: revert to normally being appended to a message,
1355 only prepend for the Resent-* case. Fixes regression introduced in
1356 Exim 4.70 by NM/22 for Bugzilla 607.
1358 PP/02 Include check_rfc2047_length in configure.default because we're seeing
1359 increasing numbers of administrators be bitten by this.
1361 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
1363 PP/03 Bugzilla 994: added openssl_options main configuration option.
1365 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
1367 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
1369 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
1371 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
1372 without a peer certificate, leading to a segfault because of an
1373 assumption that peers always have certificates. Be a little more
1374 paranoid. Problem reported by Martin Tscholak.
1376 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
1377 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
1378 NB: ClamAV planning to remove STREAM in "middle of 2010".
1379 CL also introduces -bmalware, various -d+acl logging additions and
1380 more caution in buffer sizes.
1382 PP/09 Implemented reverse_ip expansion operator.
1384 PP/10 Bugzilla 937: provide a "debug" ACL control.
1386 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
1388 PP/12 Bugzilla 973: Implement --version.
1390 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
1392 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
1394 PP/15 Bugzilla 816: support multiple condition rules on Routers.
1396 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
1397 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
1398 ignore trailing whitespace.
1400 JJ/02 prevent non-panic DKIM error from being sent to paniclog
1402 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
1405 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
1406 Notification from Dr Andrew Aitchison.
1408 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
1409 ExtendedDetectionInfo response format.
1410 Notification from John Horne.
1412 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
1415 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
1416 XSL and documented dependency on system catalogs, with examples of how
1419 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
1422 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
1423 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
1424 configuration file which is writeable by the Exim user or group.
1426 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
1427 of configuration files to cover files specified with the -C option if
1428 they are going to be used with root privileges, not just the default
1431 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
1432 option (effectively making it always true).
1434 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
1435 files to be used while preserving root privileges.
1437 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
1438 that rogue child processes cannot use them.
1440 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
1441 run-time user, instead of root.
1443 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
1444 Exim run-time user without dropping privileges.
1446 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
1447 result string, instead of calling string_vformat() twice with the same
1450 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
1451 for other users. Others should always drop root privileges if they use
1452 -C on the command line, even for a whitelisted configure file.
1454 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
1456 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
1462 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
1463 $data_path, and $header_path variables; fixed documentation bugs and
1466 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
1467 exipick to access non-standard spools, including the "frozen" queue
1470 NM/01 Bugzilla 965: Support mysql stored procedures.
1471 Patch from Alain Williams
1473 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
1475 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
1476 Patch from Andreas Metzler
1478 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
1479 Patch from Kirill Miazine
1481 NM/05 Bugzilla 671: Added umask to procmail example.
1483 JJ/03 installed exipick 20100323.0, fixing doc bug
1485 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
1486 directory. Notification and patch from Dan Rosenberg.
1488 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
1490 TK/02 Improve log output when DKIM signing operation fails.
1492 MH/01 Treat the transport option dkim_domain as a colon separated
1493 list, not as a single string, and sign the message with each element,
1494 omitting multiple occurences of the same signer.
1496 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
1497 Bugzilla 985, 986. Patch by Simon Arlott
1499 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
1500 Patch by Simon Arlott
1502 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
1503 MBX locking. Notification from Dan Rosenberg.
1509 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
1511 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
1513 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
1515 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
1517 NM/04 Bugzilla 915: Flag broken perl installation during build.
1523 TK/01 Added patch by Johannes Berg that expands the main option
1524 "spamd_address" if it starts with a dollar sign.
1526 TK/02 Write list of recipients to X-Envelope-Sender header when building
1527 the mbox-format spool file for content scanning (suggested by Jakob
1530 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
1531 (http://www.dcc-servers.net/) support via dccifd. Activated by
1532 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
1534 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
1535 by Mark Daniel Reidel <mr@df.eu>.
1537 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
1538 When building exim an external PCRE library is now needed -
1539 PCRE is a system library on the majority of modern systems.
1540 See entry on PCRE_LIBS in EDITME file.
1542 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
1543 conversation. Added nologin parameter to request.
1544 Patch contributed by Kirill Miazine.
1546 TF/01 Do not log submission mode rewrites if they do not change the address.
1548 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
1550 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
1551 log files in place. Contributed by Roberto Lima.
1553 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
1555 TF/03 Bugzilla 615: When checking the local_parts router precondition
1556 after a local_part_suffix or local_part_prefix option, Exim now
1557 does not use the address's named list lookup cache, since this
1558 contains cached lookups for the whole local part.
1560 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
1561 Robert Millan. Documentation is in experimental-spec.txt.
1563 TF/04 Bugzilla 668: Fix parallel build (make -j).
1565 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
1567 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
1568 Patch provided by Jan Srzednicki.
1570 TF/05 Leading white space used to be stripped from $spam_report which
1571 wrecked the formatting. Now it is preserved.
1573 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
1574 that they are available at delivery time.
1576 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
1578 TF/08 TLS error reporting now respects the incoming_interface and
1579 incoming_port log selectors.
1581 TF/09 Produce a more useful error message if an SMTP transport's hosts
1582 setting expands to an empty string.
1584 NM/06 Bugzilla 744: EXPN did not work under TLS.
1585 Patch provided by Phil Pennock.
1587 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
1588 Patch provided by Richard Godbee.
1590 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
1591 acl_smtp_notquit, added index entry.
1593 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
1594 Patch provided by Eugene Bujak.
1596 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
1597 accept(). Patch provided by Maxim Dounin.
1599 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
1600 Patch provided by Phil Pennock.
1602 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
1604 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
1605 Patch provided by Brad "anomie" Jorsch.
1607 NM/14 Bugzilla 622: Added timeout setting to transport filter.
1608 Patch provided by Dean Brooks.
1610 TK/05 Add native DKIM support (does not depend on external libraries).
1612 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
1613 Patch provided by Graeme Fowler.
1615 NM/16 Bugzilla 851: Documentation example syntax fix.
1617 NM/17 Changed NOTICE file to remove references to embedded PCRE.
1619 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
1622 NM/19 Bugzilla 745: TLS version reporting.
1623 Patch provided by Phil Pennock.
1625 NM/20 Bugzilla 167: bool: condition support.
1626 Patch provided by Phil Pennock.
1628 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
1629 clients. Patch provided by Phil Pennock.
1631 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
1632 Patch provided by Brad "anomie" Jorsch.
1634 NM/23 Bugzilla 687: Fix misparses in eximstats.
1635 Patch provided by Heiko Schlittermann.
1637 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
1638 Patch provided by Heiko Schlittermann.
1640 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
1641 plus update to original patch.
1643 NM/26 Bugzilla 799: Documentation correction for ratelimit.
1645 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
1646 Patch provided by David Brownlee.
1648 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
1650 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
1652 NM/30 Bugzilla 888: TLS documentation bugfixes.
1654 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
1656 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
1657 Unlike the original bugzilla I have changed all shell scripts in src tree.
1659 NM/33 Bugzilla 898: Transport filter timeout fix.
1660 Patch by Todd Rinaldo.
1662 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
1663 Patch by Serge Demonchaux.
1665 NM/35 Bugzilla 39: Base64 decode bug fixes.
1666 Patch by Jakob Hirsch.
1668 NM/36 Bugzilla 909: Correct connect() call in dcc code.
1670 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
1672 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
1674 NM/39 Bugzilla 911: Fixed MakeLinks build script.
1680 TK/01 Add preliminary DKIM support. Currently requires a forked version of
1681 ALT-N's libdkim that I have put here:
1682 http://duncanthrax.net/exim-experimental/
1684 Note to Michael Haardt: I had to rename some vars in sieve.c. They
1685 were called 'true' and it seems that C99 defines that as a reserved
1686 keyword to be used with 'bool' variable types. That means you could
1687 not include C99-style headers which use bools without triggering
1688 build errors in sieve.c.
1690 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
1691 as mailq or other aliases. Changed the --help handling significantly
1692 to do whats expected. exim_usage() emits usage/help information.
1694 SC/01 Added the -bylocaldomain option to eximstats.
1696 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
1698 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
1700 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
1706 PH/01 Another patch from the Sieve maintainer.
1708 PH/02 When an IPv6 address is converted to a string for single-key lookup
1709 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
1710 dots are used instead of colons so that keys in lsearch files need not
1711 contain colons. This was done some time before quoting was made available
1712 in lsearch files. However, iplsearch files do require colons in IPv6 keys
1713 (notated using the quote facility) so as to distinguish them from IPv4
1714 keys. This meant that lookups for IP addresses in host lists did not work
1715 for iplsearch lookups.
1717 This has been fixed by arranging for IPv6 addresses to be expressed with
1718 colons if the lookup type is iplsearch. This is not incompatible, because
1719 previously such lookups could never work.
1721 The situation is now rather anomolous, since one *can* have colons in
1722 ordinary lsearch keys. However, making the change in all cases is
1723 incompatible and would probably break a number of configurations.
1725 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
1728 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
1729 conversion specification without a maximum field width, thereby enabling
1730 a rogue spamd server to cause a buffer overflow. While nobody in their
1731 right mind would setup Exim to query an untrusted spamd server, an
1732 attacker that gains access to a server running spamd could potentially
1733 exploit this vulnerability to run arbitrary code as the Exim user.
1735 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
1736 $primary_hostname instead of what libspf2 thinks the hosts name is.
1738 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
1739 a directory entry by the name of the lookup key. Previously, if a
1740 symlink pointed to a non-existing file or a file in a directory that
1741 Exim lacked permissions to read, a lookup for a key matching that
1742 symlink would fail. Now it is enough that a matching directory entry
1743 exists, symlink or not. (Bugzilla 503.)
1745 PH/03 The body_linecount and body_zerocount variables are now exported in the
1748 PH/04 Added the $dnslist_matched variable.
1750 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
1751 This means they are set thereafter only if the connection becomes
1754 PH/06 Added the client_condition to authenticators so that some can be skipped
1755 by clients under certain conditions.
1757 PH/07 The error message for a badly-placed control=no_multiline_responses left
1758 "_responses" off the end of the name.
1760 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
1762 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
1763 (without spaces) instead of just copying the configuration text.
1765 PH/10 Added the /noupdate option to the ratelimit ACL condition.
1767 PH/11 Added $max_received_linelength.
1769 PH/12 Added +ignore_defer and +include_defer to host lists.
1771 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
1772 way in which PCRE > 7.0 is built.
1774 PH/14 Implemented queue_only_load_latch.
1776 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
1777 MAIL command. The effect was to mangle the value on 64-bit systems.
1779 PH/16 Another patch from the Sieve maintainer.
1781 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
1783 PH/18 If a system quota error occurred while trying to create the file for
1784 a maildir delivery, the message "Mailbox is full" was not appended to the
1785 bounce if the delivery eventually timed out. Change 4.67/27 below applied
1786 only to a quota excession during the actual writing of the file.
1788 PH/19 It seems that peer DN values may contain newlines (and other non-printing
1789 characters?) which causes problems in log lines. The DN values are now
1790 passed through string_printing() before being added to log lines.
1792 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
1793 and InterBase are left for another time.)
1795 PH/21 Added message_body_newlines option.
1797 PH/22 Guard against possible overflow in moan_check_errorcopy().
1799 PH/23 POSIX allows open() to be a macro; guard against that.
1801 PH/24 If the recipient of an error message contained an @ in the local part
1802 (suitably quoted, of course), incorrect values were put in $domain and
1803 $local_part during the evaluation of errors_copy.
1809 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
1810 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
1813 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
1814 issue a MAIL command.
1816 PH/02 In an ACL statement such as
1818 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
1820 if a client was not listed at all, or was listed with a value other than
1821 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
1822 the condition was not true (as it should be), so access was not denied.
1823 The bug was that the ! inversion was incorrectly passed on to the second
1824 item. This has been fixed.
1826 PH/03 Added additional dnslists conditions == and =& which are different from
1827 = and & when the dns lookup returns more than one IP address.
1829 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
1830 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
1832 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
1833 FSYNC, which compiles an option called disable_fsync that allows for
1834 bypassing fsync(). The documentation is heavily laced with warnings.
1836 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
1838 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
1839 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
1840 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
1841 including adding "make clean"; (3) Added -fPIC when compiling the test
1842 dynamically loaded module, to get rid of a warning.
1844 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
1845 message fails, move_frozen_messages = true and ignore_bounce_errors_after
1846 = 0s. The bug is otherwise harmless.
1848 PH/07 There was a bug in the dovecot authenticator such that the value of
1849 $auth1 could be overwritten, and so not correctly preserved, after a
1850 successful authentication. This usually meant that the value preserved by
1851 the server_setid option was incorrect.
1853 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
1855 PH/09 Installed PCRE release 7.0.
1857 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
1858 run for batched SMTP input. It is now run at the start of every message
1859 in the batch. While fixing this I discovered that the process information
1860 (output by running exiwhat) was not always getting set for -bs and -bS
1861 input. This is fixed, and it now also says "batched" for BSMTP.
1863 PH/11 Added control=no_pipelining.
1865 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
1866 patch, slightly modified), and move the expansion of helo_data till after
1867 the connection is made in the smtp transport (so it can use these
1870 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
1872 PH/14 Added log_selector = +pid.
1874 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
1876 PH/16 Add ${if forany and ${if forall.
1878 PH/17 Added dsn_from option to vary the From: line in DSNs.
1880 PH/18 Flush SMTP output before performing a callout, unless control =
1881 no_callout_flush is set.
1883 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
1884 was true (the default) a successful delivery failed to delete the retry
1885 item, thus causing premature timeout of the address. The bug is now
1888 PH/20 Added hosts_avoid_pipelining to the smtp transport.
1890 PH/21 Long custom messages for fakedefer and fakereject are now split up
1891 into multiline reponses in the same way that messages for "deny" and
1892 other ACL rejections are.
1894 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
1895 with slight modification.
1897 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
1898 draft, changing the syntax and factoring some duplicate code".
1900 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
1901 for deliveries of the second and subsequent messages over the same SMTP
1904 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
1905 ${reduce, with only minor "tidies".
1907 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
1909 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
1910 expansion side effects.
1912 PH/27 When a message times out after an over-quota error from an Exim-imposed
1913 quota, the bounce message says "mailbox is full". This message was not
1914 being given when it was a system quota that was exceeded. It now should
1917 MH/03 Made $recipients available in local_scan(). local_scan() already has
1918 better access to the recipient list through recipients_list[], but
1919 $recipients can be useful in postmaster-provided expansion strings.
1921 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
1922 in the case of a MAIL command with additional options following the
1923 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
1924 were accidentally chopped off.
1926 PH/29 SMTP synchronization checks are implemented when a command is read -
1927 there is a check that no more input is waiting when there shouldn't be
1928 any. However, for some commands, a delay in an ACL can mean that it is
1929 some time before the response is written. In this time, more input might
1930 arrive, invalidly. So now there are extra checks after an ACL has run for
1931 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
1932 pipelining has not been advertised.
1934 PH/30 MH's patch to allow iscntrl() characters to be list separators.
1936 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
1937 returned in the SMTP response when smtp_return_error_details was false.
1938 This has been fixed.
1940 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
1941 instead of the C I/O that was originally supplied, because problems were
1942 reported on Solaris.
1944 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
1945 Exim which did not show up earlier: it was assuming that a call to
1946 SSL_CTX_set_info_callback() might give an error value. In fact, there is
1947 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
1948 was a macro that became an assignment, so it seemed to work. This has
1949 changed to a proper function call with a void return, hence the compile
1950 error. Exim's code has been fixed.
1952 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
1955 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
1957 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
1959 PH/37 If a message is not accepted after it has had an id assigned (e.g.
1960 because it turns out to be too big or there is a timeout) there is no
1961 "Completed" line in the log. When some messages of this type were
1962 selected by exigrep, they were listed as "not completed". Others were
1963 picked up by some special patterns. I have improved the selection
1964 criteria to be more general.
1966 PH/38 The host_find_failed option in the manualroute router can now be set
1967 to "ignore", to completely ignore a host whose IP address cannot be
1968 found. If all hosts are ignored, the behaviour is controlled by the new
1969 host_all_ignored option.
1971 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
1972 homing or because of multiple MX records with /mx) generated more than
1973 one IP address, and the following item turned out to be the local host,
1974 all the secondary addresses of the first item were incorrectly removed
1975 from the list, along with the local host and any following hosts (which
1976 is what is supposed to happen).
1978 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
1979 whoever called Exim into the -H file. In the case of the daemon it was
1980 behaving confusingly. When first started, it used values for whoever
1981 started the daemon, but after a SIGHUP it used the Exim user (because it
1982 calls itself on a restart). I have changed the code so that it now always
1985 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
1986 message are rejected with the same error (e.g. no authentication or bad
1987 sender address), and a DATA command is nevertheless sent (as can happen
1988 with PIPELINING or a stupid MUA), the error message that was given to the
1989 RCPT commands is included in the rejection of the DATA command. This is
1990 intended to be helpful for MUAs that show only the final error to their
1993 PH/42 Another patch from the Sieve maintainer.
1995 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
1996 Eximstats - Fixed some broken HTML links and added missing column headers
1998 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
1999 columns for Rejects, Temp Rejects, Ham, and Spam rows.
2001 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
2003 PH/43 Yet another patch from the Sieve maintainer.
2005 PH/44 I found a way to check for a TCP/IP connection going away before sending
2006 the response to the final '.' that terminates a message, but only in the
2007 case where the client has not sent further data following the '.'
2008 (unfortunately, this is allowed). However, in many cases there won't be
2009 any further data because there won't be any more messages to send. A call
2010 to select() can be used: if it shows that the input is "ready", there is
2011 either input waiting, or the socket has been closed. An attempt to read
2012 the next input character can distinguish the two cases. Previously, Exim
2013 would have sent an OK response which the client would never have see.
2014 This could lead to message repetition. This fix should cure that, at
2015 least in a lot of common cases.
2017 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
2018 advertised in response to EHLO.
2024 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
2025 fixed by 4.65/MH/01 (is this a record?) are fixed:
2027 (i) An empty string was always treated as zero by the numeric comparison
2028 operators. This behaviour has been restored.
2030 (ii) It is documented that the numeric comparison operators always treat
2031 their arguments as decimal numbers. This was broken in that numbers
2032 starting with 0 were being interpreted as octal.
2034 While fixing these problems I realized that there was another issue that
2035 hadn't been noticed. Values of message_size_limit (both the global option
2036 and the transport option) were treated as octal if they started with 0.
2037 The documentation was vague. These values are now always treated as
2038 decimal, and I will make that clear in the documentation.
2044 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
2045 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
2048 MH/01 Don't check that the operands of numeric comparison operators are
2049 integers when their expansion is in "skipping" mode (fixes bug
2050 introduced by 4.64-PH/07).
2052 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
2053 child addresses, Exim now panics and dies. Previously, because the count
2054 is held in a short int, deliveries were likely to be lost. As such a
2055 large number of recipients for a single message is ridiculous
2056 (performance will be very, very poor), I have chosen to impose a limit
2057 rather than extend the field.
2063 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
2064 leftover -K file (the existence of which was triggered by #402).
2065 While we were at it, introduced process PID as part of the -K
2066 filename. This should rule out race conditions when creating
2069 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
2070 processing considerably. Previous code took too long for large mails,
2071 triggering a timeout which in turn triggers #401.
2073 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
2074 in the DK code in transports.c. sendfile() is not really portable,
2075 hence the _LINUX specificness.
2077 TF/01 In the add_headers option to the mail command in an Exim filter,
2078 there was a bug that Exim would claim a syntax error in any
2079 header after the first one which had an odd number of characters
2082 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
2083 callout verification, Exim cached a "reject" for the entire domain. This
2084 is correct for most verifications, but it is not correct for a recipient
2085 verification with use_sender or use_postmaster set, because in that case
2086 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
2087 case of MAIL FROM:<> rejection from other early rejections (e.g.
2088 rejection of HELO). When verifying a recipient using a non-null MAIL
2089 address, the cache is ignored if it shows MAIL FROM:<> rejection.
2090 Whatever the result of the callout, the value of the domain cache is
2091 left unchanged (for any other kind of callout, getting as far as trying
2092 RCPT means that the domain itself is ok).
2094 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
2097 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
2098 manifest itself as EPIPE rather than ECONNECT. When tidying away a
2099 session, the daemon ignores ECONNECT errors and logs others; it now
2100 ignores EPIPE as well.
2102 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
2103 (quoted-printable decoding).
2105 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
2106 later the small subsequent patch to fix an introduced bug.
2108 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
2110 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
2112 PH/08 An error is now given if message_size_limit is specified negative.
2114 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
2115 to be given (somewhat) arbitrary names.
2117 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
2120 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
2121 miscellaneous code fixes
2123 PH/10 Added the log_reject_target ACL modifier to specify where to log
2126 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
2127 hostname. This is wrong, because it relates to the incoming message (and
2128 probably the interface on which it is arriving) and not to the outgoing
2129 callout (which could be using a different interface). This has been
2130 changed to use the value of the helo_data option from the smtp transport
2131 instead - this is what is used when a message is actually being sent. If
2132 there is no remote transport (possible with a router that sets up host
2133 addresses), $smtp_active_hostname is used.
2135 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
2136 tweaks were necessary in order to get it to work (see also 21 below):
2137 (a) The code assumed that strncpy() returns a negative number on buffer
2138 overflow, which isn't the case. Replaced with Exim's string_format()
2140 (b) There were several signed/unsigned issues. I just did the minimum
2141 hacking in of casts. There is scope for a larger refactoring.
2142 (c) The code used strcasecmp() which is not a standard C function.
2143 Replaced with Exim's strcmpic() function.
2144 (d) The code set only $1; it now sets $auth1 as well.
2145 (e) A simple test gave the error "authentication client didn't specify
2146 service in request". It would seem that Dovecot has changed its
2147 interface. Fortunately there's a specification; I followed it and
2148 changed what the client sends and it appears to be working now.
2150 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
2153 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
2154 address A is aliased to B and C, where B exists and C does not. Without
2155 -v the output is "A verified" because verification stops after a
2156 successful redirection if more than one address is generated. However,
2157 with -v the child addresses are also verified. Exim was outputting "A
2158 failed to verify" and then showing the successful verification for C,
2159 with its parentage. It now outputs "B failed to verify", showing B's
2160 parentage before showing the successful verification of C.
2162 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
2163 look up a TXT record in a specific list after matching in a combined
2166 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
2167 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
2168 they consult the DNS. I had assumed they would set it the way they
2169 wanted; and indeed my experiments on Linux seem to show that in some
2170 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
2171 To be on the safe side, however, I have now made the interface to
2172 host_find_byname() similar to host_find_bydns(), with an argument
2173 containing the DNS resolver options. The host_find_byname() function now
2174 sets these options at its start, just as host_find_bydns() does. The smtp
2175 transport options dns_qualify_single and dns_search_parents are passed to
2176 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
2177 of host_find_byname() use the default settings of RES_DEFNAMES
2178 (qualify_single) but not RES_DNSRCH (search_parents).
2180 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
2181 spool_read_header() do less string testing, by means of a preliminary
2182 switch on the second character of optional "-foo" lines. (This is
2183 overdue, caused by the large number of possibilities that now exist.
2184 Originally there were few.) While I was there, I also converted the
2185 str(n)cmp tests so they don't re-test the leading "-" and the first
2186 character, in the hope this might squeeze out yet more improvement.
2188 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
2189 flag allowing group syntax was set by the header_syntax check but not
2190 turned off, possible causing trouble later; (2) The flag was not being
2191 set at all for the header_verify test, causing "group"-style headers to
2192 be rejected. I have now set it in this case, and also caused header_
2193 verify to ignore an empty address taken from a group. While doing this, I
2194 came across some other cases where the code for allowing group syntax
2195 while scanning a header line wasn't quite right (mostly, not resetting
2196 the flag correctly in the right place). These bugs could have caused
2197 trouble for malformed header lines. I hope it is now all correct.
2199 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
2200 with the "reply" argument non-NULL. The code, however (which originally
2201 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
2202 but it didn't always do it. This confused somebody who was copying the
2203 code for some other use. I have removed all the tests.
2205 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
2206 feature that was used to support insecure browsers during the U.S. crypto
2207 embargo. It requires special client support, and Exim is probably the
2208 only MTA that supported it -- and would never use it because real RSA is
2209 always available. This code has been removed, because it had the bad
2210 effect of slowing Exim down by computing (never used) parameters for the
2211 RSA_EXPORT functionality.
2213 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
2214 authenticator to fail if there's a tab character in the incoming data
2215 (there should never be unless someone is messing about, as it's supposed
2216 to be base64-encoded). Also added, on Timo's advice, the "secured" option
2217 if the connection is using TLS or if the remote IP is the same as the
2218 local IP, and the "valid-client-cert option" if a client certificate has
2221 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
2222 authenticators. This can be used for authorization after authentication
2223 succeeds. (In the case of plaintext, it servers for both authentication
2226 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
2227 if any retry times were supplied.
2229 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
2230 connection, where there is no client IP address to check. In this
2231 situation, the verify now always succeeds.
2233 PH/25 Applied John Jetmore's -Mset patch.
2235 PH/26 Added -bem to be like -Mset, but loading a message from a file.
2237 PH/27 In a string expansion for a processed (not raw) header when multiple
2238 headers of the same name were present, leading whitespace was being
2239 removed from all of them, but trailing whitespace was being removed only
2240 from the last one. Now trailing whitespace is removed from each header
2241 before concatenation. Completely empty headers in a concatenation (as
2242 before) are ignored.
2244 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
2245 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
2247 PH/29 [Removed. This was a change that I later backed out, and forgot to
2248 correct the ChangeLog entry (that I had efficiently created) before
2249 committing the later change.]
2251 PH/30 Exim was sometimes attempting to deliver messages that had suffered
2252 address errors (4xx response to RCPT) over the same connection as other
2253 messages routed to the same hosts. Such deliveries are always "forced",
2254 so retry times are not inspected. This resulted in far too many retries
2255 for the affected addresses. The effect occurred only when there were more
2256 hosts than the hosts_max_try setting in the smtp transport when it had
2257 the 4xx errors. Those hosts that it had tried were not added to the list
2258 of hosts for which the message was waiting, so if all were tried, there
2259 was no problem. Two fixes have been applied:
2261 (i) If there are any address or message errors in an SMTP delivery, none
2262 of the hosts (tried or untried) are now added to the list of hosts
2263 for which the message is waiting, so the message should not be a
2264 candidate for sending over the same connection that was used for a
2265 successful delivery of some other message. This seems entirely
2266 reasonable: after all the message is NOT "waiting for some host".
2267 This is so "obvious" that I'm not sure why it wasn't done
2268 previously. Hope I haven't missed anything, but it can't do any
2269 harm, as the worst effect is to miss an optimization.
2271 (ii) If, despite (i), such a delivery is accidentally attempted, the
2272 routing retry time is respected, so at least it doesn't keep
2273 hammering the server.
2275 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
2276 in ${readsocket because some servers need this prod.
2278 PH/32 Added some extra debug output when updating a wait-xxx database.
2280 PH/33 The hint "could be header name not terminated by colon", which has been
2281 given for certain expansion errors for a long time, was not being given
2282 for the ${if def:h_colon_omitted{... case.
2284 PH/34 The spec says: "With one important exception, whenever a domain list is
2285 being scanned, $domain contains the subject domain." There was at least
2286 one case where this was not true.
2288 PH/35 The error "getsockname() failed: connection reset by peer" was being
2289 written to the panic log as well as the main log, but it isn't really
2290 panic-worthy as it just means the connection died rather early on. I have
2291 removed the panic log writing for the ECONNRESET error when getsockname()
2294 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
2295 runs only) independently of the message's sender address. This meant
2296 that, if the 4xx error was in fact related to the sender, a different
2297 message to the same recipient with a different sender could confuse
2298 things. In particualar, this can happen when sending to a greylisting
2299 server, but other circumstances could also provoke similar problems.
2300 I have changed the default so that the retry time for these errors is now
2301 based a combination of the sender and recipient addresses. This change
2302 can be overridden by setting address_retry_include_sender=false in the
2305 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
2306 remote server are returned as part of bounce messages. This was not
2307 happening for LMTP over a pipe (the lmtp transport), but now it is the
2308 same for both kinds of LMTP.
2310 PH/38 Despite being documented as not happening, Exim was rewriting addresses
2311 in header lines that were in fact CNAMEs. This is no longer the case.
2313 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
2314 and queue runs started by the daemon processed all messages. This has
2315 been fixed so that -R and -S can now usefully be given with -q<time>.
2317 PH/40 Import PCRE release 6.7 (fixes some bugs).
2319 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
2321 PH/42 Give an error if -q is specified more than once.
2323 PH/43 Renamed the variables $interface_address and $interface_port as
2324 $received_ip_address and $received_port, to make it clear that these
2325 values apply to message reception, and not to the outgoing interface when
2326 a message is delivered. (The old names remain recognized, of course.)
2328 PH/44 There was no timeout on the connect() call when using a Unix domain
2329 socket in the ${readsocket expansion. There now is.
2331 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
2332 be meaningful with "accept".
2334 SC/01 Eximstats V1.43
2335 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
2337 SC/02 Eximstats V1.44
2338 Use a glob alias rather than an array ref in the generated
2339 parser. This improves both readability and performance.
2341 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
2342 Collect SpamAssassin and rejection statistics.
2343 Don't display local sender or destination tables unless
2344 there is data to show.
2345 Added average volumes into the top table text output.
2347 SC/04 Eximstats V1.46
2348 Collect data on the number of addresses (recipients)
2349 as well as the number of messages.
2351 SC/05 Eximstats V1.47
2352 Added 'Message too big' to the list of mail rejection
2353 reasons (thanks to Marco Gaiarin).
2355 SC/06 Eximstats V1.48
2356 Mainlog lines which have GMT offsets and are too short to
2357 have a flag are now skipped.
2359 SC/07 Eximstats V1.49 (Alain Williams)
2360 Added the -emptyok flag.
2362 SC/08 Eximstats V1.50
2363 Fixes for obtaining the IP address from reject messages.
2365 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
2366 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
2367 whitesspace changes from 4.64-PH/27
2369 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
2372 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
2373 are found, allow negative numbers in numeric criteria)
2375 JJ/06 exipick.20061117.2, added new $message_body_missing variable
2377 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
2378 to match changes made in 4.64-PH/43
2380 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
2382 PH/47 Put in an explicit test for a DNS lookup of an address record where the
2383 "domain" is actually an IP address, and force a failure. This locks out
2384 those revolvers/nameservers that support "A-for-A" lookups, in
2385 contravention of the specifications.
2387 PH/48 When a host name was looked up from an IP address, and the subsequent
2388 forward lookup of the name timed out, the host name was left in
2389 $sender_host_name, contrary to the specification.
2391 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
2392 restricted to single-key lookups, Exim was not diagnosing an error if
2393 * or *@ was used with a query-style lookup.
2395 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
2397 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
2398 long ago, but noone interested enough thought of it. Let's just say that
2399 the "1.1" means that there are some new functions that weren't there at
2400 some point in the past.
2402 PH/51 Error processing for expansion failure of helo_data from an smtp
2403 transport during callout processing was broken.
2405 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
2406 tested/used via the -bh/-bhc/-bs options.
2408 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
2409 bug, fixed in subsequent PCRE releases).
2411 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
2412 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
2414 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
2420 SC/01 Use a glob alias rather than an array ref in eximstats generated
2421 parser. This improves both readability and performance.
2423 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
2424 Don't display local sender or destination tables in eximstats unless
2425 there is data to show.
2426 Added average volumes into the eximstats top table text output.
2428 SC/03 Collect data on the number of addresses (recipients) as well
2429 as the number of messages in eximstats.
2431 TF/01 Correct an error in the documentation for the redirect router. Exim
2432 does (usually) call initgroups() when daemonizing.
2434 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
2435 with consistent privilege compared to when running as a daemon.
2437 TF/03 Note in the spec that $authenticated_id is not set for local
2438 submissions from trusted users.
2440 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
2441 Thanks to Dean Brooks <dean@iglou.com> for the patch.
2443 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
2444 by adding some example configuration directives to the default
2445 configuration file. A little bit of work is required to uncomment the
2446 directives and define how usernames and passwords are checked, but
2447 there is now a framework to start from.
2449 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
2450 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
2451 without this. I don't know how relevant this is to other LDAP libraries.
2453 PH/02 Add the verb name to the "unknown ACL verb" error.
2455 PH/03 Magnus Holmgren's patch for filter_prepend_home.
2457 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
2459 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
2460 directory not expanded when it should be if an expanded home directory
2461 was set for the address (which is overridden by the transport).
2463 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
2466 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
2467 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
2468 because it is too late at that time, and has no effect.
2470 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
2471 security issue with \' (bugzilla #107). I could not use the
2472 PQescapeStringConn() function, because it needs a PGconn value as one of
2475 PH/08 When testing addresses using -bt, indicate those final addresses that
2476 are duplicates that would not cause an additional delivery. At least one
2477 person was confused, thinking that -bt output corresponded to deliveries.
2478 (Suppressing duplicates isn't a good idea as you lose the information
2479 about possibly different redirections that led to the duplicates.)
2481 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
2482 systems where poll() doesn't work, in particular OS X.
2484 PH/10 Added more information to debugging output for retry time not reached.
2486 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
2487 operations in malware.c.
2489 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
2492 PH/13 If write_rejectlog was set false when logging was sent to syslog with
2493 syslog_duplication set false, log lines that would normally be written
2494 both the the main log and to the reject log were not written to syslog at
2497 PH/14 In the default configuration, change the use of "message" in ACL warn
2498 statements to "add_header".
2500 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
2501 not followed by a command (e.g. "seen endif").
2503 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
2504 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
2507 PH/17 Added extra conditions to the default value of delay_warning_condition
2511 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
2512 { match{$h_precedence:}{(?i)bulk|list|junk} } \
2513 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
2516 The Auto-Submitted: and various List- headers are standardised, whereas I
2517 don't think Precedence: ever was.
2519 PH/18 Refactored debugging code in route_finduser() to show more information,
2520 in particular, the error code if getpwnam() issues one.
2522 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
2523 This is apparently needed in addition to the PH/07 change above to avoid
2524 any possible encoding problems.
2526 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
2527 but not after initializing Perl.
2529 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
2530 output them only if debugging. By default they are written stderr,
2531 apparently, which is not desirable.
2533 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
2536 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
2539 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
2541 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
2542 authenticated or an ident call has been made. Suppress the default
2543 values for $authenticated_id and $authenticated_sender (but permit -oMai
2544 and -oMas) when testing with -bh.
2546 PH/24 Re-jigged the order of the tests in the default configuration so that the
2547 tests for valid domains and recipients precede the DNS black list and CSA
2548 tests, on the grounds that those ones are more expensive.
2550 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
2551 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
2552 This bug exists in every version of Exim that I still have, right back to
2555 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
2556 However, an attempt to turn on case-sensitivity in a regex key by
2557 including (?-i) didn't work because the subject string was already
2558 lowercased, and the effects were non-intuitive. It turns out that a
2559 one-line patch can be used to allow (?-i) to work as expected.
2565 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
2566 other effects) broke the use of negated acl sub-conditions.
2568 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
2571 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
2572 "Deny" causes Exim to reject the incoming connection with a 554 error.
2573 Unfortunately, if there is a major crisis, such as a disk failure,
2574 tcp-wrappers gives "deny", whereas what one would like would be some
2575 kind of temporary error. A kludge has been added to help with this.
2576 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
2577 554 error is used if errno is still zero or contains ENOENT (which occurs
2578 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
2581 PH/03 Add -lutil to the default FreeBSD LIBS setting.
2583 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
2584 errors. Otherwise a message that provokes a temporary error (when other
2585 messages do not) can cause a whole host to time out.
2587 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
2588 addresses were routed directly to files or pipes from a redirect router.
2589 File deliveries just didn't batch; pipe deliveries might have suffered
2592 PH/06 A failure to get a lock for a hints database would erroneously always say
2593 "Failed to get write lock", even when it was really a read lock.
2595 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
2596 of 0600. This has been changed to use the value of the lockfile_mode
2597 option (which defaults to 0600).
2599 PH/08 Applied small patch from the Sieve maintainer.
2601 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
2602 folder from quota calculations, a direct delivery into this folder messed
2603 up the contents of the maildirsize file. This was because the regex was
2604 used only to exclude .Trash (or whatever) when the size of the mailbox
2605 was calculated. There was no check that a delivery was happening into an
2606 excluded directory. This bug has been fixed by ignoring all quota
2607 processing for deliveries into excluded directories.
2609 PH/10 Added the maildirfolder_create_regex option to appendfile.
2615 PH/01 The code for finding all the local interface addresses on a FreeBSD
2616 system running IPv6 was broken. This may well have applied to all BSD
2617 systems, as well as to others that have similar system calls. The broken
2618 code found IPv4 interfaces correctly, but gave incorrect values for the
2619 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
2620 that it would not match correctly against @[] and not recognize the IPv6
2623 PH/02 The ipliteral router was not recognizing addresses of the form user@
2624 [ipv6:....] because it didn't know about the "ipv6:" prefix.
2626 PH/03 Added disable_ipv6.
2628 PH/04 Changed $reply_address to use the raw form of the headers instead of the
2629 decoded form, because it is most often used to construct To: headers
2630 lines in autoreplies, and the decoded form may well be syntactically
2631 invalid. However, $reply_address has leading white space removed, and all
2632 newlines turned into spaces so that the autoreply transport does not
2635 PH/05 If group was specified without a user on a router, and no group or user
2636 was specified on a transport, the group from the router was ignored.
2638 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
2639 for visible compile-time settings that can be used to change these
2640 numbers, for those that want even more. Backwards compatibility with old
2641 spool files has been maintained. However, going back to a previous Exim
2642 release will lost any variables that are in spool files.
2644 PH/07 Two small changes when running in the test harness: increase delay when
2645 passing a TCP/IP connection to a new process, in case the original
2646 process has to generate a bounce, and remove special handling of
2647 127.0.0.2 (sic), which is no longer necessary.
2649 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
2650 be the same on different OS.
2652 PH/09 Moved a debug statement in filter processing to avoid a race problem when
2655 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
2656 whether --show-vars was specified or not
2658 JJ/02 exipick: Added support for new ACL variable spool format introduced
2661 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
2662 syntactically invalid From: or Reply-to: line, and a filter used this to
2663 generate an autoreply, and therefore failed to obtain an address for the
2664 autoreply, Exim could try to deliver to a non-existent relative file
2665 name, causing unrelated and misleading errors. What now happens is that
2666 it logs this as a hard delivery error, but does not attempt to create a
2669 PH/11 The exinext utility has a -C option for testing purposes, but although
2670 the given file was scanned by exinext itself; it wasn't being passed on
2671 when Exim was called.
2673 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
2674 an end-of-file indication when reading a command response.
2676 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
2677 compiled. In many other places in Exim, IPv6 addresses are always
2678 recognized, so I have changed this. It also means that IPv4 domain
2679 literals of the form [IPV4:n.n.n.n] are now always recognized.
2681 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
2682 used if the router is not running as root, for example, when verifying at
2683 ACL time, or when using -bh. The debugging output from this situation was
2684 non-existent - all you got was a failure to exec. I have made two
2687 (a) Failures to set uid/gid, the current directory, or a process leader
2688 in a subprocess such as that created by queryprogram now generate
2689 suitable debugging ouput when -d is set.
2691 (b) The queryprogram router detects when it is not running as root,
2692 outputs suitable debugging information if -d is set, and then runs
2693 the subprocess without attempting to change uid/gid.
2695 PH/15 Minor change to Makefile for building test_host (undocumented testing
2698 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
2699 additional section of a DNS packet that returns MX or SRV records.
2700 Instead, it always explicitly searches for A/AAAA records. This avoids
2701 major problems that occur when a DNS server includes only records of one
2702 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
2703 fixed another bug: if SRV records were looked up and the corresponding
2704 address records were *not* found in the additional section, the port
2705 values from the SRV records were lost.
2707 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
2708 using the correct key (the original address) when searching the retry
2709 rules in order to find which one to use for generating the retry hint.
2711 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
2712 adding the default one. Similarly, if it contains a Reply-To: header, the
2713 errors_reply_to option, if set, is not used.
2715 PH/19 When calculating a retry time, Exim used to measure the "time since
2716 failure" by looking at the "first failed" field in the retry record. Now
2717 it does not use this if it is later than than the arrival time of the
2718 message. Instead it uses the arrival time. This makes for better
2719 behaviour in cases where some deliveries succeed, thus re-setting the
2720 "first failed" field. An example is a quota failure for a huge message
2721 when small messages continue to be delivered. Without this change, the
2722 "time since failure" will always be short, possible causing more frequent
2723 delivery attempts for the huge message than are intended.
2724 [Note: This change was subsequently modified - see PH/04 for 4.62.]
2726 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
2727 $1, $2, $3) because the numerical variables can be reset during some
2728 expansion items (e.g. "match"), thereby losing the authentication data.
2730 PH/21 Make -bV show the size of off_t variables so that the test suite can
2731 decide whether to run tests for quotas > 2G.
2733 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
2734 mailbox_size, and mailbox_filecount in the appendfile transport. If a
2735 filecount value is greater than 2G or if a quota value is greater than 2G
2736 on a system where the size of off_t is not greater than 4, a panic error
2739 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
2740 never match. The debug and -bh output now contains an explicit error
2741 message indicating a malformed IPv4 address or mask.
2743 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
2744 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
2745 PH/23 above applies.
2747 PH/25 Do not write to syslog when running in the test harness. The only
2748 occasion when this arises is a failure to open the main or panic logs
2749 (for which there is an explicit test).
2751 PH/26 Added the /no_tell option to "control=freeze".
2753 PH/27 If a host name lookup failed very early in a connection, for example, if
2754 the IP address matched host_lookup and the reverse lookup yielded a name
2755 that did not have a forward lookup, an error message of the form "no IP
2756 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
2757 could be logged. Now it outputs the IP address instead of "NULL".
2759 PH/28 An enabling patch from MH: add new function child_open_exim2() which
2760 allows the sender and the authenticated sender to be set when
2761 submitting a message from within Exim. Since child_open_exim() is
2762 documented for local_scan(), the new function should be too.
2764 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
2765 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
2766 results in an empty string is now treated as unset.
2768 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
2770 PH/31 Added sender_verify_fail logging option.
2772 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
2773 needed by RFC 822 but not by RFC 2822 was commented out. I have now
2774 tidied the source and removed it altogether.
2776 PH/33 When a queue run was abandoned because the load average was too high, a
2777 log line was always written; now it is written only if the queue_run log
2778 selector is set. In addition, the log line for abandonment now contains
2779 information about the queue run such as the pid. This is always present
2780 in "start" and "stop" lines but was omitted from the "abandon" line.
2782 PH/34 Omit spaces between a header name and the colon in the error message that
2783 is given when verify = headers_syntax fails (if there are lots of them,
2784 the message gets confusing).
2786 PH/35 Change the default for dns_check_names_pattern to allow slashes within
2787 names, as there are now some PTR records that contain slashes. This check
2788 is only to protect against broken name servers that fall over on strange
2789 characters, so the fact that it applies to all lookups doesn't matter.
2791 PH/36 Now that the new test suite is complete, we can remove some of the
2792 special code in Exim that was needed for the old test suite. For example,
2793 sorting DNS records because real resolvers return them in an arbitrary
2794 order. The new test suite's fake resolver always returns records in the
2797 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
2798 bounces) except when queue_only is set, to avoid logging races between
2799 the different processes.
2801 PH/38 Panic-die if .include specifies a non-absolute path.
2803 PH/39 A tweak to the "H" retry rule from its user.
2805 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
2806 a label. They prevented compilation on older perls.
2808 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
2809 a warning to be raised on newish perls.
2811 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
2812 on queue. Changes to match documented behaviour of showing count of
2813 messages matching specified criteria.
2815 PH/40 Changed the default ident timeout from 30s to 5s.
2817 PH/41 Added support for the use of login_cap features, on those BSD systems
2818 that have them, for controlling the resources used by pipe deliveries.
2820 PH/42 The content-scanning code uses fopen() to create files in which to put
2821 message data. Previously it was not paying any attention to the mode of
2822 the files. Exim runs with umask(0) because the rest of the code creates
2823 files with open(), and sets the required mode explicitly. Thus, these
2824 files were ending up world-writeable. This was not a big issue, because,
2825 being within the spool directory, they were not world-accessible. I have
2826 created a function called modefopen, which takes an additional mode
2827 argument. It sets umask(777), creates the file, chmods it to the required
2828 mode, then resets the umask. All the relevant calls to fopen() in the
2829 content scanning code have been changed to use this function.
2831 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
2832 to 24 hours. This avoids potential overflow problems when processing G
2833 and H retry rules. I suspect nobody ever tinkers with this value.
2835 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
2837 PH/45 When the plaintext authenticator is running as a client, the server's
2838 challenges are checked to ensure they are valid base64 strings. By
2839 default, the authentication attempt is cancelled if an invalid string is
2840 received. Setting client_ignore_invalid_base64 true ignores these errors.
2841 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
2842 they are received. Thus, the responses can be made to depend on the
2843 challenges. If an invalid string is ignored, an empty string is placed in
2846 PH/46 Messages that are created by the autoreply transport now contains a
2847 References: header, in accordance with RFCs 2822 and 3834.
2849 PH/47 Added authenticated_sender_force to the smtp transport.
2851 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
2853 PH/49 Installed latest patch from the Sieve maintainer.
2855 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
2856 was also set, the appendfile transport was unnecessarily scanning a
2857 directory of message files (e.g. for maildir delivery) to find the count
2858 of files (along with the size), even though it did not need this
2859 information. It now does the scan only if it needs to find either the
2860 size of the count of files.
2862 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
2864 PH/52 Two bugs concerned with error handling when the smtp transport is
2867 (i) Exim was not creating retry information for temporary errors given
2868 for individual recipients after the DATA command when the smtp transport
2869 was used in LMTP mode. This meant that they could be retried too
2870 frequently, and not timed out correctly.
2872 (ii) Exim was setting the flag that allows error details to be returned
2873 for LMTP errors on RCPT commands, but not for LMTP errors for individual
2874 recipients that were returned after the DATA command.
2876 PH/53 This is related to PH/52, but is more general: for any failing address,
2877 when detailed error information was permitted to be returned to the
2878 sender, but the error was temporary, then after the final timeout, only
2879 "retry timeout exceeded" was returned. Now it returns the full error as
2880 well as "retry timeout exceeded".
2882 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
2883 do this, and (what is worse) MTAs that accept it.
2885 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
2886 will now be deprecated.
2888 PH/56 New os.c-cygwin from the Cygwin maintainer.
2890 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
2891 formats (previously only available in exim formats via -bpr, -bpru,
2892 and -bpra. Now also available in native and exiqgrep formats)
2894 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
2895 with very large, slow to parse queues
2897 JJ/08 exipick: added ! as generic prefix to negate any criteria format
2899 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
2901 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
2902 responses to authentication challenges, though it was showing the
2903 challenges; (ii) I've removed the CR characters from the debug output for
2906 PH/58 Allow for the insertion of a newline as well as a space when a string
2907 is turned into more than one encoded-word during RFC 2047 encoding. The
2908 Sieve code now uses this.
2910 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
2911 data_4xx, lost_connection, tls_required.
2913 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
2914 message was being sent as an SMTP response.
2916 PH/61 Add -l and -k options to exicyclog.
2918 PH/62 When verifying, if an address was redirected to one new address, so that
2919 verification continued, and the new address failed or deferred after
2920 having set something in $address_data, the value of $address_data was not
2921 passed back to the ACL. This was different to the case when no
2922 redirection occurred. The value is now passed back in both cases.
2924 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
2925 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
2926 use login_cap.h, so on its own it isn't the distinguishing feature. The
2927 new name refers directly to the setclassresources() function.
2929 PH/65 Added configuration files for NetBSD3.
2931 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
2933 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
2934 is preferred over IPv4.
2936 PH/68 The bounce_return_message and bounce_return_body options were not being
2937 honoured for bounces generated during the reception of non-SMTP messages.
2938 In particular, this applied to messages rejected by the ACL. This bug has
2939 been fixed. However, if bounce_return_message is true and bounce_return_
2940 body is false, the headers that are returned for a non-SMTP message
2941 include only those that have been read before the error was detected.
2942 (In the case of an ACL rejection, they have all been read.)
2944 PH/69 The HTML version of the specification is now built in a directory called
2945 spec_html instead of spec.html, because the latter looks like a path with
2946 a MIME-type, and this confuses some software.
2948 PH/70 Catch two compiler warnings in sieve.c.
2950 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
2951 function verify_get_ident() calls ip_connect() to connect a socket, but
2952 if the "connect()" function timed out, ip_connect() used to close the
2953 socket. However, verify_get_ident() also closes the socket later, and in
2954 between Exim writes to the log, which may get opened at this point. When
2955 the socket was closed in ip_connect(), the log could get the same file
2956 descriptor number as the socket. This naturally causes chaos. The fix is
2957 not to close the socket in ip_connect(); the socket should be closed by
2958 the function that creates it. There was only one place in the code where
2959 this was missing, in the iplookup router, which I don't think anybody now
2960 uses, but I've fixed it anyway.
2962 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
2963 well as to direct DNS lookups. Otherwise the handling of names in host
2964 lists is inconsistent and therefore confusing.
2970 PH/01 Two changes to the default runtime configuration:
2972 (1) Move the checks for relay_from_hosts and authenticated clients from
2973 after to before the (commented out) DNS black list checks.
2975 (2) Add control=submission to the relay_from_hosts and authenticated
2976 clients checks, on the grounds that messages accepted by these
2977 statements are most likely to be submissions.
2979 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
2981 (1) Generate an error if the third argument for the ${prvs expansion is
2984 (2) Treat a missing third argument of ${prvscheck as if it were an empty
2987 (3) Reset the variables that are obtained from the first argument of
2988 ${prvscheck and used in the second argument before leaving the code,
2989 because their memory is reclaimed, so using them afterwards may do
2992 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
2993 one (it's much easier than Tom thought :-).
2995 (5) Because of (4), we can now allow for the use of $prvscheck_result
2996 inside the third argument.
2998 PH/03 For some reason, the default setting of PATH when running a command from
2999 a pipe transport was just "/usr/bin". I have changed it to
3002 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
3003 anything to be listed in the output from -bV.
3005 PH/05 When a filter generated an autoreply, the entire To: header line was
3006 quoted in the delivery log line, like this:
3008 => >A.N.Other <ano@some.domain> <original@ddress> ...
3010 This has been changed so that it extracts the operative address. There
3011 may be more than one such address. If so, they are comma-separated, like
3014 => >ano@some.domain,ona@other.domain <original@ddress> ...
3016 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
3017 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
3018 not being looked up in the rDNS to get a host name, Exim was showing the
3019 IP address twice in Received: lines, even though the IP addresses were
3020 identical. For example:
3022 Received: from [1.2.3.4] (helo=[1.2.3.4])
3024 However, if the real host name was known, it was omitting the HELO data
3025 if it matched the actual IP address. This has been tidied up so that it
3026 doesn't show the same IP address twice.
3028 PH/07 When both +timestamp and +memory debugging was on, the value given by
3029 $tod_xxx expansions could be wrong, because the tod_stamp() function was
3030 called by the debug printing, thereby overwriting the timestamp buffer.
3031 Debugging no longer uses the tod_stamp() function when +timestamp is set.
3033 PH/08 When the original message was included in an autoreply transport, it
3034 always said "this is a copy of the message, including all the headers",
3035 even if body_only or headers_only was set. It now gives an appropriate
3038 PH/09 Applied a patch from the Sieve maintainer which:
3040 o fixes some comments
3041 o adds the (disabled) notify extension core
3042 o adds some debug output for the result of if/elsif tests
3043 o points to the current vacation draft in the documentation
3044 and documents the missing references header update
3048 o fixes a bug in processing the envelope test (when testing
3049 multiple envelope elements, the last element determinted the
3052 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
3053 Electronic Mail") by including:
3055 Auto-submitted: auto-generated
3057 in the messages that it generates (bounce messages and others, such as
3058 warnings). In the case of bounce messages for non-SMTP mesages, there was
3059 also a typo: it was using "Auto_submitted" (underscore instead of
3060 hyphen). Since every message generated by Exim is necessarily in response
3061 to another message, thes have all been changed to:
3063 Auto-Submitted: auto-replied
3065 in accordance with these statements in the RFC:
3067 The auto-replied keyword:
3069 - SHOULD be used on messages sent in direct response to another
3070 message by an automatic process,
3072 - MUST NOT be used on manually-generated messages,
3074 - MAY be used on Delivery Status Notifications (DSNs) and Message
3075 Disposition Notifications (MDNs),
3077 - MUST NOT be used on messages generated by automatic or periodic
3078 processes, except for messages which are automatic responses to
3081 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
3082 to the default Received: header definition.
3084 PH/12 Added log selector acl_warn_skipped (default on).
3086 PH/13 After a successful wildlsearch lookup, discard the values of numeric
3087 variables because (a) they are in the wrong storage pool and (b) even if
3088 they were copied, it wouldn't work properly because of the caching.
3090 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
3091 checking when decoding. Apparently there are clients that generate
3092 overlong encoded strings. Why am I not surprised?
3094 PH/15 If the first argument of "${if match_address" was not empty, but did not
3095 contain an "@" character, Exim crashed. Now it writes a panic log message
3096 and treats the condition as false.
3098 PH/16 In autoreply, treat an empty string for "once" the same as unset.
3100 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
3101 extension "envelope-auth". The code is finished and in agreement with
3102 other implementations, but there is no documentation so far and in fact,
3103 nobody wrote the draft yet. This extension is currently #undef'ed, thus
3104 not changing the active code.
3106 Print executed "if" and "elsif" statements when debugging is used. This
3107 helps a great deal to understand what a filter does.
3109 Document more things not specified clearly in RFC3028. I had all this
3110 sorted out, when out of a sudden new issues came to my mind. Oops."
3112 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
3115 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
3116 canonical form (as documented). However, after a host name lookup from
3117 the IP address, check_host() was doing a simple string comparison with
3118 addresses acquired from the DNS when checking that the found name did
3119 have the original IP as one of its addresses. Since any found IPv6
3120 addresses are likely to be in abbreviated form, the comparison could
3121 fail. Luckily, there already exists a function for doing the comparison
3122 by converting both addresses to binary, so now that is used instead of
3123 the text comparison.
3125 PH/20 There was another similar case to PH/19, when a complete host name was
3126 given in a host list; looking up its IP address could give an abbreviated
3127 form, whereas the current host's name might or might not be abbreviated.
3128 The same fix has been applied.
3134 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
3135 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
3138 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
3139 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
3141 PH/03 Typo: missing ".o" in src/pcre/Makefile.
3143 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
3144 header line, restrict the check to what is listed in RFCs 2369 and 2929.
3145 Also, for "Auto-Submitted", treat anything other than "no" as
3146 non-personal, in accordance with RFC 3834. (Previously it treated
3147 anything starting "auto-" as non-personal.)
3149 TF/01 The control=submission/name=... option had a problem with syntax
3150 errors if the name included a slash character. The /name= option
3151 now slurps the rest of the string, so it can include any characters
3152 but it must come last in the list of options (after /sender_retain
3155 PH/05 Some modifications to the interface to the fake nameserver for the new
3163 TK/01 Added the "success_on_redirect" address verification option. See
3164 NewStuff for rationale and an example.
3166 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
3168 PH/02 Patch to exigrep to allow it to work on syslog lines.
3170 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
3171 fread() to skip over the body file's header line, because in Cygwin the
3172 header line is locked and is inaccessible.
3174 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
3175 co-exist for some time) to make it clear that it is the Exim ID that is
3176 referenced, not the Message-ID: header line.
3178 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
3179 string_format() function, because snprintf() does not exist on all
3182 PH/06 The use of forbid_filter_existstest now also locks out the use of the
3183 ${stat: expansion item.
3185 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
3186 protocol synchronization error", to keep the pedants happy.
3188 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
3189 well as for IRIX systems, when gcc is being used. See the host.c source
3192 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
3194 PH/10 Named domain lists were not working if used in a queue_smtp_domains
3197 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
3198 transport and to the smtp transport in LMTP mode.
3200 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
3202 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
3203 run a filter in a subprocess. This could lead to confusion in subsequent
3204 lookups in the parent process. There should also be a search_tidyup() at
3205 the end of the subprocess.
3207 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
3208 only if the host matched helo_try_verify_hosts, which caused the
3209 verification to occur when the EHLO/HELO command was issued. The ACL just
3210 tested the remembered result. Now, if a previous verification attempt has
3211 not happened, "verify = helo" does it there and then.
3213 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
3215 TK/03 Fix log output including CR from clamd.
3217 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
3218 exist provoked a memory error which could cause a segfault.
3220 PH/15 Installed PCRE 6.2
3222 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
3224 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
3225 of the problem. Specifically, suggested +O2 rather than +O1 for the
3228 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
3230 PH/20 If a delivery was routed to a non-standard port by means of an SRV
3231 record, the port was not correctly logged when the outgoing_port log
3232 selector was set (it logged the transort's default port).
3234 PH/21 Added support for host-specific ports to manualroute, queryprogram,
3235 fallback_hosts, and "hosts" in the smtp transport.
3237 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
3238 host errors such as "Connection refused".
3240 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
3241 authentication with radiusclient 0.4.9:
3243 - Error returned from rc_read_config was caught wrongly
3244 - Username/password not passed on to radius server due to wrong length.
3246 The presumption is that some radiusclient API changes for 4.51/PH/17
3247 were not taken care of correctly. The code is still untested by me (my
3248 Linux distribution still has 0.3.2 of radiusclient), but it was
3249 contributed by a Radius user.
3251 PH/24 When doing a callout, the value of $domain wasn't set correctly when
3252 expanding the "port" option of the smtp transport.
3254 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
3255 while reading a MIME header. Thanks to Tom Hughes for a patch.
3257 PH/24 Include config.h inside local_scan.h so that configuration settings are
3260 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
3261 that in an ACL for RCPT (for example), you can examine exactly what was
3264 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
3265 commands, but it was not correctly comparing the address with the actual
3266 client host address. Thus, it would show the EHLO address in Received:
3267 header lines when this was not necessary.
3269 PH/27 Added the % operator to ${eval:}.
3271 PH/28 Exim tries to create and chdir to its spool directory when it starts;
3272 it should be ignoring failures (because with -C, for example, it has lost
3273 privilege). It wasn't ignoring creation failures other than "already
3276 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
3279 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
3280 because an input line was too long, either on its own, or by virtue of
3281 too many continuations, the temporary file was not being removed, and the
3282 return code was incorrect.
3284 PH/31 Missing "BOOL" in function definition in filtertest.c.
3286 PH/32 Applied Sieve patches from the maintainer.
3288 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
3290 PH/33 Added "verify = not_blind".
3292 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
3293 Local/Makefile (with some defaults set). These are used in built scripts
3294 such as exicyclog, but they have never been used in the exim_install
3295 script (though there are many overriding facilities there). I have
3296 arranged that the exim_install script now takes note of these two
3299 PH/35 Installed configuration files for Dragonfly.
3301 PH/36 When a locally submitted message by a trusted user did not contain a
3302 From: header, and the sender address was obtained from -f or from an SMTP
3303 MAIL command, and the trusted user did not use -F to supply a sender
3304 name, $originator_name was incorrectly used when constructing a From:
3305 header. Furthermore, $originator_name was used for submission mode
3306 messages from external hosts without From: headers in a similar way,
3307 which is clearly wrong.
3309 PH/37 Added control=suppress_local_fixups.
3311 PH/38 When log_selector = +received_sender was set, and the addition of the
3312 sender made the log line's construction buffer exactly full, or one byte
3313 less than full, an overflow happened when the terminating "\n" was
3316 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
3317 when the result of a list match is failure because a DNS lookup failed.
3319 PH/40 RM_COMMAND is now used in the building process.
3321 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
3322 the "build-* directories that it finds.
3324 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
3325 address in a domain literal was a prefix of an interface address.
3327 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
3328 when verifying a sender address, unless rewrite_headers is false.
3330 PH/44 Wrote a long comment about why errors_to addresses are verified as
3331 recipients, not senders.
3333 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
3334 the ratelimit ACL was added.
3336 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
3338 PH/46 Added extra information about PostgreSQL errors to the error string.
3340 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
3341 avoiding the need to install special zones in a real server. This is
3342 backwards compatible; if it can't find the fake resolver, it drops back.
3343 Thus, both old and new test suites can be run.
3345 TF/02 Added util/ratelimit.pl
3347 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
3348 clock is set back in time.
3350 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
3351 Candler <B.Candler@pobox.com>.
3353 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
3354 OK for addresses that are the result of redirections.
3356 PH/48 A number of further additions for the benefit of the new test suite,
3357 including a fake gethostbyname() that interfaces to the fake DNS resolver
3360 TF/06 The fix for widen_domains has also been applied to qualify_single and
3361 search_parents which are the other dnslookup options that can cause
3364 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
3367 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
3369 TF/07 Exim produced the error message "an SRV record indicated no SMTP
3370 service" if it encountered an MX record with an empty target hostname.
3371 The message is now "an MX or SRV record indicated no SMTP service".
3373 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
3374 if the DNS of the sending site is misconfigured. This is quite a
3375 common situation. This change restores the behaviour of treating a
3376 helo verification defer as a failure.
3378 PH/51 If self=fail was set on a router, the bounce message did not include the
3379 actual error message.
3385 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
3387 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
3388 command itself ended in error, the underlying message about the transport
3389 filter timeout was being overwritten with the pipe command error. Now the
3390 underlying error message should be appended to the second error message.
3392 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
3394 PH/02 Reduce the amount of output that "make" produces by default. Full output
3395 can still be requested.
3397 PH/03 The warning log line about a condition test deferring for a "warn" verb
3398 was being output only once per connection, rather than after each
3399 occurrence (because it was using the same function as for successful
3400 "warn" verbs). This seems wrong, so I have changed it.
3402 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
3403 it should not have, which might have caused a crash in the right
3404 circumstances, but probably never did.
3406 PH/04 Installed a modified version of Tony Finch's patch to make submission
3407 mode fix the return path as well as the Sender: header line, and to
3408 add a /name= option so that you can make the user's friendly name appear
3411 TF/03 Added the control = fakedefer ACL modifier.
3413 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
3414 Mark Lowes for thorough testing.
3416 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
3418 TK/03 Merged latest SRS patch from Miles Wilton.
3420 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
3421 with the definition in sysexits.h (which is #included earlier).
3422 Fortunately, Exim does not actually use EX_OK. The code used to try to
3423 preserve the sysexits.h value, by assumimg that macro definitions were
3424 scanned for macro replacements. I have been disabused of this notion,
3425 so now the code just undefines EX_OK before #including unistd.h.
3427 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
3428 in the smtp transport. When a block could not be written in a single
3429 write() function, the timeout was being re-applied to each part-write.
3430 This seems wrong - if the receiver was accepting one byte at a time it
3431 would take for ever. The timeout is now adjusted when this happens. It
3432 doesn't have to be particularly precise.
3434 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
3435 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
3437 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
3438 without a domain if the check to <postmaster@domain> fails.
3440 SC/01 Eximstats: added -xls and the ability to specify output files
3441 (patch written by Frank Heydlauf).
3443 SC/02 Eximstats: use FileHandles for outputing results.
3445 SC/03 Eximstats: allow any combination of xls, txt, and html output.
3447 SC/04 Eximstats: fixed display of large numbers with -nvr option
3449 SC/05 Eximstats: fixed merging of reports with empty tables.
3451 SC/06 Eximstats: added the -include_original_destination flag
3453 SC/07 Eximstats: removed tabs and trailing whitespace.
3455 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
3457 TK/06 MBOX spool code: Add real "From " MBOX separator line
3458 so the .eml file is really in mbox format (even though
3459 most programs do not really care). Patch from Alex Miller.
3461 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
3462 The latter is generated from $received_to and is only set if the
3463 message has one envelope recipient. SA can use these headers,
3464 obviously out-of-the-box. Patch from Alex Miller.
3466 PH/08 The ${def test on a variable was returning false if the variable's
3467 value was "0", contrary to what the specification has always said!
3468 The result should be true unless the variable is empty.
3470 PH/09 The syntax error of a character other than { following "${if
3471 def:variable_name" (after optional whitespace) was not being diagnosed.
3472 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
3473 accidental colon was present, for example, could give incorrect results.
3475 PH/10 Tidied the code in a number of places where the st_size field of a stat()
3476 result is used (not including appendfile, where other changes are about
3479 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
3480 This involved changing a lot of size variables from int to off_t. It
3481 should work with maildirs and everything.
3483 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
3484 spamd dying while we are connected to it.
3486 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
3489 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
3490 function for BDB 4.3.
3492 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
3494 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
3495 the sources for PCRE have been re-arranged and more files are now
3498 PH/15 The code I had for printing potentially long long variables in PH/11
3499 above was not the best (it lost precision). The length of off_t variables
3500 is now inspected at build time, and an appropriate printing format (%ld
3501 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
3502 to be "long long int" or "long int". This is needed for the internal
3503 formatting function string_vformat().
3505 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
3506 the configuration file to be ":syslog", then the script "guesses" where
3507 the logs files are, rather than using the compiled in default. In our
3508 case the guess is not the same as the compiled default, so the script
3509 suddenly stopped working when I started to use syslog. The patch checks
3510 to see if log_file_path is "". If so, it attempts to read it from exim
3511 with no configuration file to get the compiled in version, before it
3512 falls back to the previous guessing code."
3514 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
3515 implementing BATV in an Exim configuration. See NewStuff for the gory
3518 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
3519 Makefile that are specific to HP-UX.
3521 PH/18 If the "use_postmaster" option was set for a recipient callout together
3522 with the "random" option, the postmaster address was used as the MAIL
3523 FROM address for the random test, but not for the subsequent recipient
3524 test. It is now used for both.
3526 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
3527 patch removes a few documentation additions to RFC 3028, because the
3528 latest draft now contains them. It adds the new en;ascii-case comparator
3529 and a new error check for 8bit text in MIME parts. Comparator and
3530 require names are now matched exactly. I enabled the subaddress
3531 extension, but it is not well tested yet (read: it works for me)."
3533 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
3534 rework some of the code of TK/09 above to avoid the hardwired use of
3535 "%lld" and "long long". Replaced the call to snprintf() with a call to
3538 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
3539 records point to non-existent hosts", "retry timeout exceeded", and
3540 "retry time not reached for any host after a long failure period".
3542 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
3543 experimental DomainKeys support:
3545 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
3546 (2) On an error such as an illegally used "control", the wrong name for
3547 the control was given.
3549 These problems did NOT occur unless DomainKeys support was compiled.
3551 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
3553 PH/24 Added ${if match_ip condition.
3555 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
3556 databases so that it will be absolutely obvious if a crash occurs in the
3557 DB library. This is a regular occurrence (often caused by mis-matched
3560 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
3561 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
3562 picked up on a user's system that detects such things. There doesn't seem
3563 to be a gcc warning option for this - only an attribute that has to be
3564 put on the function's prototype. It seems that in Fedora Core 4 they have
3565 set this on a number of new functions. No doubt there will be more in due
3568 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
3569 specify a transport. However, if an address that was verified by such a
3570 router was the subject of a callout, Exim crashed because it tried to
3571 read the rcpt_include_affixes from the non-existent transport. Now it
3572 just assumes that the setting of that option is false. This bug was
3573 introduced by 4.51/PH/31.
3575 PH/28 Changed -d+all to exclude +memory, because that information is very
3576 rarely of interest, but it makes the output a lot bigger. People tend to
3577 do -d+all out of habit.
3579 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
3580 code in os-type was giving problems when libc.so lives in lib64, like on
3583 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
3584 aren't the modern standard, and it seems that some systems' include files
3585 don't always have them. Exim was already checking for some of the newer
3586 ones like T_AAAA, and defining it itself. I've added checks for all the
3587 record types that Exim uses.
3589 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
3590 not automatically generating a new one, as it is supposed to. This
3591 prevented TLS from working. If the file did exist, but contained invalid
3592 data, a new version was generated, as expected. It was only the case of a
3593 non-existent file that was broken.
3595 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
3596 with a change in libdomainkeys > 0.64.
3598 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
3599 from DNS. If the selector record carries the flag, it now has
3600 precedence over the domain-wide flag.
3602 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
3604 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
3605 the use of a port name that isn't defined in /etc/services) occurred, the
3606 message was deferred as in a normal delivery, and thus remained on the
3607 spool, instead of being failed because of the mua_wrapper setting. This
3608 is now fixed, and I tidied up some of the mua_wrapper messages at the
3611 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
3612 the messages in a hash of arrays rather than using individual hashes.
3613 This is a bit cleaner and results in dramatic memory savings, albeit
3614 at a slight CPU cost.
3616 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
3617 as requested by Marc Sherman.
3619 SC/10 Eximstats: added histograms for user specified patterns as requested
3622 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
3624 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
3625 fopen() in the content-scanning modules that did not already have it.
3631 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
3632 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
3634 TK/02 Fix ACL "control" statement not being available in MIME ACL.
3636 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
3638 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
3639 to test Sieve filters that use "vacation".
3641 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
3642 that changes the way the GnuTLS parameters are stored in the cache file.
3643 The new format can be generated externally. For backward compatibility,
3644 if the data in the cache doesn't make sense, Exim assumes it has read an
3645 old-format file, and it generates new data and writes a new file. This
3646 means that you can't go back to an older release without removing the
3649 PH/03 A redirect router that has both "unseen" and "one_time" set does not
3650 work if there are any delivery delays because "one_time" forces the
3651 parent to be marked "delivered", so its unseen clone is never tried
3652 again. For this reason, Exim now forbids the simultaneous setting of
3655 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
3656 redirected to themselves ("homonym" addresses). Read the long ChangeLog
3657 entry if you want to know the details. The fix, however, neglected to
3658 consider the case when local delivery batching is involved. The test for
3659 "previously delivered" was not happening when checking to see if an
3660 address could be batched with a previous (undelivered) one; under
3661 certain circumstances this could lead to multiple deliveries to the same
3664 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
3665 in its include files, and this causes problems building Exim.
3667 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
3668 header_syntax) but Exim was just ignoring anything given after a slash.
3669 In particular, this caused confusion with an attempt to use "verify =
3670 reverse_host_lookup/defer_ok". An error is now given when options are
3671 supplied for verify items that do not have them. (Maybe reverse_host_
3672 lookup should have a defer_ok option, but that's a different point.)
3674 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
3675 defined by RFC 821) to 2048, because there were problems with some AUTH
3676 commands, and RFC 1869 says the size should be increased for extended
3677 SMTP commands that take arguments.
3679 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
3682 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
3683 "unknown" error; now it says that the functionality isn't in the binary.
3685 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
3686 an address' error message when a string expansion fails (syntax or
3687 whatever). Otherwise the password may appear in the log. Following change
3688 PH/42 below, there is no longer a chance of it appearing in a bounce
3691 PH/11 Installed exipick version 20050225.0 from John Jetmore.
3693 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
3694 of its addresses was ever tried. (Bugzilla bug #2.)
3696 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
3697 the result incorrectly in the debug output. (It correctly added a newline
3698 to what was transported.)
3700 TF/01 Added $received_time.
3702 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
3703 commented out examples of how to interface to a virus scanner and to
3704 SpamAssassin. Also added commented examples of av_scanner and
3705 spamd_address settings.
3707 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
3708 and controls are allowed in which ACLs. There were a couple of minor
3709 errors. Some of the entries in the conditions table (which is a table of
3710 where they are NOT allowed) were getting very unwieldy; rewrote them as a
3711 negation of where the condition IS allowed.
3713 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
3715 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
3716 header file does not have a version number, so I've had to invent a new
3717 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
3718 API. The code is untested by me (my Linux distribution still has 0.3.2 of
3719 radiusclient), but it was contributed by a Radius user.
3721 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
3722 files or directories, for OpenSSL.
3724 PH/19 When an Exim process that is running as root has to create an Exim log
3725 file, it does so in a subprocess that runs as exim:exim so as to get the
3726 ownership right at creation (otherwise, other Exim processes might see
3727 the file with the wrong ownership). There was no test for failure of this
3728 fork() call, which would lead to the process getting stuck as it waited
3729 for a non-existent subprocess. Forks do occasionally fail when resources
3730 run out. I reviewed all the other calls to fork(); they all seem to check
3733 PH/20 When checking for unexpected SMTP input at connect time (before writing
3734 the banner), Exim was not dealing correctly with a non-positive return
3735 from the read() function. If the client had disconnected by this time,
3736 the result was a log entry for a synchronization error with an empty
3737 string after "input=" when read() returned zero. If read() returned -1
3738 (an event I could not check), uninitialized data bytes were printed.
3739 There were reports of junk text (parts of files, etc) appearing after
3742 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
3744 PH/22 Added support for macro redefinition, and (re)definition in between
3745 driver and ACL definitions.
3747 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
3748 forgetting to use the resulting value; it was using the unexpanded value.
3750 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
3751 hadn't been configured. The fix is from Juergen Kreileder, who
3752 understands it better than I do:
3754 "Here's what I see happening with three configured cyrus_sasl
3755 authenticators configured (plain, login, cram-md5):
3757 On startup auth_cyrus_sasl_init() gets called for each of these.
3758 This means three calls to sasl_listmech() without a specified mech_list.
3759 => SASL tests which mechs of all available mechs actually work
3760 => three warnings about OTP not working
3761 => the returned list contains: plain, login, cram-md5, digest-md5, ...
3763 With the patch, sasl_listmech() also gets called three times. But now
3764 SASL's mech_list option is set to the server_mech specified in the the
3765 authenticator. Or in other words, the answer from sasl_listmech()
3766 gets limited to just the mech you're testing for (which is different
3768 => the return list contains just 'plain' or 'login', 'cram-md5' or
3769 nothing depending on the value of ob->server_mech.
3771 I've just tested the patch: Authentication still works fine,
3772 unavailable mechs specified in the exim configuration are still
3773 caught, and the auth.log warnings about OTP are gone."
3775 PH/25 When debugging is enabled, the contents of the command line are added
3776 to the debugging output, even when log_selector=+arguments is not
3779 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
3780 answer is "GNU", and only if the return is "GNU/something" is the answer
3783 PH/27 $acl_verify_message is now set immediately after the failure of a
3784 verification in an ACL, and so is available in subsequent modifiers. In
3785 particular, the message can be preserved by coding like this:
3787 warn !verify = sender
3788 set acl_m0 = $acl_verify_message
3790 Previously, $acl_verify_message was set only while expanding "message"
3791 and "log_message" when a very denied access.
3793 PH/28 Modified OS/os.c-Linux with
3795 -#ifndef OS_LOAD_AVERAGE
3796 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
3798 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
3799 nomenclature these days.)
3801 PH/29 Installed patch from the Sieve maintainer that adds the options
3802 sieve_useraddress and sieve_subaddress to the redirect router.
3804 PH/30 In these circumstances:
3805 . Two addresses routed to the same list of hosts;
3806 . First host does not offer TLS;
3807 . First host accepts first address;
3808 . First host gives temporary error to second address;
3809 . Second host offers TLS and a TLS session is established;
3810 . Second host accepts second address.
3811 Exim incorrectly logged both deliveries with the TLS parameters (cipher
3812 and peerdn, if requested) that were in fact used only for the second
3815 PH/31 When doing a callout as part of verifying an address, Exim was not paying
3816 attention to any local part prefix or suffix that was matched by the
3817 router that accepted the address. It now behaves in the same way as it
3818 does for delivery: the affixes are removed from the local part unless
3819 rcpt_include_affixes is set on the transport.
3821 PH/32 Add the sender address, as F=<...>, to the log line when logging a
3822 timeout during the DATA phase of an incoming message.
3824 PH/33 Sieve envelope tests were broken for match types other than :is. I have
3825 applied a patch sanctioned by the Sieve maintainer.
3827 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
3828 the uid or gid is negative. A case of a negative gid caused this to be
3829 noticed. The fix allows for either to be negative.
3831 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
3832 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
3835 PH/36 The change PH/12 above was broken. Fixed it.
3837 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
3838 the grounds that routing the same address twice would always produce the
3839 same answer. This might have been true once, but it is certainly no
3840 longer true now. Routing a child address may depend on the previous
3841 routing that produced that child. Some complicated redirection strategies
3842 went wrong when messages had multiple recipients, and made Exim's
3843 behaviour dependent on the order in which the addresses were given.
3845 I have moved the duplicate checking until after the routing is complete.
3846 Exim scans the addresses that are assigned to local and remote
3847 transports, and removes any duplicates. This means that more work will be
3848 done, as duplicates will always all be routed, but duplicates are
3849 presumably rare, so I don't expect this is of any significance.
3851 For deliveries to pipes, files, and autoreplies, the duplicate checking
3852 still happens during the routing process, since they are not going to be
3855 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
3856 It corrects a timeout issue with spamd. This is Ian's comment: "The
3857 background is that sometimes spamd either never reads data from a
3858 connection it has accepted, or it never writes response data. The exiscan
3859 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
3860 blindly assumes that writes won't block so it may never time out."
3862 PH/39 Allow G after quota size as well as K and M.
3864 PH/40 The value set for $authenticated_id in an authenticator may not contain
3865 binary zeroes or newlines because the value is written to log lines and
3866 to spool files. There was no check on this. Now the value is run through
3867 the string_printing() function so that such characters are converted to
3868 printable escape sequences.
3870 PH/41 $message_linecount is a new variable that contains the total number of
3871 lines in the message. Compare $body_linecount, which is the count for the
3874 PH/42 Exim no longer gives details of delivery errors for specific addresses in
3875 bounce and delay warning messages, except in certain special cases, which
3878 (a) An SMTP error message from a remote host;
3879 (b) A message specified in a :fail: redirection;
3880 (c) A message specified in a "fail" command in a system filter;
3881 (d) A message specified in a FAIL return from the queryprogram router;
3882 (e) A message specified by the cannot_route_message router option.
3884 In these cases only, Exim does include the error details in bounce and
3885 warning messages. There are also a few cases where bland messages such
3886 as "unrouteable address" or "local delivery error" are given.
3888 PH/43 $value is now also set for the "else" part of a ${run expansion.
3890 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
3891 being worked on, but at least Exim now implements the latest version to
3894 PH/45 In a pipe transport, although a timeout while waiting for the pipe
3895 process to complete was treated as a delivery failure, a timeout while
3896 writing the message to the pipe was logged, but erroneously treated as a
3897 successful delivery. Such timeouts include transport filter timeouts. For
3898 consistency with the overall process timeout, these timeouts are now
3899 treated as errors, giving rise to delivery failures by default. However,
3900 there is now a new Boolean option for the pipe transport called
3901 timeout_defer, which, if set TRUE, converts the failures into defers for
3902 both kinds of timeout. A transport filter timeout is now identified in
3905 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
3906 systems where "make" and "gmake" are different, calling "gmake" at top
3907 level broke things. I've arranged for the value of $(MAKE) to be passed
3908 from the Makefile to this script so that it can call the same version of
3912 A note about Exim versions 4.44 and 4.50
3913 ----------------------------------------
3915 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
3916 changes of various kinds. As a consequence, a big documentation update was
3917 needed. This delayed the release for rather longer than seemed good, especially
3918 in the light of a couple of (minor) security issues. Therefore, the changes
3919 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
3920 release. So 4.44 and 4.50 are in effect two different branches that both start
3923 I have left the 4.50 change log unchanged; it contains all the changes since
3924 4.43. The change log for 4.44 is below; many of its items are identical to
3925 those for 4.50. This seems to be the most sensible way to preserve the
3926 historical information.
3932 1. Minor wording change to the doc/README.SIEVE file.
3934 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
3935 computation of the current number of files was incorrect.
3937 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
3938 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
3941 4. Give more explanation in the error message when the command for a transport
3942 filter fails to execute.
3944 5. There are several places where Exim runs a non-Exim command in a
3945 subprocess. The SIGUSR1 signal should be disabled for these processes. This
3946 was being done only for the command run by the queryprogram router. It is
3947 now done for all such subprocesses. The other cases are: ${run, transport
3948 filters, and the commands run by the lmtp and pipe transports.
3950 6. Added CONFIGURE_GROUP build-time option.
3952 7. Some older OS have a limit of 256 on the maximum number of file
3953 descriptors. Exim was using setrlimit() to set 1000 as a large value
3954 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
3955 systems. I've change it so that if it can't get 1000, it tries for 256.
3957 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
3958 was an oversight, and furthermore, ever since the addition of extra
3959 controls (e.g. 4.43/32), the checks on when to allow different forms of
3960 "control" were broken. There should now be diagnostics for all cases when a
3961 control that does not make sense is encountered.
3963 9. Added the /retain_sender option to "control=submission".
3965 10. $recipients is now available in the predata ACL (oversight).
3967 11. Tidy the search cache before the fork to do a delivery from a message
3968 received from the command line. Otherwise the child will trigger a lookup
3969 failure and thereby defer the delivery if it tries to use (for example) a
3970 cached ldap connection that the parent has called unbind on.
3972 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
3973 of $address_data from the recipient verification was clobbered by the
3974 sender verification.
3976 13. The value of address_data from a sender verification is now available in
3977 $sender_address_data in subsequent conditions in the ACL statement.
3979 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
3981 15. Added a new option "connect=<time>" to callout options, to set a different
3984 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
3985 was its contents. (It was OK if the option was not defined at all.)
3987 17. A "Completed" log line is now written for messages that are removed from
3988 the spool by the -Mrm option.
3990 18. New variables $sender_verify_failure and $recipient_verify_failure contain
3991 information about exactly what failed.
3993 19. Added -dd to debug only the daemon process.
3995 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
3996 handles timeouts, both on the server side and network timeouts. Renamed the
3997 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
3999 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
4000 It is now set to "smtps".
4002 22. $host_address is now set to the target address during the checking of
4003 ignore_target_hosts.
4005 23. When checking ignore_target_hosts for an ipliteral router, no host name was
4006 being passed; this would have caused $sender_host_name to have been used if
4007 matching the list had actually called for a host name (not very likely,
4008 since this list is usually IP addresses). A host name is now passed as
4011 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4012 code that specifies a non-restarting handler (typically sigaction() in
4013 modern systems) in an attempt to fix a rare and obscure crash bug.
4015 25. Narrowed the window for a race in the daemon that could cause it to ignore
4016 SIGCHLD signals. This is not a major problem, because they are used only to
4017 wake it up if nothing else does.
4019 26. A malformed maildirsize file could cause Exim to calculate negative values
4020 for the mailbox size or file count. Odd effects could occur as a result.
4021 The maildirsize information is now recalculated if the size or filecount
4024 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4025 support for a long time. Removed HAVE_SYS_VFS_H.
4027 28. Installed the latest version of exipick from John Jetmore.
4029 29. In an address list, if the pattern was not a regular expression, an empty
4030 subject address (from a bounce message) matched only if the pattern was an
4031 empty string. Non-empty patterns were not even tested. This was the wrong
4032 because it is perfectly reasonable to use an empty address as part of a
4033 database query. An empty address is now tested by patterns that are
4034 lookups. However, all the other forms of pattern expect the subject to
4035 contain a local part and a domain, and therefore, for them, an empty
4036 address still always fails if the pattern is not itself empty.
4038 30. Exim went into a mad DNS loop when attempting to do a callout where the
4039 host was specified on an smtp transport, and looking it up yielded more
4040 than one IP address.
4042 31. Re-factored the code for checking spool and log partition space into a
4043 function that finds that data and another that does the check. The former
4044 is then used to implement four new variables: $spool_space, $log_space,
4045 $spool_inodes, and $log_inodes.
4047 32. The RFC2047 encoding function was originally intended for short strings
4048 such as real names; it was not keeping to the 75-character limit for
4049 encoded words that the RFC imposes. It now respects the limit, and
4050 generates multiple encoded words if necessary. To be on the safe side, I
4051 have increased the buffer size for the ${rfc2047: expansion operator from
4054 33. It is now permitted to omit both strings after an "if" condition; if the
4055 condition is true, the result is "true". As before, when the second string
4056 is omitted, a false condition yields an empty string. This makes it less
4057 cumbersome to write custom ACL and router conditions.
4059 34. Failure to deliver a bounce message always caused it to be frozen, even if
4060 there was an errors_to setting on the router. The errors_to setting is now
4063 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4064 canonical form (fully expanded) before being placed in
4065 $sender_host_address.
4067 36. The table in the code that translates DNS record types into text (T_A to
4068 "A" for instance) was missing entries for NS and CNAME. It is just possible
4069 that this could have caused confusion if both these types were looked up
4070 for the same domain, because the text type is used as part of Exim's
4071 per-process caching. But the chance of anyone hitting this buglet seems
4074 37. The dnsdb lookup has been extended in a number of ways.
4076 (1) There is a new type, "zns", which walks up the domain tree until it
4077 finds some nameserver records. It should be used with care.
4079 (2) There is a new type, "mxh", which is like "mx" except that it returns
4080 just the host names, not the priorities.
4082 (3) It is now possible to give a list of domains (or IP addresses) to be
4083 looked up. The behaviour when one of the lookups defers can be
4084 controlled by a keyword.
4086 (4) It is now possible to specify the separator character for use when
4087 multiple records are returned.
4089 38. The dnslists ACL condition has been extended: it is now possible to supply
4090 a list of IP addresses and/or domains to be looked up in a particular DNS
4093 39. Added log_selector=+queue_time_overall.
4095 40. When running the queue in the test harness, wait just a tad after forking a
4096 delivery process, to get repeatability of debugging output.
4098 41. Include certificate and key file names in error message when GnuTLS fails
4099 to set them up, because the GnuTLS error message doesn't include the name
4100 of the failing file when there is a problem reading it.
4102 42. Allow both -bf and -bF in the same test run.
4104 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
4106 44. The "Exiscan patch" is now merged into the mainline Exim source.
4108 45. Sometimes the final signoff response after QUIT could fail to get
4109 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4110 < 0 before doing a fflush(). This bug looks as though it goes back to the
4111 introduction of TLS in release 3.20, but "sometimes" must have been rare
4112 because the tests only now provoked it.
4114 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
4115 (this can affect the format of dates).
4117 47. exim_tidydb, when checking for the continued existence of a message for
4118 which it has found a message-specific retry record, was not finding
4119 messages that were in split spool directories. Consequently, it was
4120 deleting retry records that should have stayed in existence.
4122 48. Steve fixed some bugs in eximstats.
4124 49. The SPA authentication driver was not abandoning authentication and moving
4125 on to the next authenticator when an expansion was forced to fail,
4126 contradicting the general specification for all authenticators. Instead it
4127 was generating a temporary error. It now behaves as specified.
4129 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
4130 (the order specifies the preference for clients). The order is now AES256,
4131 AES128, 3DES, ARCFOUR128.
4133 51. Small patch to Sieve code - explicitly set From: when generating an
4136 52. Exim crashed if a remote delivery caused a very long error message to be
4137 recorded - for instance if somebody sent an entire SpamAssassin report back
4138 as a large number of 550 error lines. This bug was coincidentally fixed by
4139 increasing the size of one of Exim's internal buffers (big_buffer) that
4140 happened as part of the Exiscan merge. However, to be on the safe side, I
4141 have made the code more robust (and fixed the comments that describe what
4144 53. Now that there can be additional text after "Completed" in log lines (if
4145 the queue_time_overall log selector is set), a one-byte patch to exigrep
4146 was needed to allow it to recognize "Completed" as not the last thing in
4149 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
4150 patch that reportedly fixes this has been added. I am not expert enough to
4151 create a test for it. This is what the patch creator wrote:
4153 "I found a little strange behaviour of ldap code when working with
4154 Windows 2003 AD Domain, where users was placed in more than one
4155 Organization Units. When I tried to give exim partial DN, the exit code
4156 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
4157 But simultaneously result of request was absolutely normal ldap result,
4158 so I produce this patch..."
4160 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
4161 REFERENCE, so I have modified the code to exclude the patch when that macro
4164 55. Some experimental protocols are using DNS PTR records for new purposes. The
4165 keys for these records are domain names, not reversed IP addresses. The
4166 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4167 leaves it alone. Component reversal etc. now happens only for IP addresses.
4170 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
4172 57. Double the size of the debug message buffer (to 2048) so that more of very
4173 long debug lines gets shown.
4175 58. The exicyclog utility now does better if the number of log files to keep
4176 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
4178 59. Two changes related to the smtp_active_hostname option:
4180 (1) $smtp_active_hostname is now available as a variable.
4181 (2) The default for smtp_banner uses $smtp_active_hostname instead
4182 of $primary_hostname.
4184 60. The host_aton() function is supposed to be passed a string that is known
4185 to be a valid IP address. However, in the case of IPv6 addresses, it was
4186 not checking this. This is a hostage to fortune. Exim now panics and dies
4187 if the condition is not met. A case was found where this could be provoked
4188 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4189 components; fortuitously, this particular loophole had already been fixed
4190 by change 4.50/55 above.
4192 If there are any other similar loopholes, the new check in host_aton()
4193 itself should stop them being exploited. The report I received stated that
4194 data on the command line could provoke the exploit when Exim was running as
4195 exim, but did not say which command line option was involved. All I could
4196 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4197 running as the user.
4200 61. There was a buffer overflow vulnerability in the SPA authentication code
4201 (which came originally from the Samba project). I have added a test to the
4202 spa_base64_to_bits() function which I hope fixes it.
4205 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
4206 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
4208 63. The daemon start-up calls getloadavg() while still root for those OS that
4209 need the first call to be done as root, but it missed one case: when
4210 deliver_queue_load_max is set with deliver_drop_privilege. This is
4211 necessary for the benefit of the queue runner, because there is no re-exec
4212 when deliver_drop_privilege is set.
4214 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
4215 This has been fixed.
4217 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4218 was in use, was not putting the data itself into the right store pool;
4219 consequently, it could be overwritten for a subsequent message in the same
4220 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4223 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
4225 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
4226 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
4227 the function were treating the return as a boolean value, which happened to
4228 work because 0=false and not-0=true, but is not correct code.
4230 68. The host_aton() function was not handling scoped IPv6 addresses (those
4231 with, for example, "%eth0" on the end) correctly.
4233 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
4234 negated items (that is, ~something) in unsigned ints. Some compilers
4235 apparently mutter when there is no cast.
4237 70. If an address verification called from an ACL failed, and did not produce a
4238 user-specific message (i.e. there was only a "system" message), nothing was
4239 put in $acl_verify_message. In this situation, it now puts the system
4242 71. Change 4.23/11 added synchronization checking at the start of an SMTP
4243 session; change 4.31/43 added the unwanted input to the log line - except
4244 that it did not do this in the start of session case. It now does.
4246 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
4247 This is wrong and can cause the other end to generate a synchronization
4248 error if it is another Exim or anything else that does the synchronization
4249 check. A QUIT command is no longer sent after a timeout.
4251 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
4252 during host lookups.
4254 74. The defer_ok option of callout verification was not working if it was used
4255 when verifying addresses in header lines, that is, for this case:
4257 verify = header_sender/callout=defer_ok
4259 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
4260 those file descriptors could be used for SMTP connections. If anything
4261 wrote to stderr (the example that came up was "warn" in embedded Perl), it
4262 could be sent to the SMTP client, causing chaos. The daemon now opens
4263 stdin, stdout, and stderr to /dev/null when it puts itself into the
4266 76. Arrange for output from Perl's "warn" command to be written to Exim's main
4267 log by default. The user can override this with suitable Perl magic.
4269 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
4270 the log message when discard triggers, was not working for the DATA ACL or
4271 for the non-SMTP ACL.
4273 78. Error message wording change in sieve.c.
4275 79. If smtp_accept_max_per_host was set, the number of connections could be
4276 restricted to fewer than expected, because the daemon was trying to set up
4277 a new connection before checking whether the processes handling previous
4278 connections had finished. The check for completed processes is now done
4279 earlier. On busy systems, this bug wouldn't be noticed because something
4280 else would have woken the daemon, and it would have reaped the completed
4283 80. If a message was submitted locally by a user whose login name contained one
4284 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
4285 It caused a spool format error. I have fixed the spool reading code. A
4286 related problem was that the "from" clause in the Received: line became
4287 illegal because of the space(s). It is now covered by ${quote_local_part.
4289 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
4292 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
4294 83. Patch from Sieve maintainer to fix unterminated string problem in
4295 "vacation" handling.
4297 84. Some minor changes to the Linux configuration files to help with other
4298 OS variants using glibc.
4300 85. One more patch for Sieve to update vacation handling to latest spec.
4303 ----------------------------------------------------
4304 See the note above about the 4.44 and 4.50 releases.
4305 ----------------------------------------------------
4311 1. Change 4.43/35 introduced a bug that caused file counts to be
4312 incorrectly computed when quota_filecount was set in an appendfile
4315 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4316 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4319 3. Give more explanation in the error message when the command for a transport
4320 filter fails to execute.
4322 4. There are several places where Exim runs a non-Exim command in a
4323 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4324 was being done only for the command run by the queryprogram router. It is
4325 now done for all such subprocesses. The other cases are: ${run, transport
4326 filters, and the commands run by the lmtp and pipe transports.
4328 5. Some older OS have a limit of 256 on the maximum number of file
4329 descriptors. Exim was using setrlimit() to set 1000 as a large value
4330 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4331 systems. I've change it so that if it can't get 1000, it tries for 256.
4333 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4334 was an oversight, and furthermore, ever since the addition of extra
4335 controls (e.g. 4.43/32), the checks on when to allow different forms of
4336 "control" were broken. There should now be diagnostics for all cases when a
4337 control that does not make sense is encountered.
4339 7. $recipients is now available in the predata ACL (oversight).
4341 8. Tidy the search cache before the fork to do a delivery from a message
4342 received from the command line. Otherwise the child will trigger a lookup
4343 failure and thereby defer the delivery if it tries to use (for example) a
4344 cached ldap connection that the parent has called unbind on.
4346 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4347 of $address_data from the recipient verification was clobbered by the
4348 sender verification.
4350 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4351 was its contents. (It was OK if the option was not defined at all.)
4353 11. A "Completed" log line is now written for messages that are removed from
4354 the spool by the -Mrm option.
4356 12. $host_address is now set to the target address during the checking of
4357 ignore_target_hosts.
4359 13. When checking ignore_target_hosts for an ipliteral router, no host name was
4360 being passed; this would have caused $sender_host_name to have been used if
4361 matching the list had actually called for a host name (not very likely,
4362 since this list is usually IP addresses). A host name is now passed as
4365 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4366 code that specifies a non-restarting handler (typically sigaction() in
4367 modern systems) in an attempt to fix a rare and obscure crash bug.
4369 15. Narrowed the window for a race in the daemon that could cause it to ignore
4370 SIGCHLD signals. This is not a major problem, because they are used only to
4371 wake it up if nothing else does.
4373 16. A malformed maildirsize file could cause Exim to calculate negative values
4374 for the mailbox size or file count. Odd effects could occur as a result.
4375 The maildirsize information is now recalculated if the size or filecount
4378 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4379 support for a long time. Removed HAVE_SYS_VFS_H.
4381 18. Updated exipick to current release from John Jetmore.
4383 19. Allow an empty sender to be matched against a lookup in an address list.
4384 Previously the only cases considered were a regular expression, or an
4387 20. Exim went into a mad DNS lookup loop when doing a callout where the
4388 host was specified on the transport, if the DNS lookup yielded more than
4391 21. The RFC2047 encoding function was originally intended for short strings
4392 such as real names; it was not keeping to the 75-character limit for
4393 encoded words that the RFC imposes. It now respects the limit, and
4394 generates multiple encoded words if necessary. To be on the safe side, I
4395 have increased the buffer size for the ${rfc2047: expansion operator from
4398 22. Failure to deliver a bounce message always caused it to be frozen, even if
4399 there was an errors_to setting on the router. The errors_to setting is now
4402 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4403 canonical form (fully expanded) before being placed in
4404 $sender_host_address.
4406 24. Updated eximstats to version 1.33
4408 25. Include certificate and key file names in error message when GnuTLS fails
4409 to set them up, because the GnuTLS error message doesn't include the name
4410 of the failing file when there is a problem reading it.
4412 26. Expand error message when OpenSSL has problems setting up cert/key files.
4415 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
4416 (this can affect the format of dates).
4418 28. exim_tidydb, when checking for the continued existence of a message for
4419 which it has found a message-specific retry record, was not finding
4420 messages that were in split spool directories. Consequently, it was
4421 deleting retry records that should have stayed in existence.
4423 29. eximstats updated to version 1.35
4424 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
4425 1.35 - bugfix such that pie charts by volume are generated correctly
4427 30. The SPA authentication driver was not abandoning authentication and moving
4428 on to the next authenticator when an expansion was forced to fail,
4429 contradicting the general specification for all authenticators. Instead it
4430 was generating a temporary error. It now behaves as specified.
4432 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
4433 (the order specifies the preference for clients). The order is now AES256,
4434 AES128, 3DES, ARCFOUR128.
4436 31. Small patch to Sieve code - explicitly set From: when generating an
4439 32. Exim crashed if a remote delivery caused a very long error message to be
4440 recorded - for instance if somebody sent an entire SpamAssassin report back
4441 as a large number of 550 error lines. This bug was coincidentally fixed by
4442 increasing the size of one of Exim's internal buffers (big_buffer) that
4443 happened as part of the Exiscan merge. However, to be on the safe side, I
4444 have made the code more robust (and fixed the comments that describe what
4447 33. Some experimental protocols are using DNS PTR records for new purposes. The
4448 keys for these records are domain names, not reversed IP addresses. The
4449 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4450 leaves it alone. Component reversal etc. now happens only for IP addresses.
4453 34. The host_aton() function is supposed to be passed a string that is known
4454 to be a valid IP address. However, in the case of IPv6 addresses, it was
4455 not checking this. This is a hostage to fortune. Exim now panics and dies
4456 if the condition is not met. A case was found where this could be provoked
4457 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4458 components; fortuitously, this particular loophole had already been fixed
4459 by change 4.50/55 or 4.44/33 above.
4461 If there are any other similar loopholes, the new check in host_aton()
4462 itself should stop them being exploited. The report I received stated that
4463 data on the command line could provoke the exploit when Exim was running as
4464 exim, but did not say which command line option was involved. All I could
4465 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4466 running as the user.
4469 35. There was a buffer overflow vulnerability in the SPA authentication code
4470 (which came originally from the Samba project). I have added a test to the
4471 spa_base64_to_bits() function which I hope fixes it.
4474 36. The daemon start-up calls getloadavg() while still root for those OS that
4475 need the first call to be done as root, but it missed one case: when
4476 deliver_queue_load_max is set with deliver_drop_privilege. This is
4477 necessary for the benefit of the queue runner, because there is no re-exec
4478 when deliver_drop_privilege is set.
4480 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4481 was in use, was not putting the data itself into the right store pool;
4482 consequently, it could be overwritten for a subsequent message in the same
4483 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4486 38. Sometimes the final signoff response after QUIT could fail to get
4487 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4488 < 0 before doing a fflush(). This bug looks as though it goes back to the
4489 introduction of TLS in release 3.20, but "sometimes" must have been rare
4490 because the tests only now provoked it.
4496 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
4497 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
4498 Now it returns an integer. A number of calls were still expecting a T/F
4499 return. Fortuitously, in all cases, the tests worked in OK situations,
4500 which is the norm. However, things would have gone wrong on any write
4501 failures on the smtp file descriptor. This function is used when sending
4502 messages over SMTP and also when doing verify callouts.
4504 2. When Exim is called to do synchronous delivery of a locally submitted
4505 message (the -odf or -odi options), it no longer closes stderr before doing
4508 3. Implemented the mua_wrapper option.
4510 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
4512 5. Implemented the functions header_remove(), header_testname(),
4513 header_add_at_position(), and receive_remove_recipient(), and exported them
4516 6. If an ACL "warn" statement specified the addition of headers, Exim already
4517 inserted X-ACL-Warn: at the start if there was no header name. However, it
4518 was not making this test for the second and subsequent header lines if
4519 there were newlines in the string. This meant that an invalid header could
4520 be inserted if Exim was badly configured.
4522 7. Allow an ACL "warn" statement to add header lines at the start or after all
4523 the Received: headers, as well as at the end.
4525 8. Added the rcpt_4xx retry error code.
4527 9. Added postmaster_mailfrom=xxx to callout verification option.
4529 10. Added mailfrom=xxxx to the callout verification option, for verify=
4532 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
4533 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
4535 12. Inserted some casts to stop certain compilers warning when using pointer
4536 differences as field lengths or precisions in printf-type calls (mostly
4537 affecting debugging statements).
4539 13. Added optional readline() support for -be (dynamically loaded).
4541 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
4542 same clock tick as a message's arrival, so that its received time was the
4543 same as the "first fail" time on the retry record, and that message
4544 remained on the queue past the ultimate address timeout, every queue runner
4545 would try a delivery (because it was past the ultimate address timeout) but
4546 after another failure, the ultimate address timeout, which should have then
4547 bounced the address, did not kick in. This was a "< instead of <=" error;
4548 in most cases the first failure would have been in the next clock tick
4549 after the received time, and all would be well.
4551 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
4552 being recognized when the domain list was tested by the match_domain
4553 condition in an expansion string.
4555 16. Added the ${str2b64: operator.
4557 17. Exim was always calling setrlimit() to set a large limit for the number of
4558 processes, without checking whether the existing limit was already
4559 adequate. (It did check for the limit on file descriptors.) Furthermore,
4560 errors from getrlimit() and setrlimit() were being ignored. Now they are
4561 logged to the main and panic logs, but Exim does carry on, to try to do its
4562 job under whatever limits there are.
4564 18. Imported PCRE 5.0.
4566 19. Trivial typo in log message " temporarily refused connection" (the leading
4569 20. If the log selector return_path_on_delivery was set and an address was
4570 redirected to /dev/null, the delivery process crashed because it assumed
4571 that a return path would always be set for a "successful" delivery. In this
4572 case, the whole delivery is bypassed as an optimization, and therefore no
4575 21. Internal re-arrangement: the function for sending a challenge and reading
4576 a response while authentication was assuming a zero-terminated challenge
4577 string. It's now changed to take a pointer and a length, to allow for
4578 binary data in such strings.
4580 22. Added the cyrus_sasl authenticator (code supplied by MBM).
4582 23. Exim was not respecting finduser_retries when seeking the login of the
4583 uid under which it was called; it was always trying 10 times. (The default
4584 setting of finduser_retries is zero.) Also, it was sleeping after the final
4585 failure, which is pointless.
4587 24. Implemented tls_on_connect_ports.
4589 25. Implemented acl_smtp_predata.
4591 26. If the domain in control=submission is set empty, Exim assumes that the
4592 authenticated id is a complete email address when it generates From: or
4593 Sender: header lines.
4595 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
4596 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
4597 chown and chgrp in /bin and hostname in /usr/bin.
4599 28. Exim was keeping the "process log" file open after each use, just as it
4600 does for the main log. This opens the possibility of it remaining open for
4601 long periods when the USR1 signal hits a daemon. Occasional processlog
4602 errors were reported, that could have been caused by this. Anyway, it seems
4603 much more sensible not to leave this file open at all, so that is what now
4606 29. The long-running daemon process does not normally write to the log once it
4607 has entered its main loop, and it closes the log before doing so. This is
4608 so that log files can straightforwardly be renamed and moved. However,
4609 there are a couple of unusual error situations where the daemon does write
4610 log entries, and I had neglected to close the log afterwards.
4612 30. The text of an SMTP error response that was received during a remote
4613 delivery was being truncated at 512 bytes. This is too short for some of
4614 the long messages that one sometimes sees. I've increased the limit to
4617 31. It is now possible to make retry rules that apply only when a message has a
4618 specific sender, in particular, an empty sender.
4620 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
4621 it possible to be selective about when SMTP synchronization is enforced.
4623 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
4625 32. Implemented hosts_connection_nolog.
4627 33. Added an ACL for QUIT.
4629 34. Setting "delay_warning=" to disable warnings was not working; it gave a
4632 35. Added mailbox_size and mailbox_filecount to appendfile.
4634 36. Added control = no_multiline_responses to ACLs.
4636 37. There was a bug in the logic of the code that waits for the clock to tick
4637 in the case where the clock went backwards by a substantial amount such
4638 that the microsecond fraction of "now" was more than the microsecond
4639 fraction of "then" (but the whole seconds number was less).
4641 38. Added support for the libradius Radius client library this is found on
4642 FreeBSD (previously only the radiusclient library was supported).
4648 1. When certain lookups returned multiple values in the form name=value, the
4649 quoting of the values was not always being done properly. Specifically:
4650 (a) If the value started with a double quote, but contained no whitespace,
4652 (b) If the value contained whitespace other than a space character (i.e.
4653 tabs or newlines or carriage returns) it was not quoted.
4654 This fix has been applied to the mysql and pgsql lookups by writing a
4655 separate quoting function and calling it from the lookup code. The fix
4656 should probably also be applied to nisplus, ibase and oracle lookups, but
4657 since I cannot test any of those, I have not disturbed their existing code.
4659 2. A hit in the callout cache for a specific address caused a log line with no
4660 reason for rejecting RCPT. Now it says "Previous (cached) callout
4661 verification failure".
4663 3. There was an off-by-one bug in the queryprogram router. An over-long
4664 return line was truncated at 256 instead of 255 characters, thereby
4665 overflowing its buffer with the terminating zero. As well as fixing this, I
4666 have increased the buffer size to 1024 (and made a note to document this).
4668 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
4669 when Exim is waiting for an SMTP response from a remote server, Exim
4670 restarts its select() call on the socket, thereby resetting its timeout.
4671 This is not a problem when such interrupts are rare. Somebody set up a cron
4672 job to run exiwhat every 2 minutes, which is less than the normal select()
4673 timeout (5 or 10 minutes). This meant that the select() timeout never
4674 kicked in because it was always reset. I have fixed this by comparing the
4675 time when an interrupt arrives with the time at the start of the first call
4676 to select(). If more time than the timeout has elapsed, the interrupt is
4677 treated as a timeout.
4679 5. Some internal re-factoring in preparation for the addition of Sieve
4680 extensions (by MH). In particular, the "personal" test is moved to a
4681 separate function, and given an option for scanning Cc: and Bcc: (which is
4682 not set for Exim filters).
4684 6. When Exim created an email address using the login of the caller as the
4685 local part (e.g. when creating a From: or Sender: header line), it was not
4686 quoting the local part when it contained special characters such as @.
4688 7. Installed new OpenBSD configuration files.
4690 8. Reworded some messages for syntax errors in "and" and "or" conditions to
4691 try to make them clearer.
4693 9. Callout options, other than the timeout value, were being ignored when
4694 verifying sender addresses in header lines. For example, when using
4696 verify = header_sender/callout=no_cache
4698 the cache was (incorrectly) being used.
4700 10. Added a missing instance of ${EXE} to the exim_install script; this affects
4701 only the Cygwin environment.
4703 11. When return_path_on_delivery was set as a log selector, if different remote
4704 addresses in the same message used different return paths and parallel
4705 remote delivery occurred, the wrong values would sometimes be logged.
4706 (Whenever a remote delivery process finished, the return path value from
4707 the most recently started remote delivery process was logged.)
4709 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
4710 lines when AUTH and/or TLS are in use. This is the "received protocol"
4711 field. Exim used to use "asmtp" for authenticated SMTP, without any
4712 indication (in the protocol name) for TLS use. Now it follows the RFC and
4713 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
4714 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
4715 names appear in log lines as well as in Received: header lines.
4717 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
4718 extensions, and comparison tests, and to fix some bugs.
4720 14. Changes to the "personal" filter test:
4722 (1) The test was buggy in that it was just doing the equivalent of
4723 "contains" tests on header lines. For example, if a user's address was
4724 anne@some.where, the "personal" test would incorrectly be true for
4726 To: susanne@some.where
4728 This test is now done by extracting each address from the header in turn,
4729 and checking the entire address. Other tests that are part of "personal"
4730 are now done using regular expressions (for example, to check local parts
4731 of addresses in From: header lines).
4733 (2) The list of non-personal local parts in From: addresses has been
4734 extended to include "listserv", "majordomo", "*-request", and "owner-*",
4735 taken from the Sieve specification recommendations.
4737 (3) If the message contains any header line starting with "List-" it is
4738 treated as non-personal.
4740 (4) The test for "circular" in the Subject: header line has been removed
4741 because it now seems ill-conceived.
4743 15. Minor typos in src/EDITME comments corrected.
4745 16. Installed latest exipick from John Jetmore.
4747 17. If headers_add on a router specified a text string that was too long for
4748 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
4749 of string_sprintf() is now avoided.
4751 18. $message_body_size was not set (it was always zero) when running the DATA
4752 ACL and the local_scan() function.
4754 19. For the "mail" command in an Exim filter, no default was being set for
4755 the once_repeat time, causing a random time value to be used if "once" was
4756 specified. (If the value happened to be <= 0, no repeat happened.) The
4757 default is now 0s, meaning "never repeat". The "vacation" command was OK
4758 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
4759 (I found it when inspecting the code).
4761 20. There is now an overall timeout for performing a callout verification. It
4762 defaults to 4 times the callout timeout, which applies to individual SMTP
4763 commands during the callout. The overall timeout applies when there is more
4764 than one host that can be tried. The timeout is checked before trying the
4765 next host. This prevents very long delays if there are a large number of
4766 hosts and all are timing out (e.g. when the network connections are timing
4767 out). The value of the overall timeout can be changed by specifying an
4768 additional sub-option for "callout", called "maxwait". For example:
4770 verify = sender/callout=5s,maxwait=20s
4772 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
4773 to the end before writing, but this should make it even safer).
4775 22. Exim was forgetting that it had advertised PIPELINING for the second and
4776 subsequent messages on an SMTP connection. It was also not resetting its
4777 memory on STARTTLS and an internal HELO.
4779 23. When Exim logs an SMTP synchronization error within a session, it now
4780 records whether PIPELINING has been advertised or not.
4782 24. Added 3 instances of "(long int)" casts to time_t variables that were being
4783 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
4784 rather than long int.
4786 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
4788 26. Added the never_mail option to autoreply.
4794 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
4795 crash if the getsockname() call failed; this can happen if a connection is
4796 closed very soon after it is established. The problem was simply in the
4797 order in which certain operations were done, causing Exim to try to write
4798 to the SMTP stream before it had set up the file descriptor. The bug has
4799 been fixed by making things happen in the correct order.
4805 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
4806 before the connection was closed, thus losing the rejection response.
4808 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
4809 some early Solaris releases, but causes trouble in current releases where
4810 socklen_t is defined.
4812 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
4815 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
4818 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
4819 that releasing the top of it at the end releases what was used for sub-
4820 expansions (unless the block got too big). However, discard this block if
4821 the first thing is a variable or header, so that we can use its block when
4822 it is dynamic (useful for very large $message_headers, for example).
4824 6. Lookups now cache *every* query, not just the most recent. A new, separate
4825 store pool is used for this. It can be recovered when all lookup caches are
4826 flushed. Lookups now release memory at the end of their result strings.
4827 This has involved some general refactoring of the lookup sources.
4829 7. Some code has been added to the store_xxx() functions to reduce the amount
4830 of flapping under certain conditions.
4832 8. log_incoming_interface used to affect only the <= reception log lines. Now
4833 it causes the local interface and port to be added to several more SMTP log
4834 lines, for example "SMTP connection from", and rejection lines.
4836 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
4838 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
4840 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
4841 could be overwritten at the end of the current message (or the start of a
4842 new message if it was set in a HELO ACL). The value is now preserved for
4843 the duration of the SMTP connection.
4845 12. If a transport had a headers_rewrite setting, and a matching header line
4846 contained an unqualified address, that address was qualified, even if it
4847 did not match any rewriting rules. The underlying bug was that the values
4848 of the flags that permit the existence of unqualified sender and recipient
4849 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
4850 non-local messages, and by -bnq for local messages) were not being
4851 preserved with the message after it was received.
4853 13. When Exim was logging an SMTP synchronization error, it could sometimes log
4854 "next input=" as part of the text comprising the host identity instead of
4855 the correct text. The code was using the same buffer for two different
4856 strings. However, depending on which order the printing function evaluated
4857 its arguments, the bug did not always show up. Under Linux, for example, my
4858 test suite worked just fine.
4860 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
4861 doesn't work with some older versions of Perl. It has been changed to "my",
4862 which in any case is probably the better facility to use.
4864 15. A really picky compiler found some instances of statements for creating
4865 error messages that either had too many or two few arguments for the format
4868 16. The size of the buffer for calls to the DNS resolver has been increased
4869 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
4870 for addresses that have a lot of PTR records. This alleviates a problem; it
4871 does not fully solve it.
4873 17. A dnsdb lookup for PTR records that receives more data than will fit in the
4874 buffer now truncates the list and logs the incident, which is the same
4875 action as happens when Exim is looking up a host name and its aliases.
4876 Previously in this situation something unpredictable would happen;
4877 sometimes it was "internal error: store_reset failed".
4879 18. If a server dropped the connection unexpectedly when an Exim client was
4880 using GnuTLS and trying to read a response, the client delivery process
4881 crashed while trying to generate an error log message.
4883 19. If a "warn" verb in an ACL added multiple headers to a message in a single
4884 string, for example:
4886 warn message = H1: something\nH2: something
4888 the text was added as a single header line from Exim's point of view
4889 though it ended up OK in the delivered message. However, searching for the
4890 second and subsequent header lines using $h_h2: did not work. This has been
4891 fixed. Similarly, if a system filter added multiple headers in this way,
4892 the routers could not see them.
4894 20. Expanded the error message when iplsearch is called with an invalid key to
4895 suggest using net-iplsearch in a host list.
4897 21. When running tests using -bh, any delays imposed by "delay" modifiers in
4898 ACLs are no longer actually imposed (and a message to that effect is
4901 22. If a "gecos" field in a passwd entry contained escaped characters, in
4902 particular, if it contained a \" sequence, Exim got it wrong when building
4903 a From: or a Sender: header from that name. A second bug also caused
4904 incorrect handling when an unquoted " was present following a character
4905 that needed quoting.
4907 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
4908 was not being matched caselessly.
4910 24. Arranged for all hyphens in the exim.8 source to be escaped with
4913 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
4914 properly. Recipient callout cache records were still being keyed to include
4915 the sender, even when use_sender was set false. This led to far more
4916 callouts that were necessary. The sender is no longer included in the key
4917 when use_sender is false.
4919 26. Added "control = submission" modifier to ACLs.
4921 27. Added the ${base62d: operator to decode base 62 numbers.
4923 28. dnsdb lookups can now access SRV records.
4925 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
4926 the configuration file.
4928 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
4929 (-v) mode. This makes the output for a verbose queue run more intelligible.
4931 31. Added a use_postmaster feature to recipient callouts.
4933 32. Added the $body_zerocount variable, containing the number of binary zero
4934 bytes in the message body.
4936 33. The time of last modification of the "new" subdirectory is now used as the
4937 "mailbox time last read" when there is a quota error for a maildir
4940 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
4942 35. Added +ignore_unknown as a special item in host lists.
4944 36. Code for decoding IPv6 addresses in host lists is now included, even if
4945 IPv6 support is not being compiled. This fixes a bug in which an IPv6
4946 address was recognized as an IP address, but was then not correctly decoded
4947 into binary, causing unexpected and incorrect effects when compared with
4954 1. Very minor rewording of debugging text in manualroute to say "list of
4955 hosts" instead of "hostlist".
4957 2. If verify=header_syntax was set, and a header line with an unqualified
4958 address (no domain) and a large number of spaces between the end of the
4959 name and the colon was received, the reception process suffered a buffer
4960 overflow, and (when I tested it) crashed. This was caused by some obsolete
4961 code that should have been removed. The fix is to remove it!
4963 3. When running in the test harness, delay a bit after writing a bounce
4964 message to get a bit more predictability in the log output.
4966 4. Added a call to search_tidyup() just before forking a reception process. In
4967 theory, someone could use a lookup in the expansion of smtp_accept_max_
4968 per_host which, without the tidyup, could leave open a database connection.
4970 5. Added the variables $recipient_data and $sender_data which get set from a
4971 lookup success in an ACL "recipients" or "senders" condition, or a router
4972 "senders" option, similar to $domain_data and $local_part_data.
4974 6. Moved the writing of debug_print from before to after the "senders" test
4977 7. Change 4.31/66 (moving the time when the Received: is generated) caused
4978 problems for message scanning, either using a data ACL, or using
4979 local_scan() because the Received: header was not generated till after they
4980 were called (in order to set the time as the time of reception completion).
4981 I have revised the way this works. The header is now generated after the
4982 body is received, but before the ACL or local_scan() are called. After they
4983 are run, the timestamp in the header is updated.
4989 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
4990 before starting a queue runner without re-exec. This happened only when
4991 deliver_drop_privilege was set or when the Exim user was set to root. The
4992 effect of the bug was that timeouts during subsequent deliveries caused
4993 crashes instead of being properly handled. The handler is now left at its
4994 default (and expected) setting.
4996 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
4997 message, again when deliver_drop_privilege is set or Exim is run as root.
4998 The bug described in (1) was not present in this case, but the tidying up
4999 of the other signals was missing. I have made the two cases consistent.
5001 3. The ignore_target_hosts setting on a manualroute router was being ignored
5002 for hosts that were looked up using the /MX notation.
5004 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
5007 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
5008 operated on the sender address. After changing the $sender_address to <>
5009 for the sender address verify, Exim was re-instated it as the original
5010 (before rewriting) address, but remembering that it had rewritten it, so it
5011 wasn't rewriting it again. This bug also had the effect of breaking the
5012 sender address verification caching when the sender address was rewritten.
5014 6. The ignore_target_hosts option was being ignored by the ipliteral router.
5015 This has been changed so that if the ip literal address matches
5016 ignore_target_hosts, the router declines.
5018 7. Added expansion conditions match_domain, match_address, and match_local_
5019 part (NOT match_host).
5021 8. The placeholder for the Received: header didn't have a length field set.
5023 9. Added code to Exim itself and to exim_lock to test for a specific race
5024 condition that could lead to file corruption when using MBX delivery. The
5025 issue is with the lockfile that is created in /tmp. If this file is removed
5026 after a process has opened it but before that process has acquired a lock,
5027 there is the potential for a second process to recreate the file and also
5028 acquire a lock. This could lead to two Exim processes writing to the file
5029 at the same time. The added code performs the same test as UW imapd; it
5030 checks after acquiring the lock that its file descriptor still refers to
5031 the same named file.
5033 10. The buffer for building added header lines was of fixed size, 8192 bytes.
5034 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
5037 11. Added the smtp_active_hostname option. If used, this will typically be made
5038 to depend on the incoming interface address. Because $interface_address is
5039 not set up until the daemon has forked a reception process, error responses
5040 that can happen earlier (such as "too many connections") no longer contain
5043 12. If an expansion in a condition on a "warn" statement fails because a lookup
5044 defers, the "warn" statement is abandoned, and the next ACL statement is
5045 processed. Previously this caused the whole ACL to be aborted.
5047 13. Added the iplsearch lookup type.
5049 14. Added ident_timeout as a log selector.
5051 15. Added tls_certificate_verified as a log selector.
5053 16. Added a global option tls_require_ciphers (compare the smtp transport
5054 option of the same name). This controls incoming TLS connections.
5056 17. I finally figured out how to make tls_require_ciphers do a similar thing
5057 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
5058 before starting the TLS session.
5060 18. Tabs are now shown as \t in -bP output.
5062 19. If the log selector return_path_on_delivery was set, Exim crashed when
5063 bouncing a message because it had too many Received: header lines.
5065 20. If two routers both had headers_remove settings, and the first one included
5066 a superfluous trailing colon, the final name in the first list and the
5067 first name in the second list were incorrectly joined into one item (with a
5068 colon in the middle).
5074 1. Added -C and -D options to the exinext utility, mainly to make it easier
5075 to include in the automated testing, but these could be helpful when
5076 multiple configurations are in use.
5078 2. The exinext utility was not formatting the output nicely when there was
5079 an alternate port involved in the retry record key, nor when there was a
5080 message id as well (for retries that were specific to a specific message
5081 and a specific host). It was also confused by IPv6 addresses, because of
5082 the additional colons they contain. I have fixed the IPv4 problem, and
5083 patched it up to do a reasonable job for IPv6.
5085 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
5086 delivery, the log line now contains "pipelined" if PIPELINING was used.
5088 4. An SMTP transport process used to panic and die if the bind() call to set
5089 an explicit outgoing interface failed. This has been changed; it is now
5090 treated in the same way as a connect() failure.
5092 5. A reference to $sender_host_name in the part of a conditional expansion
5093 that was being skipped was still causing a DNS lookup. This no longer
5096 6. The def: expansion condition was not recognizing references to header lines
5097 that used bh_ and bheader_.
5099 7. Added the _cache feature to named lists.
5101 8. The code for checking quota_filecount in the appendfile transport was
5102 allowing one more file than it should have been.
5104 9. For compatibility with Sendmail, the command line option
5112 and sets the incoming protocol and host name (for trusted callers). The
5113 host name and its colon can be omitted when only the protocol is to be set.
5114 Note the Exim already has two private options, -pd and -ps, that refer to
5115 embedded Perl. It is therefore impossible to set a protocol value of "d" or
5116 "s", but I don't think that's a major issue.
5118 10. A number of refactoring changes to the code, none of which should affect
5121 (a) The number of logging options was getting close to filling up the
5122 32-bit word that was used as a bit map. I have split them into two classes:
5123 those that are passed in the argument to log_write(), and those that are
5124 only ever tested independently outside of that function. These are now in
5125 separate 32-bit words, so there is plenty of room for expansion again.
5126 There is no change in the user interface or the logging behaviour.
5128 (b) When building, for example, log lines, the code previously used a
5129 macro that called string_cat() twice, in order to add two strings. This is
5130 not really sufficiently general. Furthermore, there was one instance where
5131 it was actually wrong because one of the argument was used twice, and in
5132 one call a function was used. (As it happened, calling the function twice
5133 did not affect the overall behaviour.) The macro has been replaced by a
5134 function that can join an arbitrary number of extra strings onto a growing
5137 (c) The code for expansion conditions now uses a table and a binary chop
5138 instead of a serial search (which was left over from when there were very
5139 few conditions). Also, it now recognizes conditions like "pam" even when
5140 the relevant support is not compiled in: a suitably worded error message is
5141 given if an attempt is made to use such a condition.
5143 11. Added ${time_interval:xxxxx}.
5145 12. A bug was causing one of the ddress fields not to be passed back correctly
5146 from remote delivery subprocesses. The field in question was not being
5147 subsequently used, so this caused to problems in practice.
5149 13. Added new log selectors queue_time and deliver_time.
5151 14. Might have fixed a bug in maildirsizefile handling that threw up
5152 "unexpected character" debug warnings, and recalculated the data
5153 unnecessarily. In any case, I expanded the warning message to give more
5156 15. Added the message "Restricted characters in address" to the statements in
5157 the default ACL that block characters like @ and % in local parts.
5159 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
5160 Three changes have been made:
5162 (a) There was a serious bug; a negative response to MAIL caused the whole
5163 recipient domain to be cached as invalid, thereby blocking all messages
5164 to all local parts at the same domain, from all senders. This bug has
5165 been fixed. The domain is no longer cached after a negative response to
5166 MAIL if the sender used is not empty.
5168 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
5171 (c) A new callout option, "use_sender" has been added for people who want
5172 the modified behaviour.
5178 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
5181 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
5182 indeed breaks things for older releases.
5184 3. Added additional logging to the case where there is a problem reading data
5185 from a filter that is running in a subprocess using a pipe, in order to
5186 try to track down a specific problem.
5188 4. Testing facility fudge: when running in the test harness and attempting
5189 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
5190 getting "No route to host". Convert this to a timeout.
5192 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
5195 6. Some OS don't have socklen_t but use size_t instead. This affects the
5196 fifth argument of getsockopt() amongst other things. This is now
5197 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
5198 can be set for individual OS. I have set it for SunOS5, OSF1, and
5199 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
5200 some earlier ones do not.
5202 7. Change 4.30/15 was not doing the test caselessly.
5204 8. The standard form for an IPv6 address literal was being rejected by address
5205 parsing in, for example, MAIL and RCPT commands. An example of this kind of
5206 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5207 this, as well as the form without the "IPv6" on the front (but only when
5208 address literals are enabled, of course).
5210 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
5212 10. Exim crashed if a message with an empty sender address specified by -f
5213 encountered a router with an errors_to setting. This could be provoked only
5214 by a command such as
5218 where an empty string was supplied; "<>" did not hit this bug.
5220 11. Installed PCRE release 4.5.
5222 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
5223 remained set. It is now erased.
5225 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
5226 times from message ids (which are base 36 rather than the normal 62).
5228 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
5229 were being counted as actual protocol errors, and logged if the log
5230 selector +smtp_protocol_error was set. One cannot be perfect in this test,
5231 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
5232 and DATA following a set of rejected RCPTs do not count as protocol errors.
5233 In other words, Exim assumes they were pipelined, though this may not
5234 actually be the case. Of course, in all cases the client gets an
5235 appropriate error code.
5237 15. If a lookup fails in an ACL condition, a message about the failure may
5238 be available; it is used if testing the ACL cannot continue, because most
5239 such messages specify what the cause of the deferral is. However, some
5240 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
5241 that caused an old message to be retained and used if a later statement
5242 caused a defer, replacing the real cause of the deferral.
5244 16. If an IP address had so many PTR records that the DNS lookup buffer
5245 was not large enough to hold them, Exim could crash while trying to process
5246 the truncated data. It now detects and logs this case.
5248 17. Further to 4.21/58, another change has been made: if (and only if) the
5249 first line of a message (the first header line) ends with CRLF, a bare LF
5250 in a subsequent header line has a space inserted after it, so as not to
5251 terminate the header.
5253 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
5254 unnecessarily, used atoi() instead of strtol(), and didn't check the
5255 termination when getting file sizes from file names by regex.
5257 19. Completely re-implemented the support for maildirsize files, in the light
5258 of a number of problems with the previous contributed implementation
5259 (4.30/29). In particular:
5261 . If the quota is zero, the maildirsize file is maintained, but no quota is
5264 . If the maildir directory does not exist, it is created before any attempt
5265 to write a maildirsize file.
5267 . The quota value in the file is just a cache; if the quota is changed in
5268 the transport, the new value overrides.
5270 . A regular expression is available for excluding directories from the
5273 20. The autoreply transport checks the characters in options that define the
5274 message's headers; it allows continued headers, but it was checking with
5275 isspace() after an embedded newline instead of explicitly looking for a
5278 21. If all the "regular" hosts to which an address was routed had passed their
5279 expiry times, and had not reached their retry times, the address was
5280 bounced, even if fallback hosts were defined. Now Exim should go on to try
5283 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
5284 equivalent code in the SMTP transport. Some hosts send humungous responses
5285 to HELO/EHLO, more than 1024 it seems.
5287 23. Refactoring: code in filter.c used (void *) for "any old type" but this
5288 gives compiler warnings in some environments. I've now done it "properly",
5291 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
5292 (because of problems with the built-in one) was declared to return uschar *
5293 instead of char *, causing compiler failure.
5295 25. Fixed a file descriptor leak when processing alias/forward files.
5297 26. Fixed a minor format string issue in dbfn.c.
5299 27. Typo in exim.c: ("dmbnz" for "dbmnz").
5301 28. If a filter file refered to $h_xxx or $message_headers, and the headers
5302 contained RFC 2047 "words", Exim's memory could, under certain conditions,
5305 29. When a sender address is verified, it is cached, to save repeating the test
5306 when there is more than one recipient in a message. However, when the
5307 verification involves a callout, it is possible for different callout
5308 options to be set for different recipients. It is too complicated to keep
5309 track of this in the cache, so now Exim always runs a verification when a
5310 callout is required, relying on the callout cache for the optimization.
5311 The overhead is duplication of the address routing, but this should not be
5314 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
5315 to be verified with callout=postmaster, and the main callout worked but the
5316 postmaster check failed, the verification correctly failed. However, if a
5317 subsequent RCPT command asked for sender verification *without* the
5318 postmaster check, incorrect caching caused this verification also to fail,
5321 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
5322 it was not caching the DNS options (qualify_single, search_parents) that
5323 were used when the lookup failed. A subsequent lookup with different
5324 options therefore always gave the same answer, though there were cases
5325 where it should not have. (Example: a "domains = !$mx_any" option on a
5326 dnslookup router: the "domains" option is always processed without any
5327 widening, but the router might have qualify_single set.) Now Exim uses the
5328 cached value only when the same options are set.
5330 32. Added John Jetmore's "exipick" utility to the distribution.
5332 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
5333 than a timeout (e.g. a certificate is required, and is not provided), an
5334 Exim server now closes the connection immediately. Previously it waited for
5335 the client to close - but if the client is SSL, it seems that they each
5336 wait for each other, leading to a delay before one of them times out.
5338 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
5339 maintained 0.8.x compatibility because I don't think many are using it, and
5340 it is clearly obsolete.
5342 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
5345 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
5346 client certificate was expired. A simple patch fixes this, though I don't
5347 understand the full logic of why the verify callback is called multiple
5350 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
5351 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
5352 which causes problems with some clients (such as the Certicom SSL Plus
5353 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
5354 disables the coutermeasure allowing Eudora to connect."
5356 38. Exim was not checking that a write() to a log file succeeded. This could
5357 lead to Bad Things if a log got too big, in particular if it hit a file
5358 size limit. Exim now panics and dies if it cannot write to a log file, just
5359 as it does if it cannot open a log file.
5361 39. Modified OS/Makefile-Linux so that it now contains
5363 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
5365 The two -D definitions ensure that Exim is compiled with large file
5366 support, which makes it possible to handle log files that are bigger than
5369 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
5370 instance) a domain was checked against a named list that involved a lookup,
5371 causing $domain_data to be set, then another domain was checked against the
5372 same list, then the first domain was re-checked, the value of $domain_data
5373 after the final check could be wrong. In particular, if the second check
5374 failed, it could be set empty. This bug probably also applied to
5377 41. The strip_trailing_dot option was not being applied to the address given
5378 with the -f command-line option.
5380 42. The code for reading a message's header from the spool was incrementing
5381 $received_count, but never initializing it. This meant that the value was
5382 incorrect (doubled) while delivering a message in the same process in which
5383 it was received. In the most common configuration of Exim, this never
5384 happens - a fresh exec is done - but it can happen when
5385 deliver_drop_privilege is set.
5387 43. When Exim logs an SMTP synchronization error - client data sent too soon -
5388 it now includes up to 150 characters of the unexpected data in the log
5391 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
5392 and building data strings. The size of both of these buffers was 10 000
5393 bytes - far larger than anybody would *ever* want, thought I. Needless to
5394 say, somebody hit the limit. I have increased the maximum line length to
5395 20 000 and the maximum data length of concatenated lines to 100 000. I have
5396 also fixed two bugs, because there was no checking on these buffers. Tsk,
5397 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
5398 buffer is too small.
5400 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
5401 lsearch lookups. Now it does.
5403 46. When parsing a route_list item in a manualroute router, a fixed-length
5404 buffer was used for the list of hosts. I made this 1024 bytes long,
5405 thinking that nobody would ever have a list of hosts that long. Wrong.
5406 Somebody had a whole pile of complicated expansion conditions, and the
5407 string was silently truncated, leading to an expansion error. It turns out
5408 that it is easier to change to an unlimited length (owing to other changes
5409 that have happened since this code was originally written) than to build
5410 structure for giving a limitation error. The length of the item that
5411 expands into the list of hosts is now unlimited.
5413 47. The lsearch lookup could not handle data where the length of text line was
5414 more than 4095 characters. Such lines were truncated, leading to shortened
5415 data being returned. It should now handle lines of any length.
5417 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
5418 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
5421 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
5422 replacing strings such as BIN_DIRECTORY in a source file: the replacement
5423 no longer happens in comment lines. A list of replacements is now placed
5424 at the head of all of the source files, except those whose only change is
5425 to replace PERL_COMMAND in the very first #! line.
5427 50. Replaced the slow insertion sort in queue.c, for sorting the list of
5428 messages on the queue, with a bottom-up merge sort, using code contributed
5429 by Michael Haardt. This should make operations like -bp somewhat faster on
5430 large queues. It won't affect queue runners, except when queue_run_in_order
5433 51. Installed eximstats 1.31 in the distribution.
5435 52. Added support for SRV lookups to the dnslookup router.
5437 53. If an ACL referred to $message_body or $message_body_end, the value was not
5438 reset for any messages that followed in the same SMTP session.
5440 54. The store-handling optimization for building very long strings was not
5441 differentiating between the different store pools. I don't think this
5442 actually made any difference in practice, but I've tidied it.
5444 55. While running the routers to verify a sender address, $sender_address
5445 was still set to the sender address. This is wrong, because when routing to
5446 send a bounce to the sender, it would be empty. Therefore, I have changed
5447 it so that, while verifying a sender address, $sender_address is set to <>.
5448 (There is no change to what happens when verifying a recipient address.)
5450 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
5451 target A or AAAA records (if not already returned) without resetting the
5452 qualify_single or search_parents options of the DNS resolver. These are
5453 inappropriate in this case because the targets of MX and SRV records must
5454 be FQDNs. A broken DNS record could cause trouble if it happened to have a
5455 target that, when qualified, matched something in the local domain. These
5456 two options are now turned off when doing these lookups.
5458 57. It seems that at least some releases of Reiserfs (which does not have the
5459 concept of a fixed number of inodes) returns zero and not -1 for the
5460 number of available inodes. This interacted badly with check_spool_inodes,
5461 which assumed that -1 was the "no such thing" setting. What I have done is
5462 to check that the total number of inodes is greater than zero before doing
5463 the test of how many are available.
5465 58. When a "warn" ACL statement has a log_message modifier, the message is
5466 remembered, and not repeated. This is to avoid a lot of repetition when a
5467 message has many recipients that cause the same warning to be written.
5468 Howewer, Exim was preserving the list of already written lines for an
5469 entire SMTP session, which doesn't seem right. The memory is now reset if a
5470 new message is started.
5472 59. The "rewrite" debugging flag was not showing the result of rewriting in the
5473 debugging output unless log_rewrite was also set.
5475 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
5476 of (int)(handle) when we know that handle contains (void *)(-1).
5478 61. The Exim daemon panic-logs an error return when it closes the incoming
5479 connection. However "connection reset by peer" seems to be common, and
5480 isn't really an error worthy of noting specially, so that particular error
5483 62. When Exim is trying to find all the local interfaces, it used to panic and
5484 die if the ioctl to get the interface flags failed. However, it seems that
5485 on at least one OS (Solaris 9) it is possible to have an interface that is
5486 included in the list of interfaces, but for which you get a failure error
5487 for this call. This happens when the interface is not "plumbed" into a
5488 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
5489 failure of the "get flags" call assumes that the interface is down.
5491 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
5492 makes life easier for people who are doing arithmetic on fields extracted
5493 from dates, where you often get leading zeros that should not be
5494 interpreted as octal.
5496 64. Added qualify_domain to the redirect router, to override the global
5499 65. If a pathologically long header line contained very many addresses (the
5500 report of this problem mentioned 10 000) and each of them was rewritten,
5501 Exim could use up a very large amount of memory. (It kept on making new
5502 copies of the header line as it rewrote, and never released the old ones.)
5503 At the expense of a bit more processing, the header rewriting function has
5504 been changed so that it no longer eats memory in this way.
5506 66. The generation of the Received: header has been moved from the time that a
5507 message starts to be received, to the time that it finishes. The timestamp
5508 in the Received: header should now be very close to that of the <= log
5509 line. There are two side-effects of this change:
5511 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
5512 logged header lines no longer include the local Received: line, because
5513 it has not yet been created. The same applies to a copy of the message
5514 that is returned to a non-SMTP sender when a message is rejected.
5516 (b) When a filter file is tested using -bf, no additional Received: header
5517 is added to the test message. After some thought, I decided that this
5520 This change does not affect the value of $received_for. It is still set
5521 after address rewriting, but before local_scan() is called.
5523 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
5525 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
5526 gave an unhelpful panic error message, and a defer error. I have managed to
5527 change this behaviour so that it now rejects any supplied certificate,
5528 which seems right, as the list of acceptable certificates is empty.
5530 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
5531 gave an unhelpful defer error. I have not managed to make this reject any
5532 supplied certificates, but the error message it gives is "no certificate
5533 supplied", which is not helpful.
5535 70. exigrep's output now also includes lines that are not associated with any
5536 message, but which match the given pattern. Implemented by a patch from
5537 Martin Sluka, which also tidied up the Perl a bit.
5539 71. Recipient callout verification, like sender verification, was using <> in
5540 the MAIL FROM command. This isn't really the right thing, since the actual
5541 sender may affect whether the remote host accepts the recipient or not. I
5542 have changed it to use the actual sender in the callout; this means that
5543 the cache record is now keyed on a recipient/sender pair, not just the
5544 recipient address. There doesn't seem to be a real danger of callout loops,
5545 since a callout by the remote host to check the sender would use <>.
5546 [SEE ABOVE: changed after hitting problems.]
5548 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
5549 temporary errors. However, in the case of such a code being given after
5550 the end of a data transmission (i.e. after ".") Exim was failing to write
5551 a retry record for the message. (Yes, there was some broken host that was
5552 actually sending 8xx at this point.)
5554 73. An unknown lookup type in a host list could cause Exim to panic-die when
5555 the list was checked. (An example that provoked this was putting <; in the
5556 middle of a list instead of at the start.) If this happened during a DATA
5557 ACL check, a -D file could be left lying around. This kind of configuration
5558 error no longer causes Exim to die; instead it causes a defer errror. The
5559 incident is still logged to the main and panic logs.
5561 74. Buglet left over from Exim 3 conversion. The message "too many messages
5562 in one connection" was written to the rejectlog but not the mainlog, except
5563 when address rewriting (yes!) was being logged.
5565 75. Added write_rejectlog option.
5567 76. When a system filter was run not as root (that is, when system_filter_user
5568 was set), the values of the $n variables were not being returned to the
5569 main process; thus, they were not subsequently available in the $sn
5572 77. Added +return_path_on_delivery log selector.
5574 78. A connection timeout was being treated differently from recipients deferred
5575 when testing hosts_max_try with a message that was older than the host's
5576 retry timeout. (The host should not be counted, thus allowing all hosts to
5577 be tried at least once before bouncing.) This may have been the cause of an
5578 occasionally reported bug whereby a message would remain on the queue
5579 longer than the retry timeout, but would be bounced if a delivery was
5580 forced. I say "may" because I never totally pinned down the problem;
5581 setting up timeout/retry tests is difficult. See also the next item.
5583 79. The ultimate address timeout was not being applied to errors that involved
5584 a combination of host plus message (for example, a timeout on a MAIL
5585 command). When an address resolved to a number of possible hosts, and they
5586 were not all tried for each delivery (e.g. because of hosts_max_try), a
5587 message could remain on the queue longer than the retry timeout.
5589 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
5592 81. Fixed an obscure SMTP outgoing bug which required at least the following
5593 conditions: (a) there was another message waiting for the same server;
5594 (b) the server returned 5xx to all RCPT commands in the first message so
5595 that the message was not completed; (c) the server dropped the connection
5596 or gave a negative response to the RSET that Exim sends to abort the
5597 transaction. The observed case was a dropped connection after DATA that had
5598 been sent in pipelining mode. That is, the server had advertised PIPELINING
5599 but was not implementing it correctly. The effect of the bug was incorrect
5600 behaviour, such as trying another host, and this could lead to a crash.
5606 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
5607 and daemon.c were passed as pointers to ints; they should have been
5608 pointers to socklen_t variables (which are typically unsigned ints).
5610 2. Some signed/unsigned type warnings in the os.c file for Linux have been
5613 3. Fixed a really odd bug that affected only the testing scheme; patching a
5614 certain fixed string in the binary changed the value of another string that
5615 happened to be identical to the end of the original first string.
5617 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
5618 name", it returns that address as the IP address. On some operating
5619 systems (e.g. Solaris), it also passes back the IP address string as the
5620 "host name". However, on others (e.g. Linux), it passes back an empty
5621 string. Exim wasn't checking for this, and was changing the host name to an
5622 empty string, assuming it had been canonicized.
5624 5. Although rare, it is permitted to have more than one PTR record for a given
5625 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
5626 all the names associated with an address, because they do in Solaris.
5627 However, it seems that they do not in Linux for data that comes from the
5628 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
5629 I found this out when I moved to a new Linux workstation and tried to run
5630 the Exim test suite.
5632 To get round this problem I have changed the code so that it now does its
5633 own call to the DNS to look up PTR records when searching for a host name.
5634 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
5635 addresses that are only in /etc/hosts are still found.
5637 This behaviour is, however, controlled by an option called host_lookup_
5638 order, which defaults to "bydns:byaddr". If people want to use the other
5639 order, or indeed, just use one or the other means of lookup, they can
5640 specify it in this variable.
5642 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
5643 some operating systems, this comes back from gethostbyaddr() as an empty
5644 string, and this is what Exim used to test for. However, it seems that in
5645 other systems, "." is yielded. Exim now tests for this case too.
5647 7. The values of check_spool_space and check_log_space are now held internally
5648 as a number of kilobytes instead of an absolute number of bytes. If a
5649 numbers is specified without 'K' or 'M', it is rounded up to the nearest
5650 kilobyte. This means that much larger values can be stored.
5652 8. Exim monitor: an attempt to get the action menu when not actually pointing
5653 at a message produces an empty menu entitled "No message selected". This
5654 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
5655 no entries in it ("Shell widget menu has zero width and/or height"). So I
5656 have added a single, blank menu entry in this case.
5658 9. Added ${quote_local_part.
5660 10. MIME decoding is now applied to the contents of Subject: header lines when
5663 11. Now that a reference to $sender_host_address automatically causes a reverse
5664 lookup to occur if necessary (4.13/18), there is no need to arrange for a
5665 host lookup before query-style lookups in lists that might use this
5666 variable. This has therefore been abolished, and the "net-" prefix is no
5667 longer necessary for query-style lookups.
5669 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
5670 have been a typo for LFLAGS, so it has been changed.
5672 13. The install script calls Exim with "-C /dev/null" in order to find the
5673 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
5674 to be output. Howeve, since Exim outputs its version number before the
5675 error, it didn't break the script. It just looked ugly. I fixed this by
5676 always allowing "-C /dev/null" if the caller is root.
5678 14. Ignore overlarge ACL variable number when reading spool file - insurance
5679 against a later release with more variables having written the file.
5681 15. The standard form for an IPv6 address literal was being rejected by EHLO.
5682 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5683 this, as well as the form without the "IPv6" on the front.
5685 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
5686 OS/Makefile-Darwin file.
5688 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
5689 only to LDAP libraries that do not have LDAP_OPT_DEREF.
5691 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
5692 variable. However, some OS use ints for inodes. I've added cast to long int
5693 to get rid of the compiler warning.
5695 19. I had forgotten to lock out "/../" in configuration file names when
5696 ALT_CONFIG_PREFIX was set.
5698 20. Routers used for verification do not need to specify transports. However,
5699 if such a router generated a host list, and callout was configured, Exim
5700 crashed, because it could not find a port number from the (non-existent)
5701 transport. It now assumes port 25 in this circumstance.
5703 21. Added the -t option to exigrep.
5705 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
5706 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
5707 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
5708 them is set, LOOKUP_LSEARCH is forced.
5710 23. "exim -bV" now outputs a list of lookups that are included in the binary.
5712 24. Added sender and host information to the "rejected by local_scan()" log
5713 line; previously there was no indication of these.
5715 25. Added .include_if_exists.
5717 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
5718 Linux. It turns out that not all file systems support this. Apparently some
5719 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
5720 do it.) To cope with this, the error EINVAL, which means that sync-ing is
5721 not supported on the file descriptor, is now ignored when Exim is trying to
5722 sync a directory. This applies only to Linux.
5724 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
5726 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
5727 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
5730 29. Added support for maildirsize files from supplied patch (modified a bit).
5732 30. The use of :fail: followed by an empty string could lead Exim to respond to
5733 sender verification failures with (e.g.):
5735 550 Verification failed for <xxx>
5736 550 Sender verify failed
5738 where the first response line was missing the '-' that indicates it is not
5739 the final line of the response.
5741 31. The loop for finding the name of the user that called Exim had a hardwired
5742 limit of 10; it now uses the value of finduser_retries, which is used for
5743 all other user lookups.
5745 32. Added $received_count variable, available in data and not_smtp ACLs, and at
5748 33. Exim was neglecting to zero errno before one call of strtol() when
5749 expanding a string and expecting an integer value. On some systems this
5750 resulted in spurious "integer overflow" errors. Also, it was casting the
5751 result into an int without checking.
5753 34. Testing for a connection timeout using "timeout_connect" in the retry rules
5754 did not work. The code looks as if it has *never* worked, though it appears
5755 to have been documented since at least releast 1.62. I have made it work.
5757 35. The "timeout_DNS" error in retry rules, also documented since at least
5758 1.62, also never worked. As it isn't clear exactly what this means, and
5759 clearly it isn't a major issue, I have abolished the feature by treating it
5760 as "timeout", and writing a warning to the main and panic logs.
5762 36. The display of retry rules for -brt wasn't always showing the error code
5765 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
5766 timeout_connect_A, timeout_connect_MX.
5768 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
5769 to the empty sender.
5771 39. The daemon was not analysing the content of -oX till after it had closed
5772 stderr and disconnected from the controlling terminal. This meant that any
5773 syntax errors were only noted on the panic log, and the return code from
5774 the command was 0. By re-arranging the code a little, I've made the
5775 decoding happen first, so such errors now appear on stderr, and the return
5776 code is 1. However, the actual setting up of the sockets still happens in
5777 the disconnected process, so errors there are still only recorded on the
5780 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
5781 connections (as happens on some IP stacks) was logged at start up time as
5782 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
5783 it from "IPv6 and IPv4", which means that two separate sockets are being
5786 41. The debug output for gethostbyname2() or getipnodebyname() failures now
5787 says whether AF_INET or AF_INET6 was passed as an argument.
5789 42. Exiwhat output was messed up when time zones were included in log
5792 43. Exiwhat now gives more information about the daemon's listening ports,
5793 and whether -tls-on-connect was used.
5795 44. The "port" option of the smtp transport is now expanded.
5797 45. A "message" modifier in a "warn" statement in a non-message ACL was being
5798 silently ignored. Now an error message is written to the main and panic
5801 46. There's a new ACL modifier called "logwrite" which writes to a log file
5802 as soon as it is encountered.
5804 47. Added $local_user_uid and $local_user_gid at routing time.
5806 48. Exim crashed when trying to verify a sender address that was being
5809 49. Exim was recognizing only a space character after ".include". It now also
5810 recognizes a tab character.
5812 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
5813 extracting the relevant information from the specification. The man page no
5814 longer contains scrambled data for the -d option, and I've added a section
5815 at the front about calling Exim under different names.
5817 51. Added "extra_headers" argument to the "mail" command in filter files.
5819 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
5822 53. Installed eximstats 1.29.
5824 54. Added transport_filter_timeout as a generic transport option.
5826 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
5827 Cc: header lines. This was required by RFC 822, but it not required by RFC
5830 56. Exim used to add From:, Date:, and Message-Id: header lines to any
5831 incoming messages that did not have them. Now it does so only if the
5832 message originates locally, that is, if there is no associated remote host
5833 address. When Resent- header lines are present, this applies to the Resent-
5834 lines rather than the non-Resent- lines.
5836 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
5837 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
5839 58. Messages for configuration errors now include the name of the main
5840 configuration file - useful now that there may be more than one file in a
5841 list (.included file names were always shown).
5843 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
5844 for those rare installations that do not start the daemon as root or run it
5845 setuid root. I've cut out the call to initgroups() if the daemon is not
5848 60. The Exim user and group can now be bound into the binary as text strings
5849 that are looked up at the start of Exim's processing.
5851 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
5853 62. Added $mailstore_basename variable.
5855 63. Installed patch to sieve.c from Michael Haardt.
5857 64. When Exim failed to open the panic log after failing to open the main log,
5858 the original message it was trying to log was written to stderr and debug
5859 output, but if they were not available (the usual case in production), it
5860 was lost. Now it is written to syslog before the two lines that record the
5861 failures to open the logs.
5863 65. Users' Exim filters run in subprocesses under the user's uid. It is
5864 possible for a "deliver" command or an alias in a "personal" command to
5865 provoke an address rewrite. If logging of address rewriting is configured,
5866 this fails because the process is not running as root or exim. There may be
5867 a better way of dealing with this, but for the moment (because 4.30 needs
5868 to be released), I have disabled address rewrite logging when running a
5869 filter in a non-root, non-exim process.
5875 1. The buildconfig auxiliary program wasn't quoting the value set for
5876 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
5877 not defined. This bug was masked in 4.22 by the effect that was fixed in
5880 2. Some messages that were rejected after a message id was allocated were
5881 shown as "incomplete" by exigrep. It no longer does this for messages that
5882 are rejected by local_scan() or the DATA or non-SMTP ACLs.
5884 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
5885 have allow_domain_literals set, the ID did not get logged in the <= line.
5886 Domain literals are now always recognized in Message-ID: header lines.
5888 4. The first argument for a ${extract expansion item is the key name or field
5889 number. Leading and trailing spaces in this item were not being ignored,
5890 causing some misleading effects.
5892 5. When deliver_drop_privilege was set, single queue runner processes started
5893 manually (i.e. by the command "exim -q") or by the daemon (which uses the
5894 same command in the process it spins off) were not dropping privilege.
5896 6. When the daemon running as "exim" started a queue runner, it always
5897 re-executed Exim in the spun-off process. This is a waste of effort when
5898 deliver_drop_privilege is set. The new process now just calls the
5899 queue-runner function directly.
5905 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
5908 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
5909 ignored. Though the use of -oP was forcing the writing of a pid file, it
5910 was always written to the default place.
5912 3. If the message "no IP address found for host xxxx" is generated during
5913 incoming verification, it is now followed by identification of the incoming
5914 connection (so you can more easily find what provoked it).
5916 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
5918 5. Added some features to "harden" Exim a bit more against certain attacks:
5920 (a) There is now a build-time option called FIXED_NEVER_USERS that can
5921 be put in Local/Makefile. This is like the never_users runtime option,
5922 but it cannot be overridden. The default setting is "root".
5924 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
5925 prefix string with which any file named in a -C command line option
5928 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
5929 is retained for -C and -D only if the caller of Exim is root. Without
5930 it, the exim user may also use -C and -D and retain privilege.
5932 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
5933 command line option is disabled.
5935 6. Macro names set by the -D option must start with an upper case letter, just
5936 like macro names defined in the configuration file.
5938 7. Added "dereference=" facility to LDAP.
5940 8. Two instances of the typo "uknown" in the source files are fixed.
5942 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
5943 the Configure-Makefile script screwed up while processing it.
5945 10. Incorporated PCRE 4.4.
5947 11. The SMTP synchronization check was not operating right at the start of an
5948 SMTP session. For example, it could not catch a HELO sent before the client
5949 waited for the greeting. There is now a check for outstanding input at the
5950 point when the greeting is written. Because of the duplex, asynchronous
5951 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
5952 way, but not yet received, when the check is performed.
5954 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
5955 on TCP/IP sockets, because this apparently causes some broken clients to
5958 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
5959 unchanged) from the Cygwin maintainer.
5961 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
5962 was supported instead of testing for mbx. Effectively a typo.
5964 15. The spa authenticator server code was not checking that the input it
5965 received was valid base64.
5967 16. The debug output line for the "set" modifier in ACLs was not showing the
5968 name of the variable that was being set.
5970 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
5972 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
5973 lookup on its own. Something else was needed to trigger the lookup. For
5974 example, a match in host_lookup or the need for a host name in a host list.
5975 Now, if $sender_host_name is referenced and the host name has not yet been
5976 looked up, a lookup is performed. If the lookup fails, the variable remains
5977 empty, and $host_lookup_failed is set to "1".
5979 19. Added "eqi" as a case-independent comparison operator.
5981 20. The saslauthd authentication condition could segfault if neither service
5982 nor realm was specified.
5984 21. If an overflowing value such as "2048M" was set for message_size_limit, the
5985 error message that was logged was misleading, and incoming SMTP
5986 connections were dropped. The message is now more accurate, and temporary
5987 errors are given to SMTP connections.
5989 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
5990 (except RSET) with a 421 error, until QUIT is received. However, it was
5991 failing to send a response to QUIT.
5993 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
5994 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
5995 ACL is now run after the helo_try_verify_hosts code.
5997 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
5998 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
5999 case-independent, so other case variants are also recognized). Apparently
6000 some systems use these upper case variants.
6002 25. If more than two messages were waiting for the same host, and a transport
6003 filter was specified for the transport, Exim sent two messages over the
6004 same TCP/IP connection, and then failed with "socket operation on non-
6005 socket" when it tried to send the third.
6007 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
6009 27. The extern definition of crypt16() in expand.c was not being excluded when
6010 the OS had its own crypt16() function.
6012 28. Added bounce_return_body as a new option, and bounce_return_size_limit
6013 as a preferred synonym for return_size_limit, both as an option and as an
6016 29. Added LIBS=-liconv to OS/Makefile-OSF1.
6018 30. Changed the default configuration ACL to relax the local part checking rule
6019 for addresses that are not in any local domains. For these addresses,
6020 slashes and pipe symbols are allowed within local parts, but the sequence
6021 /../ is explicitly forbidden.
6023 31. SPA server authentication was not clearing the challenge buffer before
6026 32. log_message in a "warn" ACL statement was writing to the reject log as
6027 well as to the main log, which contradicts the documentation and doesn't
6028 seem right (because no rejection is happening). So I have stopped it.
6030 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
6031 However, I am unable to do any testing of this.
6033 34. Fixed an infelicity in the appendfile transport. When checking directories
6034 for a mailbox, to see if any needed to be created, it was accidentally
6035 using path names with one or more superfluous leading slashes; tracing
6036 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
6038 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
6039 contents are added to the log line that is written for every discarded
6040 recipient. (Previously a log_message setting was ignored.)
6042 36. The ${quote: operator now quotes the string if it is empty.
6044 37. The install script runs exim in order to find its version number. If for
6045 some reason other than non-existence or emptiness, which it checks, it
6046 could not run './exim', it was installing it with an empty version number,
6047 i.e. as "exim-". This error state is now caught, and the installation is
6050 38. An argument was missing from the function that creates an error message
6051 when Exim fails to connect to the socket for saslauthd authentication.
6052 This could cause Exim to crash, or give a corrupted message.
6054 39. Added isip, isip4, and isip6 to ${if conditions.
6056 40. The ACL variables $acl_xx are now saved with the message, and can be
6057 accessed later in routers, transports, and filters.
6059 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
6060 strings in the file are not string-expanded.
6062 42. If a MAIL command specified a SIZE value that was too large to fit into an
6063 int variable, the check against message_size_limit failed. Such values are
6064 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
6065 day this will have to be increased, but I don't think I want to be around
6066 when emails are that large.
6073 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
6074 iconv() is not standard in FreeBSD.
6076 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
6077 IPv6 enabled. The observed symptom was a segmentation fault on return from
6078 the function os_common_find_running_interfaces() in src/os.c.
6080 3. In the check_special_case() function in daemon.c I had used "errno" as an
6081 argument name, which causes warnings on some systems. This was basically a
6082 typo, since it was named "eno" in the comments!
6084 4. The code that waits for the clock to tick (at a resolution of some fraction
6085 of a second) so as to ensure message-id uniqueness was always waiting for
6086 at least one whole tick, when it could have waited for less. [This is
6087 almost certainly not relevant at current processor speeds, where it is
6088 unlikely to ever wait at all. But we try to future-proof.]
6090 5. The function that sleeps for a time interval that includes fractions of a
6091 second contained a race. It did not block SIGALRM between setting the
6092 timer, and suspending (a couple of lines later). If the interval was short
6093 and the sigsuspend() was delayed until after it had expired, the suspension
6094 never ended. On busy systems this could lead to processes getting stuck for
6097 6. Some uncommon configurations may cause a lookup to happen in a queue runner
6098 process, before it forks any delivery processes. The open lookup caching
6099 mechanism meant that the open file or database connection was passed into
6100 the delivery process. The problem was that delivery processes always tidy
6101 up cached lookup data. This could cause a problem for the next delivery
6102 process started by the queue runner, because the external queue runner
6103 process does not know about the closure. So the next delivery process
6104 still has data in the lookup cache. In the case of a file lookup, there was
6105 no problem because closing a file descriptor in a subprocess doesn't affect
6106 the parent. However, if the lookup was caching a connection to a database,
6107 the connection was closed, and the second delivery process was likely to
6108 see errors such as "PGSQL: query failed: server closed the connection
6109 unexpectedly". The problem has been fixed by closing all cached lookups
6110 in a queue runner before running a delivery process.
6112 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
6113 seem to have the "const" qualifier which it has on other OS. I've
6116 8. Change 4.21/2 was too strict. It is only if there are two authenticators
6117 *of the same type* (client or server) with the same public name that an
6118 error should be diagnosed.
6120 9. When Exim looked up a host name for an IP address, but failed to find the
6121 original IP address when looking up the host name (a safety check), it
6122 output the message "<ip address> does not match any IP for NULL", which was
6123 confusing, to say the least. The bug was that the host name should have
6124 appeared instead of "NULL".
6126 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
6127 user that is built into the binary, and the -C or -D options is used, root
6128 privilege is dropped before the configuration file is read. In addition,
6129 logging is switched to stderr instead of the normal log files. If the
6130 configuration then re-defines the Exim user, the unprivileged environment
6131 is probably not what is expected, so Exim logs a panic warning message (but
6134 However, if deliver_drop_privilege is set, the unprivileged state may well
6135 be exactly what is intended, so the warning has been cut out in that case,
6136 and Exim is allowed to try to write to its normal log files.
6142 1. smtp_return_error_details was not giving details for temporary sender
6143 or receiver verification errors.
6145 2. Diagnose a configuration error if two authenticators have the same public
6148 3. Exim used not to create the message log file for a message until the first
6149 delivery attempt. This could be confusing when incoming messages were held
6150 for policy or load reasons. The message log file is now created at the time
6151 the message is received, and an initial "Received" line is written to it.
6153 4. The automatically generated man page for command line options had a minor
6154 bug that caused no ill effects; however, a more serious problem was that
6155 the procedure for building the man page automatically didn't always
6156 operate. Consequently, release 4.20 contains an out-of-date version. This
6157 shouldn't happen again.
6159 5. When building Exim with embedded Perl support, the script that builds the
6160 Makefile was calling 'perl' to find its compile-time parameters, ignoring
6161 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
6163 6. The freeze_tell option was not being used for messages that were frozen on
6164 arrival, either by an ACL or by local_scan().
6166 7. Added the smtp_incomplete_transaction log selector.
6168 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
6169 was accepting AUTH without a new EHLO.
6171 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
6172 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
6175 10. Logging of TCP/IP connections (when configured) now happens in the main
6176 daemon process instead of the child process, so that the TCP/IP connection
6177 count is more accurate (but it can never be perfect).
6179 11. The use of "drop" in a nested ACL was not being handled correctly in the
6180 outer ACL. Now, if condition failure induced by the nested "drop" causes
6181 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
6182 or "require"), the connection is dropped.
6184 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
6185 that yield "discard" can now be used with an "accept" or a "discard" verb,
6186 but an error is generated for any others (because I can't see a useful way
6187 to define what should happen).
6189 13. When an ACL is read dynamically from a file (or anywhere else), the lines
6190 are now processed in the same way as lines in the Exim configuration file.
6191 In particular, continuation lines are supported.
6193 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
6195 15. Added -ti meaning -t -i.
6197 16. Check for letters, digits, hyphens, and dots in the names of dnslist
6198 domains, and warn by logging if others are found.
6200 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
6201 returned from GIFCONF when Exim is trying to find the list of interfaces on
6202 a host. The code in os.c has been modified to copy each ifreq to an aligned
6203 structure in all cases.
6205 Also, in some cases, the returned ifreq's were being copied to a 'struct
6206 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
6207 means the last couple of bytes of an IPv6 address could be chopped if the
6208 ifreq contained only a normal sockaddr (14 bytes storage).
6210 18. Named domain lists were not supported in the hosts_treat_as_local option.
6211 An entry such as +xxxx was not recognized, and was treated as a literal
6214 19. Ensure that header lines added by a DATA ACL are included in the reject log
6215 if the ACL subsequently rejects the message.
6217 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
6218 MD5 (which is deprecated).
6220 21. When testing a filter file using -bf, Exim was writing a message when it
6221 took the sender from a "From " line in the message, but it was not doing so
6222 when it took $return_path from a Return-Path: header line. It now does.
6224 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
6225 with a valid header line field name (a series of printing characters
6226 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
6228 23. Changed "disc" in the source to "disk" to conform to the documentation and
6229 the book and for uniformity.
6231 24. Ignore Sendmail's -Ooption=value command line item.
6233 25. When execve() failed while trying to run a command in a pipe transport,
6234 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
6235 could be confused with a return value of 69 from the command itself. This
6236 has been changed to 127, the value the shell returns if it is asked to run
6237 a non-existent command. The wording for the related log line suggests a
6238 non-existent command as the problem.
6240 26. If received_header_text expands to an empty string, do not add a Received:
6241 header line to the message. (Well, it adds a token one on the spool, but
6242 marks it "old" so that it doesn't get used or transmitted.)
6244 27. Installed eximstats 1.28 (addition of -nt option).
6246 28. There was no check for failure on the call to getsockname() in the daemon
6247 code. This can fail if there is a shortage of resources on the system, with
6248 ENOMEM, for example. A temporary error is now given on failure.
6250 29. Contrary to the C standard, it seems that in some environments, the
6251 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
6252 program. Exim now does this explicitly; it affects the formatting of
6253 timestamps using strftime().
6255 30. If exiqsumm was given junk data, it threw up some uninitialized variable
6256 complaints. I've now initialized all the variables, to avoid this.
6258 32. Header lines added by a system filter were not being "seen" during
6259 transport-time rewrites.
6261 33. The info_callback() function passed to OpenSSL is set up with type void
6262 (*)(SSL *, int, int), as described somewhere. However, when calling the
6263 function (actually a macro) that sets it up, the type void(*)() is
6264 expected. I've put in a cast to prevent warnings from picky compilers.
6266 34. If a DNS black list lookup found a CNAME record, but there were no A
6267 records associated with the domain it pointed at, Exim crashed.
6269 35. If a DNS black list lookup returned more than one A record, Exim ignored
6270 all but the first. It now scans all returned addresses if a particular IP
6271 value is being sought. In this situation, the contents of the
6272 $dnslist_value variable are a list of all the addresses, separated by a
6275 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
6276 to accept a host name and all its aliases if the forward lookup for any of
6277 them yielded the IP address of the incoming connection. Now it accepts only
6278 those names whose forward lookup yields the correct IP address. Any other
6279 names are discarded. This closes a loophole whereby a rogue DNS
6280 administrator could create reverse DNS records to break through a
6281 wildcarded host restriction in an ACL.
6283 37. If a user filter or a system filter that ran in a subprocess used any of
6284 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
6285 the wrong values were passed to the pipe command ($thisaddress had the
6286 value of $0, $0 had the value of $1, etc). This bug was introduced by
6287 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
6289 38. Improved the line breaking for long SMTP error messages from ACLs.
6290 Previously, if there was no break point between 40 and 75 characters, Exim
6291 left the rest of the message alone. Two changes have been made: (a) I've
6292 reduced the minimum length to 35 characters; (b) if it can't find a break
6293 point between 35 and 75 characters, it looks ahead and uses the first one
6294 that it finds. This may give the occasional overlong line, but at least the
6295 remaining text gets split now.
6297 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
6298 file descriptors might be low, and that setting 1000 would always raise it.
6299 It turns out that in some environments, the limit is already over 1000 and
6300 that lowering it causes trouble. So now Exim takes care not to decrease it.
6302 40. When delivering a message, the value of $return_path is set to $sender_
6303 address at the start of routing (routers may change the value). By an
6304 oversight, this default was not being set up when an address was tested by
6305 -bt or -bv, which affected the outcome if any router or filter referred to
6308 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
6309 log only when "message" or "log_message" are set. However, if one of the
6310 conditions was an address verification, or a call to a nested ACL, the
6311 messages generated by the underlying test were being passed through. This
6312 no longer happens. The underlying message is available in $acl_verify_
6313 message for both "message" and "log_message" expansions, so it can be
6314 passed through if needed.
6316 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
6317 new expansion $bh_ to give the encoded byte string without charset
6318 translation. Translation happens only if iconv() is available; HAVE_ICONV
6319 indicates this at build time. HEADERS_CHARSET gives the charset to
6320 translate to; headers_charset can change it in the configuration, and
6321 "headers charset" can change it in an individual filter file.
6323 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
6324 that creates RFC 2047 encoded "words" labels them as that charset instead
6325 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
6326 expansion operator; (ii) when Exim creates a From: line for a local
6327 message; (iii) when a header line is rewritten to include a "phrase" part.
6329 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
6330 buggy, causing it to skip the first lines of messages whose message ID
6331 ended in 'D'. This would not have bitten before Exim release 4.14, because
6332 message IDs were unlikely to end in 'D' before then. The effect was to have
6333 incorrect size information for certain domains.
6335 45. #include "config.h" was missing at the start of the crypt16.c module. This
6336 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
6339 46. If there was a timeout during a "random" callout check, Exim treated it as
6340 a failure of the random address, and carried on sending RSET and the real
6341 address. If the delay was just some slowness somewhere, the response to the
6342 original RCPT would be taken as a response to RSET and so on, causing
6343 mayhem of various kinds.
6345 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
6346 when I implemented it. It didn't allow for the fact that some option values
6347 may legitimatetly be negative (e.g. size_addition), and it didn't even do
6348 the right test for positive values.
6350 48. Domain names in DNS records are case-independent. Exim always looks them up
6351 in lower case. Some resolvers return domain names in exactly the case they
6352 appear in the zone file, that is, they may contain uppercase letters. Not
6353 all resolvers do this - some return always lower case. Exim was treating a
6354 change of case by a resolver as a change of domain, similar to a widening
6355 of a domain abbreviation. This triggered its re-routing code and so it was
6356 trying to route what was effectively the same domain again. This normally
6357 caused routing to fail (because the router wouldn't handle the domain
6358 twice). Now Exim checks for this case specially, and just changes the
6359 casing of the domain that it ultimately uses when it transmits the message
6362 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
6365 50. If a filter generated a file delivery with a non-absolute name (possible if
6366 no home directory exists for the router), the forbid_file option was not
6369 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
6370 the existing equality matching.
6372 52. Exim was using ints instead of ino_t variables in some places where it was
6373 dealing with inode numbers.
6375 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
6376 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
6377 called TMPDIR, and if it finds it is different, it changes its value.
6379 54. The smtp_printf() function is now made available to local_scan() so
6380 additional output lines can be written before returning. There is also an
6381 smtp_fflush() function to enable the detection of a dropped connection.
6382 The variables smtp_input and smtp_batched_input are exported to
6385 55. Changed the default runtime configuration: the message "Unknown user"
6386 has been removed from the ACL, and instead placed on the localuser router,
6387 using the cannot_route_message feature. This means that any verification
6388 failures that generate their own messages won't get overridden. Similarly,
6389 the "Unrouteable address" message that was in the ACL for unverifiable
6390 relay addresses has also been removed.
6392 56. Added hosts_avoid_esmtp to the smtp transport.
6394 57. The exicyclog script was not checking for the esoteric option
6395 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
6396 will work only if exicyclog is run under the appropriate euid.
6398 58. Following a discussion on the list, the rules by which Exim recognises line
6399 endings on incoming messages have been changed. The -dropcr and drop_cr
6400 options are now no-ops, retained only for backwards compatibility. The
6401 following line terminators are recognized: LF CRLF CR. However, special
6402 processing applies to CR:
6404 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
6405 nor a local message in the state where . is a terminator.
6407 (ii) If a bare CR is encountered in a header line, an extra space is added
6408 after the line terminator so as not to end the header. The reasoning
6409 behind this is that bare CRs in header lines are most likely either
6410 to be mistakes, or people trying to play silly games.
6412 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
6413 was being incorrectly given as 18 bytes larger than it should have been.
6414 This is a VOB (very old bug).
6416 60. This may never have affected anything current, but just in case it has:
6417 When the local host is found other than at the start of a list of hosts,
6418 the local host, those with the same MX, and any that follow, are discarded.
6419 When the list in question was part of a longer list of hosts, the following
6420 hosts (not currently being processed) were also being discarded. This no
6421 longer happens. I'm not sure if this situation could ever has previously
6424 61. Added the "/MX" feature to lists of hosts in the manualroute and query
6427 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
6428 header. This is something that is recommended in a new Internet Draft, and
6429 is something that is documented as being done by Sendmail. There are two
6430 possible values. For messages generated by the autoreply transport, Exim
6433 Auto-Submitted: auto-replied
6435 whereas for all other generated messages (e.g. bounces) it adds
6437 Auto-Submitted: auto-generated
6439 63. The "personal" condition in filters now includes a test for the
6440 Auto-Submitted: header. If it contains the string "auto-" the message it
6441 not considered personal.
6443 64. Added rcpt_include_affixes as a generic transport option.
6445 65. Added queue_only_override (default true).
6447 66. Added the syslog_duplication option.
6449 67. If what should have been the first header line of a message consisted of
6450 a space followed by a colon, Exim was mis-interpreting it as a header line.
6451 It isn't of course - it is syntactically invalid and should therefore be
6452 treated as the start of the message body. The misbehaviour could have
6453 caused a number of strange effects, including loss of data in subsequent
6454 header lines, and spool format errors.
6456 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
6457 client host had authenticated. This control can now be exercised by an ACL
6458 for more flexibility.
6460 69. By default, callouts do not happen when testing with -bh. There is now a
6461 variant, -bhc, which does actually run the callout code, including
6462 consulting and updating the callout cache.
6464 70. Added support for saslauthd authentication, courtesy of Alexander
6467 71. If statvfs() failed on the spool or log directories while checking their
6468 size for availability, Exim confusingly gave the error "space shortage".
6469 Furthermore, in debugging mode it crashed with a floating point exception.
6470 These checks are done if check_{spool,log}_{space,inodes} are set, and when
6471 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
6472 serious problem, Exim now writes to the main and panic logs when this
6473 happens, with details of the failure. It then refuses to accept the
6474 incoming message, giving the message "spool directory problem" or "log
6475 directory problem" with a 421 code for SMTP messages.
6477 72. When Exim is about to re-exec itself, it ensures that the file descriptors
6478 0, 1, and 2 exist, because some OS complain for execs without them (see
6479 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
6480 descriptors. However, the code omitted to check that the open succeeded,
6481 causing mysterious errors if for some reason the permissions on /dev/null
6482 got screwed. Now Exim writes a message to the main and panic logs, and
6483 bombs out if it can't open /dev/null.
6485 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
6486 interact so that it is all more flexible. It is supposed to remain
6487 backwards compatible. Also added extra_local_interfaces.
6489 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
6490 to bomb out with an assertion failure - to the client this appears as a
6491 connection drop. This problem occurs in the part of the code that was taken
6492 from the Samba project. Fortunately, the assertion is in a very simple
6493 function, so I have fixed this by reproducing the function inline in the
6494 one place where it is called, and arranging for authentication to fail
6495 instead of killing the process with assert().
6497 75. The SPA client code was not working when the server requested OEM rather
6498 than Unicode encoding.
6500 76. Added code to make require_files with a specific uid setting more usable in
6501 the case where statting the file as root fails - usually a non-root-mounted
6502 NFS file system. When this happens and the failure is EACCES, Exim now
6503 forks a subprocess and does the per-uid checking as the relevant uid.
6505 77. Added process_log_path.
6507 78. If log_file_path was not explicitly set, a setting of check_log_space or
6508 check_log_inodes was ignored.
6510 79. If a space check for the spool or log partitions fails, the incident is now
6511 logged. Of course, in the latter case the data may get lost...
6513 80. Added the %p formatting code to string_format() so that it can be used to
6514 print addresses in debug_print(). Adjusted all the address printing in the
6515 debugging in store.c to use %p rather than %d.
6517 81. There was a concern that a line of code in smtp_in.c could overflow a
6518 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
6519 initially expressed, the concern was not well-founded, because trailing
6520 spaces are removed early. However, if the trailing spaces were followed by
6521 a NULL, they did not get removed, so the overflow was possible. Two fixes
6524 (a) I re-wrote the offending code in a cleaner fashion.
6525 (b) If an incoming SMTP command contains a NULL character, it is rejected
6528 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
6529 runs initgroups(), so that if the Exim user is in any additional groups,
6530 they will be used during message reception.
6536 The change log for 4.20 and earlier releases has been archived.
projects / exim.git / blob