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.
32 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
33 and 3.4.4 - once the server is enabled to respond to an OCSP request
34 it does even when not requested, resulting in a stapling non-aware
35 client dropping the TLS connection.
37 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
38 support variable-length bit vectors. No functional change.
40 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
41 The I= interface field on outgoing lines is now after the H= remote
42 host field, same as incoming lines. There is a separate
43 outgoing_interface log selector which allows you to disable the
46 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
47 If not running log_selector +smtp_connection the mainlog would be held
48 open indefinitely after a "too many connections" event, including to a
49 deleted file after a log rotate. Leave the per net connection logging
50 leaving it open for efficiency as that will be quickly detected by the
51 check on the next write.
53 HS/01 Bug 1671: Fix post transport crash.
54 Processing the wait-<transport> messages could crash the delivery
55 process if the message IDs didn't exist for some reason. When
56 using 'split_spool_directory=yes' the construction of the spool
57 file name failed already, exposing the same netto behaviour.
59 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
60 mime_regex ACL conditions.
62 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
63 to DSN fail messages (bounces): remote IP, remote greeting, remote response
64 to HELO, local diagnostic string.
66 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
67 log line to debug. Even when configured with a tls authenticator many
68 client connections are expected to not authenticate in this way, so
69 an authenticate fail is not an error.
71 HS/02 Add the Exim version string to the process info. This way exiwhat
72 gives some more detail about the running daemon.
74 JH/06 Bug 1395: time-limit cacheing of DNS lookups, to the TTL value. This may
75 matter for fast-change records such as DNSBLs.
77 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
78 retry record, even if constant. There may be multiple transports with
79 different interface settings and the retry behaviour needs to be kept
82 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
84 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
86 JH/10 Bug 840: fix log_defer_output option of pipe transport
88 JH/11 Bug 830: use same host for all RCPTS of a message, even under
89 hosts_randomize. This matters a lot when combined with mua_wrapper.
91 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
92 ${quote_pgsql:<string>} operator.
94 JH/13 Bug 1708: avoid misaligned access in cached lookup.
96 JH/14 Change header file name for freeradius-client. Relevant if compiling
97 with Radius support; from the Gentoo tree and checked under Fedora.
99 JH/15 Bug 1712: Introduce $prdr_requested flag variable
101 JH/16 Bug 1714: Permit an empty string as expansion result for transport
102 option transport_filter, meaning no filtering.
104 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
106 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
107 defaults to "*" (all hosts). The variable is now available when not built
108 with TLS, default unset, mainly to enable keeping the testuite sane.
109 If a server certificate is not supplied (via tls_certificate) an error is
110 logged, and clients will find TLS connections fail on startup. Presumably
111 they will retry in-clear.
112 Packagers of Exim are strongly encouraged to create a server certificate
113 at installation time.
115 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
116 with the $config_file variable.
118 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
119 in transport context, after the attempt, and per-recipient. The latter type
120 is per host attempted. The event data is the error message, and the errno
121 information encodes the lookup type (A vs. MX) used for the (first) host,
122 and the trailing two digits of the smtp 4xx reponse.
124 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
125 to write to mainlog (or rejectlog, paniclog) in the window between file
126 creation and permissions/ownership being changed. Particularly affects
127 installations where exicyclog is run as root, rather than exim user;
128 result is that the running daemon panics and dies.
130 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
132 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
133 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
134 "pri" and "weight". Note that the previous implicit priority given by the
135 list order is no longer honoured.
137 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalisation
140 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
141 by defining SUPPORT_SOCKS.
143 JH/26 Move PROXY support from Experimental to mainline, enabled for a build
144 by defining SUPPORT_PROXY. Note that the proxy_required_hosts option
145 is renamed to hosts_proxy, and the proxy_{host,target}_{address,port}.
146 variables are renamed to proxy_{local,external}_{address,port}.
148 JH/27 Move Internationalisation support from Experimental to mainline, enabled
149 for a build by defining SUPPORT_I18N
151 JH/28 Bug 1745: Fix redis lookups to handle (quoted) spaces embedded in parts
152 of the query string, and make ${quote_redis:} do that quoting.
154 JH/29 Move Events support from Experimental to mainline, enabled by default
155 and removable for a build by defining DISABLE_EVENT.
157 JH/30 Updated DANE implementation code to current from Viktor Dukhovni.
159 JH/31 Fix bug with hosts_connection_nolog and named-lists which were wrongly
160 cached by the daemon.
162 JH/32 Move Redis support from Experimental to mainline, enabled for a build
163 by defining LOOKUP_REDIS. The libhiredis library is required.
165 JH/33 Bug 1748: Permit ACL dnslists= condition in non-smtp ACLs if explicit
166 keys are given for lookup.
168 JH/34 Bug 1192: replace the embedded copy of PolarSSL RSA routines in the DKIM
169 support, by using OpenSSL or GnuTLS library ones. This means DKIM is
170 only supported when built with TLS support. The PolarSSL SHA routines
171 are still used when the TLS library is too old for convenient support.
173 JH/35 Require SINGLE_DH_USE by default in OpenSSL (main config option
174 openssl_options), for security. OpenSSL forces this from version 1.1.0
175 server-side so match that on older versions.
177 JH/36 Bug 1778: longstanding bug in memory use by the ${run } expansion: A fresh
178 allocation for $value could be released as the expansion processing
179 concluded, but leaving the global pointer active for it.
181 JH/37 Bug 1769: Permit a VRFY ACL to override the default 252 response,
182 and to use the domains and local_parts ACL conditions.
184 JH/38 Fix cutthrough bug with body lines having a single dot. The dot was
185 incorrectly not doubled on cutthrough transmission, hence seen as a
186 body-termination at the receiving system - resulting in truncated mails.
187 Commonly the sender saw a TCP-level error, and retransmitted the message
188 via the normal store-and-forward channel. This could result in duplicates
189 received - but deduplicating mailstores were liable to retain only the
190 initial truncated version.
192 JH/39 Bug 1781: Fix use of DKIM private-keys having trailing '=' in the base-64.
194 JH/40 Fix crash in queryprogram router when compiled with EXPERIMENTAL_SRS.
196 JH/41 Bug 1792: Fix selection of headers to sign for DKIM: bottom-up. While
197 we're in there, support oversigning also; bug 1309.
199 JH/42 Bug 1796: Fix error logged on a malware scanner connection failure.
201 HS/04 Add support for keep_environment and add_environment options.
203 JH/43 Tidy coding issues detected by gcc --fsanitize=undefined. Some remain;
204 either intentional arithmetic overflow during PRNG, or testing config-
207 JH/44 Bug 1800: The combination of a -bhc commandline option and cutthrough
208 delivery resulted in actual delivery. Cancel cutthrough before DATA
211 JH/45 Fix cutthrough, when connection not opened by verify and target hard-
212 rejects a recipient: pass the reject to the originator.
214 JH/46 Multiple issues raised by Coverity. Some were obvious or plausible bugs.
215 Many were false-positives and ignorable, but it's worth fixing the
218 JH/47 Fix build on HP-UX and older Solaris, which need (un)setenv now also
219 for the new environment-manipulation done at startup. Move the routines
220 from being local to tls.c to being global via the os.c file.
222 JH/48 Bug 1807: Fix ${extract } for the numeric/3-string case. While preparsing
223 an extract embedded as result-arg for a map, the first arg for extract
224 is unavailable so we cannot tell if this is a numbered or keyed
225 extraction. Accept either.
231 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
234 JH/02 The smtp transport option "multi_domain" is now expanded.
236 JH/03 The smtp transport now requests PRDR by default, if the server offers
239 JH/04 Certificate name checking on server certificates, when exim is a client,
240 is now done by default. The transport option tls_verify_cert_hostnames
241 can be used to disable this per-host. The build option
242 EXPERIMENTAL_CERTNAMES is withdrawn.
244 JH/05 The value of the tls_verify_certificates smtp transport and main options
245 default to the word "system" to access the system default CA bundle.
246 For GnuTLS, only version 3.0.20 or later.
248 JH/06 Verification of the server certificate for a TLS connection is now tried
249 (but not required) by default. The verification status is now logged by
250 default, for both outbound TLS and client-certificate supplying inbound
253 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
256 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
257 Status Notification (bounce) messages are now MIME format per RFC 3464.
258 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
259 under the control of the dsn_advertise_hosts option, and routers may
260 have a dsn_lasthop option.
262 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
263 default, modifiable by a malware= option. The list separator for
264 the options can now be changed in the usual way. Bug 68.
266 JH/10 The smtp_receive_timeout main option is now expanded before use.
268 JH/11 The incoming_interface log option now also enables logging of the
269 local interface on delivery outgoing connections.
271 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
272 if the interface and destination host and port all match.
274 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
277 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
278 Patch from Andrew Lewis.
280 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
281 now supports optional time-restrictions, weighting, and priority
282 modifiers per server. Patch originally by <rommer@active.by>.
284 JH/16 The spamd_address main option now supports a mixed list of local
285 and remote servers. Remote servers can be IPv6 addresses, and
286 specify a port-range.
288 JH/17 Bug 68: The spamd_address main option now supports an optional
289 timeout value per server.
291 JH/18 Bug 1581: Router and transport options headers_add/remove can
292 now have the list separator specified.
294 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
297 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
300 JH/21 Support for the A6 type of dns record is withdrawn.
302 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
303 rather than the verbs used.
305 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
306 from 255 to 1024 chars.
308 JH/24 Verification callouts now attempt to use TLS by default.
310 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
311 are generic router options now. The defaults didn't change.
313 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
314 Original patch from Alexander Shikoff, worked over by JH.
316 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
317 files if ZCAT_COMMAND is not executable.
319 JH/26 Bug 1539: Add timout/retry options on dnsdb lookups.
321 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
323 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
324 Normally benign, it bites when the pair was led to by a CNAME;
325 modern usage is to not canoicalize the domain to a CNAME target
326 (and we were inconsistent anyway for A-only vs AAAA+A).
328 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
330 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
331 when evaluating $sender_host_dnssec.
333 JH/31 Check the HELO verification lookup for DNSSEC, adding new
334 $sender_helo_dnssec variable.
336 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
338 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
340 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
342 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
343 documented as working, but never had. Support all but $spam_report.
345 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
346 added for tls authenticator.
348 HS/03 Add perl_taintmode main config option
353 TL/01 When running the test suite, the README says that variables such as
354 no_msglog_check are global and can be placed anywhere in a specific
355 test's script, however it was observed that placement needed to be near
356 the beginning for it to behave that way. Changed the runtest perl
357 script to read through the entire script once to detect and set these
358 variables, reset to the beginning of the script, and then run through
359 the script parsing/test process like normal.
361 TL/02 The BSD's have an arc4random API. One of the functions to induce
362 adding randomness was arc4random_stir(), but it has been removed in
363 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
364 function when detected.
366 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
367 cause callback expansion.
369 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
370 syntax errors in an expansion can be treated as a string instead of
371 logging or causing an error, due to the internal use of bool_lax
372 instead of bool when processing it.
374 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
375 server certificates when making smtp deliveries.
377 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
379 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
381 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
383 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
384 Merged patch from Sebastian Wiedenroth.
386 JH/05 Fix results-pipe from transport process. Several recipients, combined
387 with certificate use, exposed issues where response data items split
388 over buffer boundaries were not parsed properly. This eventually
389 resulted in duplicates being sent. This issue only became common enough
390 to notice due to the introduction of conection certificate information,
391 the item size being so much larger. Found and fixed by Wolfgang Breyha.
393 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
394 size buffer was used, resulting in syntax errors when an expansion
397 JH/07 Add support for directories of certificates when compiled with a GnuTLS
398 version 3.3.6 or later.
400 JH/08 Rename the TPDA expermimental facility to Event Actions. The #ifdef
401 is EXPERIMENTAL_EVENT, the main-configuration and transport options
402 both become "event_action", the variables become $event_name, $event_data
403 and $event_defer_errno. There is a new variable $verify_mode, usable in
404 routers, transports and related events. The tls:cert event is now also
405 raised for inbound connections, if the main configuration event_action
408 TL/06 In test suite, disable OCSP for old versions of openssl which contained
409 early OCSP support, but no stapling (appears to be less than 1.0.0).
411 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
412 server certificate names available under the smtp transport option
413 "tls_verify_cert_hostname" now do not permit multi-component wildcard
416 JH/10 Time-related extraction expansions from certificates now use the main
417 option "timezone" setting for output formatting, and are consistent
418 between OpenSSL and GnuTLS compilations. Bug 1541.
420 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
421 encoded parameter in the incoming message. Bug 1558.
423 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
424 include certificate info, eximon was claiming there were spoolfile
427 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
429 JH/14 Log delivery-related information more consistently, using the sequence
430 "H=<name> [<ip>]" wherever possible.
432 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
433 are problematic for Debian distribution, omit them from the release
436 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
438 JH/16 Fix string representation of time values on 64bit time_t anchitectures.
441 JH/17 Fix a null-indirection in certextract expansions when a nondefault
442 output list separator was used.
447 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
448 checkers that were complaining about end of non-void function with no
451 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
452 This was a regression intruduced in 4.83 by another bugfix.
454 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
456 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
457 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
463 TF/01 Correctly close the server side of TLS when forking for delivery.
465 When a message was received over SMTP with TLS, Exim failed to clear up
466 the incoming connection properly after forking off the child process to
467 deliver the message. In some situations the subsequent outgoing
468 delivery connection happened to have the same fd number as the incoming
469 connection previously had. Exim would try to use TLS and fail, logging
470 a "Bad file descriptor" error.
472 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
473 utilities have not been installed.
475 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
476 temporary space as the ACL may create new global variables.
478 TL/01 LDAP support uses per connection or global context settings, depending
479 upon the detected version of the libraries at build time.
481 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
482 to extract and use the src ip:port in logging and expansions as if it
483 were a direct connection from the outside internet. PPv2 support was
484 updated based on HAProxy spec change in May 2014.
486 JH/02 Add ${listextract {number}{list}{success}{fail}}.
488 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
489 Properly escape header and check for NULL return.
491 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
494 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
496 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
497 characters in header names, implemented as a verify condition.
498 Contributed by Michael Fischer v. Mollard.
500 TL/05 Rename SPF condition results err_perm and err_temp to standardized
501 results permerror and temperror. Previous values are deprecated but
502 still accepted. In a future release, err_perm and err_temp will be
503 completely removed, which will be a backward incompatibility if the
504 ACL tests for either of these two old results. Patch contributed by
505 user bes-internal on the mailing list.
507 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
509 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
510 selectors, in both main and reject logs.
512 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
515 JH/07 Add malware type "sock" for talking to simple daemon.
517 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
519 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
520 routers/transports under cutthrough routing.
522 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
523 numbers. Touch up "bool" conditional to keep the same definition.
525 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
527 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
529 JH/12 Expand items in router/transport headers_add or headers_remove lists
530 individually rather than the list as a whole. Bug 1452.
532 Required for reasonable handling of multiple headers_ options when
533 they may be empty; requires that headers_remove items with embedded
534 colons must have them doubled (or the list-separator changed).
536 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
537 view the policy declared in the DMARC record. Currently, $dmarc_status
538 is a combined value of both the record presence and the result of the
541 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
543 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
544 dnslookup router and the smtp transport (applying to the forward
547 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
548 of ldap servers used for a specific lookup. Patch provided by Heiko
551 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
552 New variable $lookup_dnssec_authenticated for observability.
554 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
555 Patch submitted by Lars Timman.
557 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
559 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
560 Requires trusted mode and valid format message id, aborts otherwise.
561 Patch contributed by Heiko Schlichting.
563 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
564 certextract with support for various fields. Bug 1358.
566 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
567 is requested by default, modifiable by smtp transport option
570 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
571 operate on certificate variables to give certificate fingerprints
572 Also new ${sha256:cert_variable}.
574 JH/23 The PRDR feature is moved from being Experimental into the mainline.
576 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
577 Christian Aistleitner.
579 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
581 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
582 file. Patch from Wolfgang Breyha.
584 JH/25 Expand the coverage of the delivery $host and $host_address to
585 client authenticators run in verify callout. Bug 1476.
587 JH/26 Port service names are now accepted for tls_on_connect_ports, to
588 align with daemon_smtp_ports. Bug 72.
590 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
591 support and error reporting did not work properly.
593 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
594 and is readable. Patch from Andrew Colin Kissa.
596 TL/14 Enhance documentation of ${run expansion and how it parses the
597 commandline after expansion, particularly in the case when an
598 unquoted variable expansion results in an empty value.
600 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
602 PP/02 Fix internal collision of T_APL on systems which support RFC3123
603 by renaming away from it. Addresses GH issue 15, reported by
606 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
608 TL/15 SECURITY: prevent double expansion in math comparison functions
609 (can expand unsanitized data). Not remotely exploitable.
616 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
618 PP/02 Make -n do something, by making it not do something.
619 When combined with -bP, the name of an option is not output.
621 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
624 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
625 $sender_host_name and config options to manage this, and basic check
628 PP/05 DSCP support for outbound connections and control modifier for inbound.
630 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
631 (Only plugin which currently uses this is kerberos4, which nobody should
632 be using, but we should make it available and other future plugins might
633 conceivably use it, even though it would break NAT; stuff *should* be
634 using channel bindings instead).
636 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
637 name; added for Sendmail compatibility; requires admin caller.
638 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
639 just ignore it); requires trusted caller.
640 Also parse but ignore: -Ac -Am -X<logfile>
643 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
645 TL/02 Add +smtp_confirmation as a default logging option.
647 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
648 Patch by Magnus Holmgren from 2007-02-20.
650 TL/04 Bugzilla 1281 - Spec typo.
651 Bugzilla 1283 - Spec typo.
652 Bugzilla 1290 - Spec grammar fixes.
654 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
656 TL/06 Add Experimental DMARC support using libopendmarc libraries.
658 TL/07 Fix an out of order global option causing a segfault. Reported to dev
659 mailing list by by Dmitry Isaikin.
661 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
663 JH/02 Support "G" suffix to numbers in ${if comparisons.
665 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
667 NM/01 Bugzilla 1197 - Spec typo
668 Bugzilla 1196 - Spec examples corrections
670 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
672 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
673 gnutls_enable_pkcs11, but renamed to more accurately indicate its
676 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
677 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
679 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
680 "acl {{name}{arg}...}", and optional args on acl condition
683 JH/05 Permit multiple router/transport headers_add/remove lines.
685 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
687 JH/07 Avoid using a waiting database for a single-message-only transport.
688 Performance patch from Paul Fisher. Bugzilla 1262.
690 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
693 JH/09 Add $headers_added variable, with content from use of ACL modifier
694 add_header (but not yet added to the message). Bugzilla 199.
696 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
697 Pulled from Bugzilla 817 by Wolfgang Breyha.
699 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
701 (nb: this is the same fix as in Exim 4.80.1)
703 JH/11 Add A= logging on delivery lines, and a client_set_id option on
706 JH/12 Add optional authenticated_sender logging to A= and a log_selector
709 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
711 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
712 advertise SMTP AUTH mechanism to us, instead of a generic
713 protocol violation error. Also, make Exim more robust to bad
714 data from the Dovecot auth socket.
716 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
718 When a queue runner is handling a message, Exim first routes the
719 recipient addresses, during which it prunes them based on the retry
720 hints database. After that it attempts to deliver the message to
721 any remaining recipients. It then updates the hints database using
724 So if a recipient address works intermittently, it can get repeatedly
725 deferred at routing time. The retry hints record remains fresh so the
726 address never reaches the final cutoff time.
728 This is a fairly common occurrence when a user is bumping up against
729 their storage quota. Exim had some logic in its local delivery code
730 to deal with this. However it did not apply to per-recipient defers
731 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
733 This change adds a proper retry rule check during routing so that the
734 final cutoff time is checked against the message's age. We only do
735 this check if there is an address retry record and there is not a
736 domain retry record; this implies that previous attempts to handle
737 the address had the retry_use_local_parts option turned on. We use
738 this as an approximation for the destination being like a local
739 delivery, as in LMTP.
741 I suspect this new check makes the old local delivery cutoff check
742 redundant, but I have not verified this so I left the code in place.
744 TF/02 Correct gecos expansion when From: is a prefix of the username.
746 Test 0254 submits a message to Exim with the header
750 When I ran the test suite under the user fanf2, Exim expanded
751 the header to contain my full name, whereas it should have added
752 a Resent-Sender: header. It erroneously treats any prefix of the
753 username as equal to the username.
755 This change corrects that bug.
757 GF/01 DCC debug and logging tidyup
758 Error conditions log to paniclog rather than rejectlog.
759 Debug lines prefixed by "DCC: " to remove any ambiguity.
761 TF/03 Avoid unnecessary rebuilds of lookup-related code.
763 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
764 Bug spotted by Jeremy Harris; was flawed since initial commit.
765 Would have resulted in OCSP responses post-SNI triggering an Exim
766 NULL dereference and crash.
768 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
770 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
771 Bug detection, analysis and fix by Samuel Thibault.
772 Bugzilla 1331, Debian bug #698092.
774 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
776 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
777 Server implementation by Todd Lyons, client by JH.
778 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
779 config variable "prdr_enable" controls whether the server
780 advertises the facility. If the client requests PRDR a new
781 acl_data_smtp_prdr ACL is called once for each recipient, after
782 the body content is received and before the acl_smtp_data ACL.
783 The client is controlled by bolth of: a hosts_try_prdr option
784 on the smtp transport, and the server advertisement.
785 Default client logging of deliveries and rejections involving
786 PRDR are flagged with the string "PRDR".
788 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
789 fclose(). Diagnosis by Todd Lyons.
791 PP/17 Update configure.default to handle IPv6 localhost better.
792 Patch by Alain Williams (plus minor tweaks).
795 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
796 This is now consistent with GnuTLS, and is now documented: the
797 previous undocumented portable approach to treating the option as
798 unset was to force an expansion failure. That still works, and
799 an empty string is now equivalent.
801 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
802 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
803 not performing validation itself.
805 PP/20 Added force_command boolean option to pipe transport.
806 Patch from Nick Koston, of cPanel Inc.
808 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
811 TF/04 Added udpsend ACL modifer and hexquote expansion operator
813 PP/21 Fix eximon continuous updating with timestamped log-files.
814 Broken in a format-string cleanup in 4.80, missed when I repaired the
815 other false fix of the same issue.
816 Report and fix from Heiko Schlichting.
819 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
820 Report from Prashanth Katuri.
822 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
823 It's SecureTransport, so affects any MacOS clients which use the
824 system-integrated TLS libraries, including email clients.
826 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
827 using a MIME ACL for non-SMTP local injection.
828 Report and assistance in diagnosis by Warren Baker.
830 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
832 JH/16 Fix comparisons for 64b. Bugzilla 1385.
834 TL/09 Add expansion variable $authenticated_fail_id to keep track of
835 last id that failed so it may be referenced in subsequent ACL's.
837 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
840 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
841 ldap library initialization, allowing self-signed CA's to be
842 used. Also properly sets require_cert option later in code by
843 using NULL (global ldap config) instead of ldap handle (per
844 session). Bug diagnosis and testing by alxgomz.
846 TL/12 Enhanced documentation in the ratelimit.pl script provided in
847 the src/util/ subdirectory.
849 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
850 renamed to Transport Post Delivery Action by Jeremy Harris, as
853 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
854 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
855 redis_servers = needs to be configured which will be used by the redis
856 lookup. Patch from Warren Baker, of The Packet Hub.
858 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
860 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
861 hostname or reverse DNS when processing a host list. Used suggestions
862 from multiple comments on this bug.
864 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
866 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
867 Missed a few lines, added it to make the runtest require no keyboard
870 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
871 contains upper case chars. Make router use caseful_local_part.
873 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
874 support when GnuTLS has been built with p11-kit.
880 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
882 This, or similar/improved, will also be change PP/11 of 4.82.
888 PP/01 Handle short writes when writing local log-files.
889 In practice, only affects FreeBSD (8 onwards).
890 Bugzilla 1053, with thanks to Dmitry Isaikin.
892 NM/01 Bugzilla 949 - Documentation tweak
894 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
897 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
899 PP/02 Implemented gsasl authenticator.
901 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
903 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
904 `pkg-config foo` for cflags/libs.
906 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
907 with rest of GSASL and with heimdal_gssapi.
909 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
910 `pkg-config foo` for cflags/libs for the TLS implementation.
912 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
913 properties get this fed in as external SSF. A number of robustness
914 and debugging improvements to the cyrus_sasl authenticator.
916 PP/08 cyrus_sasl server now expands the server_realm option.
918 PP/09 Bugzilla 1214 - Log authentication information in reject log.
919 Patch by Jeremy Harris.
921 PP/10 Added dbmjz lookup type.
923 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
925 PP/12 MAIL args handles TAB as well as SP, for better interop with
926 non-compliant senders.
927 Analysis and variant patch by Todd Lyons.
929 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
930 Bug report from Lars Müller <lars@samba.org> (via SUSE),
931 Patch from Dirk Mueller <dmueller@suse.com>
933 PP/13 tls_peerdn now print-escaped for spool files.
934 Observed some $tls_peerdn in wild which contained \n, which resulted
935 in spool file corruption.
937 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
938 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
939 or write after TLS renegotiation, which otherwise led to messages
942 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
943 as a tracking header (ie: a signed header comes before the signature).
944 Patch from Wolfgang Breyha.
946 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
947 comma-sep list; embedded commas doubled.
949 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
951 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
953 Report and patch from Dmitry Banschikov.
955 PP/16 Removed "dont_insert_empty_fragments" fron "openssl_options".
956 Removed SSL_clear() after SSL_new() which led to protocol negotiation
957 failures. We appear to now support TLS1.1+ with Exim.
959 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
960 lets Exim select keys and certificates based upon TLS SNI from client.
961 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
962 before an outbound SMTP session. New log_selector, +tls_sni.
964 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
965 NULL dereference. Report and patch from Alun Jones.
967 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
968 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
969 Not seeing resolver debug output on NetBSD, but suspect this is a
970 resolver implementation change.
972 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
973 Left warnings. Added "eximon gdb" invocation mode.
975 PP/21 Defaulting "accept_8bitmime" to true, not false.
977 PP/22 Added -bw for inetd wait mode support.
979 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
980 locate the relevant includes and libraries. Made this the default.
982 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
983 Bugzilla 1246, report and most of solution from Tomasz Kusy.
985 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
986 This may cause build issues on older platforms.
988 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
989 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
990 gnutls_require_mac & gnutls_require_protocols (no longer supported).
991 Added SNI support via GnuTLS too.
992 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
994 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
996 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
997 Applied second patch from Janne, implementing suggestion to default
998 multiple-strings-in-record handling to match SPF spec.
1000 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
1002 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
1003 read-only, out of scope).
1004 Patch from Wolfgang Breyha, report from Stuart Northfield.
1006 PP/29 Fix three issues highlighted by clang analyser static analysis.
1007 Only crash-plausible issue would require the Cambridge-specific
1008 iplookup router and a misconfiguration.
1009 Report from Marcin Mirosław.
1011 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
1013 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
1014 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
1015 As part of this, removing so much warning spew let me fix some minor
1016 real issues in debug logging.
1018 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
1019 assignment on my part. Fixed.
1021 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
1022 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
1023 Janne Snabb (who went above and beyond: thank you).
1025 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
1026 string otherwise requires a connection and a bunch more work and it's
1027 relatively easy to get wrong. Should also expose TLS library linkage
1030 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
1031 64-bit ${eval} (JH/03).
1033 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
1034 GNU libc to support some of the 64-bit stuff, should not lead to
1035 conflicts. Defined before os.h is pulled in, so if a given platform
1036 needs to override this, it can.
1038 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
1039 protection layer was required, which is not implemented.
1040 Bugzilla 1254, patch from Wolfgang Breyha.
1042 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
1043 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
1044 tls_dhparam take prime identifiers. Also unbreak combination of
1045 OpenSSL+DH_params+TLSSNI.
1047 PP/39 Disable SSLv2 by default in OpenSSL support.
1053 PP/01 Solaris build fix for Oracle's LDAP libraries.
1054 Bugzilla 1109, patch from Stephen Usher.
1056 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
1058 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
1061 TF/02 Fix a couple more cases where we did not log the error message
1062 when unlink() failed. See also change 4.74-TF/03.
1064 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
1065 lock up or crash if it happened to be inside a call to libc when it
1066 got a SIGUSR1 from exiwhat.
1068 The SIGUSR1 handler appends the current process status to the process
1069 log which is later printed by exiwhat. It used to use the general
1070 purpose logging code to do this, but several functions it calls are
1071 not safe for signals.
1073 The new output code in the SIGUSR1 handler is specific to the process
1074 log, and simple enough that it's easy to inspect for signal safety.
1075 Removing some special cases also simplifies the general logging code.
1076 Removing the spurious timestamps from the process log simplifies
1079 TF/04 Improved ratelimit ACL condition.
1081 The /noupdate option has been deprecated in favour of /readonly which
1082 has clearer semantics. The /leaky, /strict, and /readonly update modes
1083 are mutually exclusive. The update mode is no longer included in the
1084 database key; it just determines when the database is updated. (This
1085 means that when you upgrde Exim will forget old rate measurements.)
1087 Exim now checks that the per_* options are used with an update mode that
1088 makes sense for the current ACL. For example, when Exim is processing a
1089 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
1090 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
1091 must specify per_mail/readonly. If you omit the update mode it defaults to
1092 /leaky where that makes sense (as before) or /readonly where required.
1094 The /noupdate option is now undocumented but still supported for
1095 backwards compatibility. It is equivalent to /readonly except that in
1096 ACLs where /readonly is required you may specify /leaky/noupdate or
1097 /strict/noupdate which are treated the same as /readonly.
1099 A useful new feature is the /count= option. This is a generalization
1100 of the per_byte option, so that you can measure the throughput of other
1101 aggregate values. For example, the per_byte option is now equivalent
1102 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
1104 The per_rcpt option has been generalized using the /count= mechanism
1105 (though it's more complicated than the per_byte equivalence). When it is
1106 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
1107 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
1108 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
1109 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
1110 Note that using per_rcpt with a non-readonly update mode in more than
1111 one ACL will cause the recipients to be double-counted. (The per_mail
1112 and per_byte options don't have this problem.)
1114 The handling of very low rates has changed slightly. If the computed rate
1115 is less than the event's count (usually one) then this event is the first
1116 after a long gap. In this case the rate is set to the same as this event's
1117 count, so that the first message of a spam run is counted properly.
1119 The major new feature is a mechanism for counting the rate of unique
1120 events. The new per_addr option counts the number of different
1121 recipients that someone has sent messages to in the last time period. It
1122 behaves like per_rcpt if all the recipient addresses are different, but
1123 duplicate recipient addresses do not increase the measured rate. Like
1124 the /count= option this is a general mechanism, so the per_addr option
1125 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
1126 example, measure the rate that a client uses different sender addresses
1127 with the options per_mail/unique=$sender_address. There are further
1128 details in the main documentation.
1130 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
1132 TF/06 Removed a few PCRE remnants.
1134 TF/07 Automatically extract Exim's version number from tags in the git
1135 repository when doing development or release builds.
1137 PP/02 Raise smtp_cmd_buffer_size to 16kB.
1138 Bugzilla 879. Patch from Paul Fisher.
1140 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
1141 Heavily based on revision 40f9a89a from Simon Arlott's tree.
1144 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
1146 PP/05 Variable $av_failed, true if the AV scanner deferred.
1147 Bugzilla 1078. Patch from John Horne.
1149 PP/06 Stop make process more reliably on build failure.
1150 Bugzilla 1087. Patch from Heiko Schlittermann.
1152 PP/07 Make maildir_use_size_file an _expandable_ boolean.
1153 Bugzilla 1089. Patch from Heiko Schlittermann.
1155 PP/08 Handle ${run} returning more data than OS pipe buffer size.
1156 Bugzilla 1131. Patch from Holger Weiß.
1158 PP/09 Handle IPv6 addresses with SPF.
1159 Bugzilla 860. Patch from Wolfgang Breyha.
1161 PP/10 GnuTLS: support TLS 1.2 & 1.1.
1163 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
1166 PP/11 match_* no longer expand right-hand-side by default.
1167 New compile-time build option, EXPAND_LISTMATCH_RHS.
1168 New expansion conditions, "inlist", "inlisti".
1170 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
1172 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
1174 PP/14 fix log_write() format string regression from TF/03.
1175 Bugzilla 1152. Patch from Dmitry Isaikin.
1181 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
1183 PP/02 Harmonised TLS library version reporting; only show if debugging.
1184 Layout now matches that introduced for other libraries in 4.74 PP/03.
1186 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
1188 PP/04 New "dns_use_edns0" global option.
1190 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
1193 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
1194 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
1196 TK/01 Updated PolarSSL code to 0.14.2.
1197 Bugzilla 1097. Patch from Andreas Metzler.
1199 PP/07 Catch divide-by-zero in ${eval:...}.
1200 Fixes bugzilla 1102.
1202 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
1205 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
1206 format-string attack -- SECURITY: remote arbitrary code execution.
1208 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
1209 time unintentionally subject to list matching rules, letting the header
1210 cause arbitrary Exim lookups (of items which can occur in lists, *not*
1211 arbitrary string expansion). This allowed for information disclosure.
1213 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
1214 INT_MIN/-1 -- value coerced to INT_MAX.
1220 NM/01 Workround for PCRE version dependency in version reporting
1223 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
1224 This fixes portability to compilers other than gcc, notably
1225 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
1227 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
1228 makefiles for portability to HP-UX and POSIX correctness.
1230 PP/01 Permit LOOKUP_foo enabling on the make command-line.
1231 Also via indented variable definition in the Makefile.
1232 (Debugging by Oliver Heesakkers).
1234 PP/02 Restore caching of spamd results with expanded spamd_address.
1235 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
1237 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
1238 Improves build reliability. Fix from: Frank Elsner
1240 NM/02 Fix wide character breakage in the rfc2047 coding
1241 Fixes bug 1064. Patch from Andrey N. Oktyabrski
1243 NM/03 Allow underscore in dnslist lookups
1244 Fixes bug 1026. Patch from Graeme Fowler
1246 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
1247 Code patches from Adam Ciarcinski of NetBSD.
1249 NM/04 Fixed exiqgrep to cope with mailq missing size issue
1252 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
1253 is logged, to avoid truncation. Patch from John Horne.
1255 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
1256 Patch from Jakob Hirsch.
1258 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
1259 SQL string expansion failure details.
1260 Patch from Andrey Oktyabrski.
1262 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
1263 Patch from Simon Arlott.
1265 PP/09 New lookups functionality failed to compile on old gcc which rejects
1266 extern declarations in function scope.
1267 Patch from Oliver Fleischmann
1269 PP/10 Use sig_atomic_t for flags set from signal handlers.
1270 Check getgroups() return and improve debugging.
1271 Fixed developed for diagnosis in bug 927 (which turned out to be
1274 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
1275 Patch from Mark Zealey.
1277 PP/12 Bugzilla 1056: Improved spamd server selection.
1278 Patch from Mark Zealey.
1280 PP/13 Bugzilla 1086: Deal with maildir quota file races.
1281 Based on patch from Heiko Schlittermann.
1283 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
1284 Patch from Uwe Doering, sign-off by Michael Haardt.
1286 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
1287 variable declaration deep within a block. Bug and patch from
1290 PP/15 lookups-Makefile IRIX compatibilty coercion.
1292 PP/16 Make DISABLE_DKIM build knob functional.
1294 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
1295 Patch by Simon Arlott
1297 TF/03 Fix valgrind.h portability to C89 compilers that do not support
1298 variable argument macros. Our copy now differs from upstream.
1304 TF/01 Failure to get a lock on a hints database can have serious
1305 consequences so log it to the panic log.
1307 TF/02 Log LMTP confirmation messages in the same way as SMTP,
1308 controlled using the smtp_confirmation log selector.
1310 TF/03 Include the error message when we fail to unlink a spool file.
1312 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
1313 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
1314 for maintaining out-of-tree patches for some time.
1316 PP/01 Bugzilla 139: Documentation and portability issues.
1317 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
1318 Handle per-OS dynamic-module compilation flags.
1320 PP/02 Let /dev/null have normal permissions.
1321 The 4.73 fixes were a little too stringent and complained about the
1322 permissions on /dev/null. Exempt it from some checks.
1323 Reported by Andreas M. Kirchwitz.
1325 PP/03 Report version information for many libraries, including
1326 Exim version information for dynamically loaded libraries. Created
1327 version.h, now support a version extension string for distributors
1328 who patch heavily. Dynamic module ABI change.
1330 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
1331 privilege escalation vulnerability whereby the Exim run-time user
1332 can cause root to append content of the attacker's choosing to
1335 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
1338 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
1339 If dropping privileges for untrusted macros, we disabled normal logging
1340 on the basis that it would fail; for the Exim run-time user, this is not
1341 the case, and it resulted in successful deliveries going unlogged.
1342 Fixed. Reported by Andreas Metzler.
1348 PP/01 Date: & Message-Id: revert to normally being appended to a message,
1349 only prepend for the Resent-* case. Fixes regression introduced in
1350 Exim 4.70 by NM/22 for Bugzilla 607.
1352 PP/02 Include check_rfc2047_length in configure.default because we're seeing
1353 increasing numbers of administrators be bitten by this.
1355 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
1357 PP/03 Bugzilla 994: added openssl_options main configuration option.
1359 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
1361 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
1363 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
1365 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
1366 without a peer certificate, leading to a segfault because of an
1367 assumption that peers always have certificates. Be a little more
1368 paranoid. Problem reported by Martin Tscholak.
1370 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
1371 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
1372 NB: ClamAV planning to remove STREAM in "middle of 2010".
1373 CL also introduces -bmalware, various -d+acl logging additions and
1374 more caution in buffer sizes.
1376 PP/09 Implemented reverse_ip expansion operator.
1378 PP/10 Bugzilla 937: provide a "debug" ACL control.
1380 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
1382 PP/12 Bugzilla 973: Implement --version.
1384 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
1386 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
1388 PP/15 Bugzilla 816: support multiple condition rules on Routers.
1390 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
1391 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
1392 ignore trailing whitespace.
1394 JJ/02 prevent non-panic DKIM error from being sent to paniclog
1396 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
1399 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
1400 Notification from Dr Andrew Aitchison.
1402 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
1403 ExtendedDetectionInfo response format.
1404 Notification from John Horne.
1406 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
1409 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
1410 XSL and documented dependency on system catalogs, with examples of how
1413 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
1416 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
1417 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
1418 configuration file which is writeable by the Exim user or group.
1420 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
1421 of configuration files to cover files specified with the -C option if
1422 they are going to be used with root privileges, not just the default
1425 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
1426 option (effectively making it always true).
1428 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
1429 files to be used while preserving root privileges.
1431 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
1432 that rogue child processes cannot use them.
1434 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
1435 run-time user, instead of root.
1437 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
1438 Exim run-time user without dropping privileges.
1440 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
1441 result string, instead of calling string_vformat() twice with the same
1444 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
1445 for other users. Others should always drop root privileges if they use
1446 -C on the command line, even for a whitelisted configure file.
1448 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
1450 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
1456 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
1457 $data_path, and $header_path variables; fixed documentation bugs and
1460 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
1461 exipick to access non-standard spools, including the "frozen" queue
1464 NM/01 Bugzilla 965: Support mysql stored procedures.
1465 Patch from Alain Williams
1467 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
1469 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
1470 Patch from Andreas Metzler
1472 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
1473 Patch from Kirill Miazine
1475 NM/05 Bugzilla 671: Added umask to procmail example.
1477 JJ/03 installed exipick 20100323.0, fixing doc bug
1479 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
1480 directory. Notification and patch from Dan Rosenberg.
1482 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
1484 TK/02 Improve log output when DKIM signing operation fails.
1486 MH/01 Treat the transport option dkim_domain as a colon separated
1487 list, not as a single string, and sign the message with each element,
1488 omitting multiple occurences of the same signer.
1490 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
1491 Bugzilla 985, 986. Patch by Simon Arlott
1493 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
1494 Patch by Simon Arlott
1496 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
1497 MBX locking. Notification from Dan Rosenberg.
1503 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
1505 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
1507 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
1509 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
1511 NM/04 Bugzilla 915: Flag broken perl installation during build.
1517 TK/01 Added patch by Johannes Berg that expands the main option
1518 "spamd_address" if it starts with a dollar sign.
1520 TK/02 Write list of recipients to X-Envelope-Sender header when building
1521 the mbox-format spool file for content scanning (suggested by Jakob
1524 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
1525 (http://www.dcc-servers.net/) support via dccifd. Activated by
1526 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
1528 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
1529 by Mark Daniel Reidel <mr@df.eu>.
1531 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
1532 When building exim an external PCRE library is now needed -
1533 PCRE is a system library on the majority of modern systems.
1534 See entry on PCRE_LIBS in EDITME file.
1536 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
1537 conversation. Added nologin parameter to request.
1538 Patch contributed by Kirill Miazine.
1540 TF/01 Do not log submission mode rewrites if they do not change the address.
1542 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
1544 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
1545 log files in place. Contributed by Roberto Lima.
1547 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
1549 TF/03 Bugzilla 615: When checking the local_parts router precondition
1550 after a local_part_suffix or local_part_prefix option, Exim now
1551 does not use the address's named list lookup cache, since this
1552 contains cached lookups for the whole local part.
1554 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
1555 Robert Millan. Documentation is in experimental-spec.txt.
1557 TF/04 Bugzilla 668: Fix parallel build (make -j).
1559 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
1561 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
1562 Patch provided by Jan Srzednicki.
1564 TF/05 Leading white space used to be stripped from $spam_report which
1565 wrecked the formatting. Now it is preserved.
1567 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
1568 that they are available at delivery time.
1570 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
1572 TF/08 TLS error reporting now respects the incoming_interface and
1573 incoming_port log selectors.
1575 TF/09 Produce a more useful error message if an SMTP transport's hosts
1576 setting expands to an empty string.
1578 NM/06 Bugzilla 744: EXPN did not work under TLS.
1579 Patch provided by Phil Pennock.
1581 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
1582 Patch provided by Richard Godbee.
1584 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
1585 acl_smtp_notquit, added index entry.
1587 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
1588 Patch provided by Eugene Bujak.
1590 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
1591 accept(). Patch provided by Maxim Dounin.
1593 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
1594 Patch provided by Phil Pennock.
1596 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
1598 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
1599 Patch provided by Brad "anomie" Jorsch.
1601 NM/14 Bugzilla 622: Added timeout setting to transport filter.
1602 Patch provided by Dean Brooks.
1604 TK/05 Add native DKIM support (does not depend on external libraries).
1606 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
1607 Patch provided by Graeme Fowler.
1609 NM/16 Bugzilla 851: Documentation example syntax fix.
1611 NM/17 Changed NOTICE file to remove references to embedded PCRE.
1613 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
1616 NM/19 Bugzilla 745: TLS version reporting.
1617 Patch provided by Phil Pennock.
1619 NM/20 Bugzilla 167: bool: condition support.
1620 Patch provided by Phil Pennock.
1622 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
1623 clients. Patch provided by Phil Pennock.
1625 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
1626 Patch provided by Brad "anomie" Jorsch.
1628 NM/23 Bugzilla 687: Fix misparses in eximstats.
1629 Patch provided by Heiko Schlittermann.
1631 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
1632 Patch provided by Heiko Schlittermann.
1634 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
1635 plus update to original patch.
1637 NM/26 Bugzilla 799: Documentation correction for ratelimit.
1639 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
1640 Patch provided by David Brownlee.
1642 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
1644 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
1646 NM/30 Bugzilla 888: TLS documentation bugfixes.
1648 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
1650 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
1651 Unlike the original bugzilla I have changed all shell scripts in src tree.
1653 NM/33 Bugzilla 898: Transport filter timeout fix.
1654 Patch by Todd Rinaldo.
1656 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
1657 Patch by Serge Demonchaux.
1659 NM/35 Bugzilla 39: Base64 decode bug fixes.
1660 Patch by Jakob Hirsch.
1662 NM/36 Bugzilla 909: Correct connect() call in dcc code.
1664 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
1666 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
1668 NM/39 Bugzilla 911: Fixed MakeLinks build script.
1674 TK/01 Add preliminary DKIM support. Currently requires a forked version of
1675 ALT-N's libdkim that I have put here:
1676 http://duncanthrax.net/exim-experimental/
1678 Note to Michael Haardt: I had to rename some vars in sieve.c. They
1679 were called 'true' and it seems that C99 defines that as a reserved
1680 keyword to be used with 'bool' variable types. That means you could
1681 not include C99-style headers which use bools without triggering
1682 build errors in sieve.c.
1684 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
1685 as mailq or other aliases. Changed the --help handling significantly
1686 to do whats expected. exim_usage() emits usage/help information.
1688 SC/01 Added the -bylocaldomain option to eximstats.
1690 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
1692 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
1694 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
1700 PH/01 Another patch from the Sieve maintainer.
1702 PH/02 When an IPv6 address is converted to a string for single-key lookup
1703 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
1704 dots are used instead of colons so that keys in lsearch files need not
1705 contain colons. This was done some time before quoting was made available
1706 in lsearch files. However, iplsearch files do require colons in IPv6 keys
1707 (notated using the quote facility) so as to distinguish them from IPv4
1708 keys. This meant that lookups for IP addresses in host lists did not work
1709 for iplsearch lookups.
1711 This has been fixed by arranging for IPv6 addresses to be expressed with
1712 colons if the lookup type is iplsearch. This is not incompatible, because
1713 previously such lookups could never work.
1715 The situation is now rather anomolous, since one *can* have colons in
1716 ordinary lsearch keys. However, making the change in all cases is
1717 incompatible and would probably break a number of configurations.
1719 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
1722 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
1723 conversion specification without a maximum field width, thereby enabling
1724 a rogue spamd server to cause a buffer overflow. While nobody in their
1725 right mind would setup Exim to query an untrusted spamd server, an
1726 attacker that gains access to a server running spamd could potentially
1727 exploit this vulnerability to run arbitrary code as the Exim user.
1729 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
1730 $primary_hostname instead of what libspf2 thinks the hosts name is.
1732 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
1733 a directory entry by the name of the lookup key. Previously, if a
1734 symlink pointed to a non-existing file or a file in a directory that
1735 Exim lacked permissions to read, a lookup for a key matching that
1736 symlink would fail. Now it is enough that a matching directory entry
1737 exists, symlink or not. (Bugzilla 503.)
1739 PH/03 The body_linecount and body_zerocount variables are now exported in the
1742 PH/04 Added the $dnslist_matched variable.
1744 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
1745 This means they are set thereafter only if the connection becomes
1748 PH/06 Added the client_condition to authenticators so that some can be skipped
1749 by clients under certain conditions.
1751 PH/07 The error message for a badly-placed control=no_multiline_responses left
1752 "_responses" off the end of the name.
1754 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
1756 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
1757 (without spaces) instead of just copying the configuration text.
1759 PH/10 Added the /noupdate option to the ratelimit ACL condition.
1761 PH/11 Added $max_received_linelength.
1763 PH/12 Added +ignore_defer and +include_defer to host lists.
1765 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
1766 way in which PCRE > 7.0 is built.
1768 PH/14 Implemented queue_only_load_latch.
1770 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
1771 MAIL command. The effect was to mangle the value on 64-bit systems.
1773 PH/16 Another patch from the Sieve maintainer.
1775 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
1777 PH/18 If a system quota error occurred while trying to create the file for
1778 a maildir delivery, the message "Mailbox is full" was not appended to the
1779 bounce if the delivery eventually timed out. Change 4.67/27 below applied
1780 only to a quota excession during the actual writing of the file.
1782 PH/19 It seems that peer DN values may contain newlines (and other non-printing
1783 characters?) which causes problems in log lines. The DN values are now
1784 passed through string_printing() before being added to log lines.
1786 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
1787 and InterBase are left for another time.)
1789 PH/21 Added message_body_newlines option.
1791 PH/22 Guard against possible overflow in moan_check_errorcopy().
1793 PH/23 POSIX allows open() to be a macro; guard against that.
1795 PH/24 If the recipient of an error message contained an @ in the local part
1796 (suitably quoted, of course), incorrect values were put in $domain and
1797 $local_part during the evaluation of errors_copy.
1803 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
1804 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
1807 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
1808 issue a MAIL command.
1810 PH/02 In an ACL statement such as
1812 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
1814 if a client was not listed at all, or was listed with a value other than
1815 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
1816 the condition was not true (as it should be), so access was not denied.
1817 The bug was that the ! inversion was incorrectly passed on to the second
1818 item. This has been fixed.
1820 PH/03 Added additional dnslists conditions == and =& which are different from
1821 = and & when the dns lookup returns more than one IP address.
1823 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
1824 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
1826 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
1827 FSYNC, which compiles an option called disable_fsync that allows for
1828 bypassing fsync(). The documentation is heavily laced with warnings.
1830 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
1832 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
1833 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
1834 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
1835 including adding "make clean"; (3) Added -fPIC when compiling the test
1836 dynamically loaded module, to get rid of a warning.
1838 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
1839 message fails, move_frozen_messages = true and ignore_bounce_errors_after
1840 = 0s. The bug is otherwise harmless.
1842 PH/07 There was a bug in the dovecot authenticator such that the value of
1843 $auth1 could be overwritten, and so not correctly preserved, after a
1844 successful authentication. This usually meant that the value preserved by
1845 the server_setid option was incorrect.
1847 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
1849 PH/09 Installed PCRE release 7.0.
1851 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
1852 run for batched SMTP input. It is now run at the start of every message
1853 in the batch. While fixing this I discovered that the process information
1854 (output by running exiwhat) was not always getting set for -bs and -bS
1855 input. This is fixed, and it now also says "batched" for BSMTP.
1857 PH/11 Added control=no_pipelining.
1859 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
1860 patch, slightly modified), and move the expansion of helo_data till after
1861 the connection is made in the smtp transport (so it can use these
1864 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
1866 PH/14 Added log_selector = +pid.
1868 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
1870 PH/16 Add ${if forany and ${if forall.
1872 PH/17 Added dsn_from option to vary the From: line in DSNs.
1874 PH/18 Flush SMTP output before performing a callout, unless control =
1875 no_callout_flush is set.
1877 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
1878 was true (the default) a successful delivery failed to delete the retry
1879 item, thus causing premature timeout of the address. The bug is now
1882 PH/20 Added hosts_avoid_pipelining to the smtp transport.
1884 PH/21 Long custom messages for fakedefer and fakereject are now split up
1885 into multiline reponses in the same way that messages for "deny" and
1886 other ACL rejections are.
1888 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
1889 with slight modification.
1891 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
1892 draft, changing the syntax and factoring some duplicate code".
1894 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
1895 for deliveries of the second and subsequent messages over the same SMTP
1898 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
1899 ${reduce, with only minor "tidies".
1901 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
1903 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
1904 expansion side effects.
1906 PH/27 When a message times out after an over-quota error from an Exim-imposed
1907 quota, the bounce message says "mailbox is full". This message was not
1908 being given when it was a system quota that was exceeded. It now should
1911 MH/03 Made $recipients available in local_scan(). local_scan() already has
1912 better access to the recipient list through recipients_list[], but
1913 $recipients can be useful in postmaster-provided expansion strings.
1915 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
1916 in the case of a MAIL command with additional options following the
1917 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
1918 were accidentally chopped off.
1920 PH/29 SMTP synchronization checks are implemented when a command is read -
1921 there is a check that no more input is waiting when there shouldn't be
1922 any. However, for some commands, a delay in an ACL can mean that it is
1923 some time before the response is written. In this time, more input might
1924 arrive, invalidly. So now there are extra checks after an ACL has run for
1925 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
1926 pipelining has not been advertised.
1928 PH/30 MH's patch to allow iscntrl() characters to be list separators.
1930 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
1931 returned in the SMTP response when smtp_return_error_details was false.
1932 This has been fixed.
1934 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
1935 instead of the C I/O that was originally supplied, because problems were
1936 reported on Solaris.
1938 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
1939 Exim which did not show up earlier: it was assuming that a call to
1940 SSL_CTX_set_info_callback() might give an error value. In fact, there is
1941 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
1942 was a macro that became an assignment, so it seemed to work. This has
1943 changed to a proper function call with a void return, hence the compile
1944 error. Exim's code has been fixed.
1946 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
1949 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
1951 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
1953 PH/37 If a message is not accepted after it has had an id assigned (e.g.
1954 because it turns out to be too big or there is a timeout) there is no
1955 "Completed" line in the log. When some messages of this type were
1956 selected by exigrep, they were listed as "not completed". Others were
1957 picked up by some special patterns. I have improved the selection
1958 criteria to be more general.
1960 PH/38 The host_find_failed option in the manualroute router can now be set
1961 to "ignore", to completely ignore a host whose IP address cannot be
1962 found. If all hosts are ignored, the behaviour is controlled by the new
1963 host_all_ignored option.
1965 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
1966 homing or because of multiple MX records with /mx) generated more than
1967 one IP address, and the following item turned out to be the local host,
1968 all the secondary addresses of the first item were incorrectly removed
1969 from the list, along with the local host and any following hosts (which
1970 is what is supposed to happen).
1972 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
1973 whoever called Exim into the -H file. In the case of the daemon it was
1974 behaving confusingly. When first started, it used values for whoever
1975 started the daemon, but after a SIGHUP it used the Exim user (because it
1976 calls itself on a restart). I have changed the code so that it now always
1979 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
1980 message are rejected with the same error (e.g. no authentication or bad
1981 sender address), and a DATA command is nevertheless sent (as can happen
1982 with PIPELINING or a stupid MUA), the error message that was given to the
1983 RCPT commands is included in the rejection of the DATA command. This is
1984 intended to be helpful for MUAs that show only the final error to their
1987 PH/42 Another patch from the Sieve maintainer.
1989 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
1990 Eximstats - Fixed some broken HTML links and added missing column headers
1992 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
1993 columns for Rejects, Temp Rejects, Ham, and Spam rows.
1995 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
1997 PH/43 Yet another patch from the Sieve maintainer.
1999 PH/44 I found a way to check for a TCP/IP connection going away before sending
2000 the response to the final '.' that terminates a message, but only in the
2001 case where the client has not sent further data following the '.'
2002 (unfortunately, this is allowed). However, in many cases there won't be
2003 any further data because there won't be any more messages to send. A call
2004 to select() can be used: if it shows that the input is "ready", there is
2005 either input waiting, or the socket has been closed. An attempt to read
2006 the next input character can distinguish the two cases. Previously, Exim
2007 would have sent an OK response which the client would never have see.
2008 This could lead to message repetition. This fix should cure that, at
2009 least in a lot of common cases.
2011 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
2012 advertised in response to EHLO.
2018 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
2019 fixed by 4.65/MH/01 (is this a record?) are fixed:
2021 (i) An empty string was always treated as zero by the numeric comparison
2022 operators. This behaviour has been restored.
2024 (ii) It is documented that the numeric comparison operators always treat
2025 their arguments as decimal numbers. This was broken in that numbers
2026 starting with 0 were being interpreted as octal.
2028 While fixing these problems I realized that there was another issue that
2029 hadn't been noticed. Values of message_size_limit (both the global option
2030 and the transport option) were treated as octal if they started with 0.
2031 The documentation was vague. These values are now always treated as
2032 decimal, and I will make that clear in the documentation.
2038 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
2039 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
2042 MH/01 Don't check that the operands of numeric comparison operators are
2043 integers when their expansion is in "skipping" mode (fixes bug
2044 introduced by 4.64-PH/07).
2046 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
2047 child addresses, Exim now panics and dies. Previously, because the count
2048 is held in a short int, deliveries were likely to be lost. As such a
2049 large number of recipients for a single message is ridiculous
2050 (performance will be very, very poor), I have chosen to impose a limit
2051 rather than extend the field.
2057 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
2058 leftover -K file (the existence of which was triggered by #402).
2059 While we were at it, introduced process PID as part of the -K
2060 filename. This should rule out race conditions when creating
2063 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
2064 processing considerably. Previous code took too long for large mails,
2065 triggering a timeout which in turn triggers #401.
2067 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
2068 in the DK code in transports.c. sendfile() is not really portable,
2069 hence the _LINUX specificness.
2071 TF/01 In the add_headers option to the mail command in an Exim filter,
2072 there was a bug that Exim would claim a syntax error in any
2073 header after the first one which had an odd number of characters
2076 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
2077 callout verification, Exim cached a "reject" for the entire domain. This
2078 is correct for most verifications, but it is not correct for a recipient
2079 verification with use_sender or use_postmaster set, because in that case
2080 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
2081 case of MAIL FROM:<> rejection from other early rejections (e.g.
2082 rejection of HELO). When verifying a recipient using a non-null MAIL
2083 address, the cache is ignored if it shows MAIL FROM:<> rejection.
2084 Whatever the result of the callout, the value of the domain cache is
2085 left unchanged (for any other kind of callout, getting as far as trying
2086 RCPT means that the domain itself is ok).
2088 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
2091 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
2092 manifest itself as EPIPE rather than ECONNECT. When tidying away a
2093 session, the daemon ignores ECONNECT errors and logs others; it now
2094 ignores EPIPE as well.
2096 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
2097 (quoted-printable decoding).
2099 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
2100 later the small subsequent patch to fix an introduced bug.
2102 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
2104 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
2106 PH/08 An error is now given if message_size_limit is specified negative.
2108 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
2109 to be given (somewhat) arbitrary names.
2111 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
2114 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
2115 miscellaneous code fixes
2117 PH/10 Added the log_reject_target ACL modifier to specify where to log
2120 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
2121 hostname. This is wrong, because it relates to the incoming message (and
2122 probably the interface on which it is arriving) and not to the outgoing
2123 callout (which could be using a different interface). This has been
2124 changed to use the value of the helo_data option from the smtp transport
2125 instead - this is what is used when a message is actually being sent. If
2126 there is no remote transport (possible with a router that sets up host
2127 addresses), $smtp_active_hostname is used.
2129 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
2130 tweaks were necessary in order to get it to work (see also 21 below):
2131 (a) The code assumed that strncpy() returns a negative number on buffer
2132 overflow, which isn't the case. Replaced with Exim's string_format()
2134 (b) There were several signed/unsigned issues. I just did the minimum
2135 hacking in of casts. There is scope for a larger refactoring.
2136 (c) The code used strcasecmp() which is not a standard C function.
2137 Replaced with Exim's strcmpic() function.
2138 (d) The code set only $1; it now sets $auth1 as well.
2139 (e) A simple test gave the error "authentication client didn't specify
2140 service in request". It would seem that Dovecot has changed its
2141 interface. Fortunately there's a specification; I followed it and
2142 changed what the client sends and it appears to be working now.
2144 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
2147 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
2148 address A is aliased to B and C, where B exists and C does not. Without
2149 -v the output is "A verified" because verification stops after a
2150 successful redirection if more than one address is generated. However,
2151 with -v the child addresses are also verified. Exim was outputting "A
2152 failed to verify" and then showing the successful verification for C,
2153 with its parentage. It now outputs "B failed to verify", showing B's
2154 parentage before showing the successful verification of C.
2156 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
2157 look up a TXT record in a specific list after matching in a combined
2160 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
2161 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
2162 they consult the DNS. I had assumed they would set it the way they
2163 wanted; and indeed my experiments on Linux seem to show that in some
2164 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
2165 To be on the safe side, however, I have now made the interface to
2166 host_find_byname() similar to host_find_bydns(), with an argument
2167 containing the DNS resolver options. The host_find_byname() function now
2168 sets these options at its start, just as host_find_bydns() does. The smtp
2169 transport options dns_qualify_single and dns_search_parents are passed to
2170 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
2171 of host_find_byname() use the default settings of RES_DEFNAMES
2172 (qualify_single) but not RES_DNSRCH (search_parents).
2174 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
2175 spool_read_header() do less string testing, by means of a preliminary
2176 switch on the second character of optional "-foo" lines. (This is
2177 overdue, caused by the large number of possibilities that now exist.
2178 Originally there were few.) While I was there, I also converted the
2179 str(n)cmp tests so they don't re-test the leading "-" and the first
2180 character, in the hope this might squeeze out yet more improvement.
2182 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
2183 flag allowing group syntax was set by the header_syntax check but not
2184 turned off, possible causing trouble later; (2) The flag was not being
2185 set at all for the header_verify test, causing "group"-style headers to
2186 be rejected. I have now set it in this case, and also caused header_
2187 verify to ignore an empty address taken from a group. While doing this, I
2188 came across some other cases where the code for allowing group syntax
2189 while scanning a header line wasn't quite right (mostly, not resetting
2190 the flag correctly in the right place). These bugs could have caused
2191 trouble for malformed header lines. I hope it is now all correct.
2193 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
2194 with the "reply" argument non-NULL. The code, however (which originally
2195 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
2196 but it didn't always do it. This confused somebody who was copying the
2197 code for some other use. I have removed all the tests.
2199 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
2200 feature that was used to support insecure browsers during the U.S. crypto
2201 embargo. It requires special client support, and Exim is probably the
2202 only MTA that supported it -- and would never use it because real RSA is
2203 always available. This code has been removed, because it had the bad
2204 effect of slowing Exim down by computing (never used) parameters for the
2205 RSA_EXPORT functionality.
2207 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
2208 authenticator to fail if there's a tab character in the incoming data
2209 (there should never be unless someone is messing about, as it's supposed
2210 to be base64-encoded). Also added, on Timo's advice, the "secured" option
2211 if the connection is using TLS or if the remote IP is the same as the
2212 local IP, and the "valid-client-cert option" if a client certificate has
2215 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
2216 authenticators. This can be used for authorization after authentication
2217 succeeds. (In the case of plaintext, it servers for both authentication
2220 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
2221 if any retry times were supplied.
2223 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
2224 connection, where there is no client IP address to check. In this
2225 situation, the verify now always succeeds.
2227 PH/25 Applied John Jetmore's -Mset patch.
2229 PH/26 Added -bem to be like -Mset, but loading a message from a file.
2231 PH/27 In a string expansion for a processed (not raw) header when multiple
2232 headers of the same name were present, leading whitespace was being
2233 removed from all of them, but trailing whitespace was being removed only
2234 from the last one. Now trailing whitespace is removed from each header
2235 before concatenation. Completely empty headers in a concatenation (as
2236 before) are ignored.
2238 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
2239 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
2241 PH/29 [Removed. This was a change that I later backed out, and forgot to
2242 correct the ChangeLog entry (that I had efficiently created) before
2243 committing the later change.]
2245 PH/30 Exim was sometimes attempting to deliver messages that had suffered
2246 address errors (4xx response to RCPT) over the same connection as other
2247 messages routed to the same hosts. Such deliveries are always "forced",
2248 so retry times are not inspected. This resulted in far too many retries
2249 for the affected addresses. The effect occurred only when there were more
2250 hosts than the hosts_max_try setting in the smtp transport when it had
2251 the 4xx errors. Those hosts that it had tried were not added to the list
2252 of hosts for which the message was waiting, so if all were tried, there
2253 was no problem. Two fixes have been applied:
2255 (i) If there are any address or message errors in an SMTP delivery, none
2256 of the hosts (tried or untried) are now added to the list of hosts
2257 for which the message is waiting, so the message should not be a
2258 candidate for sending over the same connection that was used for a
2259 successful delivery of some other message. This seems entirely
2260 reasonable: after all the message is NOT "waiting for some host".
2261 This is so "obvious" that I'm not sure why it wasn't done
2262 previously. Hope I haven't missed anything, but it can't do any
2263 harm, as the worst effect is to miss an optimization.
2265 (ii) If, despite (i), such a delivery is accidentally attempted, the
2266 routing retry time is respected, so at least it doesn't keep
2267 hammering the server.
2269 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
2270 in ${readsocket because some servers need this prod.
2272 PH/32 Added some extra debug output when updating a wait-xxx database.
2274 PH/33 The hint "could be header name not terminated by colon", which has been
2275 given for certain expansion errors for a long time, was not being given
2276 for the ${if def:h_colon_omitted{... case.
2278 PH/34 The spec says: "With one important exception, whenever a domain list is
2279 being scanned, $domain contains the subject domain." There was at least
2280 one case where this was not true.
2282 PH/35 The error "getsockname() failed: connection reset by peer" was being
2283 written to the panic log as well as the main log, but it isn't really
2284 panic-worthy as it just means the connection died rather early on. I have
2285 removed the panic log writing for the ECONNRESET error when getsockname()
2288 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
2289 runs only) independently of the message's sender address. This meant
2290 that, if the 4xx error was in fact related to the sender, a different
2291 message to the same recipient with a different sender could confuse
2292 things. In particualar, this can happen when sending to a greylisting
2293 server, but other circumstances could also provoke similar problems.
2294 I have changed the default so that the retry time for these errors is now
2295 based a combination of the sender and recipient addresses. This change
2296 can be overridden by setting address_retry_include_sender=false in the
2299 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
2300 remote server are returned as part of bounce messages. This was not
2301 happening for LMTP over a pipe (the lmtp transport), but now it is the
2302 same for both kinds of LMTP.
2304 PH/38 Despite being documented as not happening, Exim was rewriting addresses
2305 in header lines that were in fact CNAMEs. This is no longer the case.
2307 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
2308 and queue runs started by the daemon processed all messages. This has
2309 been fixed so that -R and -S can now usefully be given with -q<time>.
2311 PH/40 Import PCRE release 6.7 (fixes some bugs).
2313 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
2315 PH/42 Give an error if -q is specified more than once.
2317 PH/43 Renamed the variables $interface_address and $interface_port as
2318 $received_ip_address and $received_port, to make it clear that these
2319 values apply to message reception, and not to the outgoing interface when
2320 a message is delivered. (The old names remain recognized, of course.)
2322 PH/44 There was no timeout on the connect() call when using a Unix domain
2323 socket in the ${readsocket expansion. There now is.
2325 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
2326 be meaningful with "accept".
2328 SC/01 Eximstats V1.43
2329 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
2331 SC/02 Eximstats V1.44
2332 Use a glob alias rather than an array ref in the generated
2333 parser. This improves both readability and performance.
2335 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
2336 Collect SpamAssassin and rejection statistics.
2337 Don't display local sender or destination tables unless
2338 there is data to show.
2339 Added average volumes into the top table text output.
2341 SC/04 Eximstats V1.46
2342 Collect data on the number of addresses (recipients)
2343 as well as the number of messages.
2345 SC/05 Eximstats V1.47
2346 Added 'Message too big' to the list of mail rejection
2347 reasons (thanks to Marco Gaiarin).
2349 SC/06 Eximstats V1.48
2350 Mainlog lines which have GMT offsets and are too short to
2351 have a flag are now skipped.
2353 SC/07 Eximstats V1.49 (Alain Williams)
2354 Added the -emptyok flag.
2356 SC/08 Eximstats V1.50
2357 Fixes for obtaining the IP address from reject messages.
2359 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
2360 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
2361 whitesspace changes from 4.64-PH/27
2363 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
2366 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
2367 are found, allow negative numbers in numeric criteria)
2369 JJ/06 exipick.20061117.2, added new $message_body_missing variable
2371 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
2372 to match changes made in 4.64-PH/43
2374 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
2376 PH/47 Put in an explicit test for a DNS lookup of an address record where the
2377 "domain" is actually an IP address, and force a failure. This locks out
2378 those revolvers/nameservers that support "A-for-A" lookups, in
2379 contravention of the specifications.
2381 PH/48 When a host name was looked up from an IP address, and the subsequent
2382 forward lookup of the name timed out, the host name was left in
2383 $sender_host_name, contrary to the specification.
2385 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
2386 restricted to single-key lookups, Exim was not diagnosing an error if
2387 * or *@ was used with a query-style lookup.
2389 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
2391 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
2392 long ago, but noone interested enough thought of it. Let's just say that
2393 the "1.1" means that there are some new functions that weren't there at
2394 some point in the past.
2396 PH/51 Error processing for expansion failure of helo_data from an smtp
2397 transport during callout processing was broken.
2399 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
2400 tested/used via the -bh/-bhc/-bs options.
2402 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
2403 bug, fixed in subsequent PCRE releases).
2405 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
2406 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
2408 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
2414 SC/01 Use a glob alias rather than an array ref in eximstats generated
2415 parser. This improves both readability and performance.
2417 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
2418 Don't display local sender or destination tables in eximstats unless
2419 there is data to show.
2420 Added average volumes into the eximstats top table text output.
2422 SC/03 Collect data on the number of addresses (recipients) as well
2423 as the number of messages in eximstats.
2425 TF/01 Correct an error in the documentation for the redirect router. Exim
2426 does (usually) call initgroups() when daemonizing.
2428 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
2429 with consistent privilege compared to when running as a daemon.
2431 TF/03 Note in the spec that $authenticated_id is not set for local
2432 submissions from trusted users.
2434 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
2435 Thanks to Dean Brooks <dean@iglou.com> for the patch.
2437 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
2438 by adding some example configuration directives to the default
2439 configuration file. A little bit of work is required to uncomment the
2440 directives and define how usernames and passwords are checked, but
2441 there is now a framework to start from.
2443 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
2444 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
2445 without this. I don't know how relevant this is to other LDAP libraries.
2447 PH/02 Add the verb name to the "unknown ACL verb" error.
2449 PH/03 Magnus Holmgren's patch for filter_prepend_home.
2451 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
2453 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
2454 directory not expanded when it should be if an expanded home directory
2455 was set for the address (which is overridden by the transport).
2457 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
2460 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
2461 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
2462 because it is too late at that time, and has no effect.
2464 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
2465 security issue with \' (bugzilla #107). I could not use the
2466 PQescapeStringConn() function, because it needs a PGconn value as one of
2469 PH/08 When testing addresses using -bt, indicate those final addresses that
2470 are duplicates that would not cause an additional delivery. At least one
2471 person was confused, thinking that -bt output corresponded to deliveries.
2472 (Suppressing duplicates isn't a good idea as you lose the information
2473 about possibly different redirections that led to the duplicates.)
2475 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
2476 systems where poll() doesn't work, in particular OS X.
2478 PH/10 Added more information to debugging output for retry time not reached.
2480 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
2481 operations in malware.c.
2483 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
2486 PH/13 If write_rejectlog was set false when logging was sent to syslog with
2487 syslog_duplication set false, log lines that would normally be written
2488 both the the main log and to the reject log were not written to syslog at
2491 PH/14 In the default configuration, change the use of "message" in ACL warn
2492 statements to "add_header".
2494 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
2495 not followed by a command (e.g. "seen endif").
2497 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
2498 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
2501 PH/17 Added extra conditions to the default value of delay_warning_condition
2505 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
2506 { match{$h_precedence:}{(?i)bulk|list|junk} } \
2507 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
2510 The Auto-Submitted: and various List- headers are standardised, whereas I
2511 don't think Precedence: ever was.
2513 PH/18 Refactored debugging code in route_finduser() to show more information,
2514 in particular, the error code if getpwnam() issues one.
2516 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
2517 This is apparently needed in addition to the PH/07 change above to avoid
2518 any possible encoding problems.
2520 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
2521 but not after initializing Perl.
2523 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
2524 output them only if debugging. By default they are written stderr,
2525 apparently, which is not desirable.
2527 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
2530 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
2533 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
2535 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
2536 authenticated or an ident call has been made. Suppress the default
2537 values for $authenticated_id and $authenticated_sender (but permit -oMai
2538 and -oMas) when testing with -bh.
2540 PH/24 Re-jigged the order of the tests in the default configuration so that the
2541 tests for valid domains and recipients precede the DNS black list and CSA
2542 tests, on the grounds that those ones are more expensive.
2544 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
2545 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
2546 This bug exists in every version of Exim that I still have, right back to
2549 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
2550 However, an attempt to turn on case-sensitivity in a regex key by
2551 including (?-i) didn't work because the subject string was already
2552 lowercased, and the effects were non-intuitive. It turns out that a
2553 one-line patch can be used to allow (?-i) to work as expected.
2559 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
2560 other effects) broke the use of negated acl sub-conditions.
2562 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
2565 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
2566 "Deny" causes Exim to reject the incoming connection with a 554 error.
2567 Unfortunately, if there is a major crisis, such as a disk failure,
2568 tcp-wrappers gives "deny", whereas what one would like would be some
2569 kind of temporary error. A kludge has been added to help with this.
2570 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
2571 554 error is used if errno is still zero or contains ENOENT (which occurs
2572 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
2575 PH/03 Add -lutil to the default FreeBSD LIBS setting.
2577 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
2578 errors. Otherwise a message that provokes a temporary error (when other
2579 messages do not) can cause a whole host to time out.
2581 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
2582 addresses were routed directly to files or pipes from a redirect router.
2583 File deliveries just didn't batch; pipe deliveries might have suffered
2586 PH/06 A failure to get a lock for a hints database would erroneously always say
2587 "Failed to get write lock", even when it was really a read lock.
2589 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
2590 of 0600. This has been changed to use the value of the lockfile_mode
2591 option (which defaults to 0600).
2593 PH/08 Applied small patch from the Sieve maintainer.
2595 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
2596 folder from quota calculations, a direct delivery into this folder messed
2597 up the contents of the maildirsize file. This was because the regex was
2598 used only to exclude .Trash (or whatever) when the size of the mailbox
2599 was calculated. There was no check that a delivery was happening into an
2600 excluded directory. This bug has been fixed by ignoring all quota
2601 processing for deliveries into excluded directories.
2603 PH/10 Added the maildirfolder_create_regex option to appendfile.
2609 PH/01 The code for finding all the local interface addresses on a FreeBSD
2610 system running IPv6 was broken. This may well have applied to all BSD
2611 systems, as well as to others that have similar system calls. The broken
2612 code found IPv4 interfaces correctly, but gave incorrect values for the
2613 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
2614 that it would not match correctly against @[] and not recognize the IPv6
2617 PH/02 The ipliteral router was not recognizing addresses of the form user@
2618 [ipv6:....] because it didn't know about the "ipv6:" prefix.
2620 PH/03 Added disable_ipv6.
2622 PH/04 Changed $reply_address to use the raw form of the headers instead of the
2623 decoded form, because it is most often used to construct To: headers
2624 lines in autoreplies, and the decoded form may well be syntactically
2625 invalid. However, $reply_address has leading white space removed, and all
2626 newlines turned into spaces so that the autoreply transport does not
2629 PH/05 If group was specified without a user on a router, and no group or user
2630 was specified on a transport, the group from the router was ignored.
2632 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
2633 for visible compile-time settings that can be used to change these
2634 numbers, for those that want even more. Backwards compatibility with old
2635 spool files has been maintained. However, going back to a previous Exim
2636 release will lost any variables that are in spool files.
2638 PH/07 Two small changes when running in the test harness: increase delay when
2639 passing a TCP/IP connection to a new process, in case the original
2640 process has to generate a bounce, and remove special handling of
2641 127.0.0.2 (sic), which is no longer necessary.
2643 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
2644 be the same on different OS.
2646 PH/09 Moved a debug statement in filter processing to avoid a race problem when
2649 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
2650 whether --show-vars was specified or not
2652 JJ/02 exipick: Added support for new ACL variable spool format introduced
2655 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
2656 syntactically invalid From: or Reply-to: line, and a filter used this to
2657 generate an autoreply, and therefore failed to obtain an address for the
2658 autoreply, Exim could try to deliver to a non-existent relative file
2659 name, causing unrelated and misleading errors. What now happens is that
2660 it logs this as a hard delivery error, but does not attempt to create a
2663 PH/11 The exinext utility has a -C option for testing purposes, but although
2664 the given file was scanned by exinext itself; it wasn't being passed on
2665 when Exim was called.
2667 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
2668 an end-of-file indication when reading a command response.
2670 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
2671 compiled. In many other places in Exim, IPv6 addresses are always
2672 recognized, so I have changed this. It also means that IPv4 domain
2673 literals of the form [IPV4:n.n.n.n] are now always recognized.
2675 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
2676 used if the router is not running as root, for example, when verifying at
2677 ACL time, or when using -bh. The debugging output from this situation was
2678 non-existent - all you got was a failure to exec. I have made two
2681 (a) Failures to set uid/gid, the current directory, or a process leader
2682 in a subprocess such as that created by queryprogram now generate
2683 suitable debugging ouput when -d is set.
2685 (b) The queryprogram router detects when it is not running as root,
2686 outputs suitable debugging information if -d is set, and then runs
2687 the subprocess without attempting to change uid/gid.
2689 PH/15 Minor change to Makefile for building test_host (undocumented testing
2692 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
2693 additional section of a DNS packet that returns MX or SRV records.
2694 Instead, it always explicitly searches for A/AAAA records. This avoids
2695 major problems that occur when a DNS server includes only records of one
2696 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
2697 fixed another bug: if SRV records were looked up and the corresponding
2698 address records were *not* found in the additional section, the port
2699 values from the SRV records were lost.
2701 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
2702 using the correct key (the original address) when searching the retry
2703 rules in order to find which one to use for generating the retry hint.
2705 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
2706 adding the default one. Similarly, if it contains a Reply-To: header, the
2707 errors_reply_to option, if set, is not used.
2709 PH/19 When calculating a retry time, Exim used to measure the "time since
2710 failure" by looking at the "first failed" field in the retry record. Now
2711 it does not use this if it is later than than the arrival time of the
2712 message. Instead it uses the arrival time. This makes for better
2713 behaviour in cases where some deliveries succeed, thus re-setting the
2714 "first failed" field. An example is a quota failure for a huge message
2715 when small messages continue to be delivered. Without this change, the
2716 "time since failure" will always be short, possible causing more frequent
2717 delivery attempts for the huge message than are intended.
2718 [Note: This change was subsequently modified - see PH/04 for 4.62.]
2720 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
2721 $1, $2, $3) because the numerical variables can be reset during some
2722 expansion items (e.g. "match"), thereby losing the authentication data.
2724 PH/21 Make -bV show the size of off_t variables so that the test suite can
2725 decide whether to run tests for quotas > 2G.
2727 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
2728 mailbox_size, and mailbox_filecount in the appendfile transport. If a
2729 filecount value is greater than 2G or if a quota value is greater than 2G
2730 on a system where the size of off_t is not greater than 4, a panic error
2733 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
2734 never match. The debug and -bh output now contains an explicit error
2735 message indicating a malformed IPv4 address or mask.
2737 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
2738 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
2739 PH/23 above applies.
2741 PH/25 Do not write to syslog when running in the test harness. The only
2742 occasion when this arises is a failure to open the main or panic logs
2743 (for which there is an explicit test).
2745 PH/26 Added the /no_tell option to "control=freeze".
2747 PH/27 If a host name lookup failed very early in a connection, for example, if
2748 the IP address matched host_lookup and the reverse lookup yielded a name
2749 that did not have a forward lookup, an error message of the form "no IP
2750 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
2751 could be logged. Now it outputs the IP address instead of "NULL".
2753 PH/28 An enabling patch from MH: add new function child_open_exim2() which
2754 allows the sender and the authenticated sender to be set when
2755 submitting a message from within Exim. Since child_open_exim() is
2756 documented for local_scan(), the new function should be too.
2758 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
2759 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
2760 results in an empty string is now treated as unset.
2762 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
2764 PH/31 Added sender_verify_fail logging option.
2766 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
2767 needed by RFC 822 but not by RFC 2822 was commented out. I have now
2768 tidied the source and removed it altogether.
2770 PH/33 When a queue run was abandoned because the load average was too high, a
2771 log line was always written; now it is written only if the queue_run log
2772 selector is set. In addition, the log line for abandonment now contains
2773 information about the queue run such as the pid. This is always present
2774 in "start" and "stop" lines but was omitted from the "abandon" line.
2776 PH/34 Omit spaces between a header name and the colon in the error message that
2777 is given when verify = headers_syntax fails (if there are lots of them,
2778 the message gets confusing).
2780 PH/35 Change the default for dns_check_names_pattern to allow slashes within
2781 names, as there are now some PTR records that contain slashes. This check
2782 is only to protect against broken name servers that fall over on strange
2783 characters, so the fact that it applies to all lookups doesn't matter.
2785 PH/36 Now that the new test suite is complete, we can remove some of the
2786 special code in Exim that was needed for the old test suite. For example,
2787 sorting DNS records because real resolvers return them in an arbitrary
2788 order. The new test suite's fake resolver always returns records in the
2791 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
2792 bounces) except when queue_only is set, to avoid logging races between
2793 the different processes.
2795 PH/38 Panic-die if .include specifies a non-absolute path.
2797 PH/39 A tweak to the "H" retry rule from its user.
2799 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
2800 a label. They prevented compilation on older perls.
2802 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
2803 a warning to be raised on newish perls.
2805 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
2806 on queue. Changes to match documented behaviour of showing count of
2807 messages matching specified criteria.
2809 PH/40 Changed the default ident timeout from 30s to 5s.
2811 PH/41 Added support for the use of login_cap features, on those BSD systems
2812 that have them, for controlling the resources used by pipe deliveries.
2814 PH/42 The content-scanning code uses fopen() to create files in which to put
2815 message data. Previously it was not paying any attention to the mode of
2816 the files. Exim runs with umask(0) because the rest of the code creates
2817 files with open(), and sets the required mode explicitly. Thus, these
2818 files were ending up world-writeable. This was not a big issue, because,
2819 being within the spool directory, they were not world-accessible. I have
2820 created a function called modefopen, which takes an additional mode
2821 argument. It sets umask(777), creates the file, chmods it to the required
2822 mode, then resets the umask. All the relevant calls to fopen() in the
2823 content scanning code have been changed to use this function.
2825 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
2826 to 24 hours. This avoids potential overflow problems when processing G
2827 and H retry rules. I suspect nobody ever tinkers with this value.
2829 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
2831 PH/45 When the plaintext authenticator is running as a client, the server's
2832 challenges are checked to ensure they are valid base64 strings. By
2833 default, the authentication attempt is cancelled if an invalid string is
2834 received. Setting client_ignore_invalid_base64 true ignores these errors.
2835 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
2836 they are received. Thus, the responses can be made to depend on the
2837 challenges. If an invalid string is ignored, an empty string is placed in
2840 PH/46 Messages that are created by the autoreply transport now contains a
2841 References: header, in accordance with RFCs 2822 and 3834.
2843 PH/47 Added authenticated_sender_force to the smtp transport.
2845 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
2847 PH/49 Installed latest patch from the Sieve maintainer.
2849 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
2850 was also set, the appendfile transport was unnecessarily scanning a
2851 directory of message files (e.g. for maildir delivery) to find the count
2852 of files (along with the size), even though it did not need this
2853 information. It now does the scan only if it needs to find either the
2854 size of the count of files.
2856 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
2858 PH/52 Two bugs concerned with error handling when the smtp transport is
2861 (i) Exim was not creating retry information for temporary errors given
2862 for individual recipients after the DATA command when the smtp transport
2863 was used in LMTP mode. This meant that they could be retried too
2864 frequently, and not timed out correctly.
2866 (ii) Exim was setting the flag that allows error details to be returned
2867 for LMTP errors on RCPT commands, but not for LMTP errors for individual
2868 recipients that were returned after the DATA command.
2870 PH/53 This is related to PH/52, but is more general: for any failing address,
2871 when detailed error information was permitted to be returned to the
2872 sender, but the error was temporary, then after the final timeout, only
2873 "retry timeout exceeded" was returned. Now it returns the full error as
2874 well as "retry timeout exceeded".
2876 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
2877 do this, and (what is worse) MTAs that accept it.
2879 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
2880 will now be deprecated.
2882 PH/56 New os.c-cygwin from the Cygwin maintainer.
2884 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
2885 formats (previously only available in exim formats via -bpr, -bpru,
2886 and -bpra. Now also available in native and exiqgrep formats)
2888 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
2889 with very large, slow to parse queues
2891 JJ/08 exipick: added ! as generic prefix to negate any criteria format
2893 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
2895 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
2896 responses to authentication challenges, though it was showing the
2897 challenges; (ii) I've removed the CR characters from the debug output for
2900 PH/58 Allow for the insertion of a newline as well as a space when a string
2901 is turned into more than one encoded-word during RFC 2047 encoding. The
2902 Sieve code now uses this.
2904 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
2905 data_4xx, lost_connection, tls_required.
2907 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
2908 message was being sent as an SMTP response.
2910 PH/61 Add -l and -k options to exicyclog.
2912 PH/62 When verifying, if an address was redirected to one new address, so that
2913 verification continued, and the new address failed or deferred after
2914 having set something in $address_data, the value of $address_data was not
2915 passed back to the ACL. This was different to the case when no
2916 redirection occurred. The value is now passed back in both cases.
2918 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
2919 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
2920 use login_cap.h, so on its own it isn't the distinguishing feature. The
2921 new name refers directly to the setclassresources() function.
2923 PH/65 Added configuration files for NetBSD3.
2925 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
2927 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
2928 is preferred over IPv4.
2930 PH/68 The bounce_return_message and bounce_return_body options were not being
2931 honoured for bounces generated during the reception of non-SMTP messages.
2932 In particular, this applied to messages rejected by the ACL. This bug has
2933 been fixed. However, if bounce_return_message is true and bounce_return_
2934 body is false, the headers that are returned for a non-SMTP message
2935 include only those that have been read before the error was detected.
2936 (In the case of an ACL rejection, they have all been read.)
2938 PH/69 The HTML version of the specification is now built in a directory called
2939 spec_html instead of spec.html, because the latter looks like a path with
2940 a MIME-type, and this confuses some software.
2942 PH/70 Catch two compiler warnings in sieve.c.
2944 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
2945 function verify_get_ident() calls ip_connect() to connect a socket, but
2946 if the "connect()" function timed out, ip_connect() used to close the
2947 socket. However, verify_get_ident() also closes the socket later, and in
2948 between Exim writes to the log, which may get opened at this point. When
2949 the socket was closed in ip_connect(), the log could get the same file
2950 descriptor number as the socket. This naturally causes chaos. The fix is
2951 not to close the socket in ip_connect(); the socket should be closed by
2952 the function that creates it. There was only one place in the code where
2953 this was missing, in the iplookup router, which I don't think anybody now
2954 uses, but I've fixed it anyway.
2956 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
2957 well as to direct DNS lookups. Otherwise the handling of names in host
2958 lists is inconsistent and therefore confusing.
2964 PH/01 Two changes to the default runtime configuration:
2966 (1) Move the checks for relay_from_hosts and authenticated clients from
2967 after to before the (commented out) DNS black list checks.
2969 (2) Add control=submission to the relay_from_hosts and authenticated
2970 clients checks, on the grounds that messages accepted by these
2971 statements are most likely to be submissions.
2973 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
2975 (1) Generate an error if the third argument for the ${prvs expansion is
2978 (2) Treat a missing third argument of ${prvscheck as if it were an empty
2981 (3) Reset the variables that are obtained from the first argument of
2982 ${prvscheck and used in the second argument before leaving the code,
2983 because their memory is reclaimed, so using them afterwards may do
2986 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
2987 one (it's much easier than Tom thought :-).
2989 (5) Because of (4), we can now allow for the use of $prvscheck_result
2990 inside the third argument.
2992 PH/03 For some reason, the default setting of PATH when running a command from
2993 a pipe transport was just "/usr/bin". I have changed it to
2996 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
2997 anything to be listed in the output from -bV.
2999 PH/05 When a filter generated an autoreply, the entire To: header line was
3000 quoted in the delivery log line, like this:
3002 => >A.N.Other <ano@some.domain> <original@ddress> ...
3004 This has been changed so that it extracts the operative address. There
3005 may be more than one such address. If so, they are comma-separated, like
3008 => >ano@some.domain,ona@other.domain <original@ddress> ...
3010 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
3011 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
3012 not being looked up in the rDNS to get a host name, Exim was showing the
3013 IP address twice in Received: lines, even though the IP addresses were
3014 identical. For example:
3016 Received: from [1.2.3.4] (helo=[1.2.3.4])
3018 However, if the real host name was known, it was omitting the HELO data
3019 if it matched the actual IP address. This has been tidied up so that it
3020 doesn't show the same IP address twice.
3022 PH/07 When both +timestamp and +memory debugging was on, the value given by
3023 $tod_xxx expansions could be wrong, because the tod_stamp() function was
3024 called by the debug printing, thereby overwriting the timestamp buffer.
3025 Debugging no longer uses the tod_stamp() function when +timestamp is set.
3027 PH/08 When the original message was included in an autoreply transport, it
3028 always said "this is a copy of the message, including all the headers",
3029 even if body_only or headers_only was set. It now gives an appropriate
3032 PH/09 Applied a patch from the Sieve maintainer which:
3034 o fixes some comments
3035 o adds the (disabled) notify extension core
3036 o adds some debug output for the result of if/elsif tests
3037 o points to the current vacation draft in the documentation
3038 and documents the missing references header update
3042 o fixes a bug in processing the envelope test (when testing
3043 multiple envelope elements, the last element determinted the
3046 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
3047 Electronic Mail") by including:
3049 Auto-submitted: auto-generated
3051 in the messages that it generates (bounce messages and others, such as
3052 warnings). In the case of bounce messages for non-SMTP mesages, there was
3053 also a typo: it was using "Auto_submitted" (underscore instead of
3054 hyphen). Since every message generated by Exim is necessarily in response
3055 to another message, thes have all been changed to:
3057 Auto-Submitted: auto-replied
3059 in accordance with these statements in the RFC:
3061 The auto-replied keyword:
3063 - SHOULD be used on messages sent in direct response to another
3064 message by an automatic process,
3066 - MUST NOT be used on manually-generated messages,
3068 - MAY be used on Delivery Status Notifications (DSNs) and Message
3069 Disposition Notifications (MDNs),
3071 - MUST NOT be used on messages generated by automatic or periodic
3072 processes, except for messages which are automatic responses to
3075 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
3076 to the default Received: header definition.
3078 PH/12 Added log selector acl_warn_skipped (default on).
3080 PH/13 After a successful wildlsearch lookup, discard the values of numeric
3081 variables because (a) they are in the wrong storage pool and (b) even if
3082 they were copied, it wouldn't work properly because of the caching.
3084 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
3085 checking when decoding. Apparently there are clients that generate
3086 overlong encoded strings. Why am I not surprised?
3088 PH/15 If the first argument of "${if match_address" was not empty, but did not
3089 contain an "@" character, Exim crashed. Now it writes a panic log message
3090 and treats the condition as false.
3092 PH/16 In autoreply, treat an empty string for "once" the same as unset.
3094 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
3095 extension "envelope-auth". The code is finished and in agreement with
3096 other implementations, but there is no documentation so far and in fact,
3097 nobody wrote the draft yet. This extension is currently #undef'ed, thus
3098 not changing the active code.
3100 Print executed "if" and "elsif" statements when debugging is used. This
3101 helps a great deal to understand what a filter does.
3103 Document more things not specified clearly in RFC3028. I had all this
3104 sorted out, when out of a sudden new issues came to my mind. Oops."
3106 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
3109 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
3110 canonical form (as documented). However, after a host name lookup from
3111 the IP address, check_host() was doing a simple string comparison with
3112 addresses acquired from the DNS when checking that the found name did
3113 have the original IP as one of its addresses. Since any found IPv6
3114 addresses are likely to be in abbreviated form, the comparison could
3115 fail. Luckily, there already exists a function for doing the comparison
3116 by converting both addresses to binary, so now that is used instead of
3117 the text comparison.
3119 PH/20 There was another similar case to PH/19, when a complete host name was
3120 given in a host list; looking up its IP address could give an abbreviated
3121 form, whereas the current host's name might or might not be abbreviated.
3122 The same fix has been applied.
3128 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
3129 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
3132 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
3133 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
3135 PH/03 Typo: missing ".o" in src/pcre/Makefile.
3137 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
3138 header line, restrict the check to what is listed in RFCs 2369 and 2929.
3139 Also, for "Auto-Submitted", treat anything other than "no" as
3140 non-personal, in accordance with RFC 3834. (Previously it treated
3141 anything starting "auto-" as non-personal.)
3143 TF/01 The control=submission/name=... option had a problem with syntax
3144 errors if the name included a slash character. The /name= option
3145 now slurps the rest of the string, so it can include any characters
3146 but it must come last in the list of options (after /sender_retain
3149 PH/05 Some modifications to the interface to the fake nameserver for the new
3157 TK/01 Added the "success_on_redirect" address verification option. See
3158 NewStuff for rationale and an example.
3160 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
3162 PH/02 Patch to exigrep to allow it to work on syslog lines.
3164 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
3165 fread() to skip over the body file's header line, because in Cygwin the
3166 header line is locked and is inaccessible.
3168 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
3169 co-exist for some time) to make it clear that it is the Exim ID that is
3170 referenced, not the Message-ID: header line.
3172 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
3173 string_format() function, because snprintf() does not exist on all
3176 PH/06 The use of forbid_filter_existstest now also locks out the use of the
3177 ${stat: expansion item.
3179 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
3180 protocol synchronization error", to keep the pedants happy.
3182 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
3183 well as for IRIX systems, when gcc is being used. See the host.c source
3186 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
3188 PH/10 Named domain lists were not working if used in a queue_smtp_domains
3191 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
3192 transport and to the smtp transport in LMTP mode.
3194 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
3196 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
3197 run a filter in a subprocess. This could lead to confusion in subsequent
3198 lookups in the parent process. There should also be a search_tidyup() at
3199 the end of the subprocess.
3201 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
3202 only if the host matched helo_try_verify_hosts, which caused the
3203 verification to occur when the EHLO/HELO command was issued. The ACL just
3204 tested the remembered result. Now, if a previous verification attempt has
3205 not happened, "verify = helo" does it there and then.
3207 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
3209 TK/03 Fix log output including CR from clamd.
3211 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
3212 exist provoked a memory error which could cause a segfault.
3214 PH/15 Installed PCRE 6.2
3216 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
3218 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
3219 of the problem. Specifically, suggested +O2 rather than +O1 for the
3222 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
3224 PH/20 If a delivery was routed to a non-standard port by means of an SRV
3225 record, the port was not correctly logged when the outgoing_port log
3226 selector was set (it logged the transort's default port).
3228 PH/21 Added support for host-specific ports to manualroute, queryprogram,
3229 fallback_hosts, and "hosts" in the smtp transport.
3231 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
3232 host errors such as "Connection refused".
3234 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
3235 authentication with radiusclient 0.4.9:
3237 - Error returned from rc_read_config was caught wrongly
3238 - Username/password not passed on to radius server due to wrong length.
3240 The presumption is that some radiusclient API changes for 4.51/PH/17
3241 were not taken care of correctly. The code is still untested by me (my
3242 Linux distribution still has 0.3.2 of radiusclient), but it was
3243 contributed by a Radius user.
3245 PH/24 When doing a callout, the value of $domain wasn't set correctly when
3246 expanding the "port" option of the smtp transport.
3248 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
3249 while reading a MIME header. Thanks to Tom Hughes for a patch.
3251 PH/24 Include config.h inside local_scan.h so that configuration settings are
3254 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
3255 that in an ACL for RCPT (for example), you can examine exactly what was
3258 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
3259 commands, but it was not correctly comparing the address with the actual
3260 client host address. Thus, it would show the EHLO address in Received:
3261 header lines when this was not necessary.
3263 PH/27 Added the % operator to ${eval:}.
3265 PH/28 Exim tries to create and chdir to its spool directory when it starts;
3266 it should be ignoring failures (because with -C, for example, it has lost
3267 privilege). It wasn't ignoring creation failures other than "already
3270 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
3273 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
3274 because an input line was too long, either on its own, or by virtue of
3275 too many continuations, the temporary file was not being removed, and the
3276 return code was incorrect.
3278 PH/31 Missing "BOOL" in function definition in filtertest.c.
3280 PH/32 Applied Sieve patches from the maintainer.
3282 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
3284 PH/33 Added "verify = not_blind".
3286 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
3287 Local/Makefile (with some defaults set). These are used in built scripts
3288 such as exicyclog, but they have never been used in the exim_install
3289 script (though there are many overriding facilities there). I have
3290 arranged that the exim_install script now takes note of these two
3293 PH/35 Installed configuration files for Dragonfly.
3295 PH/36 When a locally submitted message by a trusted user did not contain a
3296 From: header, and the sender address was obtained from -f or from an SMTP
3297 MAIL command, and the trusted user did not use -F to supply a sender
3298 name, $originator_name was incorrectly used when constructing a From:
3299 header. Furthermore, $originator_name was used for submission mode
3300 messages from external hosts without From: headers in a similar way,
3301 which is clearly wrong.
3303 PH/37 Added control=suppress_local_fixups.
3305 PH/38 When log_selector = +received_sender was set, and the addition of the
3306 sender made the log line's construction buffer exactly full, or one byte
3307 less than full, an overflow happened when the terminating "\n" was
3310 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
3311 when the result of a list match is failure because a DNS lookup failed.
3313 PH/40 RM_COMMAND is now used in the building process.
3315 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
3316 the "build-* directories that it finds.
3318 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
3319 address in a domain literal was a prefix of an interface address.
3321 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
3322 when verifying a sender address, unless rewrite_headers is false.
3324 PH/44 Wrote a long comment about why errors_to addresses are verified as
3325 recipients, not senders.
3327 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
3328 the ratelimit ACL was added.
3330 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
3332 PH/46 Added extra information about PostgreSQL errors to the error string.
3334 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
3335 avoiding the need to install special zones in a real server. This is
3336 backwards compatible; if it can't find the fake resolver, it drops back.
3337 Thus, both old and new test suites can be run.
3339 TF/02 Added util/ratelimit.pl
3341 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
3342 clock is set back in time.
3344 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
3345 Candler <B.Candler@pobox.com>.
3347 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
3348 OK for addresses that are the result of redirections.
3350 PH/48 A number of further additions for the benefit of the new test suite,
3351 including a fake gethostbyname() that interfaces to the fake DNS resolver
3354 TF/06 The fix for widen_domains has also been applied to qualify_single and
3355 search_parents which are the other dnslookup options that can cause
3358 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
3361 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
3363 TF/07 Exim produced the error message "an SRV record indicated no SMTP
3364 service" if it encountered an MX record with an empty target hostname.
3365 The message is now "an MX or SRV record indicated no SMTP service".
3367 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
3368 if the DNS of the sending site is misconfigured. This is quite a
3369 common situation. This change restores the behaviour of treating a
3370 helo verification defer as a failure.
3372 PH/51 If self=fail was set on a router, the bounce message did not include the
3373 actual error message.
3379 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
3381 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
3382 command itself ended in error, the underlying message about the transport
3383 filter timeout was being overwritten with the pipe command error. Now the
3384 underlying error message should be appended to the second error message.
3386 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
3388 PH/02 Reduce the amount of output that "make" produces by default. Full output
3389 can still be requested.
3391 PH/03 The warning log line about a condition test deferring for a "warn" verb
3392 was being output only once per connection, rather than after each
3393 occurrence (because it was using the same function as for successful
3394 "warn" verbs). This seems wrong, so I have changed it.
3396 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
3397 it should not have, which might have caused a crash in the right
3398 circumstances, but probably never did.
3400 PH/04 Installed a modified version of Tony Finch's patch to make submission
3401 mode fix the return path as well as the Sender: header line, and to
3402 add a /name= option so that you can make the user's friendly name appear
3405 TF/03 Added the control = fakedefer ACL modifier.
3407 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
3408 Mark Lowes for thorough testing.
3410 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
3412 TK/03 Merged latest SRS patch from Miles Wilton.
3414 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
3415 with the definition in sysexits.h (which is #included earlier).
3416 Fortunately, Exim does not actually use EX_OK. The code used to try to
3417 preserve the sysexits.h value, by assumimg that macro definitions were
3418 scanned for macro replacements. I have been disabused of this notion,
3419 so now the code just undefines EX_OK before #including unistd.h.
3421 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
3422 in the smtp transport. When a block could not be written in a single
3423 write() function, the timeout was being re-applied to each part-write.
3424 This seems wrong - if the receiver was accepting one byte at a time it
3425 would take for ever. The timeout is now adjusted when this happens. It
3426 doesn't have to be particularly precise.
3428 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
3429 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
3431 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
3432 without a domain if the check to <postmaster@domain> fails.
3434 SC/01 Eximstats: added -xls and the ability to specify output files
3435 (patch written by Frank Heydlauf).
3437 SC/02 Eximstats: use FileHandles for outputing results.
3439 SC/03 Eximstats: allow any combination of xls, txt, and html output.
3441 SC/04 Eximstats: fixed display of large numbers with -nvr option
3443 SC/05 Eximstats: fixed merging of reports with empty tables.
3445 SC/06 Eximstats: added the -include_original_destination flag
3447 SC/07 Eximstats: removed tabs and trailing whitespace.
3449 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
3451 TK/06 MBOX spool code: Add real "From " MBOX separator line
3452 so the .eml file is really in mbox format (even though
3453 most programs do not really care). Patch from Alex Miller.
3455 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
3456 The latter is generated from $received_to and is only set if the
3457 message has one envelope recipient. SA can use these headers,
3458 obviously out-of-the-box. Patch from Alex Miller.
3460 PH/08 The ${def test on a variable was returning false if the variable's
3461 value was "0", contrary to what the specification has always said!
3462 The result should be true unless the variable is empty.
3464 PH/09 The syntax error of a character other than { following "${if
3465 def:variable_name" (after optional whitespace) was not being diagnosed.
3466 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
3467 accidental colon was present, for example, could give incorrect results.
3469 PH/10 Tidied the code in a number of places where the st_size field of a stat()
3470 result is used (not including appendfile, where other changes are about
3473 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
3474 This involved changing a lot of size variables from int to off_t. It
3475 should work with maildirs and everything.
3477 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
3478 spamd dying while we are connected to it.
3480 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
3483 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
3484 function for BDB 4.3.
3486 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
3488 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
3489 the sources for PCRE have been re-arranged and more files are now
3492 PH/15 The code I had for printing potentially long long variables in PH/11
3493 above was not the best (it lost precision). The length of off_t variables
3494 is now inspected at build time, and an appropriate printing format (%ld
3495 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
3496 to be "long long int" or "long int". This is needed for the internal
3497 formatting function string_vformat().
3499 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
3500 the configuration file to be ":syslog", then the script "guesses" where
3501 the logs files are, rather than using the compiled in default. In our
3502 case the guess is not the same as the compiled default, so the script
3503 suddenly stopped working when I started to use syslog. The patch checks
3504 to see if log_file_path is "". If so, it attempts to read it from exim
3505 with no configuration file to get the compiled in version, before it
3506 falls back to the previous guessing code."
3508 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
3509 implementing BATV in an Exim configuration. See NewStuff for the gory
3512 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
3513 Makefile that are specific to HP-UX.
3515 PH/18 If the "use_postmaster" option was set for a recipient callout together
3516 with the "random" option, the postmaster address was used as the MAIL
3517 FROM address for the random test, but not for the subsequent recipient
3518 test. It is now used for both.
3520 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
3521 patch removes a few documentation additions to RFC 3028, because the
3522 latest draft now contains them. It adds the new en;ascii-case comparator
3523 and a new error check for 8bit text in MIME parts. Comparator and
3524 require names are now matched exactly. I enabled the subaddress
3525 extension, but it is not well tested yet (read: it works for me)."
3527 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
3528 rework some of the code of TK/09 above to avoid the hardwired use of
3529 "%lld" and "long long". Replaced the call to snprintf() with a call to
3532 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
3533 records point to non-existent hosts", "retry timeout exceeded", and
3534 "retry time not reached for any host after a long failure period".
3536 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
3537 experimental DomainKeys support:
3539 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
3540 (2) On an error such as an illegally used "control", the wrong name for
3541 the control was given.
3543 These problems did NOT occur unless DomainKeys support was compiled.
3545 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
3547 PH/24 Added ${if match_ip condition.
3549 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
3550 databases so that it will be absolutely obvious if a crash occurs in the
3551 DB library. This is a regular occurrence (often caused by mis-matched
3554 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
3555 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
3556 picked up on a user's system that detects such things. There doesn't seem
3557 to be a gcc warning option for this - only an attribute that has to be
3558 put on the function's prototype. It seems that in Fedora Core 4 they have
3559 set this on a number of new functions. No doubt there will be more in due
3562 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
3563 specify a transport. However, if an address that was verified by such a
3564 router was the subject of a callout, Exim crashed because it tried to
3565 read the rcpt_include_affixes from the non-existent transport. Now it
3566 just assumes that the setting of that option is false. This bug was
3567 introduced by 4.51/PH/31.
3569 PH/28 Changed -d+all to exclude +memory, because that information is very
3570 rarely of interest, but it makes the output a lot bigger. People tend to
3571 do -d+all out of habit.
3573 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
3574 code in os-type was giving problems when libc.so lives in lib64, like on
3577 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
3578 aren't the modern standard, and it seems that some systems' include files
3579 don't always have them. Exim was already checking for some of the newer
3580 ones like T_AAAA, and defining it itself. I've added checks for all the
3581 record types that Exim uses.
3583 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
3584 not automatically generating a new one, as it is supposed to. This
3585 prevented TLS from working. If the file did exist, but contained invalid
3586 data, a new version was generated, as expected. It was only the case of a
3587 non-existent file that was broken.
3589 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
3590 with a change in libdomainkeys > 0.64.
3592 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
3593 from DNS. If the selector record carries the flag, it now has
3594 precedence over the domain-wide flag.
3596 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
3598 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
3599 the use of a port name that isn't defined in /etc/services) occurred, the
3600 message was deferred as in a normal delivery, and thus remained on the
3601 spool, instead of being failed because of the mua_wrapper setting. This
3602 is now fixed, and I tidied up some of the mua_wrapper messages at the
3605 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
3606 the messages in a hash of arrays rather than using individual hashes.
3607 This is a bit cleaner and results in dramatic memory savings, albeit
3608 at a slight CPU cost.
3610 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
3611 as requested by Marc Sherman.
3613 SC/10 Eximstats: added histograms for user specified patterns as requested
3616 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
3618 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
3619 fopen() in the content-scanning modules that did not already have it.
3625 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
3626 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
3628 TK/02 Fix ACL "control" statement not being available in MIME ACL.
3630 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
3632 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
3633 to test Sieve filters that use "vacation".
3635 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
3636 that changes the way the GnuTLS parameters are stored in the cache file.
3637 The new format can be generated externally. For backward compatibility,
3638 if the data in the cache doesn't make sense, Exim assumes it has read an
3639 old-format file, and it generates new data and writes a new file. This
3640 means that you can't go back to an older release without removing the
3643 PH/03 A redirect router that has both "unseen" and "one_time" set does not
3644 work if there are any delivery delays because "one_time" forces the
3645 parent to be marked "delivered", so its unseen clone is never tried
3646 again. For this reason, Exim now forbids the simultaneous setting of
3649 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
3650 redirected to themselves ("homonym" addresses). Read the long ChangeLog
3651 entry if you want to know the details. The fix, however, neglected to
3652 consider the case when local delivery batching is involved. The test for
3653 "previously delivered" was not happening when checking to see if an
3654 address could be batched with a previous (undelivered) one; under
3655 certain circumstances this could lead to multiple deliveries to the same
3658 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
3659 in its include files, and this causes problems building Exim.
3661 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
3662 header_syntax) but Exim was just ignoring anything given after a slash.
3663 In particular, this caused confusion with an attempt to use "verify =
3664 reverse_host_lookup/defer_ok". An error is now given when options are
3665 supplied for verify items that do not have them. (Maybe reverse_host_
3666 lookup should have a defer_ok option, but that's a different point.)
3668 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
3669 defined by RFC 821) to 2048, because there were problems with some AUTH
3670 commands, and RFC 1869 says the size should be increased for extended
3671 SMTP commands that take arguments.
3673 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
3676 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
3677 "unknown" error; now it says that the functionality isn't in the binary.
3679 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
3680 an address' error message when a string expansion fails (syntax or
3681 whatever). Otherwise the password may appear in the log. Following change
3682 PH/42 below, there is no longer a chance of it appearing in a bounce
3685 PH/11 Installed exipick version 20050225.0 from John Jetmore.
3687 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
3688 of its addresses was ever tried. (Bugzilla bug #2.)
3690 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
3691 the result incorrectly in the debug output. (It correctly added a newline
3692 to what was transported.)
3694 TF/01 Added $received_time.
3696 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
3697 commented out examples of how to interface to a virus scanner and to
3698 SpamAssassin. Also added commented examples of av_scanner and
3699 spamd_address settings.
3701 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
3702 and controls are allowed in which ACLs. There were a couple of minor
3703 errors. Some of the entries in the conditions table (which is a table of
3704 where they are NOT allowed) were getting very unwieldy; rewrote them as a
3705 negation of where the condition IS allowed.
3707 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
3709 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
3710 header file does not have a version number, so I've had to invent a new
3711 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
3712 API. The code is untested by me (my Linux distribution still has 0.3.2 of
3713 radiusclient), but it was contributed by a Radius user.
3715 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
3716 files or directories, for OpenSSL.
3718 PH/19 When an Exim process that is running as root has to create an Exim log
3719 file, it does so in a subprocess that runs as exim:exim so as to get the
3720 ownership right at creation (otherwise, other Exim processes might see
3721 the file with the wrong ownership). There was no test for failure of this
3722 fork() call, which would lead to the process getting stuck as it waited
3723 for a non-existent subprocess. Forks do occasionally fail when resources
3724 run out. I reviewed all the other calls to fork(); they all seem to check
3727 PH/20 When checking for unexpected SMTP input at connect time (before writing
3728 the banner), Exim was not dealing correctly with a non-positive return
3729 from the read() function. If the client had disconnected by this time,
3730 the result was a log entry for a synchronization error with an empty
3731 string after "input=" when read() returned zero. If read() returned -1
3732 (an event I could not check), uninitialized data bytes were printed.
3733 There were reports of junk text (parts of files, etc) appearing after
3736 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
3738 PH/22 Added support for macro redefinition, and (re)definition in between
3739 driver and ACL definitions.
3741 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
3742 forgetting to use the resulting value; it was using the unexpanded value.
3744 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
3745 hadn't been configured. The fix is from Juergen Kreileder, who
3746 understands it better than I do:
3748 "Here's what I see happening with three configured cyrus_sasl
3749 authenticators configured (plain, login, cram-md5):
3751 On startup auth_cyrus_sasl_init() gets called for each of these.
3752 This means three calls to sasl_listmech() without a specified mech_list.
3753 => SASL tests which mechs of all available mechs actually work
3754 => three warnings about OTP not working
3755 => the returned list contains: plain, login, cram-md5, digest-md5, ...
3757 With the patch, sasl_listmech() also gets called three times. But now
3758 SASL's mech_list option is set to the server_mech specified in the the
3759 authenticator. Or in other words, the answer from sasl_listmech()
3760 gets limited to just the mech you're testing for (which is different
3762 => the return list contains just 'plain' or 'login', 'cram-md5' or
3763 nothing depending on the value of ob->server_mech.
3765 I've just tested the patch: Authentication still works fine,
3766 unavailable mechs specified in the exim configuration are still
3767 caught, and the auth.log warnings about OTP are gone."
3769 PH/25 When debugging is enabled, the contents of the command line are added
3770 to the debugging output, even when log_selector=+arguments is not
3773 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
3774 answer is "GNU", and only if the return is "GNU/something" is the answer
3777 PH/27 $acl_verify_message is now set immediately after the failure of a
3778 verification in an ACL, and so is available in subsequent modifiers. In
3779 particular, the message can be preserved by coding like this:
3781 warn !verify = sender
3782 set acl_m0 = $acl_verify_message
3784 Previously, $acl_verify_message was set only while expanding "message"
3785 and "log_message" when a very denied access.
3787 PH/28 Modified OS/os.c-Linux with
3789 -#ifndef OS_LOAD_AVERAGE
3790 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
3792 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
3793 nomenclature these days.)
3795 PH/29 Installed patch from the Sieve maintainer that adds the options
3796 sieve_useraddress and sieve_subaddress to the redirect router.
3798 PH/30 In these circumstances:
3799 . Two addresses routed to the same list of hosts;
3800 . First host does not offer TLS;
3801 . First host accepts first address;
3802 . First host gives temporary error to second address;
3803 . Second host offers TLS and a TLS session is established;
3804 . Second host accepts second address.
3805 Exim incorrectly logged both deliveries with the TLS parameters (cipher
3806 and peerdn, if requested) that were in fact used only for the second
3809 PH/31 When doing a callout as part of verifying an address, Exim was not paying
3810 attention to any local part prefix or suffix that was matched by the
3811 router that accepted the address. It now behaves in the same way as it
3812 does for delivery: the affixes are removed from the local part unless
3813 rcpt_include_affixes is set on the transport.
3815 PH/32 Add the sender address, as F=<...>, to the log line when logging a
3816 timeout during the DATA phase of an incoming message.
3818 PH/33 Sieve envelope tests were broken for match types other than :is. I have
3819 applied a patch sanctioned by the Sieve maintainer.
3821 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
3822 the uid or gid is negative. A case of a negative gid caused this to be
3823 noticed. The fix allows for either to be negative.
3825 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
3826 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
3829 PH/36 The change PH/12 above was broken. Fixed it.
3831 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
3832 the grounds that routing the same address twice would always produce the
3833 same answer. This might have been true once, but it is certainly no
3834 longer true now. Routing a child address may depend on the previous
3835 routing that produced that child. Some complicated redirection strategies
3836 went wrong when messages had multiple recipients, and made Exim's
3837 behaviour dependent on the order in which the addresses were given.
3839 I have moved the duplicate checking until after the routing is complete.
3840 Exim scans the addresses that are assigned to local and remote
3841 transports, and removes any duplicates. This means that more work will be
3842 done, as duplicates will always all be routed, but duplicates are
3843 presumably rare, so I don't expect this is of any significance.
3845 For deliveries to pipes, files, and autoreplies, the duplicate checking
3846 still happens during the routing process, since they are not going to be
3849 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
3850 It corrects a timeout issue with spamd. This is Ian's comment: "The
3851 background is that sometimes spamd either never reads data from a
3852 connection it has accepted, or it never writes response data. The exiscan
3853 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
3854 blindly assumes that writes won't block so it may never time out."
3856 PH/39 Allow G after quota size as well as K and M.
3858 PH/40 The value set for $authenticated_id in an authenticator may not contain
3859 binary zeroes or newlines because the value is written to log lines and
3860 to spool files. There was no check on this. Now the value is run through
3861 the string_printing() function so that such characters are converted to
3862 printable escape sequences.
3864 PH/41 $message_linecount is a new variable that contains the total number of
3865 lines in the message. Compare $body_linecount, which is the count for the
3868 PH/42 Exim no longer gives details of delivery errors for specific addresses in
3869 bounce and delay warning messages, except in certain special cases, which
3872 (a) An SMTP error message from a remote host;
3873 (b) A message specified in a :fail: redirection;
3874 (c) A message specified in a "fail" command in a system filter;
3875 (d) A message specified in a FAIL return from the queryprogram router;
3876 (e) A message specified by the cannot_route_message router option.
3878 In these cases only, Exim does include the error details in bounce and
3879 warning messages. There are also a few cases where bland messages such
3880 as "unrouteable address" or "local delivery error" are given.
3882 PH/43 $value is now also set for the "else" part of a ${run expansion.
3884 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
3885 being worked on, but at least Exim now implements the latest version to
3888 PH/45 In a pipe transport, although a timeout while waiting for the pipe
3889 process to complete was treated as a delivery failure, a timeout while
3890 writing the message to the pipe was logged, but erroneously treated as a
3891 successful delivery. Such timeouts include transport filter timeouts. For
3892 consistency with the overall process timeout, these timeouts are now
3893 treated as errors, giving rise to delivery failures by default. However,
3894 there is now a new Boolean option for the pipe transport called
3895 timeout_defer, which, if set TRUE, converts the failures into defers for
3896 both kinds of timeout. A transport filter timeout is now identified in
3899 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
3900 systems where "make" and "gmake" are different, calling "gmake" at top
3901 level broke things. I've arranged for the value of $(MAKE) to be passed
3902 from the Makefile to this script so that it can call the same version of
3906 A note about Exim versions 4.44 and 4.50
3907 ----------------------------------------
3909 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
3910 changes of various kinds. As a consequence, a big documentation update was
3911 needed. This delayed the release for rather longer than seemed good, especially
3912 in the light of a couple of (minor) security issues. Therefore, the changes
3913 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
3914 release. So 4.44 and 4.50 are in effect two different branches that both start
3917 I have left the 4.50 change log unchanged; it contains all the changes since
3918 4.43. The change log for 4.44 is below; many of its items are identical to
3919 those for 4.50. This seems to be the most sensible way to preserve the
3920 historical information.
3926 1. Minor wording change to the doc/README.SIEVE file.
3928 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
3929 computation of the current number of files was incorrect.
3931 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
3932 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
3935 4. Give more explanation in the error message when the command for a transport
3936 filter fails to execute.
3938 5. There are several places where Exim runs a non-Exim command in a
3939 subprocess. The SIGUSR1 signal should be disabled for these processes. This
3940 was being done only for the command run by the queryprogram router. It is
3941 now done for all such subprocesses. The other cases are: ${run, transport
3942 filters, and the commands run by the lmtp and pipe transports.
3944 6. Added CONFIGURE_GROUP build-time option.
3946 7. Some older OS have a limit of 256 on the maximum number of file
3947 descriptors. Exim was using setrlimit() to set 1000 as a large value
3948 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
3949 systems. I've change it so that if it can't get 1000, it tries for 256.
3951 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
3952 was an oversight, and furthermore, ever since the addition of extra
3953 controls (e.g. 4.43/32), the checks on when to allow different forms of
3954 "control" were broken. There should now be diagnostics for all cases when a
3955 control that does not make sense is encountered.
3957 9. Added the /retain_sender option to "control=submission".
3959 10. $recipients is now available in the predata ACL (oversight).
3961 11. Tidy the search cache before the fork to do a delivery from a message
3962 received from the command line. Otherwise the child will trigger a lookup
3963 failure and thereby defer the delivery if it tries to use (for example) a
3964 cached ldap connection that the parent has called unbind on.
3966 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
3967 of $address_data from the recipient verification was clobbered by the
3968 sender verification.
3970 13. The value of address_data from a sender verification is now available in
3971 $sender_address_data in subsequent conditions in the ACL statement.
3973 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
3975 15. Added a new option "connect=<time>" to callout options, to set a different
3978 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
3979 was its contents. (It was OK if the option was not defined at all.)
3981 17. A "Completed" log line is now written for messages that are removed from
3982 the spool by the -Mrm option.
3984 18. New variables $sender_verify_failure and $recipient_verify_failure contain
3985 information about exactly what failed.
3987 19. Added -dd to debug only the daemon process.
3989 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
3990 handles timeouts, both on the server side and network timeouts. Renamed the
3991 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
3993 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
3994 It is now set to "smtps".
3996 22. $host_address is now set to the target address during the checking of
3997 ignore_target_hosts.
3999 23. When checking ignore_target_hosts for an ipliteral router, no host name was
4000 being passed; this would have caused $sender_host_name to have been used if
4001 matching the list had actually called for a host name (not very likely,
4002 since this list is usually IP addresses). A host name is now passed as
4005 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4006 code that specifies a non-restarting handler (typically sigaction() in
4007 modern systems) in an attempt to fix a rare and obscure crash bug.
4009 25. Narrowed the window for a race in the daemon that could cause it to ignore
4010 SIGCHLD signals. This is not a major problem, because they are used only to
4011 wake it up if nothing else does.
4013 26. A malformed maildirsize file could cause Exim to calculate negative values
4014 for the mailbox size or file count. Odd effects could occur as a result.
4015 The maildirsize information is now recalculated if the size or filecount
4018 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4019 support for a long time. Removed HAVE_SYS_VFS_H.
4021 28. Installed the latest version of exipick from John Jetmore.
4023 29. In an address list, if the pattern was not a regular expression, an empty
4024 subject address (from a bounce message) matched only if the pattern was an
4025 empty string. Non-empty patterns were not even tested. This was the wrong
4026 because it is perfectly reasonable to use an empty address as part of a
4027 database query. An empty address is now tested by patterns that are
4028 lookups. However, all the other forms of pattern expect the subject to
4029 contain a local part and a domain, and therefore, for them, an empty
4030 address still always fails if the pattern is not itself empty.
4032 30. Exim went into a mad DNS loop when attempting to do a callout where the
4033 host was specified on an smtp transport, and looking it up yielded more
4034 than one IP address.
4036 31. Re-factored the code for checking spool and log partition space into a
4037 function that finds that data and another that does the check. The former
4038 is then used to implement four new variables: $spool_space, $log_space,
4039 $spool_inodes, and $log_inodes.
4041 32. The RFC2047 encoding function was originally intended for short strings
4042 such as real names; it was not keeping to the 75-character limit for
4043 encoded words that the RFC imposes. It now respects the limit, and
4044 generates multiple encoded words if necessary. To be on the safe side, I
4045 have increased the buffer size for the ${rfc2047: expansion operator from
4048 33. It is now permitted to omit both strings after an "if" condition; if the
4049 condition is true, the result is "true". As before, when the second string
4050 is omitted, a false condition yields an empty string. This makes it less
4051 cumbersome to write custom ACL and router conditions.
4053 34. Failure to deliver a bounce message always caused it to be frozen, even if
4054 there was an errors_to setting on the router. The errors_to setting is now
4057 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4058 canonical form (fully expanded) before being placed in
4059 $sender_host_address.
4061 36. The table in the code that translates DNS record types into text (T_A to
4062 "A" for instance) was missing entries for NS and CNAME. It is just possible
4063 that this could have caused confusion if both these types were looked up
4064 for the same domain, because the text type is used as part of Exim's
4065 per-process caching. But the chance of anyone hitting this buglet seems
4068 37. The dnsdb lookup has been extended in a number of ways.
4070 (1) There is a new type, "zns", which walks up the domain tree until it
4071 finds some nameserver records. It should be used with care.
4073 (2) There is a new type, "mxh", which is like "mx" except that it returns
4074 just the host names, not the priorities.
4076 (3) It is now possible to give a list of domains (or IP addresses) to be
4077 looked up. The behaviour when one of the lookups defers can be
4078 controlled by a keyword.
4080 (4) It is now possible to specify the separator character for use when
4081 multiple records are returned.
4083 38. The dnslists ACL condition has been extended: it is now possible to supply
4084 a list of IP addresses and/or domains to be looked up in a particular DNS
4087 39. Added log_selector=+queue_time_overall.
4089 40. When running the queue in the test harness, wait just a tad after forking a
4090 delivery process, to get repeatability of debugging output.
4092 41. Include certificate and key file names in error message when GnuTLS fails
4093 to set them up, because the GnuTLS error message doesn't include the name
4094 of the failing file when there is a problem reading it.
4096 42. Allow both -bf and -bF in the same test run.
4098 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
4100 44. The "Exiscan patch" is now merged into the mainline Exim source.
4102 45. Sometimes the final signoff response after QUIT could fail to get
4103 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4104 < 0 before doing a fflush(). This bug looks as though it goes back to the
4105 introduction of TLS in release 3.20, but "sometimes" must have been rare
4106 because the tests only now provoked it.
4108 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
4109 (this can affect the format of dates).
4111 47. exim_tidydb, when checking for the continued existence of a message for
4112 which it has found a message-specific retry record, was not finding
4113 messages that were in split spool directories. Consequently, it was
4114 deleting retry records that should have stayed in existence.
4116 48. Steve fixed some bugs in eximstats.
4118 49. The SPA authentication driver was not abandoning authentication and moving
4119 on to the next authenticator when an expansion was forced to fail,
4120 contradicting the general specification for all authenticators. Instead it
4121 was generating a temporary error. It now behaves as specified.
4123 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
4124 (the order specifies the preference for clients). The order is now AES256,
4125 AES128, 3DES, ARCFOUR128.
4127 51. Small patch to Sieve code - explicitly set From: when generating an
4130 52. Exim crashed if a remote delivery caused a very long error message to be
4131 recorded - for instance if somebody sent an entire SpamAssassin report back
4132 as a large number of 550 error lines. This bug was coincidentally fixed by
4133 increasing the size of one of Exim's internal buffers (big_buffer) that
4134 happened as part of the Exiscan merge. However, to be on the safe side, I
4135 have made the code more robust (and fixed the comments that describe what
4138 53. Now that there can be additional text after "Completed" in log lines (if
4139 the queue_time_overall log selector is set), a one-byte patch to exigrep
4140 was needed to allow it to recognize "Completed" as not the last thing in
4143 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
4144 patch that reportedly fixes this has been added. I am not expert enough to
4145 create a test for it. This is what the patch creator wrote:
4147 "I found a little strange behaviour of ldap code when working with
4148 Windows 2003 AD Domain, where users was placed in more than one
4149 Organization Units. When I tried to give exim partial DN, the exit code
4150 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
4151 But simultaneously result of request was absolutely normal ldap result,
4152 so I produce this patch..."
4154 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
4155 REFERENCE, so I have modified the code to exclude the patch when that macro
4158 55. Some experimental protocols are using DNS PTR records for new purposes. The
4159 keys for these records are domain names, not reversed IP addresses. The
4160 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4161 leaves it alone. Component reversal etc. now happens only for IP addresses.
4164 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
4166 57. Double the size of the debug message buffer (to 2048) so that more of very
4167 long debug lines gets shown.
4169 58. The exicyclog utility now does better if the number of log files to keep
4170 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
4172 59. Two changes related to the smtp_active_hostname option:
4174 (1) $smtp_active_hostname is now available as a variable.
4175 (2) The default for smtp_banner uses $smtp_active_hostname instead
4176 of $primary_hostname.
4178 60. The host_aton() function is supposed to be passed a string that is known
4179 to be a valid IP address. However, in the case of IPv6 addresses, it was
4180 not checking this. This is a hostage to fortune. Exim now panics and dies
4181 if the condition is not met. A case was found where this could be provoked
4182 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4183 components; fortuitously, this particular loophole had already been fixed
4184 by change 4.50/55 above.
4186 If there are any other similar loopholes, the new check in host_aton()
4187 itself should stop them being exploited. The report I received stated that
4188 data on the command line could provoke the exploit when Exim was running as
4189 exim, but did not say which command line option was involved. All I could
4190 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4191 running as the user.
4194 61. There was a buffer overflow vulnerability in the SPA authentication code
4195 (which came originally from the Samba project). I have added a test to the
4196 spa_base64_to_bits() function which I hope fixes it.
4199 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
4200 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
4202 63. The daemon start-up calls getloadavg() while still root for those OS that
4203 need the first call to be done as root, but it missed one case: when
4204 deliver_queue_load_max is set with deliver_drop_privilege. This is
4205 necessary for the benefit of the queue runner, because there is no re-exec
4206 when deliver_drop_privilege is set.
4208 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
4209 This has been fixed.
4211 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4212 was in use, was not putting the data itself into the right store pool;
4213 consequently, it could be overwritten for a subsequent message in the same
4214 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4217 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
4219 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
4220 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
4221 the function were treating the return as a boolean value, which happened to
4222 work because 0=false and not-0=true, but is not correct code.
4224 68. The host_aton() function was not handling scoped IPv6 addresses (those
4225 with, for example, "%eth0" on the end) correctly.
4227 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
4228 negated items (that is, ~something) in unsigned ints. Some compilers
4229 apparently mutter when there is no cast.
4231 70. If an address verification called from an ACL failed, and did not produce a
4232 user-specific message (i.e. there was only a "system" message), nothing was
4233 put in $acl_verify_message. In this situation, it now puts the system
4236 71. Change 4.23/11 added synchronization checking at the start of an SMTP
4237 session; change 4.31/43 added the unwanted input to the log line - except
4238 that it did not do this in the start of session case. It now does.
4240 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
4241 This is wrong and can cause the other end to generate a synchronization
4242 error if it is another Exim or anything else that does the synchronization
4243 check. A QUIT command is no longer sent after a timeout.
4245 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
4246 during host lookups.
4248 74. The defer_ok option of callout verification was not working if it was used
4249 when verifying addresses in header lines, that is, for this case:
4251 verify = header_sender/callout=defer_ok
4253 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
4254 those file descriptors could be used for SMTP connections. If anything
4255 wrote to stderr (the example that came up was "warn" in embedded Perl), it
4256 could be sent to the SMTP client, causing chaos. The daemon now opens
4257 stdin, stdout, and stderr to /dev/null when it puts itself into the
4260 76. Arrange for output from Perl's "warn" command to be written to Exim's main
4261 log by default. The user can override this with suitable Perl magic.
4263 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
4264 the log message when discard triggers, was not working for the DATA ACL or
4265 for the non-SMTP ACL.
4267 78. Error message wording change in sieve.c.
4269 79. If smtp_accept_max_per_host was set, the number of connections could be
4270 restricted to fewer than expected, because the daemon was trying to set up
4271 a new connection before checking whether the processes handling previous
4272 connections had finished. The check for completed processes is now done
4273 earlier. On busy systems, this bug wouldn't be noticed because something
4274 else would have woken the daemon, and it would have reaped the completed
4277 80. If a message was submitted locally by a user whose login name contained one
4278 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
4279 It caused a spool format error. I have fixed the spool reading code. A
4280 related problem was that the "from" clause in the Received: line became
4281 illegal because of the space(s). It is now covered by ${quote_local_part.
4283 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
4286 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
4288 83. Patch from Sieve maintainer to fix unterminated string problem in
4289 "vacation" handling.
4291 84. Some minor changes to the Linux configuration files to help with other
4292 OS variants using glibc.
4294 85. One more patch for Sieve to update vacation handling to latest spec.
4297 ----------------------------------------------------
4298 See the note above about the 4.44 and 4.50 releases.
4299 ----------------------------------------------------
4305 1. Change 4.43/35 introduced a bug that caused file counts to be
4306 incorrectly computed when quota_filecount was set in an appendfile
4309 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4310 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4313 3. Give more explanation in the error message when the command for a transport
4314 filter fails to execute.
4316 4. There are several places where Exim runs a non-Exim command in a
4317 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4318 was being done only for the command run by the queryprogram router. It is
4319 now done for all such subprocesses. The other cases are: ${run, transport
4320 filters, and the commands run by the lmtp and pipe transports.
4322 5. Some older OS have a limit of 256 on the maximum number of file
4323 descriptors. Exim was using setrlimit() to set 1000 as a large value
4324 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4325 systems. I've change it so that if it can't get 1000, it tries for 256.
4327 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4328 was an oversight, and furthermore, ever since the addition of extra
4329 controls (e.g. 4.43/32), the checks on when to allow different forms of
4330 "control" were broken. There should now be diagnostics for all cases when a
4331 control that does not make sense is encountered.
4333 7. $recipients is now available in the predata ACL (oversight).
4335 8. Tidy the search cache before the fork to do a delivery from a message
4336 received from the command line. Otherwise the child will trigger a lookup
4337 failure and thereby defer the delivery if it tries to use (for example) a
4338 cached ldap connection that the parent has called unbind on.
4340 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4341 of $address_data from the recipient verification was clobbered by the
4342 sender verification.
4344 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4345 was its contents. (It was OK if the option was not defined at all.)
4347 11. A "Completed" log line is now written for messages that are removed from
4348 the spool by the -Mrm option.
4350 12. $host_address is now set to the target address during the checking of
4351 ignore_target_hosts.
4353 13. When checking ignore_target_hosts for an ipliteral router, no host name was
4354 being passed; this would have caused $sender_host_name to have been used if
4355 matching the list had actually called for a host name (not very likely,
4356 since this list is usually IP addresses). A host name is now passed as
4359 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4360 code that specifies a non-restarting handler (typically sigaction() in
4361 modern systems) in an attempt to fix a rare and obscure crash bug.
4363 15. Narrowed the window for a race in the daemon that could cause it to ignore
4364 SIGCHLD signals. This is not a major problem, because they are used only to
4365 wake it up if nothing else does.
4367 16. A malformed maildirsize file could cause Exim to calculate negative values
4368 for the mailbox size or file count. Odd effects could occur as a result.
4369 The maildirsize information is now recalculated if the size or filecount
4372 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4373 support for a long time. Removed HAVE_SYS_VFS_H.
4375 18. Updated exipick to current release from John Jetmore.
4377 19. Allow an empty sender to be matched against a lookup in an address list.
4378 Previously the only cases considered were a regular expression, or an
4381 20. Exim went into a mad DNS lookup loop when doing a callout where the
4382 host was specified on the transport, if the DNS lookup yielded more than
4385 21. The RFC2047 encoding function was originally intended for short strings
4386 such as real names; it was not keeping to the 75-character limit for
4387 encoded words that the RFC imposes. It now respects the limit, and
4388 generates multiple encoded words if necessary. To be on the safe side, I
4389 have increased the buffer size for the ${rfc2047: expansion operator from
4392 22. Failure to deliver a bounce message always caused it to be frozen, even if
4393 there was an errors_to setting on the router. The errors_to setting is now
4396 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4397 canonical form (fully expanded) before being placed in
4398 $sender_host_address.
4400 24. Updated eximstats to version 1.33
4402 25. Include certificate and key file names in error message when GnuTLS fails
4403 to set them up, because the GnuTLS error message doesn't include the name
4404 of the failing file when there is a problem reading it.
4406 26. Expand error message when OpenSSL has problems setting up cert/key files.
4409 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
4410 (this can affect the format of dates).
4412 28. exim_tidydb, when checking for the continued existence of a message for
4413 which it has found a message-specific retry record, was not finding
4414 messages that were in split spool directories. Consequently, it was
4415 deleting retry records that should have stayed in existence.
4417 29. eximstats updated to version 1.35
4418 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
4419 1.35 - bugfix such that pie charts by volume are generated correctly
4421 30. The SPA authentication driver was not abandoning authentication and moving
4422 on to the next authenticator when an expansion was forced to fail,
4423 contradicting the general specification for all authenticators. Instead it
4424 was generating a temporary error. It now behaves as specified.
4426 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
4427 (the order specifies the preference for clients). The order is now AES256,
4428 AES128, 3DES, ARCFOUR128.
4430 31. Small patch to Sieve code - explicitly set From: when generating an
4433 32. Exim crashed if a remote delivery caused a very long error message to be
4434 recorded - for instance if somebody sent an entire SpamAssassin report back
4435 as a large number of 550 error lines. This bug was coincidentally fixed by
4436 increasing the size of one of Exim's internal buffers (big_buffer) that
4437 happened as part of the Exiscan merge. However, to be on the safe side, I
4438 have made the code more robust (and fixed the comments that describe what
4441 33. Some experimental protocols are using DNS PTR records for new purposes. The
4442 keys for these records are domain names, not reversed IP addresses. The
4443 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4444 leaves it alone. Component reversal etc. now happens only for IP addresses.
4447 34. The host_aton() function is supposed to be passed a string that is known
4448 to be a valid IP address. However, in the case of IPv6 addresses, it was
4449 not checking this. This is a hostage to fortune. Exim now panics and dies
4450 if the condition is not met. A case was found where this could be provoked
4451 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4452 components; fortuitously, this particular loophole had already been fixed
4453 by change 4.50/55 or 4.44/33 above.
4455 If there are any other similar loopholes, the new check in host_aton()
4456 itself should stop them being exploited. The report I received stated that
4457 data on the command line could provoke the exploit when Exim was running as
4458 exim, but did not say which command line option was involved. All I could
4459 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4460 running as the user.
4463 35. There was a buffer overflow vulnerability in the SPA authentication code
4464 (which came originally from the Samba project). I have added a test to the
4465 spa_base64_to_bits() function which I hope fixes it.
4468 36. The daemon start-up calls getloadavg() while still root for those OS that
4469 need the first call to be done as root, but it missed one case: when
4470 deliver_queue_load_max is set with deliver_drop_privilege. This is
4471 necessary for the benefit of the queue runner, because there is no re-exec
4472 when deliver_drop_privilege is set.
4474 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4475 was in use, was not putting the data itself into the right store pool;
4476 consequently, it could be overwritten for a subsequent message in the same
4477 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4480 38. Sometimes the final signoff response after QUIT could fail to get
4481 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4482 < 0 before doing a fflush(). This bug looks as though it goes back to the
4483 introduction of TLS in release 3.20, but "sometimes" must have been rare
4484 because the tests only now provoked it.
4490 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
4491 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
4492 Now it returns an integer. A number of calls were still expecting a T/F
4493 return. Fortuitously, in all cases, the tests worked in OK situations,
4494 which is the norm. However, things would have gone wrong on any write
4495 failures on the smtp file descriptor. This function is used when sending
4496 messages over SMTP and also when doing verify callouts.
4498 2. When Exim is called to do synchronous delivery of a locally submitted
4499 message (the -odf or -odi options), it no longer closes stderr before doing
4502 3. Implemented the mua_wrapper option.
4504 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
4506 5. Implemented the functions header_remove(), header_testname(),
4507 header_add_at_position(), and receive_remove_recipient(), and exported them
4510 6. If an ACL "warn" statement specified the addition of headers, Exim already
4511 inserted X-ACL-Warn: at the start if there was no header name. However, it
4512 was not making this test for the second and subsequent header lines if
4513 there were newlines in the string. This meant that an invalid header could
4514 be inserted if Exim was badly configured.
4516 7. Allow an ACL "warn" statement to add header lines at the start or after all
4517 the Received: headers, as well as at the end.
4519 8. Added the rcpt_4xx retry error code.
4521 9. Added postmaster_mailfrom=xxx to callout verification option.
4523 10. Added mailfrom=xxxx to the callout verification option, for verify=
4526 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
4527 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
4529 12. Inserted some casts to stop certain compilers warning when using pointer
4530 differences as field lengths or precisions in printf-type calls (mostly
4531 affecting debugging statements).
4533 13. Added optional readline() support for -be (dynamically loaded).
4535 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
4536 same clock tick as a message's arrival, so that its received time was the
4537 same as the "first fail" time on the retry record, and that message
4538 remained on the queue past the ultimate address timeout, every queue runner
4539 would try a delivery (because it was past the ultimate address timeout) but
4540 after another failure, the ultimate address timeout, which should have then
4541 bounced the address, did not kick in. This was a "< instead of <=" error;
4542 in most cases the first failure would have been in the next clock tick
4543 after the received time, and all would be well.
4545 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
4546 being recognized when the domain list was tested by the match_domain
4547 condition in an expansion string.
4549 16. Added the ${str2b64: operator.
4551 17. Exim was always calling setrlimit() to set a large limit for the number of
4552 processes, without checking whether the existing limit was already
4553 adequate. (It did check for the limit on file descriptors.) Furthermore,
4554 errors from getrlimit() and setrlimit() were being ignored. Now they are
4555 logged to the main and panic logs, but Exim does carry on, to try to do its
4556 job under whatever limits there are.
4558 18. Imported PCRE 5.0.
4560 19. Trivial typo in log message " temporarily refused connection" (the leading
4563 20. If the log selector return_path_on_delivery was set and an address was
4564 redirected to /dev/null, the delivery process crashed because it assumed
4565 that a return path would always be set for a "successful" delivery. In this
4566 case, the whole delivery is bypassed as an optimization, and therefore no
4569 21. Internal re-arrangement: the function for sending a challenge and reading
4570 a response while authentication was assuming a zero-terminated challenge
4571 string. It's now changed to take a pointer and a length, to allow for
4572 binary data in such strings.
4574 22. Added the cyrus_sasl authenticator (code supplied by MBM).
4576 23. Exim was not respecting finduser_retries when seeking the login of the
4577 uid under which it was called; it was always trying 10 times. (The default
4578 setting of finduser_retries is zero.) Also, it was sleeping after the final
4579 failure, which is pointless.
4581 24. Implemented tls_on_connect_ports.
4583 25. Implemented acl_smtp_predata.
4585 26. If the domain in control=submission is set empty, Exim assumes that the
4586 authenticated id is a complete email address when it generates From: or
4587 Sender: header lines.
4589 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
4590 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
4591 chown and chgrp in /bin and hostname in /usr/bin.
4593 28. Exim was keeping the "process log" file open after each use, just as it
4594 does for the main log. This opens the possibility of it remaining open for
4595 long periods when the USR1 signal hits a daemon. Occasional processlog
4596 errors were reported, that could have been caused by this. Anyway, it seems
4597 much more sensible not to leave this file open at all, so that is what now
4600 29. The long-running daemon process does not normally write to the log once it
4601 has entered its main loop, and it closes the log before doing so. This is
4602 so that log files can straightforwardly be renamed and moved. However,
4603 there are a couple of unusual error situations where the daemon does write
4604 log entries, and I had neglected to close the log afterwards.
4606 30. The text of an SMTP error response that was received during a remote
4607 delivery was being truncated at 512 bytes. This is too short for some of
4608 the long messages that one sometimes sees. I've increased the limit to
4611 31. It is now possible to make retry rules that apply only when a message has a
4612 specific sender, in particular, an empty sender.
4614 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
4615 it possible to be selective about when SMTP synchronization is enforced.
4617 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
4619 32. Implemented hosts_connection_nolog.
4621 33. Added an ACL for QUIT.
4623 34. Setting "delay_warning=" to disable warnings was not working; it gave a
4626 35. Added mailbox_size and mailbox_filecount to appendfile.
4628 36. Added control = no_multiline_responses to ACLs.
4630 37. There was a bug in the logic of the code that waits for the clock to tick
4631 in the case where the clock went backwards by a substantial amount such
4632 that the microsecond fraction of "now" was more than the microsecond
4633 fraction of "then" (but the whole seconds number was less).
4635 38. Added support for the libradius Radius client library this is found on
4636 FreeBSD (previously only the radiusclient library was supported).
4642 1. When certain lookups returned multiple values in the form name=value, the
4643 quoting of the values was not always being done properly. Specifically:
4644 (a) If the value started with a double quote, but contained no whitespace,
4646 (b) If the value contained whitespace other than a space character (i.e.
4647 tabs or newlines or carriage returns) it was not quoted.
4648 This fix has been applied to the mysql and pgsql lookups by writing a
4649 separate quoting function and calling it from the lookup code. The fix
4650 should probably also be applied to nisplus, ibase and oracle lookups, but
4651 since I cannot test any of those, I have not disturbed their existing code.
4653 2. A hit in the callout cache for a specific address caused a log line with no
4654 reason for rejecting RCPT. Now it says "Previous (cached) callout
4655 verification failure".
4657 3. There was an off-by-one bug in the queryprogram router. An over-long
4658 return line was truncated at 256 instead of 255 characters, thereby
4659 overflowing its buffer with the terminating zero. As well as fixing this, I
4660 have increased the buffer size to 1024 (and made a note to document this).
4662 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
4663 when Exim is waiting for an SMTP response from a remote server, Exim
4664 restarts its select() call on the socket, thereby resetting its timeout.
4665 This is not a problem when such interrupts are rare. Somebody set up a cron
4666 job to run exiwhat every 2 minutes, which is less than the normal select()
4667 timeout (5 or 10 minutes). This meant that the select() timeout never
4668 kicked in because it was always reset. I have fixed this by comparing the
4669 time when an interrupt arrives with the time at the start of the first call
4670 to select(). If more time than the timeout has elapsed, the interrupt is
4671 treated as a timeout.
4673 5. Some internal re-factoring in preparation for the addition of Sieve
4674 extensions (by MH). In particular, the "personal" test is moved to a
4675 separate function, and given an option for scanning Cc: and Bcc: (which is
4676 not set for Exim filters).
4678 6. When Exim created an email address using the login of the caller as the
4679 local part (e.g. when creating a From: or Sender: header line), it was not
4680 quoting the local part when it contained special characters such as @.
4682 7. Installed new OpenBSD configuration files.
4684 8. Reworded some messages for syntax errors in "and" and "or" conditions to
4685 try to make them clearer.
4687 9. Callout options, other than the timeout value, were being ignored when
4688 verifying sender addresses in header lines. For example, when using
4690 verify = header_sender/callout=no_cache
4692 the cache was (incorrectly) being used.
4694 10. Added a missing instance of ${EXE} to the exim_install script; this affects
4695 only the Cygwin environment.
4697 11. When return_path_on_delivery was set as a log selector, if different remote
4698 addresses in the same message used different return paths and parallel
4699 remote delivery occurred, the wrong values would sometimes be logged.
4700 (Whenever a remote delivery process finished, the return path value from
4701 the most recently started remote delivery process was logged.)
4703 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
4704 lines when AUTH and/or TLS are in use. This is the "received protocol"
4705 field. Exim used to use "asmtp" for authenticated SMTP, without any
4706 indication (in the protocol name) for TLS use. Now it follows the RFC and
4707 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
4708 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
4709 names appear in log lines as well as in Received: header lines.
4711 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
4712 extensions, and comparison tests, and to fix some bugs.
4714 14. Changes to the "personal" filter test:
4716 (1) The test was buggy in that it was just doing the equivalent of
4717 "contains" tests on header lines. For example, if a user's address was
4718 anne@some.where, the "personal" test would incorrectly be true for
4720 To: susanne@some.where
4722 This test is now done by extracting each address from the header in turn,
4723 and checking the entire address. Other tests that are part of "personal"
4724 are now done using regular expressions (for example, to check local parts
4725 of addresses in From: header lines).
4727 (2) The list of non-personal local parts in From: addresses has been
4728 extended to include "listserv", "majordomo", "*-request", and "owner-*",
4729 taken from the Sieve specification recommendations.
4731 (3) If the message contains any header line starting with "List-" it is
4732 treated as non-personal.
4734 (4) The test for "circular" in the Subject: header line has been removed
4735 because it now seems ill-conceived.
4737 15. Minor typos in src/EDITME comments corrected.
4739 16. Installed latest exipick from John Jetmore.
4741 17. If headers_add on a router specified a text string that was too long for
4742 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
4743 of string_sprintf() is now avoided.
4745 18. $message_body_size was not set (it was always zero) when running the DATA
4746 ACL and the local_scan() function.
4748 19. For the "mail" command in an Exim filter, no default was being set for
4749 the once_repeat time, causing a random time value to be used if "once" was
4750 specified. (If the value happened to be <= 0, no repeat happened.) The
4751 default is now 0s, meaning "never repeat". The "vacation" command was OK
4752 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
4753 (I found it when inspecting the code).
4755 20. There is now an overall timeout for performing a callout verification. It
4756 defaults to 4 times the callout timeout, which applies to individual SMTP
4757 commands during the callout. The overall timeout applies when there is more
4758 than one host that can be tried. The timeout is checked before trying the
4759 next host. This prevents very long delays if there are a large number of
4760 hosts and all are timing out (e.g. when the network connections are timing
4761 out). The value of the overall timeout can be changed by specifying an
4762 additional sub-option for "callout", called "maxwait". For example:
4764 verify = sender/callout=5s,maxwait=20s
4766 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
4767 to the end before writing, but this should make it even safer).
4769 22. Exim was forgetting that it had advertised PIPELINING for the second and
4770 subsequent messages on an SMTP connection. It was also not resetting its
4771 memory on STARTTLS and an internal HELO.
4773 23. When Exim logs an SMTP synchronization error within a session, it now
4774 records whether PIPELINING has been advertised or not.
4776 24. Added 3 instances of "(long int)" casts to time_t variables that were being
4777 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
4778 rather than long int.
4780 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
4782 26. Added the never_mail option to autoreply.
4788 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
4789 crash if the getsockname() call failed; this can happen if a connection is
4790 closed very soon after it is established. The problem was simply in the
4791 order in which certain operations were done, causing Exim to try to write
4792 to the SMTP stream before it had set up the file descriptor. The bug has
4793 been fixed by making things happen in the correct order.
4799 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
4800 before the connection was closed, thus losing the rejection response.
4802 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
4803 some early Solaris releases, but causes trouble in current releases where
4804 socklen_t is defined.
4806 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
4809 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
4812 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
4813 that releasing the top of it at the end releases what was used for sub-
4814 expansions (unless the block got too big). However, discard this block if
4815 the first thing is a variable or header, so that we can use its block when
4816 it is dynamic (useful for very large $message_headers, for example).
4818 6. Lookups now cache *every* query, not just the most recent. A new, separate
4819 store pool is used for this. It can be recovered when all lookup caches are
4820 flushed. Lookups now release memory at the end of their result strings.
4821 This has involved some general refactoring of the lookup sources.
4823 7. Some code has been added to the store_xxx() functions to reduce the amount
4824 of flapping under certain conditions.
4826 8. log_incoming_interface used to affect only the <= reception log lines. Now
4827 it causes the local interface and port to be added to several more SMTP log
4828 lines, for example "SMTP connection from", and rejection lines.
4830 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
4832 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
4834 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
4835 could be overwritten at the end of the current message (or the start of a
4836 new message if it was set in a HELO ACL). The value is now preserved for
4837 the duration of the SMTP connection.
4839 12. If a transport had a headers_rewrite setting, and a matching header line
4840 contained an unqualified address, that address was qualified, even if it
4841 did not match any rewriting rules. The underlying bug was that the values
4842 of the flags that permit the existence of unqualified sender and recipient
4843 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
4844 non-local messages, and by -bnq for local messages) were not being
4845 preserved with the message after it was received.
4847 13. When Exim was logging an SMTP synchronization error, it could sometimes log
4848 "next input=" as part of the text comprising the host identity instead of
4849 the correct text. The code was using the same buffer for two different
4850 strings. However, depending on which order the printing function evaluated
4851 its arguments, the bug did not always show up. Under Linux, for example, my
4852 test suite worked just fine.
4854 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
4855 doesn't work with some older versions of Perl. It has been changed to "my",
4856 which in any case is probably the better facility to use.
4858 15. A really picky compiler found some instances of statements for creating
4859 error messages that either had too many or two few arguments for the format
4862 16. The size of the buffer for calls to the DNS resolver has been increased
4863 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
4864 for addresses that have a lot of PTR records. This alleviates a problem; it
4865 does not fully solve it.
4867 17. A dnsdb lookup for PTR records that receives more data than will fit in the
4868 buffer now truncates the list and logs the incident, which is the same
4869 action as happens when Exim is looking up a host name and its aliases.
4870 Previously in this situation something unpredictable would happen;
4871 sometimes it was "internal error: store_reset failed".
4873 18. If a server dropped the connection unexpectedly when an Exim client was
4874 using GnuTLS and trying to read a response, the client delivery process
4875 crashed while trying to generate an error log message.
4877 19. If a "warn" verb in an ACL added multiple headers to a message in a single
4878 string, for example:
4880 warn message = H1: something\nH2: something
4882 the text was added as a single header line from Exim's point of view
4883 though it ended up OK in the delivered message. However, searching for the
4884 second and subsequent header lines using $h_h2: did not work. This has been
4885 fixed. Similarly, if a system filter added multiple headers in this way,
4886 the routers could not see them.
4888 20. Expanded the error message when iplsearch is called with an invalid key to
4889 suggest using net-iplsearch in a host list.
4891 21. When running tests using -bh, any delays imposed by "delay" modifiers in
4892 ACLs are no longer actually imposed (and a message to that effect is
4895 22. If a "gecos" field in a passwd entry contained escaped characters, in
4896 particular, if it contained a \" sequence, Exim got it wrong when building
4897 a From: or a Sender: header from that name. A second bug also caused
4898 incorrect handling when an unquoted " was present following a character
4899 that needed quoting.
4901 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
4902 was not being matched caselessly.
4904 24. Arranged for all hyphens in the exim.8 source to be escaped with
4907 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
4908 properly. Recipient callout cache records were still being keyed to include
4909 the sender, even when use_sender was set false. This led to far more
4910 callouts that were necessary. The sender is no longer included in the key
4911 when use_sender is false.
4913 26. Added "control = submission" modifier to ACLs.
4915 27. Added the ${base62d: operator to decode base 62 numbers.
4917 28. dnsdb lookups can now access SRV records.
4919 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
4920 the configuration file.
4922 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
4923 (-v) mode. This makes the output for a verbose queue run more intelligible.
4925 31. Added a use_postmaster feature to recipient callouts.
4927 32. Added the $body_zerocount variable, containing the number of binary zero
4928 bytes in the message body.
4930 33. The time of last modification of the "new" subdirectory is now used as the
4931 "mailbox time last read" when there is a quota error for a maildir
4934 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
4936 35. Added +ignore_unknown as a special item in host lists.
4938 36. Code for decoding IPv6 addresses in host lists is now included, even if
4939 IPv6 support is not being compiled. This fixes a bug in which an IPv6
4940 address was recognized as an IP address, but was then not correctly decoded
4941 into binary, causing unexpected and incorrect effects when compared with
4948 1. Very minor rewording of debugging text in manualroute to say "list of
4949 hosts" instead of "hostlist".
4951 2. If verify=header_syntax was set, and a header line with an unqualified
4952 address (no domain) and a large number of spaces between the end of the
4953 name and the colon was received, the reception process suffered a buffer
4954 overflow, and (when I tested it) crashed. This was caused by some obsolete
4955 code that should have been removed. The fix is to remove it!
4957 3. When running in the test harness, delay a bit after writing a bounce
4958 message to get a bit more predictability in the log output.
4960 4. Added a call to search_tidyup() just before forking a reception process. In
4961 theory, someone could use a lookup in the expansion of smtp_accept_max_
4962 per_host which, without the tidyup, could leave open a database connection.
4964 5. Added the variables $recipient_data and $sender_data which get set from a
4965 lookup success in an ACL "recipients" or "senders" condition, or a router
4966 "senders" option, similar to $domain_data and $local_part_data.
4968 6. Moved the writing of debug_print from before to after the "senders" test
4971 7. Change 4.31/66 (moving the time when the Received: is generated) caused
4972 problems for message scanning, either using a data ACL, or using
4973 local_scan() because the Received: header was not generated till after they
4974 were called (in order to set the time as the time of reception completion).
4975 I have revised the way this works. The header is now generated after the
4976 body is received, but before the ACL or local_scan() are called. After they
4977 are run, the timestamp in the header is updated.
4983 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
4984 before starting a queue runner without re-exec. This happened only when
4985 deliver_drop_privilege was set or when the Exim user was set to root. The
4986 effect of the bug was that timeouts during subsequent deliveries caused
4987 crashes instead of being properly handled. The handler is now left at its
4988 default (and expected) setting.
4990 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
4991 message, again when deliver_drop_privilege is set or Exim is run as root.
4992 The bug described in (1) was not present in this case, but the tidying up
4993 of the other signals was missing. I have made the two cases consistent.
4995 3. The ignore_target_hosts setting on a manualroute router was being ignored
4996 for hosts that were looked up using the /MX notation.
4998 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
5001 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
5002 operated on the sender address. After changing the $sender_address to <>
5003 for the sender address verify, Exim was re-instated it as the original
5004 (before rewriting) address, but remembering that it had rewritten it, so it
5005 wasn't rewriting it again. This bug also had the effect of breaking the
5006 sender address verification caching when the sender address was rewritten.
5008 6. The ignore_target_hosts option was being ignored by the ipliteral router.
5009 This has been changed so that if the ip literal address matches
5010 ignore_target_hosts, the router declines.
5012 7. Added expansion conditions match_domain, match_address, and match_local_
5013 part (NOT match_host).
5015 8. The placeholder for the Received: header didn't have a length field set.
5017 9. Added code to Exim itself and to exim_lock to test for a specific race
5018 condition that could lead to file corruption when using MBX delivery. The
5019 issue is with the lockfile that is created in /tmp. If this file is removed
5020 after a process has opened it but before that process has acquired a lock,
5021 there is the potential for a second process to recreate the file and also
5022 acquire a lock. This could lead to two Exim processes writing to the file
5023 at the same time. The added code performs the same test as UW imapd; it
5024 checks after acquiring the lock that its file descriptor still refers to
5025 the same named file.
5027 10. The buffer for building added header lines was of fixed size, 8192 bytes.
5028 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
5031 11. Added the smtp_active_hostname option. If used, this will typically be made
5032 to depend on the incoming interface address. Because $interface_address is
5033 not set up until the daemon has forked a reception process, error responses
5034 that can happen earlier (such as "too many connections") no longer contain
5037 12. If an expansion in a condition on a "warn" statement fails because a lookup
5038 defers, the "warn" statement is abandoned, and the next ACL statement is
5039 processed. Previously this caused the whole ACL to be aborted.
5041 13. Added the iplsearch lookup type.
5043 14. Added ident_timeout as a log selector.
5045 15. Added tls_certificate_verified as a log selector.
5047 16. Added a global option tls_require_ciphers (compare the smtp transport
5048 option of the same name). This controls incoming TLS connections.
5050 17. I finally figured out how to make tls_require_ciphers do a similar thing
5051 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
5052 before starting the TLS session.
5054 18. Tabs are now shown as \t in -bP output.
5056 19. If the log selector return_path_on_delivery was set, Exim crashed when
5057 bouncing a message because it had too many Received: header lines.
5059 20. If two routers both had headers_remove settings, and the first one included
5060 a superfluous trailing colon, the final name in the first list and the
5061 first name in the second list were incorrectly joined into one item (with a
5062 colon in the middle).
5068 1. Added -C and -D options to the exinext utility, mainly to make it easier
5069 to include in the automated testing, but these could be helpful when
5070 multiple configurations are in use.
5072 2. The exinext utility was not formatting the output nicely when there was
5073 an alternate port involved in the retry record key, nor when there was a
5074 message id as well (for retries that were specific to a specific message
5075 and a specific host). It was also confused by IPv6 addresses, because of
5076 the additional colons they contain. I have fixed the IPv4 problem, and
5077 patched it up to do a reasonable job for IPv6.
5079 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
5080 delivery, the log line now contains "pipelined" if PIPELINING was used.
5082 4. An SMTP transport process used to panic and die if the bind() call to set
5083 an explicit outgoing interface failed. This has been changed; it is now
5084 treated in the same way as a connect() failure.
5086 5. A reference to $sender_host_name in the part of a conditional expansion
5087 that was being skipped was still causing a DNS lookup. This no longer
5090 6. The def: expansion condition was not recognizing references to header lines
5091 that used bh_ and bheader_.
5093 7. Added the _cache feature to named lists.
5095 8. The code for checking quota_filecount in the appendfile transport was
5096 allowing one more file than it should have been.
5098 9. For compatibility with Sendmail, the command line option
5106 and sets the incoming protocol and host name (for trusted callers). The
5107 host name and its colon can be omitted when only the protocol is to be set.
5108 Note the Exim already has two private options, -pd and -ps, that refer to
5109 embedded Perl. It is therefore impossible to set a protocol value of "d" or
5110 "s", but I don't think that's a major issue.
5112 10. A number of refactoring changes to the code, none of which should affect
5115 (a) The number of logging options was getting close to filling up the
5116 32-bit word that was used as a bit map. I have split them into two classes:
5117 those that are passed in the argument to log_write(), and those that are
5118 only ever tested independently outside of that function. These are now in
5119 separate 32-bit words, so there is plenty of room for expansion again.
5120 There is no change in the user interface or the logging behaviour.
5122 (b) When building, for example, log lines, the code previously used a
5123 macro that called string_cat() twice, in order to add two strings. This is
5124 not really sufficiently general. Furthermore, there was one instance where
5125 it was actually wrong because one of the argument was used twice, and in
5126 one call a function was used. (As it happened, calling the function twice
5127 did not affect the overall behaviour.) The macro has been replaced by a
5128 function that can join an arbitrary number of extra strings onto a growing
5131 (c) The code for expansion conditions now uses a table and a binary chop
5132 instead of a serial search (which was left over from when there were very
5133 few conditions). Also, it now recognizes conditions like "pam" even when
5134 the relevant support is not compiled in: a suitably worded error message is
5135 given if an attempt is made to use such a condition.
5137 11. Added ${time_interval:xxxxx}.
5139 12. A bug was causing one of the ddress fields not to be passed back correctly
5140 from remote delivery subprocesses. The field in question was not being
5141 subsequently used, so this caused to problems in practice.
5143 13. Added new log selectors queue_time and deliver_time.
5145 14. Might have fixed a bug in maildirsizefile handling that threw up
5146 "unexpected character" debug warnings, and recalculated the data
5147 unnecessarily. In any case, I expanded the warning message to give more
5150 15. Added the message "Restricted characters in address" to the statements in
5151 the default ACL that block characters like @ and % in local parts.
5153 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
5154 Three changes have been made:
5156 (a) There was a serious bug; a negative response to MAIL caused the whole
5157 recipient domain to be cached as invalid, thereby blocking all messages
5158 to all local parts at the same domain, from all senders. This bug has
5159 been fixed. The domain is no longer cached after a negative response to
5160 MAIL if the sender used is not empty.
5162 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
5165 (c) A new callout option, "use_sender" has been added for people who want
5166 the modified behaviour.
5172 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
5175 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
5176 indeed breaks things for older releases.
5178 3. Added additional logging to the case where there is a problem reading data
5179 from a filter that is running in a subprocess using a pipe, in order to
5180 try to track down a specific problem.
5182 4. Testing facility fudge: when running in the test harness and attempting
5183 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
5184 getting "No route to host". Convert this to a timeout.
5186 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
5189 6. Some OS don't have socklen_t but use size_t instead. This affects the
5190 fifth argument of getsockopt() amongst other things. This is now
5191 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
5192 can be set for individual OS. I have set it for SunOS5, OSF1, and
5193 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
5194 some earlier ones do not.
5196 7. Change 4.30/15 was not doing the test caselessly.
5198 8. The standard form for an IPv6 address literal was being rejected by address
5199 parsing in, for example, MAIL and RCPT commands. An example of this kind of
5200 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5201 this, as well as the form without the "IPv6" on the front (but only when
5202 address literals are enabled, of course).
5204 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
5206 10. Exim crashed if a message with an empty sender address specified by -f
5207 encountered a router with an errors_to setting. This could be provoked only
5208 by a command such as
5212 where an empty string was supplied; "<>" did not hit this bug.
5214 11. Installed PCRE release 4.5.
5216 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
5217 remained set. It is now erased.
5219 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
5220 times from message ids (which are base 36 rather than the normal 62).
5222 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
5223 were being counted as actual protocol errors, and logged if the log
5224 selector +smtp_protocol_error was set. One cannot be perfect in this test,
5225 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
5226 and DATA following a set of rejected RCPTs do not count as protocol errors.
5227 In other words, Exim assumes they were pipelined, though this may not
5228 actually be the case. Of course, in all cases the client gets an
5229 appropriate error code.
5231 15. If a lookup fails in an ACL condition, a message about the failure may
5232 be available; it is used if testing the ACL cannot continue, because most
5233 such messages specify what the cause of the deferral is. However, some
5234 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
5235 that caused an old message to be retained and used if a later statement
5236 caused a defer, replacing the real cause of the deferral.
5238 16. If an IP address had so many PTR records that the DNS lookup buffer
5239 was not large enough to hold them, Exim could crash while trying to process
5240 the truncated data. It now detects and logs this case.
5242 17. Further to 4.21/58, another change has been made: if (and only if) the
5243 first line of a message (the first header line) ends with CRLF, a bare LF
5244 in a subsequent header line has a space inserted after it, so as not to
5245 terminate the header.
5247 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
5248 unnecessarily, used atoi() instead of strtol(), and didn't check the
5249 termination when getting file sizes from file names by regex.
5251 19. Completely re-implemented the support for maildirsize files, in the light
5252 of a number of problems with the previous contributed implementation
5253 (4.30/29). In particular:
5255 . If the quota is zero, the maildirsize file is maintained, but no quota is
5258 . If the maildir directory does not exist, it is created before any attempt
5259 to write a maildirsize file.
5261 . The quota value in the file is just a cache; if the quota is changed in
5262 the transport, the new value overrides.
5264 . A regular expression is available for excluding directories from the
5267 20. The autoreply transport checks the characters in options that define the
5268 message's headers; it allows continued headers, but it was checking with
5269 isspace() after an embedded newline instead of explicitly looking for a
5272 21. If all the "regular" hosts to which an address was routed had passed their
5273 expiry times, and had not reached their retry times, the address was
5274 bounced, even if fallback hosts were defined. Now Exim should go on to try
5277 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
5278 equivalent code in the SMTP transport. Some hosts send humungous responses
5279 to HELO/EHLO, more than 1024 it seems.
5281 23. Refactoring: code in filter.c used (void *) for "any old type" but this
5282 gives compiler warnings in some environments. I've now done it "properly",
5285 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
5286 (because of problems with the built-in one) was declared to return uschar *
5287 instead of char *, causing compiler failure.
5289 25. Fixed a file descriptor leak when processing alias/forward files.
5291 26. Fixed a minor format string issue in dbfn.c.
5293 27. Typo in exim.c: ("dmbnz" for "dbmnz").
5295 28. If a filter file refered to $h_xxx or $message_headers, and the headers
5296 contained RFC 2047 "words", Exim's memory could, under certain conditions,
5299 29. When a sender address is verified, it is cached, to save repeating the test
5300 when there is more than one recipient in a message. However, when the
5301 verification involves a callout, it is possible for different callout
5302 options to be set for different recipients. It is too complicated to keep
5303 track of this in the cache, so now Exim always runs a verification when a
5304 callout is required, relying on the callout cache for the optimization.
5305 The overhead is duplication of the address routing, but this should not be
5308 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
5309 to be verified with callout=postmaster, and the main callout worked but the
5310 postmaster check failed, the verification correctly failed. However, if a
5311 subsequent RCPT command asked for sender verification *without* the
5312 postmaster check, incorrect caching caused this verification also to fail,
5315 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
5316 it was not caching the DNS options (qualify_single, search_parents) that
5317 were used when the lookup failed. A subsequent lookup with different
5318 options therefore always gave the same answer, though there were cases
5319 where it should not have. (Example: a "domains = !$mx_any" option on a
5320 dnslookup router: the "domains" option is always processed without any
5321 widening, but the router might have qualify_single set.) Now Exim uses the
5322 cached value only when the same options are set.
5324 32. Added John Jetmore's "exipick" utility to the distribution.
5326 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
5327 than a timeout (e.g. a certificate is required, and is not provided), an
5328 Exim server now closes the connection immediately. Previously it waited for
5329 the client to close - but if the client is SSL, it seems that they each
5330 wait for each other, leading to a delay before one of them times out.
5332 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
5333 maintained 0.8.x compatibility because I don't think many are using it, and
5334 it is clearly obsolete.
5336 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
5339 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
5340 client certificate was expired. A simple patch fixes this, though I don't
5341 understand the full logic of why the verify callback is called multiple
5344 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
5345 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
5346 which causes problems with some clients (such as the Certicom SSL Plus
5347 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
5348 disables the coutermeasure allowing Eudora to connect."
5350 38. Exim was not checking that a write() to a log file succeeded. This could
5351 lead to Bad Things if a log got too big, in particular if it hit a file
5352 size limit. Exim now panics and dies if it cannot write to a log file, just
5353 as it does if it cannot open a log file.
5355 39. Modified OS/Makefile-Linux so that it now contains
5357 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
5359 The two -D definitions ensure that Exim is compiled with large file
5360 support, which makes it possible to handle log files that are bigger than
5363 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
5364 instance) a domain was checked against a named list that involved a lookup,
5365 causing $domain_data to be set, then another domain was checked against the
5366 same list, then the first domain was re-checked, the value of $domain_data
5367 after the final check could be wrong. In particular, if the second check
5368 failed, it could be set empty. This bug probably also applied to
5371 41. The strip_trailing_dot option was not being applied to the address given
5372 with the -f command-line option.
5374 42. The code for reading a message's header from the spool was incrementing
5375 $received_count, but never initializing it. This meant that the value was
5376 incorrect (doubled) while delivering a message in the same process in which
5377 it was received. In the most common configuration of Exim, this never
5378 happens - a fresh exec is done - but it can happen when
5379 deliver_drop_privilege is set.
5381 43. When Exim logs an SMTP synchronization error - client data sent too soon -
5382 it now includes up to 150 characters of the unexpected data in the log
5385 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
5386 and building data strings. The size of both of these buffers was 10 000
5387 bytes - far larger than anybody would *ever* want, thought I. Needless to
5388 say, somebody hit the limit. I have increased the maximum line length to
5389 20 000 and the maximum data length of concatenated lines to 100 000. I have
5390 also fixed two bugs, because there was no checking on these buffers. Tsk,
5391 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
5392 buffer is too small.
5394 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
5395 lsearch lookups. Now it does.
5397 46. When parsing a route_list item in a manualroute router, a fixed-length
5398 buffer was used for the list of hosts. I made this 1024 bytes long,
5399 thinking that nobody would ever have a list of hosts that long. Wrong.
5400 Somebody had a whole pile of complicated expansion conditions, and the
5401 string was silently truncated, leading to an expansion error. It turns out
5402 that it is easier to change to an unlimited length (owing to other changes
5403 that have happened since this code was originally written) than to build
5404 structure for giving a limitation error. The length of the item that
5405 expands into the list of hosts is now unlimited.
5407 47. The lsearch lookup could not handle data where the length of text line was
5408 more than 4095 characters. Such lines were truncated, leading to shortened
5409 data being returned. It should now handle lines of any length.
5411 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
5412 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
5415 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
5416 replacing strings such as BIN_DIRECTORY in a source file: the replacement
5417 no longer happens in comment lines. A list of replacements is now placed
5418 at the head of all of the source files, except those whose only change is
5419 to replace PERL_COMMAND in the very first #! line.
5421 50. Replaced the slow insertion sort in queue.c, for sorting the list of
5422 messages on the queue, with a bottom-up merge sort, using code contributed
5423 by Michael Haardt. This should make operations like -bp somewhat faster on
5424 large queues. It won't affect queue runners, except when queue_run_in_order
5427 51. Installed eximstats 1.31 in the distribution.
5429 52. Added support for SRV lookups to the dnslookup router.
5431 53. If an ACL referred to $message_body or $message_body_end, the value was not
5432 reset for any messages that followed in the same SMTP session.
5434 54. The store-handling optimization for building very long strings was not
5435 differentiating between the different store pools. I don't think this
5436 actually made any difference in practice, but I've tidied it.
5438 55. While running the routers to verify a sender address, $sender_address
5439 was still set to the sender address. This is wrong, because when routing to
5440 send a bounce to the sender, it would be empty. Therefore, I have changed
5441 it so that, while verifying a sender address, $sender_address is set to <>.
5442 (There is no change to what happens when verifying a recipient address.)
5444 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
5445 target A or AAAA records (if not already returned) without resetting the
5446 qualify_single or search_parents options of the DNS resolver. These are
5447 inappropriate in this case because the targets of MX and SRV records must
5448 be FQDNs. A broken DNS record could cause trouble if it happened to have a
5449 target that, when qualified, matched something in the local domain. These
5450 two options are now turned off when doing these lookups.
5452 57. It seems that at least some releases of Reiserfs (which does not have the
5453 concept of a fixed number of inodes) returns zero and not -1 for the
5454 number of available inodes. This interacted badly with check_spool_inodes,
5455 which assumed that -1 was the "no such thing" setting. What I have done is
5456 to check that the total number of inodes is greater than zero before doing
5457 the test of how many are available.
5459 58. When a "warn" ACL statement has a log_message modifier, the message is
5460 remembered, and not repeated. This is to avoid a lot of repetition when a
5461 message has many recipients that cause the same warning to be written.
5462 Howewer, Exim was preserving the list of already written lines for an
5463 entire SMTP session, which doesn't seem right. The memory is now reset if a
5464 new message is started.
5466 59. The "rewrite" debugging flag was not showing the result of rewriting in the
5467 debugging output unless log_rewrite was also set.
5469 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
5470 of (int)(handle) when we know that handle contains (void *)(-1).
5472 61. The Exim daemon panic-logs an error return when it closes the incoming
5473 connection. However "connection reset by peer" seems to be common, and
5474 isn't really an error worthy of noting specially, so that particular error
5477 62. When Exim is trying to find all the local interfaces, it used to panic and
5478 die if the ioctl to get the interface flags failed. However, it seems that
5479 on at least one OS (Solaris 9) it is possible to have an interface that is
5480 included in the list of interfaces, but for which you get a failure error
5481 for this call. This happens when the interface is not "plumbed" into a
5482 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
5483 failure of the "get flags" call assumes that the interface is down.
5485 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
5486 makes life easier for people who are doing arithmetic on fields extracted
5487 from dates, where you often get leading zeros that should not be
5488 interpreted as octal.
5490 64. Added qualify_domain to the redirect router, to override the global
5493 65. If a pathologically long header line contained very many addresses (the
5494 report of this problem mentioned 10 000) and each of them was rewritten,
5495 Exim could use up a very large amount of memory. (It kept on making new
5496 copies of the header line as it rewrote, and never released the old ones.)
5497 At the expense of a bit more processing, the header rewriting function has
5498 been changed so that it no longer eats memory in this way.
5500 66. The generation of the Received: header has been moved from the time that a
5501 message starts to be received, to the time that it finishes. The timestamp
5502 in the Received: header should now be very close to that of the <= log
5503 line. There are two side-effects of this change:
5505 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
5506 logged header lines no longer include the local Received: line, because
5507 it has not yet been created. The same applies to a copy of the message
5508 that is returned to a non-SMTP sender when a message is rejected.
5510 (b) When a filter file is tested using -bf, no additional Received: header
5511 is added to the test message. After some thought, I decided that this
5514 This change does not affect the value of $received_for. It is still set
5515 after address rewriting, but before local_scan() is called.
5517 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
5519 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
5520 gave an unhelpful panic error message, and a defer error. I have managed to
5521 change this behaviour so that it now rejects any supplied certificate,
5522 which seems right, as the list of acceptable certificates is empty.
5524 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
5525 gave an unhelpful defer error. I have not managed to make this reject any
5526 supplied certificates, but the error message it gives is "no certificate
5527 supplied", which is not helpful.
5529 70. exigrep's output now also includes lines that are not associated with any
5530 message, but which match the given pattern. Implemented by a patch from
5531 Martin Sluka, which also tidied up the Perl a bit.
5533 71. Recipient callout verification, like sender verification, was using <> in
5534 the MAIL FROM command. This isn't really the right thing, since the actual
5535 sender may affect whether the remote host accepts the recipient or not. I
5536 have changed it to use the actual sender in the callout; this means that
5537 the cache record is now keyed on a recipient/sender pair, not just the
5538 recipient address. There doesn't seem to be a real danger of callout loops,
5539 since a callout by the remote host to check the sender would use <>.
5540 [SEE ABOVE: changed after hitting problems.]
5542 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
5543 temporary errors. However, in the case of such a code being given after
5544 the end of a data transmission (i.e. after ".") Exim was failing to write
5545 a retry record for the message. (Yes, there was some broken host that was
5546 actually sending 8xx at this point.)
5548 73. An unknown lookup type in a host list could cause Exim to panic-die when
5549 the list was checked. (An example that provoked this was putting <; in the
5550 middle of a list instead of at the start.) If this happened during a DATA
5551 ACL check, a -D file could be left lying around. This kind of configuration
5552 error no longer causes Exim to die; instead it causes a defer errror. The
5553 incident is still logged to the main and panic logs.
5555 74. Buglet left over from Exim 3 conversion. The message "too many messages
5556 in one connection" was written to the rejectlog but not the mainlog, except
5557 when address rewriting (yes!) was being logged.
5559 75. Added write_rejectlog option.
5561 76. When a system filter was run not as root (that is, when system_filter_user
5562 was set), the values of the $n variables were not being returned to the
5563 main process; thus, they were not subsequently available in the $sn
5566 77. Added +return_path_on_delivery log selector.
5568 78. A connection timeout was being treated differently from recipients deferred
5569 when testing hosts_max_try with a message that was older than the host's
5570 retry timeout. (The host should not be counted, thus allowing all hosts to
5571 be tried at least once before bouncing.) This may have been the cause of an
5572 occasionally reported bug whereby a message would remain on the queue
5573 longer than the retry timeout, but would be bounced if a delivery was
5574 forced. I say "may" because I never totally pinned down the problem;
5575 setting up timeout/retry tests is difficult. See also the next item.
5577 79. The ultimate address timeout was not being applied to errors that involved
5578 a combination of host plus message (for example, a timeout on a MAIL
5579 command). When an address resolved to a number of possible hosts, and they
5580 were not all tried for each delivery (e.g. because of hosts_max_try), a
5581 message could remain on the queue longer than the retry timeout.
5583 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
5586 81. Fixed an obscure SMTP outgoing bug which required at least the following
5587 conditions: (a) there was another message waiting for the same server;
5588 (b) the server returned 5xx to all RCPT commands in the first message so
5589 that the message was not completed; (c) the server dropped the connection
5590 or gave a negative response to the RSET that Exim sends to abort the
5591 transaction. The observed case was a dropped connection after DATA that had
5592 been sent in pipelining mode. That is, the server had advertised PIPELINING
5593 but was not implementing it correctly. The effect of the bug was incorrect
5594 behaviour, such as trying another host, and this could lead to a crash.
5600 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
5601 and daemon.c were passed as pointers to ints; they should have been
5602 pointers to socklen_t variables (which are typically unsigned ints).
5604 2. Some signed/unsigned type warnings in the os.c file for Linux have been
5607 3. Fixed a really odd bug that affected only the testing scheme; patching a
5608 certain fixed string in the binary changed the value of another string that
5609 happened to be identical to the end of the original first string.
5611 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
5612 name", it returns that address as the IP address. On some operating
5613 systems (e.g. Solaris), it also passes back the IP address string as the
5614 "host name". However, on others (e.g. Linux), it passes back an empty
5615 string. Exim wasn't checking for this, and was changing the host name to an
5616 empty string, assuming it had been canonicized.
5618 5. Although rare, it is permitted to have more than one PTR record for a given
5619 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
5620 all the names associated with an address, because they do in Solaris.
5621 However, it seems that they do not in Linux for data that comes from the
5622 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
5623 I found this out when I moved to a new Linux workstation and tried to run
5624 the Exim test suite.
5626 To get round this problem I have changed the code so that it now does its
5627 own call to the DNS to look up PTR records when searching for a host name.
5628 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
5629 addresses that are only in /etc/hosts are still found.
5631 This behaviour is, however, controlled by an option called host_lookup_
5632 order, which defaults to "bydns:byaddr". If people want to use the other
5633 order, or indeed, just use one or the other means of lookup, they can
5634 specify it in this variable.
5636 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
5637 some operating systems, this comes back from gethostbyaddr() as an empty
5638 string, and this is what Exim used to test for. However, it seems that in
5639 other systems, "." is yielded. Exim now tests for this case too.
5641 7. The values of check_spool_space and check_log_space are now held internally
5642 as a number of kilobytes instead of an absolute number of bytes. If a
5643 numbers is specified without 'K' or 'M', it is rounded up to the nearest
5644 kilobyte. This means that much larger values can be stored.
5646 8. Exim monitor: an attempt to get the action menu when not actually pointing
5647 at a message produces an empty menu entitled "No message selected". This
5648 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
5649 no entries in it ("Shell widget menu has zero width and/or height"). So I
5650 have added a single, blank menu entry in this case.
5652 9. Added ${quote_local_part.
5654 10. MIME decoding is now applied to the contents of Subject: header lines when
5657 11. Now that a reference to $sender_host_address automatically causes a reverse
5658 lookup to occur if necessary (4.13/18), there is no need to arrange for a
5659 host lookup before query-style lookups in lists that might use this
5660 variable. This has therefore been abolished, and the "net-" prefix is no
5661 longer necessary for query-style lookups.
5663 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
5664 have been a typo for LFLAGS, so it has been changed.
5666 13. The install script calls Exim with "-C /dev/null" in order to find the
5667 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
5668 to be output. Howeve, since Exim outputs its version number before the
5669 error, it didn't break the script. It just looked ugly. I fixed this by
5670 always allowing "-C /dev/null" if the caller is root.
5672 14. Ignore overlarge ACL variable number when reading spool file - insurance
5673 against a later release with more variables having written the file.
5675 15. The standard form for an IPv6 address literal was being rejected by EHLO.
5676 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5677 this, as well as the form without the "IPv6" on the front.
5679 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
5680 OS/Makefile-Darwin file.
5682 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
5683 only to LDAP libraries that do not have LDAP_OPT_DEREF.
5685 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
5686 variable. However, some OS use ints for inodes. I've added cast to long int
5687 to get rid of the compiler warning.
5689 19. I had forgotten to lock out "/../" in configuration file names when
5690 ALT_CONFIG_PREFIX was set.
5692 20. Routers used for verification do not need to specify transports. However,
5693 if such a router generated a host list, and callout was configured, Exim
5694 crashed, because it could not find a port number from the (non-existent)
5695 transport. It now assumes port 25 in this circumstance.
5697 21. Added the -t option to exigrep.
5699 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
5700 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
5701 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
5702 them is set, LOOKUP_LSEARCH is forced.
5704 23. "exim -bV" now outputs a list of lookups that are included in the binary.
5706 24. Added sender and host information to the "rejected by local_scan()" log
5707 line; previously there was no indication of these.
5709 25. Added .include_if_exists.
5711 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
5712 Linux. It turns out that not all file systems support this. Apparently some
5713 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
5714 do it.) To cope with this, the error EINVAL, which means that sync-ing is
5715 not supported on the file descriptor, is now ignored when Exim is trying to
5716 sync a directory. This applies only to Linux.
5718 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
5720 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
5721 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
5724 29. Added support for maildirsize files from supplied patch (modified a bit).
5726 30. The use of :fail: followed by an empty string could lead Exim to respond to
5727 sender verification failures with (e.g.):
5729 550 Verification failed for <xxx>
5730 550 Sender verify failed
5732 where the first response line was missing the '-' that indicates it is not
5733 the final line of the response.
5735 31. The loop for finding the name of the user that called Exim had a hardwired
5736 limit of 10; it now uses the value of finduser_retries, which is used for
5737 all other user lookups.
5739 32. Added $received_count variable, available in data and not_smtp ACLs, and at
5742 33. Exim was neglecting to zero errno before one call of strtol() when
5743 expanding a string and expecting an integer value. On some systems this
5744 resulted in spurious "integer overflow" errors. Also, it was casting the
5745 result into an int without checking.
5747 34. Testing for a connection timeout using "timeout_connect" in the retry rules
5748 did not work. The code looks as if it has *never* worked, though it appears
5749 to have been documented since at least releast 1.62. I have made it work.
5751 35. The "timeout_DNS" error in retry rules, also documented since at least
5752 1.62, also never worked. As it isn't clear exactly what this means, and
5753 clearly it isn't a major issue, I have abolished the feature by treating it
5754 as "timeout", and writing a warning to the main and panic logs.
5756 36. The display of retry rules for -brt wasn't always showing the error code
5759 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
5760 timeout_connect_A, timeout_connect_MX.
5762 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
5763 to the empty sender.
5765 39. The daemon was not analysing the content of -oX till after it had closed
5766 stderr and disconnected from the controlling terminal. This meant that any
5767 syntax errors were only noted on the panic log, and the return code from
5768 the command was 0. By re-arranging the code a little, I've made the
5769 decoding happen first, so such errors now appear on stderr, and the return
5770 code is 1. However, the actual setting up of the sockets still happens in
5771 the disconnected process, so errors there are still only recorded on the
5774 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
5775 connections (as happens on some IP stacks) was logged at start up time as
5776 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
5777 it from "IPv6 and IPv4", which means that two separate sockets are being
5780 41. The debug output for gethostbyname2() or getipnodebyname() failures now
5781 says whether AF_INET or AF_INET6 was passed as an argument.
5783 42. Exiwhat output was messed up when time zones were included in log
5786 43. Exiwhat now gives more information about the daemon's listening ports,
5787 and whether -tls-on-connect was used.
5789 44. The "port" option of the smtp transport is now expanded.
5791 45. A "message" modifier in a "warn" statement in a non-message ACL was being
5792 silently ignored. Now an error message is written to the main and panic
5795 46. There's a new ACL modifier called "logwrite" which writes to a log file
5796 as soon as it is encountered.
5798 47. Added $local_user_uid and $local_user_gid at routing time.
5800 48. Exim crashed when trying to verify a sender address that was being
5803 49. Exim was recognizing only a space character after ".include". It now also
5804 recognizes a tab character.
5806 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
5807 extracting the relevant information from the specification. The man page no
5808 longer contains scrambled data for the -d option, and I've added a section
5809 at the front about calling Exim under different names.
5811 51. Added "extra_headers" argument to the "mail" command in filter files.
5813 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
5816 53. Installed eximstats 1.29.
5818 54. Added transport_filter_timeout as a generic transport option.
5820 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
5821 Cc: header lines. This was required by RFC 822, but it not required by RFC
5824 56. Exim used to add From:, Date:, and Message-Id: header lines to any
5825 incoming messages that did not have them. Now it does so only if the
5826 message originates locally, that is, if there is no associated remote host
5827 address. When Resent- header lines are present, this applies to the Resent-
5828 lines rather than the non-Resent- lines.
5830 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
5831 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
5833 58. Messages for configuration errors now include the name of the main
5834 configuration file - useful now that there may be more than one file in a
5835 list (.included file names were always shown).
5837 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
5838 for those rare installations that do not start the daemon as root or run it
5839 setuid root. I've cut out the call to initgroups() if the daemon is not
5842 60. The Exim user and group can now be bound into the binary as text strings
5843 that are looked up at the start of Exim's processing.
5845 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
5847 62. Added $mailstore_basename variable.
5849 63. Installed patch to sieve.c from Michael Haardt.
5851 64. When Exim failed to open the panic log after failing to open the main log,
5852 the original message it was trying to log was written to stderr and debug
5853 output, but if they were not available (the usual case in production), it
5854 was lost. Now it is written to syslog before the two lines that record the
5855 failures to open the logs.
5857 65. Users' Exim filters run in subprocesses under the user's uid. It is
5858 possible for a "deliver" command or an alias in a "personal" command to
5859 provoke an address rewrite. If logging of address rewriting is configured,
5860 this fails because the process is not running as root or exim. There may be
5861 a better way of dealing with this, but for the moment (because 4.30 needs
5862 to be released), I have disabled address rewrite logging when running a
5863 filter in a non-root, non-exim process.
5869 1. The buildconfig auxiliary program wasn't quoting the value set for
5870 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
5871 not defined. This bug was masked in 4.22 by the effect that was fixed in
5874 2. Some messages that were rejected after a message id was allocated were
5875 shown as "incomplete" by exigrep. It no longer does this for messages that
5876 are rejected by local_scan() or the DATA or non-SMTP ACLs.
5878 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
5879 have allow_domain_literals set, the ID did not get logged in the <= line.
5880 Domain literals are now always recognized in Message-ID: header lines.
5882 4. The first argument for a ${extract expansion item is the key name or field
5883 number. Leading and trailing spaces in this item were not being ignored,
5884 causing some misleading effects.
5886 5. When deliver_drop_privilege was set, single queue runner processes started
5887 manually (i.e. by the command "exim -q") or by the daemon (which uses the
5888 same command in the process it spins off) were not dropping privilege.
5890 6. When the daemon running as "exim" started a queue runner, it always
5891 re-executed Exim in the spun-off process. This is a waste of effort when
5892 deliver_drop_privilege is set. The new process now just calls the
5893 queue-runner function directly.
5899 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
5902 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
5903 ignored. Though the use of -oP was forcing the writing of a pid file, it
5904 was always written to the default place.
5906 3. If the message "no IP address found for host xxxx" is generated during
5907 incoming verification, it is now followed by identification of the incoming
5908 connection (so you can more easily find what provoked it).
5910 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
5912 5. Added some features to "harden" Exim a bit more against certain attacks:
5914 (a) There is now a build-time option called FIXED_NEVER_USERS that can
5915 be put in Local/Makefile. This is like the never_users runtime option,
5916 but it cannot be overridden. The default setting is "root".
5918 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
5919 prefix string with which any file named in a -C command line option
5922 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
5923 is retained for -C and -D only if the caller of Exim is root. Without
5924 it, the exim user may also use -C and -D and retain privilege.
5926 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
5927 command line option is disabled.
5929 6. Macro names set by the -D option must start with an upper case letter, just
5930 like macro names defined in the configuration file.
5932 7. Added "dereference=" facility to LDAP.
5934 8. Two instances of the typo "uknown" in the source files are fixed.
5936 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
5937 the Configure-Makefile script screwed up while processing it.
5939 10. Incorporated PCRE 4.4.
5941 11. The SMTP synchronization check was not operating right at the start of an
5942 SMTP session. For example, it could not catch a HELO sent before the client
5943 waited for the greeting. There is now a check for outstanding input at the
5944 point when the greeting is written. Because of the duplex, asynchronous
5945 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
5946 way, but not yet received, when the check is performed.
5948 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
5949 on TCP/IP sockets, because this apparently causes some broken clients to
5952 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
5953 unchanged) from the Cygwin maintainer.
5955 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
5956 was supported instead of testing for mbx. Effectively a typo.
5958 15. The spa authenticator server code was not checking that the input it
5959 received was valid base64.
5961 16. The debug output line for the "set" modifier in ACLs was not showing the
5962 name of the variable that was being set.
5964 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
5966 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
5967 lookup on its own. Something else was needed to trigger the lookup. For
5968 example, a match in host_lookup or the need for a host name in a host list.
5969 Now, if $sender_host_name is referenced and the host name has not yet been
5970 looked up, a lookup is performed. If the lookup fails, the variable remains
5971 empty, and $host_lookup_failed is set to "1".
5973 19. Added "eqi" as a case-independent comparison operator.
5975 20. The saslauthd authentication condition could segfault if neither service
5976 nor realm was specified.
5978 21. If an overflowing value such as "2048M" was set for message_size_limit, the
5979 error message that was logged was misleading, and incoming SMTP
5980 connections were dropped. The message is now more accurate, and temporary
5981 errors are given to SMTP connections.
5983 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
5984 (except RSET) with a 421 error, until QUIT is received. However, it was
5985 failing to send a response to QUIT.
5987 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
5988 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
5989 ACL is now run after the helo_try_verify_hosts code.
5991 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
5992 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
5993 case-independent, so other case variants are also recognized). Apparently
5994 some systems use these upper case variants.
5996 25. If more than two messages were waiting for the same host, and a transport
5997 filter was specified for the transport, Exim sent two messages over the
5998 same TCP/IP connection, and then failed with "socket operation on non-
5999 socket" when it tried to send the third.
6001 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
6003 27. The extern definition of crypt16() in expand.c was not being excluded when
6004 the OS had its own crypt16() function.
6006 28. Added bounce_return_body as a new option, and bounce_return_size_limit
6007 as a preferred synonym for return_size_limit, both as an option and as an
6010 29. Added LIBS=-liconv to OS/Makefile-OSF1.
6012 30. Changed the default configuration ACL to relax the local part checking rule
6013 for addresses that are not in any local domains. For these addresses,
6014 slashes and pipe symbols are allowed within local parts, but the sequence
6015 /../ is explicitly forbidden.
6017 31. SPA server authentication was not clearing the challenge buffer before
6020 32. log_message in a "warn" ACL statement was writing to the reject log as
6021 well as to the main log, which contradicts the documentation and doesn't
6022 seem right (because no rejection is happening). So I have stopped it.
6024 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
6025 However, I am unable to do any testing of this.
6027 34. Fixed an infelicity in the appendfile transport. When checking directories
6028 for a mailbox, to see if any needed to be created, it was accidentally
6029 using path names with one or more superfluous leading slashes; tracing
6030 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
6032 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
6033 contents are added to the log line that is written for every discarded
6034 recipient. (Previously a log_message setting was ignored.)
6036 36. The ${quote: operator now quotes the string if it is empty.
6038 37. The install script runs exim in order to find its version number. If for
6039 some reason other than non-existence or emptiness, which it checks, it
6040 could not run './exim', it was installing it with an empty version number,
6041 i.e. as "exim-". This error state is now caught, and the installation is
6044 38. An argument was missing from the function that creates an error message
6045 when Exim fails to connect to the socket for saslauthd authentication.
6046 This could cause Exim to crash, or give a corrupted message.
6048 39. Added isip, isip4, and isip6 to ${if conditions.
6050 40. The ACL variables $acl_xx are now saved with the message, and can be
6051 accessed later in routers, transports, and filters.
6053 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
6054 strings in the file are not string-expanded.
6056 42. If a MAIL command specified a SIZE value that was too large to fit into an
6057 int variable, the check against message_size_limit failed. Such values are
6058 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
6059 day this will have to be increased, but I don't think I want to be around
6060 when emails are that large.
6067 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
6068 iconv() is not standard in FreeBSD.
6070 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
6071 IPv6 enabled. The observed symptom was a segmentation fault on return from
6072 the function os_common_find_running_interfaces() in src/os.c.
6074 3. In the check_special_case() function in daemon.c I had used "errno" as an
6075 argument name, which causes warnings on some systems. This was basically a
6076 typo, since it was named "eno" in the comments!
6078 4. The code that waits for the clock to tick (at a resolution of some fraction
6079 of a second) so as to ensure message-id uniqueness was always waiting for
6080 at least one whole tick, when it could have waited for less. [This is
6081 almost certainly not relevant at current processor speeds, where it is
6082 unlikely to ever wait at all. But we try to future-proof.]
6084 5. The function that sleeps for a time interval that includes fractions of a
6085 second contained a race. It did not block SIGALRM between setting the
6086 timer, and suspending (a couple of lines later). If the interval was short
6087 and the sigsuspend() was delayed until after it had expired, the suspension
6088 never ended. On busy systems this could lead to processes getting stuck for
6091 6. Some uncommon configurations may cause a lookup to happen in a queue runner
6092 process, before it forks any delivery processes. The open lookup caching
6093 mechanism meant that the open file or database connection was passed into
6094 the delivery process. The problem was that delivery processes always tidy
6095 up cached lookup data. This could cause a problem for the next delivery
6096 process started by the queue runner, because the external queue runner
6097 process does not know about the closure. So the next delivery process
6098 still has data in the lookup cache. In the case of a file lookup, there was
6099 no problem because closing a file descriptor in a subprocess doesn't affect
6100 the parent. However, if the lookup was caching a connection to a database,
6101 the connection was closed, and the second delivery process was likely to
6102 see errors such as "PGSQL: query failed: server closed the connection
6103 unexpectedly". The problem has been fixed by closing all cached lookups
6104 in a queue runner before running a delivery process.
6106 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
6107 seem to have the "const" qualifier which it has on other OS. I've
6110 8. Change 4.21/2 was too strict. It is only if there are two authenticators
6111 *of the same type* (client or server) with the same public name that an
6112 error should be diagnosed.
6114 9. When Exim looked up a host name for an IP address, but failed to find the
6115 original IP address when looking up the host name (a safety check), it
6116 output the message "<ip address> does not match any IP for NULL", which was
6117 confusing, to say the least. The bug was that the host name should have
6118 appeared instead of "NULL".
6120 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
6121 user that is built into the binary, and the -C or -D options is used, root
6122 privilege is dropped before the configuration file is read. In addition,
6123 logging is switched to stderr instead of the normal log files. If the
6124 configuration then re-defines the Exim user, the unprivileged environment
6125 is probably not what is expected, so Exim logs a panic warning message (but
6128 However, if deliver_drop_privilege is set, the unprivileged state may well
6129 be exactly what is intended, so the warning has been cut out in that case,
6130 and Exim is allowed to try to write to its normal log files.
6136 1. smtp_return_error_details was not giving details for temporary sender
6137 or receiver verification errors.
6139 2. Diagnose a configuration error if two authenticators have the same public
6142 3. Exim used not to create the message log file for a message until the first
6143 delivery attempt. This could be confusing when incoming messages were held
6144 for policy or load reasons. The message log file is now created at the time
6145 the message is received, and an initial "Received" line is written to it.
6147 4. The automatically generated man page for command line options had a minor
6148 bug that caused no ill effects; however, a more serious problem was that
6149 the procedure for building the man page automatically didn't always
6150 operate. Consequently, release 4.20 contains an out-of-date version. This
6151 shouldn't happen again.
6153 5. When building Exim with embedded Perl support, the script that builds the
6154 Makefile was calling 'perl' to find its compile-time parameters, ignoring
6155 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
6157 6. The freeze_tell option was not being used for messages that were frozen on
6158 arrival, either by an ACL or by local_scan().
6160 7. Added the smtp_incomplete_transaction log selector.
6162 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
6163 was accepting AUTH without a new EHLO.
6165 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
6166 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
6169 10. Logging of TCP/IP connections (when configured) now happens in the main
6170 daemon process instead of the child process, so that the TCP/IP connection
6171 count is more accurate (but it can never be perfect).
6173 11. The use of "drop" in a nested ACL was not being handled correctly in the
6174 outer ACL. Now, if condition failure induced by the nested "drop" causes
6175 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
6176 or "require"), the connection is dropped.
6178 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
6179 that yield "discard" can now be used with an "accept" or a "discard" verb,
6180 but an error is generated for any others (because I can't see a useful way
6181 to define what should happen).
6183 13. When an ACL is read dynamically from a file (or anywhere else), the lines
6184 are now processed in the same way as lines in the Exim configuration file.
6185 In particular, continuation lines are supported.
6187 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
6189 15. Added -ti meaning -t -i.
6191 16. Check for letters, digits, hyphens, and dots in the names of dnslist
6192 domains, and warn by logging if others are found.
6194 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
6195 returned from GIFCONF when Exim is trying to find the list of interfaces on
6196 a host. The code in os.c has been modified to copy each ifreq to an aligned
6197 structure in all cases.
6199 Also, in some cases, the returned ifreq's were being copied to a 'struct
6200 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
6201 means the last couple of bytes of an IPv6 address could be chopped if the
6202 ifreq contained only a normal sockaddr (14 bytes storage).
6204 18. Named domain lists were not supported in the hosts_treat_as_local option.
6205 An entry such as +xxxx was not recognized, and was treated as a literal
6208 19. Ensure that header lines added by a DATA ACL are included in the reject log
6209 if the ACL subsequently rejects the message.
6211 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
6212 MD5 (which is deprecated).
6214 21. When testing a filter file using -bf, Exim was writing a message when it
6215 took the sender from a "From " line in the message, but it was not doing so
6216 when it took $return_path from a Return-Path: header line. It now does.
6218 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
6219 with a valid header line field name (a series of printing characters
6220 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
6222 23. Changed "disc" in the source to "disk" to conform to the documentation and
6223 the book and for uniformity.
6225 24. Ignore Sendmail's -Ooption=value command line item.
6227 25. When execve() failed while trying to run a command in a pipe transport,
6228 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
6229 could be confused with a return value of 69 from the command itself. This
6230 has been changed to 127, the value the shell returns if it is asked to run
6231 a non-existent command. The wording for the related log line suggests a
6232 non-existent command as the problem.
6234 26. If received_header_text expands to an empty string, do not add a Received:
6235 header line to the message. (Well, it adds a token one on the spool, but
6236 marks it "old" so that it doesn't get used or transmitted.)
6238 27. Installed eximstats 1.28 (addition of -nt option).
6240 28. There was no check for failure on the call to getsockname() in the daemon
6241 code. This can fail if there is a shortage of resources on the system, with
6242 ENOMEM, for example. A temporary error is now given on failure.
6244 29. Contrary to the C standard, it seems that in some environments, the
6245 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
6246 program. Exim now does this explicitly; it affects the formatting of
6247 timestamps using strftime().
6249 30. If exiqsumm was given junk data, it threw up some uninitialized variable
6250 complaints. I've now initialized all the variables, to avoid this.
6252 32. Header lines added by a system filter were not being "seen" during
6253 transport-time rewrites.
6255 33. The info_callback() function passed to OpenSSL is set up with type void
6256 (*)(SSL *, int, int), as described somewhere. However, when calling the
6257 function (actually a macro) that sets it up, the type void(*)() is
6258 expected. I've put in a cast to prevent warnings from picky compilers.
6260 34. If a DNS black list lookup found a CNAME record, but there were no A
6261 records associated with the domain it pointed at, Exim crashed.
6263 35. If a DNS black list lookup returned more than one A record, Exim ignored
6264 all but the first. It now scans all returned addresses if a particular IP
6265 value is being sought. In this situation, the contents of the
6266 $dnslist_value variable are a list of all the addresses, separated by a
6269 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
6270 to accept a host name and all its aliases if the forward lookup for any of
6271 them yielded the IP address of the incoming connection. Now it accepts only
6272 those names whose forward lookup yields the correct IP address. Any other
6273 names are discarded. This closes a loophole whereby a rogue DNS
6274 administrator could create reverse DNS records to break through a
6275 wildcarded host restriction in an ACL.
6277 37. If a user filter or a system filter that ran in a subprocess used any of
6278 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
6279 the wrong values were passed to the pipe command ($thisaddress had the
6280 value of $0, $0 had the value of $1, etc). This bug was introduced by
6281 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
6283 38. Improved the line breaking for long SMTP error messages from ACLs.
6284 Previously, if there was no break point between 40 and 75 characters, Exim
6285 left the rest of the message alone. Two changes have been made: (a) I've
6286 reduced the minimum length to 35 characters; (b) if it can't find a break
6287 point between 35 and 75 characters, it looks ahead and uses the first one
6288 that it finds. This may give the occasional overlong line, but at least the
6289 remaining text gets split now.
6291 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
6292 file descriptors might be low, and that setting 1000 would always raise it.
6293 It turns out that in some environments, the limit is already over 1000 and
6294 that lowering it causes trouble. So now Exim takes care not to decrease it.
6296 40. When delivering a message, the value of $return_path is set to $sender_
6297 address at the start of routing (routers may change the value). By an
6298 oversight, this default was not being set up when an address was tested by
6299 -bt or -bv, which affected the outcome if any router or filter referred to
6302 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
6303 log only when "message" or "log_message" are set. However, if one of the
6304 conditions was an address verification, or a call to a nested ACL, the
6305 messages generated by the underlying test were being passed through. This
6306 no longer happens. The underlying message is available in $acl_verify_
6307 message for both "message" and "log_message" expansions, so it can be
6308 passed through if needed.
6310 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
6311 new expansion $bh_ to give the encoded byte string without charset
6312 translation. Translation happens only if iconv() is available; HAVE_ICONV
6313 indicates this at build time. HEADERS_CHARSET gives the charset to
6314 translate to; headers_charset can change it in the configuration, and
6315 "headers charset" can change it in an individual filter file.
6317 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
6318 that creates RFC 2047 encoded "words" labels them as that charset instead
6319 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
6320 expansion operator; (ii) when Exim creates a From: line for a local
6321 message; (iii) when a header line is rewritten to include a "phrase" part.
6323 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
6324 buggy, causing it to skip the first lines of messages whose message ID
6325 ended in 'D'. This would not have bitten before Exim release 4.14, because
6326 message IDs were unlikely to end in 'D' before then. The effect was to have
6327 incorrect size information for certain domains.
6329 45. #include "config.h" was missing at the start of the crypt16.c module. This
6330 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
6333 46. If there was a timeout during a "random" callout check, Exim treated it as
6334 a failure of the random address, and carried on sending RSET and the real
6335 address. If the delay was just some slowness somewhere, the response to the
6336 original RCPT would be taken as a response to RSET and so on, causing
6337 mayhem of various kinds.
6339 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
6340 when I implemented it. It didn't allow for the fact that some option values
6341 may legitimatetly be negative (e.g. size_addition), and it didn't even do
6342 the right test for positive values.
6344 48. Domain names in DNS records are case-independent. Exim always looks them up
6345 in lower case. Some resolvers return domain names in exactly the case they
6346 appear in the zone file, that is, they may contain uppercase letters. Not
6347 all resolvers do this - some return always lower case. Exim was treating a
6348 change of case by a resolver as a change of domain, similar to a widening
6349 of a domain abbreviation. This triggered its re-routing code and so it was
6350 trying to route what was effectively the same domain again. This normally
6351 caused routing to fail (because the router wouldn't handle the domain
6352 twice). Now Exim checks for this case specially, and just changes the
6353 casing of the domain that it ultimately uses when it transmits the message
6356 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
6359 50. If a filter generated a file delivery with a non-absolute name (possible if
6360 no home directory exists for the router), the forbid_file option was not
6363 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
6364 the existing equality matching.
6366 52. Exim was using ints instead of ino_t variables in some places where it was
6367 dealing with inode numbers.
6369 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
6370 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
6371 called TMPDIR, and if it finds it is different, it changes its value.
6373 54. The smtp_printf() function is now made available to local_scan() so
6374 additional output lines can be written before returning. There is also an
6375 smtp_fflush() function to enable the detection of a dropped connection.
6376 The variables smtp_input and smtp_batched_input are exported to
6379 55. Changed the default runtime configuration: the message "Unknown user"
6380 has been removed from the ACL, and instead placed on the localuser router,
6381 using the cannot_route_message feature. This means that any verification
6382 failures that generate their own messages won't get overridden. Similarly,
6383 the "Unrouteable address" message that was in the ACL for unverifiable
6384 relay addresses has also been removed.
6386 56. Added hosts_avoid_esmtp to the smtp transport.
6388 57. The exicyclog script was not checking for the esoteric option
6389 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
6390 will work only if exicyclog is run under the appropriate euid.
6392 58. Following a discussion on the list, the rules by which Exim recognises line
6393 endings on incoming messages have been changed. The -dropcr and drop_cr
6394 options are now no-ops, retained only for backwards compatibility. The
6395 following line terminators are recognized: LF CRLF CR. However, special
6396 processing applies to CR:
6398 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
6399 nor a local message in the state where . is a terminator.
6401 (ii) If a bare CR is encountered in a header line, an extra space is added
6402 after the line terminator so as not to end the header. The reasoning
6403 behind this is that bare CRs in header lines are most likely either
6404 to be mistakes, or people trying to play silly games.
6406 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
6407 was being incorrectly given as 18 bytes larger than it should have been.
6408 This is a VOB (very old bug).
6410 60. This may never have affected anything current, but just in case it has:
6411 When the local host is found other than at the start of a list of hosts,
6412 the local host, those with the same MX, and any that follow, are discarded.
6413 When the list in question was part of a longer list of hosts, the following
6414 hosts (not currently being processed) were also being discarded. This no
6415 longer happens. I'm not sure if this situation could ever has previously
6418 61. Added the "/MX" feature to lists of hosts in the manualroute and query
6421 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
6422 header. This is something that is recommended in a new Internet Draft, and
6423 is something that is documented as being done by Sendmail. There are two
6424 possible values. For messages generated by the autoreply transport, Exim
6427 Auto-Submitted: auto-replied
6429 whereas for all other generated messages (e.g. bounces) it adds
6431 Auto-Submitted: auto-generated
6433 63. The "personal" condition in filters now includes a test for the
6434 Auto-Submitted: header. If it contains the string "auto-" the message it
6435 not considered personal.
6437 64. Added rcpt_include_affixes as a generic transport option.
6439 65. Added queue_only_override (default true).
6441 66. Added the syslog_duplication option.
6443 67. If what should have been the first header line of a message consisted of
6444 a space followed by a colon, Exim was mis-interpreting it as a header line.
6445 It isn't of course - it is syntactically invalid and should therefore be
6446 treated as the start of the message body. The misbehaviour could have
6447 caused a number of strange effects, including loss of data in subsequent
6448 header lines, and spool format errors.
6450 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
6451 client host had authenticated. This control can now be exercised by an ACL
6452 for more flexibility.
6454 69. By default, callouts do not happen when testing with -bh. There is now a
6455 variant, -bhc, which does actually run the callout code, including
6456 consulting and updating the callout cache.
6458 70. Added support for saslauthd authentication, courtesy of Alexander
6461 71. If statvfs() failed on the spool or log directories while checking their
6462 size for availability, Exim confusingly gave the error "space shortage".
6463 Furthermore, in debugging mode it crashed with a floating point exception.
6464 These checks are done if check_{spool,log}_{space,inodes} are set, and when
6465 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
6466 serious problem, Exim now writes to the main and panic logs when this
6467 happens, with details of the failure. It then refuses to accept the
6468 incoming message, giving the message "spool directory problem" or "log
6469 directory problem" with a 421 code for SMTP messages.
6471 72. When Exim is about to re-exec itself, it ensures that the file descriptors
6472 0, 1, and 2 exist, because some OS complain for execs without them (see
6473 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
6474 descriptors. However, the code omitted to check that the open succeeded,
6475 causing mysterious errors if for some reason the permissions on /dev/null
6476 got screwed. Now Exim writes a message to the main and panic logs, and
6477 bombs out if it can't open /dev/null.
6479 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
6480 interact so that it is all more flexible. It is supposed to remain
6481 backwards compatible. Also added extra_local_interfaces.
6483 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
6484 to bomb out with an assertion failure - to the client this appears as a
6485 connection drop. This problem occurs in the part of the code that was taken
6486 from the Samba project. Fortunately, the assertion is in a very simple
6487 function, so I have fixed this by reproducing the function inline in the
6488 one place where it is called, and arranging for authentication to fail
6489 instead of killing the process with assert().
6491 75. The SPA client code was not working when the server requested OEM rather
6492 than Unicode encoding.
6494 76. Added code to make require_files with a specific uid setting more usable in
6495 the case where statting the file as root fails - usually a non-root-mounted
6496 NFS file system. When this happens and the failure is EACCES, Exim now
6497 forks a subprocess and does the per-uid checking as the relevant uid.
6499 77. Added process_log_path.
6501 78. If log_file_path was not explicitly set, a setting of check_log_space or
6502 check_log_inodes was ignored.
6504 79. If a space check for the spool or log partitions fails, the incident is now
6505 logged. Of course, in the latter case the data may get lost...
6507 80. Added the %p formatting code to string_format() so that it can be used to
6508 print addresses in debug_print(). Adjusted all the address printing in the
6509 debugging in store.c to use %p rather than %d.
6511 81. There was a concern that a line of code in smtp_in.c could overflow a
6512 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
6513 initially expressed, the concern was not well-founded, because trailing
6514 spaces are removed early. However, if the trailing spaces were followed by
6515 a NULL, they did not get removed, so the overflow was possible. Two fixes
6518 (a) I re-wrote the offending code in a cleaner fashion.
6519 (b) If an incoming SMTP command contains a NULL character, it is rejected
6522 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
6523 runs initgroups(), so that if the Exim user is in any additional groups,
6524 they will be used during message reception.
6530 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob