1 Change log file for Exim from version 4.21
2 -------------------------------------------
6 JH/01 Use SIZE on MAIL FROM in a cutthrough connection, if the destination
7 supports it and a size is available (ie. the sending peer gave us one).
9 JH/02 The obsolete acl condition "demime" is removed (finally, after ten
10 years of being deprecated). The replacements are the ACLs
11 acl_smtp_mime and acl_not_smtp_mime.
13 JH/03 Upgrade security requirements imposed for hosts_try_dane: previously
14 a downgraded non-dane trust-anchor for the TLS connection (CA-style)
15 or even an in-clear connection were permitted. Now, if the host lookup
16 was dnssec and dane was requested then the host is only used if the
17 TLSA lookup succeeds and is dnssec. Further hosts (eg. lower priority
18 MXs) will be tried (for hosts_try_dane though not for hosts_require_dane)
19 if one fails this test.
20 This means that a poorly-configured remote DNS will make it incommunicado;
21 but it protects against a DNS-interception attack on it.
23 JH/04 Bug 1810: make continued-use of an open smtp transport connection
24 non-noisy when a race steals the message being considered.
26 JH/05 If main configuration option tls_certificate is unset, generate a
27 selfsigned certificate for inbound TLS connections.
29 JH/06 Bug 165: hide more cases of password exposure - this time in expansions
30 in rewrites and routers.
32 JH/07 Retire gnutls_require_mac et.al. These were nonfunctional since 4.80
33 and logged a warning sing 4.83; now they are a configuration file error.
35 JH/08 Bug 1836: Fix crash in VRFY handling when handed an unqualified name
36 (lacking @domain). Apply the same qualification processing as RCPT.
38 JH/09 Bug 1804: Avoid writing msglog files when in -bh or -bhc mode.
40 JH/10 Support ${sha256:} applied to a string (as well as the previous
46 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
47 and 3.4.4 - once the server is enabled to respond to an OCSP request
48 it does even when not requested, resulting in a stapling non-aware
49 client dropping the TLS connection.
51 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
52 support variable-length bit vectors. No functional change.
54 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
55 The I= interface field on outgoing lines is now after the H= remote
56 host field, same as incoming lines. There is a separate
57 outgoing_interface log selector which allows you to disable the
60 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
61 If not running log_selector +smtp_connection the mainlog would be held
62 open indefinitely after a "too many connections" event, including to a
63 deleted file after a log rotate. Leave the per net connection logging
64 leaving it open for efficiency as that will be quickly detected by the
65 check on the next write.
67 HS/01 Bug 1671: Fix post transport crash.
68 Processing the wait-<transport> messages could crash the delivery
69 process if the message IDs didn't exist for some reason. When
70 using 'split_spool_directory=yes' the construction of the spool
71 file name failed already, exposing the same netto behaviour.
73 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
74 mime_regex ACL conditions.
76 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
77 to DSN fail messages (bounces): remote IP, remote greeting, remote response
78 to HELO, local diagnostic string.
80 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
81 log line to debug. Even when configured with a tls authenticator many
82 client connections are expected to not authenticate in this way, so
83 an authenticate fail is not an error.
85 HS/02 Add the Exim version string to the process info. This way exiwhat
86 gives some more detail about the running daemon.
88 JH/06 Bug 1395: time-limit cacheing of DNS lookups, to the TTL value. This may
89 matter for fast-change records such as DNSBLs.
91 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
92 retry record, even if constant. There may be multiple transports with
93 different interface settings and the retry behaviour needs to be kept
96 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
98 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
100 JH/10 Bug 840: fix log_defer_output option of pipe transport
102 JH/11 Bug 830: use same host for all RCPTS of a message, even under
103 hosts_randomize. This matters a lot when combined with mua_wrapper.
105 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
106 ${quote_pgsql:<string>} operator.
108 JH/13 Bug 1708: avoid misaligned access in cached lookup.
110 JH/14 Change header file name for freeradius-client. Relevant if compiling
111 with Radius support; from the Gentoo tree and checked under Fedora.
113 JH/15 Bug 1712: Introduce $prdr_requested flag variable
115 JH/16 Bug 1714: Permit an empty string as expansion result for transport
116 option transport_filter, meaning no filtering.
118 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
120 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
121 defaults to "*" (all hosts). The variable is now available when not built
122 with TLS, default unset, mainly to enable keeping the testuite sane.
123 If a server certificate is not supplied (via tls_certificate) an error is
124 logged, and clients will find TLS connections fail on startup. Presumably
125 they will retry in-clear.
126 Packagers of Exim are strongly encouraged to create a server certificate
127 at installation time.
129 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
130 with the $config_file variable.
132 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
133 in transport context, after the attempt, and per-recipient. The latter type
134 is per host attempted. The event data is the error message, and the errno
135 information encodes the lookup type (A vs. MX) used for the (first) host,
136 and the trailing two digits of the smtp 4xx reponse.
138 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
139 to write to mainlog (or rejectlog, paniclog) in the window between file
140 creation and permissions/ownership being changed. Particularly affects
141 installations where exicyclog is run as root, rather than exim user;
142 result is that the running daemon panics and dies.
144 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
146 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
147 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
148 "pri" and "weight". Note that the previous implicit priority given by the
149 list order is no longer honoured.
151 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalisation
154 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
155 by defining SUPPORT_SOCKS.
157 JH/26 Move PROXY support from Experimental to mainline, enabled for a build
158 by defining SUPPORT_PROXY. Note that the proxy_required_hosts option
159 is renamed to hosts_proxy, and the proxy_{host,target}_{address,port}.
160 variables are renamed to proxy_{local,external}_{address,port}.
162 JH/27 Move Internationalisation support from Experimental to mainline, enabled
163 for a build by defining SUPPORT_I18N
165 JH/28 Bug 1745: Fix redis lookups to handle (quoted) spaces embedded in parts
166 of the query string, and make ${quote_redis:} do that quoting.
168 JH/29 Move Events support from Experimental to mainline, enabled by default
169 and removable for a build by defining DISABLE_EVENT.
171 JH/30 Updated DANE implementation code to current from Viktor Dukhovni.
173 JH/31 Fix bug with hosts_connection_nolog and named-lists which were wrongly
174 cached by the daemon.
176 JH/32 Move Redis support from Experimental to mainline, enabled for a build
177 by defining LOOKUP_REDIS. The libhiredis library is required.
179 JH/33 Bug 1748: Permit ACL dnslists= condition in non-smtp ACLs if explicit
180 keys are given for lookup.
182 JH/34 Bug 1192: replace the embedded copy of PolarSSL RSA routines in the DKIM
183 support, by using OpenSSL or GnuTLS library ones. This means DKIM is
184 only supported when built with TLS support. The PolarSSL SHA routines
185 are still used when the TLS library is too old for convenient support.
187 JH/35 Require SINGLE_DH_USE by default in OpenSSL (main config option
188 openssl_options), for security. OpenSSL forces this from version 1.1.0
189 server-side so match that on older versions.
191 JH/36 Bug 1778: longstanding bug in memory use by the ${run } expansion: A fresh
192 allocation for $value could be released as the expansion processing
193 concluded, but leaving the global pointer active for it.
195 JH/37 Bug 1769: Permit a VRFY ACL to override the default 252 response,
196 and to use the domains and local_parts ACL conditions.
198 JH/38 Fix cutthrough bug with body lines having a single dot. The dot was
199 incorrectly not doubled on cutthrough transmission, hence seen as a
200 body-termination at the receiving system - resulting in truncated mails.
201 Commonly the sender saw a TCP-level error, and retransmitted the message
202 via the normal store-and-forward channel. This could result in duplicates
203 received - but deduplicating mailstores were liable to retain only the
204 initial truncated version.
206 JH/39 Bug 1781: Fix use of DKIM private-keys having trailing '=' in the base-64.
208 JH/40 Fix crash in queryprogram router when compiled with EXPERIMENTAL_SRS.
210 JH/41 Bug 1792: Fix selection of headers to sign for DKIM: bottom-up. While
211 we're in there, support oversigning also; bug 1309.
213 JH/42 Bug 1796: Fix error logged on a malware scanner connection failure.
215 HS/04 Add support for keep_environment and add_environment options.
217 JH/43 Tidy coding issues detected by gcc --fsanitize=undefined. Some remain;
218 either intentional arithmetic overflow during PRNG, or testing config-
221 JH/44 Bug 1800: The combination of a -bhc commandline option and cutthrough
222 delivery resulted in actual delivery. Cancel cutthrough before DATA
225 JH/45 Fix cutthrough, when connection not opened by verify and target hard-
226 rejects a recipient: pass the reject to the originator.
228 JH/46 Multiple issues raised by Coverity. Some were obvious or plausible bugs.
229 Many were false-positives and ignorable, but it's worth fixing the
232 JH/47 Fix build on HP-UX and older Solaris, which need (un)setenv now also
233 for the new environment-manipulation done at startup. Move the routines
234 from being local to tls.c to being global via the os.c file.
236 JH/48 Bug 1807: Fix ${extract } for the numeric/3-string case. While preparsing
237 an extract embedded as result-arg for a map, the first arg for extract
238 is unavailable so we cannot tell if this is a numbered or keyed
239 extraction. Accept either.
245 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
248 JH/02 The smtp transport option "multi_domain" is now expanded.
250 JH/03 The smtp transport now requests PRDR by default, if the server offers
253 JH/04 Certificate name checking on server certificates, when exim is a client,
254 is now done by default. The transport option tls_verify_cert_hostnames
255 can be used to disable this per-host. The build option
256 EXPERIMENTAL_CERTNAMES is withdrawn.
258 JH/05 The value of the tls_verify_certificates smtp transport and main options
259 default to the word "system" to access the system default CA bundle.
260 For GnuTLS, only version 3.0.20 or later.
262 JH/06 Verification of the server certificate for a TLS connection is now tried
263 (but not required) by default. The verification status is now logged by
264 default, for both outbound TLS and client-certificate supplying inbound
267 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
270 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
271 Status Notification (bounce) messages are now MIME format per RFC 3464.
272 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
273 under the control of the dsn_advertise_hosts option, and routers may
274 have a dsn_lasthop option.
276 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
277 default, modifiable by a malware= option. The list separator for
278 the options can now be changed in the usual way. Bug 68.
280 JH/10 The smtp_receive_timeout main option is now expanded before use.
282 JH/11 The incoming_interface log option now also enables logging of the
283 local interface on delivery outgoing connections.
285 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
286 if the interface and destination host and port all match.
288 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
291 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
292 Patch from Andrew Lewis.
294 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
295 now supports optional time-restrictions, weighting, and priority
296 modifiers per server. Patch originally by <rommer@active.by>.
298 JH/16 The spamd_address main option now supports a mixed list of local
299 and remote servers. Remote servers can be IPv6 addresses, and
300 specify a port-range.
302 JH/17 Bug 68: The spamd_address main option now supports an optional
303 timeout value per server.
305 JH/18 Bug 1581: Router and transport options headers_add/remove can
306 now have the list separator specified.
308 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
311 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
314 JH/21 Support for the A6 type of dns record is withdrawn.
316 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
317 rather than the verbs used.
319 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
320 from 255 to 1024 chars.
322 JH/24 Verification callouts now attempt to use TLS by default.
324 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
325 are generic router options now. The defaults didn't change.
327 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
328 Original patch from Alexander Shikoff, worked over by JH.
330 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
331 files if ZCAT_COMMAND is not executable.
333 JH/26 Bug 1539: Add timout/retry options on dnsdb lookups.
335 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
337 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
338 Normally benign, it bites when the pair was led to by a CNAME;
339 modern usage is to not canoicalize the domain to a CNAME target
340 (and we were inconsistent anyway for A-only vs AAAA+A).
342 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
344 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
345 when evaluating $sender_host_dnssec.
347 JH/31 Check the HELO verification lookup for DNSSEC, adding new
348 $sender_helo_dnssec variable.
350 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
352 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
354 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
356 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
357 documented as working, but never had. Support all but $spam_report.
359 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
360 added for tls authenticator.
362 HS/03 Add perl_taintmode main config option
367 TL/01 When running the test suite, the README says that variables such as
368 no_msglog_check are global and can be placed anywhere in a specific
369 test's script, however it was observed that placement needed to be near
370 the beginning for it to behave that way. Changed the runtest perl
371 script to read through the entire script once to detect and set these
372 variables, reset to the beginning of the script, and then run through
373 the script parsing/test process like normal.
375 TL/02 The BSD's have an arc4random API. One of the functions to induce
376 adding randomness was arc4random_stir(), but it has been removed in
377 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
378 function when detected.
380 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
381 cause callback expansion.
383 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
384 syntax errors in an expansion can be treated as a string instead of
385 logging or causing an error, due to the internal use of bool_lax
386 instead of bool when processing it.
388 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
389 server certificates when making smtp deliveries.
391 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
393 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
395 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
397 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
398 Merged patch from Sebastian Wiedenroth.
400 JH/05 Fix results-pipe from transport process. Several recipients, combined
401 with certificate use, exposed issues where response data items split
402 over buffer boundaries were not parsed properly. This eventually
403 resulted in duplicates being sent. This issue only became common enough
404 to notice due to the introduction of conection certificate information,
405 the item size being so much larger. Found and fixed by Wolfgang Breyha.
407 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
408 size buffer was used, resulting in syntax errors when an expansion
411 JH/07 Add support for directories of certificates when compiled with a GnuTLS
412 version 3.3.6 or later.
414 JH/08 Rename the TPDA expermimental facility to Event Actions. The #ifdef
415 is EXPERIMENTAL_EVENT, the main-configuration and transport options
416 both become "event_action", the variables become $event_name, $event_data
417 and $event_defer_errno. There is a new variable $verify_mode, usable in
418 routers, transports and related events. The tls:cert event is now also
419 raised for inbound connections, if the main configuration event_action
422 TL/06 In test suite, disable OCSP for old versions of openssl which contained
423 early OCSP support, but no stapling (appears to be less than 1.0.0).
425 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
426 server certificate names available under the smtp transport option
427 "tls_verify_cert_hostname" now do not permit multi-component wildcard
430 JH/10 Time-related extraction expansions from certificates now use the main
431 option "timezone" setting for output formatting, and are consistent
432 between OpenSSL and GnuTLS compilations. Bug 1541.
434 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
435 encoded parameter in the incoming message. Bug 1558.
437 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
438 include certificate info, eximon was claiming there were spoolfile
441 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
443 JH/14 Log delivery-related information more consistently, using the sequence
444 "H=<name> [<ip>]" wherever possible.
446 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
447 are problematic for Debian distribution, omit them from the release
450 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
452 JH/16 Fix string representation of time values on 64bit time_t anchitectures.
455 JH/17 Fix a null-indirection in certextract expansions when a nondefault
456 output list separator was used.
461 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
462 checkers that were complaining about end of non-void function with no
465 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
466 This was a regression intruduced in 4.83 by another bugfix.
468 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
470 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
471 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
477 TF/01 Correctly close the server side of TLS when forking for delivery.
479 When a message was received over SMTP with TLS, Exim failed to clear up
480 the incoming connection properly after forking off the child process to
481 deliver the message. In some situations the subsequent outgoing
482 delivery connection happened to have the same fd number as the incoming
483 connection previously had. Exim would try to use TLS and fail, logging
484 a "Bad file descriptor" error.
486 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
487 utilities have not been installed.
489 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
490 temporary space as the ACL may create new global variables.
492 TL/01 LDAP support uses per connection or global context settings, depending
493 upon the detected version of the libraries at build time.
495 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
496 to extract and use the src ip:port in logging and expansions as if it
497 were a direct connection from the outside internet. PPv2 support was
498 updated based on HAProxy spec change in May 2014.
500 JH/02 Add ${listextract {number}{list}{success}{fail}}.
502 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
503 Properly escape header and check for NULL return.
505 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
508 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
510 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
511 characters in header names, implemented as a verify condition.
512 Contributed by Michael Fischer v. Mollard.
514 TL/05 Rename SPF condition results err_perm and err_temp to standardized
515 results permerror and temperror. Previous values are deprecated but
516 still accepted. In a future release, err_perm and err_temp will be
517 completely removed, which will be a backward incompatibility if the
518 ACL tests for either of these two old results. Patch contributed by
519 user bes-internal on the mailing list.
521 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
523 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
524 selectors, in both main and reject logs.
526 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
529 JH/07 Add malware type "sock" for talking to simple daemon.
531 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
533 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
534 routers/transports under cutthrough routing.
536 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
537 numbers. Touch up "bool" conditional to keep the same definition.
539 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
541 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
543 JH/12 Expand items in router/transport headers_add or headers_remove lists
544 individually rather than the list as a whole. Bug 1452.
546 Required for reasonable handling of multiple headers_ options when
547 they may be empty; requires that headers_remove items with embedded
548 colons must have them doubled (or the list-separator changed).
550 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
551 view the policy declared in the DMARC record. Currently, $dmarc_status
552 is a combined value of both the record presence and the result of the
555 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
557 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
558 dnslookup router and the smtp transport (applying to the forward
561 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
562 of ldap servers used for a specific lookup. Patch provided by Heiko
565 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
566 New variable $lookup_dnssec_authenticated for observability.
568 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
569 Patch submitted by Lars Timman.
571 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
573 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
574 Requires trusted mode and valid format message id, aborts otherwise.
575 Patch contributed by Heiko Schlichting.
577 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
578 certextract with support for various fields. Bug 1358.
580 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
581 is requested by default, modifiable by smtp transport option
584 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
585 operate on certificate variables to give certificate fingerprints
586 Also new ${sha256:cert_variable}.
588 JH/23 The PRDR feature is moved from being Experimental into the mainline.
590 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
591 Christian Aistleitner.
593 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
595 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
596 file. Patch from Wolfgang Breyha.
598 JH/25 Expand the coverage of the delivery $host and $host_address to
599 client authenticators run in verify callout. Bug 1476.
601 JH/26 Port service names are now accepted for tls_on_connect_ports, to
602 align with daemon_smtp_ports. Bug 72.
604 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
605 support and error reporting did not work properly.
607 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
608 and is readable. Patch from Andrew Colin Kissa.
610 TL/14 Enhance documentation of ${run expansion and how it parses the
611 commandline after expansion, particularly in the case when an
612 unquoted variable expansion results in an empty value.
614 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
616 PP/02 Fix internal collision of T_APL on systems which support RFC3123
617 by renaming away from it. Addresses GH issue 15, reported by
620 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
622 TL/15 SECURITY: prevent double expansion in math comparison functions
623 (can expand unsanitized data). Not remotely exploitable.
630 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
632 PP/02 Make -n do something, by making it not do something.
633 When combined with -bP, the name of an option is not output.
635 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
638 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
639 $sender_host_name and config options to manage this, and basic check
642 PP/05 DSCP support for outbound connections and control modifier for inbound.
644 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
645 (Only plugin which currently uses this is kerberos4, which nobody should
646 be using, but we should make it available and other future plugins might
647 conceivably use it, even though it would break NAT; stuff *should* be
648 using channel bindings instead).
650 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
651 name; added for Sendmail compatibility; requires admin caller.
652 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
653 just ignore it); requires trusted caller.
654 Also parse but ignore: -Ac -Am -X<logfile>
657 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
659 TL/02 Add +smtp_confirmation as a default logging option.
661 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
662 Patch by Magnus Holmgren from 2007-02-20.
664 TL/04 Bugzilla 1281 - Spec typo.
665 Bugzilla 1283 - Spec typo.
666 Bugzilla 1290 - Spec grammar fixes.
668 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
670 TL/06 Add Experimental DMARC support using libopendmarc libraries.
672 TL/07 Fix an out of order global option causing a segfault. Reported to dev
673 mailing list by by Dmitry Isaikin.
675 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
677 JH/02 Support "G" suffix to numbers in ${if comparisons.
679 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
681 NM/01 Bugzilla 1197 - Spec typo
682 Bugzilla 1196 - Spec examples corrections
684 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
686 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
687 gnutls_enable_pkcs11, but renamed to more accurately indicate its
690 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
691 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
693 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
694 "acl {{name}{arg}...}", and optional args on acl condition
697 JH/05 Permit multiple router/transport headers_add/remove lines.
699 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
701 JH/07 Avoid using a waiting database for a single-message-only transport.
702 Performance patch from Paul Fisher. Bugzilla 1262.
704 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
707 JH/09 Add $headers_added variable, with content from use of ACL modifier
708 add_header (but not yet added to the message). Bugzilla 199.
710 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
711 Pulled from Bugzilla 817 by Wolfgang Breyha.
713 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
715 (nb: this is the same fix as in Exim 4.80.1)
717 JH/11 Add A= logging on delivery lines, and a client_set_id option on
720 JH/12 Add optional authenticated_sender logging to A= and a log_selector
723 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
725 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
726 advertise SMTP AUTH mechanism to us, instead of a generic
727 protocol violation error. Also, make Exim more robust to bad
728 data from the Dovecot auth socket.
730 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
732 When a queue runner is handling a message, Exim first routes the
733 recipient addresses, during which it prunes them based on the retry
734 hints database. After that it attempts to deliver the message to
735 any remaining recipients. It then updates the hints database using
738 So if a recipient address works intermittently, it can get repeatedly
739 deferred at routing time. The retry hints record remains fresh so the
740 address never reaches the final cutoff time.
742 This is a fairly common occurrence when a user is bumping up against
743 their storage quota. Exim had some logic in its local delivery code
744 to deal with this. However it did not apply to per-recipient defers
745 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
747 This change adds a proper retry rule check during routing so that the
748 final cutoff time is checked against the message's age. We only do
749 this check if there is an address retry record and there is not a
750 domain retry record; this implies that previous attempts to handle
751 the address had the retry_use_local_parts option turned on. We use
752 this as an approximation for the destination being like a local
753 delivery, as in LMTP.
755 I suspect this new check makes the old local delivery cutoff check
756 redundant, but I have not verified this so I left the code in place.
758 TF/02 Correct gecos expansion when From: is a prefix of the username.
760 Test 0254 submits a message to Exim with the header
764 When I ran the test suite under the user fanf2, Exim expanded
765 the header to contain my full name, whereas it should have added
766 a Resent-Sender: header. It erroneously treats any prefix of the
767 username as equal to the username.
769 This change corrects that bug.
771 GF/01 DCC debug and logging tidyup
772 Error conditions log to paniclog rather than rejectlog.
773 Debug lines prefixed by "DCC: " to remove any ambiguity.
775 TF/03 Avoid unnecessary rebuilds of lookup-related code.
777 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
778 Bug spotted by Jeremy Harris; was flawed since initial commit.
779 Would have resulted in OCSP responses post-SNI triggering an Exim
780 NULL dereference and crash.
782 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
784 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
785 Bug detection, analysis and fix by Samuel Thibault.
786 Bugzilla 1331, Debian bug #698092.
788 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
790 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
791 Server implementation by Todd Lyons, client by JH.
792 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
793 config variable "prdr_enable" controls whether the server
794 advertises the facility. If the client requests PRDR a new
795 acl_data_smtp_prdr ACL is called once for each recipient, after
796 the body content is received and before the acl_smtp_data ACL.
797 The client is controlled by bolth of: a hosts_try_prdr option
798 on the smtp transport, and the server advertisement.
799 Default client logging of deliveries and rejections involving
800 PRDR are flagged with the string "PRDR".
802 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
803 fclose(). Diagnosis by Todd Lyons.
805 PP/17 Update configure.default to handle IPv6 localhost better.
806 Patch by Alain Williams (plus minor tweaks).
809 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
810 This is now consistent with GnuTLS, and is now documented: the
811 previous undocumented portable approach to treating the option as
812 unset was to force an expansion failure. That still works, and
813 an empty string is now equivalent.
815 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
816 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
817 not performing validation itself.
819 PP/20 Added force_command boolean option to pipe transport.
820 Patch from Nick Koston, of cPanel Inc.
822 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
825 TF/04 Added udpsend ACL modifer and hexquote expansion operator
827 PP/21 Fix eximon continuous updating with timestamped log-files.
828 Broken in a format-string cleanup in 4.80, missed when I repaired the
829 other false fix of the same issue.
830 Report and fix from Heiko Schlichting.
833 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
834 Report from Prashanth Katuri.
836 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
837 It's SecureTransport, so affects any MacOS clients which use the
838 system-integrated TLS libraries, including email clients.
840 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
841 using a MIME ACL for non-SMTP local injection.
842 Report and assistance in diagnosis by Warren Baker.
844 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
846 JH/16 Fix comparisons for 64b. Bugzilla 1385.
848 TL/09 Add expansion variable $authenticated_fail_id to keep track of
849 last id that failed so it may be referenced in subsequent ACL's.
851 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
854 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
855 ldap library initialization, allowing self-signed CA's to be
856 used. Also properly sets require_cert option later in code by
857 using NULL (global ldap config) instead of ldap handle (per
858 session). Bug diagnosis and testing by alxgomz.
860 TL/12 Enhanced documentation in the ratelimit.pl script provided in
861 the src/util/ subdirectory.
863 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
864 renamed to Transport Post Delivery Action by Jeremy Harris, as
867 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
868 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
869 redis_servers = needs to be configured which will be used by the redis
870 lookup. Patch from Warren Baker, of The Packet Hub.
872 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
874 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
875 hostname or reverse DNS when processing a host list. Used suggestions
876 from multiple comments on this bug.
878 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
880 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
881 Missed a few lines, added it to make the runtest require no keyboard
884 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
885 contains upper case chars. Make router use caseful_local_part.
887 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
888 support when GnuTLS has been built with p11-kit.
894 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
896 This, or similar/improved, will also be change PP/11 of 4.82.
902 PP/01 Handle short writes when writing local log-files.
903 In practice, only affects FreeBSD (8 onwards).
904 Bugzilla 1053, with thanks to Dmitry Isaikin.
906 NM/01 Bugzilla 949 - Documentation tweak
908 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
911 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
913 PP/02 Implemented gsasl authenticator.
915 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
917 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
918 `pkg-config foo` for cflags/libs.
920 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
921 with rest of GSASL and with heimdal_gssapi.
923 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
924 `pkg-config foo` for cflags/libs for the TLS implementation.
926 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
927 properties get this fed in as external SSF. A number of robustness
928 and debugging improvements to the cyrus_sasl authenticator.
930 PP/08 cyrus_sasl server now expands the server_realm option.
932 PP/09 Bugzilla 1214 - Log authentication information in reject log.
933 Patch by Jeremy Harris.
935 PP/10 Added dbmjz lookup type.
937 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
939 PP/12 MAIL args handles TAB as well as SP, for better interop with
940 non-compliant senders.
941 Analysis and variant patch by Todd Lyons.
943 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
944 Bug report from Lars Müller <lars@samba.org> (via SUSE),
945 Patch from Dirk Mueller <dmueller@suse.com>
947 PP/13 tls_peerdn now print-escaped for spool files.
948 Observed some $tls_peerdn in wild which contained \n, which resulted
949 in spool file corruption.
951 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
952 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
953 or write after TLS renegotiation, which otherwise led to messages
956 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
957 as a tracking header (ie: a signed header comes before the signature).
958 Patch from Wolfgang Breyha.
960 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
961 comma-sep list; embedded commas doubled.
963 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
965 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
967 Report and patch from Dmitry Banschikov.
969 PP/16 Removed "dont_insert_empty_fragments" fron "openssl_options".
970 Removed SSL_clear() after SSL_new() which led to protocol negotiation
971 failures. We appear to now support TLS1.1+ with Exim.
973 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
974 lets Exim select keys and certificates based upon TLS SNI from client.
975 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
976 before an outbound SMTP session. New log_selector, +tls_sni.
978 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
979 NULL dereference. Report and patch from Alun Jones.
981 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
982 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
983 Not seeing resolver debug output on NetBSD, but suspect this is a
984 resolver implementation change.
986 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
987 Left warnings. Added "eximon gdb" invocation mode.
989 PP/21 Defaulting "accept_8bitmime" to true, not false.
991 PP/22 Added -bw for inetd wait mode support.
993 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
994 locate the relevant includes and libraries. Made this the default.
996 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
997 Bugzilla 1246, report and most of solution from Tomasz Kusy.
999 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
1000 This may cause build issues on older platforms.
1002 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
1003 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
1004 gnutls_require_mac & gnutls_require_protocols (no longer supported).
1005 Added SNI support via GnuTLS too.
1006 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
1008 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
1010 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
1011 Applied second patch from Janne, implementing suggestion to default
1012 multiple-strings-in-record handling to match SPF spec.
1014 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
1016 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
1017 read-only, out of scope).
1018 Patch from Wolfgang Breyha, report from Stuart Northfield.
1020 PP/29 Fix three issues highlighted by clang analyser static analysis.
1021 Only crash-plausible issue would require the Cambridge-specific
1022 iplookup router and a misconfiguration.
1023 Report from Marcin Mirosław.
1025 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
1027 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
1028 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
1029 As part of this, removing so much warning spew let me fix some minor
1030 real issues in debug logging.
1032 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
1033 assignment on my part. Fixed.
1035 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
1036 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
1037 Janne Snabb (who went above and beyond: thank you).
1039 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
1040 string otherwise requires a connection and a bunch more work and it's
1041 relatively easy to get wrong. Should also expose TLS library linkage
1044 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
1045 64-bit ${eval} (JH/03).
1047 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
1048 GNU libc to support some of the 64-bit stuff, should not lead to
1049 conflicts. Defined before os.h is pulled in, so if a given platform
1050 needs to override this, it can.
1052 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
1053 protection layer was required, which is not implemented.
1054 Bugzilla 1254, patch from Wolfgang Breyha.
1056 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
1057 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
1058 tls_dhparam take prime identifiers. Also unbreak combination of
1059 OpenSSL+DH_params+TLSSNI.
1061 PP/39 Disable SSLv2 by default in OpenSSL support.
1067 PP/01 Solaris build fix for Oracle's LDAP libraries.
1068 Bugzilla 1109, patch from Stephen Usher.
1070 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
1072 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
1075 TF/02 Fix a couple more cases where we did not log the error message
1076 when unlink() failed. See also change 4.74-TF/03.
1078 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
1079 lock up or crash if it happened to be inside a call to libc when it
1080 got a SIGUSR1 from exiwhat.
1082 The SIGUSR1 handler appends the current process status to the process
1083 log which is later printed by exiwhat. It used to use the general
1084 purpose logging code to do this, but several functions it calls are
1085 not safe for signals.
1087 The new output code in the SIGUSR1 handler is specific to the process
1088 log, and simple enough that it's easy to inspect for signal safety.
1089 Removing some special cases also simplifies the general logging code.
1090 Removing the spurious timestamps from the process log simplifies
1093 TF/04 Improved ratelimit ACL condition.
1095 The /noupdate option has been deprecated in favour of /readonly which
1096 has clearer semantics. The /leaky, /strict, and /readonly update modes
1097 are mutually exclusive. The update mode is no longer included in the
1098 database key; it just determines when the database is updated. (This
1099 means that when you upgrde Exim will forget old rate measurements.)
1101 Exim now checks that the per_* options are used with an update mode that
1102 makes sense for the current ACL. For example, when Exim is processing a
1103 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
1104 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
1105 must specify per_mail/readonly. If you omit the update mode it defaults to
1106 /leaky where that makes sense (as before) or /readonly where required.
1108 The /noupdate option is now undocumented but still supported for
1109 backwards compatibility. It is equivalent to /readonly except that in
1110 ACLs where /readonly is required you may specify /leaky/noupdate or
1111 /strict/noupdate which are treated the same as /readonly.
1113 A useful new feature is the /count= option. This is a generalization
1114 of the per_byte option, so that you can measure the throughput of other
1115 aggregate values. For example, the per_byte option is now equivalent
1116 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
1118 The per_rcpt option has been generalized using the /count= mechanism
1119 (though it's more complicated than the per_byte equivalence). When it is
1120 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
1121 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
1122 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
1123 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
1124 Note that using per_rcpt with a non-readonly update mode in more than
1125 one ACL will cause the recipients to be double-counted. (The per_mail
1126 and per_byte options don't have this problem.)
1128 The handling of very low rates has changed slightly. If the computed rate
1129 is less than the event's count (usually one) then this event is the first
1130 after a long gap. In this case the rate is set to the same as this event's
1131 count, so that the first message of a spam run is counted properly.
1133 The major new feature is a mechanism for counting the rate of unique
1134 events. The new per_addr option counts the number of different
1135 recipients that someone has sent messages to in the last time period. It
1136 behaves like per_rcpt if all the recipient addresses are different, but
1137 duplicate recipient addresses do not increase the measured rate. Like
1138 the /count= option this is a general mechanism, so the per_addr option
1139 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
1140 example, measure the rate that a client uses different sender addresses
1141 with the options per_mail/unique=$sender_address. There are further
1142 details in the main documentation.
1144 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
1146 TF/06 Removed a few PCRE remnants.
1148 TF/07 Automatically extract Exim's version number from tags in the git
1149 repository when doing development or release builds.
1151 PP/02 Raise smtp_cmd_buffer_size to 16kB.
1152 Bugzilla 879. Patch from Paul Fisher.
1154 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
1155 Heavily based on revision 40f9a89a from Simon Arlott's tree.
1158 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
1160 PP/05 Variable $av_failed, true if the AV scanner deferred.
1161 Bugzilla 1078. Patch from John Horne.
1163 PP/06 Stop make process more reliably on build failure.
1164 Bugzilla 1087. Patch from Heiko Schlittermann.
1166 PP/07 Make maildir_use_size_file an _expandable_ boolean.
1167 Bugzilla 1089. Patch from Heiko Schlittermann.
1169 PP/08 Handle ${run} returning more data than OS pipe buffer size.
1170 Bugzilla 1131. Patch from Holger Weiß.
1172 PP/09 Handle IPv6 addresses with SPF.
1173 Bugzilla 860. Patch from Wolfgang Breyha.
1175 PP/10 GnuTLS: support TLS 1.2 & 1.1.
1177 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
1180 PP/11 match_* no longer expand right-hand-side by default.
1181 New compile-time build option, EXPAND_LISTMATCH_RHS.
1182 New expansion conditions, "inlist", "inlisti".
1184 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
1186 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
1188 PP/14 fix log_write() format string regression from TF/03.
1189 Bugzilla 1152. Patch from Dmitry Isaikin.
1195 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
1197 PP/02 Harmonised TLS library version reporting; only show if debugging.
1198 Layout now matches that introduced for other libraries in 4.74 PP/03.
1200 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
1202 PP/04 New "dns_use_edns0" global option.
1204 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
1207 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
1208 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
1210 TK/01 Updated PolarSSL code to 0.14.2.
1211 Bugzilla 1097. Patch from Andreas Metzler.
1213 PP/07 Catch divide-by-zero in ${eval:...}.
1214 Fixes bugzilla 1102.
1216 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
1219 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
1220 format-string attack -- SECURITY: remote arbitrary code execution.
1222 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
1223 time unintentionally subject to list matching rules, letting the header
1224 cause arbitrary Exim lookups (of items which can occur in lists, *not*
1225 arbitrary string expansion). This allowed for information disclosure.
1227 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
1228 INT_MIN/-1 -- value coerced to INT_MAX.
1234 NM/01 Workround for PCRE version dependency in version reporting
1237 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
1238 This fixes portability to compilers other than gcc, notably
1239 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
1241 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
1242 makefiles for portability to HP-UX and POSIX correctness.
1244 PP/01 Permit LOOKUP_foo enabling on the make command-line.
1245 Also via indented variable definition in the Makefile.
1246 (Debugging by Oliver Heesakkers).
1248 PP/02 Restore caching of spamd results with expanded spamd_address.
1249 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
1251 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
1252 Improves build reliability. Fix from: Frank Elsner
1254 NM/02 Fix wide character breakage in the rfc2047 coding
1255 Fixes bug 1064. Patch from Andrey N. Oktyabrski
1257 NM/03 Allow underscore in dnslist lookups
1258 Fixes bug 1026. Patch from Graeme Fowler
1260 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
1261 Code patches from Adam Ciarcinski of NetBSD.
1263 NM/04 Fixed exiqgrep to cope with mailq missing size issue
1266 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
1267 is logged, to avoid truncation. Patch from John Horne.
1269 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
1270 Patch from Jakob Hirsch.
1272 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
1273 SQL string expansion failure details.
1274 Patch from Andrey Oktyabrski.
1276 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
1277 Patch from Simon Arlott.
1279 PP/09 New lookups functionality failed to compile on old gcc which rejects
1280 extern declarations in function scope.
1281 Patch from Oliver Fleischmann
1283 PP/10 Use sig_atomic_t for flags set from signal handlers.
1284 Check getgroups() return and improve debugging.
1285 Fixed developed for diagnosis in bug 927 (which turned out to be
1288 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
1289 Patch from Mark Zealey.
1291 PP/12 Bugzilla 1056: Improved spamd server selection.
1292 Patch from Mark Zealey.
1294 PP/13 Bugzilla 1086: Deal with maildir quota file races.
1295 Based on patch from Heiko Schlittermann.
1297 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
1298 Patch from Uwe Doering, sign-off by Michael Haardt.
1300 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
1301 variable declaration deep within a block. Bug and patch from
1304 PP/15 lookups-Makefile IRIX compatibilty coercion.
1306 PP/16 Make DISABLE_DKIM build knob functional.
1308 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
1309 Patch by Simon Arlott
1311 TF/03 Fix valgrind.h portability to C89 compilers that do not support
1312 variable argument macros. Our copy now differs from upstream.
1318 TF/01 Failure to get a lock on a hints database can have serious
1319 consequences so log it to the panic log.
1321 TF/02 Log LMTP confirmation messages in the same way as SMTP,
1322 controlled using the smtp_confirmation log selector.
1324 TF/03 Include the error message when we fail to unlink a spool file.
1326 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
1327 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
1328 for maintaining out-of-tree patches for some time.
1330 PP/01 Bugzilla 139: Documentation and portability issues.
1331 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
1332 Handle per-OS dynamic-module compilation flags.
1334 PP/02 Let /dev/null have normal permissions.
1335 The 4.73 fixes were a little too stringent and complained about the
1336 permissions on /dev/null. Exempt it from some checks.
1337 Reported by Andreas M. Kirchwitz.
1339 PP/03 Report version information for many libraries, including
1340 Exim version information for dynamically loaded libraries. Created
1341 version.h, now support a version extension string for distributors
1342 who patch heavily. Dynamic module ABI change.
1344 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
1345 privilege escalation vulnerability whereby the Exim run-time user
1346 can cause root to append content of the attacker's choosing to
1349 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
1352 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
1353 If dropping privileges for untrusted macros, we disabled normal logging
1354 on the basis that it would fail; for the Exim run-time user, this is not
1355 the case, and it resulted in successful deliveries going unlogged.
1356 Fixed. Reported by Andreas Metzler.
1362 PP/01 Date: & Message-Id: revert to normally being appended to a message,
1363 only prepend for the Resent-* case. Fixes regression introduced in
1364 Exim 4.70 by NM/22 for Bugzilla 607.
1366 PP/02 Include check_rfc2047_length in configure.default because we're seeing
1367 increasing numbers of administrators be bitten by this.
1369 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
1371 PP/03 Bugzilla 994: added openssl_options main configuration option.
1373 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
1375 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
1377 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
1379 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
1380 without a peer certificate, leading to a segfault because of an
1381 assumption that peers always have certificates. Be a little more
1382 paranoid. Problem reported by Martin Tscholak.
1384 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
1385 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
1386 NB: ClamAV planning to remove STREAM in "middle of 2010".
1387 CL also introduces -bmalware, various -d+acl logging additions and
1388 more caution in buffer sizes.
1390 PP/09 Implemented reverse_ip expansion operator.
1392 PP/10 Bugzilla 937: provide a "debug" ACL control.
1394 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
1396 PP/12 Bugzilla 973: Implement --version.
1398 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
1400 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
1402 PP/15 Bugzilla 816: support multiple condition rules on Routers.
1404 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
1405 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
1406 ignore trailing whitespace.
1408 JJ/02 prevent non-panic DKIM error from being sent to paniclog
1410 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
1413 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
1414 Notification from Dr Andrew Aitchison.
1416 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
1417 ExtendedDetectionInfo response format.
1418 Notification from John Horne.
1420 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
1423 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
1424 XSL and documented dependency on system catalogs, with examples of how
1427 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
1430 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
1431 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
1432 configuration file which is writeable by the Exim user or group.
1434 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
1435 of configuration files to cover files specified with the -C option if
1436 they are going to be used with root privileges, not just the default
1439 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
1440 option (effectively making it always true).
1442 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
1443 files to be used while preserving root privileges.
1445 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
1446 that rogue child processes cannot use them.
1448 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
1449 run-time user, instead of root.
1451 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
1452 Exim run-time user without dropping privileges.
1454 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
1455 result string, instead of calling string_vformat() twice with the same
1458 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
1459 for other users. Others should always drop root privileges if they use
1460 -C on the command line, even for a whitelisted configure file.
1462 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
1464 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
1470 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
1471 $data_path, and $header_path variables; fixed documentation bugs and
1474 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
1475 exipick to access non-standard spools, including the "frozen" queue
1478 NM/01 Bugzilla 965: Support mysql stored procedures.
1479 Patch from Alain Williams
1481 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
1483 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
1484 Patch from Andreas Metzler
1486 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
1487 Patch from Kirill Miazine
1489 NM/05 Bugzilla 671: Added umask to procmail example.
1491 JJ/03 installed exipick 20100323.0, fixing doc bug
1493 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
1494 directory. Notification and patch from Dan Rosenberg.
1496 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
1498 TK/02 Improve log output when DKIM signing operation fails.
1500 MH/01 Treat the transport option dkim_domain as a colon separated
1501 list, not as a single string, and sign the message with each element,
1502 omitting multiple occurences of the same signer.
1504 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
1505 Bugzilla 985, 986. Patch by Simon Arlott
1507 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
1508 Patch by Simon Arlott
1510 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
1511 MBX locking. Notification from Dan Rosenberg.
1517 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
1519 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
1521 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
1523 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
1525 NM/04 Bugzilla 915: Flag broken perl installation during build.
1531 TK/01 Added patch by Johannes Berg that expands the main option
1532 "spamd_address" if it starts with a dollar sign.
1534 TK/02 Write list of recipients to X-Envelope-Sender header when building
1535 the mbox-format spool file for content scanning (suggested by Jakob
1538 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
1539 (http://www.dcc-servers.net/) support via dccifd. Activated by
1540 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
1542 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
1543 by Mark Daniel Reidel <mr@df.eu>.
1545 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
1546 When building exim an external PCRE library is now needed -
1547 PCRE is a system library on the majority of modern systems.
1548 See entry on PCRE_LIBS in EDITME file.
1550 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
1551 conversation. Added nologin parameter to request.
1552 Patch contributed by Kirill Miazine.
1554 TF/01 Do not log submission mode rewrites if they do not change the address.
1556 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
1558 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
1559 log files in place. Contributed by Roberto Lima.
1561 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
1563 TF/03 Bugzilla 615: When checking the local_parts router precondition
1564 after a local_part_suffix or local_part_prefix option, Exim now
1565 does not use the address's named list lookup cache, since this
1566 contains cached lookups for the whole local part.
1568 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
1569 Robert Millan. Documentation is in experimental-spec.txt.
1571 TF/04 Bugzilla 668: Fix parallel build (make -j).
1573 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
1575 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
1576 Patch provided by Jan Srzednicki.
1578 TF/05 Leading white space used to be stripped from $spam_report which
1579 wrecked the formatting. Now it is preserved.
1581 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
1582 that they are available at delivery time.
1584 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
1586 TF/08 TLS error reporting now respects the incoming_interface and
1587 incoming_port log selectors.
1589 TF/09 Produce a more useful error message if an SMTP transport's hosts
1590 setting expands to an empty string.
1592 NM/06 Bugzilla 744: EXPN did not work under TLS.
1593 Patch provided by Phil Pennock.
1595 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
1596 Patch provided by Richard Godbee.
1598 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
1599 acl_smtp_notquit, added index entry.
1601 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
1602 Patch provided by Eugene Bujak.
1604 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
1605 accept(). Patch provided by Maxim Dounin.
1607 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
1608 Patch provided by Phil Pennock.
1610 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
1612 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
1613 Patch provided by Brad "anomie" Jorsch.
1615 NM/14 Bugzilla 622: Added timeout setting to transport filter.
1616 Patch provided by Dean Brooks.
1618 TK/05 Add native DKIM support (does not depend on external libraries).
1620 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
1621 Patch provided by Graeme Fowler.
1623 NM/16 Bugzilla 851: Documentation example syntax fix.
1625 NM/17 Changed NOTICE file to remove references to embedded PCRE.
1627 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
1630 NM/19 Bugzilla 745: TLS version reporting.
1631 Patch provided by Phil Pennock.
1633 NM/20 Bugzilla 167: bool: condition support.
1634 Patch provided by Phil Pennock.
1636 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
1637 clients. Patch provided by Phil Pennock.
1639 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
1640 Patch provided by Brad "anomie" Jorsch.
1642 NM/23 Bugzilla 687: Fix misparses in eximstats.
1643 Patch provided by Heiko Schlittermann.
1645 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
1646 Patch provided by Heiko Schlittermann.
1648 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
1649 plus update to original patch.
1651 NM/26 Bugzilla 799: Documentation correction for ratelimit.
1653 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
1654 Patch provided by David Brownlee.
1656 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
1658 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
1660 NM/30 Bugzilla 888: TLS documentation bugfixes.
1662 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
1664 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
1665 Unlike the original bugzilla I have changed all shell scripts in src tree.
1667 NM/33 Bugzilla 898: Transport filter timeout fix.
1668 Patch by Todd Rinaldo.
1670 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
1671 Patch by Serge Demonchaux.
1673 NM/35 Bugzilla 39: Base64 decode bug fixes.
1674 Patch by Jakob Hirsch.
1676 NM/36 Bugzilla 909: Correct connect() call in dcc code.
1678 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
1680 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
1682 NM/39 Bugzilla 911: Fixed MakeLinks build script.
1688 TK/01 Add preliminary DKIM support. Currently requires a forked version of
1689 ALT-N's libdkim that I have put here:
1690 http://duncanthrax.net/exim-experimental/
1692 Note to Michael Haardt: I had to rename some vars in sieve.c. They
1693 were called 'true' and it seems that C99 defines that as a reserved
1694 keyword to be used with 'bool' variable types. That means you could
1695 not include C99-style headers which use bools without triggering
1696 build errors in sieve.c.
1698 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
1699 as mailq or other aliases. Changed the --help handling significantly
1700 to do whats expected. exim_usage() emits usage/help information.
1702 SC/01 Added the -bylocaldomain option to eximstats.
1704 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
1706 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
1708 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
1714 PH/01 Another patch from the Sieve maintainer.
1716 PH/02 When an IPv6 address is converted to a string for single-key lookup
1717 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
1718 dots are used instead of colons so that keys in lsearch files need not
1719 contain colons. This was done some time before quoting was made available
1720 in lsearch files. However, iplsearch files do require colons in IPv6 keys
1721 (notated using the quote facility) so as to distinguish them from IPv4
1722 keys. This meant that lookups for IP addresses in host lists did not work
1723 for iplsearch lookups.
1725 This has been fixed by arranging for IPv6 addresses to be expressed with
1726 colons if the lookup type is iplsearch. This is not incompatible, because
1727 previously such lookups could never work.
1729 The situation is now rather anomolous, since one *can* have colons in
1730 ordinary lsearch keys. However, making the change in all cases is
1731 incompatible and would probably break a number of configurations.
1733 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
1736 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
1737 conversion specification without a maximum field width, thereby enabling
1738 a rogue spamd server to cause a buffer overflow. While nobody in their
1739 right mind would setup Exim to query an untrusted spamd server, an
1740 attacker that gains access to a server running spamd could potentially
1741 exploit this vulnerability to run arbitrary code as the Exim user.
1743 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
1744 $primary_hostname instead of what libspf2 thinks the hosts name is.
1746 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
1747 a directory entry by the name of the lookup key. Previously, if a
1748 symlink pointed to a non-existing file or a file in a directory that
1749 Exim lacked permissions to read, a lookup for a key matching that
1750 symlink would fail. Now it is enough that a matching directory entry
1751 exists, symlink or not. (Bugzilla 503.)
1753 PH/03 The body_linecount and body_zerocount variables are now exported in the
1756 PH/04 Added the $dnslist_matched variable.
1758 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
1759 This means they are set thereafter only if the connection becomes
1762 PH/06 Added the client_condition to authenticators so that some can be skipped
1763 by clients under certain conditions.
1765 PH/07 The error message for a badly-placed control=no_multiline_responses left
1766 "_responses" off the end of the name.
1768 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
1770 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
1771 (without spaces) instead of just copying the configuration text.
1773 PH/10 Added the /noupdate option to the ratelimit ACL condition.
1775 PH/11 Added $max_received_linelength.
1777 PH/12 Added +ignore_defer and +include_defer to host lists.
1779 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
1780 way in which PCRE > 7.0 is built.
1782 PH/14 Implemented queue_only_load_latch.
1784 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
1785 MAIL command. The effect was to mangle the value on 64-bit systems.
1787 PH/16 Another patch from the Sieve maintainer.
1789 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
1791 PH/18 If a system quota error occurred while trying to create the file for
1792 a maildir delivery, the message "Mailbox is full" was not appended to the
1793 bounce if the delivery eventually timed out. Change 4.67/27 below applied
1794 only to a quota excession during the actual writing of the file.
1796 PH/19 It seems that peer DN values may contain newlines (and other non-printing
1797 characters?) which causes problems in log lines. The DN values are now
1798 passed through string_printing() before being added to log lines.
1800 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
1801 and InterBase are left for another time.)
1803 PH/21 Added message_body_newlines option.
1805 PH/22 Guard against possible overflow in moan_check_errorcopy().
1807 PH/23 POSIX allows open() to be a macro; guard against that.
1809 PH/24 If the recipient of an error message contained an @ in the local part
1810 (suitably quoted, of course), incorrect values were put in $domain and
1811 $local_part during the evaluation of errors_copy.
1817 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
1818 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
1821 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
1822 issue a MAIL command.
1824 PH/02 In an ACL statement such as
1826 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
1828 if a client was not listed at all, or was listed with a value other than
1829 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
1830 the condition was not true (as it should be), so access was not denied.
1831 The bug was that the ! inversion was incorrectly passed on to the second
1832 item. This has been fixed.
1834 PH/03 Added additional dnslists conditions == and =& which are different from
1835 = and & when the dns lookup returns more than one IP address.
1837 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
1838 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
1840 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
1841 FSYNC, which compiles an option called disable_fsync that allows for
1842 bypassing fsync(). The documentation is heavily laced with warnings.
1844 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
1846 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
1847 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
1848 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
1849 including adding "make clean"; (3) Added -fPIC when compiling the test
1850 dynamically loaded module, to get rid of a warning.
1852 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
1853 message fails, move_frozen_messages = true and ignore_bounce_errors_after
1854 = 0s. The bug is otherwise harmless.
1856 PH/07 There was a bug in the dovecot authenticator such that the value of
1857 $auth1 could be overwritten, and so not correctly preserved, after a
1858 successful authentication. This usually meant that the value preserved by
1859 the server_setid option was incorrect.
1861 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
1863 PH/09 Installed PCRE release 7.0.
1865 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
1866 run for batched SMTP input. It is now run at the start of every message
1867 in the batch. While fixing this I discovered that the process information
1868 (output by running exiwhat) was not always getting set for -bs and -bS
1869 input. This is fixed, and it now also says "batched" for BSMTP.
1871 PH/11 Added control=no_pipelining.
1873 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
1874 patch, slightly modified), and move the expansion of helo_data till after
1875 the connection is made in the smtp transport (so it can use these
1878 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
1880 PH/14 Added log_selector = +pid.
1882 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
1884 PH/16 Add ${if forany and ${if forall.
1886 PH/17 Added dsn_from option to vary the From: line in DSNs.
1888 PH/18 Flush SMTP output before performing a callout, unless control =
1889 no_callout_flush is set.
1891 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
1892 was true (the default) a successful delivery failed to delete the retry
1893 item, thus causing premature timeout of the address. The bug is now
1896 PH/20 Added hosts_avoid_pipelining to the smtp transport.
1898 PH/21 Long custom messages for fakedefer and fakereject are now split up
1899 into multiline reponses in the same way that messages for "deny" and
1900 other ACL rejections are.
1902 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
1903 with slight modification.
1905 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
1906 draft, changing the syntax and factoring some duplicate code".
1908 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
1909 for deliveries of the second and subsequent messages over the same SMTP
1912 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
1913 ${reduce, with only minor "tidies".
1915 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
1917 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
1918 expansion side effects.
1920 PH/27 When a message times out after an over-quota error from an Exim-imposed
1921 quota, the bounce message says "mailbox is full". This message was not
1922 being given when it was a system quota that was exceeded. It now should
1925 MH/03 Made $recipients available in local_scan(). local_scan() already has
1926 better access to the recipient list through recipients_list[], but
1927 $recipients can be useful in postmaster-provided expansion strings.
1929 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
1930 in the case of a MAIL command with additional options following the
1931 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
1932 were accidentally chopped off.
1934 PH/29 SMTP synchronization checks are implemented when a command is read -
1935 there is a check that no more input is waiting when there shouldn't be
1936 any. However, for some commands, a delay in an ACL can mean that it is
1937 some time before the response is written. In this time, more input might
1938 arrive, invalidly. So now there are extra checks after an ACL has run for
1939 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
1940 pipelining has not been advertised.
1942 PH/30 MH's patch to allow iscntrl() characters to be list separators.
1944 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
1945 returned in the SMTP response when smtp_return_error_details was false.
1946 This has been fixed.
1948 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
1949 instead of the C I/O that was originally supplied, because problems were
1950 reported on Solaris.
1952 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
1953 Exim which did not show up earlier: it was assuming that a call to
1954 SSL_CTX_set_info_callback() might give an error value. In fact, there is
1955 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
1956 was a macro that became an assignment, so it seemed to work. This has
1957 changed to a proper function call with a void return, hence the compile
1958 error. Exim's code has been fixed.
1960 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
1963 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
1965 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
1967 PH/37 If a message is not accepted after it has had an id assigned (e.g.
1968 because it turns out to be too big or there is a timeout) there is no
1969 "Completed" line in the log. When some messages of this type were
1970 selected by exigrep, they were listed as "not completed". Others were
1971 picked up by some special patterns. I have improved the selection
1972 criteria to be more general.
1974 PH/38 The host_find_failed option in the manualroute router can now be set
1975 to "ignore", to completely ignore a host whose IP address cannot be
1976 found. If all hosts are ignored, the behaviour is controlled by the new
1977 host_all_ignored option.
1979 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
1980 homing or because of multiple MX records with /mx) generated more than
1981 one IP address, and the following item turned out to be the local host,
1982 all the secondary addresses of the first item were incorrectly removed
1983 from the list, along with the local host and any following hosts (which
1984 is what is supposed to happen).
1986 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
1987 whoever called Exim into the -H file. In the case of the daemon it was
1988 behaving confusingly. When first started, it used values for whoever
1989 started the daemon, but after a SIGHUP it used the Exim user (because it
1990 calls itself on a restart). I have changed the code so that it now always
1993 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
1994 message are rejected with the same error (e.g. no authentication or bad
1995 sender address), and a DATA command is nevertheless sent (as can happen
1996 with PIPELINING or a stupid MUA), the error message that was given to the
1997 RCPT commands is included in the rejection of the DATA command. This is
1998 intended to be helpful for MUAs that show only the final error to their
2001 PH/42 Another patch from the Sieve maintainer.
2003 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
2004 Eximstats - Fixed some broken HTML links and added missing column headers
2006 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
2007 columns for Rejects, Temp Rejects, Ham, and Spam rows.
2009 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
2011 PH/43 Yet another patch from the Sieve maintainer.
2013 PH/44 I found a way to check for a TCP/IP connection going away before sending
2014 the response to the final '.' that terminates a message, but only in the
2015 case where the client has not sent further data following the '.'
2016 (unfortunately, this is allowed). However, in many cases there won't be
2017 any further data because there won't be any more messages to send. A call
2018 to select() can be used: if it shows that the input is "ready", there is
2019 either input waiting, or the socket has been closed. An attempt to read
2020 the next input character can distinguish the two cases. Previously, Exim
2021 would have sent an OK response which the client would never have see.
2022 This could lead to message repetition. This fix should cure that, at
2023 least in a lot of common cases.
2025 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
2026 advertised in response to EHLO.
2032 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
2033 fixed by 4.65/MH/01 (is this a record?) are fixed:
2035 (i) An empty string was always treated as zero by the numeric comparison
2036 operators. This behaviour has been restored.
2038 (ii) It is documented that the numeric comparison operators always treat
2039 their arguments as decimal numbers. This was broken in that numbers
2040 starting with 0 were being interpreted as octal.
2042 While fixing these problems I realized that there was another issue that
2043 hadn't been noticed. Values of message_size_limit (both the global option
2044 and the transport option) were treated as octal if they started with 0.
2045 The documentation was vague. These values are now always treated as
2046 decimal, and I will make that clear in the documentation.
2052 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
2053 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
2056 MH/01 Don't check that the operands of numeric comparison operators are
2057 integers when their expansion is in "skipping" mode (fixes bug
2058 introduced by 4.64-PH/07).
2060 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
2061 child addresses, Exim now panics and dies. Previously, because the count
2062 is held in a short int, deliveries were likely to be lost. As such a
2063 large number of recipients for a single message is ridiculous
2064 (performance will be very, very poor), I have chosen to impose a limit
2065 rather than extend the field.
2071 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
2072 leftover -K file (the existence of which was triggered by #402).
2073 While we were at it, introduced process PID as part of the -K
2074 filename. This should rule out race conditions when creating
2077 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
2078 processing considerably. Previous code took too long for large mails,
2079 triggering a timeout which in turn triggers #401.
2081 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
2082 in the DK code in transports.c. sendfile() is not really portable,
2083 hence the _LINUX specificness.
2085 TF/01 In the add_headers option to the mail command in an Exim filter,
2086 there was a bug that Exim would claim a syntax error in any
2087 header after the first one which had an odd number of characters
2090 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
2091 callout verification, Exim cached a "reject" for the entire domain. This
2092 is correct for most verifications, but it is not correct for a recipient
2093 verification with use_sender or use_postmaster set, because in that case
2094 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
2095 case of MAIL FROM:<> rejection from other early rejections (e.g.
2096 rejection of HELO). When verifying a recipient using a non-null MAIL
2097 address, the cache is ignored if it shows MAIL FROM:<> rejection.
2098 Whatever the result of the callout, the value of the domain cache is
2099 left unchanged (for any other kind of callout, getting as far as trying
2100 RCPT means that the domain itself is ok).
2102 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
2105 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
2106 manifest itself as EPIPE rather than ECONNECT. When tidying away a
2107 session, the daemon ignores ECONNECT errors and logs others; it now
2108 ignores EPIPE as well.
2110 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
2111 (quoted-printable decoding).
2113 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
2114 later the small subsequent patch to fix an introduced bug.
2116 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
2118 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
2120 PH/08 An error is now given if message_size_limit is specified negative.
2122 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
2123 to be given (somewhat) arbitrary names.
2125 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
2128 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
2129 miscellaneous code fixes
2131 PH/10 Added the log_reject_target ACL modifier to specify where to log
2134 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
2135 hostname. This is wrong, because it relates to the incoming message (and
2136 probably the interface on which it is arriving) and not to the outgoing
2137 callout (which could be using a different interface). This has been
2138 changed to use the value of the helo_data option from the smtp transport
2139 instead - this is what is used when a message is actually being sent. If
2140 there is no remote transport (possible with a router that sets up host
2141 addresses), $smtp_active_hostname is used.
2143 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
2144 tweaks were necessary in order to get it to work (see also 21 below):
2145 (a) The code assumed that strncpy() returns a negative number on buffer
2146 overflow, which isn't the case. Replaced with Exim's string_format()
2148 (b) There were several signed/unsigned issues. I just did the minimum
2149 hacking in of casts. There is scope for a larger refactoring.
2150 (c) The code used strcasecmp() which is not a standard C function.
2151 Replaced with Exim's strcmpic() function.
2152 (d) The code set only $1; it now sets $auth1 as well.
2153 (e) A simple test gave the error "authentication client didn't specify
2154 service in request". It would seem that Dovecot has changed its
2155 interface. Fortunately there's a specification; I followed it and
2156 changed what the client sends and it appears to be working now.
2158 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
2161 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
2162 address A is aliased to B and C, where B exists and C does not. Without
2163 -v the output is "A verified" because verification stops after a
2164 successful redirection if more than one address is generated. However,
2165 with -v the child addresses are also verified. Exim was outputting "A
2166 failed to verify" and then showing the successful verification for C,
2167 with its parentage. It now outputs "B failed to verify", showing B's
2168 parentage before showing the successful verification of C.
2170 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
2171 look up a TXT record in a specific list after matching in a combined
2174 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
2175 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
2176 they consult the DNS. I had assumed they would set it the way they
2177 wanted; and indeed my experiments on Linux seem to show that in some
2178 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
2179 To be on the safe side, however, I have now made the interface to
2180 host_find_byname() similar to host_find_bydns(), with an argument
2181 containing the DNS resolver options. The host_find_byname() function now
2182 sets these options at its start, just as host_find_bydns() does. The smtp
2183 transport options dns_qualify_single and dns_search_parents are passed to
2184 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
2185 of host_find_byname() use the default settings of RES_DEFNAMES
2186 (qualify_single) but not RES_DNSRCH (search_parents).
2188 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
2189 spool_read_header() do less string testing, by means of a preliminary
2190 switch on the second character of optional "-foo" lines. (This is
2191 overdue, caused by the large number of possibilities that now exist.
2192 Originally there were few.) While I was there, I also converted the
2193 str(n)cmp tests so they don't re-test the leading "-" and the first
2194 character, in the hope this might squeeze out yet more improvement.
2196 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
2197 flag allowing group syntax was set by the header_syntax check but not
2198 turned off, possible causing trouble later; (2) The flag was not being
2199 set at all for the header_verify test, causing "group"-style headers to
2200 be rejected. I have now set it in this case, and also caused header_
2201 verify to ignore an empty address taken from a group. While doing this, I
2202 came across some other cases where the code for allowing group syntax
2203 while scanning a header line wasn't quite right (mostly, not resetting
2204 the flag correctly in the right place). These bugs could have caused
2205 trouble for malformed header lines. I hope it is now all correct.
2207 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
2208 with the "reply" argument non-NULL. The code, however (which originally
2209 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
2210 but it didn't always do it. This confused somebody who was copying the
2211 code for some other use. I have removed all the tests.
2213 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
2214 feature that was used to support insecure browsers during the U.S. crypto
2215 embargo. It requires special client support, and Exim is probably the
2216 only MTA that supported it -- and would never use it because real RSA is
2217 always available. This code has been removed, because it had the bad
2218 effect of slowing Exim down by computing (never used) parameters for the
2219 RSA_EXPORT functionality.
2221 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
2222 authenticator to fail if there's a tab character in the incoming data
2223 (there should never be unless someone is messing about, as it's supposed
2224 to be base64-encoded). Also added, on Timo's advice, the "secured" option
2225 if the connection is using TLS or if the remote IP is the same as the
2226 local IP, and the "valid-client-cert option" if a client certificate has
2229 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
2230 authenticators. This can be used for authorization after authentication
2231 succeeds. (In the case of plaintext, it servers for both authentication
2234 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
2235 if any retry times were supplied.
2237 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
2238 connection, where there is no client IP address to check. In this
2239 situation, the verify now always succeeds.
2241 PH/25 Applied John Jetmore's -Mset patch.
2243 PH/26 Added -bem to be like -Mset, but loading a message from a file.
2245 PH/27 In a string expansion for a processed (not raw) header when multiple
2246 headers of the same name were present, leading whitespace was being
2247 removed from all of them, but trailing whitespace was being removed only
2248 from the last one. Now trailing whitespace is removed from each header
2249 before concatenation. Completely empty headers in a concatenation (as
2250 before) are ignored.
2252 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
2253 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
2255 PH/29 [Removed. This was a change that I later backed out, and forgot to
2256 correct the ChangeLog entry (that I had efficiently created) before
2257 committing the later change.]
2259 PH/30 Exim was sometimes attempting to deliver messages that had suffered
2260 address errors (4xx response to RCPT) over the same connection as other
2261 messages routed to the same hosts. Such deliveries are always "forced",
2262 so retry times are not inspected. This resulted in far too many retries
2263 for the affected addresses. The effect occurred only when there were more
2264 hosts than the hosts_max_try setting in the smtp transport when it had
2265 the 4xx errors. Those hosts that it had tried were not added to the list
2266 of hosts for which the message was waiting, so if all were tried, there
2267 was no problem. Two fixes have been applied:
2269 (i) If there are any address or message errors in an SMTP delivery, none
2270 of the hosts (tried or untried) are now added to the list of hosts
2271 for which the message is waiting, so the message should not be a
2272 candidate for sending over the same connection that was used for a
2273 successful delivery of some other message. This seems entirely
2274 reasonable: after all the message is NOT "waiting for some host".
2275 This is so "obvious" that I'm not sure why it wasn't done
2276 previously. Hope I haven't missed anything, but it can't do any
2277 harm, as the worst effect is to miss an optimization.
2279 (ii) If, despite (i), such a delivery is accidentally attempted, the
2280 routing retry time is respected, so at least it doesn't keep
2281 hammering the server.
2283 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
2284 in ${readsocket because some servers need this prod.
2286 PH/32 Added some extra debug output when updating a wait-xxx database.
2288 PH/33 The hint "could be header name not terminated by colon", which has been
2289 given for certain expansion errors for a long time, was not being given
2290 for the ${if def:h_colon_omitted{... case.
2292 PH/34 The spec says: "With one important exception, whenever a domain list is
2293 being scanned, $domain contains the subject domain." There was at least
2294 one case where this was not true.
2296 PH/35 The error "getsockname() failed: connection reset by peer" was being
2297 written to the panic log as well as the main log, but it isn't really
2298 panic-worthy as it just means the connection died rather early on. I have
2299 removed the panic log writing for the ECONNRESET error when getsockname()
2302 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
2303 runs only) independently of the message's sender address. This meant
2304 that, if the 4xx error was in fact related to the sender, a different
2305 message to the same recipient with a different sender could confuse
2306 things. In particualar, this can happen when sending to a greylisting
2307 server, but other circumstances could also provoke similar problems.
2308 I have changed the default so that the retry time for these errors is now
2309 based a combination of the sender and recipient addresses. This change
2310 can be overridden by setting address_retry_include_sender=false in the
2313 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
2314 remote server are returned as part of bounce messages. This was not
2315 happening for LMTP over a pipe (the lmtp transport), but now it is the
2316 same for both kinds of LMTP.
2318 PH/38 Despite being documented as not happening, Exim was rewriting addresses
2319 in header lines that were in fact CNAMEs. This is no longer the case.
2321 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
2322 and queue runs started by the daemon processed all messages. This has
2323 been fixed so that -R and -S can now usefully be given with -q<time>.
2325 PH/40 Import PCRE release 6.7 (fixes some bugs).
2327 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
2329 PH/42 Give an error if -q is specified more than once.
2331 PH/43 Renamed the variables $interface_address and $interface_port as
2332 $received_ip_address and $received_port, to make it clear that these
2333 values apply to message reception, and not to the outgoing interface when
2334 a message is delivered. (The old names remain recognized, of course.)
2336 PH/44 There was no timeout on the connect() call when using a Unix domain
2337 socket in the ${readsocket expansion. There now is.
2339 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
2340 be meaningful with "accept".
2342 SC/01 Eximstats V1.43
2343 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
2345 SC/02 Eximstats V1.44
2346 Use a glob alias rather than an array ref in the generated
2347 parser. This improves both readability and performance.
2349 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
2350 Collect SpamAssassin and rejection statistics.
2351 Don't display local sender or destination tables unless
2352 there is data to show.
2353 Added average volumes into the top table text output.
2355 SC/04 Eximstats V1.46
2356 Collect data on the number of addresses (recipients)
2357 as well as the number of messages.
2359 SC/05 Eximstats V1.47
2360 Added 'Message too big' to the list of mail rejection
2361 reasons (thanks to Marco Gaiarin).
2363 SC/06 Eximstats V1.48
2364 Mainlog lines which have GMT offsets and are too short to
2365 have a flag are now skipped.
2367 SC/07 Eximstats V1.49 (Alain Williams)
2368 Added the -emptyok flag.
2370 SC/08 Eximstats V1.50
2371 Fixes for obtaining the IP address from reject messages.
2373 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
2374 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
2375 whitesspace changes from 4.64-PH/27
2377 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
2380 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
2381 are found, allow negative numbers in numeric criteria)
2383 JJ/06 exipick.20061117.2, added new $message_body_missing variable
2385 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
2386 to match changes made in 4.64-PH/43
2388 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
2390 PH/47 Put in an explicit test for a DNS lookup of an address record where the
2391 "domain" is actually an IP address, and force a failure. This locks out
2392 those revolvers/nameservers that support "A-for-A" lookups, in
2393 contravention of the specifications.
2395 PH/48 When a host name was looked up from an IP address, and the subsequent
2396 forward lookup of the name timed out, the host name was left in
2397 $sender_host_name, contrary to the specification.
2399 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
2400 restricted to single-key lookups, Exim was not diagnosing an error if
2401 * or *@ was used with a query-style lookup.
2403 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
2405 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
2406 long ago, but noone interested enough thought of it. Let's just say that
2407 the "1.1" means that there are some new functions that weren't there at
2408 some point in the past.
2410 PH/51 Error processing for expansion failure of helo_data from an smtp
2411 transport during callout processing was broken.
2413 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
2414 tested/used via the -bh/-bhc/-bs options.
2416 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
2417 bug, fixed in subsequent PCRE releases).
2419 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
2420 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
2422 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
2428 SC/01 Use a glob alias rather than an array ref in eximstats generated
2429 parser. This improves both readability and performance.
2431 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
2432 Don't display local sender or destination tables in eximstats unless
2433 there is data to show.
2434 Added average volumes into the eximstats top table text output.
2436 SC/03 Collect data on the number of addresses (recipients) as well
2437 as the number of messages in eximstats.
2439 TF/01 Correct an error in the documentation for the redirect router. Exim
2440 does (usually) call initgroups() when daemonizing.
2442 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
2443 with consistent privilege compared to when running as a daemon.
2445 TF/03 Note in the spec that $authenticated_id is not set for local
2446 submissions from trusted users.
2448 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
2449 Thanks to Dean Brooks <dean@iglou.com> for the patch.
2451 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
2452 by adding some example configuration directives to the default
2453 configuration file. A little bit of work is required to uncomment the
2454 directives and define how usernames and passwords are checked, but
2455 there is now a framework to start from.
2457 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
2458 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
2459 without this. I don't know how relevant this is to other LDAP libraries.
2461 PH/02 Add the verb name to the "unknown ACL verb" error.
2463 PH/03 Magnus Holmgren's patch for filter_prepend_home.
2465 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
2467 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
2468 directory not expanded when it should be if an expanded home directory
2469 was set for the address (which is overridden by the transport).
2471 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
2474 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
2475 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
2476 because it is too late at that time, and has no effect.
2478 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
2479 security issue with \' (bugzilla #107). I could not use the
2480 PQescapeStringConn() function, because it needs a PGconn value as one of
2483 PH/08 When testing addresses using -bt, indicate those final addresses that
2484 are duplicates that would not cause an additional delivery. At least one
2485 person was confused, thinking that -bt output corresponded to deliveries.
2486 (Suppressing duplicates isn't a good idea as you lose the information
2487 about possibly different redirections that led to the duplicates.)
2489 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
2490 systems where poll() doesn't work, in particular OS X.
2492 PH/10 Added more information to debugging output for retry time not reached.
2494 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
2495 operations in malware.c.
2497 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
2500 PH/13 If write_rejectlog was set false when logging was sent to syslog with
2501 syslog_duplication set false, log lines that would normally be written
2502 both the the main log and to the reject log were not written to syslog at
2505 PH/14 In the default configuration, change the use of "message" in ACL warn
2506 statements to "add_header".
2508 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
2509 not followed by a command (e.g. "seen endif").
2511 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
2512 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
2515 PH/17 Added extra conditions to the default value of delay_warning_condition
2519 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
2520 { match{$h_precedence:}{(?i)bulk|list|junk} } \
2521 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
2524 The Auto-Submitted: and various List- headers are standardised, whereas I
2525 don't think Precedence: ever was.
2527 PH/18 Refactored debugging code in route_finduser() to show more information,
2528 in particular, the error code if getpwnam() issues one.
2530 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
2531 This is apparently needed in addition to the PH/07 change above to avoid
2532 any possible encoding problems.
2534 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
2535 but not after initializing Perl.
2537 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
2538 output them only if debugging. By default they are written stderr,
2539 apparently, which is not desirable.
2541 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
2544 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
2547 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
2549 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
2550 authenticated or an ident call has been made. Suppress the default
2551 values for $authenticated_id and $authenticated_sender (but permit -oMai
2552 and -oMas) when testing with -bh.
2554 PH/24 Re-jigged the order of the tests in the default configuration so that the
2555 tests for valid domains and recipients precede the DNS black list and CSA
2556 tests, on the grounds that those ones are more expensive.
2558 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
2559 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
2560 This bug exists in every version of Exim that I still have, right back to
2563 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
2564 However, an attempt to turn on case-sensitivity in a regex key by
2565 including (?-i) didn't work because the subject string was already
2566 lowercased, and the effects were non-intuitive. It turns out that a
2567 one-line patch can be used to allow (?-i) to work as expected.
2573 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
2574 other effects) broke the use of negated acl sub-conditions.
2576 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
2579 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
2580 "Deny" causes Exim to reject the incoming connection with a 554 error.
2581 Unfortunately, if there is a major crisis, such as a disk failure,
2582 tcp-wrappers gives "deny", whereas what one would like would be some
2583 kind of temporary error. A kludge has been added to help with this.
2584 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
2585 554 error is used if errno is still zero or contains ENOENT (which occurs
2586 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
2589 PH/03 Add -lutil to the default FreeBSD LIBS setting.
2591 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
2592 errors. Otherwise a message that provokes a temporary error (when other
2593 messages do not) can cause a whole host to time out.
2595 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
2596 addresses were routed directly to files or pipes from a redirect router.
2597 File deliveries just didn't batch; pipe deliveries might have suffered
2600 PH/06 A failure to get a lock for a hints database would erroneously always say
2601 "Failed to get write lock", even when it was really a read lock.
2603 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
2604 of 0600. This has been changed to use the value of the lockfile_mode
2605 option (which defaults to 0600).
2607 PH/08 Applied small patch from the Sieve maintainer.
2609 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
2610 folder from quota calculations, a direct delivery into this folder messed
2611 up the contents of the maildirsize file. This was because the regex was
2612 used only to exclude .Trash (or whatever) when the size of the mailbox
2613 was calculated. There was no check that a delivery was happening into an
2614 excluded directory. This bug has been fixed by ignoring all quota
2615 processing for deliveries into excluded directories.
2617 PH/10 Added the maildirfolder_create_regex option to appendfile.
2623 PH/01 The code for finding all the local interface addresses on a FreeBSD
2624 system running IPv6 was broken. This may well have applied to all BSD
2625 systems, as well as to others that have similar system calls. The broken
2626 code found IPv4 interfaces correctly, but gave incorrect values for the
2627 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
2628 that it would not match correctly against @[] and not recognize the IPv6
2631 PH/02 The ipliteral router was not recognizing addresses of the form user@
2632 [ipv6:....] because it didn't know about the "ipv6:" prefix.
2634 PH/03 Added disable_ipv6.
2636 PH/04 Changed $reply_address to use the raw form of the headers instead of the
2637 decoded form, because it is most often used to construct To: headers
2638 lines in autoreplies, and the decoded form may well be syntactically
2639 invalid. However, $reply_address has leading white space removed, and all
2640 newlines turned into spaces so that the autoreply transport does not
2643 PH/05 If group was specified without a user on a router, and no group or user
2644 was specified on a transport, the group from the router was ignored.
2646 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
2647 for visible compile-time settings that can be used to change these
2648 numbers, for those that want even more. Backwards compatibility with old
2649 spool files has been maintained. However, going back to a previous Exim
2650 release will lost any variables that are in spool files.
2652 PH/07 Two small changes when running in the test harness: increase delay when
2653 passing a TCP/IP connection to a new process, in case the original
2654 process has to generate a bounce, and remove special handling of
2655 127.0.0.2 (sic), which is no longer necessary.
2657 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
2658 be the same on different OS.
2660 PH/09 Moved a debug statement in filter processing to avoid a race problem when
2663 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
2664 whether --show-vars was specified or not
2666 JJ/02 exipick: Added support for new ACL variable spool format introduced
2669 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
2670 syntactically invalid From: or Reply-to: line, and a filter used this to
2671 generate an autoreply, and therefore failed to obtain an address for the
2672 autoreply, Exim could try to deliver to a non-existent relative file
2673 name, causing unrelated and misleading errors. What now happens is that
2674 it logs this as a hard delivery error, but does not attempt to create a
2677 PH/11 The exinext utility has a -C option for testing purposes, but although
2678 the given file was scanned by exinext itself; it wasn't being passed on
2679 when Exim was called.
2681 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
2682 an end-of-file indication when reading a command response.
2684 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
2685 compiled. In many other places in Exim, IPv6 addresses are always
2686 recognized, so I have changed this. It also means that IPv4 domain
2687 literals of the form [IPV4:n.n.n.n] are now always recognized.
2689 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
2690 used if the router is not running as root, for example, when verifying at
2691 ACL time, or when using -bh. The debugging output from this situation was
2692 non-existent - all you got was a failure to exec. I have made two
2695 (a) Failures to set uid/gid, the current directory, or a process leader
2696 in a subprocess such as that created by queryprogram now generate
2697 suitable debugging ouput when -d is set.
2699 (b) The queryprogram router detects when it is not running as root,
2700 outputs suitable debugging information if -d is set, and then runs
2701 the subprocess without attempting to change uid/gid.
2703 PH/15 Minor change to Makefile for building test_host (undocumented testing
2706 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
2707 additional section of a DNS packet that returns MX or SRV records.
2708 Instead, it always explicitly searches for A/AAAA records. This avoids
2709 major problems that occur when a DNS server includes only records of one
2710 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
2711 fixed another bug: if SRV records were looked up and the corresponding
2712 address records were *not* found in the additional section, the port
2713 values from the SRV records were lost.
2715 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
2716 using the correct key (the original address) when searching the retry
2717 rules in order to find which one to use for generating the retry hint.
2719 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
2720 adding the default one. Similarly, if it contains a Reply-To: header, the
2721 errors_reply_to option, if set, is not used.
2723 PH/19 When calculating a retry time, Exim used to measure the "time since
2724 failure" by looking at the "first failed" field in the retry record. Now
2725 it does not use this if it is later than than the arrival time of the
2726 message. Instead it uses the arrival time. This makes for better
2727 behaviour in cases where some deliveries succeed, thus re-setting the
2728 "first failed" field. An example is a quota failure for a huge message
2729 when small messages continue to be delivered. Without this change, the
2730 "time since failure" will always be short, possible causing more frequent
2731 delivery attempts for the huge message than are intended.
2732 [Note: This change was subsequently modified - see PH/04 for 4.62.]
2734 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
2735 $1, $2, $3) because the numerical variables can be reset during some
2736 expansion items (e.g. "match"), thereby losing the authentication data.
2738 PH/21 Make -bV show the size of off_t variables so that the test suite can
2739 decide whether to run tests for quotas > 2G.
2741 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
2742 mailbox_size, and mailbox_filecount in the appendfile transport. If a
2743 filecount value is greater than 2G or if a quota value is greater than 2G
2744 on a system where the size of off_t is not greater than 4, a panic error
2747 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
2748 never match. The debug and -bh output now contains an explicit error
2749 message indicating a malformed IPv4 address or mask.
2751 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
2752 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
2753 PH/23 above applies.
2755 PH/25 Do not write to syslog when running in the test harness. The only
2756 occasion when this arises is a failure to open the main or panic logs
2757 (for which there is an explicit test).
2759 PH/26 Added the /no_tell option to "control=freeze".
2761 PH/27 If a host name lookup failed very early in a connection, for example, if
2762 the IP address matched host_lookup and the reverse lookup yielded a name
2763 that did not have a forward lookup, an error message of the form "no IP
2764 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
2765 could be logged. Now it outputs the IP address instead of "NULL".
2767 PH/28 An enabling patch from MH: add new function child_open_exim2() which
2768 allows the sender and the authenticated sender to be set when
2769 submitting a message from within Exim. Since child_open_exim() is
2770 documented for local_scan(), the new function should be too.
2772 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
2773 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
2774 results in an empty string is now treated as unset.
2776 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
2778 PH/31 Added sender_verify_fail logging option.
2780 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
2781 needed by RFC 822 but not by RFC 2822 was commented out. I have now
2782 tidied the source and removed it altogether.
2784 PH/33 When a queue run was abandoned because the load average was too high, a
2785 log line was always written; now it is written only if the queue_run log
2786 selector is set. In addition, the log line for abandonment now contains
2787 information about the queue run such as the pid. This is always present
2788 in "start" and "stop" lines but was omitted from the "abandon" line.
2790 PH/34 Omit spaces between a header name and the colon in the error message that
2791 is given when verify = headers_syntax fails (if there are lots of them,
2792 the message gets confusing).
2794 PH/35 Change the default for dns_check_names_pattern to allow slashes within
2795 names, as there are now some PTR records that contain slashes. This check
2796 is only to protect against broken name servers that fall over on strange
2797 characters, so the fact that it applies to all lookups doesn't matter.
2799 PH/36 Now that the new test suite is complete, we can remove some of the
2800 special code in Exim that was needed for the old test suite. For example,
2801 sorting DNS records because real resolvers return them in an arbitrary
2802 order. The new test suite's fake resolver always returns records in the
2805 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
2806 bounces) except when queue_only is set, to avoid logging races between
2807 the different processes.
2809 PH/38 Panic-die if .include specifies a non-absolute path.
2811 PH/39 A tweak to the "H" retry rule from its user.
2813 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
2814 a label. They prevented compilation on older perls.
2816 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
2817 a warning to be raised on newish perls.
2819 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
2820 on queue. Changes to match documented behaviour of showing count of
2821 messages matching specified criteria.
2823 PH/40 Changed the default ident timeout from 30s to 5s.
2825 PH/41 Added support for the use of login_cap features, on those BSD systems
2826 that have them, for controlling the resources used by pipe deliveries.
2828 PH/42 The content-scanning code uses fopen() to create files in which to put
2829 message data. Previously it was not paying any attention to the mode of
2830 the files. Exim runs with umask(0) because the rest of the code creates
2831 files with open(), and sets the required mode explicitly. Thus, these
2832 files were ending up world-writeable. This was not a big issue, because,
2833 being within the spool directory, they were not world-accessible. I have
2834 created a function called modefopen, which takes an additional mode
2835 argument. It sets umask(777), creates the file, chmods it to the required
2836 mode, then resets the umask. All the relevant calls to fopen() in the
2837 content scanning code have been changed to use this function.
2839 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
2840 to 24 hours. This avoids potential overflow problems when processing G
2841 and H retry rules. I suspect nobody ever tinkers with this value.
2843 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
2845 PH/45 When the plaintext authenticator is running as a client, the server's
2846 challenges are checked to ensure they are valid base64 strings. By
2847 default, the authentication attempt is cancelled if an invalid string is
2848 received. Setting client_ignore_invalid_base64 true ignores these errors.
2849 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
2850 they are received. Thus, the responses can be made to depend on the
2851 challenges. If an invalid string is ignored, an empty string is placed in
2854 PH/46 Messages that are created by the autoreply transport now contains a
2855 References: header, in accordance with RFCs 2822 and 3834.
2857 PH/47 Added authenticated_sender_force to the smtp transport.
2859 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
2861 PH/49 Installed latest patch from the Sieve maintainer.
2863 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
2864 was also set, the appendfile transport was unnecessarily scanning a
2865 directory of message files (e.g. for maildir delivery) to find the count
2866 of files (along with the size), even though it did not need this
2867 information. It now does the scan only if it needs to find either the
2868 size of the count of files.
2870 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
2872 PH/52 Two bugs concerned with error handling when the smtp transport is
2875 (i) Exim was not creating retry information for temporary errors given
2876 for individual recipients after the DATA command when the smtp transport
2877 was used in LMTP mode. This meant that they could be retried too
2878 frequently, and not timed out correctly.
2880 (ii) Exim was setting the flag that allows error details to be returned
2881 for LMTP errors on RCPT commands, but not for LMTP errors for individual
2882 recipients that were returned after the DATA command.
2884 PH/53 This is related to PH/52, but is more general: for any failing address,
2885 when detailed error information was permitted to be returned to the
2886 sender, but the error was temporary, then after the final timeout, only
2887 "retry timeout exceeded" was returned. Now it returns the full error as
2888 well as "retry timeout exceeded".
2890 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
2891 do this, and (what is worse) MTAs that accept it.
2893 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
2894 will now be deprecated.
2896 PH/56 New os.c-cygwin from the Cygwin maintainer.
2898 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
2899 formats (previously only available in exim formats via -bpr, -bpru,
2900 and -bpra. Now also available in native and exiqgrep formats)
2902 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
2903 with very large, slow to parse queues
2905 JJ/08 exipick: added ! as generic prefix to negate any criteria format
2907 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
2909 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
2910 responses to authentication challenges, though it was showing the
2911 challenges; (ii) I've removed the CR characters from the debug output for
2914 PH/58 Allow for the insertion of a newline as well as a space when a string
2915 is turned into more than one encoded-word during RFC 2047 encoding. The
2916 Sieve code now uses this.
2918 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
2919 data_4xx, lost_connection, tls_required.
2921 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
2922 message was being sent as an SMTP response.
2924 PH/61 Add -l and -k options to exicyclog.
2926 PH/62 When verifying, if an address was redirected to one new address, so that
2927 verification continued, and the new address failed or deferred after
2928 having set something in $address_data, the value of $address_data was not
2929 passed back to the ACL. This was different to the case when no
2930 redirection occurred. The value is now passed back in both cases.
2932 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
2933 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
2934 use login_cap.h, so on its own it isn't the distinguishing feature. The
2935 new name refers directly to the setclassresources() function.
2937 PH/65 Added configuration files for NetBSD3.
2939 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
2941 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
2942 is preferred over IPv4.
2944 PH/68 The bounce_return_message and bounce_return_body options were not being
2945 honoured for bounces generated during the reception of non-SMTP messages.
2946 In particular, this applied to messages rejected by the ACL. This bug has
2947 been fixed. However, if bounce_return_message is true and bounce_return_
2948 body is false, the headers that are returned for a non-SMTP message
2949 include only those that have been read before the error was detected.
2950 (In the case of an ACL rejection, they have all been read.)
2952 PH/69 The HTML version of the specification is now built in a directory called
2953 spec_html instead of spec.html, because the latter looks like a path with
2954 a MIME-type, and this confuses some software.
2956 PH/70 Catch two compiler warnings in sieve.c.
2958 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
2959 function verify_get_ident() calls ip_connect() to connect a socket, but
2960 if the "connect()" function timed out, ip_connect() used to close the
2961 socket. However, verify_get_ident() also closes the socket later, and in
2962 between Exim writes to the log, which may get opened at this point. When
2963 the socket was closed in ip_connect(), the log could get the same file
2964 descriptor number as the socket. This naturally causes chaos. The fix is
2965 not to close the socket in ip_connect(); the socket should be closed by
2966 the function that creates it. There was only one place in the code where
2967 this was missing, in the iplookup router, which I don't think anybody now
2968 uses, but I've fixed it anyway.
2970 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
2971 well as to direct DNS lookups. Otherwise the handling of names in host
2972 lists is inconsistent and therefore confusing.
2978 PH/01 Two changes to the default runtime configuration:
2980 (1) Move the checks for relay_from_hosts and authenticated clients from
2981 after to before the (commented out) DNS black list checks.
2983 (2) Add control=submission to the relay_from_hosts and authenticated
2984 clients checks, on the grounds that messages accepted by these
2985 statements are most likely to be submissions.
2987 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
2989 (1) Generate an error if the third argument for the ${prvs expansion is
2992 (2) Treat a missing third argument of ${prvscheck as if it were an empty
2995 (3) Reset the variables that are obtained from the first argument of
2996 ${prvscheck and used in the second argument before leaving the code,
2997 because their memory is reclaimed, so using them afterwards may do
3000 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
3001 one (it's much easier than Tom thought :-).
3003 (5) Because of (4), we can now allow for the use of $prvscheck_result
3004 inside the third argument.
3006 PH/03 For some reason, the default setting of PATH when running a command from
3007 a pipe transport was just "/usr/bin". I have changed it to
3010 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
3011 anything to be listed in the output from -bV.
3013 PH/05 When a filter generated an autoreply, the entire To: header line was
3014 quoted in the delivery log line, like this:
3016 => >A.N.Other <ano@some.domain> <original@ddress> ...
3018 This has been changed so that it extracts the operative address. There
3019 may be more than one such address. If so, they are comma-separated, like
3022 => >ano@some.domain,ona@other.domain <original@ddress> ...
3024 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
3025 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
3026 not being looked up in the rDNS to get a host name, Exim was showing the
3027 IP address twice in Received: lines, even though the IP addresses were
3028 identical. For example:
3030 Received: from [1.2.3.4] (helo=[1.2.3.4])
3032 However, if the real host name was known, it was omitting the HELO data
3033 if it matched the actual IP address. This has been tidied up so that it
3034 doesn't show the same IP address twice.
3036 PH/07 When both +timestamp and +memory debugging was on, the value given by
3037 $tod_xxx expansions could be wrong, because the tod_stamp() function was
3038 called by the debug printing, thereby overwriting the timestamp buffer.
3039 Debugging no longer uses the tod_stamp() function when +timestamp is set.
3041 PH/08 When the original message was included in an autoreply transport, it
3042 always said "this is a copy of the message, including all the headers",
3043 even if body_only or headers_only was set. It now gives an appropriate
3046 PH/09 Applied a patch from the Sieve maintainer which:
3048 o fixes some comments
3049 o adds the (disabled) notify extension core
3050 o adds some debug output for the result of if/elsif tests
3051 o points to the current vacation draft in the documentation
3052 and documents the missing references header update
3056 o fixes a bug in processing the envelope test (when testing
3057 multiple envelope elements, the last element determinted the
3060 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
3061 Electronic Mail") by including:
3063 Auto-submitted: auto-generated
3065 in the messages that it generates (bounce messages and others, such as
3066 warnings). In the case of bounce messages for non-SMTP mesages, there was
3067 also a typo: it was using "Auto_submitted" (underscore instead of
3068 hyphen). Since every message generated by Exim is necessarily in response
3069 to another message, thes have all been changed to:
3071 Auto-Submitted: auto-replied
3073 in accordance with these statements in the RFC:
3075 The auto-replied keyword:
3077 - SHOULD be used on messages sent in direct response to another
3078 message by an automatic process,
3080 - MUST NOT be used on manually-generated messages,
3082 - MAY be used on Delivery Status Notifications (DSNs) and Message
3083 Disposition Notifications (MDNs),
3085 - MUST NOT be used on messages generated by automatic or periodic
3086 processes, except for messages which are automatic responses to
3089 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
3090 to the default Received: header definition.
3092 PH/12 Added log selector acl_warn_skipped (default on).
3094 PH/13 After a successful wildlsearch lookup, discard the values of numeric
3095 variables because (a) they are in the wrong storage pool and (b) even if
3096 they were copied, it wouldn't work properly because of the caching.
3098 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
3099 checking when decoding. Apparently there are clients that generate
3100 overlong encoded strings. Why am I not surprised?
3102 PH/15 If the first argument of "${if match_address" was not empty, but did not
3103 contain an "@" character, Exim crashed. Now it writes a panic log message
3104 and treats the condition as false.
3106 PH/16 In autoreply, treat an empty string for "once" the same as unset.
3108 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
3109 extension "envelope-auth". The code is finished and in agreement with
3110 other implementations, but there is no documentation so far and in fact,
3111 nobody wrote the draft yet. This extension is currently #undef'ed, thus
3112 not changing the active code.
3114 Print executed "if" and "elsif" statements when debugging is used. This
3115 helps a great deal to understand what a filter does.
3117 Document more things not specified clearly in RFC3028. I had all this
3118 sorted out, when out of a sudden new issues came to my mind. Oops."
3120 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
3123 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
3124 canonical form (as documented). However, after a host name lookup from
3125 the IP address, check_host() was doing a simple string comparison with
3126 addresses acquired from the DNS when checking that the found name did
3127 have the original IP as one of its addresses. Since any found IPv6
3128 addresses are likely to be in abbreviated form, the comparison could
3129 fail. Luckily, there already exists a function for doing the comparison
3130 by converting both addresses to binary, so now that is used instead of
3131 the text comparison.
3133 PH/20 There was another similar case to PH/19, when a complete host name was
3134 given in a host list; looking up its IP address could give an abbreviated
3135 form, whereas the current host's name might or might not be abbreviated.
3136 The same fix has been applied.
3142 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
3143 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
3146 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
3147 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
3149 PH/03 Typo: missing ".o" in src/pcre/Makefile.
3151 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
3152 header line, restrict the check to what is listed in RFCs 2369 and 2929.
3153 Also, for "Auto-Submitted", treat anything other than "no" as
3154 non-personal, in accordance with RFC 3834. (Previously it treated
3155 anything starting "auto-" as non-personal.)
3157 TF/01 The control=submission/name=... option had a problem with syntax
3158 errors if the name included a slash character. The /name= option
3159 now slurps the rest of the string, so it can include any characters
3160 but it must come last in the list of options (after /sender_retain
3163 PH/05 Some modifications to the interface to the fake nameserver for the new
3171 TK/01 Added the "success_on_redirect" address verification option. See
3172 NewStuff for rationale and an example.
3174 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
3176 PH/02 Patch to exigrep to allow it to work on syslog lines.
3178 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
3179 fread() to skip over the body file's header line, because in Cygwin the
3180 header line is locked and is inaccessible.
3182 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
3183 co-exist for some time) to make it clear that it is the Exim ID that is
3184 referenced, not the Message-ID: header line.
3186 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
3187 string_format() function, because snprintf() does not exist on all
3190 PH/06 The use of forbid_filter_existstest now also locks out the use of the
3191 ${stat: expansion item.
3193 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
3194 protocol synchronization error", to keep the pedants happy.
3196 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
3197 well as for IRIX systems, when gcc is being used. See the host.c source
3200 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
3202 PH/10 Named domain lists were not working if used in a queue_smtp_domains
3205 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
3206 transport and to the smtp transport in LMTP mode.
3208 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
3210 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
3211 run a filter in a subprocess. This could lead to confusion in subsequent
3212 lookups in the parent process. There should also be a search_tidyup() at
3213 the end of the subprocess.
3215 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
3216 only if the host matched helo_try_verify_hosts, which caused the
3217 verification to occur when the EHLO/HELO command was issued. The ACL just
3218 tested the remembered result. Now, if a previous verification attempt has
3219 not happened, "verify = helo" does it there and then.
3221 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
3223 TK/03 Fix log output including CR from clamd.
3225 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
3226 exist provoked a memory error which could cause a segfault.
3228 PH/15 Installed PCRE 6.2
3230 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
3232 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
3233 of the problem. Specifically, suggested +O2 rather than +O1 for the
3236 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
3238 PH/20 If a delivery was routed to a non-standard port by means of an SRV
3239 record, the port was not correctly logged when the outgoing_port log
3240 selector was set (it logged the transort's default port).
3242 PH/21 Added support for host-specific ports to manualroute, queryprogram,
3243 fallback_hosts, and "hosts" in the smtp transport.
3245 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
3246 host errors such as "Connection refused".
3248 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
3249 authentication with radiusclient 0.4.9:
3251 - Error returned from rc_read_config was caught wrongly
3252 - Username/password not passed on to radius server due to wrong length.
3254 The presumption is that some radiusclient API changes for 4.51/PH/17
3255 were not taken care of correctly. The code is still untested by me (my
3256 Linux distribution still has 0.3.2 of radiusclient), but it was
3257 contributed by a Radius user.
3259 PH/24 When doing a callout, the value of $domain wasn't set correctly when
3260 expanding the "port" option of the smtp transport.
3262 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
3263 while reading a MIME header. Thanks to Tom Hughes for a patch.
3265 PH/24 Include config.h inside local_scan.h so that configuration settings are
3268 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
3269 that in an ACL for RCPT (for example), you can examine exactly what was
3272 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
3273 commands, but it was not correctly comparing the address with the actual
3274 client host address. Thus, it would show the EHLO address in Received:
3275 header lines when this was not necessary.
3277 PH/27 Added the % operator to ${eval:}.
3279 PH/28 Exim tries to create and chdir to its spool directory when it starts;
3280 it should be ignoring failures (because with -C, for example, it has lost
3281 privilege). It wasn't ignoring creation failures other than "already
3284 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
3287 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
3288 because an input line was too long, either on its own, or by virtue of
3289 too many continuations, the temporary file was not being removed, and the
3290 return code was incorrect.
3292 PH/31 Missing "BOOL" in function definition in filtertest.c.
3294 PH/32 Applied Sieve patches from the maintainer.
3296 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
3298 PH/33 Added "verify = not_blind".
3300 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
3301 Local/Makefile (with some defaults set). These are used in built scripts
3302 such as exicyclog, but they have never been used in the exim_install
3303 script (though there are many overriding facilities there). I have
3304 arranged that the exim_install script now takes note of these two
3307 PH/35 Installed configuration files for Dragonfly.
3309 PH/36 When a locally submitted message by a trusted user did not contain a
3310 From: header, and the sender address was obtained from -f or from an SMTP
3311 MAIL command, and the trusted user did not use -F to supply a sender
3312 name, $originator_name was incorrectly used when constructing a From:
3313 header. Furthermore, $originator_name was used for submission mode
3314 messages from external hosts without From: headers in a similar way,
3315 which is clearly wrong.
3317 PH/37 Added control=suppress_local_fixups.
3319 PH/38 When log_selector = +received_sender was set, and the addition of the
3320 sender made the log line's construction buffer exactly full, or one byte
3321 less than full, an overflow happened when the terminating "\n" was
3324 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
3325 when the result of a list match is failure because a DNS lookup failed.
3327 PH/40 RM_COMMAND is now used in the building process.
3329 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
3330 the "build-* directories that it finds.
3332 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
3333 address in a domain literal was a prefix of an interface address.
3335 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
3336 when verifying a sender address, unless rewrite_headers is false.
3338 PH/44 Wrote a long comment about why errors_to addresses are verified as
3339 recipients, not senders.
3341 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
3342 the ratelimit ACL was added.
3344 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
3346 PH/46 Added extra information about PostgreSQL errors to the error string.
3348 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
3349 avoiding the need to install special zones in a real server. This is
3350 backwards compatible; if it can't find the fake resolver, it drops back.
3351 Thus, both old and new test suites can be run.
3353 TF/02 Added util/ratelimit.pl
3355 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
3356 clock is set back in time.
3358 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
3359 Candler <B.Candler@pobox.com>.
3361 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
3362 OK for addresses that are the result of redirections.
3364 PH/48 A number of further additions for the benefit of the new test suite,
3365 including a fake gethostbyname() that interfaces to the fake DNS resolver
3368 TF/06 The fix for widen_domains has also been applied to qualify_single and
3369 search_parents which are the other dnslookup options that can cause
3372 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
3375 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
3377 TF/07 Exim produced the error message "an SRV record indicated no SMTP
3378 service" if it encountered an MX record with an empty target hostname.
3379 The message is now "an MX or SRV record indicated no SMTP service".
3381 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
3382 if the DNS of the sending site is misconfigured. This is quite a
3383 common situation. This change restores the behaviour of treating a
3384 helo verification defer as a failure.
3386 PH/51 If self=fail was set on a router, the bounce message did not include the
3387 actual error message.
3393 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
3395 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
3396 command itself ended in error, the underlying message about the transport
3397 filter timeout was being overwritten with the pipe command error. Now the
3398 underlying error message should be appended to the second error message.
3400 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
3402 PH/02 Reduce the amount of output that "make" produces by default. Full output
3403 can still be requested.
3405 PH/03 The warning log line about a condition test deferring for a "warn" verb
3406 was being output only once per connection, rather than after each
3407 occurrence (because it was using the same function as for successful
3408 "warn" verbs). This seems wrong, so I have changed it.
3410 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
3411 it should not have, which might have caused a crash in the right
3412 circumstances, but probably never did.
3414 PH/04 Installed a modified version of Tony Finch's patch to make submission
3415 mode fix the return path as well as the Sender: header line, and to
3416 add a /name= option so that you can make the user's friendly name appear
3419 TF/03 Added the control = fakedefer ACL modifier.
3421 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
3422 Mark Lowes for thorough testing.
3424 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
3426 TK/03 Merged latest SRS patch from Miles Wilton.
3428 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
3429 with the definition in sysexits.h (which is #included earlier).
3430 Fortunately, Exim does not actually use EX_OK. The code used to try to
3431 preserve the sysexits.h value, by assumimg that macro definitions were
3432 scanned for macro replacements. I have been disabused of this notion,
3433 so now the code just undefines EX_OK before #including unistd.h.
3435 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
3436 in the smtp transport. When a block could not be written in a single
3437 write() function, the timeout was being re-applied to each part-write.
3438 This seems wrong - if the receiver was accepting one byte at a time it
3439 would take for ever. The timeout is now adjusted when this happens. It
3440 doesn't have to be particularly precise.
3442 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
3443 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
3445 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
3446 without a domain if the check to <postmaster@domain> fails.
3448 SC/01 Eximstats: added -xls and the ability to specify output files
3449 (patch written by Frank Heydlauf).
3451 SC/02 Eximstats: use FileHandles for outputing results.
3453 SC/03 Eximstats: allow any combination of xls, txt, and html output.
3455 SC/04 Eximstats: fixed display of large numbers with -nvr option
3457 SC/05 Eximstats: fixed merging of reports with empty tables.
3459 SC/06 Eximstats: added the -include_original_destination flag
3461 SC/07 Eximstats: removed tabs and trailing whitespace.
3463 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
3465 TK/06 MBOX spool code: Add real "From " MBOX separator line
3466 so the .eml file is really in mbox format (even though
3467 most programs do not really care). Patch from Alex Miller.
3469 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
3470 The latter is generated from $received_to and is only set if the
3471 message has one envelope recipient. SA can use these headers,
3472 obviously out-of-the-box. Patch from Alex Miller.
3474 PH/08 The ${def test on a variable was returning false if the variable's
3475 value was "0", contrary to what the specification has always said!
3476 The result should be true unless the variable is empty.
3478 PH/09 The syntax error of a character other than { following "${if
3479 def:variable_name" (after optional whitespace) was not being diagnosed.
3480 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
3481 accidental colon was present, for example, could give incorrect results.
3483 PH/10 Tidied the code in a number of places where the st_size field of a stat()
3484 result is used (not including appendfile, where other changes are about
3487 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
3488 This involved changing a lot of size variables from int to off_t. It
3489 should work with maildirs and everything.
3491 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
3492 spamd dying while we are connected to it.
3494 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
3497 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
3498 function for BDB 4.3.
3500 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
3502 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
3503 the sources for PCRE have been re-arranged and more files are now
3506 PH/15 The code I had for printing potentially long long variables in PH/11
3507 above was not the best (it lost precision). The length of off_t variables
3508 is now inspected at build time, and an appropriate printing format (%ld
3509 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
3510 to be "long long int" or "long int". This is needed for the internal
3511 formatting function string_vformat().
3513 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
3514 the configuration file to be ":syslog", then the script "guesses" where
3515 the logs files are, rather than using the compiled in default. In our
3516 case the guess is not the same as the compiled default, so the script
3517 suddenly stopped working when I started to use syslog. The patch checks
3518 to see if log_file_path is "". If so, it attempts to read it from exim
3519 with no configuration file to get the compiled in version, before it
3520 falls back to the previous guessing code."
3522 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
3523 implementing BATV in an Exim configuration. See NewStuff for the gory
3526 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
3527 Makefile that are specific to HP-UX.
3529 PH/18 If the "use_postmaster" option was set for a recipient callout together
3530 with the "random" option, the postmaster address was used as the MAIL
3531 FROM address for the random test, but not for the subsequent recipient
3532 test. It is now used for both.
3534 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
3535 patch removes a few documentation additions to RFC 3028, because the
3536 latest draft now contains them. It adds the new en;ascii-case comparator
3537 and a new error check for 8bit text in MIME parts. Comparator and
3538 require names are now matched exactly. I enabled the subaddress
3539 extension, but it is not well tested yet (read: it works for me)."
3541 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
3542 rework some of the code of TK/09 above to avoid the hardwired use of
3543 "%lld" and "long long". Replaced the call to snprintf() with a call to
3546 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
3547 records point to non-existent hosts", "retry timeout exceeded", and
3548 "retry time not reached for any host after a long failure period".
3550 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
3551 experimental DomainKeys support:
3553 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
3554 (2) On an error such as an illegally used "control", the wrong name for
3555 the control was given.
3557 These problems did NOT occur unless DomainKeys support was compiled.
3559 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
3561 PH/24 Added ${if match_ip condition.
3563 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
3564 databases so that it will be absolutely obvious if a crash occurs in the
3565 DB library. This is a regular occurrence (often caused by mis-matched
3568 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
3569 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
3570 picked up on a user's system that detects such things. There doesn't seem
3571 to be a gcc warning option for this - only an attribute that has to be
3572 put on the function's prototype. It seems that in Fedora Core 4 they have
3573 set this on a number of new functions. No doubt there will be more in due
3576 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
3577 specify a transport. However, if an address that was verified by such a
3578 router was the subject of a callout, Exim crashed because it tried to
3579 read the rcpt_include_affixes from the non-existent transport. Now it
3580 just assumes that the setting of that option is false. This bug was
3581 introduced by 4.51/PH/31.
3583 PH/28 Changed -d+all to exclude +memory, because that information is very
3584 rarely of interest, but it makes the output a lot bigger. People tend to
3585 do -d+all out of habit.
3587 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
3588 code in os-type was giving problems when libc.so lives in lib64, like on
3591 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
3592 aren't the modern standard, and it seems that some systems' include files
3593 don't always have them. Exim was already checking for some of the newer
3594 ones like T_AAAA, and defining it itself. I've added checks for all the
3595 record types that Exim uses.
3597 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
3598 not automatically generating a new one, as it is supposed to. This
3599 prevented TLS from working. If the file did exist, but contained invalid
3600 data, a new version was generated, as expected. It was only the case of a
3601 non-existent file that was broken.
3603 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
3604 with a change in libdomainkeys > 0.64.
3606 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
3607 from DNS. If the selector record carries the flag, it now has
3608 precedence over the domain-wide flag.
3610 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
3612 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
3613 the use of a port name that isn't defined in /etc/services) occurred, the
3614 message was deferred as in a normal delivery, and thus remained on the
3615 spool, instead of being failed because of the mua_wrapper setting. This
3616 is now fixed, and I tidied up some of the mua_wrapper messages at the
3619 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
3620 the messages in a hash of arrays rather than using individual hashes.
3621 This is a bit cleaner and results in dramatic memory savings, albeit
3622 at a slight CPU cost.
3624 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
3625 as requested by Marc Sherman.
3627 SC/10 Eximstats: added histograms for user specified patterns as requested
3630 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
3632 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
3633 fopen() in the content-scanning modules that did not already have it.
3639 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
3640 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
3642 TK/02 Fix ACL "control" statement not being available in MIME ACL.
3644 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
3646 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
3647 to test Sieve filters that use "vacation".
3649 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
3650 that changes the way the GnuTLS parameters are stored in the cache file.
3651 The new format can be generated externally. For backward compatibility,
3652 if the data in the cache doesn't make sense, Exim assumes it has read an
3653 old-format file, and it generates new data and writes a new file. This
3654 means that you can't go back to an older release without removing the
3657 PH/03 A redirect router that has both "unseen" and "one_time" set does not
3658 work if there are any delivery delays because "one_time" forces the
3659 parent to be marked "delivered", so its unseen clone is never tried
3660 again. For this reason, Exim now forbids the simultaneous setting of
3663 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
3664 redirected to themselves ("homonym" addresses). Read the long ChangeLog
3665 entry if you want to know the details. The fix, however, neglected to
3666 consider the case when local delivery batching is involved. The test for
3667 "previously delivered" was not happening when checking to see if an
3668 address could be batched with a previous (undelivered) one; under
3669 certain circumstances this could lead to multiple deliveries to the same
3672 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
3673 in its include files, and this causes problems building Exim.
3675 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
3676 header_syntax) but Exim was just ignoring anything given after a slash.
3677 In particular, this caused confusion with an attempt to use "verify =
3678 reverse_host_lookup/defer_ok". An error is now given when options are
3679 supplied for verify items that do not have them. (Maybe reverse_host_
3680 lookup should have a defer_ok option, but that's a different point.)
3682 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
3683 defined by RFC 821) to 2048, because there were problems with some AUTH
3684 commands, and RFC 1869 says the size should be increased for extended
3685 SMTP commands that take arguments.
3687 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
3690 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
3691 "unknown" error; now it says that the functionality isn't in the binary.
3693 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
3694 an address' error message when a string expansion fails (syntax or
3695 whatever). Otherwise the password may appear in the log. Following change
3696 PH/42 below, there is no longer a chance of it appearing in a bounce
3699 PH/11 Installed exipick version 20050225.0 from John Jetmore.
3701 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
3702 of its addresses was ever tried. (Bugzilla bug #2.)
3704 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
3705 the result incorrectly in the debug output. (It correctly added a newline
3706 to what was transported.)
3708 TF/01 Added $received_time.
3710 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
3711 commented out examples of how to interface to a virus scanner and to
3712 SpamAssassin. Also added commented examples of av_scanner and
3713 spamd_address settings.
3715 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
3716 and controls are allowed in which ACLs. There were a couple of minor
3717 errors. Some of the entries in the conditions table (which is a table of
3718 where they are NOT allowed) were getting very unwieldy; rewrote them as a
3719 negation of where the condition IS allowed.
3721 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
3723 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
3724 header file does not have a version number, so I've had to invent a new
3725 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
3726 API. The code is untested by me (my Linux distribution still has 0.3.2 of
3727 radiusclient), but it was contributed by a Radius user.
3729 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
3730 files or directories, for OpenSSL.
3732 PH/19 When an Exim process that is running as root has to create an Exim log
3733 file, it does so in a subprocess that runs as exim:exim so as to get the
3734 ownership right at creation (otherwise, other Exim processes might see
3735 the file with the wrong ownership). There was no test for failure of this
3736 fork() call, which would lead to the process getting stuck as it waited
3737 for a non-existent subprocess. Forks do occasionally fail when resources
3738 run out. I reviewed all the other calls to fork(); they all seem to check
3741 PH/20 When checking for unexpected SMTP input at connect time (before writing
3742 the banner), Exim was not dealing correctly with a non-positive return
3743 from the read() function. If the client had disconnected by this time,
3744 the result was a log entry for a synchronization error with an empty
3745 string after "input=" when read() returned zero. If read() returned -1
3746 (an event I could not check), uninitialized data bytes were printed.
3747 There were reports of junk text (parts of files, etc) appearing after
3750 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
3752 PH/22 Added support for macro redefinition, and (re)definition in between
3753 driver and ACL definitions.
3755 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
3756 forgetting to use the resulting value; it was using the unexpanded value.
3758 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
3759 hadn't been configured. The fix is from Juergen Kreileder, who
3760 understands it better than I do:
3762 "Here's what I see happening with three configured cyrus_sasl
3763 authenticators configured (plain, login, cram-md5):
3765 On startup auth_cyrus_sasl_init() gets called for each of these.
3766 This means three calls to sasl_listmech() without a specified mech_list.
3767 => SASL tests which mechs of all available mechs actually work
3768 => three warnings about OTP not working
3769 => the returned list contains: plain, login, cram-md5, digest-md5, ...
3771 With the patch, sasl_listmech() also gets called three times. But now
3772 SASL's mech_list option is set to the server_mech specified in the the
3773 authenticator. Or in other words, the answer from sasl_listmech()
3774 gets limited to just the mech you're testing for (which is different
3776 => the return list contains just 'plain' or 'login', 'cram-md5' or
3777 nothing depending on the value of ob->server_mech.
3779 I've just tested the patch: Authentication still works fine,
3780 unavailable mechs specified in the exim configuration are still
3781 caught, and the auth.log warnings about OTP are gone."
3783 PH/25 When debugging is enabled, the contents of the command line are added
3784 to the debugging output, even when log_selector=+arguments is not
3787 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
3788 answer is "GNU", and only if the return is "GNU/something" is the answer
3791 PH/27 $acl_verify_message is now set immediately after the failure of a
3792 verification in an ACL, and so is available in subsequent modifiers. In
3793 particular, the message can be preserved by coding like this:
3795 warn !verify = sender
3796 set acl_m0 = $acl_verify_message
3798 Previously, $acl_verify_message was set only while expanding "message"
3799 and "log_message" when a very denied access.
3801 PH/28 Modified OS/os.c-Linux with
3803 -#ifndef OS_LOAD_AVERAGE
3804 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
3806 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
3807 nomenclature these days.)
3809 PH/29 Installed patch from the Sieve maintainer that adds the options
3810 sieve_useraddress and sieve_subaddress to the redirect router.
3812 PH/30 In these circumstances:
3813 . Two addresses routed to the same list of hosts;
3814 . First host does not offer TLS;
3815 . First host accepts first address;
3816 . First host gives temporary error to second address;
3817 . Second host offers TLS and a TLS session is established;
3818 . Second host accepts second address.
3819 Exim incorrectly logged both deliveries with the TLS parameters (cipher
3820 and peerdn, if requested) that were in fact used only for the second
3823 PH/31 When doing a callout as part of verifying an address, Exim was not paying
3824 attention to any local part prefix or suffix that was matched by the
3825 router that accepted the address. It now behaves in the same way as it
3826 does for delivery: the affixes are removed from the local part unless
3827 rcpt_include_affixes is set on the transport.
3829 PH/32 Add the sender address, as F=<...>, to the log line when logging a
3830 timeout during the DATA phase of an incoming message.
3832 PH/33 Sieve envelope tests were broken for match types other than :is. I have
3833 applied a patch sanctioned by the Sieve maintainer.
3835 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
3836 the uid or gid is negative. A case of a negative gid caused this to be
3837 noticed. The fix allows for either to be negative.
3839 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
3840 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
3843 PH/36 The change PH/12 above was broken. Fixed it.
3845 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
3846 the grounds that routing the same address twice would always produce the
3847 same answer. This might have been true once, but it is certainly no
3848 longer true now. Routing a child address may depend on the previous
3849 routing that produced that child. Some complicated redirection strategies
3850 went wrong when messages had multiple recipients, and made Exim's
3851 behaviour dependent on the order in which the addresses were given.
3853 I have moved the duplicate checking until after the routing is complete.
3854 Exim scans the addresses that are assigned to local and remote
3855 transports, and removes any duplicates. This means that more work will be
3856 done, as duplicates will always all be routed, but duplicates are
3857 presumably rare, so I don't expect this is of any significance.
3859 For deliveries to pipes, files, and autoreplies, the duplicate checking
3860 still happens during the routing process, since they are not going to be
3863 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
3864 It corrects a timeout issue with spamd. This is Ian's comment: "The
3865 background is that sometimes spamd either never reads data from a
3866 connection it has accepted, or it never writes response data. The exiscan
3867 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
3868 blindly assumes that writes won't block so it may never time out."
3870 PH/39 Allow G after quota size as well as K and M.
3872 PH/40 The value set for $authenticated_id in an authenticator may not contain
3873 binary zeroes or newlines because the value is written to log lines and
3874 to spool files. There was no check on this. Now the value is run through
3875 the string_printing() function so that such characters are converted to
3876 printable escape sequences.
3878 PH/41 $message_linecount is a new variable that contains the total number of
3879 lines in the message. Compare $body_linecount, which is the count for the
3882 PH/42 Exim no longer gives details of delivery errors for specific addresses in
3883 bounce and delay warning messages, except in certain special cases, which
3886 (a) An SMTP error message from a remote host;
3887 (b) A message specified in a :fail: redirection;
3888 (c) A message specified in a "fail" command in a system filter;
3889 (d) A message specified in a FAIL return from the queryprogram router;
3890 (e) A message specified by the cannot_route_message router option.
3892 In these cases only, Exim does include the error details in bounce and
3893 warning messages. There are also a few cases where bland messages such
3894 as "unrouteable address" or "local delivery error" are given.
3896 PH/43 $value is now also set for the "else" part of a ${run expansion.
3898 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
3899 being worked on, but at least Exim now implements the latest version to
3902 PH/45 In a pipe transport, although a timeout while waiting for the pipe
3903 process to complete was treated as a delivery failure, a timeout while
3904 writing the message to the pipe was logged, but erroneously treated as a
3905 successful delivery. Such timeouts include transport filter timeouts. For
3906 consistency with the overall process timeout, these timeouts are now
3907 treated as errors, giving rise to delivery failures by default. However,
3908 there is now a new Boolean option for the pipe transport called
3909 timeout_defer, which, if set TRUE, converts the failures into defers for
3910 both kinds of timeout. A transport filter timeout is now identified in
3913 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
3914 systems where "make" and "gmake" are different, calling "gmake" at top
3915 level broke things. I've arranged for the value of $(MAKE) to be passed
3916 from the Makefile to this script so that it can call the same version of
3920 A note about Exim versions 4.44 and 4.50
3921 ----------------------------------------
3923 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
3924 changes of various kinds. As a consequence, a big documentation update was
3925 needed. This delayed the release for rather longer than seemed good, especially
3926 in the light of a couple of (minor) security issues. Therefore, the changes
3927 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
3928 release. So 4.44 and 4.50 are in effect two different branches that both start
3931 I have left the 4.50 change log unchanged; it contains all the changes since
3932 4.43. The change log for 4.44 is below; many of its items are identical to
3933 those for 4.50. This seems to be the most sensible way to preserve the
3934 historical information.
3940 1. Minor wording change to the doc/README.SIEVE file.
3942 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
3943 computation of the current number of files was incorrect.
3945 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
3946 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
3949 4. Give more explanation in the error message when the command for a transport
3950 filter fails to execute.
3952 5. There are several places where Exim runs a non-Exim command in a
3953 subprocess. The SIGUSR1 signal should be disabled for these processes. This
3954 was being done only for the command run by the queryprogram router. It is
3955 now done for all such subprocesses. The other cases are: ${run, transport
3956 filters, and the commands run by the lmtp and pipe transports.
3958 6. Added CONFIGURE_GROUP build-time option.
3960 7. Some older OS have a limit of 256 on the maximum number of file
3961 descriptors. Exim was using setrlimit() to set 1000 as a large value
3962 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
3963 systems. I've change it so that if it can't get 1000, it tries for 256.
3965 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
3966 was an oversight, and furthermore, ever since the addition of extra
3967 controls (e.g. 4.43/32), the checks on when to allow different forms of
3968 "control" were broken. There should now be diagnostics for all cases when a
3969 control that does not make sense is encountered.
3971 9. Added the /retain_sender option to "control=submission".
3973 10. $recipients is now available in the predata ACL (oversight).
3975 11. Tidy the search cache before the fork to do a delivery from a message
3976 received from the command line. Otherwise the child will trigger a lookup
3977 failure and thereby defer the delivery if it tries to use (for example) a
3978 cached ldap connection that the parent has called unbind on.
3980 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
3981 of $address_data from the recipient verification was clobbered by the
3982 sender verification.
3984 13. The value of address_data from a sender verification is now available in
3985 $sender_address_data in subsequent conditions in the ACL statement.
3987 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
3989 15. Added a new option "connect=<time>" to callout options, to set a different
3992 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
3993 was its contents. (It was OK if the option was not defined at all.)
3995 17. A "Completed" log line is now written for messages that are removed from
3996 the spool by the -Mrm option.
3998 18. New variables $sender_verify_failure and $recipient_verify_failure contain
3999 information about exactly what failed.
4001 19. Added -dd to debug only the daemon process.
4003 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
4004 handles timeouts, both on the server side and network timeouts. Renamed the
4005 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
4007 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
4008 It is now set to "smtps".
4010 22. $host_address is now set to the target address during the checking of
4011 ignore_target_hosts.
4013 23. When checking ignore_target_hosts for an ipliteral router, no host name was
4014 being passed; this would have caused $sender_host_name to have been used if
4015 matching the list had actually called for a host name (not very likely,
4016 since this list is usually IP addresses). A host name is now passed as
4019 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4020 code that specifies a non-restarting handler (typically sigaction() in
4021 modern systems) in an attempt to fix a rare and obscure crash bug.
4023 25. Narrowed the window for a race in the daemon that could cause it to ignore
4024 SIGCHLD signals. This is not a major problem, because they are used only to
4025 wake it up if nothing else does.
4027 26. A malformed maildirsize file could cause Exim to calculate negative values
4028 for the mailbox size or file count. Odd effects could occur as a result.
4029 The maildirsize information is now recalculated if the size or filecount
4032 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4033 support for a long time. Removed HAVE_SYS_VFS_H.
4035 28. Installed the latest version of exipick from John Jetmore.
4037 29. In an address list, if the pattern was not a regular expression, an empty
4038 subject address (from a bounce message) matched only if the pattern was an
4039 empty string. Non-empty patterns were not even tested. This was the wrong
4040 because it is perfectly reasonable to use an empty address as part of a
4041 database query. An empty address is now tested by patterns that are
4042 lookups. However, all the other forms of pattern expect the subject to
4043 contain a local part and a domain, and therefore, for them, an empty
4044 address still always fails if the pattern is not itself empty.
4046 30. Exim went into a mad DNS loop when attempting to do a callout where the
4047 host was specified on an smtp transport, and looking it up yielded more
4048 than one IP address.
4050 31. Re-factored the code for checking spool and log partition space into a
4051 function that finds that data and another that does the check. The former
4052 is then used to implement four new variables: $spool_space, $log_space,
4053 $spool_inodes, and $log_inodes.
4055 32. The RFC2047 encoding function was originally intended for short strings
4056 such as real names; it was not keeping to the 75-character limit for
4057 encoded words that the RFC imposes. It now respects the limit, and
4058 generates multiple encoded words if necessary. To be on the safe side, I
4059 have increased the buffer size for the ${rfc2047: expansion operator from
4062 33. It is now permitted to omit both strings after an "if" condition; if the
4063 condition is true, the result is "true". As before, when the second string
4064 is omitted, a false condition yields an empty string. This makes it less
4065 cumbersome to write custom ACL and router conditions.
4067 34. Failure to deliver a bounce message always caused it to be frozen, even if
4068 there was an errors_to setting on the router. The errors_to setting is now
4071 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4072 canonical form (fully expanded) before being placed in
4073 $sender_host_address.
4075 36. The table in the code that translates DNS record types into text (T_A to
4076 "A" for instance) was missing entries for NS and CNAME. It is just possible
4077 that this could have caused confusion if both these types were looked up
4078 for the same domain, because the text type is used as part of Exim's
4079 per-process caching. But the chance of anyone hitting this buglet seems
4082 37. The dnsdb lookup has been extended in a number of ways.
4084 (1) There is a new type, "zns", which walks up the domain tree until it
4085 finds some nameserver records. It should be used with care.
4087 (2) There is a new type, "mxh", which is like "mx" except that it returns
4088 just the host names, not the priorities.
4090 (3) It is now possible to give a list of domains (or IP addresses) to be
4091 looked up. The behaviour when one of the lookups defers can be
4092 controlled by a keyword.
4094 (4) It is now possible to specify the separator character for use when
4095 multiple records are returned.
4097 38. The dnslists ACL condition has been extended: it is now possible to supply
4098 a list of IP addresses and/or domains to be looked up in a particular DNS
4101 39. Added log_selector=+queue_time_overall.
4103 40. When running the queue in the test harness, wait just a tad after forking a
4104 delivery process, to get repeatability of debugging output.
4106 41. Include certificate and key file names in error message when GnuTLS fails
4107 to set them up, because the GnuTLS error message doesn't include the name
4108 of the failing file when there is a problem reading it.
4110 42. Allow both -bf and -bF in the same test run.
4112 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
4114 44. The "Exiscan patch" is now merged into the mainline Exim source.
4116 45. Sometimes the final signoff response after QUIT could fail to get
4117 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4118 < 0 before doing a fflush(). This bug looks as though it goes back to the
4119 introduction of TLS in release 3.20, but "sometimes" must have been rare
4120 because the tests only now provoked it.
4122 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
4123 (this can affect the format of dates).
4125 47. exim_tidydb, when checking for the continued existence of a message for
4126 which it has found a message-specific retry record, was not finding
4127 messages that were in split spool directories. Consequently, it was
4128 deleting retry records that should have stayed in existence.
4130 48. Steve fixed some bugs in eximstats.
4132 49. The SPA authentication driver was not abandoning authentication and moving
4133 on to the next authenticator when an expansion was forced to fail,
4134 contradicting the general specification for all authenticators. Instead it
4135 was generating a temporary error. It now behaves as specified.
4137 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
4138 (the order specifies the preference for clients). The order is now AES256,
4139 AES128, 3DES, ARCFOUR128.
4141 51. Small patch to Sieve code - explicitly set From: when generating an
4144 52. Exim crashed if a remote delivery caused a very long error message to be
4145 recorded - for instance if somebody sent an entire SpamAssassin report back
4146 as a large number of 550 error lines. This bug was coincidentally fixed by
4147 increasing the size of one of Exim's internal buffers (big_buffer) that
4148 happened as part of the Exiscan merge. However, to be on the safe side, I
4149 have made the code more robust (and fixed the comments that describe what
4152 53. Now that there can be additional text after "Completed" in log lines (if
4153 the queue_time_overall log selector is set), a one-byte patch to exigrep
4154 was needed to allow it to recognize "Completed" as not the last thing in
4157 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
4158 patch that reportedly fixes this has been added. I am not expert enough to
4159 create a test for it. This is what the patch creator wrote:
4161 "I found a little strange behaviour of ldap code when working with
4162 Windows 2003 AD Domain, where users was placed in more than one
4163 Organization Units. When I tried to give exim partial DN, the exit code
4164 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
4165 But simultaneously result of request was absolutely normal ldap result,
4166 so I produce this patch..."
4168 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
4169 REFERENCE, so I have modified the code to exclude the patch when that macro
4172 55. Some experimental protocols are using DNS PTR records for new purposes. The
4173 keys for these records are domain names, not reversed IP addresses. The
4174 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4175 leaves it alone. Component reversal etc. now happens only for IP addresses.
4178 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
4180 57. Double the size of the debug message buffer (to 2048) so that more of very
4181 long debug lines gets shown.
4183 58. The exicyclog utility now does better if the number of log files to keep
4184 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
4186 59. Two changes related to the smtp_active_hostname option:
4188 (1) $smtp_active_hostname is now available as a variable.
4189 (2) The default for smtp_banner uses $smtp_active_hostname instead
4190 of $primary_hostname.
4192 60. The host_aton() function is supposed to be passed a string that is known
4193 to be a valid IP address. However, in the case of IPv6 addresses, it was
4194 not checking this. This is a hostage to fortune. Exim now panics and dies
4195 if the condition is not met. A case was found where this could be provoked
4196 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4197 components; fortuitously, this particular loophole had already been fixed
4198 by change 4.50/55 above.
4200 If there are any other similar loopholes, the new check in host_aton()
4201 itself should stop them being exploited. The report I received stated that
4202 data on the command line could provoke the exploit when Exim was running as
4203 exim, but did not say which command line option was involved. All I could
4204 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4205 running as the user.
4208 61. There was a buffer overflow vulnerability in the SPA authentication code
4209 (which came originally from the Samba project). I have added a test to the
4210 spa_base64_to_bits() function which I hope fixes it.
4213 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
4214 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
4216 63. The daemon start-up calls getloadavg() while still root for those OS that
4217 need the first call to be done as root, but it missed one case: when
4218 deliver_queue_load_max is set with deliver_drop_privilege. This is
4219 necessary for the benefit of the queue runner, because there is no re-exec
4220 when deliver_drop_privilege is set.
4222 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
4223 This has been fixed.
4225 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4226 was in use, was not putting the data itself into the right store pool;
4227 consequently, it could be overwritten for a subsequent message in the same
4228 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4231 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
4233 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
4234 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
4235 the function were treating the return as a boolean value, which happened to
4236 work because 0=false and not-0=true, but is not correct code.
4238 68. The host_aton() function was not handling scoped IPv6 addresses (those
4239 with, for example, "%eth0" on the end) correctly.
4241 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
4242 negated items (that is, ~something) in unsigned ints. Some compilers
4243 apparently mutter when there is no cast.
4245 70. If an address verification called from an ACL failed, and did not produce a
4246 user-specific message (i.e. there was only a "system" message), nothing was
4247 put in $acl_verify_message. In this situation, it now puts the system
4250 71. Change 4.23/11 added synchronization checking at the start of an SMTP
4251 session; change 4.31/43 added the unwanted input to the log line - except
4252 that it did not do this in the start of session case. It now does.
4254 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
4255 This is wrong and can cause the other end to generate a synchronization
4256 error if it is another Exim or anything else that does the synchronization
4257 check. A QUIT command is no longer sent after a timeout.
4259 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
4260 during host lookups.
4262 74. The defer_ok option of callout verification was not working if it was used
4263 when verifying addresses in header lines, that is, for this case:
4265 verify = header_sender/callout=defer_ok
4267 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
4268 those file descriptors could be used for SMTP connections. If anything
4269 wrote to stderr (the example that came up was "warn" in embedded Perl), it
4270 could be sent to the SMTP client, causing chaos. The daemon now opens
4271 stdin, stdout, and stderr to /dev/null when it puts itself into the
4274 76. Arrange for output from Perl's "warn" command to be written to Exim's main
4275 log by default. The user can override this with suitable Perl magic.
4277 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
4278 the log message when discard triggers, was not working for the DATA ACL or
4279 for the non-SMTP ACL.
4281 78. Error message wording change in sieve.c.
4283 79. If smtp_accept_max_per_host was set, the number of connections could be
4284 restricted to fewer than expected, because the daemon was trying to set up
4285 a new connection before checking whether the processes handling previous
4286 connections had finished. The check for completed processes is now done
4287 earlier. On busy systems, this bug wouldn't be noticed because something
4288 else would have woken the daemon, and it would have reaped the completed
4291 80. If a message was submitted locally by a user whose login name contained one
4292 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
4293 It caused a spool format error. I have fixed the spool reading code. A
4294 related problem was that the "from" clause in the Received: line became
4295 illegal because of the space(s). It is now covered by ${quote_local_part.
4297 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
4300 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
4302 83. Patch from Sieve maintainer to fix unterminated string problem in
4303 "vacation" handling.
4305 84. Some minor changes to the Linux configuration files to help with other
4306 OS variants using glibc.
4308 85. One more patch for Sieve to update vacation handling to latest spec.
4311 ----------------------------------------------------
4312 See the note above about the 4.44 and 4.50 releases.
4313 ----------------------------------------------------
4319 1. Change 4.43/35 introduced a bug that caused file counts to be
4320 incorrectly computed when quota_filecount was set in an appendfile
4323 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4324 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4327 3. Give more explanation in the error message when the command for a transport
4328 filter fails to execute.
4330 4. There are several places where Exim runs a non-Exim command in a
4331 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4332 was being done only for the command run by the queryprogram router. It is
4333 now done for all such subprocesses. The other cases are: ${run, transport
4334 filters, and the commands run by the lmtp and pipe transports.
4336 5. Some older OS have a limit of 256 on the maximum number of file
4337 descriptors. Exim was using setrlimit() to set 1000 as a large value
4338 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4339 systems. I've change it so that if it can't get 1000, it tries for 256.
4341 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4342 was an oversight, and furthermore, ever since the addition of extra
4343 controls (e.g. 4.43/32), the checks on when to allow different forms of
4344 "control" were broken. There should now be diagnostics for all cases when a
4345 control that does not make sense is encountered.
4347 7. $recipients is now available in the predata ACL (oversight).
4349 8. Tidy the search cache before the fork to do a delivery from a message
4350 received from the command line. Otherwise the child will trigger a lookup
4351 failure and thereby defer the delivery if it tries to use (for example) a
4352 cached ldap connection that the parent has called unbind on.
4354 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4355 of $address_data from the recipient verification was clobbered by the
4356 sender verification.
4358 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4359 was its contents. (It was OK if the option was not defined at all.)
4361 11. A "Completed" log line is now written for messages that are removed from
4362 the spool by the -Mrm option.
4364 12. $host_address is now set to the target address during the checking of
4365 ignore_target_hosts.
4367 13. When checking ignore_target_hosts for an ipliteral router, no host name was
4368 being passed; this would have caused $sender_host_name to have been used if
4369 matching the list had actually called for a host name (not very likely,
4370 since this list is usually IP addresses). A host name is now passed as
4373 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4374 code that specifies a non-restarting handler (typically sigaction() in
4375 modern systems) in an attempt to fix a rare and obscure crash bug.
4377 15. Narrowed the window for a race in the daemon that could cause it to ignore
4378 SIGCHLD signals. This is not a major problem, because they are used only to
4379 wake it up if nothing else does.
4381 16. A malformed maildirsize file could cause Exim to calculate negative values
4382 for the mailbox size or file count. Odd effects could occur as a result.
4383 The maildirsize information is now recalculated if the size or filecount
4386 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4387 support for a long time. Removed HAVE_SYS_VFS_H.
4389 18. Updated exipick to current release from John Jetmore.
4391 19. Allow an empty sender to be matched against a lookup in an address list.
4392 Previously the only cases considered were a regular expression, or an
4395 20. Exim went into a mad DNS lookup loop when doing a callout where the
4396 host was specified on the transport, if the DNS lookup yielded more than
4399 21. The RFC2047 encoding function was originally intended for short strings
4400 such as real names; it was not keeping to the 75-character limit for
4401 encoded words that the RFC imposes. It now respects the limit, and
4402 generates multiple encoded words if necessary. To be on the safe side, I
4403 have increased the buffer size for the ${rfc2047: expansion operator from
4406 22. Failure to deliver a bounce message always caused it to be frozen, even if
4407 there was an errors_to setting on the router. The errors_to setting is now
4410 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4411 canonical form (fully expanded) before being placed in
4412 $sender_host_address.
4414 24. Updated eximstats to version 1.33
4416 25. Include certificate and key file names in error message when GnuTLS fails
4417 to set them up, because the GnuTLS error message doesn't include the name
4418 of the failing file when there is a problem reading it.
4420 26. Expand error message when OpenSSL has problems setting up cert/key files.
4423 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
4424 (this can affect the format of dates).
4426 28. exim_tidydb, when checking for the continued existence of a message for
4427 which it has found a message-specific retry record, was not finding
4428 messages that were in split spool directories. Consequently, it was
4429 deleting retry records that should have stayed in existence.
4431 29. eximstats updated to version 1.35
4432 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
4433 1.35 - bugfix such that pie charts by volume are generated correctly
4435 30. The SPA authentication driver was not abandoning authentication and moving
4436 on to the next authenticator when an expansion was forced to fail,
4437 contradicting the general specification for all authenticators. Instead it
4438 was generating a temporary error. It now behaves as specified.
4440 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
4441 (the order specifies the preference for clients). The order is now AES256,
4442 AES128, 3DES, ARCFOUR128.
4444 31. Small patch to Sieve code - explicitly set From: when generating an
4447 32. Exim crashed if a remote delivery caused a very long error message to be
4448 recorded - for instance if somebody sent an entire SpamAssassin report back
4449 as a large number of 550 error lines. This bug was coincidentally fixed by
4450 increasing the size of one of Exim's internal buffers (big_buffer) that
4451 happened as part of the Exiscan merge. However, to be on the safe side, I
4452 have made the code more robust (and fixed the comments that describe what
4455 33. Some experimental protocols are using DNS PTR records for new purposes. The
4456 keys for these records are domain names, not reversed IP addresses. The
4457 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4458 leaves it alone. Component reversal etc. now happens only for IP addresses.
4461 34. The host_aton() function is supposed to be passed a string that is known
4462 to be a valid IP address. However, in the case of IPv6 addresses, it was
4463 not checking this. This is a hostage to fortune. Exim now panics and dies
4464 if the condition is not met. A case was found where this could be provoked
4465 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4466 components; fortuitously, this particular loophole had already been fixed
4467 by change 4.50/55 or 4.44/33 above.
4469 If there are any other similar loopholes, the new check in host_aton()
4470 itself should stop them being exploited. The report I received stated that
4471 data on the command line could provoke the exploit when Exim was running as
4472 exim, but did not say which command line option was involved. All I could
4473 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4474 running as the user.
4477 35. There was a buffer overflow vulnerability in the SPA authentication code
4478 (which came originally from the Samba project). I have added a test to the
4479 spa_base64_to_bits() function which I hope fixes it.
4482 36. The daemon start-up calls getloadavg() while still root for those OS that
4483 need the first call to be done as root, but it missed one case: when
4484 deliver_queue_load_max is set with deliver_drop_privilege. This is
4485 necessary for the benefit of the queue runner, because there is no re-exec
4486 when deliver_drop_privilege is set.
4488 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4489 was in use, was not putting the data itself into the right store pool;
4490 consequently, it could be overwritten for a subsequent message in the same
4491 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4494 38. Sometimes the final signoff response after QUIT could fail to get
4495 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4496 < 0 before doing a fflush(). This bug looks as though it goes back to the
4497 introduction of TLS in release 3.20, but "sometimes" must have been rare
4498 because the tests only now provoked it.
4504 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
4505 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
4506 Now it returns an integer. A number of calls were still expecting a T/F
4507 return. Fortuitously, in all cases, the tests worked in OK situations,
4508 which is the norm. However, things would have gone wrong on any write
4509 failures on the smtp file descriptor. This function is used when sending
4510 messages over SMTP and also when doing verify callouts.
4512 2. When Exim is called to do synchronous delivery of a locally submitted
4513 message (the -odf or -odi options), it no longer closes stderr before doing
4516 3. Implemented the mua_wrapper option.
4518 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
4520 5. Implemented the functions header_remove(), header_testname(),
4521 header_add_at_position(), and receive_remove_recipient(), and exported them
4524 6. If an ACL "warn" statement specified the addition of headers, Exim already
4525 inserted X-ACL-Warn: at the start if there was no header name. However, it
4526 was not making this test for the second and subsequent header lines if
4527 there were newlines in the string. This meant that an invalid header could
4528 be inserted if Exim was badly configured.
4530 7. Allow an ACL "warn" statement to add header lines at the start or after all
4531 the Received: headers, as well as at the end.
4533 8. Added the rcpt_4xx retry error code.
4535 9. Added postmaster_mailfrom=xxx to callout verification option.
4537 10. Added mailfrom=xxxx to the callout verification option, for verify=
4540 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
4541 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
4543 12. Inserted some casts to stop certain compilers warning when using pointer
4544 differences as field lengths or precisions in printf-type calls (mostly
4545 affecting debugging statements).
4547 13. Added optional readline() support for -be (dynamically loaded).
4549 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
4550 same clock tick as a message's arrival, so that its received time was the
4551 same as the "first fail" time on the retry record, and that message
4552 remained on the queue past the ultimate address timeout, every queue runner
4553 would try a delivery (because it was past the ultimate address timeout) but
4554 after another failure, the ultimate address timeout, which should have then
4555 bounced the address, did not kick in. This was a "< instead of <=" error;
4556 in most cases the first failure would have been in the next clock tick
4557 after the received time, and all would be well.
4559 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
4560 being recognized when the domain list was tested by the match_domain
4561 condition in an expansion string.
4563 16. Added the ${str2b64: operator.
4565 17. Exim was always calling setrlimit() to set a large limit for the number of
4566 processes, without checking whether the existing limit was already
4567 adequate. (It did check for the limit on file descriptors.) Furthermore,
4568 errors from getrlimit() and setrlimit() were being ignored. Now they are
4569 logged to the main and panic logs, but Exim does carry on, to try to do its
4570 job under whatever limits there are.
4572 18. Imported PCRE 5.0.
4574 19. Trivial typo in log message " temporarily refused connection" (the leading
4577 20. If the log selector return_path_on_delivery was set and an address was
4578 redirected to /dev/null, the delivery process crashed because it assumed
4579 that a return path would always be set for a "successful" delivery. In this
4580 case, the whole delivery is bypassed as an optimization, and therefore no
4583 21. Internal re-arrangement: the function for sending a challenge and reading
4584 a response while authentication was assuming a zero-terminated challenge
4585 string. It's now changed to take a pointer and a length, to allow for
4586 binary data in such strings.
4588 22. Added the cyrus_sasl authenticator (code supplied by MBM).
4590 23. Exim was not respecting finduser_retries when seeking the login of the
4591 uid under which it was called; it was always trying 10 times. (The default
4592 setting of finduser_retries is zero.) Also, it was sleeping after the final
4593 failure, which is pointless.
4595 24. Implemented tls_on_connect_ports.
4597 25. Implemented acl_smtp_predata.
4599 26. If the domain in control=submission is set empty, Exim assumes that the
4600 authenticated id is a complete email address when it generates From: or
4601 Sender: header lines.
4603 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
4604 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
4605 chown and chgrp in /bin and hostname in /usr/bin.
4607 28. Exim was keeping the "process log" file open after each use, just as it
4608 does for the main log. This opens the possibility of it remaining open for
4609 long periods when the USR1 signal hits a daemon. Occasional processlog
4610 errors were reported, that could have been caused by this. Anyway, it seems
4611 much more sensible not to leave this file open at all, so that is what now
4614 29. The long-running daemon process does not normally write to the log once it
4615 has entered its main loop, and it closes the log before doing so. This is
4616 so that log files can straightforwardly be renamed and moved. However,
4617 there are a couple of unusual error situations where the daemon does write
4618 log entries, and I had neglected to close the log afterwards.
4620 30. The text of an SMTP error response that was received during a remote
4621 delivery was being truncated at 512 bytes. This is too short for some of
4622 the long messages that one sometimes sees. I've increased the limit to
4625 31. It is now possible to make retry rules that apply only when a message has a
4626 specific sender, in particular, an empty sender.
4628 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
4629 it possible to be selective about when SMTP synchronization is enforced.
4631 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
4633 32. Implemented hosts_connection_nolog.
4635 33. Added an ACL for QUIT.
4637 34. Setting "delay_warning=" to disable warnings was not working; it gave a
4640 35. Added mailbox_size and mailbox_filecount to appendfile.
4642 36. Added control = no_multiline_responses to ACLs.
4644 37. There was a bug in the logic of the code that waits for the clock to tick
4645 in the case where the clock went backwards by a substantial amount such
4646 that the microsecond fraction of "now" was more than the microsecond
4647 fraction of "then" (but the whole seconds number was less).
4649 38. Added support for the libradius Radius client library this is found on
4650 FreeBSD (previously only the radiusclient library was supported).
4656 1. When certain lookups returned multiple values in the form name=value, the
4657 quoting of the values was not always being done properly. Specifically:
4658 (a) If the value started with a double quote, but contained no whitespace,
4660 (b) If the value contained whitespace other than a space character (i.e.
4661 tabs or newlines or carriage returns) it was not quoted.
4662 This fix has been applied to the mysql and pgsql lookups by writing a
4663 separate quoting function and calling it from the lookup code. The fix
4664 should probably also be applied to nisplus, ibase and oracle lookups, but
4665 since I cannot test any of those, I have not disturbed their existing code.
4667 2. A hit in the callout cache for a specific address caused a log line with no
4668 reason for rejecting RCPT. Now it says "Previous (cached) callout
4669 verification failure".
4671 3. There was an off-by-one bug in the queryprogram router. An over-long
4672 return line was truncated at 256 instead of 255 characters, thereby
4673 overflowing its buffer with the terminating zero. As well as fixing this, I
4674 have increased the buffer size to 1024 (and made a note to document this).
4676 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
4677 when Exim is waiting for an SMTP response from a remote server, Exim
4678 restarts its select() call on the socket, thereby resetting its timeout.
4679 This is not a problem when such interrupts are rare. Somebody set up a cron
4680 job to run exiwhat every 2 minutes, which is less than the normal select()
4681 timeout (5 or 10 minutes). This meant that the select() timeout never
4682 kicked in because it was always reset. I have fixed this by comparing the
4683 time when an interrupt arrives with the time at the start of the first call
4684 to select(). If more time than the timeout has elapsed, the interrupt is
4685 treated as a timeout.
4687 5. Some internal re-factoring in preparation for the addition of Sieve
4688 extensions (by MH). In particular, the "personal" test is moved to a
4689 separate function, and given an option for scanning Cc: and Bcc: (which is
4690 not set for Exim filters).
4692 6. When Exim created an email address using the login of the caller as the
4693 local part (e.g. when creating a From: or Sender: header line), it was not
4694 quoting the local part when it contained special characters such as @.
4696 7. Installed new OpenBSD configuration files.
4698 8. Reworded some messages for syntax errors in "and" and "or" conditions to
4699 try to make them clearer.
4701 9. Callout options, other than the timeout value, were being ignored when
4702 verifying sender addresses in header lines. For example, when using
4704 verify = header_sender/callout=no_cache
4706 the cache was (incorrectly) being used.
4708 10. Added a missing instance of ${EXE} to the exim_install script; this affects
4709 only the Cygwin environment.
4711 11. When return_path_on_delivery was set as a log selector, if different remote
4712 addresses in the same message used different return paths and parallel
4713 remote delivery occurred, the wrong values would sometimes be logged.
4714 (Whenever a remote delivery process finished, the return path value from
4715 the most recently started remote delivery process was logged.)
4717 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
4718 lines when AUTH and/or TLS are in use. This is the "received protocol"
4719 field. Exim used to use "asmtp" for authenticated SMTP, without any
4720 indication (in the protocol name) for TLS use. Now it follows the RFC and
4721 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
4722 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
4723 names appear in log lines as well as in Received: header lines.
4725 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
4726 extensions, and comparison tests, and to fix some bugs.
4728 14. Changes to the "personal" filter test:
4730 (1) The test was buggy in that it was just doing the equivalent of
4731 "contains" tests on header lines. For example, if a user's address was
4732 anne@some.where, the "personal" test would incorrectly be true for
4734 To: susanne@some.where
4736 This test is now done by extracting each address from the header in turn,
4737 and checking the entire address. Other tests that are part of "personal"
4738 are now done using regular expressions (for example, to check local parts
4739 of addresses in From: header lines).
4741 (2) The list of non-personal local parts in From: addresses has been
4742 extended to include "listserv", "majordomo", "*-request", and "owner-*",
4743 taken from the Sieve specification recommendations.
4745 (3) If the message contains any header line starting with "List-" it is
4746 treated as non-personal.
4748 (4) The test for "circular" in the Subject: header line has been removed
4749 because it now seems ill-conceived.
4751 15. Minor typos in src/EDITME comments corrected.
4753 16. Installed latest exipick from John Jetmore.
4755 17. If headers_add on a router specified a text string that was too long for
4756 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
4757 of string_sprintf() is now avoided.
4759 18. $message_body_size was not set (it was always zero) when running the DATA
4760 ACL and the local_scan() function.
4762 19. For the "mail" command in an Exim filter, no default was being set for
4763 the once_repeat time, causing a random time value to be used if "once" was
4764 specified. (If the value happened to be <= 0, no repeat happened.) The
4765 default is now 0s, meaning "never repeat". The "vacation" command was OK
4766 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
4767 (I found it when inspecting the code).
4769 20. There is now an overall timeout for performing a callout verification. It
4770 defaults to 4 times the callout timeout, which applies to individual SMTP
4771 commands during the callout. The overall timeout applies when there is more
4772 than one host that can be tried. The timeout is checked before trying the
4773 next host. This prevents very long delays if there are a large number of
4774 hosts and all are timing out (e.g. when the network connections are timing
4775 out). The value of the overall timeout can be changed by specifying an
4776 additional sub-option for "callout", called "maxwait". For example:
4778 verify = sender/callout=5s,maxwait=20s
4780 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
4781 to the end before writing, but this should make it even safer).
4783 22. Exim was forgetting that it had advertised PIPELINING for the second and
4784 subsequent messages on an SMTP connection. It was also not resetting its
4785 memory on STARTTLS and an internal HELO.
4787 23. When Exim logs an SMTP synchronization error within a session, it now
4788 records whether PIPELINING has been advertised or not.
4790 24. Added 3 instances of "(long int)" casts to time_t variables that were being
4791 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
4792 rather than long int.
4794 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
4796 26. Added the never_mail option to autoreply.
4802 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
4803 crash if the getsockname() call failed; this can happen if a connection is
4804 closed very soon after it is established. The problem was simply in the
4805 order in which certain operations were done, causing Exim to try to write
4806 to the SMTP stream before it had set up the file descriptor. The bug has
4807 been fixed by making things happen in the correct order.
4813 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
4814 before the connection was closed, thus losing the rejection response.
4816 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
4817 some early Solaris releases, but causes trouble in current releases where
4818 socklen_t is defined.
4820 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
4823 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
4826 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
4827 that releasing the top of it at the end releases what was used for sub-
4828 expansions (unless the block got too big). However, discard this block if
4829 the first thing is a variable or header, so that we can use its block when
4830 it is dynamic (useful for very large $message_headers, for example).
4832 6. Lookups now cache *every* query, not just the most recent. A new, separate
4833 store pool is used for this. It can be recovered when all lookup caches are
4834 flushed. Lookups now release memory at the end of their result strings.
4835 This has involved some general refactoring of the lookup sources.
4837 7. Some code has been added to the store_xxx() functions to reduce the amount
4838 of flapping under certain conditions.
4840 8. log_incoming_interface used to affect only the <= reception log lines. Now
4841 it causes the local interface and port to be added to several more SMTP log
4842 lines, for example "SMTP connection from", and rejection lines.
4844 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
4846 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
4848 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
4849 could be overwritten at the end of the current message (or the start of a
4850 new message if it was set in a HELO ACL). The value is now preserved for
4851 the duration of the SMTP connection.
4853 12. If a transport had a headers_rewrite setting, and a matching header line
4854 contained an unqualified address, that address was qualified, even if it
4855 did not match any rewriting rules. The underlying bug was that the values
4856 of the flags that permit the existence of unqualified sender and recipient
4857 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
4858 non-local messages, and by -bnq for local messages) were not being
4859 preserved with the message after it was received.
4861 13. When Exim was logging an SMTP synchronization error, it could sometimes log
4862 "next input=" as part of the text comprising the host identity instead of
4863 the correct text. The code was using the same buffer for two different
4864 strings. However, depending on which order the printing function evaluated
4865 its arguments, the bug did not always show up. Under Linux, for example, my
4866 test suite worked just fine.
4868 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
4869 doesn't work with some older versions of Perl. It has been changed to "my",
4870 which in any case is probably the better facility to use.
4872 15. A really picky compiler found some instances of statements for creating
4873 error messages that either had too many or two few arguments for the format
4876 16. The size of the buffer for calls to the DNS resolver has been increased
4877 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
4878 for addresses that have a lot of PTR records. This alleviates a problem; it
4879 does not fully solve it.
4881 17. A dnsdb lookup for PTR records that receives more data than will fit in the
4882 buffer now truncates the list and logs the incident, which is the same
4883 action as happens when Exim is looking up a host name and its aliases.
4884 Previously in this situation something unpredictable would happen;
4885 sometimes it was "internal error: store_reset failed".
4887 18. If a server dropped the connection unexpectedly when an Exim client was
4888 using GnuTLS and trying to read a response, the client delivery process
4889 crashed while trying to generate an error log message.
4891 19. If a "warn" verb in an ACL added multiple headers to a message in a single
4892 string, for example:
4894 warn message = H1: something\nH2: something
4896 the text was added as a single header line from Exim's point of view
4897 though it ended up OK in the delivered message. However, searching for the
4898 second and subsequent header lines using $h_h2: did not work. This has been
4899 fixed. Similarly, if a system filter added multiple headers in this way,
4900 the routers could not see them.
4902 20. Expanded the error message when iplsearch is called with an invalid key to
4903 suggest using net-iplsearch in a host list.
4905 21. When running tests using -bh, any delays imposed by "delay" modifiers in
4906 ACLs are no longer actually imposed (and a message to that effect is
4909 22. If a "gecos" field in a passwd entry contained escaped characters, in
4910 particular, if it contained a \" sequence, Exim got it wrong when building
4911 a From: or a Sender: header from that name. A second bug also caused
4912 incorrect handling when an unquoted " was present following a character
4913 that needed quoting.
4915 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
4916 was not being matched caselessly.
4918 24. Arranged for all hyphens in the exim.8 source to be escaped with
4921 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
4922 properly. Recipient callout cache records were still being keyed to include
4923 the sender, even when use_sender was set false. This led to far more
4924 callouts that were necessary. The sender is no longer included in the key
4925 when use_sender is false.
4927 26. Added "control = submission" modifier to ACLs.
4929 27. Added the ${base62d: operator to decode base 62 numbers.
4931 28. dnsdb lookups can now access SRV records.
4933 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
4934 the configuration file.
4936 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
4937 (-v) mode. This makes the output for a verbose queue run more intelligible.
4939 31. Added a use_postmaster feature to recipient callouts.
4941 32. Added the $body_zerocount variable, containing the number of binary zero
4942 bytes in the message body.
4944 33. The time of last modification of the "new" subdirectory is now used as the
4945 "mailbox time last read" when there is a quota error for a maildir
4948 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
4950 35. Added +ignore_unknown as a special item in host lists.
4952 36. Code for decoding IPv6 addresses in host lists is now included, even if
4953 IPv6 support is not being compiled. This fixes a bug in which an IPv6
4954 address was recognized as an IP address, but was then not correctly decoded
4955 into binary, causing unexpected and incorrect effects when compared with
4962 1. Very minor rewording of debugging text in manualroute to say "list of
4963 hosts" instead of "hostlist".
4965 2. If verify=header_syntax was set, and a header line with an unqualified
4966 address (no domain) and a large number of spaces between the end of the
4967 name and the colon was received, the reception process suffered a buffer
4968 overflow, and (when I tested it) crashed. This was caused by some obsolete
4969 code that should have been removed. The fix is to remove it!
4971 3. When running in the test harness, delay a bit after writing a bounce
4972 message to get a bit more predictability in the log output.
4974 4. Added a call to search_tidyup() just before forking a reception process. In
4975 theory, someone could use a lookup in the expansion of smtp_accept_max_
4976 per_host which, without the tidyup, could leave open a database connection.
4978 5. Added the variables $recipient_data and $sender_data which get set from a
4979 lookup success in an ACL "recipients" or "senders" condition, or a router
4980 "senders" option, similar to $domain_data and $local_part_data.
4982 6. Moved the writing of debug_print from before to after the "senders" test
4985 7. Change 4.31/66 (moving the time when the Received: is generated) caused
4986 problems for message scanning, either using a data ACL, or using
4987 local_scan() because the Received: header was not generated till after they
4988 were called (in order to set the time as the time of reception completion).
4989 I have revised the way this works. The header is now generated after the
4990 body is received, but before the ACL or local_scan() are called. After they
4991 are run, the timestamp in the header is updated.
4997 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
4998 before starting a queue runner without re-exec. This happened only when
4999 deliver_drop_privilege was set or when the Exim user was set to root. The
5000 effect of the bug was that timeouts during subsequent deliveries caused
5001 crashes instead of being properly handled. The handler is now left at its
5002 default (and expected) setting.
5004 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
5005 message, again when deliver_drop_privilege is set or Exim is run as root.
5006 The bug described in (1) was not present in this case, but the tidying up
5007 of the other signals was missing. I have made the two cases consistent.
5009 3. The ignore_target_hosts setting on a manualroute router was being ignored
5010 for hosts that were looked up using the /MX notation.
5012 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
5015 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
5016 operated on the sender address. After changing the $sender_address to <>
5017 for the sender address verify, Exim was re-instated it as the original
5018 (before rewriting) address, but remembering that it had rewritten it, so it
5019 wasn't rewriting it again. This bug also had the effect of breaking the
5020 sender address verification caching when the sender address was rewritten.
5022 6. The ignore_target_hosts option was being ignored by the ipliteral router.
5023 This has been changed so that if the ip literal address matches
5024 ignore_target_hosts, the router declines.
5026 7. Added expansion conditions match_domain, match_address, and match_local_
5027 part (NOT match_host).
5029 8. The placeholder for the Received: header didn't have a length field set.
5031 9. Added code to Exim itself and to exim_lock to test for a specific race
5032 condition that could lead to file corruption when using MBX delivery. The
5033 issue is with the lockfile that is created in /tmp. If this file is removed
5034 after a process has opened it but before that process has acquired a lock,
5035 there is the potential for a second process to recreate the file and also
5036 acquire a lock. This could lead to two Exim processes writing to the file
5037 at the same time. The added code performs the same test as UW imapd; it
5038 checks after acquiring the lock that its file descriptor still refers to
5039 the same named file.
5041 10. The buffer for building added header lines was of fixed size, 8192 bytes.
5042 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
5045 11. Added the smtp_active_hostname option. If used, this will typically be made
5046 to depend on the incoming interface address. Because $interface_address is
5047 not set up until the daemon has forked a reception process, error responses
5048 that can happen earlier (such as "too many connections") no longer contain
5051 12. If an expansion in a condition on a "warn" statement fails because a lookup
5052 defers, the "warn" statement is abandoned, and the next ACL statement is
5053 processed. Previously this caused the whole ACL to be aborted.
5055 13. Added the iplsearch lookup type.
5057 14. Added ident_timeout as a log selector.
5059 15. Added tls_certificate_verified as a log selector.
5061 16. Added a global option tls_require_ciphers (compare the smtp transport
5062 option of the same name). This controls incoming TLS connections.
5064 17. I finally figured out how to make tls_require_ciphers do a similar thing
5065 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
5066 before starting the TLS session.
5068 18. Tabs are now shown as \t in -bP output.
5070 19. If the log selector return_path_on_delivery was set, Exim crashed when
5071 bouncing a message because it had too many Received: header lines.
5073 20. If two routers both had headers_remove settings, and the first one included
5074 a superfluous trailing colon, the final name in the first list and the
5075 first name in the second list were incorrectly joined into one item (with a
5076 colon in the middle).
5082 1. Added -C and -D options to the exinext utility, mainly to make it easier
5083 to include in the automated testing, but these could be helpful when
5084 multiple configurations are in use.
5086 2. The exinext utility was not formatting the output nicely when there was
5087 an alternate port involved in the retry record key, nor when there was a
5088 message id as well (for retries that were specific to a specific message
5089 and a specific host). It was also confused by IPv6 addresses, because of
5090 the additional colons they contain. I have fixed the IPv4 problem, and
5091 patched it up to do a reasonable job for IPv6.
5093 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
5094 delivery, the log line now contains "pipelined" if PIPELINING was used.
5096 4. An SMTP transport process used to panic and die if the bind() call to set
5097 an explicit outgoing interface failed. This has been changed; it is now
5098 treated in the same way as a connect() failure.
5100 5. A reference to $sender_host_name in the part of a conditional expansion
5101 that was being skipped was still causing a DNS lookup. This no longer
5104 6. The def: expansion condition was not recognizing references to header lines
5105 that used bh_ and bheader_.
5107 7. Added the _cache feature to named lists.
5109 8. The code for checking quota_filecount in the appendfile transport was
5110 allowing one more file than it should have been.
5112 9. For compatibility with Sendmail, the command line option
5120 and sets the incoming protocol and host name (for trusted callers). The
5121 host name and its colon can be omitted when only the protocol is to be set.
5122 Note the Exim already has two private options, -pd and -ps, that refer to
5123 embedded Perl. It is therefore impossible to set a protocol value of "d" or
5124 "s", but I don't think that's a major issue.
5126 10. A number of refactoring changes to the code, none of which should affect
5129 (a) The number of logging options was getting close to filling up the
5130 32-bit word that was used as a bit map. I have split them into two classes:
5131 those that are passed in the argument to log_write(), and those that are
5132 only ever tested independently outside of that function. These are now in
5133 separate 32-bit words, so there is plenty of room for expansion again.
5134 There is no change in the user interface or the logging behaviour.
5136 (b) When building, for example, log lines, the code previously used a
5137 macro that called string_cat() twice, in order to add two strings. This is
5138 not really sufficiently general. Furthermore, there was one instance where
5139 it was actually wrong because one of the argument was used twice, and in
5140 one call a function was used. (As it happened, calling the function twice
5141 did not affect the overall behaviour.) The macro has been replaced by a
5142 function that can join an arbitrary number of extra strings onto a growing
5145 (c) The code for expansion conditions now uses a table and a binary chop
5146 instead of a serial search (which was left over from when there were very
5147 few conditions). Also, it now recognizes conditions like "pam" even when
5148 the relevant support is not compiled in: a suitably worded error message is
5149 given if an attempt is made to use such a condition.
5151 11. Added ${time_interval:xxxxx}.
5153 12. A bug was causing one of the ddress fields not to be passed back correctly
5154 from remote delivery subprocesses. The field in question was not being
5155 subsequently used, so this caused to problems in practice.
5157 13. Added new log selectors queue_time and deliver_time.
5159 14. Might have fixed a bug in maildirsizefile handling that threw up
5160 "unexpected character" debug warnings, and recalculated the data
5161 unnecessarily. In any case, I expanded the warning message to give more
5164 15. Added the message "Restricted characters in address" to the statements in
5165 the default ACL that block characters like @ and % in local parts.
5167 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
5168 Three changes have been made:
5170 (a) There was a serious bug; a negative response to MAIL caused the whole
5171 recipient domain to be cached as invalid, thereby blocking all messages
5172 to all local parts at the same domain, from all senders. This bug has
5173 been fixed. The domain is no longer cached after a negative response to
5174 MAIL if the sender used is not empty.
5176 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
5179 (c) A new callout option, "use_sender" has been added for people who want
5180 the modified behaviour.
5186 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
5189 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
5190 indeed breaks things for older releases.
5192 3. Added additional logging to the case where there is a problem reading data
5193 from a filter that is running in a subprocess using a pipe, in order to
5194 try to track down a specific problem.
5196 4. Testing facility fudge: when running in the test harness and attempting
5197 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
5198 getting "No route to host". Convert this to a timeout.
5200 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
5203 6. Some OS don't have socklen_t but use size_t instead. This affects the
5204 fifth argument of getsockopt() amongst other things. This is now
5205 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
5206 can be set for individual OS. I have set it for SunOS5, OSF1, and
5207 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
5208 some earlier ones do not.
5210 7. Change 4.30/15 was not doing the test caselessly.
5212 8. The standard form for an IPv6 address literal was being rejected by address
5213 parsing in, for example, MAIL and RCPT commands. An example of this kind of
5214 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5215 this, as well as the form without the "IPv6" on the front (but only when
5216 address literals are enabled, of course).
5218 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
5220 10. Exim crashed if a message with an empty sender address specified by -f
5221 encountered a router with an errors_to setting. This could be provoked only
5222 by a command such as
5226 where an empty string was supplied; "<>" did not hit this bug.
5228 11. Installed PCRE release 4.5.
5230 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
5231 remained set. It is now erased.
5233 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
5234 times from message ids (which are base 36 rather than the normal 62).
5236 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
5237 were being counted as actual protocol errors, and logged if the log
5238 selector +smtp_protocol_error was set. One cannot be perfect in this test,
5239 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
5240 and DATA following a set of rejected RCPTs do not count as protocol errors.
5241 In other words, Exim assumes they were pipelined, though this may not
5242 actually be the case. Of course, in all cases the client gets an
5243 appropriate error code.
5245 15. If a lookup fails in an ACL condition, a message about the failure may
5246 be available; it is used if testing the ACL cannot continue, because most
5247 such messages specify what the cause of the deferral is. However, some
5248 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
5249 that caused an old message to be retained and used if a later statement
5250 caused a defer, replacing the real cause of the deferral.
5252 16. If an IP address had so many PTR records that the DNS lookup buffer
5253 was not large enough to hold them, Exim could crash while trying to process
5254 the truncated data. It now detects and logs this case.
5256 17. Further to 4.21/58, another change has been made: if (and only if) the
5257 first line of a message (the first header line) ends with CRLF, a bare LF
5258 in a subsequent header line has a space inserted after it, so as not to
5259 terminate the header.
5261 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
5262 unnecessarily, used atoi() instead of strtol(), and didn't check the
5263 termination when getting file sizes from file names by regex.
5265 19. Completely re-implemented the support for maildirsize files, in the light
5266 of a number of problems with the previous contributed implementation
5267 (4.30/29). In particular:
5269 . If the quota is zero, the maildirsize file is maintained, but no quota is
5272 . If the maildir directory does not exist, it is created before any attempt
5273 to write a maildirsize file.
5275 . The quota value in the file is just a cache; if the quota is changed in
5276 the transport, the new value overrides.
5278 . A regular expression is available for excluding directories from the
5281 20. The autoreply transport checks the characters in options that define the
5282 message's headers; it allows continued headers, but it was checking with
5283 isspace() after an embedded newline instead of explicitly looking for a
5286 21. If all the "regular" hosts to which an address was routed had passed their
5287 expiry times, and had not reached their retry times, the address was
5288 bounced, even if fallback hosts were defined. Now Exim should go on to try
5291 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
5292 equivalent code in the SMTP transport. Some hosts send humungous responses
5293 to HELO/EHLO, more than 1024 it seems.
5295 23. Refactoring: code in filter.c used (void *) for "any old type" but this
5296 gives compiler warnings in some environments. I've now done it "properly",
5299 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
5300 (because of problems with the built-in one) was declared to return uschar *
5301 instead of char *, causing compiler failure.
5303 25. Fixed a file descriptor leak when processing alias/forward files.
5305 26. Fixed a minor format string issue in dbfn.c.
5307 27. Typo in exim.c: ("dmbnz" for "dbmnz").
5309 28. If a filter file refered to $h_xxx or $message_headers, and the headers
5310 contained RFC 2047 "words", Exim's memory could, under certain conditions,
5313 29. When a sender address is verified, it is cached, to save repeating the test
5314 when there is more than one recipient in a message. However, when the
5315 verification involves a callout, it is possible for different callout
5316 options to be set for different recipients. It is too complicated to keep
5317 track of this in the cache, so now Exim always runs a verification when a
5318 callout is required, relying on the callout cache for the optimization.
5319 The overhead is duplication of the address routing, but this should not be
5322 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
5323 to be verified with callout=postmaster, and the main callout worked but the
5324 postmaster check failed, the verification correctly failed. However, if a
5325 subsequent RCPT command asked for sender verification *without* the
5326 postmaster check, incorrect caching caused this verification also to fail,
5329 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
5330 it was not caching the DNS options (qualify_single, search_parents) that
5331 were used when the lookup failed. A subsequent lookup with different
5332 options therefore always gave the same answer, though there were cases
5333 where it should not have. (Example: a "domains = !$mx_any" option on a
5334 dnslookup router: the "domains" option is always processed without any
5335 widening, but the router might have qualify_single set.) Now Exim uses the
5336 cached value only when the same options are set.
5338 32. Added John Jetmore's "exipick" utility to the distribution.
5340 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
5341 than a timeout (e.g. a certificate is required, and is not provided), an
5342 Exim server now closes the connection immediately. Previously it waited for
5343 the client to close - but if the client is SSL, it seems that they each
5344 wait for each other, leading to a delay before one of them times out.
5346 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
5347 maintained 0.8.x compatibility because I don't think many are using it, and
5348 it is clearly obsolete.
5350 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
5353 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
5354 client certificate was expired. A simple patch fixes this, though I don't
5355 understand the full logic of why the verify callback is called multiple
5358 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
5359 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
5360 which causes problems with some clients (such as the Certicom SSL Plus
5361 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
5362 disables the coutermeasure allowing Eudora to connect."
5364 38. Exim was not checking that a write() to a log file succeeded. This could
5365 lead to Bad Things if a log got too big, in particular if it hit a file
5366 size limit. Exim now panics and dies if it cannot write to a log file, just
5367 as it does if it cannot open a log file.
5369 39. Modified OS/Makefile-Linux so that it now contains
5371 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
5373 The two -D definitions ensure that Exim is compiled with large file
5374 support, which makes it possible to handle log files that are bigger than
5377 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
5378 instance) a domain was checked against a named list that involved a lookup,
5379 causing $domain_data to be set, then another domain was checked against the
5380 same list, then the first domain was re-checked, the value of $domain_data
5381 after the final check could be wrong. In particular, if the second check
5382 failed, it could be set empty. This bug probably also applied to
5385 41. The strip_trailing_dot option was not being applied to the address given
5386 with the -f command-line option.
5388 42. The code for reading a message's header from the spool was incrementing
5389 $received_count, but never initializing it. This meant that the value was
5390 incorrect (doubled) while delivering a message in the same process in which
5391 it was received. In the most common configuration of Exim, this never
5392 happens - a fresh exec is done - but it can happen when
5393 deliver_drop_privilege is set.
5395 43. When Exim logs an SMTP synchronization error - client data sent too soon -
5396 it now includes up to 150 characters of the unexpected data in the log
5399 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
5400 and building data strings. The size of both of these buffers was 10 000
5401 bytes - far larger than anybody would *ever* want, thought I. Needless to
5402 say, somebody hit the limit. I have increased the maximum line length to
5403 20 000 and the maximum data length of concatenated lines to 100 000. I have
5404 also fixed two bugs, because there was no checking on these buffers. Tsk,
5405 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
5406 buffer is too small.
5408 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
5409 lsearch lookups. Now it does.
5411 46. When parsing a route_list item in a manualroute router, a fixed-length
5412 buffer was used for the list of hosts. I made this 1024 bytes long,
5413 thinking that nobody would ever have a list of hosts that long. Wrong.
5414 Somebody had a whole pile of complicated expansion conditions, and the
5415 string was silently truncated, leading to an expansion error. It turns out
5416 that it is easier to change to an unlimited length (owing to other changes
5417 that have happened since this code was originally written) than to build
5418 structure for giving a limitation error. The length of the item that
5419 expands into the list of hosts is now unlimited.
5421 47. The lsearch lookup could not handle data where the length of text line was
5422 more than 4095 characters. Such lines were truncated, leading to shortened
5423 data being returned. It should now handle lines of any length.
5425 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
5426 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
5429 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
5430 replacing strings such as BIN_DIRECTORY in a source file: the replacement
5431 no longer happens in comment lines. A list of replacements is now placed
5432 at the head of all of the source files, except those whose only change is
5433 to replace PERL_COMMAND in the very first #! line.
5435 50. Replaced the slow insertion sort in queue.c, for sorting the list of
5436 messages on the queue, with a bottom-up merge sort, using code contributed
5437 by Michael Haardt. This should make operations like -bp somewhat faster on
5438 large queues. It won't affect queue runners, except when queue_run_in_order
5441 51. Installed eximstats 1.31 in the distribution.
5443 52. Added support for SRV lookups to the dnslookup router.
5445 53. If an ACL referred to $message_body or $message_body_end, the value was not
5446 reset for any messages that followed in the same SMTP session.
5448 54. The store-handling optimization for building very long strings was not
5449 differentiating between the different store pools. I don't think this
5450 actually made any difference in practice, but I've tidied it.
5452 55. While running the routers to verify a sender address, $sender_address
5453 was still set to the sender address. This is wrong, because when routing to
5454 send a bounce to the sender, it would be empty. Therefore, I have changed
5455 it so that, while verifying a sender address, $sender_address is set to <>.
5456 (There is no change to what happens when verifying a recipient address.)
5458 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
5459 target A or AAAA records (if not already returned) without resetting the
5460 qualify_single or search_parents options of the DNS resolver. These are
5461 inappropriate in this case because the targets of MX and SRV records must
5462 be FQDNs. A broken DNS record could cause trouble if it happened to have a
5463 target that, when qualified, matched something in the local domain. These
5464 two options are now turned off when doing these lookups.
5466 57. It seems that at least some releases of Reiserfs (which does not have the
5467 concept of a fixed number of inodes) returns zero and not -1 for the
5468 number of available inodes. This interacted badly with check_spool_inodes,
5469 which assumed that -1 was the "no such thing" setting. What I have done is
5470 to check that the total number of inodes is greater than zero before doing
5471 the test of how many are available.
5473 58. When a "warn" ACL statement has a log_message modifier, the message is
5474 remembered, and not repeated. This is to avoid a lot of repetition when a
5475 message has many recipients that cause the same warning to be written.
5476 Howewer, Exim was preserving the list of already written lines for an
5477 entire SMTP session, which doesn't seem right. The memory is now reset if a
5478 new message is started.
5480 59. The "rewrite" debugging flag was not showing the result of rewriting in the
5481 debugging output unless log_rewrite was also set.
5483 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
5484 of (int)(handle) when we know that handle contains (void *)(-1).
5486 61. The Exim daemon panic-logs an error return when it closes the incoming
5487 connection. However "connection reset by peer" seems to be common, and
5488 isn't really an error worthy of noting specially, so that particular error
5491 62. When Exim is trying to find all the local interfaces, it used to panic and
5492 die if the ioctl to get the interface flags failed. However, it seems that
5493 on at least one OS (Solaris 9) it is possible to have an interface that is
5494 included in the list of interfaces, but for which you get a failure error
5495 for this call. This happens when the interface is not "plumbed" into a
5496 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
5497 failure of the "get flags" call assumes that the interface is down.
5499 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
5500 makes life easier for people who are doing arithmetic on fields extracted
5501 from dates, where you often get leading zeros that should not be
5502 interpreted as octal.
5504 64. Added qualify_domain to the redirect router, to override the global
5507 65. If a pathologically long header line contained very many addresses (the
5508 report of this problem mentioned 10 000) and each of them was rewritten,
5509 Exim could use up a very large amount of memory. (It kept on making new
5510 copies of the header line as it rewrote, and never released the old ones.)
5511 At the expense of a bit more processing, the header rewriting function has
5512 been changed so that it no longer eats memory in this way.
5514 66. The generation of the Received: header has been moved from the time that a
5515 message starts to be received, to the time that it finishes. The timestamp
5516 in the Received: header should now be very close to that of the <= log
5517 line. There are two side-effects of this change:
5519 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
5520 logged header lines no longer include the local Received: line, because
5521 it has not yet been created. The same applies to a copy of the message
5522 that is returned to a non-SMTP sender when a message is rejected.
5524 (b) When a filter file is tested using -bf, no additional Received: header
5525 is added to the test message. After some thought, I decided that this
5528 This change does not affect the value of $received_for. It is still set
5529 after address rewriting, but before local_scan() is called.
5531 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
5533 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
5534 gave an unhelpful panic error message, and a defer error. I have managed to
5535 change this behaviour so that it now rejects any supplied certificate,
5536 which seems right, as the list of acceptable certificates is empty.
5538 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
5539 gave an unhelpful defer error. I have not managed to make this reject any
5540 supplied certificates, but the error message it gives is "no certificate
5541 supplied", which is not helpful.
5543 70. exigrep's output now also includes lines that are not associated with any
5544 message, but which match the given pattern. Implemented by a patch from
5545 Martin Sluka, which also tidied up the Perl a bit.
5547 71. Recipient callout verification, like sender verification, was using <> in
5548 the MAIL FROM command. This isn't really the right thing, since the actual
5549 sender may affect whether the remote host accepts the recipient or not. I
5550 have changed it to use the actual sender in the callout; this means that
5551 the cache record is now keyed on a recipient/sender pair, not just the
5552 recipient address. There doesn't seem to be a real danger of callout loops,
5553 since a callout by the remote host to check the sender would use <>.
5554 [SEE ABOVE: changed after hitting problems.]
5556 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
5557 temporary errors. However, in the case of such a code being given after
5558 the end of a data transmission (i.e. after ".") Exim was failing to write
5559 a retry record for the message. (Yes, there was some broken host that was
5560 actually sending 8xx at this point.)
5562 73. An unknown lookup type in a host list could cause Exim to panic-die when
5563 the list was checked. (An example that provoked this was putting <; in the
5564 middle of a list instead of at the start.) If this happened during a DATA
5565 ACL check, a -D file could be left lying around. This kind of configuration
5566 error no longer causes Exim to die; instead it causes a defer errror. The
5567 incident is still logged to the main and panic logs.
5569 74. Buglet left over from Exim 3 conversion. The message "too many messages
5570 in one connection" was written to the rejectlog but not the mainlog, except
5571 when address rewriting (yes!) was being logged.
5573 75. Added write_rejectlog option.
5575 76. When a system filter was run not as root (that is, when system_filter_user
5576 was set), the values of the $n variables were not being returned to the
5577 main process; thus, they were not subsequently available in the $sn
5580 77. Added +return_path_on_delivery log selector.
5582 78. A connection timeout was being treated differently from recipients deferred
5583 when testing hosts_max_try with a message that was older than the host's
5584 retry timeout. (The host should not be counted, thus allowing all hosts to
5585 be tried at least once before bouncing.) This may have been the cause of an
5586 occasionally reported bug whereby a message would remain on the queue
5587 longer than the retry timeout, but would be bounced if a delivery was
5588 forced. I say "may" because I never totally pinned down the problem;
5589 setting up timeout/retry tests is difficult. See also the next item.
5591 79. The ultimate address timeout was not being applied to errors that involved
5592 a combination of host plus message (for example, a timeout on a MAIL
5593 command). When an address resolved to a number of possible hosts, and they
5594 were not all tried for each delivery (e.g. because of hosts_max_try), a
5595 message could remain on the queue longer than the retry timeout.
5597 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
5600 81. Fixed an obscure SMTP outgoing bug which required at least the following
5601 conditions: (a) there was another message waiting for the same server;
5602 (b) the server returned 5xx to all RCPT commands in the first message so
5603 that the message was not completed; (c) the server dropped the connection
5604 or gave a negative response to the RSET that Exim sends to abort the
5605 transaction. The observed case was a dropped connection after DATA that had
5606 been sent in pipelining mode. That is, the server had advertised PIPELINING
5607 but was not implementing it correctly. The effect of the bug was incorrect
5608 behaviour, such as trying another host, and this could lead to a crash.
5614 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
5615 and daemon.c were passed as pointers to ints; they should have been
5616 pointers to socklen_t variables (which are typically unsigned ints).
5618 2. Some signed/unsigned type warnings in the os.c file for Linux have been
5621 3. Fixed a really odd bug that affected only the testing scheme; patching a
5622 certain fixed string in the binary changed the value of another string that
5623 happened to be identical to the end of the original first string.
5625 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
5626 name", it returns that address as the IP address. On some operating
5627 systems (e.g. Solaris), it also passes back the IP address string as the
5628 "host name". However, on others (e.g. Linux), it passes back an empty
5629 string. Exim wasn't checking for this, and was changing the host name to an
5630 empty string, assuming it had been canonicized.
5632 5. Although rare, it is permitted to have more than one PTR record for a given
5633 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
5634 all the names associated with an address, because they do in Solaris.
5635 However, it seems that they do not in Linux for data that comes from the
5636 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
5637 I found this out when I moved to a new Linux workstation and tried to run
5638 the Exim test suite.
5640 To get round this problem I have changed the code so that it now does its
5641 own call to the DNS to look up PTR records when searching for a host name.
5642 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
5643 addresses that are only in /etc/hosts are still found.
5645 This behaviour is, however, controlled by an option called host_lookup_
5646 order, which defaults to "bydns:byaddr". If people want to use the other
5647 order, or indeed, just use one or the other means of lookup, they can
5648 specify it in this variable.
5650 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
5651 some operating systems, this comes back from gethostbyaddr() as an empty
5652 string, and this is what Exim used to test for. However, it seems that in
5653 other systems, "." is yielded. Exim now tests for this case too.
5655 7. The values of check_spool_space and check_log_space are now held internally
5656 as a number of kilobytes instead of an absolute number of bytes. If a
5657 numbers is specified without 'K' or 'M', it is rounded up to the nearest
5658 kilobyte. This means that much larger values can be stored.
5660 8. Exim monitor: an attempt to get the action menu when not actually pointing
5661 at a message produces an empty menu entitled "No message selected". This
5662 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
5663 no entries in it ("Shell widget menu has zero width and/or height"). So I
5664 have added a single, blank menu entry in this case.
5666 9. Added ${quote_local_part.
5668 10. MIME decoding is now applied to the contents of Subject: header lines when
5671 11. Now that a reference to $sender_host_address automatically causes a reverse
5672 lookup to occur if necessary (4.13/18), there is no need to arrange for a
5673 host lookup before query-style lookups in lists that might use this
5674 variable. This has therefore been abolished, and the "net-" prefix is no
5675 longer necessary for query-style lookups.
5677 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
5678 have been a typo for LFLAGS, so it has been changed.
5680 13. The install script calls Exim with "-C /dev/null" in order to find the
5681 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
5682 to be output. Howeve, since Exim outputs its version number before the
5683 error, it didn't break the script. It just looked ugly. I fixed this by
5684 always allowing "-C /dev/null" if the caller is root.
5686 14. Ignore overlarge ACL variable number when reading spool file - insurance
5687 against a later release with more variables having written the file.
5689 15. The standard form for an IPv6 address literal was being rejected by EHLO.
5690 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5691 this, as well as the form without the "IPv6" on the front.
5693 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
5694 OS/Makefile-Darwin file.
5696 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
5697 only to LDAP libraries that do not have LDAP_OPT_DEREF.
5699 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
5700 variable. However, some OS use ints for inodes. I've added cast to long int
5701 to get rid of the compiler warning.
5703 19. I had forgotten to lock out "/../" in configuration file names when
5704 ALT_CONFIG_PREFIX was set.
5706 20. Routers used for verification do not need to specify transports. However,
5707 if such a router generated a host list, and callout was configured, Exim
5708 crashed, because it could not find a port number from the (non-existent)
5709 transport. It now assumes port 25 in this circumstance.
5711 21. Added the -t option to exigrep.
5713 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
5714 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
5715 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
5716 them is set, LOOKUP_LSEARCH is forced.
5718 23. "exim -bV" now outputs a list of lookups that are included in the binary.
5720 24. Added sender and host information to the "rejected by local_scan()" log
5721 line; previously there was no indication of these.
5723 25. Added .include_if_exists.
5725 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
5726 Linux. It turns out that not all file systems support this. Apparently some
5727 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
5728 do it.) To cope with this, the error EINVAL, which means that sync-ing is
5729 not supported on the file descriptor, is now ignored when Exim is trying to
5730 sync a directory. This applies only to Linux.
5732 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
5734 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
5735 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
5738 29. Added support for maildirsize files from supplied patch (modified a bit).
5740 30. The use of :fail: followed by an empty string could lead Exim to respond to
5741 sender verification failures with (e.g.):
5743 550 Verification failed for <xxx>
5744 550 Sender verify failed
5746 where the first response line was missing the '-' that indicates it is not
5747 the final line of the response.
5749 31. The loop for finding the name of the user that called Exim had a hardwired
5750 limit of 10; it now uses the value of finduser_retries, which is used for
5751 all other user lookups.
5753 32. Added $received_count variable, available in data and not_smtp ACLs, and at
5756 33. Exim was neglecting to zero errno before one call of strtol() when
5757 expanding a string and expecting an integer value. On some systems this
5758 resulted in spurious "integer overflow" errors. Also, it was casting the
5759 result into an int without checking.
5761 34. Testing for a connection timeout using "timeout_connect" in the retry rules
5762 did not work. The code looks as if it has *never* worked, though it appears
5763 to have been documented since at least releast 1.62. I have made it work.
5765 35. The "timeout_DNS" error in retry rules, also documented since at least
5766 1.62, also never worked. As it isn't clear exactly what this means, and
5767 clearly it isn't a major issue, I have abolished the feature by treating it
5768 as "timeout", and writing a warning to the main and panic logs.
5770 36. The display of retry rules for -brt wasn't always showing the error code
5773 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
5774 timeout_connect_A, timeout_connect_MX.
5776 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
5777 to the empty sender.
5779 39. The daemon was not analysing the content of -oX till after it had closed
5780 stderr and disconnected from the controlling terminal. This meant that any
5781 syntax errors were only noted on the panic log, and the return code from
5782 the command was 0. By re-arranging the code a little, I've made the
5783 decoding happen first, so such errors now appear on stderr, and the return
5784 code is 1. However, the actual setting up of the sockets still happens in
5785 the disconnected process, so errors there are still only recorded on the
5788 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
5789 connections (as happens on some IP stacks) was logged at start up time as
5790 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
5791 it from "IPv6 and IPv4", which means that two separate sockets are being
5794 41. The debug output for gethostbyname2() or getipnodebyname() failures now
5795 says whether AF_INET or AF_INET6 was passed as an argument.
5797 42. Exiwhat output was messed up when time zones were included in log
5800 43. Exiwhat now gives more information about the daemon's listening ports,
5801 and whether -tls-on-connect was used.
5803 44. The "port" option of the smtp transport is now expanded.
5805 45. A "message" modifier in a "warn" statement in a non-message ACL was being
5806 silently ignored. Now an error message is written to the main and panic
5809 46. There's a new ACL modifier called "logwrite" which writes to a log file
5810 as soon as it is encountered.
5812 47. Added $local_user_uid and $local_user_gid at routing time.
5814 48. Exim crashed when trying to verify a sender address that was being
5817 49. Exim was recognizing only a space character after ".include". It now also
5818 recognizes a tab character.
5820 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
5821 extracting the relevant information from the specification. The man page no
5822 longer contains scrambled data for the -d option, and I've added a section
5823 at the front about calling Exim under different names.
5825 51. Added "extra_headers" argument to the "mail" command in filter files.
5827 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
5830 53. Installed eximstats 1.29.
5832 54. Added transport_filter_timeout as a generic transport option.
5834 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
5835 Cc: header lines. This was required by RFC 822, but it not required by RFC
5838 56. Exim used to add From:, Date:, and Message-Id: header lines to any
5839 incoming messages that did not have them. Now it does so only if the
5840 message originates locally, that is, if there is no associated remote host
5841 address. When Resent- header lines are present, this applies to the Resent-
5842 lines rather than the non-Resent- lines.
5844 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
5845 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
5847 58. Messages for configuration errors now include the name of the main
5848 configuration file - useful now that there may be more than one file in a
5849 list (.included file names were always shown).
5851 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
5852 for those rare installations that do not start the daemon as root or run it
5853 setuid root. I've cut out the call to initgroups() if the daemon is not
5856 60. The Exim user and group can now be bound into the binary as text strings
5857 that are looked up at the start of Exim's processing.
5859 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
5861 62. Added $mailstore_basename variable.
5863 63. Installed patch to sieve.c from Michael Haardt.
5865 64. When Exim failed to open the panic log after failing to open the main log,
5866 the original message it was trying to log was written to stderr and debug
5867 output, but if they were not available (the usual case in production), it
5868 was lost. Now it is written to syslog before the two lines that record the
5869 failures to open the logs.
5871 65. Users' Exim filters run in subprocesses under the user's uid. It is
5872 possible for a "deliver" command or an alias in a "personal" command to
5873 provoke an address rewrite. If logging of address rewriting is configured,
5874 this fails because the process is not running as root or exim. There may be
5875 a better way of dealing with this, but for the moment (because 4.30 needs
5876 to be released), I have disabled address rewrite logging when running a
5877 filter in a non-root, non-exim process.
5883 1. The buildconfig auxiliary program wasn't quoting the value set for
5884 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
5885 not defined. This bug was masked in 4.22 by the effect that was fixed in
5888 2. Some messages that were rejected after a message id was allocated were
5889 shown as "incomplete" by exigrep. It no longer does this for messages that
5890 are rejected by local_scan() or the DATA or non-SMTP ACLs.
5892 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
5893 have allow_domain_literals set, the ID did not get logged in the <= line.
5894 Domain literals are now always recognized in Message-ID: header lines.
5896 4. The first argument for a ${extract expansion item is the key name or field
5897 number. Leading and trailing spaces in this item were not being ignored,
5898 causing some misleading effects.
5900 5. When deliver_drop_privilege was set, single queue runner processes started
5901 manually (i.e. by the command "exim -q") or by the daemon (which uses the
5902 same command in the process it spins off) were not dropping privilege.
5904 6. When the daemon running as "exim" started a queue runner, it always
5905 re-executed Exim in the spun-off process. This is a waste of effort when
5906 deliver_drop_privilege is set. The new process now just calls the
5907 queue-runner function directly.
5913 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
5916 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
5917 ignored. Though the use of -oP was forcing the writing of a pid file, it
5918 was always written to the default place.
5920 3. If the message "no IP address found for host xxxx" is generated during
5921 incoming verification, it is now followed by identification of the incoming
5922 connection (so you can more easily find what provoked it).
5924 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
5926 5. Added some features to "harden" Exim a bit more against certain attacks:
5928 (a) There is now a build-time option called FIXED_NEVER_USERS that can
5929 be put in Local/Makefile. This is like the never_users runtime option,
5930 but it cannot be overridden. The default setting is "root".
5932 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
5933 prefix string with which any file named in a -C command line option
5936 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
5937 is retained for -C and -D only if the caller of Exim is root. Without
5938 it, the exim user may also use -C and -D and retain privilege.
5940 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
5941 command line option is disabled.
5943 6. Macro names set by the -D option must start with an upper case letter, just
5944 like macro names defined in the configuration file.
5946 7. Added "dereference=" facility to LDAP.
5948 8. Two instances of the typo "uknown" in the source files are fixed.
5950 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
5951 the Configure-Makefile script screwed up while processing it.
5953 10. Incorporated PCRE 4.4.
5955 11. The SMTP synchronization check was not operating right at the start of an
5956 SMTP session. For example, it could not catch a HELO sent before the client
5957 waited for the greeting. There is now a check for outstanding input at the
5958 point when the greeting is written. Because of the duplex, asynchronous
5959 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
5960 way, but not yet received, when the check is performed.
5962 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
5963 on TCP/IP sockets, because this apparently causes some broken clients to
5966 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
5967 unchanged) from the Cygwin maintainer.
5969 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
5970 was supported instead of testing for mbx. Effectively a typo.
5972 15. The spa authenticator server code was not checking that the input it
5973 received was valid base64.
5975 16. The debug output line for the "set" modifier in ACLs was not showing the
5976 name of the variable that was being set.
5978 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
5980 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
5981 lookup on its own. Something else was needed to trigger the lookup. For
5982 example, a match in host_lookup or the need for a host name in a host list.
5983 Now, if $sender_host_name is referenced and the host name has not yet been
5984 looked up, a lookup is performed. If the lookup fails, the variable remains
5985 empty, and $host_lookup_failed is set to "1".
5987 19. Added "eqi" as a case-independent comparison operator.
5989 20. The saslauthd authentication condition could segfault if neither service
5990 nor realm was specified.
5992 21. If an overflowing value such as "2048M" was set for message_size_limit, the
5993 error message that was logged was misleading, and incoming SMTP
5994 connections were dropped. The message is now more accurate, and temporary
5995 errors are given to SMTP connections.
5997 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
5998 (except RSET) with a 421 error, until QUIT is received. However, it was
5999 failing to send a response to QUIT.
6001 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
6002 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
6003 ACL is now run after the helo_try_verify_hosts code.
6005 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
6006 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
6007 case-independent, so other case variants are also recognized). Apparently
6008 some systems use these upper case variants.
6010 25. If more than two messages were waiting for the same host, and a transport
6011 filter was specified for the transport, Exim sent two messages over the
6012 same TCP/IP connection, and then failed with "socket operation on non-
6013 socket" when it tried to send the third.
6015 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
6017 27. The extern definition of crypt16() in expand.c was not being excluded when
6018 the OS had its own crypt16() function.
6020 28. Added bounce_return_body as a new option, and bounce_return_size_limit
6021 as a preferred synonym for return_size_limit, both as an option and as an
6024 29. Added LIBS=-liconv to OS/Makefile-OSF1.
6026 30. Changed the default configuration ACL to relax the local part checking rule
6027 for addresses that are not in any local domains. For these addresses,
6028 slashes and pipe symbols are allowed within local parts, but the sequence
6029 /../ is explicitly forbidden.
6031 31. SPA server authentication was not clearing the challenge buffer before
6034 32. log_message in a "warn" ACL statement was writing to the reject log as
6035 well as to the main log, which contradicts the documentation and doesn't
6036 seem right (because no rejection is happening). So I have stopped it.
6038 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
6039 However, I am unable to do any testing of this.
6041 34. Fixed an infelicity in the appendfile transport. When checking directories
6042 for a mailbox, to see if any needed to be created, it was accidentally
6043 using path names with one or more superfluous leading slashes; tracing
6044 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
6046 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
6047 contents are added to the log line that is written for every discarded
6048 recipient. (Previously a log_message setting was ignored.)
6050 36. The ${quote: operator now quotes the string if it is empty.
6052 37. The install script runs exim in order to find its version number. If for
6053 some reason other than non-existence or emptiness, which it checks, it
6054 could not run './exim', it was installing it with an empty version number,
6055 i.e. as "exim-". This error state is now caught, and the installation is
6058 38. An argument was missing from the function that creates an error message
6059 when Exim fails to connect to the socket for saslauthd authentication.
6060 This could cause Exim to crash, or give a corrupted message.
6062 39. Added isip, isip4, and isip6 to ${if conditions.
6064 40. The ACL variables $acl_xx are now saved with the message, and can be
6065 accessed later in routers, transports, and filters.
6067 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
6068 strings in the file are not string-expanded.
6070 42. If a MAIL command specified a SIZE value that was too large to fit into an
6071 int variable, the check against message_size_limit failed. Such values are
6072 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
6073 day this will have to be increased, but I don't think I want to be around
6074 when emails are that large.
6081 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
6082 iconv() is not standard in FreeBSD.
6084 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
6085 IPv6 enabled. The observed symptom was a segmentation fault on return from
6086 the function os_common_find_running_interfaces() in src/os.c.
6088 3. In the check_special_case() function in daemon.c I had used "errno" as an
6089 argument name, which causes warnings on some systems. This was basically a
6090 typo, since it was named "eno" in the comments!
6092 4. The code that waits for the clock to tick (at a resolution of some fraction
6093 of a second) so as to ensure message-id uniqueness was always waiting for
6094 at least one whole tick, when it could have waited for less. [This is
6095 almost certainly not relevant at current processor speeds, where it is
6096 unlikely to ever wait at all. But we try to future-proof.]
6098 5. The function that sleeps for a time interval that includes fractions of a
6099 second contained a race. It did not block SIGALRM between setting the
6100 timer, and suspending (a couple of lines later). If the interval was short
6101 and the sigsuspend() was delayed until after it had expired, the suspension
6102 never ended. On busy systems this could lead to processes getting stuck for
6105 6. Some uncommon configurations may cause a lookup to happen in a queue runner
6106 process, before it forks any delivery processes. The open lookup caching
6107 mechanism meant that the open file or database connection was passed into
6108 the delivery process. The problem was that delivery processes always tidy
6109 up cached lookup data. This could cause a problem for the next delivery
6110 process started by the queue runner, because the external queue runner
6111 process does not know about the closure. So the next delivery process
6112 still has data in the lookup cache. In the case of a file lookup, there was
6113 no problem because closing a file descriptor in a subprocess doesn't affect
6114 the parent. However, if the lookup was caching a connection to a database,
6115 the connection was closed, and the second delivery process was likely to
6116 see errors such as "PGSQL: query failed: server closed the connection
6117 unexpectedly". The problem has been fixed by closing all cached lookups
6118 in a queue runner before running a delivery process.
6120 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
6121 seem to have the "const" qualifier which it has on other OS. I've
6124 8. Change 4.21/2 was too strict. It is only if there are two authenticators
6125 *of the same type* (client or server) with the same public name that an
6126 error should be diagnosed.
6128 9. When Exim looked up a host name for an IP address, but failed to find the
6129 original IP address when looking up the host name (a safety check), it
6130 output the message "<ip address> does not match any IP for NULL", which was
6131 confusing, to say the least. The bug was that the host name should have
6132 appeared instead of "NULL".
6134 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
6135 user that is built into the binary, and the -C or -D options is used, root
6136 privilege is dropped before the configuration file is read. In addition,
6137 logging is switched to stderr instead of the normal log files. If the
6138 configuration then re-defines the Exim user, the unprivileged environment
6139 is probably not what is expected, so Exim logs a panic warning message (but
6142 However, if deliver_drop_privilege is set, the unprivileged state may well
6143 be exactly what is intended, so the warning has been cut out in that case,
6144 and Exim is allowed to try to write to its normal log files.
6150 1. smtp_return_error_details was not giving details for temporary sender
6151 or receiver verification errors.
6153 2. Diagnose a configuration error if two authenticators have the same public
6156 3. Exim used not to create the message log file for a message until the first
6157 delivery attempt. This could be confusing when incoming messages were held
6158 for policy or load reasons. The message log file is now created at the time
6159 the message is received, and an initial "Received" line is written to it.
6161 4. The automatically generated man page for command line options had a minor
6162 bug that caused no ill effects; however, a more serious problem was that
6163 the procedure for building the man page automatically didn't always
6164 operate. Consequently, release 4.20 contains an out-of-date version. This
6165 shouldn't happen again.
6167 5. When building Exim with embedded Perl support, the script that builds the
6168 Makefile was calling 'perl' to find its compile-time parameters, ignoring
6169 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
6171 6. The freeze_tell option was not being used for messages that were frozen on
6172 arrival, either by an ACL or by local_scan().
6174 7. Added the smtp_incomplete_transaction log selector.
6176 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
6177 was accepting AUTH without a new EHLO.
6179 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
6180 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
6183 10. Logging of TCP/IP connections (when configured) now happens in the main
6184 daemon process instead of the child process, so that the TCP/IP connection
6185 count is more accurate (but it can never be perfect).
6187 11. The use of "drop" in a nested ACL was not being handled correctly in the
6188 outer ACL. Now, if condition failure induced by the nested "drop" causes
6189 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
6190 or "require"), the connection is dropped.
6192 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
6193 that yield "discard" can now be used with an "accept" or a "discard" verb,
6194 but an error is generated for any others (because I can't see a useful way
6195 to define what should happen).
6197 13. When an ACL is read dynamically from a file (or anywhere else), the lines
6198 are now processed in the same way as lines in the Exim configuration file.
6199 In particular, continuation lines are supported.
6201 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
6203 15. Added -ti meaning -t -i.
6205 16. Check for letters, digits, hyphens, and dots in the names of dnslist
6206 domains, and warn by logging if others are found.
6208 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
6209 returned from GIFCONF when Exim is trying to find the list of interfaces on
6210 a host. The code in os.c has been modified to copy each ifreq to an aligned
6211 structure in all cases.
6213 Also, in some cases, the returned ifreq's were being copied to a 'struct
6214 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
6215 means the last couple of bytes of an IPv6 address could be chopped if the
6216 ifreq contained only a normal sockaddr (14 bytes storage).
6218 18. Named domain lists were not supported in the hosts_treat_as_local option.
6219 An entry such as +xxxx was not recognized, and was treated as a literal
6222 19. Ensure that header lines added by a DATA ACL are included in the reject log
6223 if the ACL subsequently rejects the message.
6225 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
6226 MD5 (which is deprecated).
6228 21. When testing a filter file using -bf, Exim was writing a message when it
6229 took the sender from a "From " line in the message, but it was not doing so
6230 when it took $return_path from a Return-Path: header line. It now does.
6232 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
6233 with a valid header line field name (a series of printing characters
6234 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
6236 23. Changed "disc" in the source to "disk" to conform to the documentation and
6237 the book and for uniformity.
6239 24. Ignore Sendmail's -Ooption=value command line item.
6241 25. When execve() failed while trying to run a command in a pipe transport,
6242 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
6243 could be confused with a return value of 69 from the command itself. This
6244 has been changed to 127, the value the shell returns if it is asked to run
6245 a non-existent command. The wording for the related log line suggests a
6246 non-existent command as the problem.
6248 26. If received_header_text expands to an empty string, do not add a Received:
6249 header line to the message. (Well, it adds a token one on the spool, but
6250 marks it "old" so that it doesn't get used or transmitted.)
6252 27. Installed eximstats 1.28 (addition of -nt option).
6254 28. There was no check for failure on the call to getsockname() in the daemon
6255 code. This can fail if there is a shortage of resources on the system, with
6256 ENOMEM, for example. A temporary error is now given on failure.
6258 29. Contrary to the C standard, it seems that in some environments, the
6259 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
6260 program. Exim now does this explicitly; it affects the formatting of
6261 timestamps using strftime().
6263 30. If exiqsumm was given junk data, it threw up some uninitialized variable
6264 complaints. I've now initialized all the variables, to avoid this.
6266 32. Header lines added by a system filter were not being "seen" during
6267 transport-time rewrites.
6269 33. The info_callback() function passed to OpenSSL is set up with type void
6270 (*)(SSL *, int, int), as described somewhere. However, when calling the
6271 function (actually a macro) that sets it up, the type void(*)() is
6272 expected. I've put in a cast to prevent warnings from picky compilers.
6274 34. If a DNS black list lookup found a CNAME record, but there were no A
6275 records associated with the domain it pointed at, Exim crashed.
6277 35. If a DNS black list lookup returned more than one A record, Exim ignored
6278 all but the first. It now scans all returned addresses if a particular IP
6279 value is being sought. In this situation, the contents of the
6280 $dnslist_value variable are a list of all the addresses, separated by a
6283 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
6284 to accept a host name and all its aliases if the forward lookup for any of
6285 them yielded the IP address of the incoming connection. Now it accepts only
6286 those names whose forward lookup yields the correct IP address. Any other
6287 names are discarded. This closes a loophole whereby a rogue DNS
6288 administrator could create reverse DNS records to break through a
6289 wildcarded host restriction in an ACL.
6291 37. If a user filter or a system filter that ran in a subprocess used any of
6292 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
6293 the wrong values were passed to the pipe command ($thisaddress had the
6294 value of $0, $0 had the value of $1, etc). This bug was introduced by
6295 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
6297 38. Improved the line breaking for long SMTP error messages from ACLs.
6298 Previously, if there was no break point between 40 and 75 characters, Exim
6299 left the rest of the message alone. Two changes have been made: (a) I've
6300 reduced the minimum length to 35 characters; (b) if it can't find a break
6301 point between 35 and 75 characters, it looks ahead and uses the first one
6302 that it finds. This may give the occasional overlong line, but at least the
6303 remaining text gets split now.
6305 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
6306 file descriptors might be low, and that setting 1000 would always raise it.
6307 It turns out that in some environments, the limit is already over 1000 and
6308 that lowering it causes trouble. So now Exim takes care not to decrease it.
6310 40. When delivering a message, the value of $return_path is set to $sender_
6311 address at the start of routing (routers may change the value). By an
6312 oversight, this default was not being set up when an address was tested by
6313 -bt or -bv, which affected the outcome if any router or filter referred to
6316 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
6317 log only when "message" or "log_message" are set. However, if one of the
6318 conditions was an address verification, or a call to a nested ACL, the
6319 messages generated by the underlying test were being passed through. This
6320 no longer happens. The underlying message is available in $acl_verify_
6321 message for both "message" and "log_message" expansions, so it can be
6322 passed through if needed.
6324 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
6325 new expansion $bh_ to give the encoded byte string without charset
6326 translation. Translation happens only if iconv() is available; HAVE_ICONV
6327 indicates this at build time. HEADERS_CHARSET gives the charset to
6328 translate to; headers_charset can change it in the configuration, and
6329 "headers charset" can change it in an individual filter file.
6331 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
6332 that creates RFC 2047 encoded "words" labels them as that charset instead
6333 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
6334 expansion operator; (ii) when Exim creates a From: line for a local
6335 message; (iii) when a header line is rewritten to include a "phrase" part.
6337 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
6338 buggy, causing it to skip the first lines of messages whose message ID
6339 ended in 'D'. This would not have bitten before Exim release 4.14, because
6340 message IDs were unlikely to end in 'D' before then. The effect was to have
6341 incorrect size information for certain domains.
6343 45. #include "config.h" was missing at the start of the crypt16.c module. This
6344 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
6347 46. If there was a timeout during a "random" callout check, Exim treated it as
6348 a failure of the random address, and carried on sending RSET and the real
6349 address. If the delay was just some slowness somewhere, the response to the
6350 original RCPT would be taken as a response to RSET and so on, causing
6351 mayhem of various kinds.
6353 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
6354 when I implemented it. It didn't allow for the fact that some option values
6355 may legitimatetly be negative (e.g. size_addition), and it didn't even do
6356 the right test for positive values.
6358 48. Domain names in DNS records are case-independent. Exim always looks them up
6359 in lower case. Some resolvers return domain names in exactly the case they
6360 appear in the zone file, that is, they may contain uppercase letters. Not
6361 all resolvers do this - some return always lower case. Exim was treating a
6362 change of case by a resolver as a change of domain, similar to a widening
6363 of a domain abbreviation. This triggered its re-routing code and so it was
6364 trying to route what was effectively the same domain again. This normally
6365 caused routing to fail (because the router wouldn't handle the domain
6366 twice). Now Exim checks for this case specially, and just changes the
6367 casing of the domain that it ultimately uses when it transmits the message
6370 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
6373 50. If a filter generated a file delivery with a non-absolute name (possible if
6374 no home directory exists for the router), the forbid_file option was not
6377 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
6378 the existing equality matching.
6380 52. Exim was using ints instead of ino_t variables in some places where it was
6381 dealing with inode numbers.
6383 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
6384 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
6385 called TMPDIR, and if it finds it is different, it changes its value.
6387 54. The smtp_printf() function is now made available to local_scan() so
6388 additional output lines can be written before returning. There is also an
6389 smtp_fflush() function to enable the detection of a dropped connection.
6390 The variables smtp_input and smtp_batched_input are exported to
6393 55. Changed the default runtime configuration: the message "Unknown user"
6394 has been removed from the ACL, and instead placed on the localuser router,
6395 using the cannot_route_message feature. This means that any verification
6396 failures that generate their own messages won't get overridden. Similarly,
6397 the "Unrouteable address" message that was in the ACL for unverifiable
6398 relay addresses has also been removed.
6400 56. Added hosts_avoid_esmtp to the smtp transport.
6402 57. The exicyclog script was not checking for the esoteric option
6403 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
6404 will work only if exicyclog is run under the appropriate euid.
6406 58. Following a discussion on the list, the rules by which Exim recognises line
6407 endings on incoming messages have been changed. The -dropcr and drop_cr
6408 options are now no-ops, retained only for backwards compatibility. The
6409 following line terminators are recognized: LF CRLF CR. However, special
6410 processing applies to CR:
6412 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
6413 nor a local message in the state where . is a terminator.
6415 (ii) If a bare CR is encountered in a header line, an extra space is added
6416 after the line terminator so as not to end the header. The reasoning
6417 behind this is that bare CRs in header lines are most likely either
6418 to be mistakes, or people trying to play silly games.
6420 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
6421 was being incorrectly given as 18 bytes larger than it should have been.
6422 This is a VOB (very old bug).
6424 60. This may never have affected anything current, but just in case it has:
6425 When the local host is found other than at the start of a list of hosts,
6426 the local host, those with the same MX, and any that follow, are discarded.
6427 When the list in question was part of a longer list of hosts, the following
6428 hosts (not currently being processed) were also being discarded. This no
6429 longer happens. I'm not sure if this situation could ever has previously
6432 61. Added the "/MX" feature to lists of hosts in the manualroute and query
6435 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
6436 header. This is something that is recommended in a new Internet Draft, and
6437 is something that is documented as being done by Sendmail. There are two
6438 possible values. For messages generated by the autoreply transport, Exim
6441 Auto-Submitted: auto-replied
6443 whereas for all other generated messages (e.g. bounces) it adds
6445 Auto-Submitted: auto-generated
6447 63. The "personal" condition in filters now includes a test for the
6448 Auto-Submitted: header. If it contains the string "auto-" the message it
6449 not considered personal.
6451 64. Added rcpt_include_affixes as a generic transport option.
6453 65. Added queue_only_override (default true).
6455 66. Added the syslog_duplication option.
6457 67. If what should have been the first header line of a message consisted of
6458 a space followed by a colon, Exim was mis-interpreting it as a header line.
6459 It isn't of course - it is syntactically invalid and should therefore be
6460 treated as the start of the message body. The misbehaviour could have
6461 caused a number of strange effects, including loss of data in subsequent
6462 header lines, and spool format errors.
6464 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
6465 client host had authenticated. This control can now be exercised by an ACL
6466 for more flexibility.
6468 69. By default, callouts do not happen when testing with -bh. There is now a
6469 variant, -bhc, which does actually run the callout code, including
6470 consulting and updating the callout cache.
6472 70. Added support for saslauthd authentication, courtesy of Alexander
6475 71. If statvfs() failed on the spool or log directories while checking their
6476 size for availability, Exim confusingly gave the error "space shortage".
6477 Furthermore, in debugging mode it crashed with a floating point exception.
6478 These checks are done if check_{spool,log}_{space,inodes} are set, and when
6479 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
6480 serious problem, Exim now writes to the main and panic logs when this
6481 happens, with details of the failure. It then refuses to accept the
6482 incoming message, giving the message "spool directory problem" or "log
6483 directory problem" with a 421 code for SMTP messages.
6485 72. When Exim is about to re-exec itself, it ensures that the file descriptors
6486 0, 1, and 2 exist, because some OS complain for execs without them (see
6487 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
6488 descriptors. However, the code omitted to check that the open succeeded,
6489 causing mysterious errors if for some reason the permissions on /dev/null
6490 got screwed. Now Exim writes a message to the main and panic logs, and
6491 bombs out if it can't open /dev/null.
6493 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
6494 interact so that it is all more flexible. It is supposed to remain
6495 backwards compatible. Also added extra_local_interfaces.
6497 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
6498 to bomb out with an assertion failure - to the client this appears as a
6499 connection drop. This problem occurs in the part of the code that was taken
6500 from the Samba project. Fortunately, the assertion is in a very simple
6501 function, so I have fixed this by reproducing the function inline in the
6502 one place where it is called, and arranging for authentication to fail
6503 instead of killing the process with assert().
6505 75. The SPA client code was not working when the server requested OEM rather
6506 than Unicode encoding.
6508 76. Added code to make require_files with a specific uid setting more usable in
6509 the case where statting the file as root fails - usually a non-root-mounted
6510 NFS file system. When this happens and the failure is EACCES, Exim now
6511 forks a subprocess and does the per-uid checking as the relevant uid.
6513 77. Added process_log_path.
6515 78. If log_file_path was not explicitly set, a setting of check_log_space or
6516 check_log_inodes was ignored.
6518 79. If a space check for the spool or log partitions fails, the incident is now
6519 logged. Of course, in the latter case the data may get lost...
6521 80. Added the %p formatting code to string_format() so that it can be used to
6522 print addresses in debug_print(). Adjusted all the address printing in the
6523 debugging in store.c to use %p rather than %d.
6525 81. There was a concern that a line of code in smtp_in.c could overflow a
6526 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
6527 initially expressed, the concern was not well-founded, because trailing
6528 spaces are removed early. However, if the trailing spaces were followed by
6529 a NULL, they did not get removed, so the overflow was possible. Two fixes
6532 (a) I re-wrote the offending code in a cleaner fashion.
6533 (b) If an incoming SMTP command contains a NULL character, it is rejected
6536 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
6537 runs initgroups(), so that if the Exim user is in any additional groups,
6538 they will be used during message reception.
6544 The change log for 4.20 and earlier releases has been archived.
projects / exim.git / blob