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
43 JH/11 Cutthrough: avoid using the callout hints db on a verify callout when
44 a cutthrough deliver is pending, as we always want to make a connection.
45 This also avoids re-routing the message when later placing the cutthrough
46 connection after a verify cache hit.
47 Do not update it with the verify result either.
49 JH/12 Cutthrough: disable when verify option success_on_redirect is used, and
50 when routing results in more than one destination address.
52 JH/13 Cutthrough: expand transport dkim_domain option when testing for dkim
53 signing (which inhibits the cutthrough capability). Previously only
54 the presence of an option was tested; now an expansion evaluating as
55 empty is permissible (obviously it should depend only on data available
56 when the cutthrough connection is made).
58 JH/14 Fix logging of errors under PIPELINING. Previously the log line giving
59 the relevant preceding SMTP command did not note the pipelining mode.
61 JH/15 Fix counting of empty lines in $body_linecount and $message_linecount.
62 Previously they were not counted.
64 JH/16 DANE: treat a TLSA lookup response having all non-TLSA RRs, the same
65 as one having no matching records. Previously we deferred the message
66 that needed the lookup.
68 JH/17 Fakereject: previously logged as a norml message arrival "<="; now
69 distinguished as "(=".
71 JH/18 Bug 1867: make the fail_defer_domains option on a dnslookup router work
72 for missing MX records. Previously it only worked for missing A records.
74 JH/19 Bug 1850: support Radius libraries that return REJECT_RC.
76 JH/20 Bug 1872: Ensure that acl_smtp_notquit is run when the connection drops
77 after the data-go-ahead and data-ack. Patch from Jason Betts.
82 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
83 and 3.4.4 - once the server is enabled to respond to an OCSP request
84 it does even when not requested, resulting in a stapling non-aware
85 client dropping the TLS connection.
87 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
88 support variable-length bit vectors. No functional change.
90 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
91 The I= interface field on outgoing lines is now after the H= remote
92 host field, same as incoming lines. There is a separate
93 outgoing_interface log selector which allows you to disable the
96 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
97 If not running log_selector +smtp_connection the mainlog would be held
98 open indefinitely after a "too many connections" event, including to a
99 deleted file after a log rotate. Leave the per net connection logging
100 leaving it open for efficiency as that will be quickly detected by the
101 check on the next write.
103 HS/01 Bug 1671: Fix post transport crash.
104 Processing the wait-<transport> messages could crash the delivery
105 process if the message IDs didn't exist for some reason. When
106 using 'split_spool_directory=yes' the construction of the spool
107 file name failed already, exposing the same netto behaviour.
109 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
110 mime_regex ACL conditions.
112 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
113 to DSN fail messages (bounces): remote IP, remote greeting, remote response
114 to HELO, local diagnostic string.
116 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
117 log line to debug. Even when configured with a tls authenticator many
118 client connections are expected to not authenticate in this way, so
119 an authenticate fail is not an error.
121 HS/02 Add the Exim version string to the process info. This way exiwhat
122 gives some more detail about the running daemon.
124 JH/06 Bug 1395: time-limit cacheing of DNS lookups, to the TTL value. This may
125 matter for fast-change records such as DNSBLs.
127 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
128 retry record, even if constant. There may be multiple transports with
129 different interface settings and the retry behaviour needs to be kept
132 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
134 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
136 JH/10 Bug 840: fix log_defer_output option of pipe transport
138 JH/11 Bug 830: use same host for all RCPTS of a message, even under
139 hosts_randomize. This matters a lot when combined with mua_wrapper.
141 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
142 ${quote_pgsql:<string>} operator.
144 JH/13 Bug 1708: avoid misaligned access in cached lookup.
146 JH/14 Change header file name for freeradius-client. Relevant if compiling
147 with Radius support; from the Gentoo tree and checked under Fedora.
149 JH/15 Bug 1712: Introduce $prdr_requested flag variable
151 JH/16 Bug 1714: Permit an empty string as expansion result for transport
152 option transport_filter, meaning no filtering.
154 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
156 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
157 defaults to "*" (all hosts). The variable is now available when not built
158 with TLS, default unset, mainly to enable keeping the testuite sane.
159 If a server certificate is not supplied (via tls_certificate) an error is
160 logged, and clients will find TLS connections fail on startup. Presumably
161 they will retry in-clear.
162 Packagers of Exim are strongly encouraged to create a server certificate
163 at installation time.
165 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
166 with the $config_file variable.
168 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
169 in transport context, after the attempt, and per-recipient. The latter type
170 is per host attempted. The event data is the error message, and the errno
171 information encodes the lookup type (A vs. MX) used for the (first) host,
172 and the trailing two digits of the smtp 4xx reponse.
174 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
175 to write to mainlog (or rejectlog, paniclog) in the window between file
176 creation and permissions/ownership being changed. Particularly affects
177 installations where exicyclog is run as root, rather than exim user;
178 result is that the running daemon panics and dies.
180 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
182 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
183 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
184 "pri" and "weight". Note that the previous implicit priority given by the
185 list order is no longer honoured.
187 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalisation
190 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
191 by defining SUPPORT_SOCKS.
193 JH/26 Move PROXY support from Experimental to mainline, enabled for a build
194 by defining SUPPORT_PROXY. Note that the proxy_required_hosts option
195 is renamed to hosts_proxy, and the proxy_{host,target}_{address,port}.
196 variables are renamed to proxy_{local,external}_{address,port}.
198 JH/27 Move Internationalisation support from Experimental to mainline, enabled
199 for a build by defining SUPPORT_I18N
201 JH/28 Bug 1745: Fix redis lookups to handle (quoted) spaces embedded in parts
202 of the query string, and make ${quote_redis:} do that quoting.
204 JH/29 Move Events support from Experimental to mainline, enabled by default
205 and removable for a build by defining DISABLE_EVENT.
207 JH/30 Updated DANE implementation code to current from Viktor Dukhovni.
209 JH/31 Fix bug with hosts_connection_nolog and named-lists which were wrongly
210 cached by the daemon.
212 JH/32 Move Redis support from Experimental to mainline, enabled for a build
213 by defining LOOKUP_REDIS. The libhiredis library is required.
215 JH/33 Bug 1748: Permit ACL dnslists= condition in non-smtp ACLs if explicit
216 keys are given for lookup.
218 JH/34 Bug 1192: replace the embedded copy of PolarSSL RSA routines in the DKIM
219 support, by using OpenSSL or GnuTLS library ones. This means DKIM is
220 only supported when built with TLS support. The PolarSSL SHA routines
221 are still used when the TLS library is too old for convenient support.
223 JH/35 Require SINGLE_DH_USE by default in OpenSSL (main config option
224 openssl_options), for security. OpenSSL forces this from version 1.1.0
225 server-side so match that on older versions.
227 JH/36 Bug 1778: longstanding bug in memory use by the ${run } expansion: A fresh
228 allocation for $value could be released as the expansion processing
229 concluded, but leaving the global pointer active for it.
231 JH/37 Bug 1769: Permit a VRFY ACL to override the default 252 response,
232 and to use the domains and local_parts ACL conditions.
234 JH/38 Fix cutthrough bug with body lines having a single dot. The dot was
235 incorrectly not doubled on cutthrough transmission, hence seen as a
236 body-termination at the receiving system - resulting in truncated mails.
237 Commonly the sender saw a TCP-level error, and retransmitted the message
238 via the normal store-and-forward channel. This could result in duplicates
239 received - but deduplicating mailstores were liable to retain only the
240 initial truncated version.
242 JH/39 Bug 1781: Fix use of DKIM private-keys having trailing '=' in the base-64.
244 JH/40 Fix crash in queryprogram router when compiled with EXPERIMENTAL_SRS.
246 JH/41 Bug 1792: Fix selection of headers to sign for DKIM: bottom-up. While
247 we're in there, support oversigning also; bug 1309.
249 JH/42 Bug 1796: Fix error logged on a malware scanner connection failure.
251 HS/04 Add support for keep_environment and add_environment options.
253 JH/43 Tidy coding issues detected by gcc --fsanitize=undefined. Some remain;
254 either intentional arithmetic overflow during PRNG, or testing config-
257 JH/44 Bug 1800: The combination of a -bhc commandline option and cutthrough
258 delivery resulted in actual delivery. Cancel cutthrough before DATA
261 JH/45 Fix cutthrough, when connection not opened by verify and target hard-
262 rejects a recipient: pass the reject to the originator.
264 JH/46 Multiple issues raised by Coverity. Some were obvious or plausible bugs.
265 Many were false-positives and ignorable, but it's worth fixing the
268 JH/47 Fix build on HP-UX and older Solaris, which need (un)setenv now also
269 for the new environment-manipulation done at startup. Move the routines
270 from being local to tls.c to being global via the os.c file.
272 JH/48 Bug 1807: Fix ${extract } for the numeric/3-string case. While preparsing
273 an extract embedded as result-arg for a map, the first arg for extract
274 is unavailable so we cannot tell if this is a numbered or keyed
275 extraction. Accept either.
281 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
284 JH/02 The smtp transport option "multi_domain" is now expanded.
286 JH/03 The smtp transport now requests PRDR by default, if the server offers
289 JH/04 Certificate name checking on server certificates, when exim is a client,
290 is now done by default. The transport option tls_verify_cert_hostnames
291 can be used to disable this per-host. The build option
292 EXPERIMENTAL_CERTNAMES is withdrawn.
294 JH/05 The value of the tls_verify_certificates smtp transport and main options
295 default to the word "system" to access the system default CA bundle.
296 For GnuTLS, only version 3.0.20 or later.
298 JH/06 Verification of the server certificate for a TLS connection is now tried
299 (but not required) by default. The verification status is now logged by
300 default, for both outbound TLS and client-certificate supplying inbound
303 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
306 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
307 Status Notification (bounce) messages are now MIME format per RFC 3464.
308 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
309 under the control of the dsn_advertise_hosts option, and routers may
310 have a dsn_lasthop option.
312 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
313 default, modifiable by a malware= option. The list separator for
314 the options can now be changed in the usual way. Bug 68.
316 JH/10 The smtp_receive_timeout main option is now expanded before use.
318 JH/11 The incoming_interface log option now also enables logging of the
319 local interface on delivery outgoing connections.
321 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
322 if the interface and destination host and port all match.
324 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
327 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
328 Patch from Andrew Lewis.
330 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
331 now supports optional time-restrictions, weighting, and priority
332 modifiers per server. Patch originally by <rommer@active.by>.
334 JH/16 The spamd_address main option now supports a mixed list of local
335 and remote servers. Remote servers can be IPv6 addresses, and
336 specify a port-range.
338 JH/17 Bug 68: The spamd_address main option now supports an optional
339 timeout value per server.
341 JH/18 Bug 1581: Router and transport options headers_add/remove can
342 now have the list separator specified.
344 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
347 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
350 JH/21 Support for the A6 type of dns record is withdrawn.
352 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
353 rather than the verbs used.
355 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
356 from 255 to 1024 chars.
358 JH/24 Verification callouts now attempt to use TLS by default.
360 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
361 are generic router options now. The defaults didn't change.
363 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
364 Original patch from Alexander Shikoff, worked over by JH.
366 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
367 files if ZCAT_COMMAND is not executable.
369 JH/26 Bug 1539: Add timout/retry options on dnsdb lookups.
371 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
373 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
374 Normally benign, it bites when the pair was led to by a CNAME;
375 modern usage is to not canoicalize the domain to a CNAME target
376 (and we were inconsistent anyway for A-only vs AAAA+A).
378 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
380 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
381 when evaluating $sender_host_dnssec.
383 JH/31 Check the HELO verification lookup for DNSSEC, adding new
384 $sender_helo_dnssec variable.
386 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
388 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
390 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
392 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
393 documented as working, but never had. Support all but $spam_report.
395 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
396 added for tls authenticator.
398 HS/03 Add perl_taintmode main config option
403 TL/01 When running the test suite, the README says that variables such as
404 no_msglog_check are global and can be placed anywhere in a specific
405 test's script, however it was observed that placement needed to be near
406 the beginning for it to behave that way. Changed the runtest perl
407 script to read through the entire script once to detect and set these
408 variables, reset to the beginning of the script, and then run through
409 the script parsing/test process like normal.
411 TL/02 The BSD's have an arc4random API. One of the functions to induce
412 adding randomness was arc4random_stir(), but it has been removed in
413 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
414 function when detected.
416 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
417 cause callback expansion.
419 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
420 syntax errors in an expansion can be treated as a string instead of
421 logging or causing an error, due to the internal use of bool_lax
422 instead of bool when processing it.
424 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
425 server certificates when making smtp deliveries.
427 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
429 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
431 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
433 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
434 Merged patch from Sebastian Wiedenroth.
436 JH/05 Fix results-pipe from transport process. Several recipients, combined
437 with certificate use, exposed issues where response data items split
438 over buffer boundaries were not parsed properly. This eventually
439 resulted in duplicates being sent. This issue only became common enough
440 to notice due to the introduction of conection certificate information,
441 the item size being so much larger. Found and fixed by Wolfgang Breyha.
443 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
444 size buffer was used, resulting in syntax errors when an expansion
447 JH/07 Add support for directories of certificates when compiled with a GnuTLS
448 version 3.3.6 or later.
450 JH/08 Rename the TPDA expermimental facility to Event Actions. The #ifdef
451 is EXPERIMENTAL_EVENT, the main-configuration and transport options
452 both become "event_action", the variables become $event_name, $event_data
453 and $event_defer_errno. There is a new variable $verify_mode, usable in
454 routers, transports and related events. The tls:cert event is now also
455 raised for inbound connections, if the main configuration event_action
458 TL/06 In test suite, disable OCSP for old versions of openssl which contained
459 early OCSP support, but no stapling (appears to be less than 1.0.0).
461 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
462 server certificate names available under the smtp transport option
463 "tls_verify_cert_hostname" now do not permit multi-component wildcard
466 JH/10 Time-related extraction expansions from certificates now use the main
467 option "timezone" setting for output formatting, and are consistent
468 between OpenSSL and GnuTLS compilations. Bug 1541.
470 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
471 encoded parameter in the incoming message. Bug 1558.
473 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
474 include certificate info, eximon was claiming there were spoolfile
477 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
479 JH/14 Log delivery-related information more consistently, using the sequence
480 "H=<name> [<ip>]" wherever possible.
482 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
483 are problematic for Debian distribution, omit them from the release
486 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
488 JH/16 Fix string representation of time values on 64bit time_t anchitectures.
491 JH/17 Fix a null-indirection in certextract expansions when a nondefault
492 output list separator was used.
497 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
498 checkers that were complaining about end of non-void function with no
501 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
502 This was a regression intruduced in 4.83 by another bugfix.
504 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
506 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
507 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
513 TF/01 Correctly close the server side of TLS when forking for delivery.
515 When a message was received over SMTP with TLS, Exim failed to clear up
516 the incoming connection properly after forking off the child process to
517 deliver the message. In some situations the subsequent outgoing
518 delivery connection happened to have the same fd number as the incoming
519 connection previously had. Exim would try to use TLS and fail, logging
520 a "Bad file descriptor" error.
522 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
523 utilities have not been installed.
525 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
526 temporary space as the ACL may create new global variables.
528 TL/01 LDAP support uses per connection or global context settings, depending
529 upon the detected version of the libraries at build time.
531 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
532 to extract and use the src ip:port in logging and expansions as if it
533 were a direct connection from the outside internet. PPv2 support was
534 updated based on HAProxy spec change in May 2014.
536 JH/02 Add ${listextract {number}{list}{success}{fail}}.
538 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
539 Properly escape header and check for NULL return.
541 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
544 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
546 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
547 characters in header names, implemented as a verify condition.
548 Contributed by Michael Fischer v. Mollard.
550 TL/05 Rename SPF condition results err_perm and err_temp to standardized
551 results permerror and temperror. Previous values are deprecated but
552 still accepted. In a future release, err_perm and err_temp will be
553 completely removed, which will be a backward incompatibility if the
554 ACL tests for either of these two old results. Patch contributed by
555 user bes-internal on the mailing list.
557 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
559 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
560 selectors, in both main and reject logs.
562 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
565 JH/07 Add malware type "sock" for talking to simple daemon.
567 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
569 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
570 routers/transports under cutthrough routing.
572 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
573 numbers. Touch up "bool" conditional to keep the same definition.
575 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
577 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
579 JH/12 Expand items in router/transport headers_add or headers_remove lists
580 individually rather than the list as a whole. Bug 1452.
582 Required for reasonable handling of multiple headers_ options when
583 they may be empty; requires that headers_remove items with embedded
584 colons must have them doubled (or the list-separator changed).
586 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
587 view the policy declared in the DMARC record. Currently, $dmarc_status
588 is a combined value of both the record presence and the result of the
591 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
593 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
594 dnslookup router and the smtp transport (applying to the forward
597 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
598 of ldap servers used for a specific lookup. Patch provided by Heiko
601 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
602 New variable $lookup_dnssec_authenticated for observability.
604 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
605 Patch submitted by Lars Timman.
607 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
609 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
610 Requires trusted mode and valid format message id, aborts otherwise.
611 Patch contributed by Heiko Schlichting.
613 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
614 certextract with support for various fields. Bug 1358.
616 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
617 is requested by default, modifiable by smtp transport option
620 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
621 operate on certificate variables to give certificate fingerprints
622 Also new ${sha256:cert_variable}.
624 JH/23 The PRDR feature is moved from being Experimental into the mainline.
626 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
627 Christian Aistleitner.
629 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
631 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
632 file. Patch from Wolfgang Breyha.
634 JH/25 Expand the coverage of the delivery $host and $host_address to
635 client authenticators run in verify callout. Bug 1476.
637 JH/26 Port service names are now accepted for tls_on_connect_ports, to
638 align with daemon_smtp_ports. Bug 72.
640 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
641 support and error reporting did not work properly.
643 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
644 and is readable. Patch from Andrew Colin Kissa.
646 TL/14 Enhance documentation of ${run expansion and how it parses the
647 commandline after expansion, particularly in the case when an
648 unquoted variable expansion results in an empty value.
650 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
652 PP/02 Fix internal collision of T_APL on systems which support RFC3123
653 by renaming away from it. Addresses GH issue 15, reported by
656 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
658 TL/15 SECURITY: prevent double expansion in math comparison functions
659 (can expand unsanitized data). Not remotely exploitable.
666 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
668 PP/02 Make -n do something, by making it not do something.
669 When combined with -bP, the name of an option is not output.
671 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
674 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
675 $sender_host_name and config options to manage this, and basic check
678 PP/05 DSCP support for outbound connections and control modifier for inbound.
680 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
681 (Only plugin which currently uses this is kerberos4, which nobody should
682 be using, but we should make it available and other future plugins might
683 conceivably use it, even though it would break NAT; stuff *should* be
684 using channel bindings instead).
686 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
687 name; added for Sendmail compatibility; requires admin caller.
688 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
689 just ignore it); requires trusted caller.
690 Also parse but ignore: -Ac -Am -X<logfile>
693 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
695 TL/02 Add +smtp_confirmation as a default logging option.
697 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
698 Patch by Magnus Holmgren from 2007-02-20.
700 TL/04 Bugzilla 1281 - Spec typo.
701 Bugzilla 1283 - Spec typo.
702 Bugzilla 1290 - Spec grammar fixes.
704 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
706 TL/06 Add Experimental DMARC support using libopendmarc libraries.
708 TL/07 Fix an out of order global option causing a segfault. Reported to dev
709 mailing list by by Dmitry Isaikin.
711 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
713 JH/02 Support "G" suffix to numbers in ${if comparisons.
715 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
717 NM/01 Bugzilla 1197 - Spec typo
718 Bugzilla 1196 - Spec examples corrections
720 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
722 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
723 gnutls_enable_pkcs11, but renamed to more accurately indicate its
726 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
727 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
729 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
730 "acl {{name}{arg}...}", and optional args on acl condition
733 JH/05 Permit multiple router/transport headers_add/remove lines.
735 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
737 JH/07 Avoid using a waiting database for a single-message-only transport.
738 Performance patch from Paul Fisher. Bugzilla 1262.
740 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
743 JH/09 Add $headers_added variable, with content from use of ACL modifier
744 add_header (but not yet added to the message). Bugzilla 199.
746 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
747 Pulled from Bugzilla 817 by Wolfgang Breyha.
749 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
751 (nb: this is the same fix as in Exim 4.80.1)
753 JH/11 Add A= logging on delivery lines, and a client_set_id option on
756 JH/12 Add optional authenticated_sender logging to A= and a log_selector
759 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
761 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
762 advertise SMTP AUTH mechanism to us, instead of a generic
763 protocol violation error. Also, make Exim more robust to bad
764 data from the Dovecot auth socket.
766 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
768 When a queue runner is handling a message, Exim first routes the
769 recipient addresses, during which it prunes them based on the retry
770 hints database. After that it attempts to deliver the message to
771 any remaining recipients. It then updates the hints database using
774 So if a recipient address works intermittently, it can get repeatedly
775 deferred at routing time. The retry hints record remains fresh so the
776 address never reaches the final cutoff time.
778 This is a fairly common occurrence when a user is bumping up against
779 their storage quota. Exim had some logic in its local delivery code
780 to deal with this. However it did not apply to per-recipient defers
781 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
783 This change adds a proper retry rule check during routing so that the
784 final cutoff time is checked against the message's age. We only do
785 this check if there is an address retry record and there is not a
786 domain retry record; this implies that previous attempts to handle
787 the address had the retry_use_local_parts option turned on. We use
788 this as an approximation for the destination being like a local
789 delivery, as in LMTP.
791 I suspect this new check makes the old local delivery cutoff check
792 redundant, but I have not verified this so I left the code in place.
794 TF/02 Correct gecos expansion when From: is a prefix of the username.
796 Test 0254 submits a message to Exim with the header
800 When I ran the test suite under the user fanf2, Exim expanded
801 the header to contain my full name, whereas it should have added
802 a Resent-Sender: header. It erroneously treats any prefix of the
803 username as equal to the username.
805 This change corrects that bug.
807 GF/01 DCC debug and logging tidyup
808 Error conditions log to paniclog rather than rejectlog.
809 Debug lines prefixed by "DCC: " to remove any ambiguity.
811 TF/03 Avoid unnecessary rebuilds of lookup-related code.
813 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
814 Bug spotted by Jeremy Harris; was flawed since initial commit.
815 Would have resulted in OCSP responses post-SNI triggering an Exim
816 NULL dereference and crash.
818 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
820 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
821 Bug detection, analysis and fix by Samuel Thibault.
822 Bugzilla 1331, Debian bug #698092.
824 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
826 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
827 Server implementation by Todd Lyons, client by JH.
828 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
829 config variable "prdr_enable" controls whether the server
830 advertises the facility. If the client requests PRDR a new
831 acl_data_smtp_prdr ACL is called once for each recipient, after
832 the body content is received and before the acl_smtp_data ACL.
833 The client is controlled by bolth of: a hosts_try_prdr option
834 on the smtp transport, and the server advertisement.
835 Default client logging of deliveries and rejections involving
836 PRDR are flagged with the string "PRDR".
838 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
839 fclose(). Diagnosis by Todd Lyons.
841 PP/17 Update configure.default to handle IPv6 localhost better.
842 Patch by Alain Williams (plus minor tweaks).
845 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
846 This is now consistent with GnuTLS, and is now documented: the
847 previous undocumented portable approach to treating the option as
848 unset was to force an expansion failure. That still works, and
849 an empty string is now equivalent.
851 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
852 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
853 not performing validation itself.
855 PP/20 Added force_command boolean option to pipe transport.
856 Patch from Nick Koston, of cPanel Inc.
858 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
861 TF/04 Added udpsend ACL modifer and hexquote expansion operator
863 PP/21 Fix eximon continuous updating with timestamped log-files.
864 Broken in a format-string cleanup in 4.80, missed when I repaired the
865 other false fix of the same issue.
866 Report and fix from Heiko Schlichting.
869 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
870 Report from Prashanth Katuri.
872 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
873 It's SecureTransport, so affects any MacOS clients which use the
874 system-integrated TLS libraries, including email clients.
876 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
877 using a MIME ACL for non-SMTP local injection.
878 Report and assistance in diagnosis by Warren Baker.
880 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
882 JH/16 Fix comparisons for 64b. Bugzilla 1385.
884 TL/09 Add expansion variable $authenticated_fail_id to keep track of
885 last id that failed so it may be referenced in subsequent ACL's.
887 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
890 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
891 ldap library initialization, allowing self-signed CA's to be
892 used. Also properly sets require_cert option later in code by
893 using NULL (global ldap config) instead of ldap handle (per
894 session). Bug diagnosis and testing by alxgomz.
896 TL/12 Enhanced documentation in the ratelimit.pl script provided in
897 the src/util/ subdirectory.
899 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
900 renamed to Transport Post Delivery Action by Jeremy Harris, as
903 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
904 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
905 redis_servers = needs to be configured which will be used by the redis
906 lookup. Patch from Warren Baker, of The Packet Hub.
908 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
910 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
911 hostname or reverse DNS when processing a host list. Used suggestions
912 from multiple comments on this bug.
914 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
916 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
917 Missed a few lines, added it to make the runtest require no keyboard
920 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
921 contains upper case chars. Make router use caseful_local_part.
923 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
924 support when GnuTLS has been built with p11-kit.
930 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
932 This, or similar/improved, will also be change PP/11 of 4.82.
938 PP/01 Handle short writes when writing local log-files.
939 In practice, only affects FreeBSD (8 onwards).
940 Bugzilla 1053, with thanks to Dmitry Isaikin.
942 NM/01 Bugzilla 949 - Documentation tweak
944 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
947 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
949 PP/02 Implemented gsasl authenticator.
951 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
953 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
954 `pkg-config foo` for cflags/libs.
956 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
957 with rest of GSASL and with heimdal_gssapi.
959 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
960 `pkg-config foo` for cflags/libs for the TLS implementation.
962 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
963 properties get this fed in as external SSF. A number of robustness
964 and debugging improvements to the cyrus_sasl authenticator.
966 PP/08 cyrus_sasl server now expands the server_realm option.
968 PP/09 Bugzilla 1214 - Log authentication information in reject log.
969 Patch by Jeremy Harris.
971 PP/10 Added dbmjz lookup type.
973 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
975 PP/12 MAIL args handles TAB as well as SP, for better interop with
976 non-compliant senders.
977 Analysis and variant patch by Todd Lyons.
979 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
980 Bug report from Lars Müller <lars@samba.org> (via SUSE),
981 Patch from Dirk Mueller <dmueller@suse.com>
983 PP/13 tls_peerdn now print-escaped for spool files.
984 Observed some $tls_peerdn in wild which contained \n, which resulted
985 in spool file corruption.
987 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
988 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
989 or write after TLS renegotiation, which otherwise led to messages
992 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
993 as a tracking header (ie: a signed header comes before the signature).
994 Patch from Wolfgang Breyha.
996 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
997 comma-sep list; embedded commas doubled.
999 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
1001 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
1003 Report and patch from Dmitry Banschikov.
1005 PP/16 Removed "dont_insert_empty_fragments" fron "openssl_options".
1006 Removed SSL_clear() after SSL_new() which led to protocol negotiation
1007 failures. We appear to now support TLS1.1+ with Exim.
1009 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
1010 lets Exim select keys and certificates based upon TLS SNI from client.
1011 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
1012 before an outbound SMTP session. New log_selector, +tls_sni.
1014 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
1015 NULL dereference. Report and patch from Alun Jones.
1017 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
1018 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
1019 Not seeing resolver debug output on NetBSD, but suspect this is a
1020 resolver implementation change.
1022 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
1023 Left warnings. Added "eximon gdb" invocation mode.
1025 PP/21 Defaulting "accept_8bitmime" to true, not false.
1027 PP/22 Added -bw for inetd wait mode support.
1029 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
1030 locate the relevant includes and libraries. Made this the default.
1032 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
1033 Bugzilla 1246, report and most of solution from Tomasz Kusy.
1035 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
1036 This may cause build issues on older platforms.
1038 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
1039 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
1040 gnutls_require_mac & gnutls_require_protocols (no longer supported).
1041 Added SNI support via GnuTLS too.
1042 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
1044 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
1046 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
1047 Applied second patch from Janne, implementing suggestion to default
1048 multiple-strings-in-record handling to match SPF spec.
1050 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
1052 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
1053 read-only, out of scope).
1054 Patch from Wolfgang Breyha, report from Stuart Northfield.
1056 PP/29 Fix three issues highlighted by clang analyser static analysis.
1057 Only crash-plausible issue would require the Cambridge-specific
1058 iplookup router and a misconfiguration.
1059 Report from Marcin Mirosław.
1061 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
1063 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
1064 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
1065 As part of this, removing so much warning spew let me fix some minor
1066 real issues in debug logging.
1068 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
1069 assignment on my part. Fixed.
1071 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
1072 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
1073 Janne Snabb (who went above and beyond: thank you).
1075 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
1076 string otherwise requires a connection and a bunch more work and it's
1077 relatively easy to get wrong. Should also expose TLS library linkage
1080 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
1081 64-bit ${eval} (JH/03).
1083 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
1084 GNU libc to support some of the 64-bit stuff, should not lead to
1085 conflicts. Defined before os.h is pulled in, so if a given platform
1086 needs to override this, it can.
1088 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
1089 protection layer was required, which is not implemented.
1090 Bugzilla 1254, patch from Wolfgang Breyha.
1092 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
1093 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
1094 tls_dhparam take prime identifiers. Also unbreak combination of
1095 OpenSSL+DH_params+TLSSNI.
1097 PP/39 Disable SSLv2 by default in OpenSSL support.
1103 PP/01 Solaris build fix for Oracle's LDAP libraries.
1104 Bugzilla 1109, patch from Stephen Usher.
1106 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
1108 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
1111 TF/02 Fix a couple more cases where we did not log the error message
1112 when unlink() failed. See also change 4.74-TF/03.
1114 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
1115 lock up or crash if it happened to be inside a call to libc when it
1116 got a SIGUSR1 from exiwhat.
1118 The SIGUSR1 handler appends the current process status to the process
1119 log which is later printed by exiwhat. It used to use the general
1120 purpose logging code to do this, but several functions it calls are
1121 not safe for signals.
1123 The new output code in the SIGUSR1 handler is specific to the process
1124 log, and simple enough that it's easy to inspect for signal safety.
1125 Removing some special cases also simplifies the general logging code.
1126 Removing the spurious timestamps from the process log simplifies
1129 TF/04 Improved ratelimit ACL condition.
1131 The /noupdate option has been deprecated in favour of /readonly which
1132 has clearer semantics. The /leaky, /strict, and /readonly update modes
1133 are mutually exclusive. The update mode is no longer included in the
1134 database key; it just determines when the database is updated. (This
1135 means that when you upgrde Exim will forget old rate measurements.)
1137 Exim now checks that the per_* options are used with an update mode that
1138 makes sense for the current ACL. For example, when Exim is processing a
1139 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
1140 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
1141 must specify per_mail/readonly. If you omit the update mode it defaults to
1142 /leaky where that makes sense (as before) or /readonly where required.
1144 The /noupdate option is now undocumented but still supported for
1145 backwards compatibility. It is equivalent to /readonly except that in
1146 ACLs where /readonly is required you may specify /leaky/noupdate or
1147 /strict/noupdate which are treated the same as /readonly.
1149 A useful new feature is the /count= option. This is a generalization
1150 of the per_byte option, so that you can measure the throughput of other
1151 aggregate values. For example, the per_byte option is now equivalent
1152 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
1154 The per_rcpt option has been generalized using the /count= mechanism
1155 (though it's more complicated than the per_byte equivalence). When it is
1156 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
1157 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
1158 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
1159 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
1160 Note that using per_rcpt with a non-readonly update mode in more than
1161 one ACL will cause the recipients to be double-counted. (The per_mail
1162 and per_byte options don't have this problem.)
1164 The handling of very low rates has changed slightly. If the computed rate
1165 is less than the event's count (usually one) then this event is the first
1166 after a long gap. In this case the rate is set to the same as this event's
1167 count, so that the first message of a spam run is counted properly.
1169 The major new feature is a mechanism for counting the rate of unique
1170 events. The new per_addr option counts the number of different
1171 recipients that someone has sent messages to in the last time period. It
1172 behaves like per_rcpt if all the recipient addresses are different, but
1173 duplicate recipient addresses do not increase the measured rate. Like
1174 the /count= option this is a general mechanism, so the per_addr option
1175 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
1176 example, measure the rate that a client uses different sender addresses
1177 with the options per_mail/unique=$sender_address. There are further
1178 details in the main documentation.
1180 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
1182 TF/06 Removed a few PCRE remnants.
1184 TF/07 Automatically extract Exim's version number from tags in the git
1185 repository when doing development or release builds.
1187 PP/02 Raise smtp_cmd_buffer_size to 16kB.
1188 Bugzilla 879. Patch from Paul Fisher.
1190 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
1191 Heavily based on revision 40f9a89a from Simon Arlott's tree.
1194 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
1196 PP/05 Variable $av_failed, true if the AV scanner deferred.
1197 Bugzilla 1078. Patch from John Horne.
1199 PP/06 Stop make process more reliably on build failure.
1200 Bugzilla 1087. Patch from Heiko Schlittermann.
1202 PP/07 Make maildir_use_size_file an _expandable_ boolean.
1203 Bugzilla 1089. Patch from Heiko Schlittermann.
1205 PP/08 Handle ${run} returning more data than OS pipe buffer size.
1206 Bugzilla 1131. Patch from Holger Weiß.
1208 PP/09 Handle IPv6 addresses with SPF.
1209 Bugzilla 860. Patch from Wolfgang Breyha.
1211 PP/10 GnuTLS: support TLS 1.2 & 1.1.
1213 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
1216 PP/11 match_* no longer expand right-hand-side by default.
1217 New compile-time build option, EXPAND_LISTMATCH_RHS.
1218 New expansion conditions, "inlist", "inlisti".
1220 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
1222 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
1224 PP/14 fix log_write() format string regression from TF/03.
1225 Bugzilla 1152. Patch from Dmitry Isaikin.
1231 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
1233 PP/02 Harmonised TLS library version reporting; only show if debugging.
1234 Layout now matches that introduced for other libraries in 4.74 PP/03.
1236 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
1238 PP/04 New "dns_use_edns0" global option.
1240 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
1243 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
1244 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
1246 TK/01 Updated PolarSSL code to 0.14.2.
1247 Bugzilla 1097. Patch from Andreas Metzler.
1249 PP/07 Catch divide-by-zero in ${eval:...}.
1250 Fixes bugzilla 1102.
1252 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
1255 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
1256 format-string attack -- SECURITY: remote arbitrary code execution.
1258 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
1259 time unintentionally subject to list matching rules, letting the header
1260 cause arbitrary Exim lookups (of items which can occur in lists, *not*
1261 arbitrary string expansion). This allowed for information disclosure.
1263 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
1264 INT_MIN/-1 -- value coerced to INT_MAX.
1270 NM/01 Workround for PCRE version dependency in version reporting
1273 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
1274 This fixes portability to compilers other than gcc, notably
1275 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
1277 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
1278 makefiles for portability to HP-UX and POSIX correctness.
1280 PP/01 Permit LOOKUP_foo enabling on the make command-line.
1281 Also via indented variable definition in the Makefile.
1282 (Debugging by Oliver Heesakkers).
1284 PP/02 Restore caching of spamd results with expanded spamd_address.
1285 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
1287 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
1288 Improves build reliability. Fix from: Frank Elsner
1290 NM/02 Fix wide character breakage in the rfc2047 coding
1291 Fixes bug 1064. Patch from Andrey N. Oktyabrski
1293 NM/03 Allow underscore in dnslist lookups
1294 Fixes bug 1026. Patch from Graeme Fowler
1296 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
1297 Code patches from Adam Ciarcinski of NetBSD.
1299 NM/04 Fixed exiqgrep to cope with mailq missing size issue
1302 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
1303 is logged, to avoid truncation. Patch from John Horne.
1305 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
1306 Patch from Jakob Hirsch.
1308 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
1309 SQL string expansion failure details.
1310 Patch from Andrey Oktyabrski.
1312 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
1313 Patch from Simon Arlott.
1315 PP/09 New lookups functionality failed to compile on old gcc which rejects
1316 extern declarations in function scope.
1317 Patch from Oliver Fleischmann
1319 PP/10 Use sig_atomic_t for flags set from signal handlers.
1320 Check getgroups() return and improve debugging.
1321 Fixed developed for diagnosis in bug 927 (which turned out to be
1324 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
1325 Patch from Mark Zealey.
1327 PP/12 Bugzilla 1056: Improved spamd server selection.
1328 Patch from Mark Zealey.
1330 PP/13 Bugzilla 1086: Deal with maildir quota file races.
1331 Based on patch from Heiko Schlittermann.
1333 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
1334 Patch from Uwe Doering, sign-off by Michael Haardt.
1336 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
1337 variable declaration deep within a block. Bug and patch from
1340 PP/15 lookups-Makefile IRIX compatibilty coercion.
1342 PP/16 Make DISABLE_DKIM build knob functional.
1344 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
1345 Patch by Simon Arlott
1347 TF/03 Fix valgrind.h portability to C89 compilers that do not support
1348 variable argument macros. Our copy now differs from upstream.
1354 TF/01 Failure to get a lock on a hints database can have serious
1355 consequences so log it to the panic log.
1357 TF/02 Log LMTP confirmation messages in the same way as SMTP,
1358 controlled using the smtp_confirmation log selector.
1360 TF/03 Include the error message when we fail to unlink a spool file.
1362 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
1363 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
1364 for maintaining out-of-tree patches for some time.
1366 PP/01 Bugzilla 139: Documentation and portability issues.
1367 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
1368 Handle per-OS dynamic-module compilation flags.
1370 PP/02 Let /dev/null have normal permissions.
1371 The 4.73 fixes were a little too stringent and complained about the
1372 permissions on /dev/null. Exempt it from some checks.
1373 Reported by Andreas M. Kirchwitz.
1375 PP/03 Report version information for many libraries, including
1376 Exim version information for dynamically loaded libraries. Created
1377 version.h, now support a version extension string for distributors
1378 who patch heavily. Dynamic module ABI change.
1380 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
1381 privilege escalation vulnerability whereby the Exim run-time user
1382 can cause root to append content of the attacker's choosing to
1385 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
1388 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
1389 If dropping privileges for untrusted macros, we disabled normal logging
1390 on the basis that it would fail; for the Exim run-time user, this is not
1391 the case, and it resulted in successful deliveries going unlogged.
1392 Fixed. Reported by Andreas Metzler.
1398 PP/01 Date: & Message-Id: revert to normally being appended to a message,
1399 only prepend for the Resent-* case. Fixes regression introduced in
1400 Exim 4.70 by NM/22 for Bugzilla 607.
1402 PP/02 Include check_rfc2047_length in configure.default because we're seeing
1403 increasing numbers of administrators be bitten by this.
1405 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
1407 PP/03 Bugzilla 994: added openssl_options main configuration option.
1409 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
1411 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
1413 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
1415 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
1416 without a peer certificate, leading to a segfault because of an
1417 assumption that peers always have certificates. Be a little more
1418 paranoid. Problem reported by Martin Tscholak.
1420 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
1421 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
1422 NB: ClamAV planning to remove STREAM in "middle of 2010".
1423 CL also introduces -bmalware, various -d+acl logging additions and
1424 more caution in buffer sizes.
1426 PP/09 Implemented reverse_ip expansion operator.
1428 PP/10 Bugzilla 937: provide a "debug" ACL control.
1430 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
1432 PP/12 Bugzilla 973: Implement --version.
1434 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
1436 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
1438 PP/15 Bugzilla 816: support multiple condition rules on Routers.
1440 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
1441 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
1442 ignore trailing whitespace.
1444 JJ/02 prevent non-panic DKIM error from being sent to paniclog
1446 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
1449 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
1450 Notification from Dr Andrew Aitchison.
1452 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
1453 ExtendedDetectionInfo response format.
1454 Notification from John Horne.
1456 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
1459 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
1460 XSL and documented dependency on system catalogs, with examples of how
1463 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
1466 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
1467 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
1468 configuration file which is writeable by the Exim user or group.
1470 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
1471 of configuration files to cover files specified with the -C option if
1472 they are going to be used with root privileges, not just the default
1475 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
1476 option (effectively making it always true).
1478 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
1479 files to be used while preserving root privileges.
1481 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
1482 that rogue child processes cannot use them.
1484 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
1485 run-time user, instead of root.
1487 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
1488 Exim run-time user without dropping privileges.
1490 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
1491 result string, instead of calling string_vformat() twice with the same
1494 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
1495 for other users. Others should always drop root privileges if they use
1496 -C on the command line, even for a whitelisted configure file.
1498 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
1500 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
1506 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
1507 $data_path, and $header_path variables; fixed documentation bugs and
1510 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
1511 exipick to access non-standard spools, including the "frozen" queue
1514 NM/01 Bugzilla 965: Support mysql stored procedures.
1515 Patch from Alain Williams
1517 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
1519 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
1520 Patch from Andreas Metzler
1522 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
1523 Patch from Kirill Miazine
1525 NM/05 Bugzilla 671: Added umask to procmail example.
1527 JJ/03 installed exipick 20100323.0, fixing doc bug
1529 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
1530 directory. Notification and patch from Dan Rosenberg.
1532 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
1534 TK/02 Improve log output when DKIM signing operation fails.
1536 MH/01 Treat the transport option dkim_domain as a colon separated
1537 list, not as a single string, and sign the message with each element,
1538 omitting multiple occurences of the same signer.
1540 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
1541 Bugzilla 985, 986. Patch by Simon Arlott
1543 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
1544 Patch by Simon Arlott
1546 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
1547 MBX locking. Notification from Dan Rosenberg.
1553 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
1555 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
1557 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
1559 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
1561 NM/04 Bugzilla 915: Flag broken perl installation during build.
1567 TK/01 Added patch by Johannes Berg that expands the main option
1568 "spamd_address" if it starts with a dollar sign.
1570 TK/02 Write list of recipients to X-Envelope-Sender header when building
1571 the mbox-format spool file for content scanning (suggested by Jakob
1574 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
1575 (http://www.dcc-servers.net/) support via dccifd. Activated by
1576 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
1578 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
1579 by Mark Daniel Reidel <mr@df.eu>.
1581 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
1582 When building exim an external PCRE library is now needed -
1583 PCRE is a system library on the majority of modern systems.
1584 See entry on PCRE_LIBS in EDITME file.
1586 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
1587 conversation. Added nologin parameter to request.
1588 Patch contributed by Kirill Miazine.
1590 TF/01 Do not log submission mode rewrites if they do not change the address.
1592 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
1594 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
1595 log files in place. Contributed by Roberto Lima.
1597 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
1599 TF/03 Bugzilla 615: When checking the local_parts router precondition
1600 after a local_part_suffix or local_part_prefix option, Exim now
1601 does not use the address's named list lookup cache, since this
1602 contains cached lookups for the whole local part.
1604 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
1605 Robert Millan. Documentation is in experimental-spec.txt.
1607 TF/04 Bugzilla 668: Fix parallel build (make -j).
1609 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
1611 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
1612 Patch provided by Jan Srzednicki.
1614 TF/05 Leading white space used to be stripped from $spam_report which
1615 wrecked the formatting. Now it is preserved.
1617 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
1618 that they are available at delivery time.
1620 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
1622 TF/08 TLS error reporting now respects the incoming_interface and
1623 incoming_port log selectors.
1625 TF/09 Produce a more useful error message if an SMTP transport's hosts
1626 setting expands to an empty string.
1628 NM/06 Bugzilla 744: EXPN did not work under TLS.
1629 Patch provided by Phil Pennock.
1631 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
1632 Patch provided by Richard Godbee.
1634 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
1635 acl_smtp_notquit, added index entry.
1637 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
1638 Patch provided by Eugene Bujak.
1640 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
1641 accept(). Patch provided by Maxim Dounin.
1643 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
1644 Patch provided by Phil Pennock.
1646 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
1648 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
1649 Patch provided by Brad "anomie" Jorsch.
1651 NM/14 Bugzilla 622: Added timeout setting to transport filter.
1652 Patch provided by Dean Brooks.
1654 TK/05 Add native DKIM support (does not depend on external libraries).
1656 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
1657 Patch provided by Graeme Fowler.
1659 NM/16 Bugzilla 851: Documentation example syntax fix.
1661 NM/17 Changed NOTICE file to remove references to embedded PCRE.
1663 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
1666 NM/19 Bugzilla 745: TLS version reporting.
1667 Patch provided by Phil Pennock.
1669 NM/20 Bugzilla 167: bool: condition support.
1670 Patch provided by Phil Pennock.
1672 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
1673 clients. Patch provided by Phil Pennock.
1675 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
1676 Patch provided by Brad "anomie" Jorsch.
1678 NM/23 Bugzilla 687: Fix misparses in eximstats.
1679 Patch provided by Heiko Schlittermann.
1681 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
1682 Patch provided by Heiko Schlittermann.
1684 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
1685 plus update to original patch.
1687 NM/26 Bugzilla 799: Documentation correction for ratelimit.
1689 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
1690 Patch provided by David Brownlee.
1692 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
1694 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
1696 NM/30 Bugzilla 888: TLS documentation bugfixes.
1698 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
1700 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
1701 Unlike the original bugzilla I have changed all shell scripts in src tree.
1703 NM/33 Bugzilla 898: Transport filter timeout fix.
1704 Patch by Todd Rinaldo.
1706 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
1707 Patch by Serge Demonchaux.
1709 NM/35 Bugzilla 39: Base64 decode bug fixes.
1710 Patch by Jakob Hirsch.
1712 NM/36 Bugzilla 909: Correct connect() call in dcc code.
1714 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
1716 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
1718 NM/39 Bugzilla 911: Fixed MakeLinks build script.
1724 TK/01 Add preliminary DKIM support. Currently requires a forked version of
1725 ALT-N's libdkim that I have put here:
1726 http://duncanthrax.net/exim-experimental/
1728 Note to Michael Haardt: I had to rename some vars in sieve.c. They
1729 were called 'true' and it seems that C99 defines that as a reserved
1730 keyword to be used with 'bool' variable types. That means you could
1731 not include C99-style headers which use bools without triggering
1732 build errors in sieve.c.
1734 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
1735 as mailq or other aliases. Changed the --help handling significantly
1736 to do whats expected. exim_usage() emits usage/help information.
1738 SC/01 Added the -bylocaldomain option to eximstats.
1740 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
1742 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
1744 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
1750 PH/01 Another patch from the Sieve maintainer.
1752 PH/02 When an IPv6 address is converted to a string for single-key lookup
1753 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
1754 dots are used instead of colons so that keys in lsearch files need not
1755 contain colons. This was done some time before quoting was made available
1756 in lsearch files. However, iplsearch files do require colons in IPv6 keys
1757 (notated using the quote facility) so as to distinguish them from IPv4
1758 keys. This meant that lookups for IP addresses in host lists did not work
1759 for iplsearch lookups.
1761 This has been fixed by arranging for IPv6 addresses to be expressed with
1762 colons if the lookup type is iplsearch. This is not incompatible, because
1763 previously such lookups could never work.
1765 The situation is now rather anomolous, since one *can* have colons in
1766 ordinary lsearch keys. However, making the change in all cases is
1767 incompatible and would probably break a number of configurations.
1769 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
1772 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
1773 conversion specification without a maximum field width, thereby enabling
1774 a rogue spamd server to cause a buffer overflow. While nobody in their
1775 right mind would setup Exim to query an untrusted spamd server, an
1776 attacker that gains access to a server running spamd could potentially
1777 exploit this vulnerability to run arbitrary code as the Exim user.
1779 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
1780 $primary_hostname instead of what libspf2 thinks the hosts name is.
1782 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
1783 a directory entry by the name of the lookup key. Previously, if a
1784 symlink pointed to a non-existing file or a file in a directory that
1785 Exim lacked permissions to read, a lookup for a key matching that
1786 symlink would fail. Now it is enough that a matching directory entry
1787 exists, symlink or not. (Bugzilla 503.)
1789 PH/03 The body_linecount and body_zerocount variables are now exported in the
1792 PH/04 Added the $dnslist_matched variable.
1794 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
1795 This means they are set thereafter only if the connection becomes
1798 PH/06 Added the client_condition to authenticators so that some can be skipped
1799 by clients under certain conditions.
1801 PH/07 The error message for a badly-placed control=no_multiline_responses left
1802 "_responses" off the end of the name.
1804 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
1806 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
1807 (without spaces) instead of just copying the configuration text.
1809 PH/10 Added the /noupdate option to the ratelimit ACL condition.
1811 PH/11 Added $max_received_linelength.
1813 PH/12 Added +ignore_defer and +include_defer to host lists.
1815 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
1816 way in which PCRE > 7.0 is built.
1818 PH/14 Implemented queue_only_load_latch.
1820 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
1821 MAIL command. The effect was to mangle the value on 64-bit systems.
1823 PH/16 Another patch from the Sieve maintainer.
1825 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
1827 PH/18 If a system quota error occurred while trying to create the file for
1828 a maildir delivery, the message "Mailbox is full" was not appended to the
1829 bounce if the delivery eventually timed out. Change 4.67/27 below applied
1830 only to a quota excession during the actual writing of the file.
1832 PH/19 It seems that peer DN values may contain newlines (and other non-printing
1833 characters?) which causes problems in log lines. The DN values are now
1834 passed through string_printing() before being added to log lines.
1836 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
1837 and InterBase are left for another time.)
1839 PH/21 Added message_body_newlines option.
1841 PH/22 Guard against possible overflow in moan_check_errorcopy().
1843 PH/23 POSIX allows open() to be a macro; guard against that.
1845 PH/24 If the recipient of an error message contained an @ in the local part
1846 (suitably quoted, of course), incorrect values were put in $domain and
1847 $local_part during the evaluation of errors_copy.
1853 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
1854 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
1857 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
1858 issue a MAIL command.
1860 PH/02 In an ACL statement such as
1862 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
1864 if a client was not listed at all, or was listed with a value other than
1865 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
1866 the condition was not true (as it should be), so access was not denied.
1867 The bug was that the ! inversion was incorrectly passed on to the second
1868 item. This has been fixed.
1870 PH/03 Added additional dnslists conditions == and =& which are different from
1871 = and & when the dns lookup returns more than one IP address.
1873 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
1874 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
1876 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
1877 FSYNC, which compiles an option called disable_fsync that allows for
1878 bypassing fsync(). The documentation is heavily laced with warnings.
1880 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
1882 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
1883 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
1884 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
1885 including adding "make clean"; (3) Added -fPIC when compiling the test
1886 dynamically loaded module, to get rid of a warning.
1888 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
1889 message fails, move_frozen_messages = true and ignore_bounce_errors_after
1890 = 0s. The bug is otherwise harmless.
1892 PH/07 There was a bug in the dovecot authenticator such that the value of
1893 $auth1 could be overwritten, and so not correctly preserved, after a
1894 successful authentication. This usually meant that the value preserved by
1895 the server_setid option was incorrect.
1897 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
1899 PH/09 Installed PCRE release 7.0.
1901 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
1902 run for batched SMTP input. It is now run at the start of every message
1903 in the batch. While fixing this I discovered that the process information
1904 (output by running exiwhat) was not always getting set for -bs and -bS
1905 input. This is fixed, and it now also says "batched" for BSMTP.
1907 PH/11 Added control=no_pipelining.
1909 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
1910 patch, slightly modified), and move the expansion of helo_data till after
1911 the connection is made in the smtp transport (so it can use these
1914 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
1916 PH/14 Added log_selector = +pid.
1918 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
1920 PH/16 Add ${if forany and ${if forall.
1922 PH/17 Added dsn_from option to vary the From: line in DSNs.
1924 PH/18 Flush SMTP output before performing a callout, unless control =
1925 no_callout_flush is set.
1927 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
1928 was true (the default) a successful delivery failed to delete the retry
1929 item, thus causing premature timeout of the address. The bug is now
1932 PH/20 Added hosts_avoid_pipelining to the smtp transport.
1934 PH/21 Long custom messages for fakedefer and fakereject are now split up
1935 into multiline reponses in the same way that messages for "deny" and
1936 other ACL rejections are.
1938 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
1939 with slight modification.
1941 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
1942 draft, changing the syntax and factoring some duplicate code".
1944 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
1945 for deliveries of the second and subsequent messages over the same SMTP
1948 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
1949 ${reduce, with only minor "tidies".
1951 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
1953 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
1954 expansion side effects.
1956 PH/27 When a message times out after an over-quota error from an Exim-imposed
1957 quota, the bounce message says "mailbox is full". This message was not
1958 being given when it was a system quota that was exceeded. It now should
1961 MH/03 Made $recipients available in local_scan(). local_scan() already has
1962 better access to the recipient list through recipients_list[], but
1963 $recipients can be useful in postmaster-provided expansion strings.
1965 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
1966 in the case of a MAIL command with additional options following the
1967 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
1968 were accidentally chopped off.
1970 PH/29 SMTP synchronization checks are implemented when a command is read -
1971 there is a check that no more input is waiting when there shouldn't be
1972 any. However, for some commands, a delay in an ACL can mean that it is
1973 some time before the response is written. In this time, more input might
1974 arrive, invalidly. So now there are extra checks after an ACL has run for
1975 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
1976 pipelining has not been advertised.
1978 PH/30 MH's patch to allow iscntrl() characters to be list separators.
1980 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
1981 returned in the SMTP response when smtp_return_error_details was false.
1982 This has been fixed.
1984 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
1985 instead of the C I/O that was originally supplied, because problems were
1986 reported on Solaris.
1988 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
1989 Exim which did not show up earlier: it was assuming that a call to
1990 SSL_CTX_set_info_callback() might give an error value. In fact, there is
1991 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
1992 was a macro that became an assignment, so it seemed to work. This has
1993 changed to a proper function call with a void return, hence the compile
1994 error. Exim's code has been fixed.
1996 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
1999 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
2001 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
2003 PH/37 If a message is not accepted after it has had an id assigned (e.g.
2004 because it turns out to be too big or there is a timeout) there is no
2005 "Completed" line in the log. When some messages of this type were
2006 selected by exigrep, they were listed as "not completed". Others were
2007 picked up by some special patterns. I have improved the selection
2008 criteria to be more general.
2010 PH/38 The host_find_failed option in the manualroute router can now be set
2011 to "ignore", to completely ignore a host whose IP address cannot be
2012 found. If all hosts are ignored, the behaviour is controlled by the new
2013 host_all_ignored option.
2015 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
2016 homing or because of multiple MX records with /mx) generated more than
2017 one IP address, and the following item turned out to be the local host,
2018 all the secondary addresses of the first item were incorrectly removed
2019 from the list, along with the local host and any following hosts (which
2020 is what is supposed to happen).
2022 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
2023 whoever called Exim into the -H file. In the case of the daemon it was
2024 behaving confusingly. When first started, it used values for whoever
2025 started the daemon, but after a SIGHUP it used the Exim user (because it
2026 calls itself on a restart). I have changed the code so that it now always
2029 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
2030 message are rejected with the same error (e.g. no authentication or bad
2031 sender address), and a DATA command is nevertheless sent (as can happen
2032 with PIPELINING or a stupid MUA), the error message that was given to the
2033 RCPT commands is included in the rejection of the DATA command. This is
2034 intended to be helpful for MUAs that show only the final error to their
2037 PH/42 Another patch from the Sieve maintainer.
2039 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
2040 Eximstats - Fixed some broken HTML links and added missing column headers
2042 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
2043 columns for Rejects, Temp Rejects, Ham, and Spam rows.
2045 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
2047 PH/43 Yet another patch from the Sieve maintainer.
2049 PH/44 I found a way to check for a TCP/IP connection going away before sending
2050 the response to the final '.' that terminates a message, but only in the
2051 case where the client has not sent further data following the '.'
2052 (unfortunately, this is allowed). However, in many cases there won't be
2053 any further data because there won't be any more messages to send. A call
2054 to select() can be used: if it shows that the input is "ready", there is
2055 either input waiting, or the socket has been closed. An attempt to read
2056 the next input character can distinguish the two cases. Previously, Exim
2057 would have sent an OK response which the client would never have see.
2058 This could lead to message repetition. This fix should cure that, at
2059 least in a lot of common cases.
2061 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
2062 advertised in response to EHLO.
2068 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
2069 fixed by 4.65/MH/01 (is this a record?) are fixed:
2071 (i) An empty string was always treated as zero by the numeric comparison
2072 operators. This behaviour has been restored.
2074 (ii) It is documented that the numeric comparison operators always treat
2075 their arguments as decimal numbers. This was broken in that numbers
2076 starting with 0 were being interpreted as octal.
2078 While fixing these problems I realized that there was another issue that
2079 hadn't been noticed. Values of message_size_limit (both the global option
2080 and the transport option) were treated as octal if they started with 0.
2081 The documentation was vague. These values are now always treated as
2082 decimal, and I will make that clear in the documentation.
2088 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
2089 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
2092 MH/01 Don't check that the operands of numeric comparison operators are
2093 integers when their expansion is in "skipping" mode (fixes bug
2094 introduced by 4.64-PH/07).
2096 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
2097 child addresses, Exim now panics and dies. Previously, because the count
2098 is held in a short int, deliveries were likely to be lost. As such a
2099 large number of recipients for a single message is ridiculous
2100 (performance will be very, very poor), I have chosen to impose a limit
2101 rather than extend the field.
2107 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
2108 leftover -K file (the existence of which was triggered by #402).
2109 While we were at it, introduced process PID as part of the -K
2110 filename. This should rule out race conditions when creating
2113 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
2114 processing considerably. Previous code took too long for large mails,
2115 triggering a timeout which in turn triggers #401.
2117 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
2118 in the DK code in transports.c. sendfile() is not really portable,
2119 hence the _LINUX specificness.
2121 TF/01 In the add_headers option to the mail command in an Exim filter,
2122 there was a bug that Exim would claim a syntax error in any
2123 header after the first one which had an odd number of characters
2126 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
2127 callout verification, Exim cached a "reject" for the entire domain. This
2128 is correct for most verifications, but it is not correct for a recipient
2129 verification with use_sender or use_postmaster set, because in that case
2130 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
2131 case of MAIL FROM:<> rejection from other early rejections (e.g.
2132 rejection of HELO). When verifying a recipient using a non-null MAIL
2133 address, the cache is ignored if it shows MAIL FROM:<> rejection.
2134 Whatever the result of the callout, the value of the domain cache is
2135 left unchanged (for any other kind of callout, getting as far as trying
2136 RCPT means that the domain itself is ok).
2138 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
2141 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
2142 manifest itself as EPIPE rather than ECONNECT. When tidying away a
2143 session, the daemon ignores ECONNECT errors and logs others; it now
2144 ignores EPIPE as well.
2146 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
2147 (quoted-printable decoding).
2149 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
2150 later the small subsequent patch to fix an introduced bug.
2152 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
2154 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
2156 PH/08 An error is now given if message_size_limit is specified negative.
2158 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
2159 to be given (somewhat) arbitrary names.
2161 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
2164 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
2165 miscellaneous code fixes
2167 PH/10 Added the log_reject_target ACL modifier to specify where to log
2170 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
2171 hostname. This is wrong, because it relates to the incoming message (and
2172 probably the interface on which it is arriving) and not to the outgoing
2173 callout (which could be using a different interface). This has been
2174 changed to use the value of the helo_data option from the smtp transport
2175 instead - this is what is used when a message is actually being sent. If
2176 there is no remote transport (possible with a router that sets up host
2177 addresses), $smtp_active_hostname is used.
2179 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
2180 tweaks were necessary in order to get it to work (see also 21 below):
2181 (a) The code assumed that strncpy() returns a negative number on buffer
2182 overflow, which isn't the case. Replaced with Exim's string_format()
2184 (b) There were several signed/unsigned issues. I just did the minimum
2185 hacking in of casts. There is scope for a larger refactoring.
2186 (c) The code used strcasecmp() which is not a standard C function.
2187 Replaced with Exim's strcmpic() function.
2188 (d) The code set only $1; it now sets $auth1 as well.
2189 (e) A simple test gave the error "authentication client didn't specify
2190 service in request". It would seem that Dovecot has changed its
2191 interface. Fortunately there's a specification; I followed it and
2192 changed what the client sends and it appears to be working now.
2194 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
2197 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
2198 address A is aliased to B and C, where B exists and C does not. Without
2199 -v the output is "A verified" because verification stops after a
2200 successful redirection if more than one address is generated. However,
2201 with -v the child addresses are also verified. Exim was outputting "A
2202 failed to verify" and then showing the successful verification for C,
2203 with its parentage. It now outputs "B failed to verify", showing B's
2204 parentage before showing the successful verification of C.
2206 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
2207 look up a TXT record in a specific list after matching in a combined
2210 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
2211 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
2212 they consult the DNS. I had assumed they would set it the way they
2213 wanted; and indeed my experiments on Linux seem to show that in some
2214 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
2215 To be on the safe side, however, I have now made the interface to
2216 host_find_byname() similar to host_find_bydns(), with an argument
2217 containing the DNS resolver options. The host_find_byname() function now
2218 sets these options at its start, just as host_find_bydns() does. The smtp
2219 transport options dns_qualify_single and dns_search_parents are passed to
2220 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
2221 of host_find_byname() use the default settings of RES_DEFNAMES
2222 (qualify_single) but not RES_DNSRCH (search_parents).
2224 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
2225 spool_read_header() do less string testing, by means of a preliminary
2226 switch on the second character of optional "-foo" lines. (This is
2227 overdue, caused by the large number of possibilities that now exist.
2228 Originally there were few.) While I was there, I also converted the
2229 str(n)cmp tests so they don't re-test the leading "-" and the first
2230 character, in the hope this might squeeze out yet more improvement.
2232 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
2233 flag allowing group syntax was set by the header_syntax check but not
2234 turned off, possible causing trouble later; (2) The flag was not being
2235 set at all for the header_verify test, causing "group"-style headers to
2236 be rejected. I have now set it in this case, and also caused header_
2237 verify to ignore an empty address taken from a group. While doing this, I
2238 came across some other cases where the code for allowing group syntax
2239 while scanning a header line wasn't quite right (mostly, not resetting
2240 the flag correctly in the right place). These bugs could have caused
2241 trouble for malformed header lines. I hope it is now all correct.
2243 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
2244 with the "reply" argument non-NULL. The code, however (which originally
2245 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
2246 but it didn't always do it. This confused somebody who was copying the
2247 code for some other use. I have removed all the tests.
2249 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
2250 feature that was used to support insecure browsers during the U.S. crypto
2251 embargo. It requires special client support, and Exim is probably the
2252 only MTA that supported it -- and would never use it because real RSA is
2253 always available. This code has been removed, because it had the bad
2254 effect of slowing Exim down by computing (never used) parameters for the
2255 RSA_EXPORT functionality.
2257 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
2258 authenticator to fail if there's a tab character in the incoming data
2259 (there should never be unless someone is messing about, as it's supposed
2260 to be base64-encoded). Also added, on Timo's advice, the "secured" option
2261 if the connection is using TLS or if the remote IP is the same as the
2262 local IP, and the "valid-client-cert option" if a client certificate has
2265 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
2266 authenticators. This can be used for authorization after authentication
2267 succeeds. (In the case of plaintext, it servers for both authentication
2270 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
2271 if any retry times were supplied.
2273 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
2274 connection, where there is no client IP address to check. In this
2275 situation, the verify now always succeeds.
2277 PH/25 Applied John Jetmore's -Mset patch.
2279 PH/26 Added -bem to be like -Mset, but loading a message from a file.
2281 PH/27 In a string expansion for a processed (not raw) header when multiple
2282 headers of the same name were present, leading whitespace was being
2283 removed from all of them, but trailing whitespace was being removed only
2284 from the last one. Now trailing whitespace is removed from each header
2285 before concatenation. Completely empty headers in a concatenation (as
2286 before) are ignored.
2288 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
2289 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
2291 PH/29 [Removed. This was a change that I later backed out, and forgot to
2292 correct the ChangeLog entry (that I had efficiently created) before
2293 committing the later change.]
2295 PH/30 Exim was sometimes attempting to deliver messages that had suffered
2296 address errors (4xx response to RCPT) over the same connection as other
2297 messages routed to the same hosts. Such deliveries are always "forced",
2298 so retry times are not inspected. This resulted in far too many retries
2299 for the affected addresses. The effect occurred only when there were more
2300 hosts than the hosts_max_try setting in the smtp transport when it had
2301 the 4xx errors. Those hosts that it had tried were not added to the list
2302 of hosts for which the message was waiting, so if all were tried, there
2303 was no problem. Two fixes have been applied:
2305 (i) If there are any address or message errors in an SMTP delivery, none
2306 of the hosts (tried or untried) are now added to the list of hosts
2307 for which the message is waiting, so the message should not be a
2308 candidate for sending over the same connection that was used for a
2309 successful delivery of some other message. This seems entirely
2310 reasonable: after all the message is NOT "waiting for some host".
2311 This is so "obvious" that I'm not sure why it wasn't done
2312 previously. Hope I haven't missed anything, but it can't do any
2313 harm, as the worst effect is to miss an optimization.
2315 (ii) If, despite (i), such a delivery is accidentally attempted, the
2316 routing retry time is respected, so at least it doesn't keep
2317 hammering the server.
2319 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
2320 in ${readsocket because some servers need this prod.
2322 PH/32 Added some extra debug output when updating a wait-xxx database.
2324 PH/33 The hint "could be header name not terminated by colon", which has been
2325 given for certain expansion errors for a long time, was not being given
2326 for the ${if def:h_colon_omitted{... case.
2328 PH/34 The spec says: "With one important exception, whenever a domain list is
2329 being scanned, $domain contains the subject domain." There was at least
2330 one case where this was not true.
2332 PH/35 The error "getsockname() failed: connection reset by peer" was being
2333 written to the panic log as well as the main log, but it isn't really
2334 panic-worthy as it just means the connection died rather early on. I have
2335 removed the panic log writing for the ECONNRESET error when getsockname()
2338 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
2339 runs only) independently of the message's sender address. This meant
2340 that, if the 4xx error was in fact related to the sender, a different
2341 message to the same recipient with a different sender could confuse
2342 things. In particualar, this can happen when sending to a greylisting
2343 server, but other circumstances could also provoke similar problems.
2344 I have changed the default so that the retry time for these errors is now
2345 based a combination of the sender and recipient addresses. This change
2346 can be overridden by setting address_retry_include_sender=false in the
2349 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
2350 remote server are returned as part of bounce messages. This was not
2351 happening for LMTP over a pipe (the lmtp transport), but now it is the
2352 same for both kinds of LMTP.
2354 PH/38 Despite being documented as not happening, Exim was rewriting addresses
2355 in header lines that were in fact CNAMEs. This is no longer the case.
2357 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
2358 and queue runs started by the daemon processed all messages. This has
2359 been fixed so that -R and -S can now usefully be given with -q<time>.
2361 PH/40 Import PCRE release 6.7 (fixes some bugs).
2363 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
2365 PH/42 Give an error if -q is specified more than once.
2367 PH/43 Renamed the variables $interface_address and $interface_port as
2368 $received_ip_address and $received_port, to make it clear that these
2369 values apply to message reception, and not to the outgoing interface when
2370 a message is delivered. (The old names remain recognized, of course.)
2372 PH/44 There was no timeout on the connect() call when using a Unix domain
2373 socket in the ${readsocket expansion. There now is.
2375 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
2376 be meaningful with "accept".
2378 SC/01 Eximstats V1.43
2379 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
2381 SC/02 Eximstats V1.44
2382 Use a glob alias rather than an array ref in the generated
2383 parser. This improves both readability and performance.
2385 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
2386 Collect SpamAssassin and rejection statistics.
2387 Don't display local sender or destination tables unless
2388 there is data to show.
2389 Added average volumes into the top table text output.
2391 SC/04 Eximstats V1.46
2392 Collect data on the number of addresses (recipients)
2393 as well as the number of messages.
2395 SC/05 Eximstats V1.47
2396 Added 'Message too big' to the list of mail rejection
2397 reasons (thanks to Marco Gaiarin).
2399 SC/06 Eximstats V1.48
2400 Mainlog lines which have GMT offsets and are too short to
2401 have a flag are now skipped.
2403 SC/07 Eximstats V1.49 (Alain Williams)
2404 Added the -emptyok flag.
2406 SC/08 Eximstats V1.50
2407 Fixes for obtaining the IP address from reject messages.
2409 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
2410 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
2411 whitesspace changes from 4.64-PH/27
2413 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
2416 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
2417 are found, allow negative numbers in numeric criteria)
2419 JJ/06 exipick.20061117.2, added new $message_body_missing variable
2421 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
2422 to match changes made in 4.64-PH/43
2424 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
2426 PH/47 Put in an explicit test for a DNS lookup of an address record where the
2427 "domain" is actually an IP address, and force a failure. This locks out
2428 those revolvers/nameservers that support "A-for-A" lookups, in
2429 contravention of the specifications.
2431 PH/48 When a host name was looked up from an IP address, and the subsequent
2432 forward lookup of the name timed out, the host name was left in
2433 $sender_host_name, contrary to the specification.
2435 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
2436 restricted to single-key lookups, Exim was not diagnosing an error if
2437 * or *@ was used with a query-style lookup.
2439 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
2441 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
2442 long ago, but noone interested enough thought of it. Let's just say that
2443 the "1.1" means that there are some new functions that weren't there at
2444 some point in the past.
2446 PH/51 Error processing for expansion failure of helo_data from an smtp
2447 transport during callout processing was broken.
2449 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
2450 tested/used via the -bh/-bhc/-bs options.
2452 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
2453 bug, fixed in subsequent PCRE releases).
2455 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
2456 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
2458 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
2464 SC/01 Use a glob alias rather than an array ref in eximstats generated
2465 parser. This improves both readability and performance.
2467 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
2468 Don't display local sender or destination tables in eximstats unless
2469 there is data to show.
2470 Added average volumes into the eximstats top table text output.
2472 SC/03 Collect data on the number of addresses (recipients) as well
2473 as the number of messages in eximstats.
2475 TF/01 Correct an error in the documentation for the redirect router. Exim
2476 does (usually) call initgroups() when daemonizing.
2478 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
2479 with consistent privilege compared to when running as a daemon.
2481 TF/03 Note in the spec that $authenticated_id is not set for local
2482 submissions from trusted users.
2484 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
2485 Thanks to Dean Brooks <dean@iglou.com> for the patch.
2487 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
2488 by adding some example configuration directives to the default
2489 configuration file. A little bit of work is required to uncomment the
2490 directives and define how usernames and passwords are checked, but
2491 there is now a framework to start from.
2493 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
2494 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
2495 without this. I don't know how relevant this is to other LDAP libraries.
2497 PH/02 Add the verb name to the "unknown ACL verb" error.
2499 PH/03 Magnus Holmgren's patch for filter_prepend_home.
2501 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
2503 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
2504 directory not expanded when it should be if an expanded home directory
2505 was set for the address (which is overridden by the transport).
2507 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
2510 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
2511 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
2512 because it is too late at that time, and has no effect.
2514 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
2515 security issue with \' (bugzilla #107). I could not use the
2516 PQescapeStringConn() function, because it needs a PGconn value as one of
2519 PH/08 When testing addresses using -bt, indicate those final addresses that
2520 are duplicates that would not cause an additional delivery. At least one
2521 person was confused, thinking that -bt output corresponded to deliveries.
2522 (Suppressing duplicates isn't a good idea as you lose the information
2523 about possibly different redirections that led to the duplicates.)
2525 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
2526 systems where poll() doesn't work, in particular OS X.
2528 PH/10 Added more information to debugging output for retry time not reached.
2530 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
2531 operations in malware.c.
2533 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
2536 PH/13 If write_rejectlog was set false when logging was sent to syslog with
2537 syslog_duplication set false, log lines that would normally be written
2538 both the the main log and to the reject log were not written to syslog at
2541 PH/14 In the default configuration, change the use of "message" in ACL warn
2542 statements to "add_header".
2544 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
2545 not followed by a command (e.g. "seen endif").
2547 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
2548 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
2551 PH/17 Added extra conditions to the default value of delay_warning_condition
2555 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
2556 { match{$h_precedence:}{(?i)bulk|list|junk} } \
2557 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
2560 The Auto-Submitted: and various List- headers are standardised, whereas I
2561 don't think Precedence: ever was.
2563 PH/18 Refactored debugging code in route_finduser() to show more information,
2564 in particular, the error code if getpwnam() issues one.
2566 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
2567 This is apparently needed in addition to the PH/07 change above to avoid
2568 any possible encoding problems.
2570 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
2571 but not after initializing Perl.
2573 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
2574 output them only if debugging. By default they are written stderr,
2575 apparently, which is not desirable.
2577 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
2580 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
2583 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
2585 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
2586 authenticated or an ident call has been made. Suppress the default
2587 values for $authenticated_id and $authenticated_sender (but permit -oMai
2588 and -oMas) when testing with -bh.
2590 PH/24 Re-jigged the order of the tests in the default configuration so that the
2591 tests for valid domains and recipients precede the DNS black list and CSA
2592 tests, on the grounds that those ones are more expensive.
2594 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
2595 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
2596 This bug exists in every version of Exim that I still have, right back to
2599 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
2600 However, an attempt to turn on case-sensitivity in a regex key by
2601 including (?-i) didn't work because the subject string was already
2602 lowercased, and the effects were non-intuitive. It turns out that a
2603 one-line patch can be used to allow (?-i) to work as expected.
2609 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
2610 other effects) broke the use of negated acl sub-conditions.
2612 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
2615 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
2616 "Deny" causes Exim to reject the incoming connection with a 554 error.
2617 Unfortunately, if there is a major crisis, such as a disk failure,
2618 tcp-wrappers gives "deny", whereas what one would like would be some
2619 kind of temporary error. A kludge has been added to help with this.
2620 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
2621 554 error is used if errno is still zero or contains ENOENT (which occurs
2622 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
2625 PH/03 Add -lutil to the default FreeBSD LIBS setting.
2627 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
2628 errors. Otherwise a message that provokes a temporary error (when other
2629 messages do not) can cause a whole host to time out.
2631 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
2632 addresses were routed directly to files or pipes from a redirect router.
2633 File deliveries just didn't batch; pipe deliveries might have suffered
2636 PH/06 A failure to get a lock for a hints database would erroneously always say
2637 "Failed to get write lock", even when it was really a read lock.
2639 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
2640 of 0600. This has been changed to use the value of the lockfile_mode
2641 option (which defaults to 0600).
2643 PH/08 Applied small patch from the Sieve maintainer.
2645 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
2646 folder from quota calculations, a direct delivery into this folder messed
2647 up the contents of the maildirsize file. This was because the regex was
2648 used only to exclude .Trash (or whatever) when the size of the mailbox
2649 was calculated. There was no check that a delivery was happening into an
2650 excluded directory. This bug has been fixed by ignoring all quota
2651 processing for deliveries into excluded directories.
2653 PH/10 Added the maildirfolder_create_regex option to appendfile.
2659 PH/01 The code for finding all the local interface addresses on a FreeBSD
2660 system running IPv6 was broken. This may well have applied to all BSD
2661 systems, as well as to others that have similar system calls. The broken
2662 code found IPv4 interfaces correctly, but gave incorrect values for the
2663 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
2664 that it would not match correctly against @[] and not recognize the IPv6
2667 PH/02 The ipliteral router was not recognizing addresses of the form user@
2668 [ipv6:....] because it didn't know about the "ipv6:" prefix.
2670 PH/03 Added disable_ipv6.
2672 PH/04 Changed $reply_address to use the raw form of the headers instead of the
2673 decoded form, because it is most often used to construct To: headers
2674 lines in autoreplies, and the decoded form may well be syntactically
2675 invalid. However, $reply_address has leading white space removed, and all
2676 newlines turned into spaces so that the autoreply transport does not
2679 PH/05 If group was specified without a user on a router, and no group or user
2680 was specified on a transport, the group from the router was ignored.
2682 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
2683 for visible compile-time settings that can be used to change these
2684 numbers, for those that want even more. Backwards compatibility with old
2685 spool files has been maintained. However, going back to a previous Exim
2686 release will lost any variables that are in spool files.
2688 PH/07 Two small changes when running in the test harness: increase delay when
2689 passing a TCP/IP connection to a new process, in case the original
2690 process has to generate a bounce, and remove special handling of
2691 127.0.0.2 (sic), which is no longer necessary.
2693 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
2694 be the same on different OS.
2696 PH/09 Moved a debug statement in filter processing to avoid a race problem when
2699 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
2700 whether --show-vars was specified or not
2702 JJ/02 exipick: Added support for new ACL variable spool format introduced
2705 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
2706 syntactically invalid From: or Reply-to: line, and a filter used this to
2707 generate an autoreply, and therefore failed to obtain an address for the
2708 autoreply, Exim could try to deliver to a non-existent relative file
2709 name, causing unrelated and misleading errors. What now happens is that
2710 it logs this as a hard delivery error, but does not attempt to create a
2713 PH/11 The exinext utility has a -C option for testing purposes, but although
2714 the given file was scanned by exinext itself; it wasn't being passed on
2715 when Exim was called.
2717 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
2718 an end-of-file indication when reading a command response.
2720 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
2721 compiled. In many other places in Exim, IPv6 addresses are always
2722 recognized, so I have changed this. It also means that IPv4 domain
2723 literals of the form [IPV4:n.n.n.n] are now always recognized.
2725 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
2726 used if the router is not running as root, for example, when verifying at
2727 ACL time, or when using -bh. The debugging output from this situation was
2728 non-existent - all you got was a failure to exec. I have made two
2731 (a) Failures to set uid/gid, the current directory, or a process leader
2732 in a subprocess such as that created by queryprogram now generate
2733 suitable debugging ouput when -d is set.
2735 (b) The queryprogram router detects when it is not running as root,
2736 outputs suitable debugging information if -d is set, and then runs
2737 the subprocess without attempting to change uid/gid.
2739 PH/15 Minor change to Makefile for building test_host (undocumented testing
2742 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
2743 additional section of a DNS packet that returns MX or SRV records.
2744 Instead, it always explicitly searches for A/AAAA records. This avoids
2745 major problems that occur when a DNS server includes only records of one
2746 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
2747 fixed another bug: if SRV records were looked up and the corresponding
2748 address records were *not* found in the additional section, the port
2749 values from the SRV records were lost.
2751 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
2752 using the correct key (the original address) when searching the retry
2753 rules in order to find which one to use for generating the retry hint.
2755 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
2756 adding the default one. Similarly, if it contains a Reply-To: header, the
2757 errors_reply_to option, if set, is not used.
2759 PH/19 When calculating a retry time, Exim used to measure the "time since
2760 failure" by looking at the "first failed" field in the retry record. Now
2761 it does not use this if it is later than than the arrival time of the
2762 message. Instead it uses the arrival time. This makes for better
2763 behaviour in cases where some deliveries succeed, thus re-setting the
2764 "first failed" field. An example is a quota failure for a huge message
2765 when small messages continue to be delivered. Without this change, the
2766 "time since failure" will always be short, possible causing more frequent
2767 delivery attempts for the huge message than are intended.
2768 [Note: This change was subsequently modified - see PH/04 for 4.62.]
2770 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
2771 $1, $2, $3) because the numerical variables can be reset during some
2772 expansion items (e.g. "match"), thereby losing the authentication data.
2774 PH/21 Make -bV show the size of off_t variables so that the test suite can
2775 decide whether to run tests for quotas > 2G.
2777 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
2778 mailbox_size, and mailbox_filecount in the appendfile transport. If a
2779 filecount value is greater than 2G or if a quota value is greater than 2G
2780 on a system where the size of off_t is not greater than 4, a panic error
2783 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
2784 never match. The debug and -bh output now contains an explicit error
2785 message indicating a malformed IPv4 address or mask.
2787 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
2788 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
2789 PH/23 above applies.
2791 PH/25 Do not write to syslog when running in the test harness. The only
2792 occasion when this arises is a failure to open the main or panic logs
2793 (for which there is an explicit test).
2795 PH/26 Added the /no_tell option to "control=freeze".
2797 PH/27 If a host name lookup failed very early in a connection, for example, if
2798 the IP address matched host_lookup and the reverse lookup yielded a name
2799 that did not have a forward lookup, an error message of the form "no IP
2800 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
2801 could be logged. Now it outputs the IP address instead of "NULL".
2803 PH/28 An enabling patch from MH: add new function child_open_exim2() which
2804 allows the sender and the authenticated sender to be set when
2805 submitting a message from within Exim. Since child_open_exim() is
2806 documented for local_scan(), the new function should be too.
2808 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
2809 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
2810 results in an empty string is now treated as unset.
2812 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
2814 PH/31 Added sender_verify_fail logging option.
2816 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
2817 needed by RFC 822 but not by RFC 2822 was commented out. I have now
2818 tidied the source and removed it altogether.
2820 PH/33 When a queue run was abandoned because the load average was too high, a
2821 log line was always written; now it is written only if the queue_run log
2822 selector is set. In addition, the log line for abandonment now contains
2823 information about the queue run such as the pid. This is always present
2824 in "start" and "stop" lines but was omitted from the "abandon" line.
2826 PH/34 Omit spaces between a header name and the colon in the error message that
2827 is given when verify = headers_syntax fails (if there are lots of them,
2828 the message gets confusing).
2830 PH/35 Change the default for dns_check_names_pattern to allow slashes within
2831 names, as there are now some PTR records that contain slashes. This check
2832 is only to protect against broken name servers that fall over on strange
2833 characters, so the fact that it applies to all lookups doesn't matter.
2835 PH/36 Now that the new test suite is complete, we can remove some of the
2836 special code in Exim that was needed for the old test suite. For example,
2837 sorting DNS records because real resolvers return them in an arbitrary
2838 order. The new test suite's fake resolver always returns records in the
2841 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
2842 bounces) except when queue_only is set, to avoid logging races between
2843 the different processes.
2845 PH/38 Panic-die if .include specifies a non-absolute path.
2847 PH/39 A tweak to the "H" retry rule from its user.
2849 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
2850 a label. They prevented compilation on older perls.
2852 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
2853 a warning to be raised on newish perls.
2855 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
2856 on queue. Changes to match documented behaviour of showing count of
2857 messages matching specified criteria.
2859 PH/40 Changed the default ident timeout from 30s to 5s.
2861 PH/41 Added support for the use of login_cap features, on those BSD systems
2862 that have them, for controlling the resources used by pipe deliveries.
2864 PH/42 The content-scanning code uses fopen() to create files in which to put
2865 message data. Previously it was not paying any attention to the mode of
2866 the files. Exim runs with umask(0) because the rest of the code creates
2867 files with open(), and sets the required mode explicitly. Thus, these
2868 files were ending up world-writeable. This was not a big issue, because,
2869 being within the spool directory, they were not world-accessible. I have
2870 created a function called modefopen, which takes an additional mode
2871 argument. It sets umask(777), creates the file, chmods it to the required
2872 mode, then resets the umask. All the relevant calls to fopen() in the
2873 content scanning code have been changed to use this function.
2875 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
2876 to 24 hours. This avoids potential overflow problems when processing G
2877 and H retry rules. I suspect nobody ever tinkers with this value.
2879 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
2881 PH/45 When the plaintext authenticator is running as a client, the server's
2882 challenges are checked to ensure they are valid base64 strings. By
2883 default, the authentication attempt is cancelled if an invalid string is
2884 received. Setting client_ignore_invalid_base64 true ignores these errors.
2885 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
2886 they are received. Thus, the responses can be made to depend on the
2887 challenges. If an invalid string is ignored, an empty string is placed in
2890 PH/46 Messages that are created by the autoreply transport now contains a
2891 References: header, in accordance with RFCs 2822 and 3834.
2893 PH/47 Added authenticated_sender_force to the smtp transport.
2895 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
2897 PH/49 Installed latest patch from the Sieve maintainer.
2899 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
2900 was also set, the appendfile transport was unnecessarily scanning a
2901 directory of message files (e.g. for maildir delivery) to find the count
2902 of files (along with the size), even though it did not need this
2903 information. It now does the scan only if it needs to find either the
2904 size of the count of files.
2906 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
2908 PH/52 Two bugs concerned with error handling when the smtp transport is
2911 (i) Exim was not creating retry information for temporary errors given
2912 for individual recipients after the DATA command when the smtp transport
2913 was used in LMTP mode. This meant that they could be retried too
2914 frequently, and not timed out correctly.
2916 (ii) Exim was setting the flag that allows error details to be returned
2917 for LMTP errors on RCPT commands, but not for LMTP errors for individual
2918 recipients that were returned after the DATA command.
2920 PH/53 This is related to PH/52, but is more general: for any failing address,
2921 when detailed error information was permitted to be returned to the
2922 sender, but the error was temporary, then after the final timeout, only
2923 "retry timeout exceeded" was returned. Now it returns the full error as
2924 well as "retry timeout exceeded".
2926 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
2927 do this, and (what is worse) MTAs that accept it.
2929 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
2930 will now be deprecated.
2932 PH/56 New os.c-cygwin from the Cygwin maintainer.
2934 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
2935 formats (previously only available in exim formats via -bpr, -bpru,
2936 and -bpra. Now also available in native and exiqgrep formats)
2938 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
2939 with very large, slow to parse queues
2941 JJ/08 exipick: added ! as generic prefix to negate any criteria format
2943 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
2945 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
2946 responses to authentication challenges, though it was showing the
2947 challenges; (ii) I've removed the CR characters from the debug output for
2950 PH/58 Allow for the insertion of a newline as well as a space when a string
2951 is turned into more than one encoded-word during RFC 2047 encoding. The
2952 Sieve code now uses this.
2954 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
2955 data_4xx, lost_connection, tls_required.
2957 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
2958 message was being sent as an SMTP response.
2960 PH/61 Add -l and -k options to exicyclog.
2962 PH/62 When verifying, if an address was redirected to one new address, so that
2963 verification continued, and the new address failed or deferred after
2964 having set something in $address_data, the value of $address_data was not
2965 passed back to the ACL. This was different to the case when no
2966 redirection occurred. The value is now passed back in both cases.
2968 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
2969 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
2970 use login_cap.h, so on its own it isn't the distinguishing feature. The
2971 new name refers directly to the setclassresources() function.
2973 PH/65 Added configuration files for NetBSD3.
2975 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
2977 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
2978 is preferred over IPv4.
2980 PH/68 The bounce_return_message and bounce_return_body options were not being
2981 honoured for bounces generated during the reception of non-SMTP messages.
2982 In particular, this applied to messages rejected by the ACL. This bug has
2983 been fixed. However, if bounce_return_message is true and bounce_return_
2984 body is false, the headers that are returned for a non-SMTP message
2985 include only those that have been read before the error was detected.
2986 (In the case of an ACL rejection, they have all been read.)
2988 PH/69 The HTML version of the specification is now built in a directory called
2989 spec_html instead of spec.html, because the latter looks like a path with
2990 a MIME-type, and this confuses some software.
2992 PH/70 Catch two compiler warnings in sieve.c.
2994 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
2995 function verify_get_ident() calls ip_connect() to connect a socket, but
2996 if the "connect()" function timed out, ip_connect() used to close the
2997 socket. However, verify_get_ident() also closes the socket later, and in
2998 between Exim writes to the log, which may get opened at this point. When
2999 the socket was closed in ip_connect(), the log could get the same file
3000 descriptor number as the socket. This naturally causes chaos. The fix is
3001 not to close the socket in ip_connect(); the socket should be closed by
3002 the function that creates it. There was only one place in the code where
3003 this was missing, in the iplookup router, which I don't think anybody now
3004 uses, but I've fixed it anyway.
3006 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
3007 well as to direct DNS lookups. Otherwise the handling of names in host
3008 lists is inconsistent and therefore confusing.
3014 PH/01 Two changes to the default runtime configuration:
3016 (1) Move the checks for relay_from_hosts and authenticated clients from
3017 after to before the (commented out) DNS black list checks.
3019 (2) Add control=submission to the relay_from_hosts and authenticated
3020 clients checks, on the grounds that messages accepted by these
3021 statements are most likely to be submissions.
3023 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
3025 (1) Generate an error if the third argument for the ${prvs expansion is
3028 (2) Treat a missing third argument of ${prvscheck as if it were an empty
3031 (3) Reset the variables that are obtained from the first argument of
3032 ${prvscheck and used in the second argument before leaving the code,
3033 because their memory is reclaimed, so using them afterwards may do
3036 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
3037 one (it's much easier than Tom thought :-).
3039 (5) Because of (4), we can now allow for the use of $prvscheck_result
3040 inside the third argument.
3042 PH/03 For some reason, the default setting of PATH when running a command from
3043 a pipe transport was just "/usr/bin". I have changed it to
3046 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
3047 anything to be listed in the output from -bV.
3049 PH/05 When a filter generated an autoreply, the entire To: header line was
3050 quoted in the delivery log line, like this:
3052 => >A.N.Other <ano@some.domain> <original@ddress> ...
3054 This has been changed so that it extracts the operative address. There
3055 may be more than one such address. If so, they are comma-separated, like
3058 => >ano@some.domain,ona@other.domain <original@ddress> ...
3060 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
3061 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
3062 not being looked up in the rDNS to get a host name, Exim was showing the
3063 IP address twice in Received: lines, even though the IP addresses were
3064 identical. For example:
3066 Received: from [1.2.3.4] (helo=[1.2.3.4])
3068 However, if the real host name was known, it was omitting the HELO data
3069 if it matched the actual IP address. This has been tidied up so that it
3070 doesn't show the same IP address twice.
3072 PH/07 When both +timestamp and +memory debugging was on, the value given by
3073 $tod_xxx expansions could be wrong, because the tod_stamp() function was
3074 called by the debug printing, thereby overwriting the timestamp buffer.
3075 Debugging no longer uses the tod_stamp() function when +timestamp is set.
3077 PH/08 When the original message was included in an autoreply transport, it
3078 always said "this is a copy of the message, including all the headers",
3079 even if body_only or headers_only was set. It now gives an appropriate
3082 PH/09 Applied a patch from the Sieve maintainer which:
3084 o fixes some comments
3085 o adds the (disabled) notify extension core
3086 o adds some debug output for the result of if/elsif tests
3087 o points to the current vacation draft in the documentation
3088 and documents the missing references header update
3092 o fixes a bug in processing the envelope test (when testing
3093 multiple envelope elements, the last element determinted the
3096 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
3097 Electronic Mail") by including:
3099 Auto-submitted: auto-generated
3101 in the messages that it generates (bounce messages and others, such as
3102 warnings). In the case of bounce messages for non-SMTP mesages, there was
3103 also a typo: it was using "Auto_submitted" (underscore instead of
3104 hyphen). Since every message generated by Exim is necessarily in response
3105 to another message, thes have all been changed to:
3107 Auto-Submitted: auto-replied
3109 in accordance with these statements in the RFC:
3111 The auto-replied keyword:
3113 - SHOULD be used on messages sent in direct response to another
3114 message by an automatic process,
3116 - MUST NOT be used on manually-generated messages,
3118 - MAY be used on Delivery Status Notifications (DSNs) and Message
3119 Disposition Notifications (MDNs),
3121 - MUST NOT be used on messages generated by automatic or periodic
3122 processes, except for messages which are automatic responses to
3125 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
3126 to the default Received: header definition.
3128 PH/12 Added log selector acl_warn_skipped (default on).
3130 PH/13 After a successful wildlsearch lookup, discard the values of numeric
3131 variables because (a) they are in the wrong storage pool and (b) even if
3132 they were copied, it wouldn't work properly because of the caching.
3134 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
3135 checking when decoding. Apparently there are clients that generate
3136 overlong encoded strings. Why am I not surprised?
3138 PH/15 If the first argument of "${if match_address" was not empty, but did not
3139 contain an "@" character, Exim crashed. Now it writes a panic log message
3140 and treats the condition as false.
3142 PH/16 In autoreply, treat an empty string for "once" the same as unset.
3144 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
3145 extension "envelope-auth". The code is finished and in agreement with
3146 other implementations, but there is no documentation so far and in fact,
3147 nobody wrote the draft yet. This extension is currently #undef'ed, thus
3148 not changing the active code.
3150 Print executed "if" and "elsif" statements when debugging is used. This
3151 helps a great deal to understand what a filter does.
3153 Document more things not specified clearly in RFC3028. I had all this
3154 sorted out, when out of a sudden new issues came to my mind. Oops."
3156 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
3159 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
3160 canonical form (as documented). However, after a host name lookup from
3161 the IP address, check_host() was doing a simple string comparison with
3162 addresses acquired from the DNS when checking that the found name did
3163 have the original IP as one of its addresses. Since any found IPv6
3164 addresses are likely to be in abbreviated form, the comparison could
3165 fail. Luckily, there already exists a function for doing the comparison
3166 by converting both addresses to binary, so now that is used instead of
3167 the text comparison.
3169 PH/20 There was another similar case to PH/19, when a complete host name was
3170 given in a host list; looking up its IP address could give an abbreviated
3171 form, whereas the current host's name might or might not be abbreviated.
3172 The same fix has been applied.
3178 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
3179 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
3182 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
3183 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
3185 PH/03 Typo: missing ".o" in src/pcre/Makefile.
3187 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
3188 header line, restrict the check to what is listed in RFCs 2369 and 2929.
3189 Also, for "Auto-Submitted", treat anything other than "no" as
3190 non-personal, in accordance with RFC 3834. (Previously it treated
3191 anything starting "auto-" as non-personal.)
3193 TF/01 The control=submission/name=... option had a problem with syntax
3194 errors if the name included a slash character. The /name= option
3195 now slurps the rest of the string, so it can include any characters
3196 but it must come last in the list of options (after /sender_retain
3199 PH/05 Some modifications to the interface to the fake nameserver for the new
3207 TK/01 Added the "success_on_redirect" address verification option. See
3208 NewStuff for rationale and an example.
3210 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
3212 PH/02 Patch to exigrep to allow it to work on syslog lines.
3214 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
3215 fread() to skip over the body file's header line, because in Cygwin the
3216 header line is locked and is inaccessible.
3218 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
3219 co-exist for some time) to make it clear that it is the Exim ID that is
3220 referenced, not the Message-ID: header line.
3222 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
3223 string_format() function, because snprintf() does not exist on all
3226 PH/06 The use of forbid_filter_existstest now also locks out the use of the
3227 ${stat: expansion item.
3229 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
3230 protocol synchronization error", to keep the pedants happy.
3232 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
3233 well as for IRIX systems, when gcc is being used. See the host.c source
3236 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
3238 PH/10 Named domain lists were not working if used in a queue_smtp_domains
3241 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
3242 transport and to the smtp transport in LMTP mode.
3244 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
3246 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
3247 run a filter in a subprocess. This could lead to confusion in subsequent
3248 lookups in the parent process. There should also be a search_tidyup() at
3249 the end of the subprocess.
3251 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
3252 only if the host matched helo_try_verify_hosts, which caused the
3253 verification to occur when the EHLO/HELO command was issued. The ACL just
3254 tested the remembered result. Now, if a previous verification attempt has
3255 not happened, "verify = helo" does it there and then.
3257 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
3259 TK/03 Fix log output including CR from clamd.
3261 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
3262 exist provoked a memory error which could cause a segfault.
3264 PH/15 Installed PCRE 6.2
3266 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
3268 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
3269 of the problem. Specifically, suggested +O2 rather than +O1 for the
3272 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
3274 PH/20 If a delivery was routed to a non-standard port by means of an SRV
3275 record, the port was not correctly logged when the outgoing_port log
3276 selector was set (it logged the transort's default port).
3278 PH/21 Added support for host-specific ports to manualroute, queryprogram,
3279 fallback_hosts, and "hosts" in the smtp transport.
3281 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
3282 host errors such as "Connection refused".
3284 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
3285 authentication with radiusclient 0.4.9:
3287 - Error returned from rc_read_config was caught wrongly
3288 - Username/password not passed on to radius server due to wrong length.
3290 The presumption is that some radiusclient API changes for 4.51/PH/17
3291 were not taken care of correctly. The code is still untested by me (my
3292 Linux distribution still has 0.3.2 of radiusclient), but it was
3293 contributed by a Radius user.
3295 PH/24 When doing a callout, the value of $domain wasn't set correctly when
3296 expanding the "port" option of the smtp transport.
3298 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
3299 while reading a MIME header. Thanks to Tom Hughes for a patch.
3301 PH/24 Include config.h inside local_scan.h so that configuration settings are
3304 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
3305 that in an ACL for RCPT (for example), you can examine exactly what was
3308 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
3309 commands, but it was not correctly comparing the address with the actual
3310 client host address. Thus, it would show the EHLO address in Received:
3311 header lines when this was not necessary.
3313 PH/27 Added the % operator to ${eval:}.
3315 PH/28 Exim tries to create and chdir to its spool directory when it starts;
3316 it should be ignoring failures (because with -C, for example, it has lost
3317 privilege). It wasn't ignoring creation failures other than "already
3320 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
3323 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
3324 because an input line was too long, either on its own, or by virtue of
3325 too many continuations, the temporary file was not being removed, and the
3326 return code was incorrect.
3328 PH/31 Missing "BOOL" in function definition in filtertest.c.
3330 PH/32 Applied Sieve patches from the maintainer.
3332 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
3334 PH/33 Added "verify = not_blind".
3336 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
3337 Local/Makefile (with some defaults set). These are used in built scripts
3338 such as exicyclog, but they have never been used in the exim_install
3339 script (though there are many overriding facilities there). I have
3340 arranged that the exim_install script now takes note of these two
3343 PH/35 Installed configuration files for Dragonfly.
3345 PH/36 When a locally submitted message by a trusted user did not contain a
3346 From: header, and the sender address was obtained from -f or from an SMTP
3347 MAIL command, and the trusted user did not use -F to supply a sender
3348 name, $originator_name was incorrectly used when constructing a From:
3349 header. Furthermore, $originator_name was used for submission mode
3350 messages from external hosts without From: headers in a similar way,
3351 which is clearly wrong.
3353 PH/37 Added control=suppress_local_fixups.
3355 PH/38 When log_selector = +received_sender was set, and the addition of the
3356 sender made the log line's construction buffer exactly full, or one byte
3357 less than full, an overflow happened when the terminating "\n" was
3360 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
3361 when the result of a list match is failure because a DNS lookup failed.
3363 PH/40 RM_COMMAND is now used in the building process.
3365 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
3366 the "build-* directories that it finds.
3368 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
3369 address in a domain literal was a prefix of an interface address.
3371 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
3372 when verifying a sender address, unless rewrite_headers is false.
3374 PH/44 Wrote a long comment about why errors_to addresses are verified as
3375 recipients, not senders.
3377 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
3378 the ratelimit ACL was added.
3380 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
3382 PH/46 Added extra information about PostgreSQL errors to the error string.
3384 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
3385 avoiding the need to install special zones in a real server. This is
3386 backwards compatible; if it can't find the fake resolver, it drops back.
3387 Thus, both old and new test suites can be run.
3389 TF/02 Added util/ratelimit.pl
3391 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
3392 clock is set back in time.
3394 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
3395 Candler <B.Candler@pobox.com>.
3397 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
3398 OK for addresses that are the result of redirections.
3400 PH/48 A number of further additions for the benefit of the new test suite,
3401 including a fake gethostbyname() that interfaces to the fake DNS resolver
3404 TF/06 The fix for widen_domains has also been applied to qualify_single and
3405 search_parents which are the other dnslookup options that can cause
3408 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
3411 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
3413 TF/07 Exim produced the error message "an SRV record indicated no SMTP
3414 service" if it encountered an MX record with an empty target hostname.
3415 The message is now "an MX or SRV record indicated no SMTP service".
3417 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
3418 if the DNS of the sending site is misconfigured. This is quite a
3419 common situation. This change restores the behaviour of treating a
3420 helo verification defer as a failure.
3422 PH/51 If self=fail was set on a router, the bounce message did not include the
3423 actual error message.
3429 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
3431 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
3432 command itself ended in error, the underlying message about the transport
3433 filter timeout was being overwritten with the pipe command error. Now the
3434 underlying error message should be appended to the second error message.
3436 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
3438 PH/02 Reduce the amount of output that "make" produces by default. Full output
3439 can still be requested.
3441 PH/03 The warning log line about a condition test deferring for a "warn" verb
3442 was being output only once per connection, rather than after each
3443 occurrence (because it was using the same function as for successful
3444 "warn" verbs). This seems wrong, so I have changed it.
3446 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
3447 it should not have, which might have caused a crash in the right
3448 circumstances, but probably never did.
3450 PH/04 Installed a modified version of Tony Finch's patch to make submission
3451 mode fix the return path as well as the Sender: header line, and to
3452 add a /name= option so that you can make the user's friendly name appear
3455 TF/03 Added the control = fakedefer ACL modifier.
3457 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
3458 Mark Lowes for thorough testing.
3460 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
3462 TK/03 Merged latest SRS patch from Miles Wilton.
3464 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
3465 with the definition in sysexits.h (which is #included earlier).
3466 Fortunately, Exim does not actually use EX_OK. The code used to try to
3467 preserve the sysexits.h value, by assumimg that macro definitions were
3468 scanned for macro replacements. I have been disabused of this notion,
3469 so now the code just undefines EX_OK before #including unistd.h.
3471 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
3472 in the smtp transport. When a block could not be written in a single
3473 write() function, the timeout was being re-applied to each part-write.
3474 This seems wrong - if the receiver was accepting one byte at a time it
3475 would take for ever. The timeout is now adjusted when this happens. It
3476 doesn't have to be particularly precise.
3478 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
3479 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
3481 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
3482 without a domain if the check to <postmaster@domain> fails.
3484 SC/01 Eximstats: added -xls and the ability to specify output files
3485 (patch written by Frank Heydlauf).
3487 SC/02 Eximstats: use FileHandles for outputing results.
3489 SC/03 Eximstats: allow any combination of xls, txt, and html output.
3491 SC/04 Eximstats: fixed display of large numbers with -nvr option
3493 SC/05 Eximstats: fixed merging of reports with empty tables.
3495 SC/06 Eximstats: added the -include_original_destination flag
3497 SC/07 Eximstats: removed tabs and trailing whitespace.
3499 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
3501 TK/06 MBOX spool code: Add real "From " MBOX separator line
3502 so the .eml file is really in mbox format (even though
3503 most programs do not really care). Patch from Alex Miller.
3505 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
3506 The latter is generated from $received_to and is only set if the
3507 message has one envelope recipient. SA can use these headers,
3508 obviously out-of-the-box. Patch from Alex Miller.
3510 PH/08 The ${def test on a variable was returning false if the variable's
3511 value was "0", contrary to what the specification has always said!
3512 The result should be true unless the variable is empty.
3514 PH/09 The syntax error of a character other than { following "${if
3515 def:variable_name" (after optional whitespace) was not being diagnosed.
3516 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
3517 accidental colon was present, for example, could give incorrect results.
3519 PH/10 Tidied the code in a number of places where the st_size field of a stat()
3520 result is used (not including appendfile, where other changes are about
3523 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
3524 This involved changing a lot of size variables from int to off_t. It
3525 should work with maildirs and everything.
3527 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
3528 spamd dying while we are connected to it.
3530 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
3533 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
3534 function for BDB 4.3.
3536 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
3538 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
3539 the sources for PCRE have been re-arranged and more files are now
3542 PH/15 The code I had for printing potentially long long variables in PH/11
3543 above was not the best (it lost precision). The length of off_t variables
3544 is now inspected at build time, and an appropriate printing format (%ld
3545 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
3546 to be "long long int" or "long int". This is needed for the internal
3547 formatting function string_vformat().
3549 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
3550 the configuration file to be ":syslog", then the script "guesses" where
3551 the logs files are, rather than using the compiled in default. In our
3552 case the guess is not the same as the compiled default, so the script
3553 suddenly stopped working when I started to use syslog. The patch checks
3554 to see if log_file_path is "". If so, it attempts to read it from exim
3555 with no configuration file to get the compiled in version, before it
3556 falls back to the previous guessing code."
3558 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
3559 implementing BATV in an Exim configuration. See NewStuff for the gory
3562 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
3563 Makefile that are specific to HP-UX.
3565 PH/18 If the "use_postmaster" option was set for a recipient callout together
3566 with the "random" option, the postmaster address was used as the MAIL
3567 FROM address for the random test, but not for the subsequent recipient
3568 test. It is now used for both.
3570 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
3571 patch removes a few documentation additions to RFC 3028, because the
3572 latest draft now contains them. It adds the new en;ascii-case comparator
3573 and a new error check for 8bit text in MIME parts. Comparator and
3574 require names are now matched exactly. I enabled the subaddress
3575 extension, but it is not well tested yet (read: it works for me)."
3577 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
3578 rework some of the code of TK/09 above to avoid the hardwired use of
3579 "%lld" and "long long". Replaced the call to snprintf() with a call to
3582 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
3583 records point to non-existent hosts", "retry timeout exceeded", and
3584 "retry time not reached for any host after a long failure period".
3586 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
3587 experimental DomainKeys support:
3589 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
3590 (2) On an error such as an illegally used "control", the wrong name for
3591 the control was given.
3593 These problems did NOT occur unless DomainKeys support was compiled.
3595 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
3597 PH/24 Added ${if match_ip condition.
3599 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
3600 databases so that it will be absolutely obvious if a crash occurs in the
3601 DB library. This is a regular occurrence (often caused by mis-matched
3604 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
3605 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
3606 picked up on a user's system that detects such things. There doesn't seem
3607 to be a gcc warning option for this - only an attribute that has to be
3608 put on the function's prototype. It seems that in Fedora Core 4 they have
3609 set this on a number of new functions. No doubt there will be more in due
3612 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
3613 specify a transport. However, if an address that was verified by such a
3614 router was the subject of a callout, Exim crashed because it tried to
3615 read the rcpt_include_affixes from the non-existent transport. Now it
3616 just assumes that the setting of that option is false. This bug was
3617 introduced by 4.51/PH/31.
3619 PH/28 Changed -d+all to exclude +memory, because that information is very
3620 rarely of interest, but it makes the output a lot bigger. People tend to
3621 do -d+all out of habit.
3623 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
3624 code in os-type was giving problems when libc.so lives in lib64, like on
3627 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
3628 aren't the modern standard, and it seems that some systems' include files
3629 don't always have them. Exim was already checking for some of the newer
3630 ones like T_AAAA, and defining it itself. I've added checks for all the
3631 record types that Exim uses.
3633 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
3634 not automatically generating a new one, as it is supposed to. This
3635 prevented TLS from working. If the file did exist, but contained invalid
3636 data, a new version was generated, as expected. It was only the case of a
3637 non-existent file that was broken.
3639 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
3640 with a change in libdomainkeys > 0.64.
3642 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
3643 from DNS. If the selector record carries the flag, it now has
3644 precedence over the domain-wide flag.
3646 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
3648 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
3649 the use of a port name that isn't defined in /etc/services) occurred, the
3650 message was deferred as in a normal delivery, and thus remained on the
3651 spool, instead of being failed because of the mua_wrapper setting. This
3652 is now fixed, and I tidied up some of the mua_wrapper messages at the
3655 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
3656 the messages in a hash of arrays rather than using individual hashes.
3657 This is a bit cleaner and results in dramatic memory savings, albeit
3658 at a slight CPU cost.
3660 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
3661 as requested by Marc Sherman.
3663 SC/10 Eximstats: added histograms for user specified patterns as requested
3666 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
3668 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
3669 fopen() in the content-scanning modules that did not already have it.
3675 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
3676 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
3678 TK/02 Fix ACL "control" statement not being available in MIME ACL.
3680 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
3682 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
3683 to test Sieve filters that use "vacation".
3685 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
3686 that changes the way the GnuTLS parameters are stored in the cache file.
3687 The new format can be generated externally. For backward compatibility,
3688 if the data in the cache doesn't make sense, Exim assumes it has read an
3689 old-format file, and it generates new data and writes a new file. This
3690 means that you can't go back to an older release without removing the
3693 PH/03 A redirect router that has both "unseen" and "one_time" set does not
3694 work if there are any delivery delays because "one_time" forces the
3695 parent to be marked "delivered", so its unseen clone is never tried
3696 again. For this reason, Exim now forbids the simultaneous setting of
3699 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
3700 redirected to themselves ("homonym" addresses). Read the long ChangeLog
3701 entry if you want to know the details. The fix, however, neglected to
3702 consider the case when local delivery batching is involved. The test for
3703 "previously delivered" was not happening when checking to see if an
3704 address could be batched with a previous (undelivered) one; under
3705 certain circumstances this could lead to multiple deliveries to the same
3708 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
3709 in its include files, and this causes problems building Exim.
3711 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
3712 header_syntax) but Exim was just ignoring anything given after a slash.
3713 In particular, this caused confusion with an attempt to use "verify =
3714 reverse_host_lookup/defer_ok". An error is now given when options are
3715 supplied for verify items that do not have them. (Maybe reverse_host_
3716 lookup should have a defer_ok option, but that's a different point.)
3718 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
3719 defined by RFC 821) to 2048, because there were problems with some AUTH
3720 commands, and RFC 1869 says the size should be increased for extended
3721 SMTP commands that take arguments.
3723 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
3726 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
3727 "unknown" error; now it says that the functionality isn't in the binary.
3729 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
3730 an address' error message when a string expansion fails (syntax or
3731 whatever). Otherwise the password may appear in the log. Following change
3732 PH/42 below, there is no longer a chance of it appearing in a bounce
3735 PH/11 Installed exipick version 20050225.0 from John Jetmore.
3737 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
3738 of its addresses was ever tried. (Bugzilla bug #2.)
3740 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
3741 the result incorrectly in the debug output. (It correctly added a newline
3742 to what was transported.)
3744 TF/01 Added $received_time.
3746 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
3747 commented out examples of how to interface to a virus scanner and to
3748 SpamAssassin. Also added commented examples of av_scanner and
3749 spamd_address settings.
3751 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
3752 and controls are allowed in which ACLs. There were a couple of minor
3753 errors. Some of the entries in the conditions table (which is a table of
3754 where they are NOT allowed) were getting very unwieldy; rewrote them as a
3755 negation of where the condition IS allowed.
3757 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
3759 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
3760 header file does not have a version number, so I've had to invent a new
3761 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
3762 API. The code is untested by me (my Linux distribution still has 0.3.2 of
3763 radiusclient), but it was contributed by a Radius user.
3765 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
3766 files or directories, for OpenSSL.
3768 PH/19 When an Exim process that is running as root has to create an Exim log
3769 file, it does so in a subprocess that runs as exim:exim so as to get the
3770 ownership right at creation (otherwise, other Exim processes might see
3771 the file with the wrong ownership). There was no test for failure of this
3772 fork() call, which would lead to the process getting stuck as it waited
3773 for a non-existent subprocess. Forks do occasionally fail when resources
3774 run out. I reviewed all the other calls to fork(); they all seem to check
3777 PH/20 When checking for unexpected SMTP input at connect time (before writing
3778 the banner), Exim was not dealing correctly with a non-positive return
3779 from the read() function. If the client had disconnected by this time,
3780 the result was a log entry for a synchronization error with an empty
3781 string after "input=" when read() returned zero. If read() returned -1
3782 (an event I could not check), uninitialized data bytes were printed.
3783 There were reports of junk text (parts of files, etc) appearing after
3786 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
3788 PH/22 Added support for macro redefinition, and (re)definition in between
3789 driver and ACL definitions.
3791 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
3792 forgetting to use the resulting value; it was using the unexpanded value.
3794 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
3795 hadn't been configured. The fix is from Juergen Kreileder, who
3796 understands it better than I do:
3798 "Here's what I see happening with three configured cyrus_sasl
3799 authenticators configured (plain, login, cram-md5):
3801 On startup auth_cyrus_sasl_init() gets called for each of these.
3802 This means three calls to sasl_listmech() without a specified mech_list.
3803 => SASL tests which mechs of all available mechs actually work
3804 => three warnings about OTP not working
3805 => the returned list contains: plain, login, cram-md5, digest-md5, ...
3807 With the patch, sasl_listmech() also gets called three times. But now
3808 SASL's mech_list option is set to the server_mech specified in the the
3809 authenticator. Or in other words, the answer from sasl_listmech()
3810 gets limited to just the mech you're testing for (which is different
3812 => the return list contains just 'plain' or 'login', 'cram-md5' or
3813 nothing depending on the value of ob->server_mech.
3815 I've just tested the patch: Authentication still works fine,
3816 unavailable mechs specified in the exim configuration are still
3817 caught, and the auth.log warnings about OTP are gone."
3819 PH/25 When debugging is enabled, the contents of the command line are added
3820 to the debugging output, even when log_selector=+arguments is not
3823 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
3824 answer is "GNU", and only if the return is "GNU/something" is the answer
3827 PH/27 $acl_verify_message is now set immediately after the failure of a
3828 verification in an ACL, and so is available in subsequent modifiers. In
3829 particular, the message can be preserved by coding like this:
3831 warn !verify = sender
3832 set acl_m0 = $acl_verify_message
3834 Previously, $acl_verify_message was set only while expanding "message"
3835 and "log_message" when a very denied access.
3837 PH/28 Modified OS/os.c-Linux with
3839 -#ifndef OS_LOAD_AVERAGE
3840 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
3842 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
3843 nomenclature these days.)
3845 PH/29 Installed patch from the Sieve maintainer that adds the options
3846 sieve_useraddress and sieve_subaddress to the redirect router.
3848 PH/30 In these circumstances:
3849 . Two addresses routed to the same list of hosts;
3850 . First host does not offer TLS;
3851 . First host accepts first address;
3852 . First host gives temporary error to second address;
3853 . Second host offers TLS and a TLS session is established;
3854 . Second host accepts second address.
3855 Exim incorrectly logged both deliveries with the TLS parameters (cipher
3856 and peerdn, if requested) that were in fact used only for the second
3859 PH/31 When doing a callout as part of verifying an address, Exim was not paying
3860 attention to any local part prefix or suffix that was matched by the
3861 router that accepted the address. It now behaves in the same way as it
3862 does for delivery: the affixes are removed from the local part unless
3863 rcpt_include_affixes is set on the transport.
3865 PH/32 Add the sender address, as F=<...>, to the log line when logging a
3866 timeout during the DATA phase of an incoming message.
3868 PH/33 Sieve envelope tests were broken for match types other than :is. I have
3869 applied a patch sanctioned by the Sieve maintainer.
3871 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
3872 the uid or gid is negative. A case of a negative gid caused this to be
3873 noticed. The fix allows for either to be negative.
3875 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
3876 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
3879 PH/36 The change PH/12 above was broken. Fixed it.
3881 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
3882 the grounds that routing the same address twice would always produce the
3883 same answer. This might have been true once, but it is certainly no
3884 longer true now. Routing a child address may depend on the previous
3885 routing that produced that child. Some complicated redirection strategies
3886 went wrong when messages had multiple recipients, and made Exim's
3887 behaviour dependent on the order in which the addresses were given.
3889 I have moved the duplicate checking until after the routing is complete.
3890 Exim scans the addresses that are assigned to local and remote
3891 transports, and removes any duplicates. This means that more work will be
3892 done, as duplicates will always all be routed, but duplicates are
3893 presumably rare, so I don't expect this is of any significance.
3895 For deliveries to pipes, files, and autoreplies, the duplicate checking
3896 still happens during the routing process, since they are not going to be
3899 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
3900 It corrects a timeout issue with spamd. This is Ian's comment: "The
3901 background is that sometimes spamd either never reads data from a
3902 connection it has accepted, or it never writes response data. The exiscan
3903 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
3904 blindly assumes that writes won't block so it may never time out."
3906 PH/39 Allow G after quota size as well as K and M.
3908 PH/40 The value set for $authenticated_id in an authenticator may not contain
3909 binary zeroes or newlines because the value is written to log lines and
3910 to spool files. There was no check on this. Now the value is run through
3911 the string_printing() function so that such characters are converted to
3912 printable escape sequences.
3914 PH/41 $message_linecount is a new variable that contains the total number of
3915 lines in the message. Compare $body_linecount, which is the count for the
3918 PH/42 Exim no longer gives details of delivery errors for specific addresses in
3919 bounce and delay warning messages, except in certain special cases, which
3922 (a) An SMTP error message from a remote host;
3923 (b) A message specified in a :fail: redirection;
3924 (c) A message specified in a "fail" command in a system filter;
3925 (d) A message specified in a FAIL return from the queryprogram router;
3926 (e) A message specified by the cannot_route_message router option.
3928 In these cases only, Exim does include the error details in bounce and
3929 warning messages. There are also a few cases where bland messages such
3930 as "unrouteable address" or "local delivery error" are given.
3932 PH/43 $value is now also set for the "else" part of a ${run expansion.
3934 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
3935 being worked on, but at least Exim now implements the latest version to
3938 PH/45 In a pipe transport, although a timeout while waiting for the pipe
3939 process to complete was treated as a delivery failure, a timeout while
3940 writing the message to the pipe was logged, but erroneously treated as a
3941 successful delivery. Such timeouts include transport filter timeouts. For
3942 consistency with the overall process timeout, these timeouts are now
3943 treated as errors, giving rise to delivery failures by default. However,
3944 there is now a new Boolean option for the pipe transport called
3945 timeout_defer, which, if set TRUE, converts the failures into defers for
3946 both kinds of timeout. A transport filter timeout is now identified in
3949 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
3950 systems where "make" and "gmake" are different, calling "gmake" at top
3951 level broke things. I've arranged for the value of $(MAKE) to be passed
3952 from the Makefile to this script so that it can call the same version of
3956 A note about Exim versions 4.44 and 4.50
3957 ----------------------------------------
3959 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
3960 changes of various kinds. As a consequence, a big documentation update was
3961 needed. This delayed the release for rather longer than seemed good, especially
3962 in the light of a couple of (minor) security issues. Therefore, the changes
3963 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
3964 release. So 4.44 and 4.50 are in effect two different branches that both start
3967 I have left the 4.50 change log unchanged; it contains all the changes since
3968 4.43. The change log for 4.44 is below; many of its items are identical to
3969 those for 4.50. This seems to be the most sensible way to preserve the
3970 historical information.
3976 1. Minor wording change to the doc/README.SIEVE file.
3978 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
3979 computation of the current number of files was incorrect.
3981 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
3982 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
3985 4. Give more explanation in the error message when the command for a transport
3986 filter fails to execute.
3988 5. There are several places where Exim runs a non-Exim command in a
3989 subprocess. The SIGUSR1 signal should be disabled for these processes. This
3990 was being done only for the command run by the queryprogram router. It is
3991 now done for all such subprocesses. The other cases are: ${run, transport
3992 filters, and the commands run by the lmtp and pipe transports.
3994 6. Added CONFIGURE_GROUP build-time option.
3996 7. Some older OS have a limit of 256 on the maximum number of file
3997 descriptors. Exim was using setrlimit() to set 1000 as a large value
3998 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
3999 systems. I've change it so that if it can't get 1000, it tries for 256.
4001 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4002 was an oversight, and furthermore, ever since the addition of extra
4003 controls (e.g. 4.43/32), the checks on when to allow different forms of
4004 "control" were broken. There should now be diagnostics for all cases when a
4005 control that does not make sense is encountered.
4007 9. Added the /retain_sender option to "control=submission".
4009 10. $recipients is now available in the predata ACL (oversight).
4011 11. Tidy the search cache before the fork to do a delivery from a message
4012 received from the command line. Otherwise the child will trigger a lookup
4013 failure and thereby defer the delivery if it tries to use (for example) a
4014 cached ldap connection that the parent has called unbind on.
4016 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4017 of $address_data from the recipient verification was clobbered by the
4018 sender verification.
4020 13. The value of address_data from a sender verification is now available in
4021 $sender_address_data in subsequent conditions in the ACL statement.
4023 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
4025 15. Added a new option "connect=<time>" to callout options, to set a different
4028 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4029 was its contents. (It was OK if the option was not defined at all.)
4031 17. A "Completed" log line is now written for messages that are removed from
4032 the spool by the -Mrm option.
4034 18. New variables $sender_verify_failure and $recipient_verify_failure contain
4035 information about exactly what failed.
4037 19. Added -dd to debug only the daemon process.
4039 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
4040 handles timeouts, both on the server side and network timeouts. Renamed the
4041 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
4043 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
4044 It is now set to "smtps".
4046 22. $host_address is now set to the target address during the checking of
4047 ignore_target_hosts.
4049 23. When checking ignore_target_hosts for an ipliteral router, no host name was
4050 being passed; this would have caused $sender_host_name to have been used if
4051 matching the list had actually called for a host name (not very likely,
4052 since this list is usually IP addresses). A host name is now passed as
4055 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4056 code that specifies a non-restarting handler (typically sigaction() in
4057 modern systems) in an attempt to fix a rare and obscure crash bug.
4059 25. Narrowed the window for a race in the daemon that could cause it to ignore
4060 SIGCHLD signals. This is not a major problem, because they are used only to
4061 wake it up if nothing else does.
4063 26. A malformed maildirsize file could cause Exim to calculate negative values
4064 for the mailbox size or file count. Odd effects could occur as a result.
4065 The maildirsize information is now recalculated if the size or filecount
4068 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4069 support for a long time. Removed HAVE_SYS_VFS_H.
4071 28. Installed the latest version of exipick from John Jetmore.
4073 29. In an address list, if the pattern was not a regular expression, an empty
4074 subject address (from a bounce message) matched only if the pattern was an
4075 empty string. Non-empty patterns were not even tested. This was the wrong
4076 because it is perfectly reasonable to use an empty address as part of a
4077 database query. An empty address is now tested by patterns that are
4078 lookups. However, all the other forms of pattern expect the subject to
4079 contain a local part and a domain, and therefore, for them, an empty
4080 address still always fails if the pattern is not itself empty.
4082 30. Exim went into a mad DNS loop when attempting to do a callout where the
4083 host was specified on an smtp transport, and looking it up yielded more
4084 than one IP address.
4086 31. Re-factored the code for checking spool and log partition space into a
4087 function that finds that data and another that does the check. The former
4088 is then used to implement four new variables: $spool_space, $log_space,
4089 $spool_inodes, and $log_inodes.
4091 32. The RFC2047 encoding function was originally intended for short strings
4092 such as real names; it was not keeping to the 75-character limit for
4093 encoded words that the RFC imposes. It now respects the limit, and
4094 generates multiple encoded words if necessary. To be on the safe side, I
4095 have increased the buffer size for the ${rfc2047: expansion operator from
4098 33. It is now permitted to omit both strings after an "if" condition; if the
4099 condition is true, the result is "true". As before, when the second string
4100 is omitted, a false condition yields an empty string. This makes it less
4101 cumbersome to write custom ACL and router conditions.
4103 34. Failure to deliver a bounce message always caused it to be frozen, even if
4104 there was an errors_to setting on the router. The errors_to setting is now
4107 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4108 canonical form (fully expanded) before being placed in
4109 $sender_host_address.
4111 36. The table in the code that translates DNS record types into text (T_A to
4112 "A" for instance) was missing entries for NS and CNAME. It is just possible
4113 that this could have caused confusion if both these types were looked up
4114 for the same domain, because the text type is used as part of Exim's
4115 per-process caching. But the chance of anyone hitting this buglet seems
4118 37. The dnsdb lookup has been extended in a number of ways.
4120 (1) There is a new type, "zns", which walks up the domain tree until it
4121 finds some nameserver records. It should be used with care.
4123 (2) There is a new type, "mxh", which is like "mx" except that it returns
4124 just the host names, not the priorities.
4126 (3) It is now possible to give a list of domains (or IP addresses) to be
4127 looked up. The behaviour when one of the lookups defers can be
4128 controlled by a keyword.
4130 (4) It is now possible to specify the separator character for use when
4131 multiple records are returned.
4133 38. The dnslists ACL condition has been extended: it is now possible to supply
4134 a list of IP addresses and/or domains to be looked up in a particular DNS
4137 39. Added log_selector=+queue_time_overall.
4139 40. When running the queue in the test harness, wait just a tad after forking a
4140 delivery process, to get repeatability of debugging output.
4142 41. Include certificate and key file names in error message when GnuTLS fails
4143 to set them up, because the GnuTLS error message doesn't include the name
4144 of the failing file when there is a problem reading it.
4146 42. Allow both -bf and -bF in the same test run.
4148 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
4150 44. The "Exiscan patch" is now merged into the mainline Exim source.
4152 45. Sometimes the final signoff response after QUIT could fail to get
4153 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4154 < 0 before doing a fflush(). This bug looks as though it goes back to the
4155 introduction of TLS in release 3.20, but "sometimes" must have been rare
4156 because the tests only now provoked it.
4158 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
4159 (this can affect the format of dates).
4161 47. exim_tidydb, when checking for the continued existence of a message for
4162 which it has found a message-specific retry record, was not finding
4163 messages that were in split spool directories. Consequently, it was
4164 deleting retry records that should have stayed in existence.
4166 48. Steve fixed some bugs in eximstats.
4168 49. The SPA authentication driver was not abandoning authentication and moving
4169 on to the next authenticator when an expansion was forced to fail,
4170 contradicting the general specification for all authenticators. Instead it
4171 was generating a temporary error. It now behaves as specified.
4173 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
4174 (the order specifies the preference for clients). The order is now AES256,
4175 AES128, 3DES, ARCFOUR128.
4177 51. Small patch to Sieve code - explicitly set From: when generating an
4180 52. Exim crashed if a remote delivery caused a very long error message to be
4181 recorded - for instance if somebody sent an entire SpamAssassin report back
4182 as a large number of 550 error lines. This bug was coincidentally fixed by
4183 increasing the size of one of Exim's internal buffers (big_buffer) that
4184 happened as part of the Exiscan merge. However, to be on the safe side, I
4185 have made the code more robust (and fixed the comments that describe what
4188 53. Now that there can be additional text after "Completed" in log lines (if
4189 the queue_time_overall log selector is set), a one-byte patch to exigrep
4190 was needed to allow it to recognize "Completed" as not the last thing in
4193 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
4194 patch that reportedly fixes this has been added. I am not expert enough to
4195 create a test for it. This is what the patch creator wrote:
4197 "I found a little strange behaviour of ldap code when working with
4198 Windows 2003 AD Domain, where users was placed in more than one
4199 Organization Units. When I tried to give exim partial DN, the exit code
4200 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
4201 But simultaneously result of request was absolutely normal ldap result,
4202 so I produce this patch..."
4204 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
4205 REFERENCE, so I have modified the code to exclude the patch when that macro
4208 55. Some experimental protocols are using DNS PTR records for new purposes. The
4209 keys for these records are domain names, not reversed IP addresses. The
4210 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4211 leaves it alone. Component reversal etc. now happens only for IP addresses.
4214 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
4216 57. Double the size of the debug message buffer (to 2048) so that more of very
4217 long debug lines gets shown.
4219 58. The exicyclog utility now does better if the number of log files to keep
4220 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
4222 59. Two changes related to the smtp_active_hostname option:
4224 (1) $smtp_active_hostname is now available as a variable.
4225 (2) The default for smtp_banner uses $smtp_active_hostname instead
4226 of $primary_hostname.
4228 60. The host_aton() function is supposed to be passed a string that is known
4229 to be a valid IP address. However, in the case of IPv6 addresses, it was
4230 not checking this. This is a hostage to fortune. Exim now panics and dies
4231 if the condition is not met. A case was found where this could be provoked
4232 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4233 components; fortuitously, this particular loophole had already been fixed
4234 by change 4.50/55 above.
4236 If there are any other similar loopholes, the new check in host_aton()
4237 itself should stop them being exploited. The report I received stated that
4238 data on the command line could provoke the exploit when Exim was running as
4239 exim, but did not say which command line option was involved. All I could
4240 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4241 running as the user.
4244 61. There was a buffer overflow vulnerability in the SPA authentication code
4245 (which came originally from the Samba project). I have added a test to the
4246 spa_base64_to_bits() function which I hope fixes it.
4249 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
4250 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
4252 63. The daemon start-up calls getloadavg() while still root for those OS that
4253 need the first call to be done as root, but it missed one case: when
4254 deliver_queue_load_max is set with deliver_drop_privilege. This is
4255 necessary for the benefit of the queue runner, because there is no re-exec
4256 when deliver_drop_privilege is set.
4258 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
4259 This has been fixed.
4261 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4262 was in use, was not putting the data itself into the right store pool;
4263 consequently, it could be overwritten for a subsequent message in the same
4264 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4267 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
4269 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
4270 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
4271 the function were treating the return as a boolean value, which happened to
4272 work because 0=false and not-0=true, but is not correct code.
4274 68. The host_aton() function was not handling scoped IPv6 addresses (those
4275 with, for example, "%eth0" on the end) correctly.
4277 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
4278 negated items (that is, ~something) in unsigned ints. Some compilers
4279 apparently mutter when there is no cast.
4281 70. If an address verification called from an ACL failed, and did not produce a
4282 user-specific message (i.e. there was only a "system" message), nothing was
4283 put in $acl_verify_message. In this situation, it now puts the system
4286 71. Change 4.23/11 added synchronization checking at the start of an SMTP
4287 session; change 4.31/43 added the unwanted input to the log line - except
4288 that it did not do this in the start of session case. It now does.
4290 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
4291 This is wrong and can cause the other end to generate a synchronization
4292 error if it is another Exim or anything else that does the synchronization
4293 check. A QUIT command is no longer sent after a timeout.
4295 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
4296 during host lookups.
4298 74. The defer_ok option of callout verification was not working if it was used
4299 when verifying addresses in header lines, that is, for this case:
4301 verify = header_sender/callout=defer_ok
4303 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
4304 those file descriptors could be used for SMTP connections. If anything
4305 wrote to stderr (the example that came up was "warn" in embedded Perl), it
4306 could be sent to the SMTP client, causing chaos. The daemon now opens
4307 stdin, stdout, and stderr to /dev/null when it puts itself into the
4310 76. Arrange for output from Perl's "warn" command to be written to Exim's main
4311 log by default. The user can override this with suitable Perl magic.
4313 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
4314 the log message when discard triggers, was not working for the DATA ACL or
4315 for the non-SMTP ACL.
4317 78. Error message wording change in sieve.c.
4319 79. If smtp_accept_max_per_host was set, the number of connections could be
4320 restricted to fewer than expected, because the daemon was trying to set up
4321 a new connection before checking whether the processes handling previous
4322 connections had finished. The check for completed processes is now done
4323 earlier. On busy systems, this bug wouldn't be noticed because something
4324 else would have woken the daemon, and it would have reaped the completed
4327 80. If a message was submitted locally by a user whose login name contained one
4328 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
4329 It caused a spool format error. I have fixed the spool reading code. A
4330 related problem was that the "from" clause in the Received: line became
4331 illegal because of the space(s). It is now covered by ${quote_local_part.
4333 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
4336 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
4338 83. Patch from Sieve maintainer to fix unterminated string problem in
4339 "vacation" handling.
4341 84. Some minor changes to the Linux configuration files to help with other
4342 OS variants using glibc.
4344 85. One more patch for Sieve to update vacation handling to latest spec.
4347 ----------------------------------------------------
4348 See the note above about the 4.44 and 4.50 releases.
4349 ----------------------------------------------------
4355 1. Change 4.43/35 introduced a bug that caused file counts to be
4356 incorrectly computed when quota_filecount was set in an appendfile
4359 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4360 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4363 3. Give more explanation in the error message when the command for a transport
4364 filter fails to execute.
4366 4. There are several places where Exim runs a non-Exim command in a
4367 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4368 was being done only for the command run by the queryprogram router. It is
4369 now done for all such subprocesses. The other cases are: ${run, transport
4370 filters, and the commands run by the lmtp and pipe transports.
4372 5. Some older OS have a limit of 256 on the maximum number of file
4373 descriptors. Exim was using setrlimit() to set 1000 as a large value
4374 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4375 systems. I've change it so that if it can't get 1000, it tries for 256.
4377 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4378 was an oversight, and furthermore, ever since the addition of extra
4379 controls (e.g. 4.43/32), the checks on when to allow different forms of
4380 "control" were broken. There should now be diagnostics for all cases when a
4381 control that does not make sense is encountered.
4383 7. $recipients is now available in the predata ACL (oversight).
4385 8. Tidy the search cache before the fork to do a delivery from a message
4386 received from the command line. Otherwise the child will trigger a lookup
4387 failure and thereby defer the delivery if it tries to use (for example) a
4388 cached ldap connection that the parent has called unbind on.
4390 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4391 of $address_data from the recipient verification was clobbered by the
4392 sender verification.
4394 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4395 was its contents. (It was OK if the option was not defined at all.)
4397 11. A "Completed" log line is now written for messages that are removed from
4398 the spool by the -Mrm option.
4400 12. $host_address is now set to the target address during the checking of
4401 ignore_target_hosts.
4403 13. When checking ignore_target_hosts for an ipliteral router, no host name was
4404 being passed; this would have caused $sender_host_name to have been used if
4405 matching the list had actually called for a host name (not very likely,
4406 since this list is usually IP addresses). A host name is now passed as
4409 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4410 code that specifies a non-restarting handler (typically sigaction() in
4411 modern systems) in an attempt to fix a rare and obscure crash bug.
4413 15. Narrowed the window for a race in the daemon that could cause it to ignore
4414 SIGCHLD signals. This is not a major problem, because they are used only to
4415 wake it up if nothing else does.
4417 16. A malformed maildirsize file could cause Exim to calculate negative values
4418 for the mailbox size or file count. Odd effects could occur as a result.
4419 The maildirsize information is now recalculated if the size or filecount
4422 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4423 support for a long time. Removed HAVE_SYS_VFS_H.
4425 18. Updated exipick to current release from John Jetmore.
4427 19. Allow an empty sender to be matched against a lookup in an address list.
4428 Previously the only cases considered were a regular expression, or an
4431 20. Exim went into a mad DNS lookup loop when doing a callout where the
4432 host was specified on the transport, if the DNS lookup yielded more than
4435 21. The RFC2047 encoding function was originally intended for short strings
4436 such as real names; it was not keeping to the 75-character limit for
4437 encoded words that the RFC imposes. It now respects the limit, and
4438 generates multiple encoded words if necessary. To be on the safe side, I
4439 have increased the buffer size for the ${rfc2047: expansion operator from
4442 22. Failure to deliver a bounce message always caused it to be frozen, even if
4443 there was an errors_to setting on the router. The errors_to setting is now
4446 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4447 canonical form (fully expanded) before being placed in
4448 $sender_host_address.
4450 24. Updated eximstats to version 1.33
4452 25. Include certificate and key file names in error message when GnuTLS fails
4453 to set them up, because the GnuTLS error message doesn't include the name
4454 of the failing file when there is a problem reading it.
4456 26. Expand error message when OpenSSL has problems setting up cert/key files.
4459 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
4460 (this can affect the format of dates).
4462 28. exim_tidydb, when checking for the continued existence of a message for
4463 which it has found a message-specific retry record, was not finding
4464 messages that were in split spool directories. Consequently, it was
4465 deleting retry records that should have stayed in existence.
4467 29. eximstats updated to version 1.35
4468 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
4469 1.35 - bugfix such that pie charts by volume are generated correctly
4471 30. The SPA authentication driver was not abandoning authentication and moving
4472 on to the next authenticator when an expansion was forced to fail,
4473 contradicting the general specification for all authenticators. Instead it
4474 was generating a temporary error. It now behaves as specified.
4476 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
4477 (the order specifies the preference for clients). The order is now AES256,
4478 AES128, 3DES, ARCFOUR128.
4480 31. Small patch to Sieve code - explicitly set From: when generating an
4483 32. Exim crashed if a remote delivery caused a very long error message to be
4484 recorded - for instance if somebody sent an entire SpamAssassin report back
4485 as a large number of 550 error lines. This bug was coincidentally fixed by
4486 increasing the size of one of Exim's internal buffers (big_buffer) that
4487 happened as part of the Exiscan merge. However, to be on the safe side, I
4488 have made the code more robust (and fixed the comments that describe what
4491 33. Some experimental protocols are using DNS PTR records for new purposes. The
4492 keys for these records are domain names, not reversed IP addresses. The
4493 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4494 leaves it alone. Component reversal etc. now happens only for IP addresses.
4497 34. The host_aton() function is supposed to be passed a string that is known
4498 to be a valid IP address. However, in the case of IPv6 addresses, it was
4499 not checking this. This is a hostage to fortune. Exim now panics and dies
4500 if the condition is not met. A case was found where this could be provoked
4501 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4502 components; fortuitously, this particular loophole had already been fixed
4503 by change 4.50/55 or 4.44/33 above.
4505 If there are any other similar loopholes, the new check in host_aton()
4506 itself should stop them being exploited. The report I received stated that
4507 data on the command line could provoke the exploit when Exim was running as
4508 exim, but did not say which command line option was involved. All I could
4509 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4510 running as the user.
4513 35. There was a buffer overflow vulnerability in the SPA authentication code
4514 (which came originally from the Samba project). I have added a test to the
4515 spa_base64_to_bits() function which I hope fixes it.
4518 36. The daemon start-up calls getloadavg() while still root for those OS that
4519 need the first call to be done as root, but it missed one case: when
4520 deliver_queue_load_max is set with deliver_drop_privilege. This is
4521 necessary for the benefit of the queue runner, because there is no re-exec
4522 when deliver_drop_privilege is set.
4524 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4525 was in use, was not putting the data itself into the right store pool;
4526 consequently, it could be overwritten for a subsequent message in the same
4527 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4530 38. Sometimes the final signoff response after QUIT could fail to get
4531 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4532 < 0 before doing a fflush(). This bug looks as though it goes back to the
4533 introduction of TLS in release 3.20, but "sometimes" must have been rare
4534 because the tests only now provoked it.
4540 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
4541 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
4542 Now it returns an integer. A number of calls were still expecting a T/F
4543 return. Fortuitously, in all cases, the tests worked in OK situations,
4544 which is the norm. However, things would have gone wrong on any write
4545 failures on the smtp file descriptor. This function is used when sending
4546 messages over SMTP and also when doing verify callouts.
4548 2. When Exim is called to do synchronous delivery of a locally submitted
4549 message (the -odf or -odi options), it no longer closes stderr before doing
4552 3. Implemented the mua_wrapper option.
4554 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
4556 5. Implemented the functions header_remove(), header_testname(),
4557 header_add_at_position(), and receive_remove_recipient(), and exported them
4560 6. If an ACL "warn" statement specified the addition of headers, Exim already
4561 inserted X-ACL-Warn: at the start if there was no header name. However, it
4562 was not making this test for the second and subsequent header lines if
4563 there were newlines in the string. This meant that an invalid header could
4564 be inserted if Exim was badly configured.
4566 7. Allow an ACL "warn" statement to add header lines at the start or after all
4567 the Received: headers, as well as at the end.
4569 8. Added the rcpt_4xx retry error code.
4571 9. Added postmaster_mailfrom=xxx to callout verification option.
4573 10. Added mailfrom=xxxx to the callout verification option, for verify=
4576 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
4577 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
4579 12. Inserted some casts to stop certain compilers warning when using pointer
4580 differences as field lengths or precisions in printf-type calls (mostly
4581 affecting debugging statements).
4583 13. Added optional readline() support for -be (dynamically loaded).
4585 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
4586 same clock tick as a message's arrival, so that its received time was the
4587 same as the "first fail" time on the retry record, and that message
4588 remained on the queue past the ultimate address timeout, every queue runner
4589 would try a delivery (because it was past the ultimate address timeout) but
4590 after another failure, the ultimate address timeout, which should have then
4591 bounced the address, did not kick in. This was a "< instead of <=" error;
4592 in most cases the first failure would have been in the next clock tick
4593 after the received time, and all would be well.
4595 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
4596 being recognized when the domain list was tested by the match_domain
4597 condition in an expansion string.
4599 16. Added the ${str2b64: operator.
4601 17. Exim was always calling setrlimit() to set a large limit for the number of
4602 processes, without checking whether the existing limit was already
4603 adequate. (It did check for the limit on file descriptors.) Furthermore,
4604 errors from getrlimit() and setrlimit() were being ignored. Now they are
4605 logged to the main and panic logs, but Exim does carry on, to try to do its
4606 job under whatever limits there are.
4608 18. Imported PCRE 5.0.
4610 19. Trivial typo in log message " temporarily refused connection" (the leading
4613 20. If the log selector return_path_on_delivery was set and an address was
4614 redirected to /dev/null, the delivery process crashed because it assumed
4615 that a return path would always be set for a "successful" delivery. In this
4616 case, the whole delivery is bypassed as an optimization, and therefore no
4619 21. Internal re-arrangement: the function for sending a challenge and reading
4620 a response while authentication was assuming a zero-terminated challenge
4621 string. It's now changed to take a pointer and a length, to allow for
4622 binary data in such strings.
4624 22. Added the cyrus_sasl authenticator (code supplied by MBM).
4626 23. Exim was not respecting finduser_retries when seeking the login of the
4627 uid under which it was called; it was always trying 10 times. (The default
4628 setting of finduser_retries is zero.) Also, it was sleeping after the final
4629 failure, which is pointless.
4631 24. Implemented tls_on_connect_ports.
4633 25. Implemented acl_smtp_predata.
4635 26. If the domain in control=submission is set empty, Exim assumes that the
4636 authenticated id is a complete email address when it generates From: or
4637 Sender: header lines.
4639 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
4640 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
4641 chown and chgrp in /bin and hostname in /usr/bin.
4643 28. Exim was keeping the "process log" file open after each use, just as it
4644 does for the main log. This opens the possibility of it remaining open for
4645 long periods when the USR1 signal hits a daemon. Occasional processlog
4646 errors were reported, that could have been caused by this. Anyway, it seems
4647 much more sensible not to leave this file open at all, so that is what now
4650 29. The long-running daemon process does not normally write to the log once it
4651 has entered its main loop, and it closes the log before doing so. This is
4652 so that log files can straightforwardly be renamed and moved. However,
4653 there are a couple of unusual error situations where the daemon does write
4654 log entries, and I had neglected to close the log afterwards.
4656 30. The text of an SMTP error response that was received during a remote
4657 delivery was being truncated at 512 bytes. This is too short for some of
4658 the long messages that one sometimes sees. I've increased the limit to
4661 31. It is now possible to make retry rules that apply only when a message has a
4662 specific sender, in particular, an empty sender.
4664 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
4665 it possible to be selective about when SMTP synchronization is enforced.
4667 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
4669 32. Implemented hosts_connection_nolog.
4671 33. Added an ACL for QUIT.
4673 34. Setting "delay_warning=" to disable warnings was not working; it gave a
4676 35. Added mailbox_size and mailbox_filecount to appendfile.
4678 36. Added control = no_multiline_responses to ACLs.
4680 37. There was a bug in the logic of the code that waits for the clock to tick
4681 in the case where the clock went backwards by a substantial amount such
4682 that the microsecond fraction of "now" was more than the microsecond
4683 fraction of "then" (but the whole seconds number was less).
4685 38. Added support for the libradius Radius client library this is found on
4686 FreeBSD (previously only the radiusclient library was supported).
4692 1. When certain lookups returned multiple values in the form name=value, the
4693 quoting of the values was not always being done properly. Specifically:
4694 (a) If the value started with a double quote, but contained no whitespace,
4696 (b) If the value contained whitespace other than a space character (i.e.
4697 tabs or newlines or carriage returns) it was not quoted.
4698 This fix has been applied to the mysql and pgsql lookups by writing a
4699 separate quoting function and calling it from the lookup code. The fix
4700 should probably also be applied to nisplus, ibase and oracle lookups, but
4701 since I cannot test any of those, I have not disturbed their existing code.
4703 2. A hit in the callout cache for a specific address caused a log line with no
4704 reason for rejecting RCPT. Now it says "Previous (cached) callout
4705 verification failure".
4707 3. There was an off-by-one bug in the queryprogram router. An over-long
4708 return line was truncated at 256 instead of 255 characters, thereby
4709 overflowing its buffer with the terminating zero. As well as fixing this, I
4710 have increased the buffer size to 1024 (and made a note to document this).
4712 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
4713 when Exim is waiting for an SMTP response from a remote server, Exim
4714 restarts its select() call on the socket, thereby resetting its timeout.
4715 This is not a problem when such interrupts are rare. Somebody set up a cron
4716 job to run exiwhat every 2 minutes, which is less than the normal select()
4717 timeout (5 or 10 minutes). This meant that the select() timeout never
4718 kicked in because it was always reset. I have fixed this by comparing the
4719 time when an interrupt arrives with the time at the start of the first call
4720 to select(). If more time than the timeout has elapsed, the interrupt is
4721 treated as a timeout.
4723 5. Some internal re-factoring in preparation for the addition of Sieve
4724 extensions (by MH). In particular, the "personal" test is moved to a
4725 separate function, and given an option for scanning Cc: and Bcc: (which is
4726 not set for Exim filters).
4728 6. When Exim created an email address using the login of the caller as the
4729 local part (e.g. when creating a From: or Sender: header line), it was not
4730 quoting the local part when it contained special characters such as @.
4732 7. Installed new OpenBSD configuration files.
4734 8. Reworded some messages for syntax errors in "and" and "or" conditions to
4735 try to make them clearer.
4737 9. Callout options, other than the timeout value, were being ignored when
4738 verifying sender addresses in header lines. For example, when using
4740 verify = header_sender/callout=no_cache
4742 the cache was (incorrectly) being used.
4744 10. Added a missing instance of ${EXE} to the exim_install script; this affects
4745 only the Cygwin environment.
4747 11. When return_path_on_delivery was set as a log selector, if different remote
4748 addresses in the same message used different return paths and parallel
4749 remote delivery occurred, the wrong values would sometimes be logged.
4750 (Whenever a remote delivery process finished, the return path value from
4751 the most recently started remote delivery process was logged.)
4753 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
4754 lines when AUTH and/or TLS are in use. This is the "received protocol"
4755 field. Exim used to use "asmtp" for authenticated SMTP, without any
4756 indication (in the protocol name) for TLS use. Now it follows the RFC and
4757 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
4758 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
4759 names appear in log lines as well as in Received: header lines.
4761 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
4762 extensions, and comparison tests, and to fix some bugs.
4764 14. Changes to the "personal" filter test:
4766 (1) The test was buggy in that it was just doing the equivalent of
4767 "contains" tests on header lines. For example, if a user's address was
4768 anne@some.where, the "personal" test would incorrectly be true for
4770 To: susanne@some.where
4772 This test is now done by extracting each address from the header in turn,
4773 and checking the entire address. Other tests that are part of "personal"
4774 are now done using regular expressions (for example, to check local parts
4775 of addresses in From: header lines).
4777 (2) The list of non-personal local parts in From: addresses has been
4778 extended to include "listserv", "majordomo", "*-request", and "owner-*",
4779 taken from the Sieve specification recommendations.
4781 (3) If the message contains any header line starting with "List-" it is
4782 treated as non-personal.
4784 (4) The test for "circular" in the Subject: header line has been removed
4785 because it now seems ill-conceived.
4787 15. Minor typos in src/EDITME comments corrected.
4789 16. Installed latest exipick from John Jetmore.
4791 17. If headers_add on a router specified a text string that was too long for
4792 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
4793 of string_sprintf() is now avoided.
4795 18. $message_body_size was not set (it was always zero) when running the DATA
4796 ACL and the local_scan() function.
4798 19. For the "mail" command in an Exim filter, no default was being set for
4799 the once_repeat time, causing a random time value to be used if "once" was
4800 specified. (If the value happened to be <= 0, no repeat happened.) The
4801 default is now 0s, meaning "never repeat". The "vacation" command was OK
4802 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
4803 (I found it when inspecting the code).
4805 20. There is now an overall timeout for performing a callout verification. It
4806 defaults to 4 times the callout timeout, which applies to individual SMTP
4807 commands during the callout. The overall timeout applies when there is more
4808 than one host that can be tried. The timeout is checked before trying the
4809 next host. This prevents very long delays if there are a large number of
4810 hosts and all are timing out (e.g. when the network connections are timing
4811 out). The value of the overall timeout can be changed by specifying an
4812 additional sub-option for "callout", called "maxwait". For example:
4814 verify = sender/callout=5s,maxwait=20s
4816 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
4817 to the end before writing, but this should make it even safer).
4819 22. Exim was forgetting that it had advertised PIPELINING for the second and
4820 subsequent messages on an SMTP connection. It was also not resetting its
4821 memory on STARTTLS and an internal HELO.
4823 23. When Exim logs an SMTP synchronization error within a session, it now
4824 records whether PIPELINING has been advertised or not.
4826 24. Added 3 instances of "(long int)" casts to time_t variables that were being
4827 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
4828 rather than long int.
4830 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
4832 26. Added the never_mail option to autoreply.
4838 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
4839 crash if the getsockname() call failed; this can happen if a connection is
4840 closed very soon after it is established. The problem was simply in the
4841 order in which certain operations were done, causing Exim to try to write
4842 to the SMTP stream before it had set up the file descriptor. The bug has
4843 been fixed by making things happen in the correct order.
4849 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
4850 before the connection was closed, thus losing the rejection response.
4852 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
4853 some early Solaris releases, but causes trouble in current releases where
4854 socklen_t is defined.
4856 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
4859 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
4862 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
4863 that releasing the top of it at the end releases what was used for sub-
4864 expansions (unless the block got too big). However, discard this block if
4865 the first thing is a variable or header, so that we can use its block when
4866 it is dynamic (useful for very large $message_headers, for example).
4868 6. Lookups now cache *every* query, not just the most recent. A new, separate
4869 store pool is used for this. It can be recovered when all lookup caches are
4870 flushed. Lookups now release memory at the end of their result strings.
4871 This has involved some general refactoring of the lookup sources.
4873 7. Some code has been added to the store_xxx() functions to reduce the amount
4874 of flapping under certain conditions.
4876 8. log_incoming_interface used to affect only the <= reception log lines. Now
4877 it causes the local interface and port to be added to several more SMTP log
4878 lines, for example "SMTP connection from", and rejection lines.
4880 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
4882 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
4884 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
4885 could be overwritten at the end of the current message (or the start of a
4886 new message if it was set in a HELO ACL). The value is now preserved for
4887 the duration of the SMTP connection.
4889 12. If a transport had a headers_rewrite setting, and a matching header line
4890 contained an unqualified address, that address was qualified, even if it
4891 did not match any rewriting rules. The underlying bug was that the values
4892 of the flags that permit the existence of unqualified sender and recipient
4893 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
4894 non-local messages, and by -bnq for local messages) were not being
4895 preserved with the message after it was received.
4897 13. When Exim was logging an SMTP synchronization error, it could sometimes log
4898 "next input=" as part of the text comprising the host identity instead of
4899 the correct text. The code was using the same buffer for two different
4900 strings. However, depending on which order the printing function evaluated
4901 its arguments, the bug did not always show up. Under Linux, for example, my
4902 test suite worked just fine.
4904 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
4905 doesn't work with some older versions of Perl. It has been changed to "my",
4906 which in any case is probably the better facility to use.
4908 15. A really picky compiler found some instances of statements for creating
4909 error messages that either had too many or two few arguments for the format
4912 16. The size of the buffer for calls to the DNS resolver has been increased
4913 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
4914 for addresses that have a lot of PTR records. This alleviates a problem; it
4915 does not fully solve it.
4917 17. A dnsdb lookup for PTR records that receives more data than will fit in the
4918 buffer now truncates the list and logs the incident, which is the same
4919 action as happens when Exim is looking up a host name and its aliases.
4920 Previously in this situation something unpredictable would happen;
4921 sometimes it was "internal error: store_reset failed".
4923 18. If a server dropped the connection unexpectedly when an Exim client was
4924 using GnuTLS and trying to read a response, the client delivery process
4925 crashed while trying to generate an error log message.
4927 19. If a "warn" verb in an ACL added multiple headers to a message in a single
4928 string, for example:
4930 warn message = H1: something\nH2: something
4932 the text was added as a single header line from Exim's point of view
4933 though it ended up OK in the delivered message. However, searching for the
4934 second and subsequent header lines using $h_h2: did not work. This has been
4935 fixed. Similarly, if a system filter added multiple headers in this way,
4936 the routers could not see them.
4938 20. Expanded the error message when iplsearch is called with an invalid key to
4939 suggest using net-iplsearch in a host list.
4941 21. When running tests using -bh, any delays imposed by "delay" modifiers in
4942 ACLs are no longer actually imposed (and a message to that effect is
4945 22. If a "gecos" field in a passwd entry contained escaped characters, in
4946 particular, if it contained a \" sequence, Exim got it wrong when building
4947 a From: or a Sender: header from that name. A second bug also caused
4948 incorrect handling when an unquoted " was present following a character
4949 that needed quoting.
4951 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
4952 was not being matched caselessly.
4954 24. Arranged for all hyphens in the exim.8 source to be escaped with
4957 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
4958 properly. Recipient callout cache records were still being keyed to include
4959 the sender, even when use_sender was set false. This led to far more
4960 callouts that were necessary. The sender is no longer included in the key
4961 when use_sender is false.
4963 26. Added "control = submission" modifier to ACLs.
4965 27. Added the ${base62d: operator to decode base 62 numbers.
4967 28. dnsdb lookups can now access SRV records.
4969 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
4970 the configuration file.
4972 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
4973 (-v) mode. This makes the output for a verbose queue run more intelligible.
4975 31. Added a use_postmaster feature to recipient callouts.
4977 32. Added the $body_zerocount variable, containing the number of binary zero
4978 bytes in the message body.
4980 33. The time of last modification of the "new" subdirectory is now used as the
4981 "mailbox time last read" when there is a quota error for a maildir
4984 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
4986 35. Added +ignore_unknown as a special item in host lists.
4988 36. Code for decoding IPv6 addresses in host lists is now included, even if
4989 IPv6 support is not being compiled. This fixes a bug in which an IPv6
4990 address was recognized as an IP address, but was then not correctly decoded
4991 into binary, causing unexpected and incorrect effects when compared with
4998 1. Very minor rewording of debugging text in manualroute to say "list of
4999 hosts" instead of "hostlist".
5001 2. If verify=header_syntax was set, and a header line with an unqualified
5002 address (no domain) and a large number of spaces between the end of the
5003 name and the colon was received, the reception process suffered a buffer
5004 overflow, and (when I tested it) crashed. This was caused by some obsolete
5005 code that should have been removed. The fix is to remove it!
5007 3. When running in the test harness, delay a bit after writing a bounce
5008 message to get a bit more predictability in the log output.
5010 4. Added a call to search_tidyup() just before forking a reception process. In
5011 theory, someone could use a lookup in the expansion of smtp_accept_max_
5012 per_host which, without the tidyup, could leave open a database connection.
5014 5. Added the variables $recipient_data and $sender_data which get set from a
5015 lookup success in an ACL "recipients" or "senders" condition, or a router
5016 "senders" option, similar to $domain_data and $local_part_data.
5018 6. Moved the writing of debug_print from before to after the "senders" test
5021 7. Change 4.31/66 (moving the time when the Received: is generated) caused
5022 problems for message scanning, either using a data ACL, or using
5023 local_scan() because the Received: header was not generated till after they
5024 were called (in order to set the time as the time of reception completion).
5025 I have revised the way this works. The header is now generated after the
5026 body is received, but before the ACL or local_scan() are called. After they
5027 are run, the timestamp in the header is updated.
5033 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
5034 before starting a queue runner without re-exec. This happened only when
5035 deliver_drop_privilege was set or when the Exim user was set to root. The
5036 effect of the bug was that timeouts during subsequent deliveries caused
5037 crashes instead of being properly handled. The handler is now left at its
5038 default (and expected) setting.
5040 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
5041 message, again when deliver_drop_privilege is set or Exim is run as root.
5042 The bug described in (1) was not present in this case, but the tidying up
5043 of the other signals was missing. I have made the two cases consistent.
5045 3. The ignore_target_hosts setting on a manualroute router was being ignored
5046 for hosts that were looked up using the /MX notation.
5048 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
5051 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
5052 operated on the sender address. After changing the $sender_address to <>
5053 for the sender address verify, Exim was re-instated it as the original
5054 (before rewriting) address, but remembering that it had rewritten it, so it
5055 wasn't rewriting it again. This bug also had the effect of breaking the
5056 sender address verification caching when the sender address was rewritten.
5058 6. The ignore_target_hosts option was being ignored by the ipliteral router.
5059 This has been changed so that if the ip literal address matches
5060 ignore_target_hosts, the router declines.
5062 7. Added expansion conditions match_domain, match_address, and match_local_
5063 part (NOT match_host).
5065 8. The placeholder for the Received: header didn't have a length field set.
5067 9. Added code to Exim itself and to exim_lock to test for a specific race
5068 condition that could lead to file corruption when using MBX delivery. The
5069 issue is with the lockfile that is created in /tmp. If this file is removed
5070 after a process has opened it but before that process has acquired a lock,
5071 there is the potential for a second process to recreate the file and also
5072 acquire a lock. This could lead to two Exim processes writing to the file
5073 at the same time. The added code performs the same test as UW imapd; it
5074 checks after acquiring the lock that its file descriptor still refers to
5075 the same named file.
5077 10. The buffer for building added header lines was of fixed size, 8192 bytes.
5078 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
5081 11. Added the smtp_active_hostname option. If used, this will typically be made
5082 to depend on the incoming interface address. Because $interface_address is
5083 not set up until the daemon has forked a reception process, error responses
5084 that can happen earlier (such as "too many connections") no longer contain
5087 12. If an expansion in a condition on a "warn" statement fails because a lookup
5088 defers, the "warn" statement is abandoned, and the next ACL statement is
5089 processed. Previously this caused the whole ACL to be aborted.
5091 13. Added the iplsearch lookup type.
5093 14. Added ident_timeout as a log selector.
5095 15. Added tls_certificate_verified as a log selector.
5097 16. Added a global option tls_require_ciphers (compare the smtp transport
5098 option of the same name). This controls incoming TLS connections.
5100 17. I finally figured out how to make tls_require_ciphers do a similar thing
5101 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
5102 before starting the TLS session.
5104 18. Tabs are now shown as \t in -bP output.
5106 19. If the log selector return_path_on_delivery was set, Exim crashed when
5107 bouncing a message because it had too many Received: header lines.
5109 20. If two routers both had headers_remove settings, and the first one included
5110 a superfluous trailing colon, the final name in the first list and the
5111 first name in the second list were incorrectly joined into one item (with a
5112 colon in the middle).
5118 1. Added -C and -D options to the exinext utility, mainly to make it easier
5119 to include in the automated testing, but these could be helpful when
5120 multiple configurations are in use.
5122 2. The exinext utility was not formatting the output nicely when there was
5123 an alternate port involved in the retry record key, nor when there was a
5124 message id as well (for retries that were specific to a specific message
5125 and a specific host). It was also confused by IPv6 addresses, because of
5126 the additional colons they contain. I have fixed the IPv4 problem, and
5127 patched it up to do a reasonable job for IPv6.
5129 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
5130 delivery, the log line now contains "pipelined" if PIPELINING was used.
5132 4. An SMTP transport process used to panic and die if the bind() call to set
5133 an explicit outgoing interface failed. This has been changed; it is now
5134 treated in the same way as a connect() failure.
5136 5. A reference to $sender_host_name in the part of a conditional expansion
5137 that was being skipped was still causing a DNS lookup. This no longer
5140 6. The def: expansion condition was not recognizing references to header lines
5141 that used bh_ and bheader_.
5143 7. Added the _cache feature to named lists.
5145 8. The code for checking quota_filecount in the appendfile transport was
5146 allowing one more file than it should have been.
5148 9. For compatibility with Sendmail, the command line option
5156 and sets the incoming protocol and host name (for trusted callers). The
5157 host name and its colon can be omitted when only the protocol is to be set.
5158 Note the Exim already has two private options, -pd and -ps, that refer to
5159 embedded Perl. It is therefore impossible to set a protocol value of "d" or
5160 "s", but I don't think that's a major issue.
5162 10. A number of refactoring changes to the code, none of which should affect
5165 (a) The number of logging options was getting close to filling up the
5166 32-bit word that was used as a bit map. I have split them into two classes:
5167 those that are passed in the argument to log_write(), and those that are
5168 only ever tested independently outside of that function. These are now in
5169 separate 32-bit words, so there is plenty of room for expansion again.
5170 There is no change in the user interface or the logging behaviour.
5172 (b) When building, for example, log lines, the code previously used a
5173 macro that called string_cat() twice, in order to add two strings. This is
5174 not really sufficiently general. Furthermore, there was one instance where
5175 it was actually wrong because one of the argument was used twice, and in
5176 one call a function was used. (As it happened, calling the function twice
5177 did not affect the overall behaviour.) The macro has been replaced by a
5178 function that can join an arbitrary number of extra strings onto a growing
5181 (c) The code for expansion conditions now uses a table and a binary chop
5182 instead of a serial search (which was left over from when there were very
5183 few conditions). Also, it now recognizes conditions like "pam" even when
5184 the relevant support is not compiled in: a suitably worded error message is
5185 given if an attempt is made to use such a condition.
5187 11. Added ${time_interval:xxxxx}.
5189 12. A bug was causing one of the ddress fields not to be passed back correctly
5190 from remote delivery subprocesses. The field in question was not being
5191 subsequently used, so this caused to problems in practice.
5193 13. Added new log selectors queue_time and deliver_time.
5195 14. Might have fixed a bug in maildirsizefile handling that threw up
5196 "unexpected character" debug warnings, and recalculated the data
5197 unnecessarily. In any case, I expanded the warning message to give more
5200 15. Added the message "Restricted characters in address" to the statements in
5201 the default ACL that block characters like @ and % in local parts.
5203 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
5204 Three changes have been made:
5206 (a) There was a serious bug; a negative response to MAIL caused the whole
5207 recipient domain to be cached as invalid, thereby blocking all messages
5208 to all local parts at the same domain, from all senders. This bug has
5209 been fixed. The domain is no longer cached after a negative response to
5210 MAIL if the sender used is not empty.
5212 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
5215 (c) A new callout option, "use_sender" has been added for people who want
5216 the modified behaviour.
5222 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
5225 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
5226 indeed breaks things for older releases.
5228 3. Added additional logging to the case where there is a problem reading data
5229 from a filter that is running in a subprocess using a pipe, in order to
5230 try to track down a specific problem.
5232 4. Testing facility fudge: when running in the test harness and attempting
5233 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
5234 getting "No route to host". Convert this to a timeout.
5236 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
5239 6. Some OS don't have socklen_t but use size_t instead. This affects the
5240 fifth argument of getsockopt() amongst other things. This is now
5241 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
5242 can be set for individual OS. I have set it for SunOS5, OSF1, and
5243 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
5244 some earlier ones do not.
5246 7. Change 4.30/15 was not doing the test caselessly.
5248 8. The standard form for an IPv6 address literal was being rejected by address
5249 parsing in, for example, MAIL and RCPT commands. An example of this kind of
5250 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5251 this, as well as the form without the "IPv6" on the front (but only when
5252 address literals are enabled, of course).
5254 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
5256 10. Exim crashed if a message with an empty sender address specified by -f
5257 encountered a router with an errors_to setting. This could be provoked only
5258 by a command such as
5262 where an empty string was supplied; "<>" did not hit this bug.
5264 11. Installed PCRE release 4.5.
5266 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
5267 remained set. It is now erased.
5269 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
5270 times from message ids (which are base 36 rather than the normal 62).
5272 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
5273 were being counted as actual protocol errors, and logged if the log
5274 selector +smtp_protocol_error was set. One cannot be perfect in this test,
5275 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
5276 and DATA following a set of rejected RCPTs do not count as protocol errors.
5277 In other words, Exim assumes they were pipelined, though this may not
5278 actually be the case. Of course, in all cases the client gets an
5279 appropriate error code.
5281 15. If a lookup fails in an ACL condition, a message about the failure may
5282 be available; it is used if testing the ACL cannot continue, because most
5283 such messages specify what the cause of the deferral is. However, some
5284 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
5285 that caused an old message to be retained and used if a later statement
5286 caused a defer, replacing the real cause of the deferral.
5288 16. If an IP address had so many PTR records that the DNS lookup buffer
5289 was not large enough to hold them, Exim could crash while trying to process
5290 the truncated data. It now detects and logs this case.
5292 17. Further to 4.21/58, another change has been made: if (and only if) the
5293 first line of a message (the first header line) ends with CRLF, a bare LF
5294 in a subsequent header line has a space inserted after it, so as not to
5295 terminate the header.
5297 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
5298 unnecessarily, used atoi() instead of strtol(), and didn't check the
5299 termination when getting file sizes from file names by regex.
5301 19. Completely re-implemented the support for maildirsize files, in the light
5302 of a number of problems with the previous contributed implementation
5303 (4.30/29). In particular:
5305 . If the quota is zero, the maildirsize file is maintained, but no quota is
5308 . If the maildir directory does not exist, it is created before any attempt
5309 to write a maildirsize file.
5311 . The quota value in the file is just a cache; if the quota is changed in
5312 the transport, the new value overrides.
5314 . A regular expression is available for excluding directories from the
5317 20. The autoreply transport checks the characters in options that define the
5318 message's headers; it allows continued headers, but it was checking with
5319 isspace() after an embedded newline instead of explicitly looking for a
5322 21. If all the "regular" hosts to which an address was routed had passed their
5323 expiry times, and had not reached their retry times, the address was
5324 bounced, even if fallback hosts were defined. Now Exim should go on to try
5327 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
5328 equivalent code in the SMTP transport. Some hosts send humungous responses
5329 to HELO/EHLO, more than 1024 it seems.
5331 23. Refactoring: code in filter.c used (void *) for "any old type" but this
5332 gives compiler warnings in some environments. I've now done it "properly",
5335 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
5336 (because of problems with the built-in one) was declared to return uschar *
5337 instead of char *, causing compiler failure.
5339 25. Fixed a file descriptor leak when processing alias/forward files.
5341 26. Fixed a minor format string issue in dbfn.c.
5343 27. Typo in exim.c: ("dmbnz" for "dbmnz").
5345 28. If a filter file refered to $h_xxx or $message_headers, and the headers
5346 contained RFC 2047 "words", Exim's memory could, under certain conditions,
5349 29. When a sender address is verified, it is cached, to save repeating the test
5350 when there is more than one recipient in a message. However, when the
5351 verification involves a callout, it is possible for different callout
5352 options to be set for different recipients. It is too complicated to keep
5353 track of this in the cache, so now Exim always runs a verification when a
5354 callout is required, relying on the callout cache for the optimization.
5355 The overhead is duplication of the address routing, but this should not be
5358 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
5359 to be verified with callout=postmaster, and the main callout worked but the
5360 postmaster check failed, the verification correctly failed. However, if a
5361 subsequent RCPT command asked for sender verification *without* the
5362 postmaster check, incorrect caching caused this verification also to fail,
5365 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
5366 it was not caching the DNS options (qualify_single, search_parents) that
5367 were used when the lookup failed. A subsequent lookup with different
5368 options therefore always gave the same answer, though there were cases
5369 where it should not have. (Example: a "domains = !$mx_any" option on a
5370 dnslookup router: the "domains" option is always processed without any
5371 widening, but the router might have qualify_single set.) Now Exim uses the
5372 cached value only when the same options are set.
5374 32. Added John Jetmore's "exipick" utility to the distribution.
5376 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
5377 than a timeout (e.g. a certificate is required, and is not provided), an
5378 Exim server now closes the connection immediately. Previously it waited for
5379 the client to close - but if the client is SSL, it seems that they each
5380 wait for each other, leading to a delay before one of them times out.
5382 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
5383 maintained 0.8.x compatibility because I don't think many are using it, and
5384 it is clearly obsolete.
5386 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
5389 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
5390 client certificate was expired. A simple patch fixes this, though I don't
5391 understand the full logic of why the verify callback is called multiple
5394 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
5395 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
5396 which causes problems with some clients (such as the Certicom SSL Plus
5397 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
5398 disables the coutermeasure allowing Eudora to connect."
5400 38. Exim was not checking that a write() to a log file succeeded. This could
5401 lead to Bad Things if a log got too big, in particular if it hit a file
5402 size limit. Exim now panics and dies if it cannot write to a log file, just
5403 as it does if it cannot open a log file.
5405 39. Modified OS/Makefile-Linux so that it now contains
5407 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
5409 The two -D definitions ensure that Exim is compiled with large file
5410 support, which makes it possible to handle log files that are bigger than
5413 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
5414 instance) a domain was checked against a named list that involved a lookup,
5415 causing $domain_data to be set, then another domain was checked against the
5416 same list, then the first domain was re-checked, the value of $domain_data
5417 after the final check could be wrong. In particular, if the second check
5418 failed, it could be set empty. This bug probably also applied to
5421 41. The strip_trailing_dot option was not being applied to the address given
5422 with the -f command-line option.
5424 42. The code for reading a message's header from the spool was incrementing
5425 $received_count, but never initializing it. This meant that the value was
5426 incorrect (doubled) while delivering a message in the same process in which
5427 it was received. In the most common configuration of Exim, this never
5428 happens - a fresh exec is done - but it can happen when
5429 deliver_drop_privilege is set.
5431 43. When Exim logs an SMTP synchronization error - client data sent too soon -
5432 it now includes up to 150 characters of the unexpected data in the log
5435 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
5436 and building data strings. The size of both of these buffers was 10 000
5437 bytes - far larger than anybody would *ever* want, thought I. Needless to
5438 say, somebody hit the limit. I have increased the maximum line length to
5439 20 000 and the maximum data length of concatenated lines to 100 000. I have
5440 also fixed two bugs, because there was no checking on these buffers. Tsk,
5441 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
5442 buffer is too small.
5444 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
5445 lsearch lookups. Now it does.
5447 46. When parsing a route_list item in a manualroute router, a fixed-length
5448 buffer was used for the list of hosts. I made this 1024 bytes long,
5449 thinking that nobody would ever have a list of hosts that long. Wrong.
5450 Somebody had a whole pile of complicated expansion conditions, and the
5451 string was silently truncated, leading to an expansion error. It turns out
5452 that it is easier to change to an unlimited length (owing to other changes
5453 that have happened since this code was originally written) than to build
5454 structure for giving a limitation error. The length of the item that
5455 expands into the list of hosts is now unlimited.
5457 47. The lsearch lookup could not handle data where the length of text line was
5458 more than 4095 characters. Such lines were truncated, leading to shortened
5459 data being returned. It should now handle lines of any length.
5461 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
5462 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
5465 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
5466 replacing strings such as BIN_DIRECTORY in a source file: the replacement
5467 no longer happens in comment lines. A list of replacements is now placed
5468 at the head of all of the source files, except those whose only change is
5469 to replace PERL_COMMAND in the very first #! line.
5471 50. Replaced the slow insertion sort in queue.c, for sorting the list of
5472 messages on the queue, with a bottom-up merge sort, using code contributed
5473 by Michael Haardt. This should make operations like -bp somewhat faster on
5474 large queues. It won't affect queue runners, except when queue_run_in_order
5477 51. Installed eximstats 1.31 in the distribution.
5479 52. Added support for SRV lookups to the dnslookup router.
5481 53. If an ACL referred to $message_body or $message_body_end, the value was not
5482 reset for any messages that followed in the same SMTP session.
5484 54. The store-handling optimization for building very long strings was not
5485 differentiating between the different store pools. I don't think this
5486 actually made any difference in practice, but I've tidied it.
5488 55. While running the routers to verify a sender address, $sender_address
5489 was still set to the sender address. This is wrong, because when routing to
5490 send a bounce to the sender, it would be empty. Therefore, I have changed
5491 it so that, while verifying a sender address, $sender_address is set to <>.
5492 (There is no change to what happens when verifying a recipient address.)
5494 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
5495 target A or AAAA records (if not already returned) without resetting the
5496 qualify_single or search_parents options of the DNS resolver. These are
5497 inappropriate in this case because the targets of MX and SRV records must
5498 be FQDNs. A broken DNS record could cause trouble if it happened to have a
5499 target that, when qualified, matched something in the local domain. These
5500 two options are now turned off when doing these lookups.
5502 57. It seems that at least some releases of Reiserfs (which does not have the
5503 concept of a fixed number of inodes) returns zero and not -1 for the
5504 number of available inodes. This interacted badly with check_spool_inodes,
5505 which assumed that -1 was the "no such thing" setting. What I have done is
5506 to check that the total number of inodes is greater than zero before doing
5507 the test of how many are available.
5509 58. When a "warn" ACL statement has a log_message modifier, the message is
5510 remembered, and not repeated. This is to avoid a lot of repetition when a
5511 message has many recipients that cause the same warning to be written.
5512 Howewer, Exim was preserving the list of already written lines for an
5513 entire SMTP session, which doesn't seem right. The memory is now reset if a
5514 new message is started.
5516 59. The "rewrite" debugging flag was not showing the result of rewriting in the
5517 debugging output unless log_rewrite was also set.
5519 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
5520 of (int)(handle) when we know that handle contains (void *)(-1).
5522 61. The Exim daemon panic-logs an error return when it closes the incoming
5523 connection. However "connection reset by peer" seems to be common, and
5524 isn't really an error worthy of noting specially, so that particular error
5527 62. When Exim is trying to find all the local interfaces, it used to panic and
5528 die if the ioctl to get the interface flags failed. However, it seems that
5529 on at least one OS (Solaris 9) it is possible to have an interface that is
5530 included in the list of interfaces, but for which you get a failure error
5531 for this call. This happens when the interface is not "plumbed" into a
5532 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
5533 failure of the "get flags" call assumes that the interface is down.
5535 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
5536 makes life easier for people who are doing arithmetic on fields extracted
5537 from dates, where you often get leading zeros that should not be
5538 interpreted as octal.
5540 64. Added qualify_domain to the redirect router, to override the global
5543 65. If a pathologically long header line contained very many addresses (the
5544 report of this problem mentioned 10 000) and each of them was rewritten,
5545 Exim could use up a very large amount of memory. (It kept on making new
5546 copies of the header line as it rewrote, and never released the old ones.)
5547 At the expense of a bit more processing, the header rewriting function has
5548 been changed so that it no longer eats memory in this way.
5550 66. The generation of the Received: header has been moved from the time that a
5551 message starts to be received, to the time that it finishes. The timestamp
5552 in the Received: header should now be very close to that of the <= log
5553 line. There are two side-effects of this change:
5555 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
5556 logged header lines no longer include the local Received: line, because
5557 it has not yet been created. The same applies to a copy of the message
5558 that is returned to a non-SMTP sender when a message is rejected.
5560 (b) When a filter file is tested using -bf, no additional Received: header
5561 is added to the test message. After some thought, I decided that this
5564 This change does not affect the value of $received_for. It is still set
5565 after address rewriting, but before local_scan() is called.
5567 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
5569 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
5570 gave an unhelpful panic error message, and a defer error. I have managed to
5571 change this behaviour so that it now rejects any supplied certificate,
5572 which seems right, as the list of acceptable certificates is empty.
5574 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
5575 gave an unhelpful defer error. I have not managed to make this reject any
5576 supplied certificates, but the error message it gives is "no certificate
5577 supplied", which is not helpful.
5579 70. exigrep's output now also includes lines that are not associated with any
5580 message, but which match the given pattern. Implemented by a patch from
5581 Martin Sluka, which also tidied up the Perl a bit.
5583 71. Recipient callout verification, like sender verification, was using <> in
5584 the MAIL FROM command. This isn't really the right thing, since the actual
5585 sender may affect whether the remote host accepts the recipient or not. I
5586 have changed it to use the actual sender in the callout; this means that
5587 the cache record is now keyed on a recipient/sender pair, not just the
5588 recipient address. There doesn't seem to be a real danger of callout loops,
5589 since a callout by the remote host to check the sender would use <>.
5590 [SEE ABOVE: changed after hitting problems.]
5592 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
5593 temporary errors. However, in the case of such a code being given after
5594 the end of a data transmission (i.e. after ".") Exim was failing to write
5595 a retry record for the message. (Yes, there was some broken host that was
5596 actually sending 8xx at this point.)
5598 73. An unknown lookup type in a host list could cause Exim to panic-die when
5599 the list was checked. (An example that provoked this was putting <; in the
5600 middle of a list instead of at the start.) If this happened during a DATA
5601 ACL check, a -D file could be left lying around. This kind of configuration
5602 error no longer causes Exim to die; instead it causes a defer errror. The
5603 incident is still logged to the main and panic logs.
5605 74. Buglet left over from Exim 3 conversion. The message "too many messages
5606 in one connection" was written to the rejectlog but not the mainlog, except
5607 when address rewriting (yes!) was being logged.
5609 75. Added write_rejectlog option.
5611 76. When a system filter was run not as root (that is, when system_filter_user
5612 was set), the values of the $n variables were not being returned to the
5613 main process; thus, they were not subsequently available in the $sn
5616 77. Added +return_path_on_delivery log selector.
5618 78. A connection timeout was being treated differently from recipients deferred
5619 when testing hosts_max_try with a message that was older than the host's
5620 retry timeout. (The host should not be counted, thus allowing all hosts to
5621 be tried at least once before bouncing.) This may have been the cause of an
5622 occasionally reported bug whereby a message would remain on the queue
5623 longer than the retry timeout, but would be bounced if a delivery was
5624 forced. I say "may" because I never totally pinned down the problem;
5625 setting up timeout/retry tests is difficult. See also the next item.
5627 79. The ultimate address timeout was not being applied to errors that involved
5628 a combination of host plus message (for example, a timeout on a MAIL
5629 command). When an address resolved to a number of possible hosts, and they
5630 were not all tried for each delivery (e.g. because of hosts_max_try), a
5631 message could remain on the queue longer than the retry timeout.
5633 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
5636 81. Fixed an obscure SMTP outgoing bug which required at least the following
5637 conditions: (a) there was another message waiting for the same server;
5638 (b) the server returned 5xx to all RCPT commands in the first message so
5639 that the message was not completed; (c) the server dropped the connection
5640 or gave a negative response to the RSET that Exim sends to abort the
5641 transaction. The observed case was a dropped connection after DATA that had
5642 been sent in pipelining mode. That is, the server had advertised PIPELINING
5643 but was not implementing it correctly. The effect of the bug was incorrect
5644 behaviour, such as trying another host, and this could lead to a crash.
5650 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
5651 and daemon.c were passed as pointers to ints; they should have been
5652 pointers to socklen_t variables (which are typically unsigned ints).
5654 2. Some signed/unsigned type warnings in the os.c file for Linux have been
5657 3. Fixed a really odd bug that affected only the testing scheme; patching a
5658 certain fixed string in the binary changed the value of another string that
5659 happened to be identical to the end of the original first string.
5661 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
5662 name", it returns that address as the IP address. On some operating
5663 systems (e.g. Solaris), it also passes back the IP address string as the
5664 "host name". However, on others (e.g. Linux), it passes back an empty
5665 string. Exim wasn't checking for this, and was changing the host name to an
5666 empty string, assuming it had been canonicized.
5668 5. Although rare, it is permitted to have more than one PTR record for a given
5669 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
5670 all the names associated with an address, because they do in Solaris.
5671 However, it seems that they do not in Linux for data that comes from the
5672 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
5673 I found this out when I moved to a new Linux workstation and tried to run
5674 the Exim test suite.
5676 To get round this problem I have changed the code so that it now does its
5677 own call to the DNS to look up PTR records when searching for a host name.
5678 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
5679 addresses that are only in /etc/hosts are still found.
5681 This behaviour is, however, controlled by an option called host_lookup_
5682 order, which defaults to "bydns:byaddr". If people want to use the other
5683 order, or indeed, just use one or the other means of lookup, they can
5684 specify it in this variable.
5686 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
5687 some operating systems, this comes back from gethostbyaddr() as an empty
5688 string, and this is what Exim used to test for. However, it seems that in
5689 other systems, "." is yielded. Exim now tests for this case too.
5691 7. The values of check_spool_space and check_log_space are now held internally
5692 as a number of kilobytes instead of an absolute number of bytes. If a
5693 numbers is specified without 'K' or 'M', it is rounded up to the nearest
5694 kilobyte. This means that much larger values can be stored.
5696 8. Exim monitor: an attempt to get the action menu when not actually pointing
5697 at a message produces an empty menu entitled "No message selected". This
5698 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
5699 no entries in it ("Shell widget menu has zero width and/or height"). So I
5700 have added a single, blank menu entry in this case.
5702 9. Added ${quote_local_part.
5704 10. MIME decoding is now applied to the contents of Subject: header lines when
5707 11. Now that a reference to $sender_host_address automatically causes a reverse
5708 lookup to occur if necessary (4.13/18), there is no need to arrange for a
5709 host lookup before query-style lookups in lists that might use this
5710 variable. This has therefore been abolished, and the "net-" prefix is no
5711 longer necessary for query-style lookups.
5713 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
5714 have been a typo for LFLAGS, so it has been changed.
5716 13. The install script calls Exim with "-C /dev/null" in order to find the
5717 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
5718 to be output. Howeve, since Exim outputs its version number before the
5719 error, it didn't break the script. It just looked ugly. I fixed this by
5720 always allowing "-C /dev/null" if the caller is root.
5722 14. Ignore overlarge ACL variable number when reading spool file - insurance
5723 against a later release with more variables having written the file.
5725 15. The standard form for an IPv6 address literal was being rejected by EHLO.
5726 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5727 this, as well as the form without the "IPv6" on the front.
5729 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
5730 OS/Makefile-Darwin file.
5732 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
5733 only to LDAP libraries that do not have LDAP_OPT_DEREF.
5735 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
5736 variable. However, some OS use ints for inodes. I've added cast to long int
5737 to get rid of the compiler warning.
5739 19. I had forgotten to lock out "/../" in configuration file names when
5740 ALT_CONFIG_PREFIX was set.
5742 20. Routers used for verification do not need to specify transports. However,
5743 if such a router generated a host list, and callout was configured, Exim
5744 crashed, because it could not find a port number from the (non-existent)
5745 transport. It now assumes port 25 in this circumstance.
5747 21. Added the -t option to exigrep.
5749 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
5750 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
5751 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
5752 them is set, LOOKUP_LSEARCH is forced.
5754 23. "exim -bV" now outputs a list of lookups that are included in the binary.
5756 24. Added sender and host information to the "rejected by local_scan()" log
5757 line; previously there was no indication of these.
5759 25. Added .include_if_exists.
5761 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
5762 Linux. It turns out that not all file systems support this. Apparently some
5763 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
5764 do it.) To cope with this, the error EINVAL, which means that sync-ing is
5765 not supported on the file descriptor, is now ignored when Exim is trying to
5766 sync a directory. This applies only to Linux.
5768 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
5770 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
5771 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
5774 29. Added support for maildirsize files from supplied patch (modified a bit).
5776 30. The use of :fail: followed by an empty string could lead Exim to respond to
5777 sender verification failures with (e.g.):
5779 550 Verification failed for <xxx>
5780 550 Sender verify failed
5782 where the first response line was missing the '-' that indicates it is not
5783 the final line of the response.
5785 31. The loop for finding the name of the user that called Exim had a hardwired
5786 limit of 10; it now uses the value of finduser_retries, which is used for
5787 all other user lookups.
5789 32. Added $received_count variable, available in data and not_smtp ACLs, and at
5792 33. Exim was neglecting to zero errno before one call of strtol() when
5793 expanding a string and expecting an integer value. On some systems this
5794 resulted in spurious "integer overflow" errors. Also, it was casting the
5795 result into an int without checking.
5797 34. Testing for a connection timeout using "timeout_connect" in the retry rules
5798 did not work. The code looks as if it has *never* worked, though it appears
5799 to have been documented since at least releast 1.62. I have made it work.
5801 35. The "timeout_DNS" error in retry rules, also documented since at least
5802 1.62, also never worked. As it isn't clear exactly what this means, and
5803 clearly it isn't a major issue, I have abolished the feature by treating it
5804 as "timeout", and writing a warning to the main and panic logs.
5806 36. The display of retry rules for -brt wasn't always showing the error code
5809 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
5810 timeout_connect_A, timeout_connect_MX.
5812 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
5813 to the empty sender.
5815 39. The daemon was not analysing the content of -oX till after it had closed
5816 stderr and disconnected from the controlling terminal. This meant that any
5817 syntax errors were only noted on the panic log, and the return code from
5818 the command was 0. By re-arranging the code a little, I've made the
5819 decoding happen first, so such errors now appear on stderr, and the return
5820 code is 1. However, the actual setting up of the sockets still happens in
5821 the disconnected process, so errors there are still only recorded on the
5824 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
5825 connections (as happens on some IP stacks) was logged at start up time as
5826 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
5827 it from "IPv6 and IPv4", which means that two separate sockets are being
5830 41. The debug output for gethostbyname2() or getipnodebyname() failures now
5831 says whether AF_INET or AF_INET6 was passed as an argument.
5833 42. Exiwhat output was messed up when time zones were included in log
5836 43. Exiwhat now gives more information about the daemon's listening ports,
5837 and whether -tls-on-connect was used.
5839 44. The "port" option of the smtp transport is now expanded.
5841 45. A "message" modifier in a "warn" statement in a non-message ACL was being
5842 silently ignored. Now an error message is written to the main and panic
5845 46. There's a new ACL modifier called "logwrite" which writes to a log file
5846 as soon as it is encountered.
5848 47. Added $local_user_uid and $local_user_gid at routing time.
5850 48. Exim crashed when trying to verify a sender address that was being
5853 49. Exim was recognizing only a space character after ".include". It now also
5854 recognizes a tab character.
5856 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
5857 extracting the relevant information from the specification. The man page no
5858 longer contains scrambled data for the -d option, and I've added a section
5859 at the front about calling Exim under different names.
5861 51. Added "extra_headers" argument to the "mail" command in filter files.
5863 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
5866 53. Installed eximstats 1.29.
5868 54. Added transport_filter_timeout as a generic transport option.
5870 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
5871 Cc: header lines. This was required by RFC 822, but it not required by RFC
5874 56. Exim used to add From:, Date:, and Message-Id: header lines to any
5875 incoming messages that did not have them. Now it does so only if the
5876 message originates locally, that is, if there is no associated remote host
5877 address. When Resent- header lines are present, this applies to the Resent-
5878 lines rather than the non-Resent- lines.
5880 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
5881 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
5883 58. Messages for configuration errors now include the name of the main
5884 configuration file - useful now that there may be more than one file in a
5885 list (.included file names were always shown).
5887 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
5888 for those rare installations that do not start the daemon as root or run it
5889 setuid root. I've cut out the call to initgroups() if the daemon is not
5892 60. The Exim user and group can now be bound into the binary as text strings
5893 that are looked up at the start of Exim's processing.
5895 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
5897 62. Added $mailstore_basename variable.
5899 63. Installed patch to sieve.c from Michael Haardt.
5901 64. When Exim failed to open the panic log after failing to open the main log,
5902 the original message it was trying to log was written to stderr and debug
5903 output, but if they were not available (the usual case in production), it
5904 was lost. Now it is written to syslog before the two lines that record the
5905 failures to open the logs.
5907 65. Users' Exim filters run in subprocesses under the user's uid. It is
5908 possible for a "deliver" command or an alias in a "personal" command to
5909 provoke an address rewrite. If logging of address rewriting is configured,
5910 this fails because the process is not running as root or exim. There may be
5911 a better way of dealing with this, but for the moment (because 4.30 needs
5912 to be released), I have disabled address rewrite logging when running a
5913 filter in a non-root, non-exim process.
5919 1. The buildconfig auxiliary program wasn't quoting the value set for
5920 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
5921 not defined. This bug was masked in 4.22 by the effect that was fixed in
5924 2. Some messages that were rejected after a message id was allocated were
5925 shown as "incomplete" by exigrep. It no longer does this for messages that
5926 are rejected by local_scan() or the DATA or non-SMTP ACLs.
5928 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
5929 have allow_domain_literals set, the ID did not get logged in the <= line.
5930 Domain literals are now always recognized in Message-ID: header lines.
5932 4. The first argument for a ${extract expansion item is the key name or field
5933 number. Leading and trailing spaces in this item were not being ignored,
5934 causing some misleading effects.
5936 5. When deliver_drop_privilege was set, single queue runner processes started
5937 manually (i.e. by the command "exim -q") or by the daemon (which uses the
5938 same command in the process it spins off) were not dropping privilege.
5940 6. When the daemon running as "exim" started a queue runner, it always
5941 re-executed Exim in the spun-off process. This is a waste of effort when
5942 deliver_drop_privilege is set. The new process now just calls the
5943 queue-runner function directly.
5949 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
5952 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
5953 ignored. Though the use of -oP was forcing the writing of a pid file, it
5954 was always written to the default place.
5956 3. If the message "no IP address found for host xxxx" is generated during
5957 incoming verification, it is now followed by identification of the incoming
5958 connection (so you can more easily find what provoked it).
5960 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
5962 5. Added some features to "harden" Exim a bit more against certain attacks:
5964 (a) There is now a build-time option called FIXED_NEVER_USERS that can
5965 be put in Local/Makefile. This is like the never_users runtime option,
5966 but it cannot be overridden. The default setting is "root".
5968 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
5969 prefix string with which any file named in a -C command line option
5972 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
5973 is retained for -C and -D only if the caller of Exim is root. Without
5974 it, the exim user may also use -C and -D and retain privilege.
5976 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
5977 command line option is disabled.
5979 6. Macro names set by the -D option must start with an upper case letter, just
5980 like macro names defined in the configuration file.
5982 7. Added "dereference=" facility to LDAP.
5984 8. Two instances of the typo "uknown" in the source files are fixed.
5986 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
5987 the Configure-Makefile script screwed up while processing it.
5989 10. Incorporated PCRE 4.4.
5991 11. The SMTP synchronization check was not operating right at the start of an
5992 SMTP session. For example, it could not catch a HELO sent before the client
5993 waited for the greeting. There is now a check for outstanding input at the
5994 point when the greeting is written. Because of the duplex, asynchronous
5995 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
5996 way, but not yet received, when the check is performed.
5998 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
5999 on TCP/IP sockets, because this apparently causes some broken clients to
6002 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
6003 unchanged) from the Cygwin maintainer.
6005 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
6006 was supported instead of testing for mbx. Effectively a typo.
6008 15. The spa authenticator server code was not checking that the input it
6009 received was valid base64.
6011 16. The debug output line for the "set" modifier in ACLs was not showing the
6012 name of the variable that was being set.
6014 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
6016 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
6017 lookup on its own. Something else was needed to trigger the lookup. For
6018 example, a match in host_lookup or the need for a host name in a host list.
6019 Now, if $sender_host_name is referenced and the host name has not yet been
6020 looked up, a lookup is performed. If the lookup fails, the variable remains
6021 empty, and $host_lookup_failed is set to "1".
6023 19. Added "eqi" as a case-independent comparison operator.
6025 20. The saslauthd authentication condition could segfault if neither service
6026 nor realm was specified.
6028 21. If an overflowing value such as "2048M" was set for message_size_limit, the
6029 error message that was logged was misleading, and incoming SMTP
6030 connections were dropped. The message is now more accurate, and temporary
6031 errors are given to SMTP connections.
6033 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
6034 (except RSET) with a 421 error, until QUIT is received. However, it was
6035 failing to send a response to QUIT.
6037 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
6038 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
6039 ACL is now run after the helo_try_verify_hosts code.
6041 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
6042 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
6043 case-independent, so other case variants are also recognized). Apparently
6044 some systems use these upper case variants.
6046 25. If more than two messages were waiting for the same host, and a transport
6047 filter was specified for the transport, Exim sent two messages over the
6048 same TCP/IP connection, and then failed with "socket operation on non-
6049 socket" when it tried to send the third.
6051 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
6053 27. The extern definition of crypt16() in expand.c was not being excluded when
6054 the OS had its own crypt16() function.
6056 28. Added bounce_return_body as a new option, and bounce_return_size_limit
6057 as a preferred synonym for return_size_limit, both as an option and as an
6060 29. Added LIBS=-liconv to OS/Makefile-OSF1.
6062 30. Changed the default configuration ACL to relax the local part checking rule
6063 for addresses that are not in any local domains. For these addresses,
6064 slashes and pipe symbols are allowed within local parts, but the sequence
6065 /../ is explicitly forbidden.
6067 31. SPA server authentication was not clearing the challenge buffer before
6070 32. log_message in a "warn" ACL statement was writing to the reject log as
6071 well as to the main log, which contradicts the documentation and doesn't
6072 seem right (because no rejection is happening). So I have stopped it.
6074 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
6075 However, I am unable to do any testing of this.
6077 34. Fixed an infelicity in the appendfile transport. When checking directories
6078 for a mailbox, to see if any needed to be created, it was accidentally
6079 using path names with one or more superfluous leading slashes; tracing
6080 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
6082 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
6083 contents are added to the log line that is written for every discarded
6084 recipient. (Previously a log_message setting was ignored.)
6086 36. The ${quote: operator now quotes the string if it is empty.
6088 37. The install script runs exim in order to find its version number. If for
6089 some reason other than non-existence or emptiness, which it checks, it
6090 could not run './exim', it was installing it with an empty version number,
6091 i.e. as "exim-". This error state is now caught, and the installation is
6094 38. An argument was missing from the function that creates an error message
6095 when Exim fails to connect to the socket for saslauthd authentication.
6096 This could cause Exim to crash, or give a corrupted message.
6098 39. Added isip, isip4, and isip6 to ${if conditions.
6100 40. The ACL variables $acl_xx are now saved with the message, and can be
6101 accessed later in routers, transports, and filters.
6103 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
6104 strings in the file are not string-expanded.
6106 42. If a MAIL command specified a SIZE value that was too large to fit into an
6107 int variable, the check against message_size_limit failed. Such values are
6108 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
6109 day this will have to be increased, but I don't think I want to be around
6110 when emails are that large.
6117 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
6118 iconv() is not standard in FreeBSD.
6120 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
6121 IPv6 enabled. The observed symptom was a segmentation fault on return from
6122 the function os_common_find_running_interfaces() in src/os.c.
6124 3. In the check_special_case() function in daemon.c I had used "errno" as an
6125 argument name, which causes warnings on some systems. This was basically a
6126 typo, since it was named "eno" in the comments!
6128 4. The code that waits for the clock to tick (at a resolution of some fraction
6129 of a second) so as to ensure message-id uniqueness was always waiting for
6130 at least one whole tick, when it could have waited for less. [This is
6131 almost certainly not relevant at current processor speeds, where it is
6132 unlikely to ever wait at all. But we try to future-proof.]
6134 5. The function that sleeps for a time interval that includes fractions of a
6135 second contained a race. It did not block SIGALRM between setting the
6136 timer, and suspending (a couple of lines later). If the interval was short
6137 and the sigsuspend() was delayed until after it had expired, the suspension
6138 never ended. On busy systems this could lead to processes getting stuck for
6141 6. Some uncommon configurations may cause a lookup to happen in a queue runner
6142 process, before it forks any delivery processes. The open lookup caching
6143 mechanism meant that the open file or database connection was passed into
6144 the delivery process. The problem was that delivery processes always tidy
6145 up cached lookup data. This could cause a problem for the next delivery
6146 process started by the queue runner, because the external queue runner
6147 process does not know about the closure. So the next delivery process
6148 still has data in the lookup cache. In the case of a file lookup, there was
6149 no problem because closing a file descriptor in a subprocess doesn't affect
6150 the parent. However, if the lookup was caching a connection to a database,
6151 the connection was closed, and the second delivery process was likely to
6152 see errors such as "PGSQL: query failed: server closed the connection
6153 unexpectedly". The problem has been fixed by closing all cached lookups
6154 in a queue runner before running a delivery process.
6156 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
6157 seem to have the "const" qualifier which it has on other OS. I've
6160 8. Change 4.21/2 was too strict. It is only if there are two authenticators
6161 *of the same type* (client or server) with the same public name that an
6162 error should be diagnosed.
6164 9. When Exim looked up a host name for an IP address, but failed to find the
6165 original IP address when looking up the host name (a safety check), it
6166 output the message "<ip address> does not match any IP for NULL", which was
6167 confusing, to say the least. The bug was that the host name should have
6168 appeared instead of "NULL".
6170 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
6171 user that is built into the binary, and the -C or -D options is used, root
6172 privilege is dropped before the configuration file is read. In addition,
6173 logging is switched to stderr instead of the normal log files. If the
6174 configuration then re-defines the Exim user, the unprivileged environment
6175 is probably not what is expected, so Exim logs a panic warning message (but
6178 However, if deliver_drop_privilege is set, the unprivileged state may well
6179 be exactly what is intended, so the warning has been cut out in that case,
6180 and Exim is allowed to try to write to its normal log files.
6186 1. smtp_return_error_details was not giving details for temporary sender
6187 or receiver verification errors.
6189 2. Diagnose a configuration error if two authenticators have the same public
6192 3. Exim used not to create the message log file for a message until the first
6193 delivery attempt. This could be confusing when incoming messages were held
6194 for policy or load reasons. The message log file is now created at the time
6195 the message is received, and an initial "Received" line is written to it.
6197 4. The automatically generated man page for command line options had a minor
6198 bug that caused no ill effects; however, a more serious problem was that
6199 the procedure for building the man page automatically didn't always
6200 operate. Consequently, release 4.20 contains an out-of-date version. This
6201 shouldn't happen again.
6203 5. When building Exim with embedded Perl support, the script that builds the
6204 Makefile was calling 'perl' to find its compile-time parameters, ignoring
6205 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
6207 6. The freeze_tell option was not being used for messages that were frozen on
6208 arrival, either by an ACL or by local_scan().
6210 7. Added the smtp_incomplete_transaction log selector.
6212 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
6213 was accepting AUTH without a new EHLO.
6215 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
6216 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
6219 10. Logging of TCP/IP connections (when configured) now happens in the main
6220 daemon process instead of the child process, so that the TCP/IP connection
6221 count is more accurate (but it can never be perfect).
6223 11. The use of "drop" in a nested ACL was not being handled correctly in the
6224 outer ACL. Now, if condition failure induced by the nested "drop" causes
6225 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
6226 or "require"), the connection is dropped.
6228 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
6229 that yield "discard" can now be used with an "accept" or a "discard" verb,
6230 but an error is generated for any others (because I can't see a useful way
6231 to define what should happen).
6233 13. When an ACL is read dynamically from a file (or anywhere else), the lines
6234 are now processed in the same way as lines in the Exim configuration file.
6235 In particular, continuation lines are supported.
6237 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
6239 15. Added -ti meaning -t -i.
6241 16. Check for letters, digits, hyphens, and dots in the names of dnslist
6242 domains, and warn by logging if others are found.
6244 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
6245 returned from GIFCONF when Exim is trying to find the list of interfaces on
6246 a host. The code in os.c has been modified to copy each ifreq to an aligned
6247 structure in all cases.
6249 Also, in some cases, the returned ifreq's were being copied to a 'struct
6250 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
6251 means the last couple of bytes of an IPv6 address could be chopped if the
6252 ifreq contained only a normal sockaddr (14 bytes storage).
6254 18. Named domain lists were not supported in the hosts_treat_as_local option.
6255 An entry such as +xxxx was not recognized, and was treated as a literal
6258 19. Ensure that header lines added by a DATA ACL are included in the reject log
6259 if the ACL subsequently rejects the message.
6261 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
6262 MD5 (which is deprecated).
6264 21. When testing a filter file using -bf, Exim was writing a message when it
6265 took the sender from a "From " line in the message, but it was not doing so
6266 when it took $return_path from a Return-Path: header line. It now does.
6268 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
6269 with a valid header line field name (a series of printing characters
6270 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
6272 23. Changed "disc" in the source to "disk" to conform to the documentation and
6273 the book and for uniformity.
6275 24. Ignore Sendmail's -Ooption=value command line item.
6277 25. When execve() failed while trying to run a command in a pipe transport,
6278 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
6279 could be confused with a return value of 69 from the command itself. This
6280 has been changed to 127, the value the shell returns if it is asked to run
6281 a non-existent command. The wording for the related log line suggests a
6282 non-existent command as the problem.
6284 26. If received_header_text expands to an empty string, do not add a Received:
6285 header line to the message. (Well, it adds a token one on the spool, but
6286 marks it "old" so that it doesn't get used or transmitted.)
6288 27. Installed eximstats 1.28 (addition of -nt option).
6290 28. There was no check for failure on the call to getsockname() in the daemon
6291 code. This can fail if there is a shortage of resources on the system, with
6292 ENOMEM, for example. A temporary error is now given on failure.
6294 29. Contrary to the C standard, it seems that in some environments, the
6295 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
6296 program. Exim now does this explicitly; it affects the formatting of
6297 timestamps using strftime().
6299 30. If exiqsumm was given junk data, it threw up some uninitialized variable
6300 complaints. I've now initialized all the variables, to avoid this.
6302 32. Header lines added by a system filter were not being "seen" during
6303 transport-time rewrites.
6305 33. The info_callback() function passed to OpenSSL is set up with type void
6306 (*)(SSL *, int, int), as described somewhere. However, when calling the
6307 function (actually a macro) that sets it up, the type void(*)() is
6308 expected. I've put in a cast to prevent warnings from picky compilers.
6310 34. If a DNS black list lookup found a CNAME record, but there were no A
6311 records associated with the domain it pointed at, Exim crashed.
6313 35. If a DNS black list lookup returned more than one A record, Exim ignored
6314 all but the first. It now scans all returned addresses if a particular IP
6315 value is being sought. In this situation, the contents of the
6316 $dnslist_value variable are a list of all the addresses, separated by a
6319 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
6320 to accept a host name and all its aliases if the forward lookup for any of
6321 them yielded the IP address of the incoming connection. Now it accepts only
6322 those names whose forward lookup yields the correct IP address. Any other
6323 names are discarded. This closes a loophole whereby a rogue DNS
6324 administrator could create reverse DNS records to break through a
6325 wildcarded host restriction in an ACL.
6327 37. If a user filter or a system filter that ran in a subprocess used any of
6328 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
6329 the wrong values were passed to the pipe command ($thisaddress had the
6330 value of $0, $0 had the value of $1, etc). This bug was introduced by
6331 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
6333 38. Improved the line breaking for long SMTP error messages from ACLs.
6334 Previously, if there was no break point between 40 and 75 characters, Exim
6335 left the rest of the message alone. Two changes have been made: (a) I've
6336 reduced the minimum length to 35 characters; (b) if it can't find a break
6337 point between 35 and 75 characters, it looks ahead and uses the first one
6338 that it finds. This may give the occasional overlong line, but at least the
6339 remaining text gets split now.
6341 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
6342 file descriptors might be low, and that setting 1000 would always raise it.
6343 It turns out that in some environments, the limit is already over 1000 and
6344 that lowering it causes trouble. So now Exim takes care not to decrease it.
6346 40. When delivering a message, the value of $return_path is set to $sender_
6347 address at the start of routing (routers may change the value). By an
6348 oversight, this default was not being set up when an address was tested by
6349 -bt or -bv, which affected the outcome if any router or filter referred to
6352 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
6353 log only when "message" or "log_message" are set. However, if one of the
6354 conditions was an address verification, or a call to a nested ACL, the
6355 messages generated by the underlying test were being passed through. This
6356 no longer happens. The underlying message is available in $acl_verify_
6357 message for both "message" and "log_message" expansions, so it can be
6358 passed through if needed.
6360 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
6361 new expansion $bh_ to give the encoded byte string without charset
6362 translation. Translation happens only if iconv() is available; HAVE_ICONV
6363 indicates this at build time. HEADERS_CHARSET gives the charset to
6364 translate to; headers_charset can change it in the configuration, and
6365 "headers charset" can change it in an individual filter file.
6367 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
6368 that creates RFC 2047 encoded "words" labels them as that charset instead
6369 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
6370 expansion operator; (ii) when Exim creates a From: line for a local
6371 message; (iii) when a header line is rewritten to include a "phrase" part.
6373 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
6374 buggy, causing it to skip the first lines of messages whose message ID
6375 ended in 'D'. This would not have bitten before Exim release 4.14, because
6376 message IDs were unlikely to end in 'D' before then. The effect was to have
6377 incorrect size information for certain domains.
6379 45. #include "config.h" was missing at the start of the crypt16.c module. This
6380 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
6383 46. If there was a timeout during a "random" callout check, Exim treated it as
6384 a failure of the random address, and carried on sending RSET and the real
6385 address. If the delay was just some slowness somewhere, the response to the
6386 original RCPT would be taken as a response to RSET and so on, causing
6387 mayhem of various kinds.
6389 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
6390 when I implemented it. It didn't allow for the fact that some option values
6391 may legitimatetly be negative (e.g. size_addition), and it didn't even do
6392 the right test for positive values.
6394 48. Domain names in DNS records are case-independent. Exim always looks them up
6395 in lower case. Some resolvers return domain names in exactly the case they
6396 appear in the zone file, that is, they may contain uppercase letters. Not
6397 all resolvers do this - some return always lower case. Exim was treating a
6398 change of case by a resolver as a change of domain, similar to a widening
6399 of a domain abbreviation. This triggered its re-routing code and so it was
6400 trying to route what was effectively the same domain again. This normally
6401 caused routing to fail (because the router wouldn't handle the domain
6402 twice). Now Exim checks for this case specially, and just changes the
6403 casing of the domain that it ultimately uses when it transmits the message
6406 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
6409 50. If a filter generated a file delivery with a non-absolute name (possible if
6410 no home directory exists for the router), the forbid_file option was not
6413 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
6414 the existing equality matching.
6416 52. Exim was using ints instead of ino_t variables in some places where it was
6417 dealing with inode numbers.
6419 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
6420 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
6421 called TMPDIR, and if it finds it is different, it changes its value.
6423 54. The smtp_printf() function is now made available to local_scan() so
6424 additional output lines can be written before returning. There is also an
6425 smtp_fflush() function to enable the detection of a dropped connection.
6426 The variables smtp_input and smtp_batched_input are exported to
6429 55. Changed the default runtime configuration: the message "Unknown user"
6430 has been removed from the ACL, and instead placed on the localuser router,
6431 using the cannot_route_message feature. This means that any verification
6432 failures that generate their own messages won't get overridden. Similarly,
6433 the "Unrouteable address" message that was in the ACL for unverifiable
6434 relay addresses has also been removed.
6436 56. Added hosts_avoid_esmtp to the smtp transport.
6438 57. The exicyclog script was not checking for the esoteric option
6439 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
6440 will work only if exicyclog is run under the appropriate euid.
6442 58. Following a discussion on the list, the rules by which Exim recognises line
6443 endings on incoming messages have been changed. The -dropcr and drop_cr
6444 options are now no-ops, retained only for backwards compatibility. The
6445 following line terminators are recognized: LF CRLF CR. However, special
6446 processing applies to CR:
6448 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
6449 nor a local message in the state where . is a terminator.
6451 (ii) If a bare CR is encountered in a header line, an extra space is added
6452 after the line terminator so as not to end the header. The reasoning
6453 behind this is that bare CRs in header lines are most likely either
6454 to be mistakes, or people trying to play silly games.
6456 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
6457 was being incorrectly given as 18 bytes larger than it should have been.
6458 This is a VOB (very old bug).
6460 60. This may never have affected anything current, but just in case it has:
6461 When the local host is found other than at the start of a list of hosts,
6462 the local host, those with the same MX, and any that follow, are discarded.
6463 When the list in question was part of a longer list of hosts, the following
6464 hosts (not currently being processed) were also being discarded. This no
6465 longer happens. I'm not sure if this situation could ever has previously
6468 61. Added the "/MX" feature to lists of hosts in the manualroute and query
6471 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
6472 header. This is something that is recommended in a new Internet Draft, and
6473 is something that is documented as being done by Sendmail. There are two
6474 possible values. For messages generated by the autoreply transport, Exim
6477 Auto-Submitted: auto-replied
6479 whereas for all other generated messages (e.g. bounces) it adds
6481 Auto-Submitted: auto-generated
6483 63. The "personal" condition in filters now includes a test for the
6484 Auto-Submitted: header. If it contains the string "auto-" the message it
6485 not considered personal.
6487 64. Added rcpt_include_affixes as a generic transport option.
6489 65. Added queue_only_override (default true).
6491 66. Added the syslog_duplication option.
6493 67. If what should have been the first header line of a message consisted of
6494 a space followed by a colon, Exim was mis-interpreting it as a header line.
6495 It isn't of course - it is syntactically invalid and should therefore be
6496 treated as the start of the message body. The misbehaviour could have
6497 caused a number of strange effects, including loss of data in subsequent
6498 header lines, and spool format errors.
6500 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
6501 client host had authenticated. This control can now be exercised by an ACL
6502 for more flexibility.
6504 69. By default, callouts do not happen when testing with -bh. There is now a
6505 variant, -bhc, which does actually run the callout code, including
6506 consulting and updating the callout cache.
6508 70. Added support for saslauthd authentication, courtesy of Alexander
6511 71. If statvfs() failed on the spool or log directories while checking their
6512 size for availability, Exim confusingly gave the error "space shortage".
6513 Furthermore, in debugging mode it crashed with a floating point exception.
6514 These checks are done if check_{spool,log}_{space,inodes} are set, and when
6515 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
6516 serious problem, Exim now writes to the main and panic logs when this
6517 happens, with details of the failure. It then refuses to accept the
6518 incoming message, giving the message "spool directory problem" or "log
6519 directory problem" with a 421 code for SMTP messages.
6521 72. When Exim is about to re-exec itself, it ensures that the file descriptors
6522 0, 1, and 2 exist, because some OS complain for execs without them (see
6523 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
6524 descriptors. However, the code omitted to check that the open succeeded,
6525 causing mysterious errors if for some reason the permissions on /dev/null
6526 got screwed. Now Exim writes a message to the main and panic logs, and
6527 bombs out if it can't open /dev/null.
6529 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
6530 interact so that it is all more flexible. It is supposed to remain
6531 backwards compatible. Also added extra_local_interfaces.
6533 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
6534 to bomb out with an assertion failure - to the client this appears as a
6535 connection drop. This problem occurs in the part of the code that was taken
6536 from the Samba project. Fortunately, the assertion is in a very simple
6537 function, so I have fixed this by reproducing the function inline in the
6538 one place where it is called, and arranging for authentication to fail
6539 instead of killing the process with assert().
6541 75. The SPA client code was not working when the server requested OEM rather
6542 than Unicode encoding.
6544 76. Added code to make require_files with a specific uid setting more usable in
6545 the case where statting the file as root fails - usually a non-root-mounted
6546 NFS file system. When this happens and the failure is EACCES, Exim now
6547 forks a subprocess and does the per-uid checking as the relevant uid.
6549 77. Added process_log_path.
6551 78. If log_file_path was not explicitly set, a setting of check_log_space or
6552 check_log_inodes was ignored.
6554 79. If a space check for the spool or log partitions fails, the incident is now
6555 logged. Of course, in the latter case the data may get lost...
6557 80. Added the %p formatting code to string_format() so that it can be used to
6558 print addresses in debug_print(). Adjusted all the address printing in the
6559 debugging in store.c to use %p rather than %d.
6561 81. There was a concern that a line of code in smtp_in.c could overflow a
6562 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
6563 initially expressed, the concern was not well-founded, because trailing
6564 spaces are removed early. However, if the trailing spaces were followed by
6565 a NULL, they did not get removed, so the overflow was possible. Two fixes
6568 (a) I re-wrote the offending code in a cleaner fashion.
6569 (b) If an incoming SMTP command contains a NULL character, it is rejected
6572 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
6573 runs initgroups(), so that if the Exim user is in any additional groups,
6574 they will be used during message reception.
6580 The change log for 4.20 and earlier releases has been archived.