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.
79 JH/21 Bug 1846: Send DMARC forensic reports for reject and quaratine results,
80 even for a "none" policy. Patch from Tony Meyer.
82 JH/22 Fix continued use of a connection for further deliveries. If a port was
83 specified by a router, it must also match for the delivery to be
86 JH/23 Bug 1874: fix continued use of a connection for further deliveries.
87 When one of the recipients of a message was unsuitable for the connection
88 (has no matching addresses), we lost track of needing to mark it
89 deferred. As a result mail would be lost.
91 JH/24 Bug 1832: Log EHLO response on getting conn-close response for HELO.
93 JH/25 Decoding ACL controls is now done using a binary search; the sourcecode
94 takes up less space and should be simpler to maintain. Merge the ACL
95 condition decode tables also, with similar effect.
97 JH/26 Fix problem with one_time used on a redirect router which returned the
98 parent address unchanged. A retry would see the parent address marked as
99 delivered, so not attempt the (identical) child. As a result mail would
102 JH/27 Fix a possible security hole, wherein a process operating with the Exim
103 UID can gain a root shell. Credit to http://www.halfdog.net/ for
104 discovery and writeup. Ubuntu bug 1580454; no bug raised against Exim
110 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
111 and 3.4.4 - once the server is enabled to respond to an OCSP request
112 it does even when not requested, resulting in a stapling non-aware
113 client dropping the TLS connection.
115 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
116 support variable-length bit vectors. No functional change.
118 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
119 The I= interface field on outgoing lines is now after the H= remote
120 host field, same as incoming lines. There is a separate
121 outgoing_interface log selector which allows you to disable the
124 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
125 If not running log_selector +smtp_connection the mainlog would be held
126 open indefinitely after a "too many connections" event, including to a
127 deleted file after a log rotate. Leave the per net connection logging
128 leaving it open for efficiency as that will be quickly detected by the
129 check on the next write.
131 HS/01 Bug 1671: Fix post transport crash.
132 Processing the wait-<transport> messages could crash the delivery
133 process if the message IDs didn't exist for some reason. When
134 using 'split_spool_directory=yes' the construction of the spool
135 file name failed already, exposing the same netto behaviour.
137 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
138 mime_regex ACL conditions.
140 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
141 to DSN fail messages (bounces): remote IP, remote greeting, remote response
142 to HELO, local diagnostic string.
144 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
145 log line to debug. Even when configured with a tls authenticator many
146 client connections are expected to not authenticate in this way, so
147 an authenticate fail is not an error.
149 HS/02 Add the Exim version string to the process info. This way exiwhat
150 gives some more detail about the running daemon.
152 JH/06 Bug 1395: time-limit cacheing of DNS lookups, to the TTL value. This may
153 matter for fast-change records such as DNSBLs.
155 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
156 retry record, even if constant. There may be multiple transports with
157 different interface settings and the retry behaviour needs to be kept
160 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
162 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
164 JH/10 Bug 840: fix log_defer_output option of pipe transport
166 JH/11 Bug 830: use same host for all RCPTS of a message, even under
167 hosts_randomize. This matters a lot when combined with mua_wrapper.
169 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
170 ${quote_pgsql:<string>} operator.
172 JH/13 Bug 1708: avoid misaligned access in cached lookup.
174 JH/14 Change header file name for freeradius-client. Relevant if compiling
175 with Radius support; from the Gentoo tree and checked under Fedora.
177 JH/15 Bug 1712: Introduce $prdr_requested flag variable
179 JH/16 Bug 1714: Permit an empty string as expansion result for transport
180 option transport_filter, meaning no filtering.
182 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
184 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
185 defaults to "*" (all hosts). The variable is now available when not built
186 with TLS, default unset, mainly to enable keeping the testuite sane.
187 If a server certificate is not supplied (via tls_certificate) an error is
188 logged, and clients will find TLS connections fail on startup. Presumably
189 they will retry in-clear.
190 Packagers of Exim are strongly encouraged to create a server certificate
191 at installation time.
193 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
194 with the $config_file variable.
196 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
197 in transport context, after the attempt, and per-recipient. The latter type
198 is per host attempted. The event data is the error message, and the errno
199 information encodes the lookup type (A vs. MX) used for the (first) host,
200 and the trailing two digits of the smtp 4xx reponse.
202 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
203 to write to mainlog (or rejectlog, paniclog) in the window between file
204 creation and permissions/ownership being changed. Particularly affects
205 installations where exicyclog is run as root, rather than exim user;
206 result is that the running daemon panics and dies.
208 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
210 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
211 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
212 "pri" and "weight". Note that the previous implicit priority given by the
213 list order is no longer honoured.
215 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalisation
218 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
219 by defining SUPPORT_SOCKS.
221 JH/26 Move PROXY support from Experimental to mainline, enabled for a build
222 by defining SUPPORT_PROXY. Note that the proxy_required_hosts option
223 is renamed to hosts_proxy, and the proxy_{host,target}_{address,port}.
224 variables are renamed to proxy_{local,external}_{address,port}.
226 JH/27 Move Internationalisation support from Experimental to mainline, enabled
227 for a build by defining SUPPORT_I18N
229 JH/28 Bug 1745: Fix redis lookups to handle (quoted) spaces embedded in parts
230 of the query string, and make ${quote_redis:} do that quoting.
232 JH/29 Move Events support from Experimental to mainline, enabled by default
233 and removable for a build by defining DISABLE_EVENT.
235 JH/30 Updated DANE implementation code to current from Viktor Dukhovni.
237 JH/31 Fix bug with hosts_connection_nolog and named-lists which were wrongly
238 cached by the daemon.
240 JH/32 Move Redis support from Experimental to mainline, enabled for a build
241 by defining LOOKUP_REDIS. The libhiredis library is required.
243 JH/33 Bug 1748: Permit ACL dnslists= condition in non-smtp ACLs if explicit
244 keys are given for lookup.
246 JH/34 Bug 1192: replace the embedded copy of PolarSSL RSA routines in the DKIM
247 support, by using OpenSSL or GnuTLS library ones. This means DKIM is
248 only supported when built with TLS support. The PolarSSL SHA routines
249 are still used when the TLS library is too old for convenient support.
251 JH/35 Require SINGLE_DH_USE by default in OpenSSL (main config option
252 openssl_options), for security. OpenSSL forces this from version 1.1.0
253 server-side so match that on older versions.
255 JH/36 Bug 1778: longstanding bug in memory use by the ${run } expansion: A fresh
256 allocation for $value could be released as the expansion processing
257 concluded, but leaving the global pointer active for it.
259 JH/37 Bug 1769: Permit a VRFY ACL to override the default 252 response,
260 and to use the domains and local_parts ACL conditions.
262 JH/38 Fix cutthrough bug with body lines having a single dot. The dot was
263 incorrectly not doubled on cutthrough transmission, hence seen as a
264 body-termination at the receiving system - resulting in truncated mails.
265 Commonly the sender saw a TCP-level error, and retransmitted the message
266 via the normal store-and-forward channel. This could result in duplicates
267 received - but deduplicating mailstores were liable to retain only the
268 initial truncated version.
270 JH/39 Bug 1781: Fix use of DKIM private-keys having trailing '=' in the base-64.
272 JH/40 Fix crash in queryprogram router when compiled with EXPERIMENTAL_SRS.
274 JH/41 Bug 1792: Fix selection of headers to sign for DKIM: bottom-up. While
275 we're in there, support oversigning also; bug 1309.
277 JH/42 Bug 1796: Fix error logged on a malware scanner connection failure.
279 HS/04 Add support for keep_environment and add_environment options.
281 JH/43 Tidy coding issues detected by gcc --fsanitize=undefined. Some remain;
282 either intentional arithmetic overflow during PRNG, or testing config-
285 JH/44 Bug 1800: The combination of a -bhc commandline option and cutthrough
286 delivery resulted in actual delivery. Cancel cutthrough before DATA
289 JH/45 Fix cutthrough, when connection not opened by verify and target hard-
290 rejects a recipient: pass the reject to the originator.
292 JH/46 Multiple issues raised by Coverity. Some were obvious or plausible bugs.
293 Many were false-positives and ignorable, but it's worth fixing the
296 JH/47 Fix build on HP-UX and older Solaris, which need (un)setenv now also
297 for the new environment-manipulation done at startup. Move the routines
298 from being local to tls.c to being global via the os.c file.
300 JH/48 Bug 1807: Fix ${extract } for the numeric/3-string case. While preparsing
301 an extract embedded as result-arg for a map, the first arg for extract
302 is unavailable so we cannot tell if this is a numbered or keyed
303 extraction. Accept either.
309 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
312 JH/02 The smtp transport option "multi_domain" is now expanded.
314 JH/03 The smtp transport now requests PRDR by default, if the server offers
317 JH/04 Certificate name checking on server certificates, when exim is a client,
318 is now done by default. The transport option tls_verify_cert_hostnames
319 can be used to disable this per-host. The build option
320 EXPERIMENTAL_CERTNAMES is withdrawn.
322 JH/05 The value of the tls_verify_certificates smtp transport and main options
323 default to the word "system" to access the system default CA bundle.
324 For GnuTLS, only version 3.0.20 or later.
326 JH/06 Verification of the server certificate for a TLS connection is now tried
327 (but not required) by default. The verification status is now logged by
328 default, for both outbound TLS and client-certificate supplying inbound
331 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
334 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
335 Status Notification (bounce) messages are now MIME format per RFC 3464.
336 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
337 under the control of the dsn_advertise_hosts option, and routers may
338 have a dsn_lasthop option.
340 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
341 default, modifiable by a malware= option. The list separator for
342 the options can now be changed in the usual way. Bug 68.
344 JH/10 The smtp_receive_timeout main option is now expanded before use.
346 JH/11 The incoming_interface log option now also enables logging of the
347 local interface on delivery outgoing connections.
349 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
350 if the interface and destination host and port all match.
352 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
355 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
356 Patch from Andrew Lewis.
358 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
359 now supports optional time-restrictions, weighting, and priority
360 modifiers per server. Patch originally by <rommer@active.by>.
362 JH/16 The spamd_address main option now supports a mixed list of local
363 and remote servers. Remote servers can be IPv6 addresses, and
364 specify a port-range.
366 JH/17 Bug 68: The spamd_address main option now supports an optional
367 timeout value per server.
369 JH/18 Bug 1581: Router and transport options headers_add/remove can
370 now have the list separator specified.
372 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
375 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
378 JH/21 Support for the A6 type of dns record is withdrawn.
380 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
381 rather than the verbs used.
383 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
384 from 255 to 1024 chars.
386 JH/24 Verification callouts now attempt to use TLS by default.
388 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
389 are generic router options now. The defaults didn't change.
391 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
392 Original patch from Alexander Shikoff, worked over by JH.
394 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
395 files if ZCAT_COMMAND is not executable.
397 JH/26 Bug 1539: Add timout/retry options on dnsdb lookups.
399 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
401 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
402 Normally benign, it bites when the pair was led to by a CNAME;
403 modern usage is to not canoicalize the domain to a CNAME target
404 (and we were inconsistent anyway for A-only vs AAAA+A).
406 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
408 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
409 when evaluating $sender_host_dnssec.
411 JH/31 Check the HELO verification lookup for DNSSEC, adding new
412 $sender_helo_dnssec variable.
414 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
416 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
418 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
420 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
421 documented as working, but never had. Support all but $spam_report.
423 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
424 added for tls authenticator.
426 HS/03 Add perl_taintmode main config option
431 TL/01 When running the test suite, the README says that variables such as
432 no_msglog_check are global and can be placed anywhere in a specific
433 test's script, however it was observed that placement needed to be near
434 the beginning for it to behave that way. Changed the runtest perl
435 script to read through the entire script once to detect and set these
436 variables, reset to the beginning of the script, and then run through
437 the script parsing/test process like normal.
439 TL/02 The BSD's have an arc4random API. One of the functions to induce
440 adding randomness was arc4random_stir(), but it has been removed in
441 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
442 function when detected.
444 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
445 cause callback expansion.
447 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
448 syntax errors in an expansion can be treated as a string instead of
449 logging or causing an error, due to the internal use of bool_lax
450 instead of bool when processing it.
452 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
453 server certificates when making smtp deliveries.
455 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
457 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
459 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
461 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
462 Merged patch from Sebastian Wiedenroth.
464 JH/05 Fix results-pipe from transport process. Several recipients, combined
465 with certificate use, exposed issues where response data items split
466 over buffer boundaries were not parsed properly. This eventually
467 resulted in duplicates being sent. This issue only became common enough
468 to notice due to the introduction of conection certificate information,
469 the item size being so much larger. Found and fixed by Wolfgang Breyha.
471 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
472 size buffer was used, resulting in syntax errors when an expansion
475 JH/07 Add support for directories of certificates when compiled with a GnuTLS
476 version 3.3.6 or later.
478 JH/08 Rename the TPDA expermimental facility to Event Actions. The #ifdef
479 is EXPERIMENTAL_EVENT, the main-configuration and transport options
480 both become "event_action", the variables become $event_name, $event_data
481 and $event_defer_errno. There is a new variable $verify_mode, usable in
482 routers, transports and related events. The tls:cert event is now also
483 raised for inbound connections, if the main configuration event_action
486 TL/06 In test suite, disable OCSP for old versions of openssl which contained
487 early OCSP support, but no stapling (appears to be less than 1.0.0).
489 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
490 server certificate names available under the smtp transport option
491 "tls_verify_cert_hostname" now do not permit multi-component wildcard
494 JH/10 Time-related extraction expansions from certificates now use the main
495 option "timezone" setting for output formatting, and are consistent
496 between OpenSSL and GnuTLS compilations. Bug 1541.
498 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
499 encoded parameter in the incoming message. Bug 1558.
501 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
502 include certificate info, eximon was claiming there were spoolfile
505 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
507 JH/14 Log delivery-related information more consistently, using the sequence
508 "H=<name> [<ip>]" wherever possible.
510 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
511 are problematic for Debian distribution, omit them from the release
514 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
516 JH/16 Fix string representation of time values on 64bit time_t anchitectures.
519 JH/17 Fix a null-indirection in certextract expansions when a nondefault
520 output list separator was used.
525 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
526 checkers that were complaining about end of non-void function with no
529 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
530 This was a regression intruduced in 4.83 by another bugfix.
532 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
534 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
535 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
541 TF/01 Correctly close the server side of TLS when forking for delivery.
543 When a message was received over SMTP with TLS, Exim failed to clear up
544 the incoming connection properly after forking off the child process to
545 deliver the message. In some situations the subsequent outgoing
546 delivery connection happened to have the same fd number as the incoming
547 connection previously had. Exim would try to use TLS and fail, logging
548 a "Bad file descriptor" error.
550 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
551 utilities have not been installed.
553 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
554 temporary space as the ACL may create new global variables.
556 TL/01 LDAP support uses per connection or global context settings, depending
557 upon the detected version of the libraries at build time.
559 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
560 to extract and use the src ip:port in logging and expansions as if it
561 were a direct connection from the outside internet. PPv2 support was
562 updated based on HAProxy spec change in May 2014.
564 JH/02 Add ${listextract {number}{list}{success}{fail}}.
566 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
567 Properly escape header and check for NULL return.
569 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
572 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
574 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
575 characters in header names, implemented as a verify condition.
576 Contributed by Michael Fischer v. Mollard.
578 TL/05 Rename SPF condition results err_perm and err_temp to standardized
579 results permerror and temperror. Previous values are deprecated but
580 still accepted. In a future release, err_perm and err_temp will be
581 completely removed, which will be a backward incompatibility if the
582 ACL tests for either of these two old results. Patch contributed by
583 user bes-internal on the mailing list.
585 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
587 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
588 selectors, in both main and reject logs.
590 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
593 JH/07 Add malware type "sock" for talking to simple daemon.
595 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
597 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
598 routers/transports under cutthrough routing.
600 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
601 numbers. Touch up "bool" conditional to keep the same definition.
603 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
605 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
607 JH/12 Expand items in router/transport headers_add or headers_remove lists
608 individually rather than the list as a whole. Bug 1452.
610 Required for reasonable handling of multiple headers_ options when
611 they may be empty; requires that headers_remove items with embedded
612 colons must have them doubled (or the list-separator changed).
614 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
615 view the policy declared in the DMARC record. Currently, $dmarc_status
616 is a combined value of both the record presence and the result of the
619 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
621 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
622 dnslookup router and the smtp transport (applying to the forward
625 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
626 of ldap servers used for a specific lookup. Patch provided by Heiko
629 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
630 New variable $lookup_dnssec_authenticated for observability.
632 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
633 Patch submitted by Lars Timman.
635 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
637 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
638 Requires trusted mode and valid format message id, aborts otherwise.
639 Patch contributed by Heiko Schlichting.
641 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
642 certextract with support for various fields. Bug 1358.
644 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
645 is requested by default, modifiable by smtp transport option
648 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
649 operate on certificate variables to give certificate fingerprints
650 Also new ${sha256:cert_variable}.
652 JH/23 The PRDR feature is moved from being Experimental into the mainline.
654 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
655 Christian Aistleitner.
657 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
659 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
660 file. Patch from Wolfgang Breyha.
662 JH/25 Expand the coverage of the delivery $host and $host_address to
663 client authenticators run in verify callout. Bug 1476.
665 JH/26 Port service names are now accepted for tls_on_connect_ports, to
666 align with daemon_smtp_ports. Bug 72.
668 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
669 support and error reporting did not work properly.
671 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
672 and is readable. Patch from Andrew Colin Kissa.
674 TL/14 Enhance documentation of ${run expansion and how it parses the
675 commandline after expansion, particularly in the case when an
676 unquoted variable expansion results in an empty value.
678 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
680 PP/02 Fix internal collision of T_APL on systems which support RFC3123
681 by renaming away from it. Addresses GH issue 15, reported by
684 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
686 TL/15 SECURITY: prevent double expansion in math comparison functions
687 (can expand unsanitized data). Not remotely exploitable.
694 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
696 PP/02 Make -n do something, by making it not do something.
697 When combined with -bP, the name of an option is not output.
699 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
702 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
703 $sender_host_name and config options to manage this, and basic check
706 PP/05 DSCP support for outbound connections and control modifier for inbound.
708 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
709 (Only plugin which currently uses this is kerberos4, which nobody should
710 be using, but we should make it available and other future plugins might
711 conceivably use it, even though it would break NAT; stuff *should* be
712 using channel bindings instead).
714 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
715 name; added for Sendmail compatibility; requires admin caller.
716 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
717 just ignore it); requires trusted caller.
718 Also parse but ignore: -Ac -Am -X<logfile>
721 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
723 TL/02 Add +smtp_confirmation as a default logging option.
725 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
726 Patch by Magnus Holmgren from 2007-02-20.
728 TL/04 Bugzilla 1281 - Spec typo.
729 Bugzilla 1283 - Spec typo.
730 Bugzilla 1290 - Spec grammar fixes.
732 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
734 TL/06 Add Experimental DMARC support using libopendmarc libraries.
736 TL/07 Fix an out of order global option causing a segfault. Reported to dev
737 mailing list by by Dmitry Isaikin.
739 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
741 JH/02 Support "G" suffix to numbers in ${if comparisons.
743 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
745 NM/01 Bugzilla 1197 - Spec typo
746 Bugzilla 1196 - Spec examples corrections
748 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
750 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
751 gnutls_enable_pkcs11, but renamed to more accurately indicate its
754 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
755 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
757 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
758 "acl {{name}{arg}...}", and optional args on acl condition
761 JH/05 Permit multiple router/transport headers_add/remove lines.
763 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
765 JH/07 Avoid using a waiting database for a single-message-only transport.
766 Performance patch from Paul Fisher. Bugzilla 1262.
768 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
771 JH/09 Add $headers_added variable, with content from use of ACL modifier
772 add_header (but not yet added to the message). Bugzilla 199.
774 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
775 Pulled from Bugzilla 817 by Wolfgang Breyha.
777 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
779 (nb: this is the same fix as in Exim 4.80.1)
781 JH/11 Add A= logging on delivery lines, and a client_set_id option on
784 JH/12 Add optional authenticated_sender logging to A= and a log_selector
787 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
789 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
790 advertise SMTP AUTH mechanism to us, instead of a generic
791 protocol violation error. Also, make Exim more robust to bad
792 data from the Dovecot auth socket.
794 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
796 When a queue runner is handling a message, Exim first routes the
797 recipient addresses, during which it prunes them based on the retry
798 hints database. After that it attempts to deliver the message to
799 any remaining recipients. It then updates the hints database using
802 So if a recipient address works intermittently, it can get repeatedly
803 deferred at routing time. The retry hints record remains fresh so the
804 address never reaches the final cutoff time.
806 This is a fairly common occurrence when a user is bumping up against
807 their storage quota. Exim had some logic in its local delivery code
808 to deal with this. However it did not apply to per-recipient defers
809 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
811 This change adds a proper retry rule check during routing so that the
812 final cutoff time is checked against the message's age. We only do
813 this check if there is an address retry record and there is not a
814 domain retry record; this implies that previous attempts to handle
815 the address had the retry_use_local_parts option turned on. We use
816 this as an approximation for the destination being like a local
817 delivery, as in LMTP.
819 I suspect this new check makes the old local delivery cutoff check
820 redundant, but I have not verified this so I left the code in place.
822 TF/02 Correct gecos expansion when From: is a prefix of the username.
824 Test 0254 submits a message to Exim with the header
828 When I ran the test suite under the user fanf2, Exim expanded
829 the header to contain my full name, whereas it should have added
830 a Resent-Sender: header. It erroneously treats any prefix of the
831 username as equal to the username.
833 This change corrects that bug.
835 GF/01 DCC debug and logging tidyup
836 Error conditions log to paniclog rather than rejectlog.
837 Debug lines prefixed by "DCC: " to remove any ambiguity.
839 TF/03 Avoid unnecessary rebuilds of lookup-related code.
841 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
842 Bug spotted by Jeremy Harris; was flawed since initial commit.
843 Would have resulted in OCSP responses post-SNI triggering an Exim
844 NULL dereference and crash.
846 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
848 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
849 Bug detection, analysis and fix by Samuel Thibault.
850 Bugzilla 1331, Debian bug #698092.
852 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
854 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
855 Server implementation by Todd Lyons, client by JH.
856 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
857 config variable "prdr_enable" controls whether the server
858 advertises the facility. If the client requests PRDR a new
859 acl_data_smtp_prdr ACL is called once for each recipient, after
860 the body content is received and before the acl_smtp_data ACL.
861 The client is controlled by bolth of: a hosts_try_prdr option
862 on the smtp transport, and the server advertisement.
863 Default client logging of deliveries and rejections involving
864 PRDR are flagged with the string "PRDR".
866 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
867 fclose(). Diagnosis by Todd Lyons.
869 PP/17 Update configure.default to handle IPv6 localhost better.
870 Patch by Alain Williams (plus minor tweaks).
873 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
874 This is now consistent with GnuTLS, and is now documented: the
875 previous undocumented portable approach to treating the option as
876 unset was to force an expansion failure. That still works, and
877 an empty string is now equivalent.
879 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
880 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
881 not performing validation itself.
883 PP/20 Added force_command boolean option to pipe transport.
884 Patch from Nick Koston, of cPanel Inc.
886 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
889 TF/04 Added udpsend ACL modifer and hexquote expansion operator
891 PP/21 Fix eximon continuous updating with timestamped log-files.
892 Broken in a format-string cleanup in 4.80, missed when I repaired the
893 other false fix of the same issue.
894 Report and fix from Heiko Schlichting.
897 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
898 Report from Prashanth Katuri.
900 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
901 It's SecureTransport, so affects any MacOS clients which use the
902 system-integrated TLS libraries, including email clients.
904 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
905 using a MIME ACL for non-SMTP local injection.
906 Report and assistance in diagnosis by Warren Baker.
908 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
910 JH/16 Fix comparisons for 64b. Bugzilla 1385.
912 TL/09 Add expansion variable $authenticated_fail_id to keep track of
913 last id that failed so it may be referenced in subsequent ACL's.
915 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
918 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
919 ldap library initialization, allowing self-signed CA's to be
920 used. Also properly sets require_cert option later in code by
921 using NULL (global ldap config) instead of ldap handle (per
922 session). Bug diagnosis and testing by alxgomz.
924 TL/12 Enhanced documentation in the ratelimit.pl script provided in
925 the src/util/ subdirectory.
927 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
928 renamed to Transport Post Delivery Action by Jeremy Harris, as
931 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
932 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
933 redis_servers = needs to be configured which will be used by the redis
934 lookup. Patch from Warren Baker, of The Packet Hub.
936 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
938 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
939 hostname or reverse DNS when processing a host list. Used suggestions
940 from multiple comments on this bug.
942 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
944 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
945 Missed a few lines, added it to make the runtest require no keyboard
948 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
949 contains upper case chars. Make router use caseful_local_part.
951 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
952 support when GnuTLS has been built with p11-kit.
958 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
960 This, or similar/improved, will also be change PP/11 of 4.82.
966 PP/01 Handle short writes when writing local log-files.
967 In practice, only affects FreeBSD (8 onwards).
968 Bugzilla 1053, with thanks to Dmitry Isaikin.
970 NM/01 Bugzilla 949 - Documentation tweak
972 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
975 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
977 PP/02 Implemented gsasl authenticator.
979 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
981 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
982 `pkg-config foo` for cflags/libs.
984 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
985 with rest of GSASL and with heimdal_gssapi.
987 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
988 `pkg-config foo` for cflags/libs for the TLS implementation.
990 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
991 properties get this fed in as external SSF. A number of robustness
992 and debugging improvements to the cyrus_sasl authenticator.
994 PP/08 cyrus_sasl server now expands the server_realm option.
996 PP/09 Bugzilla 1214 - Log authentication information in reject log.
997 Patch by Jeremy Harris.
999 PP/10 Added dbmjz lookup type.
1001 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
1003 PP/12 MAIL args handles TAB as well as SP, for better interop with
1004 non-compliant senders.
1005 Analysis and variant patch by Todd Lyons.
1007 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
1008 Bug report from Lars Müller <lars@samba.org> (via SUSE),
1009 Patch from Dirk Mueller <dmueller@suse.com>
1011 PP/13 tls_peerdn now print-escaped for spool files.
1012 Observed some $tls_peerdn in wild which contained \n, which resulted
1013 in spool file corruption.
1015 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
1016 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
1017 or write after TLS renegotiation, which otherwise led to messages
1020 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
1021 as a tracking header (ie: a signed header comes before the signature).
1022 Patch from Wolfgang Breyha.
1024 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
1025 comma-sep list; embedded commas doubled.
1027 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
1029 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
1031 Report and patch from Dmitry Banschikov.
1033 PP/16 Removed "dont_insert_empty_fragments" fron "openssl_options".
1034 Removed SSL_clear() after SSL_new() which led to protocol negotiation
1035 failures. We appear to now support TLS1.1+ with Exim.
1037 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
1038 lets Exim select keys and certificates based upon TLS SNI from client.
1039 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
1040 before an outbound SMTP session. New log_selector, +tls_sni.
1042 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
1043 NULL dereference. Report and patch from Alun Jones.
1045 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
1046 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
1047 Not seeing resolver debug output on NetBSD, but suspect this is a
1048 resolver implementation change.
1050 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
1051 Left warnings. Added "eximon gdb" invocation mode.
1053 PP/21 Defaulting "accept_8bitmime" to true, not false.
1055 PP/22 Added -bw for inetd wait mode support.
1057 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
1058 locate the relevant includes and libraries. Made this the default.
1060 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
1061 Bugzilla 1246, report and most of solution from Tomasz Kusy.
1063 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
1064 This may cause build issues on older platforms.
1066 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
1067 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
1068 gnutls_require_mac & gnutls_require_protocols (no longer supported).
1069 Added SNI support via GnuTLS too.
1070 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
1072 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
1074 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
1075 Applied second patch from Janne, implementing suggestion to default
1076 multiple-strings-in-record handling to match SPF spec.
1078 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
1080 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
1081 read-only, out of scope).
1082 Patch from Wolfgang Breyha, report from Stuart Northfield.
1084 PP/29 Fix three issues highlighted by clang analyser static analysis.
1085 Only crash-plausible issue would require the Cambridge-specific
1086 iplookup router and a misconfiguration.
1087 Report from Marcin Mirosław.
1089 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
1091 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
1092 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
1093 As part of this, removing so much warning spew let me fix some minor
1094 real issues in debug logging.
1096 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
1097 assignment on my part. Fixed.
1099 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
1100 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
1101 Janne Snabb (who went above and beyond: thank you).
1103 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
1104 string otherwise requires a connection and a bunch more work and it's
1105 relatively easy to get wrong. Should also expose TLS library linkage
1108 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
1109 64-bit ${eval} (JH/03).
1111 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
1112 GNU libc to support some of the 64-bit stuff, should not lead to
1113 conflicts. Defined before os.h is pulled in, so if a given platform
1114 needs to override this, it can.
1116 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
1117 protection layer was required, which is not implemented.
1118 Bugzilla 1254, patch from Wolfgang Breyha.
1120 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
1121 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
1122 tls_dhparam take prime identifiers. Also unbreak combination of
1123 OpenSSL+DH_params+TLSSNI.
1125 PP/39 Disable SSLv2 by default in OpenSSL support.
1131 PP/01 Solaris build fix for Oracle's LDAP libraries.
1132 Bugzilla 1109, patch from Stephen Usher.
1134 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
1136 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
1139 TF/02 Fix a couple more cases where we did not log the error message
1140 when unlink() failed. See also change 4.74-TF/03.
1142 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
1143 lock up or crash if it happened to be inside a call to libc when it
1144 got a SIGUSR1 from exiwhat.
1146 The SIGUSR1 handler appends the current process status to the process
1147 log which is later printed by exiwhat. It used to use the general
1148 purpose logging code to do this, but several functions it calls are
1149 not safe for signals.
1151 The new output code in the SIGUSR1 handler is specific to the process
1152 log, and simple enough that it's easy to inspect for signal safety.
1153 Removing some special cases also simplifies the general logging code.
1154 Removing the spurious timestamps from the process log simplifies
1157 TF/04 Improved ratelimit ACL condition.
1159 The /noupdate option has been deprecated in favour of /readonly which
1160 has clearer semantics. The /leaky, /strict, and /readonly update modes
1161 are mutually exclusive. The update mode is no longer included in the
1162 database key; it just determines when the database is updated. (This
1163 means that when you upgrde Exim will forget old rate measurements.)
1165 Exim now checks that the per_* options are used with an update mode that
1166 makes sense for the current ACL. For example, when Exim is processing a
1167 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
1168 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
1169 must specify per_mail/readonly. If you omit the update mode it defaults to
1170 /leaky where that makes sense (as before) or /readonly where required.
1172 The /noupdate option is now undocumented but still supported for
1173 backwards compatibility. It is equivalent to /readonly except that in
1174 ACLs where /readonly is required you may specify /leaky/noupdate or
1175 /strict/noupdate which are treated the same as /readonly.
1177 A useful new feature is the /count= option. This is a generalization
1178 of the per_byte option, so that you can measure the throughput of other
1179 aggregate values. For example, the per_byte option is now equivalent
1180 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
1182 The per_rcpt option has been generalized using the /count= mechanism
1183 (though it's more complicated than the per_byte equivalence). When it is
1184 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
1185 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
1186 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
1187 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
1188 Note that using per_rcpt with a non-readonly update mode in more than
1189 one ACL will cause the recipients to be double-counted. (The per_mail
1190 and per_byte options don't have this problem.)
1192 The handling of very low rates has changed slightly. If the computed rate
1193 is less than the event's count (usually one) then this event is the first
1194 after a long gap. In this case the rate is set to the same as this event's
1195 count, so that the first message of a spam run is counted properly.
1197 The major new feature is a mechanism for counting the rate of unique
1198 events. The new per_addr option counts the number of different
1199 recipients that someone has sent messages to in the last time period. It
1200 behaves like per_rcpt if all the recipient addresses are different, but
1201 duplicate recipient addresses do not increase the measured rate. Like
1202 the /count= option this is a general mechanism, so the per_addr option
1203 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
1204 example, measure the rate that a client uses different sender addresses
1205 with the options per_mail/unique=$sender_address. There are further
1206 details in the main documentation.
1208 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
1210 TF/06 Removed a few PCRE remnants.
1212 TF/07 Automatically extract Exim's version number from tags in the git
1213 repository when doing development or release builds.
1215 PP/02 Raise smtp_cmd_buffer_size to 16kB.
1216 Bugzilla 879. Patch from Paul Fisher.
1218 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
1219 Heavily based on revision 40f9a89a from Simon Arlott's tree.
1222 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
1224 PP/05 Variable $av_failed, true if the AV scanner deferred.
1225 Bugzilla 1078. Patch from John Horne.
1227 PP/06 Stop make process more reliably on build failure.
1228 Bugzilla 1087. Patch from Heiko Schlittermann.
1230 PP/07 Make maildir_use_size_file an _expandable_ boolean.
1231 Bugzilla 1089. Patch from Heiko Schlittermann.
1233 PP/08 Handle ${run} returning more data than OS pipe buffer size.
1234 Bugzilla 1131. Patch from Holger Weiß.
1236 PP/09 Handle IPv6 addresses with SPF.
1237 Bugzilla 860. Patch from Wolfgang Breyha.
1239 PP/10 GnuTLS: support TLS 1.2 & 1.1.
1241 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
1244 PP/11 match_* no longer expand right-hand-side by default.
1245 New compile-time build option, EXPAND_LISTMATCH_RHS.
1246 New expansion conditions, "inlist", "inlisti".
1248 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
1250 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
1252 PP/14 fix log_write() format string regression from TF/03.
1253 Bugzilla 1152. Patch from Dmitry Isaikin.
1259 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
1261 PP/02 Harmonised TLS library version reporting; only show if debugging.
1262 Layout now matches that introduced for other libraries in 4.74 PP/03.
1264 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
1266 PP/04 New "dns_use_edns0" global option.
1268 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
1271 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
1272 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
1274 TK/01 Updated PolarSSL code to 0.14.2.
1275 Bugzilla 1097. Patch from Andreas Metzler.
1277 PP/07 Catch divide-by-zero in ${eval:...}.
1278 Fixes bugzilla 1102.
1280 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
1283 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
1284 format-string attack -- SECURITY: remote arbitrary code execution.
1286 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
1287 time unintentionally subject to list matching rules, letting the header
1288 cause arbitrary Exim lookups (of items which can occur in lists, *not*
1289 arbitrary string expansion). This allowed for information disclosure.
1291 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
1292 INT_MIN/-1 -- value coerced to INT_MAX.
1298 NM/01 Workround for PCRE version dependency in version reporting
1301 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
1302 This fixes portability to compilers other than gcc, notably
1303 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
1305 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
1306 makefiles for portability to HP-UX and POSIX correctness.
1308 PP/01 Permit LOOKUP_foo enabling on the make command-line.
1309 Also via indented variable definition in the Makefile.
1310 (Debugging by Oliver Heesakkers).
1312 PP/02 Restore caching of spamd results with expanded spamd_address.
1313 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
1315 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
1316 Improves build reliability. Fix from: Frank Elsner
1318 NM/02 Fix wide character breakage in the rfc2047 coding
1319 Fixes bug 1064. Patch from Andrey N. Oktyabrski
1321 NM/03 Allow underscore in dnslist lookups
1322 Fixes bug 1026. Patch from Graeme Fowler
1324 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
1325 Code patches from Adam Ciarcinski of NetBSD.
1327 NM/04 Fixed exiqgrep to cope with mailq missing size issue
1330 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
1331 is logged, to avoid truncation. Patch from John Horne.
1333 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
1334 Patch from Jakob Hirsch.
1336 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
1337 SQL string expansion failure details.
1338 Patch from Andrey Oktyabrski.
1340 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
1341 Patch from Simon Arlott.
1343 PP/09 New lookups functionality failed to compile on old gcc which rejects
1344 extern declarations in function scope.
1345 Patch from Oliver Fleischmann
1347 PP/10 Use sig_atomic_t for flags set from signal handlers.
1348 Check getgroups() return and improve debugging.
1349 Fixed developed for diagnosis in bug 927 (which turned out to be
1352 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
1353 Patch from Mark Zealey.
1355 PP/12 Bugzilla 1056: Improved spamd server selection.
1356 Patch from Mark Zealey.
1358 PP/13 Bugzilla 1086: Deal with maildir quota file races.
1359 Based on patch from Heiko Schlittermann.
1361 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
1362 Patch from Uwe Doering, sign-off by Michael Haardt.
1364 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
1365 variable declaration deep within a block. Bug and patch from
1368 PP/15 lookups-Makefile IRIX compatibilty coercion.
1370 PP/16 Make DISABLE_DKIM build knob functional.
1372 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
1373 Patch by Simon Arlott
1375 TF/03 Fix valgrind.h portability to C89 compilers that do not support
1376 variable argument macros. Our copy now differs from upstream.
1382 TF/01 Failure to get a lock on a hints database can have serious
1383 consequences so log it to the panic log.
1385 TF/02 Log LMTP confirmation messages in the same way as SMTP,
1386 controlled using the smtp_confirmation log selector.
1388 TF/03 Include the error message when we fail to unlink a spool file.
1390 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
1391 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
1392 for maintaining out-of-tree patches for some time.
1394 PP/01 Bugzilla 139: Documentation and portability issues.
1395 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
1396 Handle per-OS dynamic-module compilation flags.
1398 PP/02 Let /dev/null have normal permissions.
1399 The 4.73 fixes were a little too stringent and complained about the
1400 permissions on /dev/null. Exempt it from some checks.
1401 Reported by Andreas M. Kirchwitz.
1403 PP/03 Report version information for many libraries, including
1404 Exim version information for dynamically loaded libraries. Created
1405 version.h, now support a version extension string for distributors
1406 who patch heavily. Dynamic module ABI change.
1408 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
1409 privilege escalation vulnerability whereby the Exim run-time user
1410 can cause root to append content of the attacker's choosing to
1413 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
1416 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
1417 If dropping privileges for untrusted macros, we disabled normal logging
1418 on the basis that it would fail; for the Exim run-time user, this is not
1419 the case, and it resulted in successful deliveries going unlogged.
1420 Fixed. Reported by Andreas Metzler.
1426 PP/01 Date: & Message-Id: revert to normally being appended to a message,
1427 only prepend for the Resent-* case. Fixes regression introduced in
1428 Exim 4.70 by NM/22 for Bugzilla 607.
1430 PP/02 Include check_rfc2047_length in configure.default because we're seeing
1431 increasing numbers of administrators be bitten by this.
1433 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
1435 PP/03 Bugzilla 994: added openssl_options main configuration option.
1437 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
1439 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
1441 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
1443 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
1444 without a peer certificate, leading to a segfault because of an
1445 assumption that peers always have certificates. Be a little more
1446 paranoid. Problem reported by Martin Tscholak.
1448 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
1449 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
1450 NB: ClamAV planning to remove STREAM in "middle of 2010".
1451 CL also introduces -bmalware, various -d+acl logging additions and
1452 more caution in buffer sizes.
1454 PP/09 Implemented reverse_ip expansion operator.
1456 PP/10 Bugzilla 937: provide a "debug" ACL control.
1458 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
1460 PP/12 Bugzilla 973: Implement --version.
1462 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
1464 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
1466 PP/15 Bugzilla 816: support multiple condition rules on Routers.
1468 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
1469 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
1470 ignore trailing whitespace.
1472 JJ/02 prevent non-panic DKIM error from being sent to paniclog
1474 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
1477 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
1478 Notification from Dr Andrew Aitchison.
1480 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
1481 ExtendedDetectionInfo response format.
1482 Notification from John Horne.
1484 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
1487 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
1488 XSL and documented dependency on system catalogs, with examples of how
1491 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
1494 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
1495 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
1496 configuration file which is writeable by the Exim user or group.
1498 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
1499 of configuration files to cover files specified with the -C option if
1500 they are going to be used with root privileges, not just the default
1503 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
1504 option (effectively making it always true).
1506 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
1507 files to be used while preserving root privileges.
1509 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
1510 that rogue child processes cannot use them.
1512 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
1513 run-time user, instead of root.
1515 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
1516 Exim run-time user without dropping privileges.
1518 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
1519 result string, instead of calling string_vformat() twice with the same
1522 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
1523 for other users. Others should always drop root privileges if they use
1524 -C on the command line, even for a whitelisted configure file.
1526 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
1528 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
1534 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
1535 $data_path, and $header_path variables; fixed documentation bugs and
1538 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
1539 exipick to access non-standard spools, including the "frozen" queue
1542 NM/01 Bugzilla 965: Support mysql stored procedures.
1543 Patch from Alain Williams
1545 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
1547 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
1548 Patch from Andreas Metzler
1550 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
1551 Patch from Kirill Miazine
1553 NM/05 Bugzilla 671: Added umask to procmail example.
1555 JJ/03 installed exipick 20100323.0, fixing doc bug
1557 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
1558 directory. Notification and patch from Dan Rosenberg.
1560 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
1562 TK/02 Improve log output when DKIM signing operation fails.
1564 MH/01 Treat the transport option dkim_domain as a colon separated
1565 list, not as a single string, and sign the message with each element,
1566 omitting multiple occurences of the same signer.
1568 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
1569 Bugzilla 985, 986. Patch by Simon Arlott
1571 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
1572 Patch by Simon Arlott
1574 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
1575 MBX locking. Notification from Dan Rosenberg.
1581 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
1583 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
1585 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
1587 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
1589 NM/04 Bugzilla 915: Flag broken perl installation during build.
1595 TK/01 Added patch by Johannes Berg that expands the main option
1596 "spamd_address" if it starts with a dollar sign.
1598 TK/02 Write list of recipients to X-Envelope-Sender header when building
1599 the mbox-format spool file for content scanning (suggested by Jakob
1602 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
1603 (http://www.dcc-servers.net/) support via dccifd. Activated by
1604 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
1606 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
1607 by Mark Daniel Reidel <mr@df.eu>.
1609 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
1610 When building exim an external PCRE library is now needed -
1611 PCRE is a system library on the majority of modern systems.
1612 See entry on PCRE_LIBS in EDITME file.
1614 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
1615 conversation. Added nologin parameter to request.
1616 Patch contributed by Kirill Miazine.
1618 TF/01 Do not log submission mode rewrites if they do not change the address.
1620 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
1622 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
1623 log files in place. Contributed by Roberto Lima.
1625 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
1627 TF/03 Bugzilla 615: When checking the local_parts router precondition
1628 after a local_part_suffix or local_part_prefix option, Exim now
1629 does not use the address's named list lookup cache, since this
1630 contains cached lookups for the whole local part.
1632 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
1633 Robert Millan. Documentation is in experimental-spec.txt.
1635 TF/04 Bugzilla 668: Fix parallel build (make -j).
1637 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
1639 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
1640 Patch provided by Jan Srzednicki.
1642 TF/05 Leading white space used to be stripped from $spam_report which
1643 wrecked the formatting. Now it is preserved.
1645 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
1646 that they are available at delivery time.
1648 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
1650 TF/08 TLS error reporting now respects the incoming_interface and
1651 incoming_port log selectors.
1653 TF/09 Produce a more useful error message if an SMTP transport's hosts
1654 setting expands to an empty string.
1656 NM/06 Bugzilla 744: EXPN did not work under TLS.
1657 Patch provided by Phil Pennock.
1659 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
1660 Patch provided by Richard Godbee.
1662 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
1663 acl_smtp_notquit, added index entry.
1665 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
1666 Patch provided by Eugene Bujak.
1668 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
1669 accept(). Patch provided by Maxim Dounin.
1671 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
1672 Patch provided by Phil Pennock.
1674 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
1676 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
1677 Patch provided by Brad "anomie" Jorsch.
1679 NM/14 Bugzilla 622: Added timeout setting to transport filter.
1680 Patch provided by Dean Brooks.
1682 TK/05 Add native DKIM support (does not depend on external libraries).
1684 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
1685 Patch provided by Graeme Fowler.
1687 NM/16 Bugzilla 851: Documentation example syntax fix.
1689 NM/17 Changed NOTICE file to remove references to embedded PCRE.
1691 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
1694 NM/19 Bugzilla 745: TLS version reporting.
1695 Patch provided by Phil Pennock.
1697 NM/20 Bugzilla 167: bool: condition support.
1698 Patch provided by Phil Pennock.
1700 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
1701 clients. Patch provided by Phil Pennock.
1703 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
1704 Patch provided by Brad "anomie" Jorsch.
1706 NM/23 Bugzilla 687: Fix misparses in eximstats.
1707 Patch provided by Heiko Schlittermann.
1709 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
1710 Patch provided by Heiko Schlittermann.
1712 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
1713 plus update to original patch.
1715 NM/26 Bugzilla 799: Documentation correction for ratelimit.
1717 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
1718 Patch provided by David Brownlee.
1720 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
1722 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
1724 NM/30 Bugzilla 888: TLS documentation bugfixes.
1726 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
1728 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
1729 Unlike the original bugzilla I have changed all shell scripts in src tree.
1731 NM/33 Bugzilla 898: Transport filter timeout fix.
1732 Patch by Todd Rinaldo.
1734 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
1735 Patch by Serge Demonchaux.
1737 NM/35 Bugzilla 39: Base64 decode bug fixes.
1738 Patch by Jakob Hirsch.
1740 NM/36 Bugzilla 909: Correct connect() call in dcc code.
1742 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
1744 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
1746 NM/39 Bugzilla 911: Fixed MakeLinks build script.
1752 TK/01 Add preliminary DKIM support. Currently requires a forked version of
1753 ALT-N's libdkim that I have put here:
1754 http://duncanthrax.net/exim-experimental/
1756 Note to Michael Haardt: I had to rename some vars in sieve.c. They
1757 were called 'true' and it seems that C99 defines that as a reserved
1758 keyword to be used with 'bool' variable types. That means you could
1759 not include C99-style headers which use bools without triggering
1760 build errors in sieve.c.
1762 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
1763 as mailq or other aliases. Changed the --help handling significantly
1764 to do whats expected. exim_usage() emits usage/help information.
1766 SC/01 Added the -bylocaldomain option to eximstats.
1768 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
1770 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
1772 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
1778 PH/01 Another patch from the Sieve maintainer.
1780 PH/02 When an IPv6 address is converted to a string for single-key lookup
1781 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
1782 dots are used instead of colons so that keys in lsearch files need not
1783 contain colons. This was done some time before quoting was made available
1784 in lsearch files. However, iplsearch files do require colons in IPv6 keys
1785 (notated using the quote facility) so as to distinguish them from IPv4
1786 keys. This meant that lookups for IP addresses in host lists did not work
1787 for iplsearch lookups.
1789 This has been fixed by arranging for IPv6 addresses to be expressed with
1790 colons if the lookup type is iplsearch. This is not incompatible, because
1791 previously such lookups could never work.
1793 The situation is now rather anomolous, since one *can* have colons in
1794 ordinary lsearch keys. However, making the change in all cases is
1795 incompatible and would probably break a number of configurations.
1797 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
1800 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
1801 conversion specification without a maximum field width, thereby enabling
1802 a rogue spamd server to cause a buffer overflow. While nobody in their
1803 right mind would setup Exim to query an untrusted spamd server, an
1804 attacker that gains access to a server running spamd could potentially
1805 exploit this vulnerability to run arbitrary code as the Exim user.
1807 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
1808 $primary_hostname instead of what libspf2 thinks the hosts name is.
1810 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
1811 a directory entry by the name of the lookup key. Previously, if a
1812 symlink pointed to a non-existing file or a file in a directory that
1813 Exim lacked permissions to read, a lookup for a key matching that
1814 symlink would fail. Now it is enough that a matching directory entry
1815 exists, symlink or not. (Bugzilla 503.)
1817 PH/03 The body_linecount and body_zerocount variables are now exported in the
1820 PH/04 Added the $dnslist_matched variable.
1822 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
1823 This means they are set thereafter only if the connection becomes
1826 PH/06 Added the client_condition to authenticators so that some can be skipped
1827 by clients under certain conditions.
1829 PH/07 The error message for a badly-placed control=no_multiline_responses left
1830 "_responses" off the end of the name.
1832 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
1834 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
1835 (without spaces) instead of just copying the configuration text.
1837 PH/10 Added the /noupdate option to the ratelimit ACL condition.
1839 PH/11 Added $max_received_linelength.
1841 PH/12 Added +ignore_defer and +include_defer to host lists.
1843 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
1844 way in which PCRE > 7.0 is built.
1846 PH/14 Implemented queue_only_load_latch.
1848 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
1849 MAIL command. The effect was to mangle the value on 64-bit systems.
1851 PH/16 Another patch from the Sieve maintainer.
1853 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
1855 PH/18 If a system quota error occurred while trying to create the file for
1856 a maildir delivery, the message "Mailbox is full" was not appended to the
1857 bounce if the delivery eventually timed out. Change 4.67/27 below applied
1858 only to a quota excession during the actual writing of the file.
1860 PH/19 It seems that peer DN values may contain newlines (and other non-printing
1861 characters?) which causes problems in log lines. The DN values are now
1862 passed through string_printing() before being added to log lines.
1864 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
1865 and InterBase are left for another time.)
1867 PH/21 Added message_body_newlines option.
1869 PH/22 Guard against possible overflow in moan_check_errorcopy().
1871 PH/23 POSIX allows open() to be a macro; guard against that.
1873 PH/24 If the recipient of an error message contained an @ in the local part
1874 (suitably quoted, of course), incorrect values were put in $domain and
1875 $local_part during the evaluation of errors_copy.
1881 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
1882 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
1885 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
1886 issue a MAIL command.
1888 PH/02 In an ACL statement such as
1890 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
1892 if a client was not listed at all, or was listed with a value other than
1893 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
1894 the condition was not true (as it should be), so access was not denied.
1895 The bug was that the ! inversion was incorrectly passed on to the second
1896 item. This has been fixed.
1898 PH/03 Added additional dnslists conditions == and =& which are different from
1899 = and & when the dns lookup returns more than one IP address.
1901 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
1902 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
1904 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
1905 FSYNC, which compiles an option called disable_fsync that allows for
1906 bypassing fsync(). The documentation is heavily laced with warnings.
1908 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
1910 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
1911 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
1912 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
1913 including adding "make clean"; (3) Added -fPIC when compiling the test
1914 dynamically loaded module, to get rid of a warning.
1916 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
1917 message fails, move_frozen_messages = true and ignore_bounce_errors_after
1918 = 0s. The bug is otherwise harmless.
1920 PH/07 There was a bug in the dovecot authenticator such that the value of
1921 $auth1 could be overwritten, and so not correctly preserved, after a
1922 successful authentication. This usually meant that the value preserved by
1923 the server_setid option was incorrect.
1925 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
1927 PH/09 Installed PCRE release 7.0.
1929 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
1930 run for batched SMTP input. It is now run at the start of every message
1931 in the batch. While fixing this I discovered that the process information
1932 (output by running exiwhat) was not always getting set for -bs and -bS
1933 input. This is fixed, and it now also says "batched" for BSMTP.
1935 PH/11 Added control=no_pipelining.
1937 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
1938 patch, slightly modified), and move the expansion of helo_data till after
1939 the connection is made in the smtp transport (so it can use these
1942 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
1944 PH/14 Added log_selector = +pid.
1946 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
1948 PH/16 Add ${if forany and ${if forall.
1950 PH/17 Added dsn_from option to vary the From: line in DSNs.
1952 PH/18 Flush SMTP output before performing a callout, unless control =
1953 no_callout_flush is set.
1955 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
1956 was true (the default) a successful delivery failed to delete the retry
1957 item, thus causing premature timeout of the address. The bug is now
1960 PH/20 Added hosts_avoid_pipelining to the smtp transport.
1962 PH/21 Long custom messages for fakedefer and fakereject are now split up
1963 into multiline reponses in the same way that messages for "deny" and
1964 other ACL rejections are.
1966 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
1967 with slight modification.
1969 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
1970 draft, changing the syntax and factoring some duplicate code".
1972 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
1973 for deliveries of the second and subsequent messages over the same SMTP
1976 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
1977 ${reduce, with only minor "tidies".
1979 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
1981 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
1982 expansion side effects.
1984 PH/27 When a message times out after an over-quota error from an Exim-imposed
1985 quota, the bounce message says "mailbox is full". This message was not
1986 being given when it was a system quota that was exceeded. It now should
1989 MH/03 Made $recipients available in local_scan(). local_scan() already has
1990 better access to the recipient list through recipients_list[], but
1991 $recipients can be useful in postmaster-provided expansion strings.
1993 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
1994 in the case of a MAIL command with additional options following the
1995 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
1996 were accidentally chopped off.
1998 PH/29 SMTP synchronization checks are implemented when a command is read -
1999 there is a check that no more input is waiting when there shouldn't be
2000 any. However, for some commands, a delay in an ACL can mean that it is
2001 some time before the response is written. In this time, more input might
2002 arrive, invalidly. So now there are extra checks after an ACL has run for
2003 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
2004 pipelining has not been advertised.
2006 PH/30 MH's patch to allow iscntrl() characters to be list separators.
2008 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
2009 returned in the SMTP response when smtp_return_error_details was false.
2010 This has been fixed.
2012 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
2013 instead of the C I/O that was originally supplied, because problems were
2014 reported on Solaris.
2016 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
2017 Exim which did not show up earlier: it was assuming that a call to
2018 SSL_CTX_set_info_callback() might give an error value. In fact, there is
2019 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
2020 was a macro that became an assignment, so it seemed to work. This has
2021 changed to a proper function call with a void return, hence the compile
2022 error. Exim's code has been fixed.
2024 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
2027 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
2029 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
2031 PH/37 If a message is not accepted after it has had an id assigned (e.g.
2032 because it turns out to be too big or there is a timeout) there is no
2033 "Completed" line in the log. When some messages of this type were
2034 selected by exigrep, they were listed as "not completed". Others were
2035 picked up by some special patterns. I have improved the selection
2036 criteria to be more general.
2038 PH/38 The host_find_failed option in the manualroute router can now be set
2039 to "ignore", to completely ignore a host whose IP address cannot be
2040 found. If all hosts are ignored, the behaviour is controlled by the new
2041 host_all_ignored option.
2043 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
2044 homing or because of multiple MX records with /mx) generated more than
2045 one IP address, and the following item turned out to be the local host,
2046 all the secondary addresses of the first item were incorrectly removed
2047 from the list, along with the local host and any following hosts (which
2048 is what is supposed to happen).
2050 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
2051 whoever called Exim into the -H file. In the case of the daemon it was
2052 behaving confusingly. When first started, it used values for whoever
2053 started the daemon, but after a SIGHUP it used the Exim user (because it
2054 calls itself on a restart). I have changed the code so that it now always
2057 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
2058 message are rejected with the same error (e.g. no authentication or bad
2059 sender address), and a DATA command is nevertheless sent (as can happen
2060 with PIPELINING or a stupid MUA), the error message that was given to the
2061 RCPT commands is included in the rejection of the DATA command. This is
2062 intended to be helpful for MUAs that show only the final error to their
2065 PH/42 Another patch from the Sieve maintainer.
2067 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
2068 Eximstats - Fixed some broken HTML links and added missing column headers
2070 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
2071 columns for Rejects, Temp Rejects, Ham, and Spam rows.
2073 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
2075 PH/43 Yet another patch from the Sieve maintainer.
2077 PH/44 I found a way to check for a TCP/IP connection going away before sending
2078 the response to the final '.' that terminates a message, but only in the
2079 case where the client has not sent further data following the '.'
2080 (unfortunately, this is allowed). However, in many cases there won't be
2081 any further data because there won't be any more messages to send. A call
2082 to select() can be used: if it shows that the input is "ready", there is
2083 either input waiting, or the socket has been closed. An attempt to read
2084 the next input character can distinguish the two cases. Previously, Exim
2085 would have sent an OK response which the client would never have see.
2086 This could lead to message repetition. This fix should cure that, at
2087 least in a lot of common cases.
2089 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
2090 advertised in response to EHLO.
2096 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
2097 fixed by 4.65/MH/01 (is this a record?) are fixed:
2099 (i) An empty string was always treated as zero by the numeric comparison
2100 operators. This behaviour has been restored.
2102 (ii) It is documented that the numeric comparison operators always treat
2103 their arguments as decimal numbers. This was broken in that numbers
2104 starting with 0 were being interpreted as octal.
2106 While fixing these problems I realized that there was another issue that
2107 hadn't been noticed. Values of message_size_limit (both the global option
2108 and the transport option) were treated as octal if they started with 0.
2109 The documentation was vague. These values are now always treated as
2110 decimal, and I will make that clear in the documentation.
2116 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
2117 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
2120 MH/01 Don't check that the operands of numeric comparison operators are
2121 integers when their expansion is in "skipping" mode (fixes bug
2122 introduced by 4.64-PH/07).
2124 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
2125 child addresses, Exim now panics and dies. Previously, because the count
2126 is held in a short int, deliveries were likely to be lost. As such a
2127 large number of recipients for a single message is ridiculous
2128 (performance will be very, very poor), I have chosen to impose a limit
2129 rather than extend the field.
2135 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
2136 leftover -K file (the existence of which was triggered by #402).
2137 While we were at it, introduced process PID as part of the -K
2138 filename. This should rule out race conditions when creating
2141 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
2142 processing considerably. Previous code took too long for large mails,
2143 triggering a timeout which in turn triggers #401.
2145 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
2146 in the DK code in transports.c. sendfile() is not really portable,
2147 hence the _LINUX specificness.
2149 TF/01 In the add_headers option to the mail command in an Exim filter,
2150 there was a bug that Exim would claim a syntax error in any
2151 header after the first one which had an odd number of characters
2154 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
2155 callout verification, Exim cached a "reject" for the entire domain. This
2156 is correct for most verifications, but it is not correct for a recipient
2157 verification with use_sender or use_postmaster set, because in that case
2158 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
2159 case of MAIL FROM:<> rejection from other early rejections (e.g.
2160 rejection of HELO). When verifying a recipient using a non-null MAIL
2161 address, the cache is ignored if it shows MAIL FROM:<> rejection.
2162 Whatever the result of the callout, the value of the domain cache is
2163 left unchanged (for any other kind of callout, getting as far as trying
2164 RCPT means that the domain itself is ok).
2166 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
2169 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
2170 manifest itself as EPIPE rather than ECONNECT. When tidying away a
2171 session, the daemon ignores ECONNECT errors and logs others; it now
2172 ignores EPIPE as well.
2174 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
2175 (quoted-printable decoding).
2177 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
2178 later the small subsequent patch to fix an introduced bug.
2180 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
2182 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
2184 PH/08 An error is now given if message_size_limit is specified negative.
2186 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
2187 to be given (somewhat) arbitrary names.
2189 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
2192 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
2193 miscellaneous code fixes
2195 PH/10 Added the log_reject_target ACL modifier to specify where to log
2198 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
2199 hostname. This is wrong, because it relates to the incoming message (and
2200 probably the interface on which it is arriving) and not to the outgoing
2201 callout (which could be using a different interface). This has been
2202 changed to use the value of the helo_data option from the smtp transport
2203 instead - this is what is used when a message is actually being sent. If
2204 there is no remote transport (possible with a router that sets up host
2205 addresses), $smtp_active_hostname is used.
2207 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
2208 tweaks were necessary in order to get it to work (see also 21 below):
2209 (a) The code assumed that strncpy() returns a negative number on buffer
2210 overflow, which isn't the case. Replaced with Exim's string_format()
2212 (b) There were several signed/unsigned issues. I just did the minimum
2213 hacking in of casts. There is scope for a larger refactoring.
2214 (c) The code used strcasecmp() which is not a standard C function.
2215 Replaced with Exim's strcmpic() function.
2216 (d) The code set only $1; it now sets $auth1 as well.
2217 (e) A simple test gave the error "authentication client didn't specify
2218 service in request". It would seem that Dovecot has changed its
2219 interface. Fortunately there's a specification; I followed it and
2220 changed what the client sends and it appears to be working now.
2222 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
2225 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
2226 address A is aliased to B and C, where B exists and C does not. Without
2227 -v the output is "A verified" because verification stops after a
2228 successful redirection if more than one address is generated. However,
2229 with -v the child addresses are also verified. Exim was outputting "A
2230 failed to verify" and then showing the successful verification for C,
2231 with its parentage. It now outputs "B failed to verify", showing B's
2232 parentage before showing the successful verification of C.
2234 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
2235 look up a TXT record in a specific list after matching in a combined
2238 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
2239 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
2240 they consult the DNS. I had assumed they would set it the way they
2241 wanted; and indeed my experiments on Linux seem to show that in some
2242 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
2243 To be on the safe side, however, I have now made the interface to
2244 host_find_byname() similar to host_find_bydns(), with an argument
2245 containing the DNS resolver options. The host_find_byname() function now
2246 sets these options at its start, just as host_find_bydns() does. The smtp
2247 transport options dns_qualify_single and dns_search_parents are passed to
2248 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
2249 of host_find_byname() use the default settings of RES_DEFNAMES
2250 (qualify_single) but not RES_DNSRCH (search_parents).
2252 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
2253 spool_read_header() do less string testing, by means of a preliminary
2254 switch on the second character of optional "-foo" lines. (This is
2255 overdue, caused by the large number of possibilities that now exist.
2256 Originally there were few.) While I was there, I also converted the
2257 str(n)cmp tests so they don't re-test the leading "-" and the first
2258 character, in the hope this might squeeze out yet more improvement.
2260 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
2261 flag allowing group syntax was set by the header_syntax check but not
2262 turned off, possible causing trouble later; (2) The flag was not being
2263 set at all for the header_verify test, causing "group"-style headers to
2264 be rejected. I have now set it in this case, and also caused header_
2265 verify to ignore an empty address taken from a group. While doing this, I
2266 came across some other cases where the code for allowing group syntax
2267 while scanning a header line wasn't quite right (mostly, not resetting
2268 the flag correctly in the right place). These bugs could have caused
2269 trouble for malformed header lines. I hope it is now all correct.
2271 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
2272 with the "reply" argument non-NULL. The code, however (which originally
2273 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
2274 but it didn't always do it. This confused somebody who was copying the
2275 code for some other use. I have removed all the tests.
2277 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
2278 feature that was used to support insecure browsers during the U.S. crypto
2279 embargo. It requires special client support, and Exim is probably the
2280 only MTA that supported it -- and would never use it because real RSA is
2281 always available. This code has been removed, because it had the bad
2282 effect of slowing Exim down by computing (never used) parameters for the
2283 RSA_EXPORT functionality.
2285 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
2286 authenticator to fail if there's a tab character in the incoming data
2287 (there should never be unless someone is messing about, as it's supposed
2288 to be base64-encoded). Also added, on Timo's advice, the "secured" option
2289 if the connection is using TLS or if the remote IP is the same as the
2290 local IP, and the "valid-client-cert option" if a client certificate has
2293 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
2294 authenticators. This can be used for authorization after authentication
2295 succeeds. (In the case of plaintext, it servers for both authentication
2298 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
2299 if any retry times were supplied.
2301 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
2302 connection, where there is no client IP address to check. In this
2303 situation, the verify now always succeeds.
2305 PH/25 Applied John Jetmore's -Mset patch.
2307 PH/26 Added -bem to be like -Mset, but loading a message from a file.
2309 PH/27 In a string expansion for a processed (not raw) header when multiple
2310 headers of the same name were present, leading whitespace was being
2311 removed from all of them, but trailing whitespace was being removed only
2312 from the last one. Now trailing whitespace is removed from each header
2313 before concatenation. Completely empty headers in a concatenation (as
2314 before) are ignored.
2316 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
2317 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
2319 PH/29 [Removed. This was a change that I later backed out, and forgot to
2320 correct the ChangeLog entry (that I had efficiently created) before
2321 committing the later change.]
2323 PH/30 Exim was sometimes attempting to deliver messages that had suffered
2324 address errors (4xx response to RCPT) over the same connection as other
2325 messages routed to the same hosts. Such deliveries are always "forced",
2326 so retry times are not inspected. This resulted in far too many retries
2327 for the affected addresses. The effect occurred only when there were more
2328 hosts than the hosts_max_try setting in the smtp transport when it had
2329 the 4xx errors. Those hosts that it had tried were not added to the list
2330 of hosts for which the message was waiting, so if all were tried, there
2331 was no problem. Two fixes have been applied:
2333 (i) If there are any address or message errors in an SMTP delivery, none
2334 of the hosts (tried or untried) are now added to the list of hosts
2335 for which the message is waiting, so the message should not be a
2336 candidate for sending over the same connection that was used for a
2337 successful delivery of some other message. This seems entirely
2338 reasonable: after all the message is NOT "waiting for some host".
2339 This is so "obvious" that I'm not sure why it wasn't done
2340 previously. Hope I haven't missed anything, but it can't do any
2341 harm, as the worst effect is to miss an optimization.
2343 (ii) If, despite (i), such a delivery is accidentally attempted, the
2344 routing retry time is respected, so at least it doesn't keep
2345 hammering the server.
2347 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
2348 in ${readsocket because some servers need this prod.
2350 PH/32 Added some extra debug output when updating a wait-xxx database.
2352 PH/33 The hint "could be header name not terminated by colon", which has been
2353 given for certain expansion errors for a long time, was not being given
2354 for the ${if def:h_colon_omitted{... case.
2356 PH/34 The spec says: "With one important exception, whenever a domain list is
2357 being scanned, $domain contains the subject domain." There was at least
2358 one case where this was not true.
2360 PH/35 The error "getsockname() failed: connection reset by peer" was being
2361 written to the panic log as well as the main log, but it isn't really
2362 panic-worthy as it just means the connection died rather early on. I have
2363 removed the panic log writing for the ECONNRESET error when getsockname()
2366 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
2367 runs only) independently of the message's sender address. This meant
2368 that, if the 4xx error was in fact related to the sender, a different
2369 message to the same recipient with a different sender could confuse
2370 things. In particualar, this can happen when sending to a greylisting
2371 server, but other circumstances could also provoke similar problems.
2372 I have changed the default so that the retry time for these errors is now
2373 based a combination of the sender and recipient addresses. This change
2374 can be overridden by setting address_retry_include_sender=false in the
2377 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
2378 remote server are returned as part of bounce messages. This was not
2379 happening for LMTP over a pipe (the lmtp transport), but now it is the
2380 same for both kinds of LMTP.
2382 PH/38 Despite being documented as not happening, Exim was rewriting addresses
2383 in header lines that were in fact CNAMEs. This is no longer the case.
2385 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
2386 and queue runs started by the daemon processed all messages. This has
2387 been fixed so that -R and -S can now usefully be given with -q<time>.
2389 PH/40 Import PCRE release 6.7 (fixes some bugs).
2391 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
2393 PH/42 Give an error if -q is specified more than once.
2395 PH/43 Renamed the variables $interface_address and $interface_port as
2396 $received_ip_address and $received_port, to make it clear that these
2397 values apply to message reception, and not to the outgoing interface when
2398 a message is delivered. (The old names remain recognized, of course.)
2400 PH/44 There was no timeout on the connect() call when using a Unix domain
2401 socket in the ${readsocket expansion. There now is.
2403 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
2404 be meaningful with "accept".
2406 SC/01 Eximstats V1.43
2407 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
2409 SC/02 Eximstats V1.44
2410 Use a glob alias rather than an array ref in the generated
2411 parser. This improves both readability and performance.
2413 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
2414 Collect SpamAssassin and rejection statistics.
2415 Don't display local sender or destination tables unless
2416 there is data to show.
2417 Added average volumes into the top table text output.
2419 SC/04 Eximstats V1.46
2420 Collect data on the number of addresses (recipients)
2421 as well as the number of messages.
2423 SC/05 Eximstats V1.47
2424 Added 'Message too big' to the list of mail rejection
2425 reasons (thanks to Marco Gaiarin).
2427 SC/06 Eximstats V1.48
2428 Mainlog lines which have GMT offsets and are too short to
2429 have a flag are now skipped.
2431 SC/07 Eximstats V1.49 (Alain Williams)
2432 Added the -emptyok flag.
2434 SC/08 Eximstats V1.50
2435 Fixes for obtaining the IP address from reject messages.
2437 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
2438 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
2439 whitesspace changes from 4.64-PH/27
2441 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
2444 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
2445 are found, allow negative numbers in numeric criteria)
2447 JJ/06 exipick.20061117.2, added new $message_body_missing variable
2449 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
2450 to match changes made in 4.64-PH/43
2452 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
2454 PH/47 Put in an explicit test for a DNS lookup of an address record where the
2455 "domain" is actually an IP address, and force a failure. This locks out
2456 those revolvers/nameservers that support "A-for-A" lookups, in
2457 contravention of the specifications.
2459 PH/48 When a host name was looked up from an IP address, and the subsequent
2460 forward lookup of the name timed out, the host name was left in
2461 $sender_host_name, contrary to the specification.
2463 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
2464 restricted to single-key lookups, Exim was not diagnosing an error if
2465 * or *@ was used with a query-style lookup.
2467 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
2469 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
2470 long ago, but noone interested enough thought of it. Let's just say that
2471 the "1.1" means that there are some new functions that weren't there at
2472 some point in the past.
2474 PH/51 Error processing for expansion failure of helo_data from an smtp
2475 transport during callout processing was broken.
2477 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
2478 tested/used via the -bh/-bhc/-bs options.
2480 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
2481 bug, fixed in subsequent PCRE releases).
2483 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
2484 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
2486 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
2492 SC/01 Use a glob alias rather than an array ref in eximstats generated
2493 parser. This improves both readability and performance.
2495 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
2496 Don't display local sender or destination tables in eximstats unless
2497 there is data to show.
2498 Added average volumes into the eximstats top table text output.
2500 SC/03 Collect data on the number of addresses (recipients) as well
2501 as the number of messages in eximstats.
2503 TF/01 Correct an error in the documentation for the redirect router. Exim
2504 does (usually) call initgroups() when daemonizing.
2506 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
2507 with consistent privilege compared to when running as a daemon.
2509 TF/03 Note in the spec that $authenticated_id is not set for local
2510 submissions from trusted users.
2512 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
2513 Thanks to Dean Brooks <dean@iglou.com> for the patch.
2515 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
2516 by adding some example configuration directives to the default
2517 configuration file. A little bit of work is required to uncomment the
2518 directives and define how usernames and passwords are checked, but
2519 there is now a framework to start from.
2521 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
2522 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
2523 without this. I don't know how relevant this is to other LDAP libraries.
2525 PH/02 Add the verb name to the "unknown ACL verb" error.
2527 PH/03 Magnus Holmgren's patch for filter_prepend_home.
2529 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
2531 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
2532 directory not expanded when it should be if an expanded home directory
2533 was set for the address (which is overridden by the transport).
2535 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
2538 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
2539 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
2540 because it is too late at that time, and has no effect.
2542 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
2543 security issue with \' (bugzilla #107). I could not use the
2544 PQescapeStringConn() function, because it needs a PGconn value as one of
2547 PH/08 When testing addresses using -bt, indicate those final addresses that
2548 are duplicates that would not cause an additional delivery. At least one
2549 person was confused, thinking that -bt output corresponded to deliveries.
2550 (Suppressing duplicates isn't a good idea as you lose the information
2551 about possibly different redirections that led to the duplicates.)
2553 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
2554 systems where poll() doesn't work, in particular OS X.
2556 PH/10 Added more information to debugging output for retry time not reached.
2558 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
2559 operations in malware.c.
2561 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
2564 PH/13 If write_rejectlog was set false when logging was sent to syslog with
2565 syslog_duplication set false, log lines that would normally be written
2566 both the the main log and to the reject log were not written to syslog at
2569 PH/14 In the default configuration, change the use of "message" in ACL warn
2570 statements to "add_header".
2572 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
2573 not followed by a command (e.g. "seen endif").
2575 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
2576 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
2579 PH/17 Added extra conditions to the default value of delay_warning_condition
2583 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
2584 { match{$h_precedence:}{(?i)bulk|list|junk} } \
2585 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
2588 The Auto-Submitted: and various List- headers are standardised, whereas I
2589 don't think Precedence: ever was.
2591 PH/18 Refactored debugging code in route_finduser() to show more information,
2592 in particular, the error code if getpwnam() issues one.
2594 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
2595 This is apparently needed in addition to the PH/07 change above to avoid
2596 any possible encoding problems.
2598 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
2599 but not after initializing Perl.
2601 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
2602 output them only if debugging. By default they are written stderr,
2603 apparently, which is not desirable.
2605 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
2608 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
2611 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
2613 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
2614 authenticated or an ident call has been made. Suppress the default
2615 values for $authenticated_id and $authenticated_sender (but permit -oMai
2616 and -oMas) when testing with -bh.
2618 PH/24 Re-jigged the order of the tests in the default configuration so that the
2619 tests for valid domains and recipients precede the DNS black list and CSA
2620 tests, on the grounds that those ones are more expensive.
2622 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
2623 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
2624 This bug exists in every version of Exim that I still have, right back to
2627 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
2628 However, an attempt to turn on case-sensitivity in a regex key by
2629 including (?-i) didn't work because the subject string was already
2630 lowercased, and the effects were non-intuitive. It turns out that a
2631 one-line patch can be used to allow (?-i) to work as expected.
2637 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
2638 other effects) broke the use of negated acl sub-conditions.
2640 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
2643 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
2644 "Deny" causes Exim to reject the incoming connection with a 554 error.
2645 Unfortunately, if there is a major crisis, such as a disk failure,
2646 tcp-wrappers gives "deny", whereas what one would like would be some
2647 kind of temporary error. A kludge has been added to help with this.
2648 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
2649 554 error is used if errno is still zero or contains ENOENT (which occurs
2650 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
2653 PH/03 Add -lutil to the default FreeBSD LIBS setting.
2655 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
2656 errors. Otherwise a message that provokes a temporary error (when other
2657 messages do not) can cause a whole host to time out.
2659 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
2660 addresses were routed directly to files or pipes from a redirect router.
2661 File deliveries just didn't batch; pipe deliveries might have suffered
2664 PH/06 A failure to get a lock for a hints database would erroneously always say
2665 "Failed to get write lock", even when it was really a read lock.
2667 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
2668 of 0600. This has been changed to use the value of the lockfile_mode
2669 option (which defaults to 0600).
2671 PH/08 Applied small patch from the Sieve maintainer.
2673 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
2674 folder from quota calculations, a direct delivery into this folder messed
2675 up the contents of the maildirsize file. This was because the regex was
2676 used only to exclude .Trash (or whatever) when the size of the mailbox
2677 was calculated. There was no check that a delivery was happening into an
2678 excluded directory. This bug has been fixed by ignoring all quota
2679 processing for deliveries into excluded directories.
2681 PH/10 Added the maildirfolder_create_regex option to appendfile.
2687 PH/01 The code for finding all the local interface addresses on a FreeBSD
2688 system running IPv6 was broken. This may well have applied to all BSD
2689 systems, as well as to others that have similar system calls. The broken
2690 code found IPv4 interfaces correctly, but gave incorrect values for the
2691 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
2692 that it would not match correctly against @[] and not recognize the IPv6
2695 PH/02 The ipliteral router was not recognizing addresses of the form user@
2696 [ipv6:....] because it didn't know about the "ipv6:" prefix.
2698 PH/03 Added disable_ipv6.
2700 PH/04 Changed $reply_address to use the raw form of the headers instead of the
2701 decoded form, because it is most often used to construct To: headers
2702 lines in autoreplies, and the decoded form may well be syntactically
2703 invalid. However, $reply_address has leading white space removed, and all
2704 newlines turned into spaces so that the autoreply transport does not
2707 PH/05 If group was specified without a user on a router, and no group or user
2708 was specified on a transport, the group from the router was ignored.
2710 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
2711 for visible compile-time settings that can be used to change these
2712 numbers, for those that want even more. Backwards compatibility with old
2713 spool files has been maintained. However, going back to a previous Exim
2714 release will lost any variables that are in spool files.
2716 PH/07 Two small changes when running in the test harness: increase delay when
2717 passing a TCP/IP connection to a new process, in case the original
2718 process has to generate a bounce, and remove special handling of
2719 127.0.0.2 (sic), which is no longer necessary.
2721 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
2722 be the same on different OS.
2724 PH/09 Moved a debug statement in filter processing to avoid a race problem when
2727 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
2728 whether --show-vars was specified or not
2730 JJ/02 exipick: Added support for new ACL variable spool format introduced
2733 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
2734 syntactically invalid From: or Reply-to: line, and a filter used this to
2735 generate an autoreply, and therefore failed to obtain an address for the
2736 autoreply, Exim could try to deliver to a non-existent relative file
2737 name, causing unrelated and misleading errors. What now happens is that
2738 it logs this as a hard delivery error, but does not attempt to create a
2741 PH/11 The exinext utility has a -C option for testing purposes, but although
2742 the given file was scanned by exinext itself; it wasn't being passed on
2743 when Exim was called.
2745 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
2746 an end-of-file indication when reading a command response.
2748 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
2749 compiled. In many other places in Exim, IPv6 addresses are always
2750 recognized, so I have changed this. It also means that IPv4 domain
2751 literals of the form [IPV4:n.n.n.n] are now always recognized.
2753 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
2754 used if the router is not running as root, for example, when verifying at
2755 ACL time, or when using -bh. The debugging output from this situation was
2756 non-existent - all you got was a failure to exec. I have made two
2759 (a) Failures to set uid/gid, the current directory, or a process leader
2760 in a subprocess such as that created by queryprogram now generate
2761 suitable debugging ouput when -d is set.
2763 (b) The queryprogram router detects when it is not running as root,
2764 outputs suitable debugging information if -d is set, and then runs
2765 the subprocess without attempting to change uid/gid.
2767 PH/15 Minor change to Makefile for building test_host (undocumented testing
2770 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
2771 additional section of a DNS packet that returns MX or SRV records.
2772 Instead, it always explicitly searches for A/AAAA records. This avoids
2773 major problems that occur when a DNS server includes only records of one
2774 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
2775 fixed another bug: if SRV records were looked up and the corresponding
2776 address records were *not* found in the additional section, the port
2777 values from the SRV records were lost.
2779 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
2780 using the correct key (the original address) when searching the retry
2781 rules in order to find which one to use for generating the retry hint.
2783 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
2784 adding the default one. Similarly, if it contains a Reply-To: header, the
2785 errors_reply_to option, if set, is not used.
2787 PH/19 When calculating a retry time, Exim used to measure the "time since
2788 failure" by looking at the "first failed" field in the retry record. Now
2789 it does not use this if it is later than than the arrival time of the
2790 message. Instead it uses the arrival time. This makes for better
2791 behaviour in cases where some deliveries succeed, thus re-setting the
2792 "first failed" field. An example is a quota failure for a huge message
2793 when small messages continue to be delivered. Without this change, the
2794 "time since failure" will always be short, possible causing more frequent
2795 delivery attempts for the huge message than are intended.
2796 [Note: This change was subsequently modified - see PH/04 for 4.62.]
2798 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
2799 $1, $2, $3) because the numerical variables can be reset during some
2800 expansion items (e.g. "match"), thereby losing the authentication data.
2802 PH/21 Make -bV show the size of off_t variables so that the test suite can
2803 decide whether to run tests for quotas > 2G.
2805 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
2806 mailbox_size, and mailbox_filecount in the appendfile transport. If a
2807 filecount value is greater than 2G or if a quota value is greater than 2G
2808 on a system where the size of off_t is not greater than 4, a panic error
2811 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
2812 never match. The debug and -bh output now contains an explicit error
2813 message indicating a malformed IPv4 address or mask.
2815 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
2816 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
2817 PH/23 above applies.
2819 PH/25 Do not write to syslog when running in the test harness. The only
2820 occasion when this arises is a failure to open the main or panic logs
2821 (for which there is an explicit test).
2823 PH/26 Added the /no_tell option to "control=freeze".
2825 PH/27 If a host name lookup failed very early in a connection, for example, if
2826 the IP address matched host_lookup and the reverse lookup yielded a name
2827 that did not have a forward lookup, an error message of the form "no IP
2828 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
2829 could be logged. Now it outputs the IP address instead of "NULL".
2831 PH/28 An enabling patch from MH: add new function child_open_exim2() which
2832 allows the sender and the authenticated sender to be set when
2833 submitting a message from within Exim. Since child_open_exim() is
2834 documented for local_scan(), the new function should be too.
2836 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
2837 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
2838 results in an empty string is now treated as unset.
2840 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
2842 PH/31 Added sender_verify_fail logging option.
2844 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
2845 needed by RFC 822 but not by RFC 2822 was commented out. I have now
2846 tidied the source and removed it altogether.
2848 PH/33 When a queue run was abandoned because the load average was too high, a
2849 log line was always written; now it is written only if the queue_run log
2850 selector is set. In addition, the log line for abandonment now contains
2851 information about the queue run such as the pid. This is always present
2852 in "start" and "stop" lines but was omitted from the "abandon" line.
2854 PH/34 Omit spaces between a header name and the colon in the error message that
2855 is given when verify = headers_syntax fails (if there are lots of them,
2856 the message gets confusing).
2858 PH/35 Change the default for dns_check_names_pattern to allow slashes within
2859 names, as there are now some PTR records that contain slashes. This check
2860 is only to protect against broken name servers that fall over on strange
2861 characters, so the fact that it applies to all lookups doesn't matter.
2863 PH/36 Now that the new test suite is complete, we can remove some of the
2864 special code in Exim that was needed for the old test suite. For example,
2865 sorting DNS records because real resolvers return them in an arbitrary
2866 order. The new test suite's fake resolver always returns records in the
2869 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
2870 bounces) except when queue_only is set, to avoid logging races between
2871 the different processes.
2873 PH/38 Panic-die if .include specifies a non-absolute path.
2875 PH/39 A tweak to the "H" retry rule from its user.
2877 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
2878 a label. They prevented compilation on older perls.
2880 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
2881 a warning to be raised on newish perls.
2883 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
2884 on queue. Changes to match documented behaviour of showing count of
2885 messages matching specified criteria.
2887 PH/40 Changed the default ident timeout from 30s to 5s.
2889 PH/41 Added support for the use of login_cap features, on those BSD systems
2890 that have them, for controlling the resources used by pipe deliveries.
2892 PH/42 The content-scanning code uses fopen() to create files in which to put
2893 message data. Previously it was not paying any attention to the mode of
2894 the files. Exim runs with umask(0) because the rest of the code creates
2895 files with open(), and sets the required mode explicitly. Thus, these
2896 files were ending up world-writeable. This was not a big issue, because,
2897 being within the spool directory, they were not world-accessible. I have
2898 created a function called modefopen, which takes an additional mode
2899 argument. It sets umask(777), creates the file, chmods it to the required
2900 mode, then resets the umask. All the relevant calls to fopen() in the
2901 content scanning code have been changed to use this function.
2903 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
2904 to 24 hours. This avoids potential overflow problems when processing G
2905 and H retry rules. I suspect nobody ever tinkers with this value.
2907 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
2909 PH/45 When the plaintext authenticator is running as a client, the server's
2910 challenges are checked to ensure they are valid base64 strings. By
2911 default, the authentication attempt is cancelled if an invalid string is
2912 received. Setting client_ignore_invalid_base64 true ignores these errors.
2913 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
2914 they are received. Thus, the responses can be made to depend on the
2915 challenges. If an invalid string is ignored, an empty string is placed in
2918 PH/46 Messages that are created by the autoreply transport now contains a
2919 References: header, in accordance with RFCs 2822 and 3834.
2921 PH/47 Added authenticated_sender_force to the smtp transport.
2923 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
2925 PH/49 Installed latest patch from the Sieve maintainer.
2927 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
2928 was also set, the appendfile transport was unnecessarily scanning a
2929 directory of message files (e.g. for maildir delivery) to find the count
2930 of files (along with the size), even though it did not need this
2931 information. It now does the scan only if it needs to find either the
2932 size of the count of files.
2934 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
2936 PH/52 Two bugs concerned with error handling when the smtp transport is
2939 (i) Exim was not creating retry information for temporary errors given
2940 for individual recipients after the DATA command when the smtp transport
2941 was used in LMTP mode. This meant that they could be retried too
2942 frequently, and not timed out correctly.
2944 (ii) Exim was setting the flag that allows error details to be returned
2945 for LMTP errors on RCPT commands, but not for LMTP errors for individual
2946 recipients that were returned after the DATA command.
2948 PH/53 This is related to PH/52, but is more general: for any failing address,
2949 when detailed error information was permitted to be returned to the
2950 sender, but the error was temporary, then after the final timeout, only
2951 "retry timeout exceeded" was returned. Now it returns the full error as
2952 well as "retry timeout exceeded".
2954 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
2955 do this, and (what is worse) MTAs that accept it.
2957 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
2958 will now be deprecated.
2960 PH/56 New os.c-cygwin from the Cygwin maintainer.
2962 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
2963 formats (previously only available in exim formats via -bpr, -bpru,
2964 and -bpra. Now also available in native and exiqgrep formats)
2966 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
2967 with very large, slow to parse queues
2969 JJ/08 exipick: added ! as generic prefix to negate any criteria format
2971 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
2973 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
2974 responses to authentication challenges, though it was showing the
2975 challenges; (ii) I've removed the CR characters from the debug output for
2978 PH/58 Allow for the insertion of a newline as well as a space when a string
2979 is turned into more than one encoded-word during RFC 2047 encoding. The
2980 Sieve code now uses this.
2982 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
2983 data_4xx, lost_connection, tls_required.
2985 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
2986 message was being sent as an SMTP response.
2988 PH/61 Add -l and -k options to exicyclog.
2990 PH/62 When verifying, if an address was redirected to one new address, so that
2991 verification continued, and the new address failed or deferred after
2992 having set something in $address_data, the value of $address_data was not
2993 passed back to the ACL. This was different to the case when no
2994 redirection occurred. The value is now passed back in both cases.
2996 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
2997 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
2998 use login_cap.h, so on its own it isn't the distinguishing feature. The
2999 new name refers directly to the setclassresources() function.
3001 PH/65 Added configuration files for NetBSD3.
3003 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
3005 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
3006 is preferred over IPv4.
3008 PH/68 The bounce_return_message and bounce_return_body options were not being
3009 honoured for bounces generated during the reception of non-SMTP messages.
3010 In particular, this applied to messages rejected by the ACL. This bug has
3011 been fixed. However, if bounce_return_message is true and bounce_return_
3012 body is false, the headers that are returned for a non-SMTP message
3013 include only those that have been read before the error was detected.
3014 (In the case of an ACL rejection, they have all been read.)
3016 PH/69 The HTML version of the specification is now built in a directory called
3017 spec_html instead of spec.html, because the latter looks like a path with
3018 a MIME-type, and this confuses some software.
3020 PH/70 Catch two compiler warnings in sieve.c.
3022 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
3023 function verify_get_ident() calls ip_connect() to connect a socket, but
3024 if the "connect()" function timed out, ip_connect() used to close the
3025 socket. However, verify_get_ident() also closes the socket later, and in
3026 between Exim writes to the log, which may get opened at this point. When
3027 the socket was closed in ip_connect(), the log could get the same file
3028 descriptor number as the socket. This naturally causes chaos. The fix is
3029 not to close the socket in ip_connect(); the socket should be closed by
3030 the function that creates it. There was only one place in the code where
3031 this was missing, in the iplookup router, which I don't think anybody now
3032 uses, but I've fixed it anyway.
3034 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
3035 well as to direct DNS lookups. Otherwise the handling of names in host
3036 lists is inconsistent and therefore confusing.
3042 PH/01 Two changes to the default runtime configuration:
3044 (1) Move the checks for relay_from_hosts and authenticated clients from
3045 after to before the (commented out) DNS black list checks.
3047 (2) Add control=submission to the relay_from_hosts and authenticated
3048 clients checks, on the grounds that messages accepted by these
3049 statements are most likely to be submissions.
3051 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
3053 (1) Generate an error if the third argument for the ${prvs expansion is
3056 (2) Treat a missing third argument of ${prvscheck as if it were an empty
3059 (3) Reset the variables that are obtained from the first argument of
3060 ${prvscheck and used in the second argument before leaving the code,
3061 because their memory is reclaimed, so using them afterwards may do
3064 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
3065 one (it's much easier than Tom thought :-).
3067 (5) Because of (4), we can now allow for the use of $prvscheck_result
3068 inside the third argument.
3070 PH/03 For some reason, the default setting of PATH when running a command from
3071 a pipe transport was just "/usr/bin". I have changed it to
3074 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
3075 anything to be listed in the output from -bV.
3077 PH/05 When a filter generated an autoreply, the entire To: header line was
3078 quoted in the delivery log line, like this:
3080 => >A.N.Other <ano@some.domain> <original@ddress> ...
3082 This has been changed so that it extracts the operative address. There
3083 may be more than one such address. If so, they are comma-separated, like
3086 => >ano@some.domain,ona@other.domain <original@ddress> ...
3088 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
3089 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
3090 not being looked up in the rDNS to get a host name, Exim was showing the
3091 IP address twice in Received: lines, even though the IP addresses were
3092 identical. For example:
3094 Received: from [1.2.3.4] (helo=[1.2.3.4])
3096 However, if the real host name was known, it was omitting the HELO data
3097 if it matched the actual IP address. This has been tidied up so that it
3098 doesn't show the same IP address twice.
3100 PH/07 When both +timestamp and +memory debugging was on, the value given by
3101 $tod_xxx expansions could be wrong, because the tod_stamp() function was
3102 called by the debug printing, thereby overwriting the timestamp buffer.
3103 Debugging no longer uses the tod_stamp() function when +timestamp is set.
3105 PH/08 When the original message was included in an autoreply transport, it
3106 always said "this is a copy of the message, including all the headers",
3107 even if body_only or headers_only was set. It now gives an appropriate
3110 PH/09 Applied a patch from the Sieve maintainer which:
3112 o fixes some comments
3113 o adds the (disabled) notify extension core
3114 o adds some debug output for the result of if/elsif tests
3115 o points to the current vacation draft in the documentation
3116 and documents the missing references header update
3120 o fixes a bug in processing the envelope test (when testing
3121 multiple envelope elements, the last element determinted the
3124 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
3125 Electronic Mail") by including:
3127 Auto-submitted: auto-generated
3129 in the messages that it generates (bounce messages and others, such as
3130 warnings). In the case of bounce messages for non-SMTP mesages, there was
3131 also a typo: it was using "Auto_submitted" (underscore instead of
3132 hyphen). Since every message generated by Exim is necessarily in response
3133 to another message, thes have all been changed to:
3135 Auto-Submitted: auto-replied
3137 in accordance with these statements in the RFC:
3139 The auto-replied keyword:
3141 - SHOULD be used on messages sent in direct response to another
3142 message by an automatic process,
3144 - MUST NOT be used on manually-generated messages,
3146 - MAY be used on Delivery Status Notifications (DSNs) and Message
3147 Disposition Notifications (MDNs),
3149 - MUST NOT be used on messages generated by automatic or periodic
3150 processes, except for messages which are automatic responses to
3153 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
3154 to the default Received: header definition.
3156 PH/12 Added log selector acl_warn_skipped (default on).
3158 PH/13 After a successful wildlsearch lookup, discard the values of numeric
3159 variables because (a) they are in the wrong storage pool and (b) even if
3160 they were copied, it wouldn't work properly because of the caching.
3162 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
3163 checking when decoding. Apparently there are clients that generate
3164 overlong encoded strings. Why am I not surprised?
3166 PH/15 If the first argument of "${if match_address" was not empty, but did not
3167 contain an "@" character, Exim crashed. Now it writes a panic log message
3168 and treats the condition as false.
3170 PH/16 In autoreply, treat an empty string for "once" the same as unset.
3172 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
3173 extension "envelope-auth". The code is finished and in agreement with
3174 other implementations, but there is no documentation so far and in fact,
3175 nobody wrote the draft yet. This extension is currently #undef'ed, thus
3176 not changing the active code.
3178 Print executed "if" and "elsif" statements when debugging is used. This
3179 helps a great deal to understand what a filter does.
3181 Document more things not specified clearly in RFC3028. I had all this
3182 sorted out, when out of a sudden new issues came to my mind. Oops."
3184 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
3187 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
3188 canonical form (as documented). However, after a host name lookup from
3189 the IP address, check_host() was doing a simple string comparison with
3190 addresses acquired from the DNS when checking that the found name did
3191 have the original IP as one of its addresses. Since any found IPv6
3192 addresses are likely to be in abbreviated form, the comparison could
3193 fail. Luckily, there already exists a function for doing the comparison
3194 by converting both addresses to binary, so now that is used instead of
3195 the text comparison.
3197 PH/20 There was another similar case to PH/19, when a complete host name was
3198 given in a host list; looking up its IP address could give an abbreviated
3199 form, whereas the current host's name might or might not be abbreviated.
3200 The same fix has been applied.
3206 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
3207 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
3210 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
3211 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
3213 PH/03 Typo: missing ".o" in src/pcre/Makefile.
3215 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
3216 header line, restrict the check to what is listed in RFCs 2369 and 2929.
3217 Also, for "Auto-Submitted", treat anything other than "no" as
3218 non-personal, in accordance with RFC 3834. (Previously it treated
3219 anything starting "auto-" as non-personal.)
3221 TF/01 The control=submission/name=... option had a problem with syntax
3222 errors if the name included a slash character. The /name= option
3223 now slurps the rest of the string, so it can include any characters
3224 but it must come last in the list of options (after /sender_retain
3227 PH/05 Some modifications to the interface to the fake nameserver for the new
3235 TK/01 Added the "success_on_redirect" address verification option. See
3236 NewStuff for rationale and an example.
3238 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
3240 PH/02 Patch to exigrep to allow it to work on syslog lines.
3242 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
3243 fread() to skip over the body file's header line, because in Cygwin the
3244 header line is locked and is inaccessible.
3246 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
3247 co-exist for some time) to make it clear that it is the Exim ID that is
3248 referenced, not the Message-ID: header line.
3250 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
3251 string_format() function, because snprintf() does not exist on all
3254 PH/06 The use of forbid_filter_existstest now also locks out the use of the
3255 ${stat: expansion item.
3257 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
3258 protocol synchronization error", to keep the pedants happy.
3260 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
3261 well as for IRIX systems, when gcc is being used. See the host.c source
3264 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
3266 PH/10 Named domain lists were not working if used in a queue_smtp_domains
3269 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
3270 transport and to the smtp transport in LMTP mode.
3272 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
3274 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
3275 run a filter in a subprocess. This could lead to confusion in subsequent
3276 lookups in the parent process. There should also be a search_tidyup() at
3277 the end of the subprocess.
3279 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
3280 only if the host matched helo_try_verify_hosts, which caused the
3281 verification to occur when the EHLO/HELO command was issued. The ACL just
3282 tested the remembered result. Now, if a previous verification attempt has
3283 not happened, "verify = helo" does it there and then.
3285 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
3287 TK/03 Fix log output including CR from clamd.
3289 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
3290 exist provoked a memory error which could cause a segfault.
3292 PH/15 Installed PCRE 6.2
3294 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
3296 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
3297 of the problem. Specifically, suggested +O2 rather than +O1 for the
3300 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
3302 PH/20 If a delivery was routed to a non-standard port by means of an SRV
3303 record, the port was not correctly logged when the outgoing_port log
3304 selector was set (it logged the transort's default port).
3306 PH/21 Added support for host-specific ports to manualroute, queryprogram,
3307 fallback_hosts, and "hosts" in the smtp transport.
3309 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
3310 host errors such as "Connection refused".
3312 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
3313 authentication with radiusclient 0.4.9:
3315 - Error returned from rc_read_config was caught wrongly
3316 - Username/password not passed on to radius server due to wrong length.
3318 The presumption is that some radiusclient API changes for 4.51/PH/17
3319 were not taken care of correctly. The code is still untested by me (my
3320 Linux distribution still has 0.3.2 of radiusclient), but it was
3321 contributed by a Radius user.
3323 PH/24 When doing a callout, the value of $domain wasn't set correctly when
3324 expanding the "port" option of the smtp transport.
3326 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
3327 while reading a MIME header. Thanks to Tom Hughes for a patch.
3329 PH/24 Include config.h inside local_scan.h so that configuration settings are
3332 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
3333 that in an ACL for RCPT (for example), you can examine exactly what was
3336 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
3337 commands, but it was not correctly comparing the address with the actual
3338 client host address. Thus, it would show the EHLO address in Received:
3339 header lines when this was not necessary.
3341 PH/27 Added the % operator to ${eval:}.
3343 PH/28 Exim tries to create and chdir to its spool directory when it starts;
3344 it should be ignoring failures (because with -C, for example, it has lost
3345 privilege). It wasn't ignoring creation failures other than "already
3348 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
3351 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
3352 because an input line was too long, either on its own, or by virtue of
3353 too many continuations, the temporary file was not being removed, and the
3354 return code was incorrect.
3356 PH/31 Missing "BOOL" in function definition in filtertest.c.
3358 PH/32 Applied Sieve patches from the maintainer.
3360 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
3362 PH/33 Added "verify = not_blind".
3364 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
3365 Local/Makefile (with some defaults set). These are used in built scripts
3366 such as exicyclog, but they have never been used in the exim_install
3367 script (though there are many overriding facilities there). I have
3368 arranged that the exim_install script now takes note of these two
3371 PH/35 Installed configuration files for Dragonfly.
3373 PH/36 When a locally submitted message by a trusted user did not contain a
3374 From: header, and the sender address was obtained from -f or from an SMTP
3375 MAIL command, and the trusted user did not use -F to supply a sender
3376 name, $originator_name was incorrectly used when constructing a From:
3377 header. Furthermore, $originator_name was used for submission mode
3378 messages from external hosts without From: headers in a similar way,
3379 which is clearly wrong.
3381 PH/37 Added control=suppress_local_fixups.
3383 PH/38 When log_selector = +received_sender was set, and the addition of the
3384 sender made the log line's construction buffer exactly full, or one byte
3385 less than full, an overflow happened when the terminating "\n" was
3388 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
3389 when the result of a list match is failure because a DNS lookup failed.
3391 PH/40 RM_COMMAND is now used in the building process.
3393 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
3394 the "build-* directories that it finds.
3396 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
3397 address in a domain literal was a prefix of an interface address.
3399 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
3400 when verifying a sender address, unless rewrite_headers is false.
3402 PH/44 Wrote a long comment about why errors_to addresses are verified as
3403 recipients, not senders.
3405 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
3406 the ratelimit ACL was added.
3408 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
3410 PH/46 Added extra information about PostgreSQL errors to the error string.
3412 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
3413 avoiding the need to install special zones in a real server. This is
3414 backwards compatible; if it can't find the fake resolver, it drops back.
3415 Thus, both old and new test suites can be run.
3417 TF/02 Added util/ratelimit.pl
3419 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
3420 clock is set back in time.
3422 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
3423 Candler <B.Candler@pobox.com>.
3425 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
3426 OK for addresses that are the result of redirections.
3428 PH/48 A number of further additions for the benefit of the new test suite,
3429 including a fake gethostbyname() that interfaces to the fake DNS resolver
3432 TF/06 The fix for widen_domains has also been applied to qualify_single and
3433 search_parents which are the other dnslookup options that can cause
3436 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
3439 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
3441 TF/07 Exim produced the error message "an SRV record indicated no SMTP
3442 service" if it encountered an MX record with an empty target hostname.
3443 The message is now "an MX or SRV record indicated no SMTP service".
3445 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
3446 if the DNS of the sending site is misconfigured. This is quite a
3447 common situation. This change restores the behaviour of treating a
3448 helo verification defer as a failure.
3450 PH/51 If self=fail was set on a router, the bounce message did not include the
3451 actual error message.
3457 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
3459 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
3460 command itself ended in error, the underlying message about the transport
3461 filter timeout was being overwritten with the pipe command error. Now the
3462 underlying error message should be appended to the second error message.
3464 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
3466 PH/02 Reduce the amount of output that "make" produces by default. Full output
3467 can still be requested.
3469 PH/03 The warning log line about a condition test deferring for a "warn" verb
3470 was being output only once per connection, rather than after each
3471 occurrence (because it was using the same function as for successful
3472 "warn" verbs). This seems wrong, so I have changed it.
3474 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
3475 it should not have, which might have caused a crash in the right
3476 circumstances, but probably never did.
3478 PH/04 Installed a modified version of Tony Finch's patch to make submission
3479 mode fix the return path as well as the Sender: header line, and to
3480 add a /name= option so that you can make the user's friendly name appear
3483 TF/03 Added the control = fakedefer ACL modifier.
3485 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
3486 Mark Lowes for thorough testing.
3488 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
3490 TK/03 Merged latest SRS patch from Miles Wilton.
3492 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
3493 with the definition in sysexits.h (which is #included earlier).
3494 Fortunately, Exim does not actually use EX_OK. The code used to try to
3495 preserve the sysexits.h value, by assumimg that macro definitions were
3496 scanned for macro replacements. I have been disabused of this notion,
3497 so now the code just undefines EX_OK before #including unistd.h.
3499 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
3500 in the smtp transport. When a block could not be written in a single
3501 write() function, the timeout was being re-applied to each part-write.
3502 This seems wrong - if the receiver was accepting one byte at a time it
3503 would take for ever. The timeout is now adjusted when this happens. It
3504 doesn't have to be particularly precise.
3506 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
3507 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
3509 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
3510 without a domain if the check to <postmaster@domain> fails.
3512 SC/01 Eximstats: added -xls and the ability to specify output files
3513 (patch written by Frank Heydlauf).
3515 SC/02 Eximstats: use FileHandles for outputing results.
3517 SC/03 Eximstats: allow any combination of xls, txt, and html output.
3519 SC/04 Eximstats: fixed display of large numbers with -nvr option
3521 SC/05 Eximstats: fixed merging of reports with empty tables.
3523 SC/06 Eximstats: added the -include_original_destination flag
3525 SC/07 Eximstats: removed tabs and trailing whitespace.
3527 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
3529 TK/06 MBOX spool code: Add real "From " MBOX separator line
3530 so the .eml file is really in mbox format (even though
3531 most programs do not really care). Patch from Alex Miller.
3533 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
3534 The latter is generated from $received_to and is only set if the
3535 message has one envelope recipient. SA can use these headers,
3536 obviously out-of-the-box. Patch from Alex Miller.
3538 PH/08 The ${def test on a variable was returning false if the variable's
3539 value was "0", contrary to what the specification has always said!
3540 The result should be true unless the variable is empty.
3542 PH/09 The syntax error of a character other than { following "${if
3543 def:variable_name" (after optional whitespace) was not being diagnosed.
3544 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
3545 accidental colon was present, for example, could give incorrect results.
3547 PH/10 Tidied the code in a number of places where the st_size field of a stat()
3548 result is used (not including appendfile, where other changes are about
3551 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
3552 This involved changing a lot of size variables from int to off_t. It
3553 should work with maildirs and everything.
3555 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
3556 spamd dying while we are connected to it.
3558 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
3561 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
3562 function for BDB 4.3.
3564 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
3566 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
3567 the sources for PCRE have been re-arranged and more files are now
3570 PH/15 The code I had for printing potentially long long variables in PH/11
3571 above was not the best (it lost precision). The length of off_t variables
3572 is now inspected at build time, and an appropriate printing format (%ld
3573 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
3574 to be "long long int" or "long int". This is needed for the internal
3575 formatting function string_vformat().
3577 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
3578 the configuration file to be ":syslog", then the script "guesses" where
3579 the logs files are, rather than using the compiled in default. In our
3580 case the guess is not the same as the compiled default, so the script
3581 suddenly stopped working when I started to use syslog. The patch checks
3582 to see if log_file_path is "". If so, it attempts to read it from exim
3583 with no configuration file to get the compiled in version, before it
3584 falls back to the previous guessing code."
3586 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
3587 implementing BATV in an Exim configuration. See NewStuff for the gory
3590 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
3591 Makefile that are specific to HP-UX.
3593 PH/18 If the "use_postmaster" option was set for a recipient callout together
3594 with the "random" option, the postmaster address was used as the MAIL
3595 FROM address for the random test, but not for the subsequent recipient
3596 test. It is now used for both.
3598 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
3599 patch removes a few documentation additions to RFC 3028, because the
3600 latest draft now contains them. It adds the new en;ascii-case comparator
3601 and a new error check for 8bit text in MIME parts. Comparator and
3602 require names are now matched exactly. I enabled the subaddress
3603 extension, but it is not well tested yet (read: it works for me)."
3605 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
3606 rework some of the code of TK/09 above to avoid the hardwired use of
3607 "%lld" and "long long". Replaced the call to snprintf() with a call to
3610 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
3611 records point to non-existent hosts", "retry timeout exceeded", and
3612 "retry time not reached for any host after a long failure period".
3614 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
3615 experimental DomainKeys support:
3617 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
3618 (2) On an error such as an illegally used "control", the wrong name for
3619 the control was given.
3621 These problems did NOT occur unless DomainKeys support was compiled.
3623 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
3625 PH/24 Added ${if match_ip condition.
3627 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
3628 databases so that it will be absolutely obvious if a crash occurs in the
3629 DB library. This is a regular occurrence (often caused by mis-matched
3632 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
3633 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
3634 picked up on a user's system that detects such things. There doesn't seem
3635 to be a gcc warning option for this - only an attribute that has to be
3636 put on the function's prototype. It seems that in Fedora Core 4 they have
3637 set this on a number of new functions. No doubt there will be more in due
3640 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
3641 specify a transport. However, if an address that was verified by such a
3642 router was the subject of a callout, Exim crashed because it tried to
3643 read the rcpt_include_affixes from the non-existent transport. Now it
3644 just assumes that the setting of that option is false. This bug was
3645 introduced by 4.51/PH/31.
3647 PH/28 Changed -d+all to exclude +memory, because that information is very
3648 rarely of interest, but it makes the output a lot bigger. People tend to
3649 do -d+all out of habit.
3651 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
3652 code in os-type was giving problems when libc.so lives in lib64, like on
3655 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
3656 aren't the modern standard, and it seems that some systems' include files
3657 don't always have them. Exim was already checking for some of the newer
3658 ones like T_AAAA, and defining it itself. I've added checks for all the
3659 record types that Exim uses.
3661 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
3662 not automatically generating a new one, as it is supposed to. This
3663 prevented TLS from working. If the file did exist, but contained invalid
3664 data, a new version was generated, as expected. It was only the case of a
3665 non-existent file that was broken.
3667 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
3668 with a change in libdomainkeys > 0.64.
3670 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
3671 from DNS. If the selector record carries the flag, it now has
3672 precedence over the domain-wide flag.
3674 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
3676 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
3677 the use of a port name that isn't defined in /etc/services) occurred, the
3678 message was deferred as in a normal delivery, and thus remained on the
3679 spool, instead of being failed because of the mua_wrapper setting. This
3680 is now fixed, and I tidied up some of the mua_wrapper messages at the
3683 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
3684 the messages in a hash of arrays rather than using individual hashes.
3685 This is a bit cleaner and results in dramatic memory savings, albeit
3686 at a slight CPU cost.
3688 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
3689 as requested by Marc Sherman.
3691 SC/10 Eximstats: added histograms for user specified patterns as requested
3694 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
3696 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
3697 fopen() in the content-scanning modules that did not already have it.
3703 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
3704 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
3706 TK/02 Fix ACL "control" statement not being available in MIME ACL.
3708 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
3710 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
3711 to test Sieve filters that use "vacation".
3713 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
3714 that changes the way the GnuTLS parameters are stored in the cache file.
3715 The new format can be generated externally. For backward compatibility,
3716 if the data in the cache doesn't make sense, Exim assumes it has read an
3717 old-format file, and it generates new data and writes a new file. This
3718 means that you can't go back to an older release without removing the
3721 PH/03 A redirect router that has both "unseen" and "one_time" set does not
3722 work if there are any delivery delays because "one_time" forces the
3723 parent to be marked "delivered", so its unseen clone is never tried
3724 again. For this reason, Exim now forbids the simultaneous setting of
3727 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
3728 redirected to themselves ("homonym" addresses). Read the long ChangeLog
3729 entry if you want to know the details. The fix, however, neglected to
3730 consider the case when local delivery batching is involved. The test for
3731 "previously delivered" was not happening when checking to see if an
3732 address could be batched with a previous (undelivered) one; under
3733 certain circumstances this could lead to multiple deliveries to the same
3736 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
3737 in its include files, and this causes problems building Exim.
3739 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
3740 header_syntax) but Exim was just ignoring anything given after a slash.
3741 In particular, this caused confusion with an attempt to use "verify =
3742 reverse_host_lookup/defer_ok". An error is now given when options are
3743 supplied for verify items that do not have them. (Maybe reverse_host_
3744 lookup should have a defer_ok option, but that's a different point.)
3746 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
3747 defined by RFC 821) to 2048, because there were problems with some AUTH
3748 commands, and RFC 1869 says the size should be increased for extended
3749 SMTP commands that take arguments.
3751 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
3754 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
3755 "unknown" error; now it says that the functionality isn't in the binary.
3757 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
3758 an address' error message when a string expansion fails (syntax or
3759 whatever). Otherwise the password may appear in the log. Following change
3760 PH/42 below, there is no longer a chance of it appearing in a bounce
3763 PH/11 Installed exipick version 20050225.0 from John Jetmore.
3765 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
3766 of its addresses was ever tried. (Bugzilla bug #2.)
3768 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
3769 the result incorrectly in the debug output. (It correctly added a newline
3770 to what was transported.)
3772 TF/01 Added $received_time.
3774 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
3775 commented out examples of how to interface to a virus scanner and to
3776 SpamAssassin. Also added commented examples of av_scanner and
3777 spamd_address settings.
3779 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
3780 and controls are allowed in which ACLs. There were a couple of minor
3781 errors. Some of the entries in the conditions table (which is a table of
3782 where they are NOT allowed) were getting very unwieldy; rewrote them as a
3783 negation of where the condition IS allowed.
3785 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
3787 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
3788 header file does not have a version number, so I've had to invent a new
3789 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
3790 API. The code is untested by me (my Linux distribution still has 0.3.2 of
3791 radiusclient), but it was contributed by a Radius user.
3793 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
3794 files or directories, for OpenSSL.
3796 PH/19 When an Exim process that is running as root has to create an Exim log
3797 file, it does so in a subprocess that runs as exim:exim so as to get the
3798 ownership right at creation (otherwise, other Exim processes might see
3799 the file with the wrong ownership). There was no test for failure of this
3800 fork() call, which would lead to the process getting stuck as it waited
3801 for a non-existent subprocess. Forks do occasionally fail when resources
3802 run out. I reviewed all the other calls to fork(); they all seem to check
3805 PH/20 When checking for unexpected SMTP input at connect time (before writing
3806 the banner), Exim was not dealing correctly with a non-positive return
3807 from the read() function. If the client had disconnected by this time,
3808 the result was a log entry for a synchronization error with an empty
3809 string after "input=" when read() returned zero. If read() returned -1
3810 (an event I could not check), uninitialized data bytes were printed.
3811 There were reports of junk text (parts of files, etc) appearing after
3814 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
3816 PH/22 Added support for macro redefinition, and (re)definition in between
3817 driver and ACL definitions.
3819 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
3820 forgetting to use the resulting value; it was using the unexpanded value.
3822 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
3823 hadn't been configured. The fix is from Juergen Kreileder, who
3824 understands it better than I do:
3826 "Here's what I see happening with three configured cyrus_sasl
3827 authenticators configured (plain, login, cram-md5):
3829 On startup auth_cyrus_sasl_init() gets called for each of these.
3830 This means three calls to sasl_listmech() without a specified mech_list.
3831 => SASL tests which mechs of all available mechs actually work
3832 => three warnings about OTP not working
3833 => the returned list contains: plain, login, cram-md5, digest-md5, ...
3835 With the patch, sasl_listmech() also gets called three times. But now
3836 SASL's mech_list option is set to the server_mech specified in the the
3837 authenticator. Or in other words, the answer from sasl_listmech()
3838 gets limited to just the mech you're testing for (which is different
3840 => the return list contains just 'plain' or 'login', 'cram-md5' or
3841 nothing depending on the value of ob->server_mech.
3843 I've just tested the patch: Authentication still works fine,
3844 unavailable mechs specified in the exim configuration are still
3845 caught, and the auth.log warnings about OTP are gone."
3847 PH/25 When debugging is enabled, the contents of the command line are added
3848 to the debugging output, even when log_selector=+arguments is not
3851 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
3852 answer is "GNU", and only if the return is "GNU/something" is the answer
3855 PH/27 $acl_verify_message is now set immediately after the failure of a
3856 verification in an ACL, and so is available in subsequent modifiers. In
3857 particular, the message can be preserved by coding like this:
3859 warn !verify = sender
3860 set acl_m0 = $acl_verify_message
3862 Previously, $acl_verify_message was set only while expanding "message"
3863 and "log_message" when a very denied access.
3865 PH/28 Modified OS/os.c-Linux with
3867 -#ifndef OS_LOAD_AVERAGE
3868 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
3870 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
3871 nomenclature these days.)
3873 PH/29 Installed patch from the Sieve maintainer that adds the options
3874 sieve_useraddress and sieve_subaddress to the redirect router.
3876 PH/30 In these circumstances:
3877 . Two addresses routed to the same list of hosts;
3878 . First host does not offer TLS;
3879 . First host accepts first address;
3880 . First host gives temporary error to second address;
3881 . Second host offers TLS and a TLS session is established;
3882 . Second host accepts second address.
3883 Exim incorrectly logged both deliveries with the TLS parameters (cipher
3884 and peerdn, if requested) that were in fact used only for the second
3887 PH/31 When doing a callout as part of verifying an address, Exim was not paying
3888 attention to any local part prefix or suffix that was matched by the
3889 router that accepted the address. It now behaves in the same way as it
3890 does for delivery: the affixes are removed from the local part unless
3891 rcpt_include_affixes is set on the transport.
3893 PH/32 Add the sender address, as F=<...>, to the log line when logging a
3894 timeout during the DATA phase of an incoming message.
3896 PH/33 Sieve envelope tests were broken for match types other than :is. I have
3897 applied a patch sanctioned by the Sieve maintainer.
3899 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
3900 the uid or gid is negative. A case of a negative gid caused this to be
3901 noticed. The fix allows for either to be negative.
3903 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
3904 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
3907 PH/36 The change PH/12 above was broken. Fixed it.
3909 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
3910 the grounds that routing the same address twice would always produce the
3911 same answer. This might have been true once, but it is certainly no
3912 longer true now. Routing a child address may depend on the previous
3913 routing that produced that child. Some complicated redirection strategies
3914 went wrong when messages had multiple recipients, and made Exim's
3915 behaviour dependent on the order in which the addresses were given.
3917 I have moved the duplicate checking until after the routing is complete.
3918 Exim scans the addresses that are assigned to local and remote
3919 transports, and removes any duplicates. This means that more work will be
3920 done, as duplicates will always all be routed, but duplicates are
3921 presumably rare, so I don't expect this is of any significance.
3923 For deliveries to pipes, files, and autoreplies, the duplicate checking
3924 still happens during the routing process, since they are not going to be
3927 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
3928 It corrects a timeout issue with spamd. This is Ian's comment: "The
3929 background is that sometimes spamd either never reads data from a
3930 connection it has accepted, or it never writes response data. The exiscan
3931 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
3932 blindly assumes that writes won't block so it may never time out."
3934 PH/39 Allow G after quota size as well as K and M.
3936 PH/40 The value set for $authenticated_id in an authenticator may not contain
3937 binary zeroes or newlines because the value is written to log lines and
3938 to spool files. There was no check on this. Now the value is run through
3939 the string_printing() function so that such characters are converted to
3940 printable escape sequences.
3942 PH/41 $message_linecount is a new variable that contains the total number of
3943 lines in the message. Compare $body_linecount, which is the count for the
3946 PH/42 Exim no longer gives details of delivery errors for specific addresses in
3947 bounce and delay warning messages, except in certain special cases, which
3950 (a) An SMTP error message from a remote host;
3951 (b) A message specified in a :fail: redirection;
3952 (c) A message specified in a "fail" command in a system filter;
3953 (d) A message specified in a FAIL return from the queryprogram router;
3954 (e) A message specified by the cannot_route_message router option.
3956 In these cases only, Exim does include the error details in bounce and
3957 warning messages. There are also a few cases where bland messages such
3958 as "unrouteable address" or "local delivery error" are given.
3960 PH/43 $value is now also set for the "else" part of a ${run expansion.
3962 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
3963 being worked on, but at least Exim now implements the latest version to
3966 PH/45 In a pipe transport, although a timeout while waiting for the pipe
3967 process to complete was treated as a delivery failure, a timeout while
3968 writing the message to the pipe was logged, but erroneously treated as a
3969 successful delivery. Such timeouts include transport filter timeouts. For
3970 consistency with the overall process timeout, these timeouts are now
3971 treated as errors, giving rise to delivery failures by default. However,
3972 there is now a new Boolean option for the pipe transport called
3973 timeout_defer, which, if set TRUE, converts the failures into defers for
3974 both kinds of timeout. A transport filter timeout is now identified in
3977 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
3978 systems where "make" and "gmake" are different, calling "gmake" at top
3979 level broke things. I've arranged for the value of $(MAKE) to be passed
3980 from the Makefile to this script so that it can call the same version of
3984 A note about Exim versions 4.44 and 4.50
3985 ----------------------------------------
3987 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
3988 changes of various kinds. As a consequence, a big documentation update was
3989 needed. This delayed the release for rather longer than seemed good, especially
3990 in the light of a couple of (minor) security issues. Therefore, the changes
3991 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
3992 release. So 4.44 and 4.50 are in effect two different branches that both start
3995 I have left the 4.50 change log unchanged; it contains all the changes since
3996 4.43. The change log for 4.44 is below; many of its items are identical to
3997 those for 4.50. This seems to be the most sensible way to preserve the
3998 historical information.
4004 1. Minor wording change to the doc/README.SIEVE file.
4006 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
4007 computation of the current number of files was incorrect.
4009 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4010 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4013 4. Give more explanation in the error message when the command for a transport
4014 filter fails to execute.
4016 5. There are several places where Exim runs a non-Exim command in a
4017 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4018 was being done only for the command run by the queryprogram router. It is
4019 now done for all such subprocesses. The other cases are: ${run, transport
4020 filters, and the commands run by the lmtp and pipe transports.
4022 6. Added CONFIGURE_GROUP build-time option.
4024 7. Some older OS have a limit of 256 on the maximum number of file
4025 descriptors. Exim was using setrlimit() to set 1000 as a large value
4026 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4027 systems. I've change it so that if it can't get 1000, it tries for 256.
4029 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4030 was an oversight, and furthermore, ever since the addition of extra
4031 controls (e.g. 4.43/32), the checks on when to allow different forms of
4032 "control" were broken. There should now be diagnostics for all cases when a
4033 control that does not make sense is encountered.
4035 9. Added the /retain_sender option to "control=submission".
4037 10. $recipients is now available in the predata ACL (oversight).
4039 11. Tidy the search cache before the fork to do a delivery from a message
4040 received from the command line. Otherwise the child will trigger a lookup
4041 failure and thereby defer the delivery if it tries to use (for example) a
4042 cached ldap connection that the parent has called unbind on.
4044 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4045 of $address_data from the recipient verification was clobbered by the
4046 sender verification.
4048 13. The value of address_data from a sender verification is now available in
4049 $sender_address_data in subsequent conditions in the ACL statement.
4051 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
4053 15. Added a new option "connect=<time>" to callout options, to set a different
4056 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4057 was its contents. (It was OK if the option was not defined at all.)
4059 17. A "Completed" log line is now written for messages that are removed from
4060 the spool by the -Mrm option.
4062 18. New variables $sender_verify_failure and $recipient_verify_failure contain
4063 information about exactly what failed.
4065 19. Added -dd to debug only the daemon process.
4067 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
4068 handles timeouts, both on the server side and network timeouts. Renamed the
4069 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
4071 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
4072 It is now set to "smtps".
4074 22. $host_address is now set to the target address during the checking of
4075 ignore_target_hosts.
4077 23. When checking ignore_target_hosts for an ipliteral router, no host name was
4078 being passed; this would have caused $sender_host_name to have been used if
4079 matching the list had actually called for a host name (not very likely,
4080 since this list is usually IP addresses). A host name is now passed as
4083 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4084 code that specifies a non-restarting handler (typically sigaction() in
4085 modern systems) in an attempt to fix a rare and obscure crash bug.
4087 25. Narrowed the window for a race in the daemon that could cause it to ignore
4088 SIGCHLD signals. This is not a major problem, because they are used only to
4089 wake it up if nothing else does.
4091 26. A malformed maildirsize file could cause Exim to calculate negative values
4092 for the mailbox size or file count. Odd effects could occur as a result.
4093 The maildirsize information is now recalculated if the size or filecount
4096 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4097 support for a long time. Removed HAVE_SYS_VFS_H.
4099 28. Installed the latest version of exipick from John Jetmore.
4101 29. In an address list, if the pattern was not a regular expression, an empty
4102 subject address (from a bounce message) matched only if the pattern was an
4103 empty string. Non-empty patterns were not even tested. This was the wrong
4104 because it is perfectly reasonable to use an empty address as part of a
4105 database query. An empty address is now tested by patterns that are
4106 lookups. However, all the other forms of pattern expect the subject to
4107 contain a local part and a domain, and therefore, for them, an empty
4108 address still always fails if the pattern is not itself empty.
4110 30. Exim went into a mad DNS loop when attempting to do a callout where the
4111 host was specified on an smtp transport, and looking it up yielded more
4112 than one IP address.
4114 31. Re-factored the code for checking spool and log partition space into a
4115 function that finds that data and another that does the check. The former
4116 is then used to implement four new variables: $spool_space, $log_space,
4117 $spool_inodes, and $log_inodes.
4119 32. The RFC2047 encoding function was originally intended for short strings
4120 such as real names; it was not keeping to the 75-character limit for
4121 encoded words that the RFC imposes. It now respects the limit, and
4122 generates multiple encoded words if necessary. To be on the safe side, I
4123 have increased the buffer size for the ${rfc2047: expansion operator from
4126 33. It is now permitted to omit both strings after an "if" condition; if the
4127 condition is true, the result is "true". As before, when the second string
4128 is omitted, a false condition yields an empty string. This makes it less
4129 cumbersome to write custom ACL and router conditions.
4131 34. Failure to deliver a bounce message always caused it to be frozen, even if
4132 there was an errors_to setting on the router. The errors_to setting is now
4135 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4136 canonical form (fully expanded) before being placed in
4137 $sender_host_address.
4139 36. The table in the code that translates DNS record types into text (T_A to
4140 "A" for instance) was missing entries for NS and CNAME. It is just possible
4141 that this could have caused confusion if both these types were looked up
4142 for the same domain, because the text type is used as part of Exim's
4143 per-process caching. But the chance of anyone hitting this buglet seems
4146 37. The dnsdb lookup has been extended in a number of ways.
4148 (1) There is a new type, "zns", which walks up the domain tree until it
4149 finds some nameserver records. It should be used with care.
4151 (2) There is a new type, "mxh", which is like "mx" except that it returns
4152 just the host names, not the priorities.
4154 (3) It is now possible to give a list of domains (or IP addresses) to be
4155 looked up. The behaviour when one of the lookups defers can be
4156 controlled by a keyword.
4158 (4) It is now possible to specify the separator character for use when
4159 multiple records are returned.
4161 38. The dnslists ACL condition has been extended: it is now possible to supply
4162 a list of IP addresses and/or domains to be looked up in a particular DNS
4165 39. Added log_selector=+queue_time_overall.
4167 40. When running the queue in the test harness, wait just a tad after forking a
4168 delivery process, to get repeatability of debugging output.
4170 41. Include certificate and key file names in error message when GnuTLS fails
4171 to set them up, because the GnuTLS error message doesn't include the name
4172 of the failing file when there is a problem reading it.
4174 42. Allow both -bf and -bF in the same test run.
4176 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
4178 44. The "Exiscan patch" is now merged into the mainline Exim source.
4180 45. Sometimes the final signoff response after QUIT could fail to get
4181 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4182 < 0 before doing a fflush(). This bug looks as though it goes back to the
4183 introduction of TLS in release 3.20, but "sometimes" must have been rare
4184 because the tests only now provoked it.
4186 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
4187 (this can affect the format of dates).
4189 47. exim_tidydb, when checking for the continued existence of a message for
4190 which it has found a message-specific retry record, was not finding
4191 messages that were in split spool directories. Consequently, it was
4192 deleting retry records that should have stayed in existence.
4194 48. Steve fixed some bugs in eximstats.
4196 49. The SPA authentication driver was not abandoning authentication and moving
4197 on to the next authenticator when an expansion was forced to fail,
4198 contradicting the general specification for all authenticators. Instead it
4199 was generating a temporary error. It now behaves as specified.
4201 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
4202 (the order specifies the preference for clients). The order is now AES256,
4203 AES128, 3DES, ARCFOUR128.
4205 51. Small patch to Sieve code - explicitly set From: when generating an
4208 52. Exim crashed if a remote delivery caused a very long error message to be
4209 recorded - for instance if somebody sent an entire SpamAssassin report back
4210 as a large number of 550 error lines. This bug was coincidentally fixed by
4211 increasing the size of one of Exim's internal buffers (big_buffer) that
4212 happened as part of the Exiscan merge. However, to be on the safe side, I
4213 have made the code more robust (and fixed the comments that describe what
4216 53. Now that there can be additional text after "Completed" in log lines (if
4217 the queue_time_overall log selector is set), a one-byte patch to exigrep
4218 was needed to allow it to recognize "Completed" as not the last thing in
4221 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
4222 patch that reportedly fixes this has been added. I am not expert enough to
4223 create a test for it. This is what the patch creator wrote:
4225 "I found a little strange behaviour of ldap code when working with
4226 Windows 2003 AD Domain, where users was placed in more than one
4227 Organization Units. When I tried to give exim partial DN, the exit code
4228 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
4229 But simultaneously result of request was absolutely normal ldap result,
4230 so I produce this patch..."
4232 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
4233 REFERENCE, so I have modified the code to exclude the patch when that macro
4236 55. Some experimental protocols are using DNS PTR records for new purposes. The
4237 keys for these records are domain names, not reversed IP addresses. The
4238 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4239 leaves it alone. Component reversal etc. now happens only for IP addresses.
4242 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
4244 57. Double the size of the debug message buffer (to 2048) so that more of very
4245 long debug lines gets shown.
4247 58. The exicyclog utility now does better if the number of log files to keep
4248 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
4250 59. Two changes related to the smtp_active_hostname option:
4252 (1) $smtp_active_hostname is now available as a variable.
4253 (2) The default for smtp_banner uses $smtp_active_hostname instead
4254 of $primary_hostname.
4256 60. The host_aton() function is supposed to be passed a string that is known
4257 to be a valid IP address. However, in the case of IPv6 addresses, it was
4258 not checking this. This is a hostage to fortune. Exim now panics and dies
4259 if the condition is not met. A case was found where this could be provoked
4260 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4261 components; fortuitously, this particular loophole had already been fixed
4262 by change 4.50/55 above.
4264 If there are any other similar loopholes, the new check in host_aton()
4265 itself should stop them being exploited. The report I received stated that
4266 data on the command line could provoke the exploit when Exim was running as
4267 exim, but did not say which command line option was involved. All I could
4268 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4269 running as the user.
4272 61. There was a buffer overflow vulnerability in the SPA authentication code
4273 (which came originally from the Samba project). I have added a test to the
4274 spa_base64_to_bits() function which I hope fixes it.
4277 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
4278 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
4280 63. The daemon start-up calls getloadavg() while still root for those OS that
4281 need the first call to be done as root, but it missed one case: when
4282 deliver_queue_load_max is set with deliver_drop_privilege. This is
4283 necessary for the benefit of the queue runner, because there is no re-exec
4284 when deliver_drop_privilege is set.
4286 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
4287 This has been fixed.
4289 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4290 was in use, was not putting the data itself into the right store pool;
4291 consequently, it could be overwritten for a subsequent message in the same
4292 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4295 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
4297 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
4298 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
4299 the function were treating the return as a boolean value, which happened to
4300 work because 0=false and not-0=true, but is not correct code.
4302 68. The host_aton() function was not handling scoped IPv6 addresses (those
4303 with, for example, "%eth0" on the end) correctly.
4305 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
4306 negated items (that is, ~something) in unsigned ints. Some compilers
4307 apparently mutter when there is no cast.
4309 70. If an address verification called from an ACL failed, and did not produce a
4310 user-specific message (i.e. there was only a "system" message), nothing was
4311 put in $acl_verify_message. In this situation, it now puts the system
4314 71. Change 4.23/11 added synchronization checking at the start of an SMTP
4315 session; change 4.31/43 added the unwanted input to the log line - except
4316 that it did not do this in the start of session case. It now does.
4318 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
4319 This is wrong and can cause the other end to generate a synchronization
4320 error if it is another Exim or anything else that does the synchronization
4321 check. A QUIT command is no longer sent after a timeout.
4323 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
4324 during host lookups.
4326 74. The defer_ok option of callout verification was not working if it was used
4327 when verifying addresses in header lines, that is, for this case:
4329 verify = header_sender/callout=defer_ok
4331 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
4332 those file descriptors could be used for SMTP connections. If anything
4333 wrote to stderr (the example that came up was "warn" in embedded Perl), it
4334 could be sent to the SMTP client, causing chaos. The daemon now opens
4335 stdin, stdout, and stderr to /dev/null when it puts itself into the
4338 76. Arrange for output from Perl's "warn" command to be written to Exim's main
4339 log by default. The user can override this with suitable Perl magic.
4341 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
4342 the log message when discard triggers, was not working for the DATA ACL or
4343 for the non-SMTP ACL.
4345 78. Error message wording change in sieve.c.
4347 79. If smtp_accept_max_per_host was set, the number of connections could be
4348 restricted to fewer than expected, because the daemon was trying to set up
4349 a new connection before checking whether the processes handling previous
4350 connections had finished. The check for completed processes is now done
4351 earlier. On busy systems, this bug wouldn't be noticed because something
4352 else would have woken the daemon, and it would have reaped the completed
4355 80. If a message was submitted locally by a user whose login name contained one
4356 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
4357 It caused a spool format error. I have fixed the spool reading code. A
4358 related problem was that the "from" clause in the Received: line became
4359 illegal because of the space(s). It is now covered by ${quote_local_part.
4361 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
4364 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
4366 83. Patch from Sieve maintainer to fix unterminated string problem in
4367 "vacation" handling.
4369 84. Some minor changes to the Linux configuration files to help with other
4370 OS variants using glibc.
4372 85. One more patch for Sieve to update vacation handling to latest spec.
4375 ----------------------------------------------------
4376 See the note above about the 4.44 and 4.50 releases.
4377 ----------------------------------------------------
4383 1. Change 4.43/35 introduced a bug that caused file counts to be
4384 incorrectly computed when quota_filecount was set in an appendfile
4387 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4388 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4391 3. Give more explanation in the error message when the command for a transport
4392 filter fails to execute.
4394 4. There are several places where Exim runs a non-Exim command in a
4395 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4396 was being done only for the command run by the queryprogram router. It is
4397 now done for all such subprocesses. The other cases are: ${run, transport
4398 filters, and the commands run by the lmtp and pipe transports.
4400 5. Some older OS have a limit of 256 on the maximum number of file
4401 descriptors. Exim was using setrlimit() to set 1000 as a large value
4402 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4403 systems. I've change it so that if it can't get 1000, it tries for 256.
4405 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4406 was an oversight, and furthermore, ever since the addition of extra
4407 controls (e.g. 4.43/32), the checks on when to allow different forms of
4408 "control" were broken. There should now be diagnostics for all cases when a
4409 control that does not make sense is encountered.
4411 7. $recipients is now available in the predata ACL (oversight).
4413 8. Tidy the search cache before the fork to do a delivery from a message
4414 received from the command line. Otherwise the child will trigger a lookup
4415 failure and thereby defer the delivery if it tries to use (for example) a
4416 cached ldap connection that the parent has called unbind on.
4418 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4419 of $address_data from the recipient verification was clobbered by the
4420 sender verification.
4422 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4423 was its contents. (It was OK if the option was not defined at all.)
4425 11. A "Completed" log line is now written for messages that are removed from
4426 the spool by the -Mrm option.
4428 12. $host_address is now set to the target address during the checking of
4429 ignore_target_hosts.
4431 13. When checking ignore_target_hosts for an ipliteral router, no host name was
4432 being passed; this would have caused $sender_host_name to have been used if
4433 matching the list had actually called for a host name (not very likely,
4434 since this list is usually IP addresses). A host name is now passed as
4437 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4438 code that specifies a non-restarting handler (typically sigaction() in
4439 modern systems) in an attempt to fix a rare and obscure crash bug.
4441 15. Narrowed the window for a race in the daemon that could cause it to ignore
4442 SIGCHLD signals. This is not a major problem, because they are used only to
4443 wake it up if nothing else does.
4445 16. A malformed maildirsize file could cause Exim to calculate negative values
4446 for the mailbox size or file count. Odd effects could occur as a result.
4447 The maildirsize information is now recalculated if the size or filecount
4450 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4451 support for a long time. Removed HAVE_SYS_VFS_H.
4453 18. Updated exipick to current release from John Jetmore.
4455 19. Allow an empty sender to be matched against a lookup in an address list.
4456 Previously the only cases considered were a regular expression, or an
4459 20. Exim went into a mad DNS lookup loop when doing a callout where the
4460 host was specified on the transport, if the DNS lookup yielded more than
4463 21. The RFC2047 encoding function was originally intended for short strings
4464 such as real names; it was not keeping to the 75-character limit for
4465 encoded words that the RFC imposes. It now respects the limit, and
4466 generates multiple encoded words if necessary. To be on the safe side, I
4467 have increased the buffer size for the ${rfc2047: expansion operator from
4470 22. Failure to deliver a bounce message always caused it to be frozen, even if
4471 there was an errors_to setting on the router. The errors_to setting is now
4474 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4475 canonical form (fully expanded) before being placed in
4476 $sender_host_address.
4478 24. Updated eximstats to version 1.33
4480 25. Include certificate and key file names in error message when GnuTLS fails
4481 to set them up, because the GnuTLS error message doesn't include the name
4482 of the failing file when there is a problem reading it.
4484 26. Expand error message when OpenSSL has problems setting up cert/key files.
4487 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
4488 (this can affect the format of dates).
4490 28. exim_tidydb, when checking for the continued existence of a message for
4491 which it has found a message-specific retry record, was not finding
4492 messages that were in split spool directories. Consequently, it was
4493 deleting retry records that should have stayed in existence.
4495 29. eximstats updated to version 1.35
4496 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
4497 1.35 - bugfix such that pie charts by volume are generated correctly
4499 30. The SPA authentication driver was not abandoning authentication and moving
4500 on to the next authenticator when an expansion was forced to fail,
4501 contradicting the general specification for all authenticators. Instead it
4502 was generating a temporary error. It now behaves as specified.
4504 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
4505 (the order specifies the preference for clients). The order is now AES256,
4506 AES128, 3DES, ARCFOUR128.
4508 31. Small patch to Sieve code - explicitly set From: when generating an
4511 32. Exim crashed if a remote delivery caused a very long error message to be
4512 recorded - for instance if somebody sent an entire SpamAssassin report back
4513 as a large number of 550 error lines. This bug was coincidentally fixed by
4514 increasing the size of one of Exim's internal buffers (big_buffer) that
4515 happened as part of the Exiscan merge. However, to be on the safe side, I
4516 have made the code more robust (and fixed the comments that describe what
4519 33. Some experimental protocols are using DNS PTR records for new purposes. The
4520 keys for these records are domain names, not reversed IP addresses. The
4521 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4522 leaves it alone. Component reversal etc. now happens only for IP addresses.
4525 34. The host_aton() function is supposed to be passed a string that is known
4526 to be a valid IP address. However, in the case of IPv6 addresses, it was
4527 not checking this. This is a hostage to fortune. Exim now panics and dies
4528 if the condition is not met. A case was found where this could be provoked
4529 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4530 components; fortuitously, this particular loophole had already been fixed
4531 by change 4.50/55 or 4.44/33 above.
4533 If there are any other similar loopholes, the new check in host_aton()
4534 itself should stop them being exploited. The report I received stated that
4535 data on the command line could provoke the exploit when Exim was running as
4536 exim, but did not say which command line option was involved. All I could
4537 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4538 running as the user.
4541 35. There was a buffer overflow vulnerability in the SPA authentication code
4542 (which came originally from the Samba project). I have added a test to the
4543 spa_base64_to_bits() function which I hope fixes it.
4546 36. The daemon start-up calls getloadavg() while still root for those OS that
4547 need the first call to be done as root, but it missed one case: when
4548 deliver_queue_load_max is set with deliver_drop_privilege. This is
4549 necessary for the benefit of the queue runner, because there is no re-exec
4550 when deliver_drop_privilege is set.
4552 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4553 was in use, was not putting the data itself into the right store pool;
4554 consequently, it could be overwritten for a subsequent message in the same
4555 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4558 38. Sometimes the final signoff response after QUIT could fail to get
4559 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4560 < 0 before doing a fflush(). This bug looks as though it goes back to the
4561 introduction of TLS in release 3.20, but "sometimes" must have been rare
4562 because the tests only now provoked it.
4568 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
4569 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
4570 Now it returns an integer. A number of calls were still expecting a T/F
4571 return. Fortuitously, in all cases, the tests worked in OK situations,
4572 which is the norm. However, things would have gone wrong on any write
4573 failures on the smtp file descriptor. This function is used when sending
4574 messages over SMTP and also when doing verify callouts.
4576 2. When Exim is called to do synchronous delivery of a locally submitted
4577 message (the -odf or -odi options), it no longer closes stderr before doing
4580 3. Implemented the mua_wrapper option.
4582 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
4584 5. Implemented the functions header_remove(), header_testname(),
4585 header_add_at_position(), and receive_remove_recipient(), and exported them
4588 6. If an ACL "warn" statement specified the addition of headers, Exim already
4589 inserted X-ACL-Warn: at the start if there was no header name. However, it
4590 was not making this test for the second and subsequent header lines if
4591 there were newlines in the string. This meant that an invalid header could
4592 be inserted if Exim was badly configured.
4594 7. Allow an ACL "warn" statement to add header lines at the start or after all
4595 the Received: headers, as well as at the end.
4597 8. Added the rcpt_4xx retry error code.
4599 9. Added postmaster_mailfrom=xxx to callout verification option.
4601 10. Added mailfrom=xxxx to the callout verification option, for verify=
4604 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
4605 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
4607 12. Inserted some casts to stop certain compilers warning when using pointer
4608 differences as field lengths or precisions in printf-type calls (mostly
4609 affecting debugging statements).
4611 13. Added optional readline() support for -be (dynamically loaded).
4613 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
4614 same clock tick as a message's arrival, so that its received time was the
4615 same as the "first fail" time on the retry record, and that message
4616 remained on the queue past the ultimate address timeout, every queue runner
4617 would try a delivery (because it was past the ultimate address timeout) but
4618 after another failure, the ultimate address timeout, which should have then
4619 bounced the address, did not kick in. This was a "< instead of <=" error;
4620 in most cases the first failure would have been in the next clock tick
4621 after the received time, and all would be well.
4623 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
4624 being recognized when the domain list was tested by the match_domain
4625 condition in an expansion string.
4627 16. Added the ${str2b64: operator.
4629 17. Exim was always calling setrlimit() to set a large limit for the number of
4630 processes, without checking whether the existing limit was already
4631 adequate. (It did check for the limit on file descriptors.) Furthermore,
4632 errors from getrlimit() and setrlimit() were being ignored. Now they are
4633 logged to the main and panic logs, but Exim does carry on, to try to do its
4634 job under whatever limits there are.
4636 18. Imported PCRE 5.0.
4638 19. Trivial typo in log message " temporarily refused connection" (the leading
4641 20. If the log selector return_path_on_delivery was set and an address was
4642 redirected to /dev/null, the delivery process crashed because it assumed
4643 that a return path would always be set for a "successful" delivery. In this
4644 case, the whole delivery is bypassed as an optimization, and therefore no
4647 21. Internal re-arrangement: the function for sending a challenge and reading
4648 a response while authentication was assuming a zero-terminated challenge
4649 string. It's now changed to take a pointer and a length, to allow for
4650 binary data in such strings.
4652 22. Added the cyrus_sasl authenticator (code supplied by MBM).
4654 23. Exim was not respecting finduser_retries when seeking the login of the
4655 uid under which it was called; it was always trying 10 times. (The default
4656 setting of finduser_retries is zero.) Also, it was sleeping after the final
4657 failure, which is pointless.
4659 24. Implemented tls_on_connect_ports.
4661 25. Implemented acl_smtp_predata.
4663 26. If the domain in control=submission is set empty, Exim assumes that the
4664 authenticated id is a complete email address when it generates From: or
4665 Sender: header lines.
4667 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
4668 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
4669 chown and chgrp in /bin and hostname in /usr/bin.
4671 28. Exim was keeping the "process log" file open after each use, just as it
4672 does for the main log. This opens the possibility of it remaining open for
4673 long periods when the USR1 signal hits a daemon. Occasional processlog
4674 errors were reported, that could have been caused by this. Anyway, it seems
4675 much more sensible not to leave this file open at all, so that is what now
4678 29. The long-running daemon process does not normally write to the log once it
4679 has entered its main loop, and it closes the log before doing so. This is
4680 so that log files can straightforwardly be renamed and moved. However,
4681 there are a couple of unusual error situations where the daemon does write
4682 log entries, and I had neglected to close the log afterwards.
4684 30. The text of an SMTP error response that was received during a remote
4685 delivery was being truncated at 512 bytes. This is too short for some of
4686 the long messages that one sometimes sees. I've increased the limit to
4689 31. It is now possible to make retry rules that apply only when a message has a
4690 specific sender, in particular, an empty sender.
4692 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
4693 it possible to be selective about when SMTP synchronization is enforced.
4695 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
4697 32. Implemented hosts_connection_nolog.
4699 33. Added an ACL for QUIT.
4701 34. Setting "delay_warning=" to disable warnings was not working; it gave a
4704 35. Added mailbox_size and mailbox_filecount to appendfile.
4706 36. Added control = no_multiline_responses to ACLs.
4708 37. There was a bug in the logic of the code that waits for the clock to tick
4709 in the case where the clock went backwards by a substantial amount such
4710 that the microsecond fraction of "now" was more than the microsecond
4711 fraction of "then" (but the whole seconds number was less).
4713 38. Added support for the libradius Radius client library this is found on
4714 FreeBSD (previously only the radiusclient library was supported).
4720 1. When certain lookups returned multiple values in the form name=value, the
4721 quoting of the values was not always being done properly. Specifically:
4722 (a) If the value started with a double quote, but contained no whitespace,
4724 (b) If the value contained whitespace other than a space character (i.e.
4725 tabs or newlines or carriage returns) it was not quoted.
4726 This fix has been applied to the mysql and pgsql lookups by writing a
4727 separate quoting function and calling it from the lookup code. The fix
4728 should probably also be applied to nisplus, ibase and oracle lookups, but
4729 since I cannot test any of those, I have not disturbed their existing code.
4731 2. A hit in the callout cache for a specific address caused a log line with no
4732 reason for rejecting RCPT. Now it says "Previous (cached) callout
4733 verification failure".
4735 3. There was an off-by-one bug in the queryprogram router. An over-long
4736 return line was truncated at 256 instead of 255 characters, thereby
4737 overflowing its buffer with the terminating zero. As well as fixing this, I
4738 have increased the buffer size to 1024 (and made a note to document this).
4740 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
4741 when Exim is waiting for an SMTP response from a remote server, Exim
4742 restarts its select() call on the socket, thereby resetting its timeout.
4743 This is not a problem when such interrupts are rare. Somebody set up a cron
4744 job to run exiwhat every 2 minutes, which is less than the normal select()
4745 timeout (5 or 10 minutes). This meant that the select() timeout never
4746 kicked in because it was always reset. I have fixed this by comparing the
4747 time when an interrupt arrives with the time at the start of the first call
4748 to select(). If more time than the timeout has elapsed, the interrupt is
4749 treated as a timeout.
4751 5. Some internal re-factoring in preparation for the addition of Sieve
4752 extensions (by MH). In particular, the "personal" test is moved to a
4753 separate function, and given an option for scanning Cc: and Bcc: (which is
4754 not set for Exim filters).
4756 6. When Exim created an email address using the login of the caller as the
4757 local part (e.g. when creating a From: or Sender: header line), it was not
4758 quoting the local part when it contained special characters such as @.
4760 7. Installed new OpenBSD configuration files.
4762 8. Reworded some messages for syntax errors in "and" and "or" conditions to
4763 try to make them clearer.
4765 9. Callout options, other than the timeout value, were being ignored when
4766 verifying sender addresses in header lines. For example, when using
4768 verify = header_sender/callout=no_cache
4770 the cache was (incorrectly) being used.
4772 10. Added a missing instance of ${EXE} to the exim_install script; this affects
4773 only the Cygwin environment.
4775 11. When return_path_on_delivery was set as a log selector, if different remote
4776 addresses in the same message used different return paths and parallel
4777 remote delivery occurred, the wrong values would sometimes be logged.
4778 (Whenever a remote delivery process finished, the return path value from
4779 the most recently started remote delivery process was logged.)
4781 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
4782 lines when AUTH and/or TLS are in use. This is the "received protocol"
4783 field. Exim used to use "asmtp" for authenticated SMTP, without any
4784 indication (in the protocol name) for TLS use. Now it follows the RFC and
4785 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
4786 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
4787 names appear in log lines as well as in Received: header lines.
4789 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
4790 extensions, and comparison tests, and to fix some bugs.
4792 14. Changes to the "personal" filter test:
4794 (1) The test was buggy in that it was just doing the equivalent of
4795 "contains" tests on header lines. For example, if a user's address was
4796 anne@some.where, the "personal" test would incorrectly be true for
4798 To: susanne@some.where
4800 This test is now done by extracting each address from the header in turn,
4801 and checking the entire address. Other tests that are part of "personal"
4802 are now done using regular expressions (for example, to check local parts
4803 of addresses in From: header lines).
4805 (2) The list of non-personal local parts in From: addresses has been
4806 extended to include "listserv", "majordomo", "*-request", and "owner-*",
4807 taken from the Sieve specification recommendations.
4809 (3) If the message contains any header line starting with "List-" it is
4810 treated as non-personal.
4812 (4) The test for "circular" in the Subject: header line has been removed
4813 because it now seems ill-conceived.
4815 15. Minor typos in src/EDITME comments corrected.
4817 16. Installed latest exipick from John Jetmore.
4819 17. If headers_add on a router specified a text string that was too long for
4820 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
4821 of string_sprintf() is now avoided.
4823 18. $message_body_size was not set (it was always zero) when running the DATA
4824 ACL and the local_scan() function.
4826 19. For the "mail" command in an Exim filter, no default was being set for
4827 the once_repeat time, causing a random time value to be used if "once" was
4828 specified. (If the value happened to be <= 0, no repeat happened.) The
4829 default is now 0s, meaning "never repeat". The "vacation" command was OK
4830 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
4831 (I found it when inspecting the code).
4833 20. There is now an overall timeout for performing a callout verification. It
4834 defaults to 4 times the callout timeout, which applies to individual SMTP
4835 commands during the callout. The overall timeout applies when there is more
4836 than one host that can be tried. The timeout is checked before trying the
4837 next host. This prevents very long delays if there are a large number of
4838 hosts and all are timing out (e.g. when the network connections are timing
4839 out). The value of the overall timeout can be changed by specifying an
4840 additional sub-option for "callout", called "maxwait". For example:
4842 verify = sender/callout=5s,maxwait=20s
4844 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
4845 to the end before writing, but this should make it even safer).
4847 22. Exim was forgetting that it had advertised PIPELINING for the second and
4848 subsequent messages on an SMTP connection. It was also not resetting its
4849 memory on STARTTLS and an internal HELO.
4851 23. When Exim logs an SMTP synchronization error within a session, it now
4852 records whether PIPELINING has been advertised or not.
4854 24. Added 3 instances of "(long int)" casts to time_t variables that were being
4855 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
4856 rather than long int.
4858 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
4860 26. Added the never_mail option to autoreply.
4866 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
4867 crash if the getsockname() call failed; this can happen if a connection is
4868 closed very soon after it is established. The problem was simply in the
4869 order in which certain operations were done, causing Exim to try to write
4870 to the SMTP stream before it had set up the file descriptor. The bug has
4871 been fixed by making things happen in the correct order.
4877 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
4878 before the connection was closed, thus losing the rejection response.
4880 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
4881 some early Solaris releases, but causes trouble in current releases where
4882 socklen_t is defined.
4884 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
4887 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
4890 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
4891 that releasing the top of it at the end releases what was used for sub-
4892 expansions (unless the block got too big). However, discard this block if
4893 the first thing is a variable or header, so that we can use its block when
4894 it is dynamic (useful for very large $message_headers, for example).
4896 6. Lookups now cache *every* query, not just the most recent. A new, separate
4897 store pool is used for this. It can be recovered when all lookup caches are
4898 flushed. Lookups now release memory at the end of their result strings.
4899 This has involved some general refactoring of the lookup sources.
4901 7. Some code has been added to the store_xxx() functions to reduce the amount
4902 of flapping under certain conditions.
4904 8. log_incoming_interface used to affect only the <= reception log lines. Now
4905 it causes the local interface and port to be added to several more SMTP log
4906 lines, for example "SMTP connection from", and rejection lines.
4908 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
4910 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
4912 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
4913 could be overwritten at the end of the current message (or the start of a
4914 new message if it was set in a HELO ACL). The value is now preserved for
4915 the duration of the SMTP connection.
4917 12. If a transport had a headers_rewrite setting, and a matching header line
4918 contained an unqualified address, that address was qualified, even if it
4919 did not match any rewriting rules. The underlying bug was that the values
4920 of the flags that permit the existence of unqualified sender and recipient
4921 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
4922 non-local messages, and by -bnq for local messages) were not being
4923 preserved with the message after it was received.
4925 13. When Exim was logging an SMTP synchronization error, it could sometimes log
4926 "next input=" as part of the text comprising the host identity instead of
4927 the correct text. The code was using the same buffer for two different
4928 strings. However, depending on which order the printing function evaluated
4929 its arguments, the bug did not always show up. Under Linux, for example, my
4930 test suite worked just fine.
4932 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
4933 doesn't work with some older versions of Perl. It has been changed to "my",
4934 which in any case is probably the better facility to use.
4936 15. A really picky compiler found some instances of statements for creating
4937 error messages that either had too many or two few arguments for the format
4940 16. The size of the buffer for calls to the DNS resolver has been increased
4941 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
4942 for addresses that have a lot of PTR records. This alleviates a problem; it
4943 does not fully solve it.
4945 17. A dnsdb lookup for PTR records that receives more data than will fit in the
4946 buffer now truncates the list and logs the incident, which is the same
4947 action as happens when Exim is looking up a host name and its aliases.
4948 Previously in this situation something unpredictable would happen;
4949 sometimes it was "internal error: store_reset failed".
4951 18. If a server dropped the connection unexpectedly when an Exim client was
4952 using GnuTLS and trying to read a response, the client delivery process
4953 crashed while trying to generate an error log message.
4955 19. If a "warn" verb in an ACL added multiple headers to a message in a single
4956 string, for example:
4958 warn message = H1: something\nH2: something
4960 the text was added as a single header line from Exim's point of view
4961 though it ended up OK in the delivered message. However, searching for the
4962 second and subsequent header lines using $h_h2: did not work. This has been
4963 fixed. Similarly, if a system filter added multiple headers in this way,
4964 the routers could not see them.
4966 20. Expanded the error message when iplsearch is called with an invalid key to
4967 suggest using net-iplsearch in a host list.
4969 21. When running tests using -bh, any delays imposed by "delay" modifiers in
4970 ACLs are no longer actually imposed (and a message to that effect is
4973 22. If a "gecos" field in a passwd entry contained escaped characters, in
4974 particular, if it contained a \" sequence, Exim got it wrong when building
4975 a From: or a Sender: header from that name. A second bug also caused
4976 incorrect handling when an unquoted " was present following a character
4977 that needed quoting.
4979 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
4980 was not being matched caselessly.
4982 24. Arranged for all hyphens in the exim.8 source to be escaped with
4985 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
4986 properly. Recipient callout cache records were still being keyed to include
4987 the sender, even when use_sender was set false. This led to far more
4988 callouts that were necessary. The sender is no longer included in the key
4989 when use_sender is false.
4991 26. Added "control = submission" modifier to ACLs.
4993 27. Added the ${base62d: operator to decode base 62 numbers.
4995 28. dnsdb lookups can now access SRV records.
4997 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
4998 the configuration file.
5000 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
5001 (-v) mode. This makes the output for a verbose queue run more intelligible.
5003 31. Added a use_postmaster feature to recipient callouts.
5005 32. Added the $body_zerocount variable, containing the number of binary zero
5006 bytes in the message body.
5008 33. The time of last modification of the "new" subdirectory is now used as the
5009 "mailbox time last read" when there is a quota error for a maildir
5012 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
5014 35. Added +ignore_unknown as a special item in host lists.
5016 36. Code for decoding IPv6 addresses in host lists is now included, even if
5017 IPv6 support is not being compiled. This fixes a bug in which an IPv6
5018 address was recognized as an IP address, but was then not correctly decoded
5019 into binary, causing unexpected and incorrect effects when compared with
5026 1. Very minor rewording of debugging text in manualroute to say "list of
5027 hosts" instead of "hostlist".
5029 2. If verify=header_syntax was set, and a header line with an unqualified
5030 address (no domain) and a large number of spaces between the end of the
5031 name and the colon was received, the reception process suffered a buffer
5032 overflow, and (when I tested it) crashed. This was caused by some obsolete
5033 code that should have been removed. The fix is to remove it!
5035 3. When running in the test harness, delay a bit after writing a bounce
5036 message to get a bit more predictability in the log output.
5038 4. Added a call to search_tidyup() just before forking a reception process. In
5039 theory, someone could use a lookup in the expansion of smtp_accept_max_
5040 per_host which, without the tidyup, could leave open a database connection.
5042 5. Added the variables $recipient_data and $sender_data which get set from a
5043 lookup success in an ACL "recipients" or "senders" condition, or a router
5044 "senders" option, similar to $domain_data and $local_part_data.
5046 6. Moved the writing of debug_print from before to after the "senders" test
5049 7. Change 4.31/66 (moving the time when the Received: is generated) caused
5050 problems for message scanning, either using a data ACL, or using
5051 local_scan() because the Received: header was not generated till after they
5052 were called (in order to set the time as the time of reception completion).
5053 I have revised the way this works. The header is now generated after the
5054 body is received, but before the ACL or local_scan() are called. After they
5055 are run, the timestamp in the header is updated.
5061 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
5062 before starting a queue runner without re-exec. This happened only when
5063 deliver_drop_privilege was set or when the Exim user was set to root. The
5064 effect of the bug was that timeouts during subsequent deliveries caused
5065 crashes instead of being properly handled. The handler is now left at its
5066 default (and expected) setting.
5068 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
5069 message, again when deliver_drop_privilege is set or Exim is run as root.
5070 The bug described in (1) was not present in this case, but the tidying up
5071 of the other signals was missing. I have made the two cases consistent.
5073 3. The ignore_target_hosts setting on a manualroute router was being ignored
5074 for hosts that were looked up using the /MX notation.
5076 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
5079 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
5080 operated on the sender address. After changing the $sender_address to <>
5081 for the sender address verify, Exim was re-instated it as the original
5082 (before rewriting) address, but remembering that it had rewritten it, so it
5083 wasn't rewriting it again. This bug also had the effect of breaking the
5084 sender address verification caching when the sender address was rewritten.
5086 6. The ignore_target_hosts option was being ignored by the ipliteral router.
5087 This has been changed so that if the ip literal address matches
5088 ignore_target_hosts, the router declines.
5090 7. Added expansion conditions match_domain, match_address, and match_local_
5091 part (NOT match_host).
5093 8. The placeholder for the Received: header didn't have a length field set.
5095 9. Added code to Exim itself and to exim_lock to test for a specific race
5096 condition that could lead to file corruption when using MBX delivery. The
5097 issue is with the lockfile that is created in /tmp. If this file is removed
5098 after a process has opened it but before that process has acquired a lock,
5099 there is the potential for a second process to recreate the file and also
5100 acquire a lock. This could lead to two Exim processes writing to the file
5101 at the same time. The added code performs the same test as UW imapd; it
5102 checks after acquiring the lock that its file descriptor still refers to
5103 the same named file.
5105 10. The buffer for building added header lines was of fixed size, 8192 bytes.
5106 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
5109 11. Added the smtp_active_hostname option. If used, this will typically be made
5110 to depend on the incoming interface address. Because $interface_address is
5111 not set up until the daemon has forked a reception process, error responses
5112 that can happen earlier (such as "too many connections") no longer contain
5115 12. If an expansion in a condition on a "warn" statement fails because a lookup
5116 defers, the "warn" statement is abandoned, and the next ACL statement is
5117 processed. Previously this caused the whole ACL to be aborted.
5119 13. Added the iplsearch lookup type.
5121 14. Added ident_timeout as a log selector.
5123 15. Added tls_certificate_verified as a log selector.
5125 16. Added a global option tls_require_ciphers (compare the smtp transport
5126 option of the same name). This controls incoming TLS connections.
5128 17. I finally figured out how to make tls_require_ciphers do a similar thing
5129 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
5130 before starting the TLS session.
5132 18. Tabs are now shown as \t in -bP output.
5134 19. If the log selector return_path_on_delivery was set, Exim crashed when
5135 bouncing a message because it had too many Received: header lines.
5137 20. If two routers both had headers_remove settings, and the first one included
5138 a superfluous trailing colon, the final name in the first list and the
5139 first name in the second list were incorrectly joined into one item (with a
5140 colon in the middle).
5146 1. Added -C and -D options to the exinext utility, mainly to make it easier
5147 to include in the automated testing, but these could be helpful when
5148 multiple configurations are in use.
5150 2. The exinext utility was not formatting the output nicely when there was
5151 an alternate port involved in the retry record key, nor when there was a
5152 message id as well (for retries that were specific to a specific message
5153 and a specific host). It was also confused by IPv6 addresses, because of
5154 the additional colons they contain. I have fixed the IPv4 problem, and
5155 patched it up to do a reasonable job for IPv6.
5157 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
5158 delivery, the log line now contains "pipelined" if PIPELINING was used.
5160 4. An SMTP transport process used to panic and die if the bind() call to set
5161 an explicit outgoing interface failed. This has been changed; it is now
5162 treated in the same way as a connect() failure.
5164 5. A reference to $sender_host_name in the part of a conditional expansion
5165 that was being skipped was still causing a DNS lookup. This no longer
5168 6. The def: expansion condition was not recognizing references to header lines
5169 that used bh_ and bheader_.
5171 7. Added the _cache feature to named lists.
5173 8. The code for checking quota_filecount in the appendfile transport was
5174 allowing one more file than it should have been.
5176 9. For compatibility with Sendmail, the command line option
5184 and sets the incoming protocol and host name (for trusted callers). The
5185 host name and its colon can be omitted when only the protocol is to be set.
5186 Note the Exim already has two private options, -pd and -ps, that refer to
5187 embedded Perl. It is therefore impossible to set a protocol value of "d" or
5188 "s", but I don't think that's a major issue.
5190 10. A number of refactoring changes to the code, none of which should affect
5193 (a) The number of logging options was getting close to filling up the
5194 32-bit word that was used as a bit map. I have split them into two classes:
5195 those that are passed in the argument to log_write(), and those that are
5196 only ever tested independently outside of that function. These are now in
5197 separate 32-bit words, so there is plenty of room for expansion again.
5198 There is no change in the user interface or the logging behaviour.
5200 (b) When building, for example, log lines, the code previously used a
5201 macro that called string_cat() twice, in order to add two strings. This is
5202 not really sufficiently general. Furthermore, there was one instance where
5203 it was actually wrong because one of the argument was used twice, and in
5204 one call a function was used. (As it happened, calling the function twice
5205 did not affect the overall behaviour.) The macro has been replaced by a
5206 function that can join an arbitrary number of extra strings onto a growing
5209 (c) The code for expansion conditions now uses a table and a binary chop
5210 instead of a serial search (which was left over from when there were very
5211 few conditions). Also, it now recognizes conditions like "pam" even when
5212 the relevant support is not compiled in: a suitably worded error message is
5213 given if an attempt is made to use such a condition.
5215 11. Added ${time_interval:xxxxx}.
5217 12. A bug was causing one of the ddress fields not to be passed back correctly
5218 from remote delivery subprocesses. The field in question was not being
5219 subsequently used, so this caused to problems in practice.
5221 13. Added new log selectors queue_time and deliver_time.
5223 14. Might have fixed a bug in maildirsizefile handling that threw up
5224 "unexpected character" debug warnings, and recalculated the data
5225 unnecessarily. In any case, I expanded the warning message to give more
5228 15. Added the message "Restricted characters in address" to the statements in
5229 the default ACL that block characters like @ and % in local parts.
5231 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
5232 Three changes have been made:
5234 (a) There was a serious bug; a negative response to MAIL caused the whole
5235 recipient domain to be cached as invalid, thereby blocking all messages
5236 to all local parts at the same domain, from all senders. This bug has
5237 been fixed. The domain is no longer cached after a negative response to
5238 MAIL if the sender used is not empty.
5240 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
5243 (c) A new callout option, "use_sender" has been added for people who want
5244 the modified behaviour.
5250 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
5253 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
5254 indeed breaks things for older releases.
5256 3. Added additional logging to the case where there is a problem reading data
5257 from a filter that is running in a subprocess using a pipe, in order to
5258 try to track down a specific problem.
5260 4. Testing facility fudge: when running in the test harness and attempting
5261 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
5262 getting "No route to host". Convert this to a timeout.
5264 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
5267 6. Some OS don't have socklen_t but use size_t instead. This affects the
5268 fifth argument of getsockopt() amongst other things. This is now
5269 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
5270 can be set for individual OS. I have set it for SunOS5, OSF1, and
5271 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
5272 some earlier ones do not.
5274 7. Change 4.30/15 was not doing the test caselessly.
5276 8. The standard form for an IPv6 address literal was being rejected by address
5277 parsing in, for example, MAIL and RCPT commands. An example of this kind of
5278 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5279 this, as well as the form without the "IPv6" on the front (but only when
5280 address literals are enabled, of course).
5282 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
5284 10. Exim crashed if a message with an empty sender address specified by -f
5285 encountered a router with an errors_to setting. This could be provoked only
5286 by a command such as
5290 where an empty string was supplied; "<>" did not hit this bug.
5292 11. Installed PCRE release 4.5.
5294 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
5295 remained set. It is now erased.
5297 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
5298 times from message ids (which are base 36 rather than the normal 62).
5300 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
5301 were being counted as actual protocol errors, and logged if the log
5302 selector +smtp_protocol_error was set. One cannot be perfect in this test,
5303 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
5304 and DATA following a set of rejected RCPTs do not count as protocol errors.
5305 In other words, Exim assumes they were pipelined, though this may not
5306 actually be the case. Of course, in all cases the client gets an
5307 appropriate error code.
5309 15. If a lookup fails in an ACL condition, a message about the failure may
5310 be available; it is used if testing the ACL cannot continue, because most
5311 such messages specify what the cause of the deferral is. However, some
5312 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
5313 that caused an old message to be retained and used if a later statement
5314 caused a defer, replacing the real cause of the deferral.
5316 16. If an IP address had so many PTR records that the DNS lookup buffer
5317 was not large enough to hold them, Exim could crash while trying to process
5318 the truncated data. It now detects and logs this case.
5320 17. Further to 4.21/58, another change has been made: if (and only if) the
5321 first line of a message (the first header line) ends with CRLF, a bare LF
5322 in a subsequent header line has a space inserted after it, so as not to
5323 terminate the header.
5325 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
5326 unnecessarily, used atoi() instead of strtol(), and didn't check the
5327 termination when getting file sizes from file names by regex.
5329 19. Completely re-implemented the support for maildirsize files, in the light
5330 of a number of problems with the previous contributed implementation
5331 (4.30/29). In particular:
5333 . If the quota is zero, the maildirsize file is maintained, but no quota is
5336 . If the maildir directory does not exist, it is created before any attempt
5337 to write a maildirsize file.
5339 . The quota value in the file is just a cache; if the quota is changed in
5340 the transport, the new value overrides.
5342 . A regular expression is available for excluding directories from the
5345 20. The autoreply transport checks the characters in options that define the
5346 message's headers; it allows continued headers, but it was checking with
5347 isspace() after an embedded newline instead of explicitly looking for a
5350 21. If all the "regular" hosts to which an address was routed had passed their
5351 expiry times, and had not reached their retry times, the address was
5352 bounced, even if fallback hosts were defined. Now Exim should go on to try
5355 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
5356 equivalent code in the SMTP transport. Some hosts send humungous responses
5357 to HELO/EHLO, more than 1024 it seems.
5359 23. Refactoring: code in filter.c used (void *) for "any old type" but this
5360 gives compiler warnings in some environments. I've now done it "properly",
5363 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
5364 (because of problems with the built-in one) was declared to return uschar *
5365 instead of char *, causing compiler failure.
5367 25. Fixed a file descriptor leak when processing alias/forward files.
5369 26. Fixed a minor format string issue in dbfn.c.
5371 27. Typo in exim.c: ("dmbnz" for "dbmnz").
5373 28. If a filter file refered to $h_xxx or $message_headers, and the headers
5374 contained RFC 2047 "words", Exim's memory could, under certain conditions,
5377 29. When a sender address is verified, it is cached, to save repeating the test
5378 when there is more than one recipient in a message. However, when the
5379 verification involves a callout, it is possible for different callout
5380 options to be set for different recipients. It is too complicated to keep
5381 track of this in the cache, so now Exim always runs a verification when a
5382 callout is required, relying on the callout cache for the optimization.
5383 The overhead is duplication of the address routing, but this should not be
5386 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
5387 to be verified with callout=postmaster, and the main callout worked but the
5388 postmaster check failed, the verification correctly failed. However, if a
5389 subsequent RCPT command asked for sender verification *without* the
5390 postmaster check, incorrect caching caused this verification also to fail,
5393 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
5394 it was not caching the DNS options (qualify_single, search_parents) that
5395 were used when the lookup failed. A subsequent lookup with different
5396 options therefore always gave the same answer, though there were cases
5397 where it should not have. (Example: a "domains = !$mx_any" option on a
5398 dnslookup router: the "domains" option is always processed without any
5399 widening, but the router might have qualify_single set.) Now Exim uses the
5400 cached value only when the same options are set.
5402 32. Added John Jetmore's "exipick" utility to the distribution.
5404 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
5405 than a timeout (e.g. a certificate is required, and is not provided), an
5406 Exim server now closes the connection immediately. Previously it waited for
5407 the client to close - but if the client is SSL, it seems that they each
5408 wait for each other, leading to a delay before one of them times out.
5410 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
5411 maintained 0.8.x compatibility because I don't think many are using it, and
5412 it is clearly obsolete.
5414 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
5417 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
5418 client certificate was expired. A simple patch fixes this, though I don't
5419 understand the full logic of why the verify callback is called multiple
5422 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
5423 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
5424 which causes problems with some clients (such as the Certicom SSL Plus
5425 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
5426 disables the coutermeasure allowing Eudora to connect."
5428 38. Exim was not checking that a write() to a log file succeeded. This could
5429 lead to Bad Things if a log got too big, in particular if it hit a file
5430 size limit. Exim now panics and dies if it cannot write to a log file, just
5431 as it does if it cannot open a log file.
5433 39. Modified OS/Makefile-Linux so that it now contains
5435 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
5437 The two -D definitions ensure that Exim is compiled with large file
5438 support, which makes it possible to handle log files that are bigger than
5441 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
5442 instance) a domain was checked against a named list that involved a lookup,
5443 causing $domain_data to be set, then another domain was checked against the
5444 same list, then the first domain was re-checked, the value of $domain_data
5445 after the final check could be wrong. In particular, if the second check
5446 failed, it could be set empty. This bug probably also applied to
5449 41. The strip_trailing_dot option was not being applied to the address given
5450 with the -f command-line option.
5452 42. The code for reading a message's header from the spool was incrementing
5453 $received_count, but never initializing it. This meant that the value was
5454 incorrect (doubled) while delivering a message in the same process in which
5455 it was received. In the most common configuration of Exim, this never
5456 happens - a fresh exec is done - but it can happen when
5457 deliver_drop_privilege is set.
5459 43. When Exim logs an SMTP synchronization error - client data sent too soon -
5460 it now includes up to 150 characters of the unexpected data in the log
5463 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
5464 and building data strings. The size of both of these buffers was 10 000
5465 bytes - far larger than anybody would *ever* want, thought I. Needless to
5466 say, somebody hit the limit. I have increased the maximum line length to
5467 20 000 and the maximum data length of concatenated lines to 100 000. I have
5468 also fixed two bugs, because there was no checking on these buffers. Tsk,
5469 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
5470 buffer is too small.
5472 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
5473 lsearch lookups. Now it does.
5475 46. When parsing a route_list item in a manualroute router, a fixed-length
5476 buffer was used for the list of hosts. I made this 1024 bytes long,
5477 thinking that nobody would ever have a list of hosts that long. Wrong.
5478 Somebody had a whole pile of complicated expansion conditions, and the
5479 string was silently truncated, leading to an expansion error. It turns out
5480 that it is easier to change to an unlimited length (owing to other changes
5481 that have happened since this code was originally written) than to build
5482 structure for giving a limitation error. The length of the item that
5483 expands into the list of hosts is now unlimited.
5485 47. The lsearch lookup could not handle data where the length of text line was
5486 more than 4095 characters. Such lines were truncated, leading to shortened
5487 data being returned. It should now handle lines of any length.
5489 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
5490 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
5493 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
5494 replacing strings such as BIN_DIRECTORY in a source file: the replacement
5495 no longer happens in comment lines. A list of replacements is now placed
5496 at the head of all of the source files, except those whose only change is
5497 to replace PERL_COMMAND in the very first #! line.
5499 50. Replaced the slow insertion sort in queue.c, for sorting the list of
5500 messages on the queue, with a bottom-up merge sort, using code contributed
5501 by Michael Haardt. This should make operations like -bp somewhat faster on
5502 large queues. It won't affect queue runners, except when queue_run_in_order
5505 51. Installed eximstats 1.31 in the distribution.
5507 52. Added support for SRV lookups to the dnslookup router.
5509 53. If an ACL referred to $message_body or $message_body_end, the value was not
5510 reset for any messages that followed in the same SMTP session.
5512 54. The store-handling optimization for building very long strings was not
5513 differentiating between the different store pools. I don't think this
5514 actually made any difference in practice, but I've tidied it.
5516 55. While running the routers to verify a sender address, $sender_address
5517 was still set to the sender address. This is wrong, because when routing to
5518 send a bounce to the sender, it would be empty. Therefore, I have changed
5519 it so that, while verifying a sender address, $sender_address is set to <>.
5520 (There is no change to what happens when verifying a recipient address.)
5522 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
5523 target A or AAAA records (if not already returned) without resetting the
5524 qualify_single or search_parents options of the DNS resolver. These are
5525 inappropriate in this case because the targets of MX and SRV records must
5526 be FQDNs. A broken DNS record could cause trouble if it happened to have a
5527 target that, when qualified, matched something in the local domain. These
5528 two options are now turned off when doing these lookups.
5530 57. It seems that at least some releases of Reiserfs (which does not have the
5531 concept of a fixed number of inodes) returns zero and not -1 for the
5532 number of available inodes. This interacted badly with check_spool_inodes,
5533 which assumed that -1 was the "no such thing" setting. What I have done is
5534 to check that the total number of inodes is greater than zero before doing
5535 the test of how many are available.
5537 58. When a "warn" ACL statement has a log_message modifier, the message is
5538 remembered, and not repeated. This is to avoid a lot of repetition when a
5539 message has many recipients that cause the same warning to be written.
5540 Howewer, Exim was preserving the list of already written lines for an
5541 entire SMTP session, which doesn't seem right. The memory is now reset if a
5542 new message is started.
5544 59. The "rewrite" debugging flag was not showing the result of rewriting in the
5545 debugging output unless log_rewrite was also set.
5547 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
5548 of (int)(handle) when we know that handle contains (void *)(-1).
5550 61. The Exim daemon panic-logs an error return when it closes the incoming
5551 connection. However "connection reset by peer" seems to be common, and
5552 isn't really an error worthy of noting specially, so that particular error
5555 62. When Exim is trying to find all the local interfaces, it used to panic and
5556 die if the ioctl to get the interface flags failed. However, it seems that
5557 on at least one OS (Solaris 9) it is possible to have an interface that is
5558 included in the list of interfaces, but for which you get a failure error
5559 for this call. This happens when the interface is not "plumbed" into a
5560 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
5561 failure of the "get flags" call assumes that the interface is down.
5563 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
5564 makes life easier for people who are doing arithmetic on fields extracted
5565 from dates, where you often get leading zeros that should not be
5566 interpreted as octal.
5568 64. Added qualify_domain to the redirect router, to override the global
5571 65. If a pathologically long header line contained very many addresses (the
5572 report of this problem mentioned 10 000) and each of them was rewritten,
5573 Exim could use up a very large amount of memory. (It kept on making new
5574 copies of the header line as it rewrote, and never released the old ones.)
5575 At the expense of a bit more processing, the header rewriting function has
5576 been changed so that it no longer eats memory in this way.
5578 66. The generation of the Received: header has been moved from the time that a
5579 message starts to be received, to the time that it finishes. The timestamp
5580 in the Received: header should now be very close to that of the <= log
5581 line. There are two side-effects of this change:
5583 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
5584 logged header lines no longer include the local Received: line, because
5585 it has not yet been created. The same applies to a copy of the message
5586 that is returned to a non-SMTP sender when a message is rejected.
5588 (b) When a filter file is tested using -bf, no additional Received: header
5589 is added to the test message. After some thought, I decided that this
5592 This change does not affect the value of $received_for. It is still set
5593 after address rewriting, but before local_scan() is called.
5595 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
5597 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
5598 gave an unhelpful panic error message, and a defer error. I have managed to
5599 change this behaviour so that it now rejects any supplied certificate,
5600 which seems right, as the list of acceptable certificates is empty.
5602 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
5603 gave an unhelpful defer error. I have not managed to make this reject any
5604 supplied certificates, but the error message it gives is "no certificate
5605 supplied", which is not helpful.
5607 70. exigrep's output now also includes lines that are not associated with any
5608 message, but which match the given pattern. Implemented by a patch from
5609 Martin Sluka, which also tidied up the Perl a bit.
5611 71. Recipient callout verification, like sender verification, was using <> in
5612 the MAIL FROM command. This isn't really the right thing, since the actual
5613 sender may affect whether the remote host accepts the recipient or not. I
5614 have changed it to use the actual sender in the callout; this means that
5615 the cache record is now keyed on a recipient/sender pair, not just the
5616 recipient address. There doesn't seem to be a real danger of callout loops,
5617 since a callout by the remote host to check the sender would use <>.
5618 [SEE ABOVE: changed after hitting problems.]
5620 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
5621 temporary errors. However, in the case of such a code being given after
5622 the end of a data transmission (i.e. after ".") Exim was failing to write
5623 a retry record for the message. (Yes, there was some broken host that was
5624 actually sending 8xx at this point.)
5626 73. An unknown lookup type in a host list could cause Exim to panic-die when
5627 the list was checked. (An example that provoked this was putting <; in the
5628 middle of a list instead of at the start.) If this happened during a DATA
5629 ACL check, a -D file could be left lying around. This kind of configuration
5630 error no longer causes Exim to die; instead it causes a defer errror. The
5631 incident is still logged to the main and panic logs.
5633 74. Buglet left over from Exim 3 conversion. The message "too many messages
5634 in one connection" was written to the rejectlog but not the mainlog, except
5635 when address rewriting (yes!) was being logged.
5637 75. Added write_rejectlog option.
5639 76. When a system filter was run not as root (that is, when system_filter_user
5640 was set), the values of the $n variables were not being returned to the
5641 main process; thus, they were not subsequently available in the $sn
5644 77. Added +return_path_on_delivery log selector.
5646 78. A connection timeout was being treated differently from recipients deferred
5647 when testing hosts_max_try with a message that was older than the host's
5648 retry timeout. (The host should not be counted, thus allowing all hosts to
5649 be tried at least once before bouncing.) This may have been the cause of an
5650 occasionally reported bug whereby a message would remain on the queue
5651 longer than the retry timeout, but would be bounced if a delivery was
5652 forced. I say "may" because I never totally pinned down the problem;
5653 setting up timeout/retry tests is difficult. See also the next item.
5655 79. The ultimate address timeout was not being applied to errors that involved
5656 a combination of host plus message (for example, a timeout on a MAIL
5657 command). When an address resolved to a number of possible hosts, and they
5658 were not all tried for each delivery (e.g. because of hosts_max_try), a
5659 message could remain on the queue longer than the retry timeout.
5661 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
5664 81. Fixed an obscure SMTP outgoing bug which required at least the following
5665 conditions: (a) there was another message waiting for the same server;
5666 (b) the server returned 5xx to all RCPT commands in the first message so
5667 that the message was not completed; (c) the server dropped the connection
5668 or gave a negative response to the RSET that Exim sends to abort the
5669 transaction. The observed case was a dropped connection after DATA that had
5670 been sent in pipelining mode. That is, the server had advertised PIPELINING
5671 but was not implementing it correctly. The effect of the bug was incorrect
5672 behaviour, such as trying another host, and this could lead to a crash.
5678 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
5679 and daemon.c were passed as pointers to ints; they should have been
5680 pointers to socklen_t variables (which are typically unsigned ints).
5682 2. Some signed/unsigned type warnings in the os.c file for Linux have been
5685 3. Fixed a really odd bug that affected only the testing scheme; patching a
5686 certain fixed string in the binary changed the value of another string that
5687 happened to be identical to the end of the original first string.
5689 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
5690 name", it returns that address as the IP address. On some operating
5691 systems (e.g. Solaris), it also passes back the IP address string as the
5692 "host name". However, on others (e.g. Linux), it passes back an empty
5693 string. Exim wasn't checking for this, and was changing the host name to an
5694 empty string, assuming it had been canonicized.
5696 5. Although rare, it is permitted to have more than one PTR record for a given
5697 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
5698 all the names associated with an address, because they do in Solaris.
5699 However, it seems that they do not in Linux for data that comes from the
5700 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
5701 I found this out when I moved to a new Linux workstation and tried to run
5702 the Exim test suite.
5704 To get round this problem I have changed the code so that it now does its
5705 own call to the DNS to look up PTR records when searching for a host name.
5706 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
5707 addresses that are only in /etc/hosts are still found.
5709 This behaviour is, however, controlled by an option called host_lookup_
5710 order, which defaults to "bydns:byaddr". If people want to use the other
5711 order, or indeed, just use one or the other means of lookup, they can
5712 specify it in this variable.
5714 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
5715 some operating systems, this comes back from gethostbyaddr() as an empty
5716 string, and this is what Exim used to test for. However, it seems that in
5717 other systems, "." is yielded. Exim now tests for this case too.
5719 7. The values of check_spool_space and check_log_space are now held internally
5720 as a number of kilobytes instead of an absolute number of bytes. If a
5721 numbers is specified without 'K' or 'M', it is rounded up to the nearest
5722 kilobyte. This means that much larger values can be stored.
5724 8. Exim monitor: an attempt to get the action menu when not actually pointing
5725 at a message produces an empty menu entitled "No message selected". This
5726 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
5727 no entries in it ("Shell widget menu has zero width and/or height"). So I
5728 have added a single, blank menu entry in this case.
5730 9. Added ${quote_local_part.
5732 10. MIME decoding is now applied to the contents of Subject: header lines when
5735 11. Now that a reference to $sender_host_address automatically causes a reverse
5736 lookup to occur if necessary (4.13/18), there is no need to arrange for a
5737 host lookup before query-style lookups in lists that might use this
5738 variable. This has therefore been abolished, and the "net-" prefix is no
5739 longer necessary for query-style lookups.
5741 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
5742 have been a typo for LFLAGS, so it has been changed.
5744 13. The install script calls Exim with "-C /dev/null" in order to find the
5745 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
5746 to be output. Howeve, since Exim outputs its version number before the
5747 error, it didn't break the script. It just looked ugly. I fixed this by
5748 always allowing "-C /dev/null" if the caller is root.
5750 14. Ignore overlarge ACL variable number when reading spool file - insurance
5751 against a later release with more variables having written the file.
5753 15. The standard form for an IPv6 address literal was being rejected by EHLO.
5754 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5755 this, as well as the form without the "IPv6" on the front.
5757 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
5758 OS/Makefile-Darwin file.
5760 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
5761 only to LDAP libraries that do not have LDAP_OPT_DEREF.
5763 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
5764 variable. However, some OS use ints for inodes. I've added cast to long int
5765 to get rid of the compiler warning.
5767 19. I had forgotten to lock out "/../" in configuration file names when
5768 ALT_CONFIG_PREFIX was set.
5770 20. Routers used for verification do not need to specify transports. However,
5771 if such a router generated a host list, and callout was configured, Exim
5772 crashed, because it could not find a port number from the (non-existent)
5773 transport. It now assumes port 25 in this circumstance.
5775 21. Added the -t option to exigrep.
5777 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
5778 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
5779 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
5780 them is set, LOOKUP_LSEARCH is forced.
5782 23. "exim -bV" now outputs a list of lookups that are included in the binary.
5784 24. Added sender and host information to the "rejected by local_scan()" log
5785 line; previously there was no indication of these.
5787 25. Added .include_if_exists.
5789 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
5790 Linux. It turns out that not all file systems support this. Apparently some
5791 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
5792 do it.) To cope with this, the error EINVAL, which means that sync-ing is
5793 not supported on the file descriptor, is now ignored when Exim is trying to
5794 sync a directory. This applies only to Linux.
5796 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
5798 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
5799 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
5802 29. Added support for maildirsize files from supplied patch (modified a bit).
5804 30. The use of :fail: followed by an empty string could lead Exim to respond to
5805 sender verification failures with (e.g.):
5807 550 Verification failed for <xxx>
5808 550 Sender verify failed
5810 where the first response line was missing the '-' that indicates it is not
5811 the final line of the response.
5813 31. The loop for finding the name of the user that called Exim had a hardwired
5814 limit of 10; it now uses the value of finduser_retries, which is used for
5815 all other user lookups.
5817 32. Added $received_count variable, available in data and not_smtp ACLs, and at
5820 33. Exim was neglecting to zero errno before one call of strtol() when
5821 expanding a string and expecting an integer value. On some systems this
5822 resulted in spurious "integer overflow" errors. Also, it was casting the
5823 result into an int without checking.
5825 34. Testing for a connection timeout using "timeout_connect" in the retry rules
5826 did not work. The code looks as if it has *never* worked, though it appears
5827 to have been documented since at least releast 1.62. I have made it work.
5829 35. The "timeout_DNS" error in retry rules, also documented since at least
5830 1.62, also never worked. As it isn't clear exactly what this means, and
5831 clearly it isn't a major issue, I have abolished the feature by treating it
5832 as "timeout", and writing a warning to the main and panic logs.
5834 36. The display of retry rules for -brt wasn't always showing the error code
5837 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
5838 timeout_connect_A, timeout_connect_MX.
5840 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
5841 to the empty sender.
5843 39. The daemon was not analysing the content of -oX till after it had closed
5844 stderr and disconnected from the controlling terminal. This meant that any
5845 syntax errors were only noted on the panic log, and the return code from
5846 the command was 0. By re-arranging the code a little, I've made the
5847 decoding happen first, so such errors now appear on stderr, and the return
5848 code is 1. However, the actual setting up of the sockets still happens in
5849 the disconnected process, so errors there are still only recorded on the
5852 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
5853 connections (as happens on some IP stacks) was logged at start up time as
5854 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
5855 it from "IPv6 and IPv4", which means that two separate sockets are being
5858 41. The debug output for gethostbyname2() or getipnodebyname() failures now
5859 says whether AF_INET or AF_INET6 was passed as an argument.
5861 42. Exiwhat output was messed up when time zones were included in log
5864 43. Exiwhat now gives more information about the daemon's listening ports,
5865 and whether -tls-on-connect was used.
5867 44. The "port" option of the smtp transport is now expanded.
5869 45. A "message" modifier in a "warn" statement in a non-message ACL was being
5870 silently ignored. Now an error message is written to the main and panic
5873 46. There's a new ACL modifier called "logwrite" which writes to a log file
5874 as soon as it is encountered.
5876 47. Added $local_user_uid and $local_user_gid at routing time.
5878 48. Exim crashed when trying to verify a sender address that was being
5881 49. Exim was recognizing only a space character after ".include". It now also
5882 recognizes a tab character.
5884 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
5885 extracting the relevant information from the specification. The man page no
5886 longer contains scrambled data for the -d option, and I've added a section
5887 at the front about calling Exim under different names.
5889 51. Added "extra_headers" argument to the "mail" command in filter files.
5891 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
5894 53. Installed eximstats 1.29.
5896 54. Added transport_filter_timeout as a generic transport option.
5898 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
5899 Cc: header lines. This was required by RFC 822, but it not required by RFC
5902 56. Exim used to add From:, Date:, and Message-Id: header lines to any
5903 incoming messages that did not have them. Now it does so only if the
5904 message originates locally, that is, if there is no associated remote host
5905 address. When Resent- header lines are present, this applies to the Resent-
5906 lines rather than the non-Resent- lines.
5908 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
5909 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
5911 58. Messages for configuration errors now include the name of the main
5912 configuration file - useful now that there may be more than one file in a
5913 list (.included file names were always shown).
5915 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
5916 for those rare installations that do not start the daemon as root or run it
5917 setuid root. I've cut out the call to initgroups() if the daemon is not
5920 60. The Exim user and group can now be bound into the binary as text strings
5921 that are looked up at the start of Exim's processing.
5923 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
5925 62. Added $mailstore_basename variable.
5927 63. Installed patch to sieve.c from Michael Haardt.
5929 64. When Exim failed to open the panic log after failing to open the main log,
5930 the original message it was trying to log was written to stderr and debug
5931 output, but if they were not available (the usual case in production), it
5932 was lost. Now it is written to syslog before the two lines that record the
5933 failures to open the logs.
5935 65. Users' Exim filters run in subprocesses under the user's uid. It is
5936 possible for a "deliver" command or an alias in a "personal" command to
5937 provoke an address rewrite. If logging of address rewriting is configured,
5938 this fails because the process is not running as root or exim. There may be
5939 a better way of dealing with this, but for the moment (because 4.30 needs
5940 to be released), I have disabled address rewrite logging when running a
5941 filter in a non-root, non-exim process.
5947 1. The buildconfig auxiliary program wasn't quoting the value set for
5948 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
5949 not defined. This bug was masked in 4.22 by the effect that was fixed in
5952 2. Some messages that were rejected after a message id was allocated were
5953 shown as "incomplete" by exigrep. It no longer does this for messages that
5954 are rejected by local_scan() or the DATA or non-SMTP ACLs.
5956 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
5957 have allow_domain_literals set, the ID did not get logged in the <= line.
5958 Domain literals are now always recognized in Message-ID: header lines.
5960 4. The first argument for a ${extract expansion item is the key name or field
5961 number. Leading and trailing spaces in this item were not being ignored,
5962 causing some misleading effects.
5964 5. When deliver_drop_privilege was set, single queue runner processes started
5965 manually (i.e. by the command "exim -q") or by the daemon (which uses the
5966 same command in the process it spins off) were not dropping privilege.
5968 6. When the daemon running as "exim" started a queue runner, it always
5969 re-executed Exim in the spun-off process. This is a waste of effort when
5970 deliver_drop_privilege is set. The new process now just calls the
5971 queue-runner function directly.
5977 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
5980 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
5981 ignored. Though the use of -oP was forcing the writing of a pid file, it
5982 was always written to the default place.
5984 3. If the message "no IP address found for host xxxx" is generated during
5985 incoming verification, it is now followed by identification of the incoming
5986 connection (so you can more easily find what provoked it).
5988 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
5990 5. Added some features to "harden" Exim a bit more against certain attacks:
5992 (a) There is now a build-time option called FIXED_NEVER_USERS that can
5993 be put in Local/Makefile. This is like the never_users runtime option,
5994 but it cannot be overridden. The default setting is "root".
5996 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
5997 prefix string with which any file named in a -C command line option
6000 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
6001 is retained for -C and -D only if the caller of Exim is root. Without
6002 it, the exim user may also use -C and -D and retain privilege.
6004 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
6005 command line option is disabled.
6007 6. Macro names set by the -D option must start with an upper case letter, just
6008 like macro names defined in the configuration file.
6010 7. Added "dereference=" facility to LDAP.
6012 8. Two instances of the typo "uknown" in the source files are fixed.
6014 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
6015 the Configure-Makefile script screwed up while processing it.
6017 10. Incorporated PCRE 4.4.
6019 11. The SMTP synchronization check was not operating right at the start of an
6020 SMTP session. For example, it could not catch a HELO sent before the client
6021 waited for the greeting. There is now a check for outstanding input at the
6022 point when the greeting is written. Because of the duplex, asynchronous
6023 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
6024 way, but not yet received, when the check is performed.
6026 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
6027 on TCP/IP sockets, because this apparently causes some broken clients to
6030 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
6031 unchanged) from the Cygwin maintainer.
6033 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
6034 was supported instead of testing for mbx. Effectively a typo.
6036 15. The spa authenticator server code was not checking that the input it
6037 received was valid base64.
6039 16. The debug output line for the "set" modifier in ACLs was not showing the
6040 name of the variable that was being set.
6042 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
6044 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
6045 lookup on its own. Something else was needed to trigger the lookup. For
6046 example, a match in host_lookup or the need for a host name in a host list.
6047 Now, if $sender_host_name is referenced and the host name has not yet been
6048 looked up, a lookup is performed. If the lookup fails, the variable remains
6049 empty, and $host_lookup_failed is set to "1".
6051 19. Added "eqi" as a case-independent comparison operator.
6053 20. The saslauthd authentication condition could segfault if neither service
6054 nor realm was specified.
6056 21. If an overflowing value such as "2048M" was set for message_size_limit, the
6057 error message that was logged was misleading, and incoming SMTP
6058 connections were dropped. The message is now more accurate, and temporary
6059 errors are given to SMTP connections.
6061 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
6062 (except RSET) with a 421 error, until QUIT is received. However, it was
6063 failing to send a response to QUIT.
6065 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
6066 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
6067 ACL is now run after the helo_try_verify_hosts code.
6069 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
6070 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
6071 case-independent, so other case variants are also recognized). Apparently
6072 some systems use these upper case variants.
6074 25. If more than two messages were waiting for the same host, and a transport
6075 filter was specified for the transport, Exim sent two messages over the
6076 same TCP/IP connection, and then failed with "socket operation on non-
6077 socket" when it tried to send the third.
6079 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
6081 27. The extern definition of crypt16() in expand.c was not being excluded when
6082 the OS had its own crypt16() function.
6084 28. Added bounce_return_body as a new option, and bounce_return_size_limit
6085 as a preferred synonym for return_size_limit, both as an option and as an
6088 29. Added LIBS=-liconv to OS/Makefile-OSF1.
6090 30. Changed the default configuration ACL to relax the local part checking rule
6091 for addresses that are not in any local domains. For these addresses,
6092 slashes and pipe symbols are allowed within local parts, but the sequence
6093 /../ is explicitly forbidden.
6095 31. SPA server authentication was not clearing the challenge buffer before
6098 32. log_message in a "warn" ACL statement was writing to the reject log as
6099 well as to the main log, which contradicts the documentation and doesn't
6100 seem right (because no rejection is happening). So I have stopped it.
6102 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
6103 However, I am unable to do any testing of this.
6105 34. Fixed an infelicity in the appendfile transport. When checking directories
6106 for a mailbox, to see if any needed to be created, it was accidentally
6107 using path names with one or more superfluous leading slashes; tracing
6108 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
6110 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
6111 contents are added to the log line that is written for every discarded
6112 recipient. (Previously a log_message setting was ignored.)
6114 36. The ${quote: operator now quotes the string if it is empty.
6116 37. The install script runs exim in order to find its version number. If for
6117 some reason other than non-existence or emptiness, which it checks, it
6118 could not run './exim', it was installing it with an empty version number,
6119 i.e. as "exim-". This error state is now caught, and the installation is
6122 38. An argument was missing from the function that creates an error message
6123 when Exim fails to connect to the socket for saslauthd authentication.
6124 This could cause Exim to crash, or give a corrupted message.
6126 39. Added isip, isip4, and isip6 to ${if conditions.
6128 40. The ACL variables $acl_xx are now saved with the message, and can be
6129 accessed later in routers, transports, and filters.
6131 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
6132 strings in the file are not string-expanded.
6134 42. If a MAIL command specified a SIZE value that was too large to fit into an
6135 int variable, the check against message_size_limit failed. Such values are
6136 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
6137 day this will have to be increased, but I don't think I want to be around
6138 when emails are that large.
6145 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
6146 iconv() is not standard in FreeBSD.
6148 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
6149 IPv6 enabled. The observed symptom was a segmentation fault on return from
6150 the function os_common_find_running_interfaces() in src/os.c.
6152 3. In the check_special_case() function in daemon.c I had used "errno" as an
6153 argument name, which causes warnings on some systems. This was basically a
6154 typo, since it was named "eno" in the comments!
6156 4. The code that waits for the clock to tick (at a resolution of some fraction
6157 of a second) so as to ensure message-id uniqueness was always waiting for
6158 at least one whole tick, when it could have waited for less. [This is
6159 almost certainly not relevant at current processor speeds, where it is
6160 unlikely to ever wait at all. But we try to future-proof.]
6162 5. The function that sleeps for a time interval that includes fractions of a
6163 second contained a race. It did not block SIGALRM between setting the
6164 timer, and suspending (a couple of lines later). If the interval was short
6165 and the sigsuspend() was delayed until after it had expired, the suspension
6166 never ended. On busy systems this could lead to processes getting stuck for
6169 6. Some uncommon configurations may cause a lookup to happen in a queue runner
6170 process, before it forks any delivery processes. The open lookup caching
6171 mechanism meant that the open file or database connection was passed into
6172 the delivery process. The problem was that delivery processes always tidy
6173 up cached lookup data. This could cause a problem for the next delivery
6174 process started by the queue runner, because the external queue runner
6175 process does not know about the closure. So the next delivery process
6176 still has data in the lookup cache. In the case of a file lookup, there was
6177 no problem because closing a file descriptor in a subprocess doesn't affect
6178 the parent. However, if the lookup was caching a connection to a database,
6179 the connection was closed, and the second delivery process was likely to
6180 see errors such as "PGSQL: query failed: server closed the connection
6181 unexpectedly". The problem has been fixed by closing all cached lookups
6182 in a queue runner before running a delivery process.
6184 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
6185 seem to have the "const" qualifier which it has on other OS. I've
6188 8. Change 4.21/2 was too strict. It is only if there are two authenticators
6189 *of the same type* (client or server) with the same public name that an
6190 error should be diagnosed.
6192 9. When Exim looked up a host name for an IP address, but failed to find the
6193 original IP address when looking up the host name (a safety check), it
6194 output the message "<ip address> does not match any IP for NULL", which was
6195 confusing, to say the least. The bug was that the host name should have
6196 appeared instead of "NULL".
6198 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
6199 user that is built into the binary, and the -C or -D options is used, root
6200 privilege is dropped before the configuration file is read. In addition,
6201 logging is switched to stderr instead of the normal log files. If the
6202 configuration then re-defines the Exim user, the unprivileged environment
6203 is probably not what is expected, so Exim logs a panic warning message (but
6206 However, if deliver_drop_privilege is set, the unprivileged state may well
6207 be exactly what is intended, so the warning has been cut out in that case,
6208 and Exim is allowed to try to write to its normal log files.
6214 1. smtp_return_error_details was not giving details for temporary sender
6215 or receiver verification errors.
6217 2. Diagnose a configuration error if two authenticators have the same public
6220 3. Exim used not to create the message log file for a message until the first
6221 delivery attempt. This could be confusing when incoming messages were held
6222 for policy or load reasons. The message log file is now created at the time
6223 the message is received, and an initial "Received" line is written to it.
6225 4. The automatically generated man page for command line options had a minor
6226 bug that caused no ill effects; however, a more serious problem was that
6227 the procedure for building the man page automatically didn't always
6228 operate. Consequently, release 4.20 contains an out-of-date version. This
6229 shouldn't happen again.
6231 5. When building Exim with embedded Perl support, the script that builds the
6232 Makefile was calling 'perl' to find its compile-time parameters, ignoring
6233 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
6235 6. The freeze_tell option was not being used for messages that were frozen on
6236 arrival, either by an ACL or by local_scan().
6238 7. Added the smtp_incomplete_transaction log selector.
6240 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
6241 was accepting AUTH without a new EHLO.
6243 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
6244 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
6247 10. Logging of TCP/IP connections (when configured) now happens in the main
6248 daemon process instead of the child process, so that the TCP/IP connection
6249 count is more accurate (but it can never be perfect).
6251 11. The use of "drop" in a nested ACL was not being handled correctly in the
6252 outer ACL. Now, if condition failure induced by the nested "drop" causes
6253 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
6254 or "require"), the connection is dropped.
6256 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
6257 that yield "discard" can now be used with an "accept" or a "discard" verb,
6258 but an error is generated for any others (because I can't see a useful way
6259 to define what should happen).
6261 13. When an ACL is read dynamically from a file (or anywhere else), the lines
6262 are now processed in the same way as lines in the Exim configuration file.
6263 In particular, continuation lines are supported.
6265 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
6267 15. Added -ti meaning -t -i.
6269 16. Check for letters, digits, hyphens, and dots in the names of dnslist
6270 domains, and warn by logging if others are found.
6272 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
6273 returned from GIFCONF when Exim is trying to find the list of interfaces on
6274 a host. The code in os.c has been modified to copy each ifreq to an aligned
6275 structure in all cases.
6277 Also, in some cases, the returned ifreq's were being copied to a 'struct
6278 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
6279 means the last couple of bytes of an IPv6 address could be chopped if the
6280 ifreq contained only a normal sockaddr (14 bytes storage).
6282 18. Named domain lists were not supported in the hosts_treat_as_local option.
6283 An entry such as +xxxx was not recognized, and was treated as a literal
6286 19. Ensure that header lines added by a DATA ACL are included in the reject log
6287 if the ACL subsequently rejects the message.
6289 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
6290 MD5 (which is deprecated).
6292 21. When testing a filter file using -bf, Exim was writing a message when it
6293 took the sender from a "From " line in the message, but it was not doing so
6294 when it took $return_path from a Return-Path: header line. It now does.
6296 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
6297 with a valid header line field name (a series of printing characters
6298 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
6300 23. Changed "disc" in the source to "disk" to conform to the documentation and
6301 the book and for uniformity.
6303 24. Ignore Sendmail's -Ooption=value command line item.
6305 25. When execve() failed while trying to run a command in a pipe transport,
6306 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
6307 could be confused with a return value of 69 from the command itself. This
6308 has been changed to 127, the value the shell returns if it is asked to run
6309 a non-existent command. The wording for the related log line suggests a
6310 non-existent command as the problem.
6312 26. If received_header_text expands to an empty string, do not add a Received:
6313 header line to the message. (Well, it adds a token one on the spool, but
6314 marks it "old" so that it doesn't get used or transmitted.)
6316 27. Installed eximstats 1.28 (addition of -nt option).
6318 28. There was no check for failure on the call to getsockname() in the daemon
6319 code. This can fail if there is a shortage of resources on the system, with
6320 ENOMEM, for example. A temporary error is now given on failure.
6322 29. Contrary to the C standard, it seems that in some environments, the
6323 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
6324 program. Exim now does this explicitly; it affects the formatting of
6325 timestamps using strftime().
6327 30. If exiqsumm was given junk data, it threw up some uninitialized variable
6328 complaints. I've now initialized all the variables, to avoid this.
6330 32. Header lines added by a system filter were not being "seen" during
6331 transport-time rewrites.
6333 33. The info_callback() function passed to OpenSSL is set up with type void
6334 (*)(SSL *, int, int), as described somewhere. However, when calling the
6335 function (actually a macro) that sets it up, the type void(*)() is
6336 expected. I've put in a cast to prevent warnings from picky compilers.
6338 34. If a DNS black list lookup found a CNAME record, but there were no A
6339 records associated with the domain it pointed at, Exim crashed.
6341 35. If a DNS black list lookup returned more than one A record, Exim ignored
6342 all but the first. It now scans all returned addresses if a particular IP
6343 value is being sought. In this situation, the contents of the
6344 $dnslist_value variable are a list of all the addresses, separated by a
6347 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
6348 to accept a host name and all its aliases if the forward lookup for any of
6349 them yielded the IP address of the incoming connection. Now it accepts only
6350 those names whose forward lookup yields the correct IP address. Any other
6351 names are discarded. This closes a loophole whereby a rogue DNS
6352 administrator could create reverse DNS records to break through a
6353 wildcarded host restriction in an ACL.
6355 37. If a user filter or a system filter that ran in a subprocess used any of
6356 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
6357 the wrong values were passed to the pipe command ($thisaddress had the
6358 value of $0, $0 had the value of $1, etc). This bug was introduced by
6359 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
6361 38. Improved the line breaking for long SMTP error messages from ACLs.
6362 Previously, if there was no break point between 40 and 75 characters, Exim
6363 left the rest of the message alone. Two changes have been made: (a) I've
6364 reduced the minimum length to 35 characters; (b) if it can't find a break
6365 point between 35 and 75 characters, it looks ahead and uses the first one
6366 that it finds. This may give the occasional overlong line, but at least the
6367 remaining text gets split now.
6369 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
6370 file descriptors might be low, and that setting 1000 would always raise it.
6371 It turns out that in some environments, the limit is already over 1000 and
6372 that lowering it causes trouble. So now Exim takes care not to decrease it.
6374 40. When delivering a message, the value of $return_path is set to $sender_
6375 address at the start of routing (routers may change the value). By an
6376 oversight, this default was not being set up when an address was tested by
6377 -bt or -bv, which affected the outcome if any router or filter referred to
6380 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
6381 log only when "message" or "log_message" are set. However, if one of the
6382 conditions was an address verification, or a call to a nested ACL, the
6383 messages generated by the underlying test were being passed through. This
6384 no longer happens. The underlying message is available in $acl_verify_
6385 message for both "message" and "log_message" expansions, so it can be
6386 passed through if needed.
6388 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
6389 new expansion $bh_ to give the encoded byte string without charset
6390 translation. Translation happens only if iconv() is available; HAVE_ICONV
6391 indicates this at build time. HEADERS_CHARSET gives the charset to
6392 translate to; headers_charset can change it in the configuration, and
6393 "headers charset" can change it in an individual filter file.
6395 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
6396 that creates RFC 2047 encoded "words" labels them as that charset instead
6397 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
6398 expansion operator; (ii) when Exim creates a From: line for a local
6399 message; (iii) when a header line is rewritten to include a "phrase" part.
6401 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
6402 buggy, causing it to skip the first lines of messages whose message ID
6403 ended in 'D'. This would not have bitten before Exim release 4.14, because
6404 message IDs were unlikely to end in 'D' before then. The effect was to have
6405 incorrect size information for certain domains.
6407 45. #include "config.h" was missing at the start of the crypt16.c module. This
6408 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
6411 46. If there was a timeout during a "random" callout check, Exim treated it as
6412 a failure of the random address, and carried on sending RSET and the real
6413 address. If the delay was just some slowness somewhere, the response to the
6414 original RCPT would be taken as a response to RSET and so on, causing
6415 mayhem of various kinds.
6417 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
6418 when I implemented it. It didn't allow for the fact that some option values
6419 may legitimatetly be negative (e.g. size_addition), and it didn't even do
6420 the right test for positive values.
6422 48. Domain names in DNS records are case-independent. Exim always looks them up
6423 in lower case. Some resolvers return domain names in exactly the case they
6424 appear in the zone file, that is, they may contain uppercase letters. Not
6425 all resolvers do this - some return always lower case. Exim was treating a
6426 change of case by a resolver as a change of domain, similar to a widening
6427 of a domain abbreviation. This triggered its re-routing code and so it was
6428 trying to route what was effectively the same domain again. This normally
6429 caused routing to fail (because the router wouldn't handle the domain
6430 twice). Now Exim checks for this case specially, and just changes the
6431 casing of the domain that it ultimately uses when it transmits the message
6434 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
6437 50. If a filter generated a file delivery with a non-absolute name (possible if
6438 no home directory exists for the router), the forbid_file option was not
6441 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
6442 the existing equality matching.
6444 52. Exim was using ints instead of ino_t variables in some places where it was
6445 dealing with inode numbers.
6447 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
6448 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
6449 called TMPDIR, and if it finds it is different, it changes its value.
6451 54. The smtp_printf() function is now made available to local_scan() so
6452 additional output lines can be written before returning. There is also an
6453 smtp_fflush() function to enable the detection of a dropped connection.
6454 The variables smtp_input and smtp_batched_input are exported to
6457 55. Changed the default runtime configuration: the message "Unknown user"
6458 has been removed from the ACL, and instead placed on the localuser router,
6459 using the cannot_route_message feature. This means that any verification
6460 failures that generate their own messages won't get overridden. Similarly,
6461 the "Unrouteable address" message that was in the ACL for unverifiable
6462 relay addresses has also been removed.
6464 56. Added hosts_avoid_esmtp to the smtp transport.
6466 57. The exicyclog script was not checking for the esoteric option
6467 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
6468 will work only if exicyclog is run under the appropriate euid.
6470 58. Following a discussion on the list, the rules by which Exim recognises line
6471 endings on incoming messages have been changed. The -dropcr and drop_cr
6472 options are now no-ops, retained only for backwards compatibility. The
6473 following line terminators are recognized: LF CRLF CR. However, special
6474 processing applies to CR:
6476 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
6477 nor a local message in the state where . is a terminator.
6479 (ii) If a bare CR is encountered in a header line, an extra space is added
6480 after the line terminator so as not to end the header. The reasoning
6481 behind this is that bare CRs in header lines are most likely either
6482 to be mistakes, or people trying to play silly games.
6484 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
6485 was being incorrectly given as 18 bytes larger than it should have been.
6486 This is a VOB (very old bug).
6488 60. This may never have affected anything current, but just in case it has:
6489 When the local host is found other than at the start of a list of hosts,
6490 the local host, those with the same MX, and any that follow, are discarded.
6491 When the list in question was part of a longer list of hosts, the following
6492 hosts (not currently being processed) were also being discarded. This no
6493 longer happens. I'm not sure if this situation could ever has previously
6496 61. Added the "/MX" feature to lists of hosts in the manualroute and query
6499 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
6500 header. This is something that is recommended in a new Internet Draft, and
6501 is something that is documented as being done by Sendmail. There are two
6502 possible values. For messages generated by the autoreply transport, Exim
6505 Auto-Submitted: auto-replied
6507 whereas for all other generated messages (e.g. bounces) it adds
6509 Auto-Submitted: auto-generated
6511 63. The "personal" condition in filters now includes a test for the
6512 Auto-Submitted: header. If it contains the string "auto-" the message it
6513 not considered personal.
6515 64. Added rcpt_include_affixes as a generic transport option.
6517 65. Added queue_only_override (default true).
6519 66. Added the syslog_duplication option.
6521 67. If what should have been the first header line of a message consisted of
6522 a space followed by a colon, Exim was mis-interpreting it as a header line.
6523 It isn't of course - it is syntactically invalid and should therefore be
6524 treated as the start of the message body. The misbehaviour could have
6525 caused a number of strange effects, including loss of data in subsequent
6526 header lines, and spool format errors.
6528 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
6529 client host had authenticated. This control can now be exercised by an ACL
6530 for more flexibility.
6532 69. By default, callouts do not happen when testing with -bh. There is now a
6533 variant, -bhc, which does actually run the callout code, including
6534 consulting and updating the callout cache.
6536 70. Added support for saslauthd authentication, courtesy of Alexander
6539 71. If statvfs() failed on the spool or log directories while checking their
6540 size for availability, Exim confusingly gave the error "space shortage".
6541 Furthermore, in debugging mode it crashed with a floating point exception.
6542 These checks are done if check_{spool,log}_{space,inodes} are set, and when
6543 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
6544 serious problem, Exim now writes to the main and panic logs when this
6545 happens, with details of the failure. It then refuses to accept the
6546 incoming message, giving the message "spool directory problem" or "log
6547 directory problem" with a 421 code for SMTP messages.
6549 72. When Exim is about to re-exec itself, it ensures that the file descriptors
6550 0, 1, and 2 exist, because some OS complain for execs without them (see
6551 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
6552 descriptors. However, the code omitted to check that the open succeeded,
6553 causing mysterious errors if for some reason the permissions on /dev/null
6554 got screwed. Now Exim writes a message to the main and panic logs, and
6555 bombs out if it can't open /dev/null.
6557 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
6558 interact so that it is all more flexible. It is supposed to remain
6559 backwards compatible. Also added extra_local_interfaces.
6561 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
6562 to bomb out with an assertion failure - to the client this appears as a
6563 connection drop. This problem occurs in the part of the code that was taken
6564 from the Samba project. Fortunately, the assertion is in a very simple
6565 function, so I have fixed this by reproducing the function inline in the
6566 one place where it is called, and arranging for authentication to fail
6567 instead of killing the process with assert().
6569 75. The SPA client code was not working when the server requested OEM rather
6570 than Unicode encoding.
6572 76. Added code to make require_files with a specific uid setting more usable in
6573 the case where statting the file as root fails - usually a non-root-mounted
6574 NFS file system. When this happens and the failure is EACCES, Exim now
6575 forks a subprocess and does the per-uid checking as the relevant uid.
6577 77. Added process_log_path.
6579 78. If log_file_path was not explicitly set, a setting of check_log_space or
6580 check_log_inodes was ignored.
6582 79. If a space check for the spool or log partitions fails, the incident is now
6583 logged. Of course, in the latter case the data may get lost...
6585 80. Added the %p formatting code to string_format() so that it can be used to
6586 print addresses in debug_print(). Adjusted all the address printing in the
6587 debugging in store.c to use %p rather than %d.
6589 81. There was a concern that a line of code in smtp_in.c could overflow a
6590 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
6591 initially expressed, the concern was not well-founded, because trailing
6592 spaces are removed early. However, if the trailing spaces were followed by
6593 a NULL, they did not get removed, so the overflow was possible. Two fixes
6596 (a) I re-wrote the offending code in a cleaner fashion.
6597 (b) If an incoming SMTP command contains a NULL character, it is rejected
6600 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
6601 runs initgroups(), so that if the Exim user is in any additional groups,
6602 they will be used during message reception.
6608 The change log for 4.20 and earlier releases has been archived.
projects / exim.git / blob