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
107 JH/28 Enable {spool,log} filesystem space and inode checks as default.
108 Main config options check_{log,spool}_{inodes,space} are now
109 100 inodes, 10MB unless set otherwise in the configuration.
114 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
115 and 3.4.4 - once the server is enabled to respond to an OCSP request
116 it does even when not requested, resulting in a stapling non-aware
117 client dropping the TLS connection.
119 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
120 support variable-length bit vectors. No functional change.
122 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
123 The I= interface field on outgoing lines is now after the H= remote
124 host field, same as incoming lines. There is a separate
125 outgoing_interface log selector which allows you to disable the
128 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
129 If not running log_selector +smtp_connection the mainlog would be held
130 open indefinitely after a "too many connections" event, including to a
131 deleted file after a log rotate. Leave the per net connection logging
132 leaving it open for efficiency as that will be quickly detected by the
133 check on the next write.
135 HS/01 Bug 1671: Fix post transport crash.
136 Processing the wait-<transport> messages could crash the delivery
137 process if the message IDs didn't exist for some reason. When
138 using 'split_spool_directory=yes' the construction of the spool
139 file name failed already, exposing the same netto behaviour.
141 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
142 mime_regex ACL conditions.
144 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
145 to DSN fail messages (bounces): remote IP, remote greeting, remote response
146 to HELO, local diagnostic string.
148 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
149 log line to debug. Even when configured with a tls authenticator many
150 client connections are expected to not authenticate in this way, so
151 an authenticate fail is not an error.
153 HS/02 Add the Exim version string to the process info. This way exiwhat
154 gives some more detail about the running daemon.
156 JH/06 Bug 1395: time-limit cacheing of DNS lookups, to the TTL value. This may
157 matter for fast-change records such as DNSBLs.
159 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
160 retry record, even if constant. There may be multiple transports with
161 different interface settings and the retry behaviour needs to be kept
164 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
166 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
168 JH/10 Bug 840: fix log_defer_output option of pipe transport
170 JH/11 Bug 830: use same host for all RCPTS of a message, even under
171 hosts_randomize. This matters a lot when combined with mua_wrapper.
173 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
174 ${quote_pgsql:<string>} operator.
176 JH/13 Bug 1708: avoid misaligned access in cached lookup.
178 JH/14 Change header file name for freeradius-client. Relevant if compiling
179 with Radius support; from the Gentoo tree and checked under Fedora.
181 JH/15 Bug 1712: Introduce $prdr_requested flag variable
183 JH/16 Bug 1714: Permit an empty string as expansion result for transport
184 option transport_filter, meaning no filtering.
186 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
188 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
189 defaults to "*" (all hosts). The variable is now available when not built
190 with TLS, default unset, mainly to enable keeping the testuite sane.
191 If a server certificate is not supplied (via tls_certificate) an error is
192 logged, and clients will find TLS connections fail on startup. Presumably
193 they will retry in-clear.
194 Packagers of Exim are strongly encouraged to create a server certificate
195 at installation time.
197 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
198 with the $config_file variable.
200 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
201 in transport context, after the attempt, and per-recipient. The latter type
202 is per host attempted. The event data is the error message, and the errno
203 information encodes the lookup type (A vs. MX) used for the (first) host,
204 and the trailing two digits of the smtp 4xx reponse.
206 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
207 to write to mainlog (or rejectlog, paniclog) in the window between file
208 creation and permissions/ownership being changed. Particularly affects
209 installations where exicyclog is run as root, rather than exim user;
210 result is that the running daemon panics and dies.
212 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
214 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
215 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
216 "pri" and "weight". Note that the previous implicit priority given by the
217 list order is no longer honoured.
219 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalisation
222 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
223 by defining SUPPORT_SOCKS.
225 JH/26 Move PROXY support from Experimental to mainline, enabled for a build
226 by defining SUPPORT_PROXY. Note that the proxy_required_hosts option
227 is renamed to hosts_proxy, and the proxy_{host,target}_{address,port}.
228 variables are renamed to proxy_{local,external}_{address,port}.
230 JH/27 Move Internationalisation support from Experimental to mainline, enabled
231 for a build by defining SUPPORT_I18N
233 JH/28 Bug 1745: Fix redis lookups to handle (quoted) spaces embedded in parts
234 of the query string, and make ${quote_redis:} do that quoting.
236 JH/29 Move Events support from Experimental to mainline, enabled by default
237 and removable for a build by defining DISABLE_EVENT.
239 JH/30 Updated DANE implementation code to current from Viktor Dukhovni.
241 JH/31 Fix bug with hosts_connection_nolog and named-lists which were wrongly
242 cached by the daemon.
244 JH/32 Move Redis support from Experimental to mainline, enabled for a build
245 by defining LOOKUP_REDIS. The libhiredis library is required.
247 JH/33 Bug 1748: Permit ACL dnslists= condition in non-smtp ACLs if explicit
248 keys are given for lookup.
250 JH/34 Bug 1192: replace the embedded copy of PolarSSL RSA routines in the DKIM
251 support, by using OpenSSL or GnuTLS library ones. This means DKIM is
252 only supported when built with TLS support. The PolarSSL SHA routines
253 are still used when the TLS library is too old for convenient support.
255 JH/35 Require SINGLE_DH_USE by default in OpenSSL (main config option
256 openssl_options), for security. OpenSSL forces this from version 1.1.0
257 server-side so match that on older versions.
259 JH/36 Bug 1778: longstanding bug in memory use by the ${run } expansion: A fresh
260 allocation for $value could be released as the expansion processing
261 concluded, but leaving the global pointer active for it.
263 JH/37 Bug 1769: Permit a VRFY ACL to override the default 252 response,
264 and to use the domains and local_parts ACL conditions.
266 JH/38 Fix cutthrough bug with body lines having a single dot. The dot was
267 incorrectly not doubled on cutthrough transmission, hence seen as a
268 body-termination at the receiving system - resulting in truncated mails.
269 Commonly the sender saw a TCP-level error, and retransmitted the message
270 via the normal store-and-forward channel. This could result in duplicates
271 received - but deduplicating mailstores were liable to retain only the
272 initial truncated version.
274 JH/39 Bug 1781: Fix use of DKIM private-keys having trailing '=' in the base-64.
276 JH/40 Fix crash in queryprogram router when compiled with EXPERIMENTAL_SRS.
278 JH/41 Bug 1792: Fix selection of headers to sign for DKIM: bottom-up. While
279 we're in there, support oversigning also; bug 1309.
281 JH/42 Bug 1796: Fix error logged on a malware scanner connection failure.
283 HS/04 Add support for keep_environment and add_environment options.
285 JH/43 Tidy coding issues detected by gcc --fsanitize=undefined. Some remain;
286 either intentional arithmetic overflow during PRNG, or testing config-
289 JH/44 Bug 1800: The combination of a -bhc commandline option and cutthrough
290 delivery resulted in actual delivery. Cancel cutthrough before DATA
293 JH/45 Fix cutthrough, when connection not opened by verify and target hard-
294 rejects a recipient: pass the reject to the originator.
296 JH/46 Multiple issues raised by Coverity. Some were obvious or plausible bugs.
297 Many were false-positives and ignorable, but it's worth fixing the
300 JH/47 Fix build on HP-UX and older Solaris, which need (un)setenv now also
301 for the new environment-manipulation done at startup. Move the routines
302 from being local to tls.c to being global via the os.c file.
304 JH/48 Bug 1807: Fix ${extract } for the numeric/3-string case. While preparsing
305 an extract embedded as result-arg for a map, the first arg for extract
306 is unavailable so we cannot tell if this is a numbered or keyed
307 extraction. Accept either.
313 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
316 JH/02 The smtp transport option "multi_domain" is now expanded.
318 JH/03 The smtp transport now requests PRDR by default, if the server offers
321 JH/04 Certificate name checking on server certificates, when exim is a client,
322 is now done by default. The transport option tls_verify_cert_hostnames
323 can be used to disable this per-host. The build option
324 EXPERIMENTAL_CERTNAMES is withdrawn.
326 JH/05 The value of the tls_verify_certificates smtp transport and main options
327 default to the word "system" to access the system default CA bundle.
328 For GnuTLS, only version 3.0.20 or later.
330 JH/06 Verification of the server certificate for a TLS connection is now tried
331 (but not required) by default. The verification status is now logged by
332 default, for both outbound TLS and client-certificate supplying inbound
335 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
338 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
339 Status Notification (bounce) messages are now MIME format per RFC 3464.
340 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
341 under the control of the dsn_advertise_hosts option, and routers may
342 have a dsn_lasthop option.
344 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
345 default, modifiable by a malware= option. The list separator for
346 the options can now be changed in the usual way. Bug 68.
348 JH/10 The smtp_receive_timeout main option is now expanded before use.
350 JH/11 The incoming_interface log option now also enables logging of the
351 local interface on delivery outgoing connections.
353 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
354 if the interface and destination host and port all match.
356 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
359 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
360 Patch from Andrew Lewis.
362 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
363 now supports optional time-restrictions, weighting, and priority
364 modifiers per server. Patch originally by <rommer@active.by>.
366 JH/16 The spamd_address main option now supports a mixed list of local
367 and remote servers. Remote servers can be IPv6 addresses, and
368 specify a port-range.
370 JH/17 Bug 68: The spamd_address main option now supports an optional
371 timeout value per server.
373 JH/18 Bug 1581: Router and transport options headers_add/remove can
374 now have the list separator specified.
376 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
379 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
382 JH/21 Support for the A6 type of dns record is withdrawn.
384 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
385 rather than the verbs used.
387 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
388 from 255 to 1024 chars.
390 JH/24 Verification callouts now attempt to use TLS by default.
392 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
393 are generic router options now. The defaults didn't change.
395 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
396 Original patch from Alexander Shikoff, worked over by JH.
398 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
399 files if ZCAT_COMMAND is not executable.
401 JH/26 Bug 1539: Add timout/retry options on dnsdb lookups.
403 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
405 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
406 Normally benign, it bites when the pair was led to by a CNAME;
407 modern usage is to not canoicalize the domain to a CNAME target
408 (and we were inconsistent anyway for A-only vs AAAA+A).
410 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
412 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
413 when evaluating $sender_host_dnssec.
415 JH/31 Check the HELO verification lookup for DNSSEC, adding new
416 $sender_helo_dnssec variable.
418 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
420 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
422 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
424 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
425 documented as working, but never had. Support all but $spam_report.
427 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
428 added for tls authenticator.
430 HS/03 Add perl_taintmode main config option
435 TL/01 When running the test suite, the README says that variables such as
436 no_msglog_check are global and can be placed anywhere in a specific
437 test's script, however it was observed that placement needed to be near
438 the beginning for it to behave that way. Changed the runtest perl
439 script to read through the entire script once to detect and set these
440 variables, reset to the beginning of the script, and then run through
441 the script parsing/test process like normal.
443 TL/02 The BSD's have an arc4random API. One of the functions to induce
444 adding randomness was arc4random_stir(), but it has been removed in
445 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
446 function when detected.
448 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
449 cause callback expansion.
451 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
452 syntax errors in an expansion can be treated as a string instead of
453 logging or causing an error, due to the internal use of bool_lax
454 instead of bool when processing it.
456 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
457 server certificates when making smtp deliveries.
459 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
461 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
463 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
465 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
466 Merged patch from Sebastian Wiedenroth.
468 JH/05 Fix results-pipe from transport process. Several recipients, combined
469 with certificate use, exposed issues where response data items split
470 over buffer boundaries were not parsed properly. This eventually
471 resulted in duplicates being sent. This issue only became common enough
472 to notice due to the introduction of conection certificate information,
473 the item size being so much larger. Found and fixed by Wolfgang Breyha.
475 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
476 size buffer was used, resulting in syntax errors when an expansion
479 JH/07 Add support for directories of certificates when compiled with a GnuTLS
480 version 3.3.6 or later.
482 JH/08 Rename the TPDA expermimental facility to Event Actions. The #ifdef
483 is EXPERIMENTAL_EVENT, the main-configuration and transport options
484 both become "event_action", the variables become $event_name, $event_data
485 and $event_defer_errno. There is a new variable $verify_mode, usable in
486 routers, transports and related events. The tls:cert event is now also
487 raised for inbound connections, if the main configuration event_action
490 TL/06 In test suite, disable OCSP for old versions of openssl which contained
491 early OCSP support, but no stapling (appears to be less than 1.0.0).
493 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
494 server certificate names available under the smtp transport option
495 "tls_verify_cert_hostname" now do not permit multi-component wildcard
498 JH/10 Time-related extraction expansions from certificates now use the main
499 option "timezone" setting for output formatting, and are consistent
500 between OpenSSL and GnuTLS compilations. Bug 1541.
502 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
503 encoded parameter in the incoming message. Bug 1558.
505 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
506 include certificate info, eximon was claiming there were spoolfile
509 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
511 JH/14 Log delivery-related information more consistently, using the sequence
512 "H=<name> [<ip>]" wherever possible.
514 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
515 are problematic for Debian distribution, omit them from the release
518 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
520 JH/16 Fix string representation of time values on 64bit time_t anchitectures.
523 JH/17 Fix a null-indirection in certextract expansions when a nondefault
524 output list separator was used.
529 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
530 checkers that were complaining about end of non-void function with no
533 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
534 This was a regression intruduced in 4.83 by another bugfix.
536 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
538 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
539 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
545 TF/01 Correctly close the server side of TLS when forking for delivery.
547 When a message was received over SMTP with TLS, Exim failed to clear up
548 the incoming connection properly after forking off the child process to
549 deliver the message. In some situations the subsequent outgoing
550 delivery connection happened to have the same fd number as the incoming
551 connection previously had. Exim would try to use TLS and fail, logging
552 a "Bad file descriptor" error.
554 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
555 utilities have not been installed.
557 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
558 temporary space as the ACL may create new global variables.
560 TL/01 LDAP support uses per connection or global context settings, depending
561 upon the detected version of the libraries at build time.
563 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
564 to extract and use the src ip:port in logging and expansions as if it
565 were a direct connection from the outside internet. PPv2 support was
566 updated based on HAProxy spec change in May 2014.
568 JH/02 Add ${listextract {number}{list}{success}{fail}}.
570 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
571 Properly escape header and check for NULL return.
573 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
576 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
578 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
579 characters in header names, implemented as a verify condition.
580 Contributed by Michael Fischer v. Mollard.
582 TL/05 Rename SPF condition results err_perm and err_temp to standardized
583 results permerror and temperror. Previous values are deprecated but
584 still accepted. In a future release, err_perm and err_temp will be
585 completely removed, which will be a backward incompatibility if the
586 ACL tests for either of these two old results. Patch contributed by
587 user bes-internal on the mailing list.
589 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
591 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
592 selectors, in both main and reject logs.
594 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
597 JH/07 Add malware type "sock" for talking to simple daemon.
599 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
601 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
602 routers/transports under cutthrough routing.
604 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
605 numbers. Touch up "bool" conditional to keep the same definition.
607 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
609 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
611 JH/12 Expand items in router/transport headers_add or headers_remove lists
612 individually rather than the list as a whole. Bug 1452.
614 Required for reasonable handling of multiple headers_ options when
615 they may be empty; requires that headers_remove items with embedded
616 colons must have them doubled (or the list-separator changed).
618 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
619 view the policy declared in the DMARC record. Currently, $dmarc_status
620 is a combined value of both the record presence and the result of the
623 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
625 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
626 dnslookup router and the smtp transport (applying to the forward
629 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
630 of ldap servers used for a specific lookup. Patch provided by Heiko
633 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
634 New variable $lookup_dnssec_authenticated for observability.
636 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
637 Patch submitted by Lars Timman.
639 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
641 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
642 Requires trusted mode and valid format message id, aborts otherwise.
643 Patch contributed by Heiko Schlichting.
645 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
646 certextract with support for various fields. Bug 1358.
648 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
649 is requested by default, modifiable by smtp transport option
652 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
653 operate on certificate variables to give certificate fingerprints
654 Also new ${sha256:cert_variable}.
656 JH/23 The PRDR feature is moved from being Experimental into the mainline.
658 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
659 Christian Aistleitner.
661 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
663 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
664 file. Patch from Wolfgang Breyha.
666 JH/25 Expand the coverage of the delivery $host and $host_address to
667 client authenticators run in verify callout. Bug 1476.
669 JH/26 Port service names are now accepted for tls_on_connect_ports, to
670 align with daemon_smtp_ports. Bug 72.
672 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
673 support and error reporting did not work properly.
675 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
676 and is readable. Patch from Andrew Colin Kissa.
678 TL/14 Enhance documentation of ${run expansion and how it parses the
679 commandline after expansion, particularly in the case when an
680 unquoted variable expansion results in an empty value.
682 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
684 PP/02 Fix internal collision of T_APL on systems which support RFC3123
685 by renaming away from it. Addresses GH issue 15, reported by
688 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
690 TL/15 SECURITY: prevent double expansion in math comparison functions
691 (can expand unsanitized data). Not remotely exploitable.
698 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
700 PP/02 Make -n do something, by making it not do something.
701 When combined with -bP, the name of an option is not output.
703 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
706 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
707 $sender_host_name and config options to manage this, and basic check
710 PP/05 DSCP support for outbound connections and control modifier for inbound.
712 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
713 (Only plugin which currently uses this is kerberos4, which nobody should
714 be using, but we should make it available and other future plugins might
715 conceivably use it, even though it would break NAT; stuff *should* be
716 using channel bindings instead).
718 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
719 name; added for Sendmail compatibility; requires admin caller.
720 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
721 just ignore it); requires trusted caller.
722 Also parse but ignore: -Ac -Am -X<logfile>
725 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
727 TL/02 Add +smtp_confirmation as a default logging option.
729 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
730 Patch by Magnus Holmgren from 2007-02-20.
732 TL/04 Bugzilla 1281 - Spec typo.
733 Bugzilla 1283 - Spec typo.
734 Bugzilla 1290 - Spec grammar fixes.
736 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
738 TL/06 Add Experimental DMARC support using libopendmarc libraries.
740 TL/07 Fix an out of order global option causing a segfault. Reported to dev
741 mailing list by by Dmitry Isaikin.
743 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
745 JH/02 Support "G" suffix to numbers in ${if comparisons.
747 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
749 NM/01 Bugzilla 1197 - Spec typo
750 Bugzilla 1196 - Spec examples corrections
752 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
754 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
755 gnutls_enable_pkcs11, but renamed to more accurately indicate its
758 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
759 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
761 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
762 "acl {{name}{arg}...}", and optional args on acl condition
765 JH/05 Permit multiple router/transport headers_add/remove lines.
767 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
769 JH/07 Avoid using a waiting database for a single-message-only transport.
770 Performance patch from Paul Fisher. Bugzilla 1262.
772 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
775 JH/09 Add $headers_added variable, with content from use of ACL modifier
776 add_header (but not yet added to the message). Bugzilla 199.
778 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
779 Pulled from Bugzilla 817 by Wolfgang Breyha.
781 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
783 (nb: this is the same fix as in Exim 4.80.1)
785 JH/11 Add A= logging on delivery lines, and a client_set_id option on
788 JH/12 Add optional authenticated_sender logging to A= and a log_selector
791 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
793 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
794 advertise SMTP AUTH mechanism to us, instead of a generic
795 protocol violation error. Also, make Exim more robust to bad
796 data from the Dovecot auth socket.
798 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
800 When a queue runner is handling a message, Exim first routes the
801 recipient addresses, during which it prunes them based on the retry
802 hints database. After that it attempts to deliver the message to
803 any remaining recipients. It then updates the hints database using
806 So if a recipient address works intermittently, it can get repeatedly
807 deferred at routing time. The retry hints record remains fresh so the
808 address never reaches the final cutoff time.
810 This is a fairly common occurrence when a user is bumping up against
811 their storage quota. Exim had some logic in its local delivery code
812 to deal with this. However it did not apply to per-recipient defers
813 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
815 This change adds a proper retry rule check during routing so that the
816 final cutoff time is checked against the message's age. We only do
817 this check if there is an address retry record and there is not a
818 domain retry record; this implies that previous attempts to handle
819 the address had the retry_use_local_parts option turned on. We use
820 this as an approximation for the destination being like a local
821 delivery, as in LMTP.
823 I suspect this new check makes the old local delivery cutoff check
824 redundant, but I have not verified this so I left the code in place.
826 TF/02 Correct gecos expansion when From: is a prefix of the username.
828 Test 0254 submits a message to Exim with the header
832 When I ran the test suite under the user fanf2, Exim expanded
833 the header to contain my full name, whereas it should have added
834 a Resent-Sender: header. It erroneously treats any prefix of the
835 username as equal to the username.
837 This change corrects that bug.
839 GF/01 DCC debug and logging tidyup
840 Error conditions log to paniclog rather than rejectlog.
841 Debug lines prefixed by "DCC: " to remove any ambiguity.
843 TF/03 Avoid unnecessary rebuilds of lookup-related code.
845 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
846 Bug spotted by Jeremy Harris; was flawed since initial commit.
847 Would have resulted in OCSP responses post-SNI triggering an Exim
848 NULL dereference and crash.
850 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
852 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
853 Bug detection, analysis and fix by Samuel Thibault.
854 Bugzilla 1331, Debian bug #698092.
856 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
858 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
859 Server implementation by Todd Lyons, client by JH.
860 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
861 config variable "prdr_enable" controls whether the server
862 advertises the facility. If the client requests PRDR a new
863 acl_data_smtp_prdr ACL is called once for each recipient, after
864 the body content is received and before the acl_smtp_data ACL.
865 The client is controlled by bolth of: a hosts_try_prdr option
866 on the smtp transport, and the server advertisement.
867 Default client logging of deliveries and rejections involving
868 PRDR are flagged with the string "PRDR".
870 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
871 fclose(). Diagnosis by Todd Lyons.
873 PP/17 Update configure.default to handle IPv6 localhost better.
874 Patch by Alain Williams (plus minor tweaks).
877 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
878 This is now consistent with GnuTLS, and is now documented: the
879 previous undocumented portable approach to treating the option as
880 unset was to force an expansion failure. That still works, and
881 an empty string is now equivalent.
883 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
884 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
885 not performing validation itself.
887 PP/20 Added force_command boolean option to pipe transport.
888 Patch from Nick Koston, of cPanel Inc.
890 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
893 TF/04 Added udpsend ACL modifer and hexquote expansion operator
895 PP/21 Fix eximon continuous updating with timestamped log-files.
896 Broken in a format-string cleanup in 4.80, missed when I repaired the
897 other false fix of the same issue.
898 Report and fix from Heiko Schlichting.
901 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
902 Report from Prashanth Katuri.
904 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
905 It's SecureTransport, so affects any MacOS clients which use the
906 system-integrated TLS libraries, including email clients.
908 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
909 using a MIME ACL for non-SMTP local injection.
910 Report and assistance in diagnosis by Warren Baker.
912 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
914 JH/16 Fix comparisons for 64b. Bugzilla 1385.
916 TL/09 Add expansion variable $authenticated_fail_id to keep track of
917 last id that failed so it may be referenced in subsequent ACL's.
919 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
922 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
923 ldap library initialization, allowing self-signed CA's to be
924 used. Also properly sets require_cert option later in code by
925 using NULL (global ldap config) instead of ldap handle (per
926 session). Bug diagnosis and testing by alxgomz.
928 TL/12 Enhanced documentation in the ratelimit.pl script provided in
929 the src/util/ subdirectory.
931 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
932 renamed to Transport Post Delivery Action by Jeremy Harris, as
935 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
936 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
937 redis_servers = needs to be configured which will be used by the redis
938 lookup. Patch from Warren Baker, of The Packet Hub.
940 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
942 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
943 hostname or reverse DNS when processing a host list. Used suggestions
944 from multiple comments on this bug.
946 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
948 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
949 Missed a few lines, added it to make the runtest require no keyboard
952 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
953 contains upper case chars. Make router use caseful_local_part.
955 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
956 support when GnuTLS has been built with p11-kit.
962 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
964 This, or similar/improved, will also be change PP/11 of 4.82.
970 PP/01 Handle short writes when writing local log-files.
971 In practice, only affects FreeBSD (8 onwards).
972 Bugzilla 1053, with thanks to Dmitry Isaikin.
974 NM/01 Bugzilla 949 - Documentation tweak
976 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
979 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
981 PP/02 Implemented gsasl authenticator.
983 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
985 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
986 `pkg-config foo` for cflags/libs.
988 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
989 with rest of GSASL and with heimdal_gssapi.
991 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
992 `pkg-config foo` for cflags/libs for the TLS implementation.
994 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
995 properties get this fed in as external SSF. A number of robustness
996 and debugging improvements to the cyrus_sasl authenticator.
998 PP/08 cyrus_sasl server now expands the server_realm option.
1000 PP/09 Bugzilla 1214 - Log authentication information in reject log.
1001 Patch by Jeremy Harris.
1003 PP/10 Added dbmjz lookup type.
1005 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
1007 PP/12 MAIL args handles TAB as well as SP, for better interop with
1008 non-compliant senders.
1009 Analysis and variant patch by Todd Lyons.
1011 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
1012 Bug report from Lars Müller <lars@samba.org> (via SUSE),
1013 Patch from Dirk Mueller <dmueller@suse.com>
1015 PP/13 tls_peerdn now print-escaped for spool files.
1016 Observed some $tls_peerdn in wild which contained \n, which resulted
1017 in spool file corruption.
1019 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
1020 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
1021 or write after TLS renegotiation, which otherwise led to messages
1024 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
1025 as a tracking header (ie: a signed header comes before the signature).
1026 Patch from Wolfgang Breyha.
1028 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
1029 comma-sep list; embedded commas doubled.
1031 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
1033 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
1035 Report and patch from Dmitry Banschikov.
1037 PP/16 Removed "dont_insert_empty_fragments" fron "openssl_options".
1038 Removed SSL_clear() after SSL_new() which led to protocol negotiation
1039 failures. We appear to now support TLS1.1+ with Exim.
1041 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
1042 lets Exim select keys and certificates based upon TLS SNI from client.
1043 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
1044 before an outbound SMTP session. New log_selector, +tls_sni.
1046 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
1047 NULL dereference. Report and patch from Alun Jones.
1049 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
1050 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
1051 Not seeing resolver debug output on NetBSD, but suspect this is a
1052 resolver implementation change.
1054 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
1055 Left warnings. Added "eximon gdb" invocation mode.
1057 PP/21 Defaulting "accept_8bitmime" to true, not false.
1059 PP/22 Added -bw for inetd wait mode support.
1061 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
1062 locate the relevant includes and libraries. Made this the default.
1064 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
1065 Bugzilla 1246, report and most of solution from Tomasz Kusy.
1067 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
1068 This may cause build issues on older platforms.
1070 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
1071 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
1072 gnutls_require_mac & gnutls_require_protocols (no longer supported).
1073 Added SNI support via GnuTLS too.
1074 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
1076 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
1078 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
1079 Applied second patch from Janne, implementing suggestion to default
1080 multiple-strings-in-record handling to match SPF spec.
1082 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
1084 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
1085 read-only, out of scope).
1086 Patch from Wolfgang Breyha, report from Stuart Northfield.
1088 PP/29 Fix three issues highlighted by clang analyser static analysis.
1089 Only crash-plausible issue would require the Cambridge-specific
1090 iplookup router and a misconfiguration.
1091 Report from Marcin Mirosław.
1093 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
1095 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
1096 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
1097 As part of this, removing so much warning spew let me fix some minor
1098 real issues in debug logging.
1100 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
1101 assignment on my part. Fixed.
1103 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
1104 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
1105 Janne Snabb (who went above and beyond: thank you).
1107 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
1108 string otherwise requires a connection and a bunch more work and it's
1109 relatively easy to get wrong. Should also expose TLS library linkage
1112 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
1113 64-bit ${eval} (JH/03).
1115 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
1116 GNU libc to support some of the 64-bit stuff, should not lead to
1117 conflicts. Defined before os.h is pulled in, so if a given platform
1118 needs to override this, it can.
1120 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
1121 protection layer was required, which is not implemented.
1122 Bugzilla 1254, patch from Wolfgang Breyha.
1124 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
1125 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
1126 tls_dhparam take prime identifiers. Also unbreak combination of
1127 OpenSSL+DH_params+TLSSNI.
1129 PP/39 Disable SSLv2 by default in OpenSSL support.
1135 PP/01 Solaris build fix for Oracle's LDAP libraries.
1136 Bugzilla 1109, patch from Stephen Usher.
1138 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
1140 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
1143 TF/02 Fix a couple more cases where we did not log the error message
1144 when unlink() failed. See also change 4.74-TF/03.
1146 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
1147 lock up or crash if it happened to be inside a call to libc when it
1148 got a SIGUSR1 from exiwhat.
1150 The SIGUSR1 handler appends the current process status to the process
1151 log which is later printed by exiwhat. It used to use the general
1152 purpose logging code to do this, but several functions it calls are
1153 not safe for signals.
1155 The new output code in the SIGUSR1 handler is specific to the process
1156 log, and simple enough that it's easy to inspect for signal safety.
1157 Removing some special cases also simplifies the general logging code.
1158 Removing the spurious timestamps from the process log simplifies
1161 TF/04 Improved ratelimit ACL condition.
1163 The /noupdate option has been deprecated in favour of /readonly which
1164 has clearer semantics. The /leaky, /strict, and /readonly update modes
1165 are mutually exclusive. The update mode is no longer included in the
1166 database key; it just determines when the database is updated. (This
1167 means that when you upgrde Exim will forget old rate measurements.)
1169 Exim now checks that the per_* options are used with an update mode that
1170 makes sense for the current ACL. For example, when Exim is processing a
1171 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
1172 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
1173 must specify per_mail/readonly. If you omit the update mode it defaults to
1174 /leaky where that makes sense (as before) or /readonly where required.
1176 The /noupdate option is now undocumented but still supported for
1177 backwards compatibility. It is equivalent to /readonly except that in
1178 ACLs where /readonly is required you may specify /leaky/noupdate or
1179 /strict/noupdate which are treated the same as /readonly.
1181 A useful new feature is the /count= option. This is a generalization
1182 of the per_byte option, so that you can measure the throughput of other
1183 aggregate values. For example, the per_byte option is now equivalent
1184 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
1186 The per_rcpt option has been generalized using the /count= mechanism
1187 (though it's more complicated than the per_byte equivalence). When it is
1188 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
1189 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
1190 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
1191 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
1192 Note that using per_rcpt with a non-readonly update mode in more than
1193 one ACL will cause the recipients to be double-counted. (The per_mail
1194 and per_byte options don't have this problem.)
1196 The handling of very low rates has changed slightly. If the computed rate
1197 is less than the event's count (usually one) then this event is the first
1198 after a long gap. In this case the rate is set to the same as this event's
1199 count, so that the first message of a spam run is counted properly.
1201 The major new feature is a mechanism for counting the rate of unique
1202 events. The new per_addr option counts the number of different
1203 recipients that someone has sent messages to in the last time period. It
1204 behaves like per_rcpt if all the recipient addresses are different, but
1205 duplicate recipient addresses do not increase the measured rate. Like
1206 the /count= option this is a general mechanism, so the per_addr option
1207 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
1208 example, measure the rate that a client uses different sender addresses
1209 with the options per_mail/unique=$sender_address. There are further
1210 details in the main documentation.
1212 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
1214 TF/06 Removed a few PCRE remnants.
1216 TF/07 Automatically extract Exim's version number from tags in the git
1217 repository when doing development or release builds.
1219 PP/02 Raise smtp_cmd_buffer_size to 16kB.
1220 Bugzilla 879. Patch from Paul Fisher.
1222 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
1223 Heavily based on revision 40f9a89a from Simon Arlott's tree.
1226 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
1228 PP/05 Variable $av_failed, true if the AV scanner deferred.
1229 Bugzilla 1078. Patch from John Horne.
1231 PP/06 Stop make process more reliably on build failure.
1232 Bugzilla 1087. Patch from Heiko Schlittermann.
1234 PP/07 Make maildir_use_size_file an _expandable_ boolean.
1235 Bugzilla 1089. Patch from Heiko Schlittermann.
1237 PP/08 Handle ${run} returning more data than OS pipe buffer size.
1238 Bugzilla 1131. Patch from Holger Weiß.
1240 PP/09 Handle IPv6 addresses with SPF.
1241 Bugzilla 860. Patch from Wolfgang Breyha.
1243 PP/10 GnuTLS: support TLS 1.2 & 1.1.
1245 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
1248 PP/11 match_* no longer expand right-hand-side by default.
1249 New compile-time build option, EXPAND_LISTMATCH_RHS.
1250 New expansion conditions, "inlist", "inlisti".
1252 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
1254 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
1256 PP/14 fix log_write() format string regression from TF/03.
1257 Bugzilla 1152. Patch from Dmitry Isaikin.
1263 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
1265 PP/02 Harmonised TLS library version reporting; only show if debugging.
1266 Layout now matches that introduced for other libraries in 4.74 PP/03.
1268 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
1270 PP/04 New "dns_use_edns0" global option.
1272 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
1275 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
1276 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
1278 TK/01 Updated PolarSSL code to 0.14.2.
1279 Bugzilla 1097. Patch from Andreas Metzler.
1281 PP/07 Catch divide-by-zero in ${eval:...}.
1282 Fixes bugzilla 1102.
1284 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
1287 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
1288 format-string attack -- SECURITY: remote arbitrary code execution.
1290 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
1291 time unintentionally subject to list matching rules, letting the header
1292 cause arbitrary Exim lookups (of items which can occur in lists, *not*
1293 arbitrary string expansion). This allowed for information disclosure.
1295 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
1296 INT_MIN/-1 -- value coerced to INT_MAX.
1302 NM/01 Workround for PCRE version dependency in version reporting
1305 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
1306 This fixes portability to compilers other than gcc, notably
1307 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
1309 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
1310 makefiles for portability to HP-UX and POSIX correctness.
1312 PP/01 Permit LOOKUP_foo enabling on the make command-line.
1313 Also via indented variable definition in the Makefile.
1314 (Debugging by Oliver Heesakkers).
1316 PP/02 Restore caching of spamd results with expanded spamd_address.
1317 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
1319 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
1320 Improves build reliability. Fix from: Frank Elsner
1322 NM/02 Fix wide character breakage in the rfc2047 coding
1323 Fixes bug 1064. Patch from Andrey N. Oktyabrski
1325 NM/03 Allow underscore in dnslist lookups
1326 Fixes bug 1026. Patch from Graeme Fowler
1328 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
1329 Code patches from Adam Ciarcinski of NetBSD.
1331 NM/04 Fixed exiqgrep to cope with mailq missing size issue
1334 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
1335 is logged, to avoid truncation. Patch from John Horne.
1337 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
1338 Patch from Jakob Hirsch.
1340 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
1341 SQL string expansion failure details.
1342 Patch from Andrey Oktyabrski.
1344 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
1345 Patch from Simon Arlott.
1347 PP/09 New lookups functionality failed to compile on old gcc which rejects
1348 extern declarations in function scope.
1349 Patch from Oliver Fleischmann
1351 PP/10 Use sig_atomic_t for flags set from signal handlers.
1352 Check getgroups() return and improve debugging.
1353 Fixed developed for diagnosis in bug 927 (which turned out to be
1356 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
1357 Patch from Mark Zealey.
1359 PP/12 Bugzilla 1056: Improved spamd server selection.
1360 Patch from Mark Zealey.
1362 PP/13 Bugzilla 1086: Deal with maildir quota file races.
1363 Based on patch from Heiko Schlittermann.
1365 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
1366 Patch from Uwe Doering, sign-off by Michael Haardt.
1368 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
1369 variable declaration deep within a block. Bug and patch from
1372 PP/15 lookups-Makefile IRIX compatibilty coercion.
1374 PP/16 Make DISABLE_DKIM build knob functional.
1376 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
1377 Patch by Simon Arlott
1379 TF/03 Fix valgrind.h portability to C89 compilers that do not support
1380 variable argument macros. Our copy now differs from upstream.
1386 TF/01 Failure to get a lock on a hints database can have serious
1387 consequences so log it to the panic log.
1389 TF/02 Log LMTP confirmation messages in the same way as SMTP,
1390 controlled using the smtp_confirmation log selector.
1392 TF/03 Include the error message when we fail to unlink a spool file.
1394 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
1395 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
1396 for maintaining out-of-tree patches for some time.
1398 PP/01 Bugzilla 139: Documentation and portability issues.
1399 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
1400 Handle per-OS dynamic-module compilation flags.
1402 PP/02 Let /dev/null have normal permissions.
1403 The 4.73 fixes were a little too stringent and complained about the
1404 permissions on /dev/null. Exempt it from some checks.
1405 Reported by Andreas M. Kirchwitz.
1407 PP/03 Report version information for many libraries, including
1408 Exim version information for dynamically loaded libraries. Created
1409 version.h, now support a version extension string for distributors
1410 who patch heavily. Dynamic module ABI change.
1412 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
1413 privilege escalation vulnerability whereby the Exim run-time user
1414 can cause root to append content of the attacker's choosing to
1417 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
1420 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
1421 If dropping privileges for untrusted macros, we disabled normal logging
1422 on the basis that it would fail; for the Exim run-time user, this is not
1423 the case, and it resulted in successful deliveries going unlogged.
1424 Fixed. Reported by Andreas Metzler.
1430 PP/01 Date: & Message-Id: revert to normally being appended to a message,
1431 only prepend for the Resent-* case. Fixes regression introduced in
1432 Exim 4.70 by NM/22 for Bugzilla 607.
1434 PP/02 Include check_rfc2047_length in configure.default because we're seeing
1435 increasing numbers of administrators be bitten by this.
1437 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
1439 PP/03 Bugzilla 994: added openssl_options main configuration option.
1441 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
1443 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
1445 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
1447 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
1448 without a peer certificate, leading to a segfault because of an
1449 assumption that peers always have certificates. Be a little more
1450 paranoid. Problem reported by Martin Tscholak.
1452 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
1453 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
1454 NB: ClamAV planning to remove STREAM in "middle of 2010".
1455 CL also introduces -bmalware, various -d+acl logging additions and
1456 more caution in buffer sizes.
1458 PP/09 Implemented reverse_ip expansion operator.
1460 PP/10 Bugzilla 937: provide a "debug" ACL control.
1462 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
1464 PP/12 Bugzilla 973: Implement --version.
1466 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
1468 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
1470 PP/15 Bugzilla 816: support multiple condition rules on Routers.
1472 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
1473 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
1474 ignore trailing whitespace.
1476 JJ/02 prevent non-panic DKIM error from being sent to paniclog
1478 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
1481 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
1482 Notification from Dr Andrew Aitchison.
1484 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
1485 ExtendedDetectionInfo response format.
1486 Notification from John Horne.
1488 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
1491 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
1492 XSL and documented dependency on system catalogs, with examples of how
1495 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
1498 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
1499 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
1500 configuration file which is writeable by the Exim user or group.
1502 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
1503 of configuration files to cover files specified with the -C option if
1504 they are going to be used with root privileges, not just the default
1507 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
1508 option (effectively making it always true).
1510 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
1511 files to be used while preserving root privileges.
1513 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
1514 that rogue child processes cannot use them.
1516 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
1517 run-time user, instead of root.
1519 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
1520 Exim run-time user without dropping privileges.
1522 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
1523 result string, instead of calling string_vformat() twice with the same
1526 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
1527 for other users. Others should always drop root privileges if they use
1528 -C on the command line, even for a whitelisted configure file.
1530 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
1532 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
1538 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
1539 $data_path, and $header_path variables; fixed documentation bugs and
1542 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
1543 exipick to access non-standard spools, including the "frozen" queue
1546 NM/01 Bugzilla 965: Support mysql stored procedures.
1547 Patch from Alain Williams
1549 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
1551 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
1552 Patch from Andreas Metzler
1554 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
1555 Patch from Kirill Miazine
1557 NM/05 Bugzilla 671: Added umask to procmail example.
1559 JJ/03 installed exipick 20100323.0, fixing doc bug
1561 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
1562 directory. Notification and patch from Dan Rosenberg.
1564 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
1566 TK/02 Improve log output when DKIM signing operation fails.
1568 MH/01 Treat the transport option dkim_domain as a colon separated
1569 list, not as a single string, and sign the message with each element,
1570 omitting multiple occurences of the same signer.
1572 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
1573 Bugzilla 985, 986. Patch by Simon Arlott
1575 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
1576 Patch by Simon Arlott
1578 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
1579 MBX locking. Notification from Dan Rosenberg.
1585 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
1587 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
1589 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
1591 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
1593 NM/04 Bugzilla 915: Flag broken perl installation during build.
1599 TK/01 Added patch by Johannes Berg that expands the main option
1600 "spamd_address" if it starts with a dollar sign.
1602 TK/02 Write list of recipients to X-Envelope-Sender header when building
1603 the mbox-format spool file for content scanning (suggested by Jakob
1606 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
1607 (http://www.dcc-servers.net/) support via dccifd. Activated by
1608 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
1610 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
1611 by Mark Daniel Reidel <mr@df.eu>.
1613 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
1614 When building exim an external PCRE library is now needed -
1615 PCRE is a system library on the majority of modern systems.
1616 See entry on PCRE_LIBS in EDITME file.
1618 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
1619 conversation. Added nologin parameter to request.
1620 Patch contributed by Kirill Miazine.
1622 TF/01 Do not log submission mode rewrites if they do not change the address.
1624 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
1626 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
1627 log files in place. Contributed by Roberto Lima.
1629 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
1631 TF/03 Bugzilla 615: When checking the local_parts router precondition
1632 after a local_part_suffix or local_part_prefix option, Exim now
1633 does not use the address's named list lookup cache, since this
1634 contains cached lookups for the whole local part.
1636 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
1637 Robert Millan. Documentation is in experimental-spec.txt.
1639 TF/04 Bugzilla 668: Fix parallel build (make -j).
1641 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
1643 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
1644 Patch provided by Jan Srzednicki.
1646 TF/05 Leading white space used to be stripped from $spam_report which
1647 wrecked the formatting. Now it is preserved.
1649 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
1650 that they are available at delivery time.
1652 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
1654 TF/08 TLS error reporting now respects the incoming_interface and
1655 incoming_port log selectors.
1657 TF/09 Produce a more useful error message if an SMTP transport's hosts
1658 setting expands to an empty string.
1660 NM/06 Bugzilla 744: EXPN did not work under TLS.
1661 Patch provided by Phil Pennock.
1663 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
1664 Patch provided by Richard Godbee.
1666 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
1667 acl_smtp_notquit, added index entry.
1669 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
1670 Patch provided by Eugene Bujak.
1672 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
1673 accept(). Patch provided by Maxim Dounin.
1675 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
1676 Patch provided by Phil Pennock.
1678 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
1680 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
1681 Patch provided by Brad "anomie" Jorsch.
1683 NM/14 Bugzilla 622: Added timeout setting to transport filter.
1684 Patch provided by Dean Brooks.
1686 TK/05 Add native DKIM support (does not depend on external libraries).
1688 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
1689 Patch provided by Graeme Fowler.
1691 NM/16 Bugzilla 851: Documentation example syntax fix.
1693 NM/17 Changed NOTICE file to remove references to embedded PCRE.
1695 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
1698 NM/19 Bugzilla 745: TLS version reporting.
1699 Patch provided by Phil Pennock.
1701 NM/20 Bugzilla 167: bool: condition support.
1702 Patch provided by Phil Pennock.
1704 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
1705 clients. Patch provided by Phil Pennock.
1707 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
1708 Patch provided by Brad "anomie" Jorsch.
1710 NM/23 Bugzilla 687: Fix misparses in eximstats.
1711 Patch provided by Heiko Schlittermann.
1713 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
1714 Patch provided by Heiko Schlittermann.
1716 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
1717 plus update to original patch.
1719 NM/26 Bugzilla 799: Documentation correction for ratelimit.
1721 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
1722 Patch provided by David Brownlee.
1724 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
1726 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
1728 NM/30 Bugzilla 888: TLS documentation bugfixes.
1730 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
1732 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
1733 Unlike the original bugzilla I have changed all shell scripts in src tree.
1735 NM/33 Bugzilla 898: Transport filter timeout fix.
1736 Patch by Todd Rinaldo.
1738 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
1739 Patch by Serge Demonchaux.
1741 NM/35 Bugzilla 39: Base64 decode bug fixes.
1742 Patch by Jakob Hirsch.
1744 NM/36 Bugzilla 909: Correct connect() call in dcc code.
1746 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
1748 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
1750 NM/39 Bugzilla 911: Fixed MakeLinks build script.
1756 TK/01 Add preliminary DKIM support. Currently requires a forked version of
1757 ALT-N's libdkim that I have put here:
1758 http://duncanthrax.net/exim-experimental/
1760 Note to Michael Haardt: I had to rename some vars in sieve.c. They
1761 were called 'true' and it seems that C99 defines that as a reserved
1762 keyword to be used with 'bool' variable types. That means you could
1763 not include C99-style headers which use bools without triggering
1764 build errors in sieve.c.
1766 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
1767 as mailq or other aliases. Changed the --help handling significantly
1768 to do whats expected. exim_usage() emits usage/help information.
1770 SC/01 Added the -bylocaldomain option to eximstats.
1772 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
1774 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
1776 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
1782 PH/01 Another patch from the Sieve maintainer.
1784 PH/02 When an IPv6 address is converted to a string for single-key lookup
1785 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
1786 dots are used instead of colons so that keys in lsearch files need not
1787 contain colons. This was done some time before quoting was made available
1788 in lsearch files. However, iplsearch files do require colons in IPv6 keys
1789 (notated using the quote facility) so as to distinguish them from IPv4
1790 keys. This meant that lookups for IP addresses in host lists did not work
1791 for iplsearch lookups.
1793 This has been fixed by arranging for IPv6 addresses to be expressed with
1794 colons if the lookup type is iplsearch. This is not incompatible, because
1795 previously such lookups could never work.
1797 The situation is now rather anomolous, since one *can* have colons in
1798 ordinary lsearch keys. However, making the change in all cases is
1799 incompatible and would probably break a number of configurations.
1801 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
1804 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
1805 conversion specification without a maximum field width, thereby enabling
1806 a rogue spamd server to cause a buffer overflow. While nobody in their
1807 right mind would setup Exim to query an untrusted spamd server, an
1808 attacker that gains access to a server running spamd could potentially
1809 exploit this vulnerability to run arbitrary code as the Exim user.
1811 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
1812 $primary_hostname instead of what libspf2 thinks the hosts name is.
1814 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
1815 a directory entry by the name of the lookup key. Previously, if a
1816 symlink pointed to a non-existing file or a file in a directory that
1817 Exim lacked permissions to read, a lookup for a key matching that
1818 symlink would fail. Now it is enough that a matching directory entry
1819 exists, symlink or not. (Bugzilla 503.)
1821 PH/03 The body_linecount and body_zerocount variables are now exported in the
1824 PH/04 Added the $dnslist_matched variable.
1826 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
1827 This means they are set thereafter only if the connection becomes
1830 PH/06 Added the client_condition to authenticators so that some can be skipped
1831 by clients under certain conditions.
1833 PH/07 The error message for a badly-placed control=no_multiline_responses left
1834 "_responses" off the end of the name.
1836 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
1838 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
1839 (without spaces) instead of just copying the configuration text.
1841 PH/10 Added the /noupdate option to the ratelimit ACL condition.
1843 PH/11 Added $max_received_linelength.
1845 PH/12 Added +ignore_defer and +include_defer to host lists.
1847 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
1848 way in which PCRE > 7.0 is built.
1850 PH/14 Implemented queue_only_load_latch.
1852 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
1853 MAIL command. The effect was to mangle the value on 64-bit systems.
1855 PH/16 Another patch from the Sieve maintainer.
1857 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
1859 PH/18 If a system quota error occurred while trying to create the file for
1860 a maildir delivery, the message "Mailbox is full" was not appended to the
1861 bounce if the delivery eventually timed out. Change 4.67/27 below applied
1862 only to a quota excession during the actual writing of the file.
1864 PH/19 It seems that peer DN values may contain newlines (and other non-printing
1865 characters?) which causes problems in log lines. The DN values are now
1866 passed through string_printing() before being added to log lines.
1868 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
1869 and InterBase are left for another time.)
1871 PH/21 Added message_body_newlines option.
1873 PH/22 Guard against possible overflow in moan_check_errorcopy().
1875 PH/23 POSIX allows open() to be a macro; guard against that.
1877 PH/24 If the recipient of an error message contained an @ in the local part
1878 (suitably quoted, of course), incorrect values were put in $domain and
1879 $local_part during the evaluation of errors_copy.
1885 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
1886 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
1889 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
1890 issue a MAIL command.
1892 PH/02 In an ACL statement such as
1894 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
1896 if a client was not listed at all, or was listed with a value other than
1897 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
1898 the condition was not true (as it should be), so access was not denied.
1899 The bug was that the ! inversion was incorrectly passed on to the second
1900 item. This has been fixed.
1902 PH/03 Added additional dnslists conditions == and =& which are different from
1903 = and & when the dns lookup returns more than one IP address.
1905 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
1906 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
1908 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
1909 FSYNC, which compiles an option called disable_fsync that allows for
1910 bypassing fsync(). The documentation is heavily laced with warnings.
1912 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
1914 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
1915 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
1916 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
1917 including adding "make clean"; (3) Added -fPIC when compiling the test
1918 dynamically loaded module, to get rid of a warning.
1920 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
1921 message fails, move_frozen_messages = true and ignore_bounce_errors_after
1922 = 0s. The bug is otherwise harmless.
1924 PH/07 There was a bug in the dovecot authenticator such that the value of
1925 $auth1 could be overwritten, and so not correctly preserved, after a
1926 successful authentication. This usually meant that the value preserved by
1927 the server_setid option was incorrect.
1929 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
1931 PH/09 Installed PCRE release 7.0.
1933 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
1934 run for batched SMTP input. It is now run at the start of every message
1935 in the batch. While fixing this I discovered that the process information
1936 (output by running exiwhat) was not always getting set for -bs and -bS
1937 input. This is fixed, and it now also says "batched" for BSMTP.
1939 PH/11 Added control=no_pipelining.
1941 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
1942 patch, slightly modified), and move the expansion of helo_data till after
1943 the connection is made in the smtp transport (so it can use these
1946 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
1948 PH/14 Added log_selector = +pid.
1950 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
1952 PH/16 Add ${if forany and ${if forall.
1954 PH/17 Added dsn_from option to vary the From: line in DSNs.
1956 PH/18 Flush SMTP output before performing a callout, unless control =
1957 no_callout_flush is set.
1959 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
1960 was true (the default) a successful delivery failed to delete the retry
1961 item, thus causing premature timeout of the address. The bug is now
1964 PH/20 Added hosts_avoid_pipelining to the smtp transport.
1966 PH/21 Long custom messages for fakedefer and fakereject are now split up
1967 into multiline reponses in the same way that messages for "deny" and
1968 other ACL rejections are.
1970 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
1971 with slight modification.
1973 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
1974 draft, changing the syntax and factoring some duplicate code".
1976 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
1977 for deliveries of the second and subsequent messages over the same SMTP
1980 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
1981 ${reduce, with only minor "tidies".
1983 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
1985 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
1986 expansion side effects.
1988 PH/27 When a message times out after an over-quota error from an Exim-imposed
1989 quota, the bounce message says "mailbox is full". This message was not
1990 being given when it was a system quota that was exceeded. It now should
1993 MH/03 Made $recipients available in local_scan(). local_scan() already has
1994 better access to the recipient list through recipients_list[], but
1995 $recipients can be useful in postmaster-provided expansion strings.
1997 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
1998 in the case of a MAIL command with additional options following the
1999 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
2000 were accidentally chopped off.
2002 PH/29 SMTP synchronization checks are implemented when a command is read -
2003 there is a check that no more input is waiting when there shouldn't be
2004 any. However, for some commands, a delay in an ACL can mean that it is
2005 some time before the response is written. In this time, more input might
2006 arrive, invalidly. So now there are extra checks after an ACL has run for
2007 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
2008 pipelining has not been advertised.
2010 PH/30 MH's patch to allow iscntrl() characters to be list separators.
2012 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
2013 returned in the SMTP response when smtp_return_error_details was false.
2014 This has been fixed.
2016 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
2017 instead of the C I/O that was originally supplied, because problems were
2018 reported on Solaris.
2020 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
2021 Exim which did not show up earlier: it was assuming that a call to
2022 SSL_CTX_set_info_callback() might give an error value. In fact, there is
2023 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
2024 was a macro that became an assignment, so it seemed to work. This has
2025 changed to a proper function call with a void return, hence the compile
2026 error. Exim's code has been fixed.
2028 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
2031 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
2033 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
2035 PH/37 If a message is not accepted after it has had an id assigned (e.g.
2036 because it turns out to be too big or there is a timeout) there is no
2037 "Completed" line in the log. When some messages of this type were
2038 selected by exigrep, they were listed as "not completed". Others were
2039 picked up by some special patterns. I have improved the selection
2040 criteria to be more general.
2042 PH/38 The host_find_failed option in the manualroute router can now be set
2043 to "ignore", to completely ignore a host whose IP address cannot be
2044 found. If all hosts are ignored, the behaviour is controlled by the new
2045 host_all_ignored option.
2047 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
2048 homing or because of multiple MX records with /mx) generated more than
2049 one IP address, and the following item turned out to be the local host,
2050 all the secondary addresses of the first item were incorrectly removed
2051 from the list, along with the local host and any following hosts (which
2052 is what is supposed to happen).
2054 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
2055 whoever called Exim into the -H file. In the case of the daemon it was
2056 behaving confusingly. When first started, it used values for whoever
2057 started the daemon, but after a SIGHUP it used the Exim user (because it
2058 calls itself on a restart). I have changed the code so that it now always
2061 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
2062 message are rejected with the same error (e.g. no authentication or bad
2063 sender address), and a DATA command is nevertheless sent (as can happen
2064 with PIPELINING or a stupid MUA), the error message that was given to the
2065 RCPT commands is included in the rejection of the DATA command. This is
2066 intended to be helpful for MUAs that show only the final error to their
2069 PH/42 Another patch from the Sieve maintainer.
2071 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
2072 Eximstats - Fixed some broken HTML links and added missing column headers
2074 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
2075 columns for Rejects, Temp Rejects, Ham, and Spam rows.
2077 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
2079 PH/43 Yet another patch from the Sieve maintainer.
2081 PH/44 I found a way to check for a TCP/IP connection going away before sending
2082 the response to the final '.' that terminates a message, but only in the
2083 case where the client has not sent further data following the '.'
2084 (unfortunately, this is allowed). However, in many cases there won't be
2085 any further data because there won't be any more messages to send. A call
2086 to select() can be used: if it shows that the input is "ready", there is
2087 either input waiting, or the socket has been closed. An attempt to read
2088 the next input character can distinguish the two cases. Previously, Exim
2089 would have sent an OK response which the client would never have see.
2090 This could lead to message repetition. This fix should cure that, at
2091 least in a lot of common cases.
2093 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
2094 advertised in response to EHLO.
2100 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
2101 fixed by 4.65/MH/01 (is this a record?) are fixed:
2103 (i) An empty string was always treated as zero by the numeric comparison
2104 operators. This behaviour has been restored.
2106 (ii) It is documented that the numeric comparison operators always treat
2107 their arguments as decimal numbers. This was broken in that numbers
2108 starting with 0 were being interpreted as octal.
2110 While fixing these problems I realized that there was another issue that
2111 hadn't been noticed. Values of message_size_limit (both the global option
2112 and the transport option) were treated as octal if they started with 0.
2113 The documentation was vague. These values are now always treated as
2114 decimal, and I will make that clear in the documentation.
2120 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
2121 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
2124 MH/01 Don't check that the operands of numeric comparison operators are
2125 integers when their expansion is in "skipping" mode (fixes bug
2126 introduced by 4.64-PH/07).
2128 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
2129 child addresses, Exim now panics and dies. Previously, because the count
2130 is held in a short int, deliveries were likely to be lost. As such a
2131 large number of recipients for a single message is ridiculous
2132 (performance will be very, very poor), I have chosen to impose a limit
2133 rather than extend the field.
2139 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
2140 leftover -K file (the existence of which was triggered by #402).
2141 While we were at it, introduced process PID as part of the -K
2142 filename. This should rule out race conditions when creating
2145 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
2146 processing considerably. Previous code took too long for large mails,
2147 triggering a timeout which in turn triggers #401.
2149 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
2150 in the DK code in transports.c. sendfile() is not really portable,
2151 hence the _LINUX specificness.
2153 TF/01 In the add_headers option to the mail command in an Exim filter,
2154 there was a bug that Exim would claim a syntax error in any
2155 header after the first one which had an odd number of characters
2158 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
2159 callout verification, Exim cached a "reject" for the entire domain. This
2160 is correct for most verifications, but it is not correct for a recipient
2161 verification with use_sender or use_postmaster set, because in that case
2162 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
2163 case of MAIL FROM:<> rejection from other early rejections (e.g.
2164 rejection of HELO). When verifying a recipient using a non-null MAIL
2165 address, the cache is ignored if it shows MAIL FROM:<> rejection.
2166 Whatever the result of the callout, the value of the domain cache is
2167 left unchanged (for any other kind of callout, getting as far as trying
2168 RCPT means that the domain itself is ok).
2170 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
2173 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
2174 manifest itself as EPIPE rather than ECONNECT. When tidying away a
2175 session, the daemon ignores ECONNECT errors and logs others; it now
2176 ignores EPIPE as well.
2178 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
2179 (quoted-printable decoding).
2181 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
2182 later the small subsequent patch to fix an introduced bug.
2184 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
2186 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
2188 PH/08 An error is now given if message_size_limit is specified negative.
2190 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
2191 to be given (somewhat) arbitrary names.
2193 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
2196 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
2197 miscellaneous code fixes
2199 PH/10 Added the log_reject_target ACL modifier to specify where to log
2202 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
2203 hostname. This is wrong, because it relates to the incoming message (and
2204 probably the interface on which it is arriving) and not to the outgoing
2205 callout (which could be using a different interface). This has been
2206 changed to use the value of the helo_data option from the smtp transport
2207 instead - this is what is used when a message is actually being sent. If
2208 there is no remote transport (possible with a router that sets up host
2209 addresses), $smtp_active_hostname is used.
2211 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
2212 tweaks were necessary in order to get it to work (see also 21 below):
2213 (a) The code assumed that strncpy() returns a negative number on buffer
2214 overflow, which isn't the case. Replaced with Exim's string_format()
2216 (b) There were several signed/unsigned issues. I just did the minimum
2217 hacking in of casts. There is scope for a larger refactoring.
2218 (c) The code used strcasecmp() which is not a standard C function.
2219 Replaced with Exim's strcmpic() function.
2220 (d) The code set only $1; it now sets $auth1 as well.
2221 (e) A simple test gave the error "authentication client didn't specify
2222 service in request". It would seem that Dovecot has changed its
2223 interface. Fortunately there's a specification; I followed it and
2224 changed what the client sends and it appears to be working now.
2226 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
2229 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
2230 address A is aliased to B and C, where B exists and C does not. Without
2231 -v the output is "A verified" because verification stops after a
2232 successful redirection if more than one address is generated. However,
2233 with -v the child addresses are also verified. Exim was outputting "A
2234 failed to verify" and then showing the successful verification for C,
2235 with its parentage. It now outputs "B failed to verify", showing B's
2236 parentage before showing the successful verification of C.
2238 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
2239 look up a TXT record in a specific list after matching in a combined
2242 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
2243 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
2244 they consult the DNS. I had assumed they would set it the way they
2245 wanted; and indeed my experiments on Linux seem to show that in some
2246 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
2247 To be on the safe side, however, I have now made the interface to
2248 host_find_byname() similar to host_find_bydns(), with an argument
2249 containing the DNS resolver options. The host_find_byname() function now
2250 sets these options at its start, just as host_find_bydns() does. The smtp
2251 transport options dns_qualify_single and dns_search_parents are passed to
2252 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
2253 of host_find_byname() use the default settings of RES_DEFNAMES
2254 (qualify_single) but not RES_DNSRCH (search_parents).
2256 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
2257 spool_read_header() do less string testing, by means of a preliminary
2258 switch on the second character of optional "-foo" lines. (This is
2259 overdue, caused by the large number of possibilities that now exist.
2260 Originally there were few.) While I was there, I also converted the
2261 str(n)cmp tests so they don't re-test the leading "-" and the first
2262 character, in the hope this might squeeze out yet more improvement.
2264 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
2265 flag allowing group syntax was set by the header_syntax check but not
2266 turned off, possible causing trouble later; (2) The flag was not being
2267 set at all for the header_verify test, causing "group"-style headers to
2268 be rejected. I have now set it in this case, and also caused header_
2269 verify to ignore an empty address taken from a group. While doing this, I
2270 came across some other cases where the code for allowing group syntax
2271 while scanning a header line wasn't quite right (mostly, not resetting
2272 the flag correctly in the right place). These bugs could have caused
2273 trouble for malformed header lines. I hope it is now all correct.
2275 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
2276 with the "reply" argument non-NULL. The code, however (which originally
2277 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
2278 but it didn't always do it. This confused somebody who was copying the
2279 code for some other use. I have removed all the tests.
2281 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
2282 feature that was used to support insecure browsers during the U.S. crypto
2283 embargo. It requires special client support, and Exim is probably the
2284 only MTA that supported it -- and would never use it because real RSA is
2285 always available. This code has been removed, because it had the bad
2286 effect of slowing Exim down by computing (never used) parameters for the
2287 RSA_EXPORT functionality.
2289 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
2290 authenticator to fail if there's a tab character in the incoming data
2291 (there should never be unless someone is messing about, as it's supposed
2292 to be base64-encoded). Also added, on Timo's advice, the "secured" option
2293 if the connection is using TLS or if the remote IP is the same as the
2294 local IP, and the "valid-client-cert option" if a client certificate has
2297 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
2298 authenticators. This can be used for authorization after authentication
2299 succeeds. (In the case of plaintext, it servers for both authentication
2302 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
2303 if any retry times were supplied.
2305 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
2306 connection, where there is no client IP address to check. In this
2307 situation, the verify now always succeeds.
2309 PH/25 Applied John Jetmore's -Mset patch.
2311 PH/26 Added -bem to be like -Mset, but loading a message from a file.
2313 PH/27 In a string expansion for a processed (not raw) header when multiple
2314 headers of the same name were present, leading whitespace was being
2315 removed from all of them, but trailing whitespace was being removed only
2316 from the last one. Now trailing whitespace is removed from each header
2317 before concatenation. Completely empty headers in a concatenation (as
2318 before) are ignored.
2320 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
2321 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
2323 PH/29 [Removed. This was a change that I later backed out, and forgot to
2324 correct the ChangeLog entry (that I had efficiently created) before
2325 committing the later change.]
2327 PH/30 Exim was sometimes attempting to deliver messages that had suffered
2328 address errors (4xx response to RCPT) over the same connection as other
2329 messages routed to the same hosts. Such deliveries are always "forced",
2330 so retry times are not inspected. This resulted in far too many retries
2331 for the affected addresses. The effect occurred only when there were more
2332 hosts than the hosts_max_try setting in the smtp transport when it had
2333 the 4xx errors. Those hosts that it had tried were not added to the list
2334 of hosts for which the message was waiting, so if all were tried, there
2335 was no problem. Two fixes have been applied:
2337 (i) If there are any address or message errors in an SMTP delivery, none
2338 of the hosts (tried or untried) are now added to the list of hosts
2339 for which the message is waiting, so the message should not be a
2340 candidate for sending over the same connection that was used for a
2341 successful delivery of some other message. This seems entirely
2342 reasonable: after all the message is NOT "waiting for some host".
2343 This is so "obvious" that I'm not sure why it wasn't done
2344 previously. Hope I haven't missed anything, but it can't do any
2345 harm, as the worst effect is to miss an optimization.
2347 (ii) If, despite (i), such a delivery is accidentally attempted, the
2348 routing retry time is respected, so at least it doesn't keep
2349 hammering the server.
2351 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
2352 in ${readsocket because some servers need this prod.
2354 PH/32 Added some extra debug output when updating a wait-xxx database.
2356 PH/33 The hint "could be header name not terminated by colon", which has been
2357 given for certain expansion errors for a long time, was not being given
2358 for the ${if def:h_colon_omitted{... case.
2360 PH/34 The spec says: "With one important exception, whenever a domain list is
2361 being scanned, $domain contains the subject domain." There was at least
2362 one case where this was not true.
2364 PH/35 The error "getsockname() failed: connection reset by peer" was being
2365 written to the panic log as well as the main log, but it isn't really
2366 panic-worthy as it just means the connection died rather early on. I have
2367 removed the panic log writing for the ECONNRESET error when getsockname()
2370 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
2371 runs only) independently of the message's sender address. This meant
2372 that, if the 4xx error was in fact related to the sender, a different
2373 message to the same recipient with a different sender could confuse
2374 things. In particualar, this can happen when sending to a greylisting
2375 server, but other circumstances could also provoke similar problems.
2376 I have changed the default so that the retry time for these errors is now
2377 based a combination of the sender and recipient addresses. This change
2378 can be overridden by setting address_retry_include_sender=false in the
2381 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
2382 remote server are returned as part of bounce messages. This was not
2383 happening for LMTP over a pipe (the lmtp transport), but now it is the
2384 same for both kinds of LMTP.
2386 PH/38 Despite being documented as not happening, Exim was rewriting addresses
2387 in header lines that were in fact CNAMEs. This is no longer the case.
2389 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
2390 and queue runs started by the daemon processed all messages. This has
2391 been fixed so that -R and -S can now usefully be given with -q<time>.
2393 PH/40 Import PCRE release 6.7 (fixes some bugs).
2395 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
2397 PH/42 Give an error if -q is specified more than once.
2399 PH/43 Renamed the variables $interface_address and $interface_port as
2400 $received_ip_address and $received_port, to make it clear that these
2401 values apply to message reception, and not to the outgoing interface when
2402 a message is delivered. (The old names remain recognized, of course.)
2404 PH/44 There was no timeout on the connect() call when using a Unix domain
2405 socket in the ${readsocket expansion. There now is.
2407 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
2408 be meaningful with "accept".
2410 SC/01 Eximstats V1.43
2411 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
2413 SC/02 Eximstats V1.44
2414 Use a glob alias rather than an array ref in the generated
2415 parser. This improves both readability and performance.
2417 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
2418 Collect SpamAssassin and rejection statistics.
2419 Don't display local sender or destination tables unless
2420 there is data to show.
2421 Added average volumes into the top table text output.
2423 SC/04 Eximstats V1.46
2424 Collect data on the number of addresses (recipients)
2425 as well as the number of messages.
2427 SC/05 Eximstats V1.47
2428 Added 'Message too big' to the list of mail rejection
2429 reasons (thanks to Marco Gaiarin).
2431 SC/06 Eximstats V1.48
2432 Mainlog lines which have GMT offsets and are too short to
2433 have a flag are now skipped.
2435 SC/07 Eximstats V1.49 (Alain Williams)
2436 Added the -emptyok flag.
2438 SC/08 Eximstats V1.50
2439 Fixes for obtaining the IP address from reject messages.
2441 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
2442 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
2443 whitesspace changes from 4.64-PH/27
2445 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
2448 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
2449 are found, allow negative numbers in numeric criteria)
2451 JJ/06 exipick.20061117.2, added new $message_body_missing variable
2453 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
2454 to match changes made in 4.64-PH/43
2456 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
2458 PH/47 Put in an explicit test for a DNS lookup of an address record where the
2459 "domain" is actually an IP address, and force a failure. This locks out
2460 those revolvers/nameservers that support "A-for-A" lookups, in
2461 contravention of the specifications.
2463 PH/48 When a host name was looked up from an IP address, and the subsequent
2464 forward lookup of the name timed out, the host name was left in
2465 $sender_host_name, contrary to the specification.
2467 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
2468 restricted to single-key lookups, Exim was not diagnosing an error if
2469 * or *@ was used with a query-style lookup.
2471 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
2473 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
2474 long ago, but noone interested enough thought of it. Let's just say that
2475 the "1.1" means that there are some new functions that weren't there at
2476 some point in the past.
2478 PH/51 Error processing for expansion failure of helo_data from an smtp
2479 transport during callout processing was broken.
2481 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
2482 tested/used via the -bh/-bhc/-bs options.
2484 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
2485 bug, fixed in subsequent PCRE releases).
2487 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
2488 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
2490 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
2496 SC/01 Use a glob alias rather than an array ref in eximstats generated
2497 parser. This improves both readability and performance.
2499 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
2500 Don't display local sender or destination tables in eximstats unless
2501 there is data to show.
2502 Added average volumes into the eximstats top table text output.
2504 SC/03 Collect data on the number of addresses (recipients) as well
2505 as the number of messages in eximstats.
2507 TF/01 Correct an error in the documentation for the redirect router. Exim
2508 does (usually) call initgroups() when daemonizing.
2510 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
2511 with consistent privilege compared to when running as a daemon.
2513 TF/03 Note in the spec that $authenticated_id is not set for local
2514 submissions from trusted users.
2516 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
2517 Thanks to Dean Brooks <dean@iglou.com> for the patch.
2519 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
2520 by adding some example configuration directives to the default
2521 configuration file. A little bit of work is required to uncomment the
2522 directives and define how usernames and passwords are checked, but
2523 there is now a framework to start from.
2525 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
2526 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
2527 without this. I don't know how relevant this is to other LDAP libraries.
2529 PH/02 Add the verb name to the "unknown ACL verb" error.
2531 PH/03 Magnus Holmgren's patch for filter_prepend_home.
2533 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
2535 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
2536 directory not expanded when it should be if an expanded home directory
2537 was set for the address (which is overridden by the transport).
2539 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
2542 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
2543 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
2544 because it is too late at that time, and has no effect.
2546 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
2547 security issue with \' (bugzilla #107). I could not use the
2548 PQescapeStringConn() function, because it needs a PGconn value as one of
2551 PH/08 When testing addresses using -bt, indicate those final addresses that
2552 are duplicates that would not cause an additional delivery. At least one
2553 person was confused, thinking that -bt output corresponded to deliveries.
2554 (Suppressing duplicates isn't a good idea as you lose the information
2555 about possibly different redirections that led to the duplicates.)
2557 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
2558 systems where poll() doesn't work, in particular OS X.
2560 PH/10 Added more information to debugging output for retry time not reached.
2562 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
2563 operations in malware.c.
2565 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
2568 PH/13 If write_rejectlog was set false when logging was sent to syslog with
2569 syslog_duplication set false, log lines that would normally be written
2570 both the the main log and to the reject log were not written to syslog at
2573 PH/14 In the default configuration, change the use of "message" in ACL warn
2574 statements to "add_header".
2576 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
2577 not followed by a command (e.g. "seen endif").
2579 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
2580 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
2583 PH/17 Added extra conditions to the default value of delay_warning_condition
2587 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
2588 { match{$h_precedence:}{(?i)bulk|list|junk} } \
2589 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
2592 The Auto-Submitted: and various List- headers are standardised, whereas I
2593 don't think Precedence: ever was.
2595 PH/18 Refactored debugging code in route_finduser() to show more information,
2596 in particular, the error code if getpwnam() issues one.
2598 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
2599 This is apparently needed in addition to the PH/07 change above to avoid
2600 any possible encoding problems.
2602 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
2603 but not after initializing Perl.
2605 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
2606 output them only if debugging. By default they are written stderr,
2607 apparently, which is not desirable.
2609 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
2612 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
2615 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
2617 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
2618 authenticated or an ident call has been made. Suppress the default
2619 values for $authenticated_id and $authenticated_sender (but permit -oMai
2620 and -oMas) when testing with -bh.
2622 PH/24 Re-jigged the order of the tests in the default configuration so that the
2623 tests for valid domains and recipients precede the DNS black list and CSA
2624 tests, on the grounds that those ones are more expensive.
2626 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
2627 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
2628 This bug exists in every version of Exim that I still have, right back to
2631 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
2632 However, an attempt to turn on case-sensitivity in a regex key by
2633 including (?-i) didn't work because the subject string was already
2634 lowercased, and the effects were non-intuitive. It turns out that a
2635 one-line patch can be used to allow (?-i) to work as expected.
2641 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
2642 other effects) broke the use of negated acl sub-conditions.
2644 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
2647 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
2648 "Deny" causes Exim to reject the incoming connection with a 554 error.
2649 Unfortunately, if there is a major crisis, such as a disk failure,
2650 tcp-wrappers gives "deny", whereas what one would like would be some
2651 kind of temporary error. A kludge has been added to help with this.
2652 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
2653 554 error is used if errno is still zero or contains ENOENT (which occurs
2654 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
2657 PH/03 Add -lutil to the default FreeBSD LIBS setting.
2659 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
2660 errors. Otherwise a message that provokes a temporary error (when other
2661 messages do not) can cause a whole host to time out.
2663 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
2664 addresses were routed directly to files or pipes from a redirect router.
2665 File deliveries just didn't batch; pipe deliveries might have suffered
2668 PH/06 A failure to get a lock for a hints database would erroneously always say
2669 "Failed to get write lock", even when it was really a read lock.
2671 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
2672 of 0600. This has been changed to use the value of the lockfile_mode
2673 option (which defaults to 0600).
2675 PH/08 Applied small patch from the Sieve maintainer.
2677 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
2678 folder from quota calculations, a direct delivery into this folder messed
2679 up the contents of the maildirsize file. This was because the regex was
2680 used only to exclude .Trash (or whatever) when the size of the mailbox
2681 was calculated. There was no check that a delivery was happening into an
2682 excluded directory. This bug has been fixed by ignoring all quota
2683 processing for deliveries into excluded directories.
2685 PH/10 Added the maildirfolder_create_regex option to appendfile.
2691 PH/01 The code for finding all the local interface addresses on a FreeBSD
2692 system running IPv6 was broken. This may well have applied to all BSD
2693 systems, as well as to others that have similar system calls. The broken
2694 code found IPv4 interfaces correctly, but gave incorrect values for the
2695 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
2696 that it would not match correctly against @[] and not recognize the IPv6
2699 PH/02 The ipliteral router was not recognizing addresses of the form user@
2700 [ipv6:....] because it didn't know about the "ipv6:" prefix.
2702 PH/03 Added disable_ipv6.
2704 PH/04 Changed $reply_address to use the raw form of the headers instead of the
2705 decoded form, because it is most often used to construct To: headers
2706 lines in autoreplies, and the decoded form may well be syntactically
2707 invalid. However, $reply_address has leading white space removed, and all
2708 newlines turned into spaces so that the autoreply transport does not
2711 PH/05 If group was specified without a user on a router, and no group or user
2712 was specified on a transport, the group from the router was ignored.
2714 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
2715 for visible compile-time settings that can be used to change these
2716 numbers, for those that want even more. Backwards compatibility with old
2717 spool files has been maintained. However, going back to a previous Exim
2718 release will lost any variables that are in spool files.
2720 PH/07 Two small changes when running in the test harness: increase delay when
2721 passing a TCP/IP connection to a new process, in case the original
2722 process has to generate a bounce, and remove special handling of
2723 127.0.0.2 (sic), which is no longer necessary.
2725 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
2726 be the same on different OS.
2728 PH/09 Moved a debug statement in filter processing to avoid a race problem when
2731 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
2732 whether --show-vars was specified or not
2734 JJ/02 exipick: Added support for new ACL variable spool format introduced
2737 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
2738 syntactically invalid From: or Reply-to: line, and a filter used this to
2739 generate an autoreply, and therefore failed to obtain an address for the
2740 autoreply, Exim could try to deliver to a non-existent relative file
2741 name, causing unrelated and misleading errors. What now happens is that
2742 it logs this as a hard delivery error, but does not attempt to create a
2745 PH/11 The exinext utility has a -C option for testing purposes, but although
2746 the given file was scanned by exinext itself; it wasn't being passed on
2747 when Exim was called.
2749 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
2750 an end-of-file indication when reading a command response.
2752 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
2753 compiled. In many other places in Exim, IPv6 addresses are always
2754 recognized, so I have changed this. It also means that IPv4 domain
2755 literals of the form [IPV4:n.n.n.n] are now always recognized.
2757 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
2758 used if the router is not running as root, for example, when verifying at
2759 ACL time, or when using -bh. The debugging output from this situation was
2760 non-existent - all you got was a failure to exec. I have made two
2763 (a) Failures to set uid/gid, the current directory, or a process leader
2764 in a subprocess such as that created by queryprogram now generate
2765 suitable debugging ouput when -d is set.
2767 (b) The queryprogram router detects when it is not running as root,
2768 outputs suitable debugging information if -d is set, and then runs
2769 the subprocess without attempting to change uid/gid.
2771 PH/15 Minor change to Makefile for building test_host (undocumented testing
2774 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
2775 additional section of a DNS packet that returns MX or SRV records.
2776 Instead, it always explicitly searches for A/AAAA records. This avoids
2777 major problems that occur when a DNS server includes only records of one
2778 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
2779 fixed another bug: if SRV records were looked up and the corresponding
2780 address records were *not* found in the additional section, the port
2781 values from the SRV records were lost.
2783 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
2784 using the correct key (the original address) when searching the retry
2785 rules in order to find which one to use for generating the retry hint.
2787 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
2788 adding the default one. Similarly, if it contains a Reply-To: header, the
2789 errors_reply_to option, if set, is not used.
2791 PH/19 When calculating a retry time, Exim used to measure the "time since
2792 failure" by looking at the "first failed" field in the retry record. Now
2793 it does not use this if it is later than than the arrival time of the
2794 message. Instead it uses the arrival time. This makes for better
2795 behaviour in cases where some deliveries succeed, thus re-setting the
2796 "first failed" field. An example is a quota failure for a huge message
2797 when small messages continue to be delivered. Without this change, the
2798 "time since failure" will always be short, possible causing more frequent
2799 delivery attempts for the huge message than are intended.
2800 [Note: This change was subsequently modified - see PH/04 for 4.62.]
2802 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
2803 $1, $2, $3) because the numerical variables can be reset during some
2804 expansion items (e.g. "match"), thereby losing the authentication data.
2806 PH/21 Make -bV show the size of off_t variables so that the test suite can
2807 decide whether to run tests for quotas > 2G.
2809 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
2810 mailbox_size, and mailbox_filecount in the appendfile transport. If a
2811 filecount value is greater than 2G or if a quota value is greater than 2G
2812 on a system where the size of off_t is not greater than 4, a panic error
2815 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
2816 never match. The debug and -bh output now contains an explicit error
2817 message indicating a malformed IPv4 address or mask.
2819 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
2820 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
2821 PH/23 above applies.
2823 PH/25 Do not write to syslog when running in the test harness. The only
2824 occasion when this arises is a failure to open the main or panic logs
2825 (for which there is an explicit test).
2827 PH/26 Added the /no_tell option to "control=freeze".
2829 PH/27 If a host name lookup failed very early in a connection, for example, if
2830 the IP address matched host_lookup and the reverse lookup yielded a name
2831 that did not have a forward lookup, an error message of the form "no IP
2832 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
2833 could be logged. Now it outputs the IP address instead of "NULL".
2835 PH/28 An enabling patch from MH: add new function child_open_exim2() which
2836 allows the sender and the authenticated sender to be set when
2837 submitting a message from within Exim. Since child_open_exim() is
2838 documented for local_scan(), the new function should be too.
2840 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
2841 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
2842 results in an empty string is now treated as unset.
2844 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
2846 PH/31 Added sender_verify_fail logging option.
2848 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
2849 needed by RFC 822 but not by RFC 2822 was commented out. I have now
2850 tidied the source and removed it altogether.
2852 PH/33 When a queue run was abandoned because the load average was too high, a
2853 log line was always written; now it is written only if the queue_run log
2854 selector is set. In addition, the log line for abandonment now contains
2855 information about the queue run such as the pid. This is always present
2856 in "start" and "stop" lines but was omitted from the "abandon" line.
2858 PH/34 Omit spaces between a header name and the colon in the error message that
2859 is given when verify = headers_syntax fails (if there are lots of them,
2860 the message gets confusing).
2862 PH/35 Change the default for dns_check_names_pattern to allow slashes within
2863 names, as there are now some PTR records that contain slashes. This check
2864 is only to protect against broken name servers that fall over on strange
2865 characters, so the fact that it applies to all lookups doesn't matter.
2867 PH/36 Now that the new test suite is complete, we can remove some of the
2868 special code in Exim that was needed for the old test suite. For example,
2869 sorting DNS records because real resolvers return them in an arbitrary
2870 order. The new test suite's fake resolver always returns records in the
2873 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
2874 bounces) except when queue_only is set, to avoid logging races between
2875 the different processes.
2877 PH/38 Panic-die if .include specifies a non-absolute path.
2879 PH/39 A tweak to the "H" retry rule from its user.
2881 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
2882 a label. They prevented compilation on older perls.
2884 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
2885 a warning to be raised on newish perls.
2887 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
2888 on queue. Changes to match documented behaviour of showing count of
2889 messages matching specified criteria.
2891 PH/40 Changed the default ident timeout from 30s to 5s.
2893 PH/41 Added support for the use of login_cap features, on those BSD systems
2894 that have them, for controlling the resources used by pipe deliveries.
2896 PH/42 The content-scanning code uses fopen() to create files in which to put
2897 message data. Previously it was not paying any attention to the mode of
2898 the files. Exim runs with umask(0) because the rest of the code creates
2899 files with open(), and sets the required mode explicitly. Thus, these
2900 files were ending up world-writeable. This was not a big issue, because,
2901 being within the spool directory, they were not world-accessible. I have
2902 created a function called modefopen, which takes an additional mode
2903 argument. It sets umask(777), creates the file, chmods it to the required
2904 mode, then resets the umask. All the relevant calls to fopen() in the
2905 content scanning code have been changed to use this function.
2907 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
2908 to 24 hours. This avoids potential overflow problems when processing G
2909 and H retry rules. I suspect nobody ever tinkers with this value.
2911 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
2913 PH/45 When the plaintext authenticator is running as a client, the server's
2914 challenges are checked to ensure they are valid base64 strings. By
2915 default, the authentication attempt is cancelled if an invalid string is
2916 received. Setting client_ignore_invalid_base64 true ignores these errors.
2917 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
2918 they are received. Thus, the responses can be made to depend on the
2919 challenges. If an invalid string is ignored, an empty string is placed in
2922 PH/46 Messages that are created by the autoreply transport now contains a
2923 References: header, in accordance with RFCs 2822 and 3834.
2925 PH/47 Added authenticated_sender_force to the smtp transport.
2927 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
2929 PH/49 Installed latest patch from the Sieve maintainer.
2931 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
2932 was also set, the appendfile transport was unnecessarily scanning a
2933 directory of message files (e.g. for maildir delivery) to find the count
2934 of files (along with the size), even though it did not need this
2935 information. It now does the scan only if it needs to find either the
2936 size of the count of files.
2938 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
2940 PH/52 Two bugs concerned with error handling when the smtp transport is
2943 (i) Exim was not creating retry information for temporary errors given
2944 for individual recipients after the DATA command when the smtp transport
2945 was used in LMTP mode. This meant that they could be retried too
2946 frequently, and not timed out correctly.
2948 (ii) Exim was setting the flag that allows error details to be returned
2949 for LMTP errors on RCPT commands, but not for LMTP errors for individual
2950 recipients that were returned after the DATA command.
2952 PH/53 This is related to PH/52, but is more general: for any failing address,
2953 when detailed error information was permitted to be returned to the
2954 sender, but the error was temporary, then after the final timeout, only
2955 "retry timeout exceeded" was returned. Now it returns the full error as
2956 well as "retry timeout exceeded".
2958 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
2959 do this, and (what is worse) MTAs that accept it.
2961 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
2962 will now be deprecated.
2964 PH/56 New os.c-cygwin from the Cygwin maintainer.
2966 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
2967 formats (previously only available in exim formats via -bpr, -bpru,
2968 and -bpra. Now also available in native and exiqgrep formats)
2970 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
2971 with very large, slow to parse queues
2973 JJ/08 exipick: added ! as generic prefix to negate any criteria format
2975 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
2977 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
2978 responses to authentication challenges, though it was showing the
2979 challenges; (ii) I've removed the CR characters from the debug output for
2982 PH/58 Allow for the insertion of a newline as well as a space when a string
2983 is turned into more than one encoded-word during RFC 2047 encoding. The
2984 Sieve code now uses this.
2986 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
2987 data_4xx, lost_connection, tls_required.
2989 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
2990 message was being sent as an SMTP response.
2992 PH/61 Add -l and -k options to exicyclog.
2994 PH/62 When verifying, if an address was redirected to one new address, so that
2995 verification continued, and the new address failed or deferred after
2996 having set something in $address_data, the value of $address_data was not
2997 passed back to the ACL. This was different to the case when no
2998 redirection occurred. The value is now passed back in both cases.
3000 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
3001 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
3002 use login_cap.h, so on its own it isn't the distinguishing feature. The
3003 new name refers directly to the setclassresources() function.
3005 PH/65 Added configuration files for NetBSD3.
3007 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
3009 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
3010 is preferred over IPv4.
3012 PH/68 The bounce_return_message and bounce_return_body options were not being
3013 honoured for bounces generated during the reception of non-SMTP messages.
3014 In particular, this applied to messages rejected by the ACL. This bug has
3015 been fixed. However, if bounce_return_message is true and bounce_return_
3016 body is false, the headers that are returned for a non-SMTP message
3017 include only those that have been read before the error was detected.
3018 (In the case of an ACL rejection, they have all been read.)
3020 PH/69 The HTML version of the specification is now built in a directory called
3021 spec_html instead of spec.html, because the latter looks like a path with
3022 a MIME-type, and this confuses some software.
3024 PH/70 Catch two compiler warnings in sieve.c.
3026 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
3027 function verify_get_ident() calls ip_connect() to connect a socket, but
3028 if the "connect()" function timed out, ip_connect() used to close the
3029 socket. However, verify_get_ident() also closes the socket later, and in
3030 between Exim writes to the log, which may get opened at this point. When
3031 the socket was closed in ip_connect(), the log could get the same file
3032 descriptor number as the socket. This naturally causes chaos. The fix is
3033 not to close the socket in ip_connect(); the socket should be closed by
3034 the function that creates it. There was only one place in the code where
3035 this was missing, in the iplookup router, which I don't think anybody now
3036 uses, but I've fixed it anyway.
3038 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
3039 well as to direct DNS lookups. Otherwise the handling of names in host
3040 lists is inconsistent and therefore confusing.
3046 PH/01 Two changes to the default runtime configuration:
3048 (1) Move the checks for relay_from_hosts and authenticated clients from
3049 after to before the (commented out) DNS black list checks.
3051 (2) Add control=submission to the relay_from_hosts and authenticated
3052 clients checks, on the grounds that messages accepted by these
3053 statements are most likely to be submissions.
3055 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
3057 (1) Generate an error if the third argument for the ${prvs expansion is
3060 (2) Treat a missing third argument of ${prvscheck as if it were an empty
3063 (3) Reset the variables that are obtained from the first argument of
3064 ${prvscheck and used in the second argument before leaving the code,
3065 because their memory is reclaimed, so using them afterwards may do
3068 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
3069 one (it's much easier than Tom thought :-).
3071 (5) Because of (4), we can now allow for the use of $prvscheck_result
3072 inside the third argument.
3074 PH/03 For some reason, the default setting of PATH when running a command from
3075 a pipe transport was just "/usr/bin". I have changed it to
3078 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
3079 anything to be listed in the output from -bV.
3081 PH/05 When a filter generated an autoreply, the entire To: header line was
3082 quoted in the delivery log line, like this:
3084 => >A.N.Other <ano@some.domain> <original@ddress> ...
3086 This has been changed so that it extracts the operative address. There
3087 may be more than one such address. If so, they are comma-separated, like
3090 => >ano@some.domain,ona@other.domain <original@ddress> ...
3092 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
3093 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
3094 not being looked up in the rDNS to get a host name, Exim was showing the
3095 IP address twice in Received: lines, even though the IP addresses were
3096 identical. For example:
3098 Received: from [1.2.3.4] (helo=[1.2.3.4])
3100 However, if the real host name was known, it was omitting the HELO data
3101 if it matched the actual IP address. This has been tidied up so that it
3102 doesn't show the same IP address twice.
3104 PH/07 When both +timestamp and +memory debugging was on, the value given by
3105 $tod_xxx expansions could be wrong, because the tod_stamp() function was
3106 called by the debug printing, thereby overwriting the timestamp buffer.
3107 Debugging no longer uses the tod_stamp() function when +timestamp is set.
3109 PH/08 When the original message was included in an autoreply transport, it
3110 always said "this is a copy of the message, including all the headers",
3111 even if body_only or headers_only was set. It now gives an appropriate
3114 PH/09 Applied a patch from the Sieve maintainer which:
3116 o fixes some comments
3117 o adds the (disabled) notify extension core
3118 o adds some debug output for the result of if/elsif tests
3119 o points to the current vacation draft in the documentation
3120 and documents the missing references header update
3124 o fixes a bug in processing the envelope test (when testing
3125 multiple envelope elements, the last element determinted the
3128 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
3129 Electronic Mail") by including:
3131 Auto-submitted: auto-generated
3133 in the messages that it generates (bounce messages and others, such as
3134 warnings). In the case of bounce messages for non-SMTP mesages, there was
3135 also a typo: it was using "Auto_submitted" (underscore instead of
3136 hyphen). Since every message generated by Exim is necessarily in response
3137 to another message, thes have all been changed to:
3139 Auto-Submitted: auto-replied
3141 in accordance with these statements in the RFC:
3143 The auto-replied keyword:
3145 - SHOULD be used on messages sent in direct response to another
3146 message by an automatic process,
3148 - MUST NOT be used on manually-generated messages,
3150 - MAY be used on Delivery Status Notifications (DSNs) and Message
3151 Disposition Notifications (MDNs),
3153 - MUST NOT be used on messages generated by automatic or periodic
3154 processes, except for messages which are automatic responses to
3157 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
3158 to the default Received: header definition.
3160 PH/12 Added log selector acl_warn_skipped (default on).
3162 PH/13 After a successful wildlsearch lookup, discard the values of numeric
3163 variables because (a) they are in the wrong storage pool and (b) even if
3164 they were copied, it wouldn't work properly because of the caching.
3166 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
3167 checking when decoding. Apparently there are clients that generate
3168 overlong encoded strings. Why am I not surprised?
3170 PH/15 If the first argument of "${if match_address" was not empty, but did not
3171 contain an "@" character, Exim crashed. Now it writes a panic log message
3172 and treats the condition as false.
3174 PH/16 In autoreply, treat an empty string for "once" the same as unset.
3176 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
3177 extension "envelope-auth". The code is finished and in agreement with
3178 other implementations, but there is no documentation so far and in fact,
3179 nobody wrote the draft yet. This extension is currently #undef'ed, thus
3180 not changing the active code.
3182 Print executed "if" and "elsif" statements when debugging is used. This
3183 helps a great deal to understand what a filter does.
3185 Document more things not specified clearly in RFC3028. I had all this
3186 sorted out, when out of a sudden new issues came to my mind. Oops."
3188 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
3191 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
3192 canonical form (as documented). However, after a host name lookup from
3193 the IP address, check_host() was doing a simple string comparison with
3194 addresses acquired from the DNS when checking that the found name did
3195 have the original IP as one of its addresses. Since any found IPv6
3196 addresses are likely to be in abbreviated form, the comparison could
3197 fail. Luckily, there already exists a function for doing the comparison
3198 by converting both addresses to binary, so now that is used instead of
3199 the text comparison.
3201 PH/20 There was another similar case to PH/19, when a complete host name was
3202 given in a host list; looking up its IP address could give an abbreviated
3203 form, whereas the current host's name might or might not be abbreviated.
3204 The same fix has been applied.
3210 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
3211 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
3214 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
3215 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
3217 PH/03 Typo: missing ".o" in src/pcre/Makefile.
3219 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
3220 header line, restrict the check to what is listed in RFCs 2369 and 2929.
3221 Also, for "Auto-Submitted", treat anything other than "no" as
3222 non-personal, in accordance with RFC 3834. (Previously it treated
3223 anything starting "auto-" as non-personal.)
3225 TF/01 The control=submission/name=... option had a problem with syntax
3226 errors if the name included a slash character. The /name= option
3227 now slurps the rest of the string, so it can include any characters
3228 but it must come last in the list of options (after /sender_retain
3231 PH/05 Some modifications to the interface to the fake nameserver for the new
3239 TK/01 Added the "success_on_redirect" address verification option. See
3240 NewStuff for rationale and an example.
3242 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
3244 PH/02 Patch to exigrep to allow it to work on syslog lines.
3246 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
3247 fread() to skip over the body file's header line, because in Cygwin the
3248 header line is locked and is inaccessible.
3250 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
3251 co-exist for some time) to make it clear that it is the Exim ID that is
3252 referenced, not the Message-ID: header line.
3254 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
3255 string_format() function, because snprintf() does not exist on all
3258 PH/06 The use of forbid_filter_existstest now also locks out the use of the
3259 ${stat: expansion item.
3261 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
3262 protocol synchronization error", to keep the pedants happy.
3264 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
3265 well as for IRIX systems, when gcc is being used. See the host.c source
3268 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
3270 PH/10 Named domain lists were not working if used in a queue_smtp_domains
3273 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
3274 transport and to the smtp transport in LMTP mode.
3276 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
3278 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
3279 run a filter in a subprocess. This could lead to confusion in subsequent
3280 lookups in the parent process. There should also be a search_tidyup() at
3281 the end of the subprocess.
3283 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
3284 only if the host matched helo_try_verify_hosts, which caused the
3285 verification to occur when the EHLO/HELO command was issued. The ACL just
3286 tested the remembered result. Now, if a previous verification attempt has
3287 not happened, "verify = helo" does it there and then.
3289 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
3291 TK/03 Fix log output including CR from clamd.
3293 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
3294 exist provoked a memory error which could cause a segfault.
3296 PH/15 Installed PCRE 6.2
3298 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
3300 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
3301 of the problem. Specifically, suggested +O2 rather than +O1 for the
3304 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
3306 PH/20 If a delivery was routed to a non-standard port by means of an SRV
3307 record, the port was not correctly logged when the outgoing_port log
3308 selector was set (it logged the transort's default port).
3310 PH/21 Added support for host-specific ports to manualroute, queryprogram,
3311 fallback_hosts, and "hosts" in the smtp transport.
3313 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
3314 host errors such as "Connection refused".
3316 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
3317 authentication with radiusclient 0.4.9:
3319 - Error returned from rc_read_config was caught wrongly
3320 - Username/password not passed on to radius server due to wrong length.
3322 The presumption is that some radiusclient API changes for 4.51/PH/17
3323 were not taken care of correctly. The code is still untested by me (my
3324 Linux distribution still has 0.3.2 of radiusclient), but it was
3325 contributed by a Radius user.
3327 PH/24 When doing a callout, the value of $domain wasn't set correctly when
3328 expanding the "port" option of the smtp transport.
3330 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
3331 while reading a MIME header. Thanks to Tom Hughes for a patch.
3333 PH/24 Include config.h inside local_scan.h so that configuration settings are
3336 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
3337 that in an ACL for RCPT (for example), you can examine exactly what was
3340 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
3341 commands, but it was not correctly comparing the address with the actual
3342 client host address. Thus, it would show the EHLO address in Received:
3343 header lines when this was not necessary.
3345 PH/27 Added the % operator to ${eval:}.
3347 PH/28 Exim tries to create and chdir to its spool directory when it starts;
3348 it should be ignoring failures (because with -C, for example, it has lost
3349 privilege). It wasn't ignoring creation failures other than "already
3352 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
3355 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
3356 because an input line was too long, either on its own, or by virtue of
3357 too many continuations, the temporary file was not being removed, and the
3358 return code was incorrect.
3360 PH/31 Missing "BOOL" in function definition in filtertest.c.
3362 PH/32 Applied Sieve patches from the maintainer.
3364 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
3366 PH/33 Added "verify = not_blind".
3368 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
3369 Local/Makefile (with some defaults set). These are used in built scripts
3370 such as exicyclog, but they have never been used in the exim_install
3371 script (though there are many overriding facilities there). I have
3372 arranged that the exim_install script now takes note of these two
3375 PH/35 Installed configuration files for Dragonfly.
3377 PH/36 When a locally submitted message by a trusted user did not contain a
3378 From: header, and the sender address was obtained from -f or from an SMTP
3379 MAIL command, and the trusted user did not use -F to supply a sender
3380 name, $originator_name was incorrectly used when constructing a From:
3381 header. Furthermore, $originator_name was used for submission mode
3382 messages from external hosts without From: headers in a similar way,
3383 which is clearly wrong.
3385 PH/37 Added control=suppress_local_fixups.
3387 PH/38 When log_selector = +received_sender was set, and the addition of the
3388 sender made the log line's construction buffer exactly full, or one byte
3389 less than full, an overflow happened when the terminating "\n" was
3392 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
3393 when the result of a list match is failure because a DNS lookup failed.
3395 PH/40 RM_COMMAND is now used in the building process.
3397 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
3398 the "build-* directories that it finds.
3400 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
3401 address in a domain literal was a prefix of an interface address.
3403 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
3404 when verifying a sender address, unless rewrite_headers is false.
3406 PH/44 Wrote a long comment about why errors_to addresses are verified as
3407 recipients, not senders.
3409 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
3410 the ratelimit ACL was added.
3412 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
3414 PH/46 Added extra information about PostgreSQL errors to the error string.
3416 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
3417 avoiding the need to install special zones in a real server. This is
3418 backwards compatible; if it can't find the fake resolver, it drops back.
3419 Thus, both old and new test suites can be run.
3421 TF/02 Added util/ratelimit.pl
3423 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
3424 clock is set back in time.
3426 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
3427 Candler <B.Candler@pobox.com>.
3429 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
3430 OK for addresses that are the result of redirections.
3432 PH/48 A number of further additions for the benefit of the new test suite,
3433 including a fake gethostbyname() that interfaces to the fake DNS resolver
3436 TF/06 The fix for widen_domains has also been applied to qualify_single and
3437 search_parents which are the other dnslookup options that can cause
3440 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
3443 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
3445 TF/07 Exim produced the error message "an SRV record indicated no SMTP
3446 service" if it encountered an MX record with an empty target hostname.
3447 The message is now "an MX or SRV record indicated no SMTP service".
3449 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
3450 if the DNS of the sending site is misconfigured. This is quite a
3451 common situation. This change restores the behaviour of treating a
3452 helo verification defer as a failure.
3454 PH/51 If self=fail was set on a router, the bounce message did not include the
3455 actual error message.
3461 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
3463 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
3464 command itself ended in error, the underlying message about the transport
3465 filter timeout was being overwritten with the pipe command error. Now the
3466 underlying error message should be appended to the second error message.
3468 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
3470 PH/02 Reduce the amount of output that "make" produces by default. Full output
3471 can still be requested.
3473 PH/03 The warning log line about a condition test deferring for a "warn" verb
3474 was being output only once per connection, rather than after each
3475 occurrence (because it was using the same function as for successful
3476 "warn" verbs). This seems wrong, so I have changed it.
3478 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
3479 it should not have, which might have caused a crash in the right
3480 circumstances, but probably never did.
3482 PH/04 Installed a modified version of Tony Finch's patch to make submission
3483 mode fix the return path as well as the Sender: header line, and to
3484 add a /name= option so that you can make the user's friendly name appear
3487 TF/03 Added the control = fakedefer ACL modifier.
3489 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
3490 Mark Lowes for thorough testing.
3492 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
3494 TK/03 Merged latest SRS patch from Miles Wilton.
3496 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
3497 with the definition in sysexits.h (which is #included earlier).
3498 Fortunately, Exim does not actually use EX_OK. The code used to try to
3499 preserve the sysexits.h value, by assumimg that macro definitions were
3500 scanned for macro replacements. I have been disabused of this notion,
3501 so now the code just undefines EX_OK before #including unistd.h.
3503 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
3504 in the smtp transport. When a block could not be written in a single
3505 write() function, the timeout was being re-applied to each part-write.
3506 This seems wrong - if the receiver was accepting one byte at a time it
3507 would take for ever. The timeout is now adjusted when this happens. It
3508 doesn't have to be particularly precise.
3510 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
3511 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
3513 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
3514 without a domain if the check to <postmaster@domain> fails.
3516 SC/01 Eximstats: added -xls and the ability to specify output files
3517 (patch written by Frank Heydlauf).
3519 SC/02 Eximstats: use FileHandles for outputing results.
3521 SC/03 Eximstats: allow any combination of xls, txt, and html output.
3523 SC/04 Eximstats: fixed display of large numbers with -nvr option
3525 SC/05 Eximstats: fixed merging of reports with empty tables.
3527 SC/06 Eximstats: added the -include_original_destination flag
3529 SC/07 Eximstats: removed tabs and trailing whitespace.
3531 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
3533 TK/06 MBOX spool code: Add real "From " MBOX separator line
3534 so the .eml file is really in mbox format (even though
3535 most programs do not really care). Patch from Alex Miller.
3537 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
3538 The latter is generated from $received_to and is only set if the
3539 message has one envelope recipient. SA can use these headers,
3540 obviously out-of-the-box. Patch from Alex Miller.
3542 PH/08 The ${def test on a variable was returning false if the variable's
3543 value was "0", contrary to what the specification has always said!
3544 The result should be true unless the variable is empty.
3546 PH/09 The syntax error of a character other than { following "${if
3547 def:variable_name" (after optional whitespace) was not being diagnosed.
3548 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
3549 accidental colon was present, for example, could give incorrect results.
3551 PH/10 Tidied the code in a number of places where the st_size field of a stat()
3552 result is used (not including appendfile, where other changes are about
3555 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
3556 This involved changing a lot of size variables from int to off_t. It
3557 should work with maildirs and everything.
3559 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
3560 spamd dying while we are connected to it.
3562 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
3565 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
3566 function for BDB 4.3.
3568 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
3570 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
3571 the sources for PCRE have been re-arranged and more files are now
3574 PH/15 The code I had for printing potentially long long variables in PH/11
3575 above was not the best (it lost precision). The length of off_t variables
3576 is now inspected at build time, and an appropriate printing format (%ld
3577 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
3578 to be "long long int" or "long int". This is needed for the internal
3579 formatting function string_vformat().
3581 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
3582 the configuration file to be ":syslog", then the script "guesses" where
3583 the logs files are, rather than using the compiled in default. In our
3584 case the guess is not the same as the compiled default, so the script
3585 suddenly stopped working when I started to use syslog. The patch checks
3586 to see if log_file_path is "". If so, it attempts to read it from exim
3587 with no configuration file to get the compiled in version, before it
3588 falls back to the previous guessing code."
3590 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
3591 implementing BATV in an Exim configuration. See NewStuff for the gory
3594 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
3595 Makefile that are specific to HP-UX.
3597 PH/18 If the "use_postmaster" option was set for a recipient callout together
3598 with the "random" option, the postmaster address was used as the MAIL
3599 FROM address for the random test, but not for the subsequent recipient
3600 test. It is now used for both.
3602 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
3603 patch removes a few documentation additions to RFC 3028, because the
3604 latest draft now contains them. It adds the new en;ascii-case comparator
3605 and a new error check for 8bit text in MIME parts. Comparator and
3606 require names are now matched exactly. I enabled the subaddress
3607 extension, but it is not well tested yet (read: it works for me)."
3609 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
3610 rework some of the code of TK/09 above to avoid the hardwired use of
3611 "%lld" and "long long". Replaced the call to snprintf() with a call to
3614 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
3615 records point to non-existent hosts", "retry timeout exceeded", and
3616 "retry time not reached for any host after a long failure period".
3618 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
3619 experimental DomainKeys support:
3621 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
3622 (2) On an error such as an illegally used "control", the wrong name for
3623 the control was given.
3625 These problems did NOT occur unless DomainKeys support was compiled.
3627 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
3629 PH/24 Added ${if match_ip condition.
3631 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
3632 databases so that it will be absolutely obvious if a crash occurs in the
3633 DB library. This is a regular occurrence (often caused by mis-matched
3636 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
3637 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
3638 picked up on a user's system that detects such things. There doesn't seem
3639 to be a gcc warning option for this - only an attribute that has to be
3640 put on the function's prototype. It seems that in Fedora Core 4 they have
3641 set this on a number of new functions. No doubt there will be more in due
3644 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
3645 specify a transport. However, if an address that was verified by such a
3646 router was the subject of a callout, Exim crashed because it tried to
3647 read the rcpt_include_affixes from the non-existent transport. Now it
3648 just assumes that the setting of that option is false. This bug was
3649 introduced by 4.51/PH/31.
3651 PH/28 Changed -d+all to exclude +memory, because that information is very
3652 rarely of interest, but it makes the output a lot bigger. People tend to
3653 do -d+all out of habit.
3655 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
3656 code in os-type was giving problems when libc.so lives in lib64, like on
3659 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
3660 aren't the modern standard, and it seems that some systems' include files
3661 don't always have them. Exim was already checking for some of the newer
3662 ones like T_AAAA, and defining it itself. I've added checks for all the
3663 record types that Exim uses.
3665 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
3666 not automatically generating a new one, as it is supposed to. This
3667 prevented TLS from working. If the file did exist, but contained invalid
3668 data, a new version was generated, as expected. It was only the case of a
3669 non-existent file that was broken.
3671 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
3672 with a change in libdomainkeys > 0.64.
3674 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
3675 from DNS. If the selector record carries the flag, it now has
3676 precedence over the domain-wide flag.
3678 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
3680 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
3681 the use of a port name that isn't defined in /etc/services) occurred, the
3682 message was deferred as in a normal delivery, and thus remained on the
3683 spool, instead of being failed because of the mua_wrapper setting. This
3684 is now fixed, and I tidied up some of the mua_wrapper messages at the
3687 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
3688 the messages in a hash of arrays rather than using individual hashes.
3689 This is a bit cleaner and results in dramatic memory savings, albeit
3690 at a slight CPU cost.
3692 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
3693 as requested by Marc Sherman.
3695 SC/10 Eximstats: added histograms for user specified patterns as requested
3698 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
3700 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
3701 fopen() in the content-scanning modules that did not already have it.
3707 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
3708 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
3710 TK/02 Fix ACL "control" statement not being available in MIME ACL.
3712 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
3714 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
3715 to test Sieve filters that use "vacation".
3717 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
3718 that changes the way the GnuTLS parameters are stored in the cache file.
3719 The new format can be generated externally. For backward compatibility,
3720 if the data in the cache doesn't make sense, Exim assumes it has read an
3721 old-format file, and it generates new data and writes a new file. This
3722 means that you can't go back to an older release without removing the
3725 PH/03 A redirect router that has both "unseen" and "one_time" set does not
3726 work if there are any delivery delays because "one_time" forces the
3727 parent to be marked "delivered", so its unseen clone is never tried
3728 again. For this reason, Exim now forbids the simultaneous setting of
3731 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
3732 redirected to themselves ("homonym" addresses). Read the long ChangeLog
3733 entry if you want to know the details. The fix, however, neglected to
3734 consider the case when local delivery batching is involved. The test for
3735 "previously delivered" was not happening when checking to see if an
3736 address could be batched with a previous (undelivered) one; under
3737 certain circumstances this could lead to multiple deliveries to the same
3740 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
3741 in its include files, and this causes problems building Exim.
3743 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
3744 header_syntax) but Exim was just ignoring anything given after a slash.
3745 In particular, this caused confusion with an attempt to use "verify =
3746 reverse_host_lookup/defer_ok". An error is now given when options are
3747 supplied for verify items that do not have them. (Maybe reverse_host_
3748 lookup should have a defer_ok option, but that's a different point.)
3750 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
3751 defined by RFC 821) to 2048, because there were problems with some AUTH
3752 commands, and RFC 1869 says the size should be increased for extended
3753 SMTP commands that take arguments.
3755 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
3758 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
3759 "unknown" error; now it says that the functionality isn't in the binary.
3761 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
3762 an address' error message when a string expansion fails (syntax or
3763 whatever). Otherwise the password may appear in the log. Following change
3764 PH/42 below, there is no longer a chance of it appearing in a bounce
3767 PH/11 Installed exipick version 20050225.0 from John Jetmore.
3769 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
3770 of its addresses was ever tried. (Bugzilla bug #2.)
3772 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
3773 the result incorrectly in the debug output. (It correctly added a newline
3774 to what was transported.)
3776 TF/01 Added $received_time.
3778 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
3779 commented out examples of how to interface to a virus scanner and to
3780 SpamAssassin. Also added commented examples of av_scanner and
3781 spamd_address settings.
3783 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
3784 and controls are allowed in which ACLs. There were a couple of minor
3785 errors. Some of the entries in the conditions table (which is a table of
3786 where they are NOT allowed) were getting very unwieldy; rewrote them as a
3787 negation of where the condition IS allowed.
3789 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
3791 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
3792 header file does not have a version number, so I've had to invent a new
3793 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
3794 API. The code is untested by me (my Linux distribution still has 0.3.2 of
3795 radiusclient), but it was contributed by a Radius user.
3797 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
3798 files or directories, for OpenSSL.
3800 PH/19 When an Exim process that is running as root has to create an Exim log
3801 file, it does so in a subprocess that runs as exim:exim so as to get the
3802 ownership right at creation (otherwise, other Exim processes might see
3803 the file with the wrong ownership). There was no test for failure of this
3804 fork() call, which would lead to the process getting stuck as it waited
3805 for a non-existent subprocess. Forks do occasionally fail when resources
3806 run out. I reviewed all the other calls to fork(); they all seem to check
3809 PH/20 When checking for unexpected SMTP input at connect time (before writing
3810 the banner), Exim was not dealing correctly with a non-positive return
3811 from the read() function. If the client had disconnected by this time,
3812 the result was a log entry for a synchronization error with an empty
3813 string after "input=" when read() returned zero. If read() returned -1
3814 (an event I could not check), uninitialized data bytes were printed.
3815 There were reports of junk text (parts of files, etc) appearing after
3818 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
3820 PH/22 Added support for macro redefinition, and (re)definition in between
3821 driver and ACL definitions.
3823 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
3824 forgetting to use the resulting value; it was using the unexpanded value.
3826 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
3827 hadn't been configured. The fix is from Juergen Kreileder, who
3828 understands it better than I do:
3830 "Here's what I see happening with three configured cyrus_sasl
3831 authenticators configured (plain, login, cram-md5):
3833 On startup auth_cyrus_sasl_init() gets called for each of these.
3834 This means three calls to sasl_listmech() without a specified mech_list.
3835 => SASL tests which mechs of all available mechs actually work
3836 => three warnings about OTP not working
3837 => the returned list contains: plain, login, cram-md5, digest-md5, ...
3839 With the patch, sasl_listmech() also gets called three times. But now
3840 SASL's mech_list option is set to the server_mech specified in the the
3841 authenticator. Or in other words, the answer from sasl_listmech()
3842 gets limited to just the mech you're testing for (which is different
3844 => the return list contains just 'plain' or 'login', 'cram-md5' or
3845 nothing depending on the value of ob->server_mech.
3847 I've just tested the patch: Authentication still works fine,
3848 unavailable mechs specified in the exim configuration are still
3849 caught, and the auth.log warnings about OTP are gone."
3851 PH/25 When debugging is enabled, the contents of the command line are added
3852 to the debugging output, even when log_selector=+arguments is not
3855 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
3856 answer is "GNU", and only if the return is "GNU/something" is the answer
3859 PH/27 $acl_verify_message is now set immediately after the failure of a
3860 verification in an ACL, and so is available in subsequent modifiers. In
3861 particular, the message can be preserved by coding like this:
3863 warn !verify = sender
3864 set acl_m0 = $acl_verify_message
3866 Previously, $acl_verify_message was set only while expanding "message"
3867 and "log_message" when a very denied access.
3869 PH/28 Modified OS/os.c-Linux with
3871 -#ifndef OS_LOAD_AVERAGE
3872 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
3874 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
3875 nomenclature these days.)
3877 PH/29 Installed patch from the Sieve maintainer that adds the options
3878 sieve_useraddress and sieve_subaddress to the redirect router.
3880 PH/30 In these circumstances:
3881 . Two addresses routed to the same list of hosts;
3882 . First host does not offer TLS;
3883 . First host accepts first address;
3884 . First host gives temporary error to second address;
3885 . Second host offers TLS and a TLS session is established;
3886 . Second host accepts second address.
3887 Exim incorrectly logged both deliveries with the TLS parameters (cipher
3888 and peerdn, if requested) that were in fact used only for the second
3891 PH/31 When doing a callout as part of verifying an address, Exim was not paying
3892 attention to any local part prefix or suffix that was matched by the
3893 router that accepted the address. It now behaves in the same way as it
3894 does for delivery: the affixes are removed from the local part unless
3895 rcpt_include_affixes is set on the transport.
3897 PH/32 Add the sender address, as F=<...>, to the log line when logging a
3898 timeout during the DATA phase of an incoming message.
3900 PH/33 Sieve envelope tests were broken for match types other than :is. I have
3901 applied a patch sanctioned by the Sieve maintainer.
3903 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
3904 the uid or gid is negative. A case of a negative gid caused this to be
3905 noticed. The fix allows for either to be negative.
3907 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
3908 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
3911 PH/36 The change PH/12 above was broken. Fixed it.
3913 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
3914 the grounds that routing the same address twice would always produce the
3915 same answer. This might have been true once, but it is certainly no
3916 longer true now. Routing a child address may depend on the previous
3917 routing that produced that child. Some complicated redirection strategies
3918 went wrong when messages had multiple recipients, and made Exim's
3919 behaviour dependent on the order in which the addresses were given.
3921 I have moved the duplicate checking until after the routing is complete.
3922 Exim scans the addresses that are assigned to local and remote
3923 transports, and removes any duplicates. This means that more work will be
3924 done, as duplicates will always all be routed, but duplicates are
3925 presumably rare, so I don't expect this is of any significance.
3927 For deliveries to pipes, files, and autoreplies, the duplicate checking
3928 still happens during the routing process, since they are not going to be
3931 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
3932 It corrects a timeout issue with spamd. This is Ian's comment: "The
3933 background is that sometimes spamd either never reads data from a
3934 connection it has accepted, or it never writes response data. The exiscan
3935 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
3936 blindly assumes that writes won't block so it may never time out."
3938 PH/39 Allow G after quota size as well as K and M.
3940 PH/40 The value set for $authenticated_id in an authenticator may not contain
3941 binary zeroes or newlines because the value is written to log lines and
3942 to spool files. There was no check on this. Now the value is run through
3943 the string_printing() function so that such characters are converted to
3944 printable escape sequences.
3946 PH/41 $message_linecount is a new variable that contains the total number of
3947 lines in the message. Compare $body_linecount, which is the count for the
3950 PH/42 Exim no longer gives details of delivery errors for specific addresses in
3951 bounce and delay warning messages, except in certain special cases, which
3954 (a) An SMTP error message from a remote host;
3955 (b) A message specified in a :fail: redirection;
3956 (c) A message specified in a "fail" command in a system filter;
3957 (d) A message specified in a FAIL return from the queryprogram router;
3958 (e) A message specified by the cannot_route_message router option.
3960 In these cases only, Exim does include the error details in bounce and
3961 warning messages. There are also a few cases where bland messages such
3962 as "unrouteable address" or "local delivery error" are given.
3964 PH/43 $value is now also set for the "else" part of a ${run expansion.
3966 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
3967 being worked on, but at least Exim now implements the latest version to
3970 PH/45 In a pipe transport, although a timeout while waiting for the pipe
3971 process to complete was treated as a delivery failure, a timeout while
3972 writing the message to the pipe was logged, but erroneously treated as a
3973 successful delivery. Such timeouts include transport filter timeouts. For
3974 consistency with the overall process timeout, these timeouts are now
3975 treated as errors, giving rise to delivery failures by default. However,
3976 there is now a new Boolean option for the pipe transport called
3977 timeout_defer, which, if set TRUE, converts the failures into defers for
3978 both kinds of timeout. A transport filter timeout is now identified in
3981 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
3982 systems where "make" and "gmake" are different, calling "gmake" at top
3983 level broke things. I've arranged for the value of $(MAKE) to be passed
3984 from the Makefile to this script so that it can call the same version of
3988 A note about Exim versions 4.44 and 4.50
3989 ----------------------------------------
3991 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
3992 changes of various kinds. As a consequence, a big documentation update was
3993 needed. This delayed the release for rather longer than seemed good, especially
3994 in the light of a couple of (minor) security issues. Therefore, the changes
3995 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
3996 release. So 4.44 and 4.50 are in effect two different branches that both start
3999 I have left the 4.50 change log unchanged; it contains all the changes since
4000 4.43. The change log for 4.44 is below; many of its items are identical to
4001 those for 4.50. This seems to be the most sensible way to preserve the
4002 historical information.
4008 1. Minor wording change to the doc/README.SIEVE file.
4010 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
4011 computation of the current number of files was incorrect.
4013 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4014 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4017 4. Give more explanation in the error message when the command for a transport
4018 filter fails to execute.
4020 5. There are several places where Exim runs a non-Exim command in a
4021 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4022 was being done only for the command run by the queryprogram router. It is
4023 now done for all such subprocesses. The other cases are: ${run, transport
4024 filters, and the commands run by the lmtp and pipe transports.
4026 6. Added CONFIGURE_GROUP build-time option.
4028 7. Some older OS have a limit of 256 on the maximum number of file
4029 descriptors. Exim was using setrlimit() to set 1000 as a large value
4030 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4031 systems. I've change it so that if it can't get 1000, it tries for 256.
4033 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4034 was an oversight, and furthermore, ever since the addition of extra
4035 controls (e.g. 4.43/32), the checks on when to allow different forms of
4036 "control" were broken. There should now be diagnostics for all cases when a
4037 control that does not make sense is encountered.
4039 9. Added the /retain_sender option to "control=submission".
4041 10. $recipients is now available in the predata ACL (oversight).
4043 11. Tidy the search cache before the fork to do a delivery from a message
4044 received from the command line. Otherwise the child will trigger a lookup
4045 failure and thereby defer the delivery if it tries to use (for example) a
4046 cached ldap connection that the parent has called unbind on.
4048 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4049 of $address_data from the recipient verification was clobbered by the
4050 sender verification.
4052 13. The value of address_data from a sender verification is now available in
4053 $sender_address_data in subsequent conditions in the ACL statement.
4055 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
4057 15. Added a new option "connect=<time>" to callout options, to set a different
4060 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4061 was its contents. (It was OK if the option was not defined at all.)
4063 17. A "Completed" log line is now written for messages that are removed from
4064 the spool by the -Mrm option.
4066 18. New variables $sender_verify_failure and $recipient_verify_failure contain
4067 information about exactly what failed.
4069 19. Added -dd to debug only the daemon process.
4071 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
4072 handles timeouts, both on the server side and network timeouts. Renamed the
4073 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
4075 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
4076 It is now set to "smtps".
4078 22. $host_address is now set to the target address during the checking of
4079 ignore_target_hosts.
4081 23. When checking ignore_target_hosts for an ipliteral router, no host name was
4082 being passed; this would have caused $sender_host_name to have been used if
4083 matching the list had actually called for a host name (not very likely,
4084 since this list is usually IP addresses). A host name is now passed as
4087 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4088 code that specifies a non-restarting handler (typically sigaction() in
4089 modern systems) in an attempt to fix a rare and obscure crash bug.
4091 25. Narrowed the window for a race in the daemon that could cause it to ignore
4092 SIGCHLD signals. This is not a major problem, because they are used only to
4093 wake it up if nothing else does.
4095 26. A malformed maildirsize file could cause Exim to calculate negative values
4096 for the mailbox size or file count. Odd effects could occur as a result.
4097 The maildirsize information is now recalculated if the size or filecount
4100 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4101 support for a long time. Removed HAVE_SYS_VFS_H.
4103 28. Installed the latest version of exipick from John Jetmore.
4105 29. In an address list, if the pattern was not a regular expression, an empty
4106 subject address (from a bounce message) matched only if the pattern was an
4107 empty string. Non-empty patterns were not even tested. This was the wrong
4108 because it is perfectly reasonable to use an empty address as part of a
4109 database query. An empty address is now tested by patterns that are
4110 lookups. However, all the other forms of pattern expect the subject to
4111 contain a local part and a domain, and therefore, for them, an empty
4112 address still always fails if the pattern is not itself empty.
4114 30. Exim went into a mad DNS loop when attempting to do a callout where the
4115 host was specified on an smtp transport, and looking it up yielded more
4116 than one IP address.
4118 31. Re-factored the code for checking spool and log partition space into a
4119 function that finds that data and another that does the check. The former
4120 is then used to implement four new variables: $spool_space, $log_space,
4121 $spool_inodes, and $log_inodes.
4123 32. The RFC2047 encoding function was originally intended for short strings
4124 such as real names; it was not keeping to the 75-character limit for
4125 encoded words that the RFC imposes. It now respects the limit, and
4126 generates multiple encoded words if necessary. To be on the safe side, I
4127 have increased the buffer size for the ${rfc2047: expansion operator from
4130 33. It is now permitted to omit both strings after an "if" condition; if the
4131 condition is true, the result is "true". As before, when the second string
4132 is omitted, a false condition yields an empty string. This makes it less
4133 cumbersome to write custom ACL and router conditions.
4135 34. Failure to deliver a bounce message always caused it to be frozen, even if
4136 there was an errors_to setting on the router. The errors_to setting is now
4139 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4140 canonical form (fully expanded) before being placed in
4141 $sender_host_address.
4143 36. The table in the code that translates DNS record types into text (T_A to
4144 "A" for instance) was missing entries for NS and CNAME. It is just possible
4145 that this could have caused confusion if both these types were looked up
4146 for the same domain, because the text type is used as part of Exim's
4147 per-process caching. But the chance of anyone hitting this buglet seems
4150 37. The dnsdb lookup has been extended in a number of ways.
4152 (1) There is a new type, "zns", which walks up the domain tree until it
4153 finds some nameserver records. It should be used with care.
4155 (2) There is a new type, "mxh", which is like "mx" except that it returns
4156 just the host names, not the priorities.
4158 (3) It is now possible to give a list of domains (or IP addresses) to be
4159 looked up. The behaviour when one of the lookups defers can be
4160 controlled by a keyword.
4162 (4) It is now possible to specify the separator character for use when
4163 multiple records are returned.
4165 38. The dnslists ACL condition has been extended: it is now possible to supply
4166 a list of IP addresses and/or domains to be looked up in a particular DNS
4169 39. Added log_selector=+queue_time_overall.
4171 40. When running the queue in the test harness, wait just a tad after forking a
4172 delivery process, to get repeatability of debugging output.
4174 41. Include certificate and key file names in error message when GnuTLS fails
4175 to set them up, because the GnuTLS error message doesn't include the name
4176 of the failing file when there is a problem reading it.
4178 42. Allow both -bf and -bF in the same test run.
4180 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
4182 44. The "Exiscan patch" is now merged into the mainline Exim source.
4184 45. Sometimes the final signoff response after QUIT could fail to get
4185 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4186 < 0 before doing a fflush(). This bug looks as though it goes back to the
4187 introduction of TLS in release 3.20, but "sometimes" must have been rare
4188 because the tests only now provoked it.
4190 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
4191 (this can affect the format of dates).
4193 47. exim_tidydb, when checking for the continued existence of a message for
4194 which it has found a message-specific retry record, was not finding
4195 messages that were in split spool directories. Consequently, it was
4196 deleting retry records that should have stayed in existence.
4198 48. Steve fixed some bugs in eximstats.
4200 49. The SPA authentication driver was not abandoning authentication and moving
4201 on to the next authenticator when an expansion was forced to fail,
4202 contradicting the general specification for all authenticators. Instead it
4203 was generating a temporary error. It now behaves as specified.
4205 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
4206 (the order specifies the preference for clients). The order is now AES256,
4207 AES128, 3DES, ARCFOUR128.
4209 51. Small patch to Sieve code - explicitly set From: when generating an
4212 52. Exim crashed if a remote delivery caused a very long error message to be
4213 recorded - for instance if somebody sent an entire SpamAssassin report back
4214 as a large number of 550 error lines. This bug was coincidentally fixed by
4215 increasing the size of one of Exim's internal buffers (big_buffer) that
4216 happened as part of the Exiscan merge. However, to be on the safe side, I
4217 have made the code more robust (and fixed the comments that describe what
4220 53. Now that there can be additional text after "Completed" in log lines (if
4221 the queue_time_overall log selector is set), a one-byte patch to exigrep
4222 was needed to allow it to recognize "Completed" as not the last thing in
4225 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
4226 patch that reportedly fixes this has been added. I am not expert enough to
4227 create a test for it. This is what the patch creator wrote:
4229 "I found a little strange behaviour of ldap code when working with
4230 Windows 2003 AD Domain, where users was placed in more than one
4231 Organization Units. When I tried to give exim partial DN, the exit code
4232 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
4233 But simultaneously result of request was absolutely normal ldap result,
4234 so I produce this patch..."
4236 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
4237 REFERENCE, so I have modified the code to exclude the patch when that macro
4240 55. Some experimental protocols are using DNS PTR records for new purposes. The
4241 keys for these records are domain names, not reversed IP addresses. The
4242 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4243 leaves it alone. Component reversal etc. now happens only for IP addresses.
4246 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
4248 57. Double the size of the debug message buffer (to 2048) so that more of very
4249 long debug lines gets shown.
4251 58. The exicyclog utility now does better if the number of log files to keep
4252 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
4254 59. Two changes related to the smtp_active_hostname option:
4256 (1) $smtp_active_hostname is now available as a variable.
4257 (2) The default for smtp_banner uses $smtp_active_hostname instead
4258 of $primary_hostname.
4260 60. The host_aton() function is supposed to be passed a string that is known
4261 to be a valid IP address. However, in the case of IPv6 addresses, it was
4262 not checking this. This is a hostage to fortune. Exim now panics and dies
4263 if the condition is not met. A case was found where this could be provoked
4264 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4265 components; fortuitously, this particular loophole had already been fixed
4266 by change 4.50/55 above.
4268 If there are any other similar loopholes, the new check in host_aton()
4269 itself should stop them being exploited. The report I received stated that
4270 data on the command line could provoke the exploit when Exim was running as
4271 exim, but did not say which command line option was involved. All I could
4272 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4273 running as the user.
4276 61. There was a buffer overflow vulnerability in the SPA authentication code
4277 (which came originally from the Samba project). I have added a test to the
4278 spa_base64_to_bits() function which I hope fixes it.
4281 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
4282 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
4284 63. The daemon start-up calls getloadavg() while still root for those OS that
4285 need the first call to be done as root, but it missed one case: when
4286 deliver_queue_load_max is set with deliver_drop_privilege. This is
4287 necessary for the benefit of the queue runner, because there is no re-exec
4288 when deliver_drop_privilege is set.
4290 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
4291 This has been fixed.
4293 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4294 was in use, was not putting the data itself into the right store pool;
4295 consequently, it could be overwritten for a subsequent message in the same
4296 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4299 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
4301 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
4302 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
4303 the function were treating the return as a boolean value, which happened to
4304 work because 0=false and not-0=true, but is not correct code.
4306 68. The host_aton() function was not handling scoped IPv6 addresses (those
4307 with, for example, "%eth0" on the end) correctly.
4309 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
4310 negated items (that is, ~something) in unsigned ints. Some compilers
4311 apparently mutter when there is no cast.
4313 70. If an address verification called from an ACL failed, and did not produce a
4314 user-specific message (i.e. there was only a "system" message), nothing was
4315 put in $acl_verify_message. In this situation, it now puts the system
4318 71. Change 4.23/11 added synchronization checking at the start of an SMTP
4319 session; change 4.31/43 added the unwanted input to the log line - except
4320 that it did not do this in the start of session case. It now does.
4322 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
4323 This is wrong and can cause the other end to generate a synchronization
4324 error if it is another Exim or anything else that does the synchronization
4325 check. A QUIT command is no longer sent after a timeout.
4327 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
4328 during host lookups.
4330 74. The defer_ok option of callout verification was not working if it was used
4331 when verifying addresses in header lines, that is, for this case:
4333 verify = header_sender/callout=defer_ok
4335 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
4336 those file descriptors could be used for SMTP connections. If anything
4337 wrote to stderr (the example that came up was "warn" in embedded Perl), it
4338 could be sent to the SMTP client, causing chaos. The daemon now opens
4339 stdin, stdout, and stderr to /dev/null when it puts itself into the
4342 76. Arrange for output from Perl's "warn" command to be written to Exim's main
4343 log by default. The user can override this with suitable Perl magic.
4345 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
4346 the log message when discard triggers, was not working for the DATA ACL or
4347 for the non-SMTP ACL.
4349 78. Error message wording change in sieve.c.
4351 79. If smtp_accept_max_per_host was set, the number of connections could be
4352 restricted to fewer than expected, because the daemon was trying to set up
4353 a new connection before checking whether the processes handling previous
4354 connections had finished. The check for completed processes is now done
4355 earlier. On busy systems, this bug wouldn't be noticed because something
4356 else would have woken the daemon, and it would have reaped the completed
4359 80. If a message was submitted locally by a user whose login name contained one
4360 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
4361 It caused a spool format error. I have fixed the spool reading code. A
4362 related problem was that the "from" clause in the Received: line became
4363 illegal because of the space(s). It is now covered by ${quote_local_part.
4365 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
4368 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
4370 83. Patch from Sieve maintainer to fix unterminated string problem in
4371 "vacation" handling.
4373 84. Some minor changes to the Linux configuration files to help with other
4374 OS variants using glibc.
4376 85. One more patch for Sieve to update vacation handling to latest spec.
4379 ----------------------------------------------------
4380 See the note above about the 4.44 and 4.50 releases.
4381 ----------------------------------------------------
4387 1. Change 4.43/35 introduced a bug that caused file counts to be
4388 incorrectly computed when quota_filecount was set in an appendfile
4391 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4392 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4395 3. Give more explanation in the error message when the command for a transport
4396 filter fails to execute.
4398 4. There are several places where Exim runs a non-Exim command in a
4399 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4400 was being done only for the command run by the queryprogram router. It is
4401 now done for all such subprocesses. The other cases are: ${run, transport
4402 filters, and the commands run by the lmtp and pipe transports.
4404 5. Some older OS have a limit of 256 on the maximum number of file
4405 descriptors. Exim was using setrlimit() to set 1000 as a large value
4406 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4407 systems. I've change it so that if it can't get 1000, it tries for 256.
4409 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4410 was an oversight, and furthermore, ever since the addition of extra
4411 controls (e.g. 4.43/32), the checks on when to allow different forms of
4412 "control" were broken. There should now be diagnostics for all cases when a
4413 control that does not make sense is encountered.
4415 7. $recipients is now available in the predata ACL (oversight).
4417 8. Tidy the search cache before the fork to do a delivery from a message
4418 received from the command line. Otherwise the child will trigger a lookup
4419 failure and thereby defer the delivery if it tries to use (for example) a
4420 cached ldap connection that the parent has called unbind on.
4422 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4423 of $address_data from the recipient verification was clobbered by the
4424 sender verification.
4426 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4427 was its contents. (It was OK if the option was not defined at all.)
4429 11. A "Completed" log line is now written for messages that are removed from
4430 the spool by the -Mrm option.
4432 12. $host_address is now set to the target address during the checking of
4433 ignore_target_hosts.
4435 13. When checking ignore_target_hosts for an ipliteral router, no host name was
4436 being passed; this would have caused $sender_host_name to have been used if
4437 matching the list had actually called for a host name (not very likely,
4438 since this list is usually IP addresses). A host name is now passed as
4441 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4442 code that specifies a non-restarting handler (typically sigaction() in
4443 modern systems) in an attempt to fix a rare and obscure crash bug.
4445 15. Narrowed the window for a race in the daemon that could cause it to ignore
4446 SIGCHLD signals. This is not a major problem, because they are used only to
4447 wake it up if nothing else does.
4449 16. A malformed maildirsize file could cause Exim to calculate negative values
4450 for the mailbox size or file count. Odd effects could occur as a result.
4451 The maildirsize information is now recalculated if the size or filecount
4454 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4455 support for a long time. Removed HAVE_SYS_VFS_H.
4457 18. Updated exipick to current release from John Jetmore.
4459 19. Allow an empty sender to be matched against a lookup in an address list.
4460 Previously the only cases considered were a regular expression, or an
4463 20. Exim went into a mad DNS lookup loop when doing a callout where the
4464 host was specified on the transport, if the DNS lookup yielded more than
4467 21. The RFC2047 encoding function was originally intended for short strings
4468 such as real names; it was not keeping to the 75-character limit for
4469 encoded words that the RFC imposes. It now respects the limit, and
4470 generates multiple encoded words if necessary. To be on the safe side, I
4471 have increased the buffer size for the ${rfc2047: expansion operator from
4474 22. Failure to deliver a bounce message always caused it to be frozen, even if
4475 there was an errors_to setting on the router. The errors_to setting is now
4478 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4479 canonical form (fully expanded) before being placed in
4480 $sender_host_address.
4482 24. Updated eximstats to version 1.33
4484 25. Include certificate and key file names in error message when GnuTLS fails
4485 to set them up, because the GnuTLS error message doesn't include the name
4486 of the failing file when there is a problem reading it.
4488 26. Expand error message when OpenSSL has problems setting up cert/key files.
4491 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
4492 (this can affect the format of dates).
4494 28. exim_tidydb, when checking for the continued existence of a message for
4495 which it has found a message-specific retry record, was not finding
4496 messages that were in split spool directories. Consequently, it was
4497 deleting retry records that should have stayed in existence.
4499 29. eximstats updated to version 1.35
4500 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
4501 1.35 - bugfix such that pie charts by volume are generated correctly
4503 30. The SPA authentication driver was not abandoning authentication and moving
4504 on to the next authenticator when an expansion was forced to fail,
4505 contradicting the general specification for all authenticators. Instead it
4506 was generating a temporary error. It now behaves as specified.
4508 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
4509 (the order specifies the preference for clients). The order is now AES256,
4510 AES128, 3DES, ARCFOUR128.
4512 31. Small patch to Sieve code - explicitly set From: when generating an
4515 32. Exim crashed if a remote delivery caused a very long error message to be
4516 recorded - for instance if somebody sent an entire SpamAssassin report back
4517 as a large number of 550 error lines. This bug was coincidentally fixed by
4518 increasing the size of one of Exim's internal buffers (big_buffer) that
4519 happened as part of the Exiscan merge. However, to be on the safe side, I
4520 have made the code more robust (and fixed the comments that describe what
4523 33. Some experimental protocols are using DNS PTR records for new purposes. The
4524 keys for these records are domain names, not reversed IP addresses. The
4525 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4526 leaves it alone. Component reversal etc. now happens only for IP addresses.
4529 34. The host_aton() function is supposed to be passed a string that is known
4530 to be a valid IP address. However, in the case of IPv6 addresses, it was
4531 not checking this. This is a hostage to fortune. Exim now panics and dies
4532 if the condition is not met. A case was found where this could be provoked
4533 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4534 components; fortuitously, this particular loophole had already been fixed
4535 by change 4.50/55 or 4.44/33 above.
4537 If there are any other similar loopholes, the new check in host_aton()
4538 itself should stop them being exploited. The report I received stated that
4539 data on the command line could provoke the exploit when Exim was running as
4540 exim, but did not say which command line option was involved. All I could
4541 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4542 running as the user.
4545 35. There was a buffer overflow vulnerability in the SPA authentication code
4546 (which came originally from the Samba project). I have added a test to the
4547 spa_base64_to_bits() function which I hope fixes it.
4550 36. The daemon start-up calls getloadavg() while still root for those OS that
4551 need the first call to be done as root, but it missed one case: when
4552 deliver_queue_load_max is set with deliver_drop_privilege. This is
4553 necessary for the benefit of the queue runner, because there is no re-exec
4554 when deliver_drop_privilege is set.
4556 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4557 was in use, was not putting the data itself into the right store pool;
4558 consequently, it could be overwritten for a subsequent message in the same
4559 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4562 38. Sometimes the final signoff response after QUIT could fail to get
4563 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4564 < 0 before doing a fflush(). This bug looks as though it goes back to the
4565 introduction of TLS in release 3.20, but "sometimes" must have been rare
4566 because the tests only now provoked it.
4572 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
4573 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
4574 Now it returns an integer. A number of calls were still expecting a T/F
4575 return. Fortuitously, in all cases, the tests worked in OK situations,
4576 which is the norm. However, things would have gone wrong on any write
4577 failures on the smtp file descriptor. This function is used when sending
4578 messages over SMTP and also when doing verify callouts.
4580 2. When Exim is called to do synchronous delivery of a locally submitted
4581 message (the -odf or -odi options), it no longer closes stderr before doing
4584 3. Implemented the mua_wrapper option.
4586 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
4588 5. Implemented the functions header_remove(), header_testname(),
4589 header_add_at_position(), and receive_remove_recipient(), and exported them
4592 6. If an ACL "warn" statement specified the addition of headers, Exim already
4593 inserted X-ACL-Warn: at the start if there was no header name. However, it
4594 was not making this test for the second and subsequent header lines if
4595 there were newlines in the string. This meant that an invalid header could
4596 be inserted if Exim was badly configured.
4598 7. Allow an ACL "warn" statement to add header lines at the start or after all
4599 the Received: headers, as well as at the end.
4601 8. Added the rcpt_4xx retry error code.
4603 9. Added postmaster_mailfrom=xxx to callout verification option.
4605 10. Added mailfrom=xxxx to the callout verification option, for verify=
4608 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
4609 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
4611 12. Inserted some casts to stop certain compilers warning when using pointer
4612 differences as field lengths or precisions in printf-type calls (mostly
4613 affecting debugging statements).
4615 13. Added optional readline() support for -be (dynamically loaded).
4617 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
4618 same clock tick as a message's arrival, so that its received time was the
4619 same as the "first fail" time on the retry record, and that message
4620 remained on the queue past the ultimate address timeout, every queue runner
4621 would try a delivery (because it was past the ultimate address timeout) but
4622 after another failure, the ultimate address timeout, which should have then
4623 bounced the address, did not kick in. This was a "< instead of <=" error;
4624 in most cases the first failure would have been in the next clock tick
4625 after the received time, and all would be well.
4627 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
4628 being recognized when the domain list was tested by the match_domain
4629 condition in an expansion string.
4631 16. Added the ${str2b64: operator.
4633 17. Exim was always calling setrlimit() to set a large limit for the number of
4634 processes, without checking whether the existing limit was already
4635 adequate. (It did check for the limit on file descriptors.) Furthermore,
4636 errors from getrlimit() and setrlimit() were being ignored. Now they are
4637 logged to the main and panic logs, but Exim does carry on, to try to do its
4638 job under whatever limits there are.
4640 18. Imported PCRE 5.0.
4642 19. Trivial typo in log message " temporarily refused connection" (the leading
4645 20. If the log selector return_path_on_delivery was set and an address was
4646 redirected to /dev/null, the delivery process crashed because it assumed
4647 that a return path would always be set for a "successful" delivery. In this
4648 case, the whole delivery is bypassed as an optimization, and therefore no
4651 21. Internal re-arrangement: the function for sending a challenge and reading
4652 a response while authentication was assuming a zero-terminated challenge
4653 string. It's now changed to take a pointer and a length, to allow for
4654 binary data in such strings.
4656 22. Added the cyrus_sasl authenticator (code supplied by MBM).
4658 23. Exim was not respecting finduser_retries when seeking the login of the
4659 uid under which it was called; it was always trying 10 times. (The default
4660 setting of finduser_retries is zero.) Also, it was sleeping after the final
4661 failure, which is pointless.
4663 24. Implemented tls_on_connect_ports.
4665 25. Implemented acl_smtp_predata.
4667 26. If the domain in control=submission is set empty, Exim assumes that the
4668 authenticated id is a complete email address when it generates From: or
4669 Sender: header lines.
4671 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
4672 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
4673 chown and chgrp in /bin and hostname in /usr/bin.
4675 28. Exim was keeping the "process log" file open after each use, just as it
4676 does for the main log. This opens the possibility of it remaining open for
4677 long periods when the USR1 signal hits a daemon. Occasional processlog
4678 errors were reported, that could have been caused by this. Anyway, it seems
4679 much more sensible not to leave this file open at all, so that is what now
4682 29. The long-running daemon process does not normally write to the log once it
4683 has entered its main loop, and it closes the log before doing so. This is
4684 so that log files can straightforwardly be renamed and moved. However,
4685 there are a couple of unusual error situations where the daemon does write
4686 log entries, and I had neglected to close the log afterwards.
4688 30. The text of an SMTP error response that was received during a remote
4689 delivery was being truncated at 512 bytes. This is too short for some of
4690 the long messages that one sometimes sees. I've increased the limit to
4693 31. It is now possible to make retry rules that apply only when a message has a
4694 specific sender, in particular, an empty sender.
4696 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
4697 it possible to be selective about when SMTP synchronization is enforced.
4699 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
4701 32. Implemented hosts_connection_nolog.
4703 33. Added an ACL for QUIT.
4705 34. Setting "delay_warning=" to disable warnings was not working; it gave a
4708 35. Added mailbox_size and mailbox_filecount to appendfile.
4710 36. Added control = no_multiline_responses to ACLs.
4712 37. There was a bug in the logic of the code that waits for the clock to tick
4713 in the case where the clock went backwards by a substantial amount such
4714 that the microsecond fraction of "now" was more than the microsecond
4715 fraction of "then" (but the whole seconds number was less).
4717 38. Added support for the libradius Radius client library this is found on
4718 FreeBSD (previously only the radiusclient library was supported).
4724 1. When certain lookups returned multiple values in the form name=value, the
4725 quoting of the values was not always being done properly. Specifically:
4726 (a) If the value started with a double quote, but contained no whitespace,
4728 (b) If the value contained whitespace other than a space character (i.e.
4729 tabs or newlines or carriage returns) it was not quoted.
4730 This fix has been applied to the mysql and pgsql lookups by writing a
4731 separate quoting function and calling it from the lookup code. The fix
4732 should probably also be applied to nisplus, ibase and oracle lookups, but
4733 since I cannot test any of those, I have not disturbed their existing code.
4735 2. A hit in the callout cache for a specific address caused a log line with no
4736 reason for rejecting RCPT. Now it says "Previous (cached) callout
4737 verification failure".
4739 3. There was an off-by-one bug in the queryprogram router. An over-long
4740 return line was truncated at 256 instead of 255 characters, thereby
4741 overflowing its buffer with the terminating zero. As well as fixing this, I
4742 have increased the buffer size to 1024 (and made a note to document this).
4744 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
4745 when Exim is waiting for an SMTP response from a remote server, Exim
4746 restarts its select() call on the socket, thereby resetting its timeout.
4747 This is not a problem when such interrupts are rare. Somebody set up a cron
4748 job to run exiwhat every 2 minutes, which is less than the normal select()
4749 timeout (5 or 10 minutes). This meant that the select() timeout never
4750 kicked in because it was always reset. I have fixed this by comparing the
4751 time when an interrupt arrives with the time at the start of the first call
4752 to select(). If more time than the timeout has elapsed, the interrupt is
4753 treated as a timeout.
4755 5. Some internal re-factoring in preparation for the addition of Sieve
4756 extensions (by MH). In particular, the "personal" test is moved to a
4757 separate function, and given an option for scanning Cc: and Bcc: (which is
4758 not set for Exim filters).
4760 6. When Exim created an email address using the login of the caller as the
4761 local part (e.g. when creating a From: or Sender: header line), it was not
4762 quoting the local part when it contained special characters such as @.
4764 7. Installed new OpenBSD configuration files.
4766 8. Reworded some messages for syntax errors in "and" and "or" conditions to
4767 try to make them clearer.
4769 9. Callout options, other than the timeout value, were being ignored when
4770 verifying sender addresses in header lines. For example, when using
4772 verify = header_sender/callout=no_cache
4774 the cache was (incorrectly) being used.
4776 10. Added a missing instance of ${EXE} to the exim_install script; this affects
4777 only the Cygwin environment.
4779 11. When return_path_on_delivery was set as a log selector, if different remote
4780 addresses in the same message used different return paths and parallel
4781 remote delivery occurred, the wrong values would sometimes be logged.
4782 (Whenever a remote delivery process finished, the return path value from
4783 the most recently started remote delivery process was logged.)
4785 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
4786 lines when AUTH and/or TLS are in use. This is the "received protocol"
4787 field. Exim used to use "asmtp" for authenticated SMTP, without any
4788 indication (in the protocol name) for TLS use. Now it follows the RFC and
4789 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
4790 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
4791 names appear in log lines as well as in Received: header lines.
4793 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
4794 extensions, and comparison tests, and to fix some bugs.
4796 14. Changes to the "personal" filter test:
4798 (1) The test was buggy in that it was just doing the equivalent of
4799 "contains" tests on header lines. For example, if a user's address was
4800 anne@some.where, the "personal" test would incorrectly be true for
4802 To: susanne@some.where
4804 This test is now done by extracting each address from the header in turn,
4805 and checking the entire address. Other tests that are part of "personal"
4806 are now done using regular expressions (for example, to check local parts
4807 of addresses in From: header lines).
4809 (2) The list of non-personal local parts in From: addresses has been
4810 extended to include "listserv", "majordomo", "*-request", and "owner-*",
4811 taken from the Sieve specification recommendations.
4813 (3) If the message contains any header line starting with "List-" it is
4814 treated as non-personal.
4816 (4) The test for "circular" in the Subject: header line has been removed
4817 because it now seems ill-conceived.
4819 15. Minor typos in src/EDITME comments corrected.
4821 16. Installed latest exipick from John Jetmore.
4823 17. If headers_add on a router specified a text string that was too long for
4824 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
4825 of string_sprintf() is now avoided.
4827 18. $message_body_size was not set (it was always zero) when running the DATA
4828 ACL and the local_scan() function.
4830 19. For the "mail" command in an Exim filter, no default was being set for
4831 the once_repeat time, causing a random time value to be used if "once" was
4832 specified. (If the value happened to be <= 0, no repeat happened.) The
4833 default is now 0s, meaning "never repeat". The "vacation" command was OK
4834 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
4835 (I found it when inspecting the code).
4837 20. There is now an overall timeout for performing a callout verification. It
4838 defaults to 4 times the callout timeout, which applies to individual SMTP
4839 commands during the callout. The overall timeout applies when there is more
4840 than one host that can be tried. The timeout is checked before trying the
4841 next host. This prevents very long delays if there are a large number of
4842 hosts and all are timing out (e.g. when the network connections are timing
4843 out). The value of the overall timeout can be changed by specifying an
4844 additional sub-option for "callout", called "maxwait". For example:
4846 verify = sender/callout=5s,maxwait=20s
4848 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
4849 to the end before writing, but this should make it even safer).
4851 22. Exim was forgetting that it had advertised PIPELINING for the second and
4852 subsequent messages on an SMTP connection. It was also not resetting its
4853 memory on STARTTLS and an internal HELO.
4855 23. When Exim logs an SMTP synchronization error within a session, it now
4856 records whether PIPELINING has been advertised or not.
4858 24. Added 3 instances of "(long int)" casts to time_t variables that were being
4859 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
4860 rather than long int.
4862 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
4864 26. Added the never_mail option to autoreply.
4870 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
4871 crash if the getsockname() call failed; this can happen if a connection is
4872 closed very soon after it is established. The problem was simply in the
4873 order in which certain operations were done, causing Exim to try to write
4874 to the SMTP stream before it had set up the file descriptor. The bug has
4875 been fixed by making things happen in the correct order.
4881 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
4882 before the connection was closed, thus losing the rejection response.
4884 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
4885 some early Solaris releases, but causes trouble in current releases where
4886 socklen_t is defined.
4888 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
4891 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
4894 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
4895 that releasing the top of it at the end releases what was used for sub-
4896 expansions (unless the block got too big). However, discard this block if
4897 the first thing is a variable or header, so that we can use its block when
4898 it is dynamic (useful for very large $message_headers, for example).
4900 6. Lookups now cache *every* query, not just the most recent. A new, separate
4901 store pool is used for this. It can be recovered when all lookup caches are
4902 flushed. Lookups now release memory at the end of their result strings.
4903 This has involved some general refactoring of the lookup sources.
4905 7. Some code has been added to the store_xxx() functions to reduce the amount
4906 of flapping under certain conditions.
4908 8. log_incoming_interface used to affect only the <= reception log lines. Now
4909 it causes the local interface and port to be added to several more SMTP log
4910 lines, for example "SMTP connection from", and rejection lines.
4912 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
4914 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
4916 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
4917 could be overwritten at the end of the current message (or the start of a
4918 new message if it was set in a HELO ACL). The value is now preserved for
4919 the duration of the SMTP connection.
4921 12. If a transport had a headers_rewrite setting, and a matching header line
4922 contained an unqualified address, that address was qualified, even if it
4923 did not match any rewriting rules. The underlying bug was that the values
4924 of the flags that permit the existence of unqualified sender and recipient
4925 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
4926 non-local messages, and by -bnq for local messages) were not being
4927 preserved with the message after it was received.
4929 13. When Exim was logging an SMTP synchronization error, it could sometimes log
4930 "next input=" as part of the text comprising the host identity instead of
4931 the correct text. The code was using the same buffer for two different
4932 strings. However, depending on which order the printing function evaluated
4933 its arguments, the bug did not always show up. Under Linux, for example, my
4934 test suite worked just fine.
4936 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
4937 doesn't work with some older versions of Perl. It has been changed to "my",
4938 which in any case is probably the better facility to use.
4940 15. A really picky compiler found some instances of statements for creating
4941 error messages that either had too many or two few arguments for the format
4944 16. The size of the buffer for calls to the DNS resolver has been increased
4945 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
4946 for addresses that have a lot of PTR records. This alleviates a problem; it
4947 does not fully solve it.
4949 17. A dnsdb lookup for PTR records that receives more data than will fit in the
4950 buffer now truncates the list and logs the incident, which is the same
4951 action as happens when Exim is looking up a host name and its aliases.
4952 Previously in this situation something unpredictable would happen;
4953 sometimes it was "internal error: store_reset failed".
4955 18. If a server dropped the connection unexpectedly when an Exim client was
4956 using GnuTLS and trying to read a response, the client delivery process
4957 crashed while trying to generate an error log message.
4959 19. If a "warn" verb in an ACL added multiple headers to a message in a single
4960 string, for example:
4962 warn message = H1: something\nH2: something
4964 the text was added as a single header line from Exim's point of view
4965 though it ended up OK in the delivered message. However, searching for the
4966 second and subsequent header lines using $h_h2: did not work. This has been
4967 fixed. Similarly, if a system filter added multiple headers in this way,
4968 the routers could not see them.
4970 20. Expanded the error message when iplsearch is called with an invalid key to
4971 suggest using net-iplsearch in a host list.
4973 21. When running tests using -bh, any delays imposed by "delay" modifiers in
4974 ACLs are no longer actually imposed (and a message to that effect is
4977 22. If a "gecos" field in a passwd entry contained escaped characters, in
4978 particular, if it contained a \" sequence, Exim got it wrong when building
4979 a From: or a Sender: header from that name. A second bug also caused
4980 incorrect handling when an unquoted " was present following a character
4981 that needed quoting.
4983 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
4984 was not being matched caselessly.
4986 24. Arranged for all hyphens in the exim.8 source to be escaped with
4989 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
4990 properly. Recipient callout cache records were still being keyed to include
4991 the sender, even when use_sender was set false. This led to far more
4992 callouts that were necessary. The sender is no longer included in the key
4993 when use_sender is false.
4995 26. Added "control = submission" modifier to ACLs.
4997 27. Added the ${base62d: operator to decode base 62 numbers.
4999 28. dnsdb lookups can now access SRV records.
5001 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
5002 the configuration file.
5004 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
5005 (-v) mode. This makes the output for a verbose queue run more intelligible.
5007 31. Added a use_postmaster feature to recipient callouts.
5009 32. Added the $body_zerocount variable, containing the number of binary zero
5010 bytes in the message body.
5012 33. The time of last modification of the "new" subdirectory is now used as the
5013 "mailbox time last read" when there is a quota error for a maildir
5016 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
5018 35. Added +ignore_unknown as a special item in host lists.
5020 36. Code for decoding IPv6 addresses in host lists is now included, even if
5021 IPv6 support is not being compiled. This fixes a bug in which an IPv6
5022 address was recognized as an IP address, but was then not correctly decoded
5023 into binary, causing unexpected and incorrect effects when compared with
5030 1. Very minor rewording of debugging text in manualroute to say "list of
5031 hosts" instead of "hostlist".
5033 2. If verify=header_syntax was set, and a header line with an unqualified
5034 address (no domain) and a large number of spaces between the end of the
5035 name and the colon was received, the reception process suffered a buffer
5036 overflow, and (when I tested it) crashed. This was caused by some obsolete
5037 code that should have been removed. The fix is to remove it!
5039 3. When running in the test harness, delay a bit after writing a bounce
5040 message to get a bit more predictability in the log output.
5042 4. Added a call to search_tidyup() just before forking a reception process. In
5043 theory, someone could use a lookup in the expansion of smtp_accept_max_
5044 per_host which, without the tidyup, could leave open a database connection.
5046 5. Added the variables $recipient_data and $sender_data which get set from a
5047 lookup success in an ACL "recipients" or "senders" condition, or a router
5048 "senders" option, similar to $domain_data and $local_part_data.
5050 6. Moved the writing of debug_print from before to after the "senders" test
5053 7. Change 4.31/66 (moving the time when the Received: is generated) caused
5054 problems for message scanning, either using a data ACL, or using
5055 local_scan() because the Received: header was not generated till after they
5056 were called (in order to set the time as the time of reception completion).
5057 I have revised the way this works. The header is now generated after the
5058 body is received, but before the ACL or local_scan() are called. After they
5059 are run, the timestamp in the header is updated.
5065 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
5066 before starting a queue runner without re-exec. This happened only when
5067 deliver_drop_privilege was set or when the Exim user was set to root. The
5068 effect of the bug was that timeouts during subsequent deliveries caused
5069 crashes instead of being properly handled. The handler is now left at its
5070 default (and expected) setting.
5072 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
5073 message, again when deliver_drop_privilege is set or Exim is run as root.
5074 The bug described in (1) was not present in this case, but the tidying up
5075 of the other signals was missing. I have made the two cases consistent.
5077 3. The ignore_target_hosts setting on a manualroute router was being ignored
5078 for hosts that were looked up using the /MX notation.
5080 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
5083 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
5084 operated on the sender address. After changing the $sender_address to <>
5085 for the sender address verify, Exim was re-instated it as the original
5086 (before rewriting) address, but remembering that it had rewritten it, so it
5087 wasn't rewriting it again. This bug also had the effect of breaking the
5088 sender address verification caching when the sender address was rewritten.
5090 6. The ignore_target_hosts option was being ignored by the ipliteral router.
5091 This has been changed so that if the ip literal address matches
5092 ignore_target_hosts, the router declines.
5094 7. Added expansion conditions match_domain, match_address, and match_local_
5095 part (NOT match_host).
5097 8. The placeholder for the Received: header didn't have a length field set.
5099 9. Added code to Exim itself and to exim_lock to test for a specific race
5100 condition that could lead to file corruption when using MBX delivery. The
5101 issue is with the lockfile that is created in /tmp. If this file is removed
5102 after a process has opened it but before that process has acquired a lock,
5103 there is the potential for a second process to recreate the file and also
5104 acquire a lock. This could lead to two Exim processes writing to the file
5105 at the same time. The added code performs the same test as UW imapd; it
5106 checks after acquiring the lock that its file descriptor still refers to
5107 the same named file.
5109 10. The buffer for building added header lines was of fixed size, 8192 bytes.
5110 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
5113 11. Added the smtp_active_hostname option. If used, this will typically be made
5114 to depend on the incoming interface address. Because $interface_address is
5115 not set up until the daemon has forked a reception process, error responses
5116 that can happen earlier (such as "too many connections") no longer contain
5119 12. If an expansion in a condition on a "warn" statement fails because a lookup
5120 defers, the "warn" statement is abandoned, and the next ACL statement is
5121 processed. Previously this caused the whole ACL to be aborted.
5123 13. Added the iplsearch lookup type.
5125 14. Added ident_timeout as a log selector.
5127 15. Added tls_certificate_verified as a log selector.
5129 16. Added a global option tls_require_ciphers (compare the smtp transport
5130 option of the same name). This controls incoming TLS connections.
5132 17. I finally figured out how to make tls_require_ciphers do a similar thing
5133 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
5134 before starting the TLS session.
5136 18. Tabs are now shown as \t in -bP output.
5138 19. If the log selector return_path_on_delivery was set, Exim crashed when
5139 bouncing a message because it had too many Received: header lines.
5141 20. If two routers both had headers_remove settings, and the first one included
5142 a superfluous trailing colon, the final name in the first list and the
5143 first name in the second list were incorrectly joined into one item (with a
5144 colon in the middle).
5150 1. Added -C and -D options to the exinext utility, mainly to make it easier
5151 to include in the automated testing, but these could be helpful when
5152 multiple configurations are in use.
5154 2. The exinext utility was not formatting the output nicely when there was
5155 an alternate port involved in the retry record key, nor when there was a
5156 message id as well (for retries that were specific to a specific message
5157 and a specific host). It was also confused by IPv6 addresses, because of
5158 the additional colons they contain. I have fixed the IPv4 problem, and
5159 patched it up to do a reasonable job for IPv6.
5161 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
5162 delivery, the log line now contains "pipelined" if PIPELINING was used.
5164 4. An SMTP transport process used to panic and die if the bind() call to set
5165 an explicit outgoing interface failed. This has been changed; it is now
5166 treated in the same way as a connect() failure.
5168 5. A reference to $sender_host_name in the part of a conditional expansion
5169 that was being skipped was still causing a DNS lookup. This no longer
5172 6. The def: expansion condition was not recognizing references to header lines
5173 that used bh_ and bheader_.
5175 7. Added the _cache feature to named lists.
5177 8. The code for checking quota_filecount in the appendfile transport was
5178 allowing one more file than it should have been.
5180 9. For compatibility with Sendmail, the command line option
5188 and sets the incoming protocol and host name (for trusted callers). The
5189 host name and its colon can be omitted when only the protocol is to be set.
5190 Note the Exim already has two private options, -pd and -ps, that refer to
5191 embedded Perl. It is therefore impossible to set a protocol value of "d" or
5192 "s", but I don't think that's a major issue.
5194 10. A number of refactoring changes to the code, none of which should affect
5197 (a) The number of logging options was getting close to filling up the
5198 32-bit word that was used as a bit map. I have split them into two classes:
5199 those that are passed in the argument to log_write(), and those that are
5200 only ever tested independently outside of that function. These are now in
5201 separate 32-bit words, so there is plenty of room for expansion again.
5202 There is no change in the user interface or the logging behaviour.
5204 (b) When building, for example, log lines, the code previously used a
5205 macro that called string_cat() twice, in order to add two strings. This is
5206 not really sufficiently general. Furthermore, there was one instance where
5207 it was actually wrong because one of the argument was used twice, and in
5208 one call a function was used. (As it happened, calling the function twice
5209 did not affect the overall behaviour.) The macro has been replaced by a
5210 function that can join an arbitrary number of extra strings onto a growing
5213 (c) The code for expansion conditions now uses a table and a binary chop
5214 instead of a serial search (which was left over from when there were very
5215 few conditions). Also, it now recognizes conditions like "pam" even when
5216 the relevant support is not compiled in: a suitably worded error message is
5217 given if an attempt is made to use such a condition.
5219 11. Added ${time_interval:xxxxx}.
5221 12. A bug was causing one of the ddress fields not to be passed back correctly
5222 from remote delivery subprocesses. The field in question was not being
5223 subsequently used, so this caused to problems in practice.
5225 13. Added new log selectors queue_time and deliver_time.
5227 14. Might have fixed a bug in maildirsizefile handling that threw up
5228 "unexpected character" debug warnings, and recalculated the data
5229 unnecessarily. In any case, I expanded the warning message to give more
5232 15. Added the message "Restricted characters in address" to the statements in
5233 the default ACL that block characters like @ and % in local parts.
5235 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
5236 Three changes have been made:
5238 (a) There was a serious bug; a negative response to MAIL caused the whole
5239 recipient domain to be cached as invalid, thereby blocking all messages
5240 to all local parts at the same domain, from all senders. This bug has
5241 been fixed. The domain is no longer cached after a negative response to
5242 MAIL if the sender used is not empty.
5244 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
5247 (c) A new callout option, "use_sender" has been added for people who want
5248 the modified behaviour.
5254 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
5257 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
5258 indeed breaks things for older releases.
5260 3. Added additional logging to the case where there is a problem reading data
5261 from a filter that is running in a subprocess using a pipe, in order to
5262 try to track down a specific problem.
5264 4. Testing facility fudge: when running in the test harness and attempting
5265 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
5266 getting "No route to host". Convert this to a timeout.
5268 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
5271 6. Some OS don't have socklen_t but use size_t instead. This affects the
5272 fifth argument of getsockopt() amongst other things. This is now
5273 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
5274 can be set for individual OS. I have set it for SunOS5, OSF1, and
5275 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
5276 some earlier ones do not.
5278 7. Change 4.30/15 was not doing the test caselessly.
5280 8. The standard form for an IPv6 address literal was being rejected by address
5281 parsing in, for example, MAIL and RCPT commands. An example of this kind of
5282 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5283 this, as well as the form without the "IPv6" on the front (but only when
5284 address literals are enabled, of course).
5286 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
5288 10. Exim crashed if a message with an empty sender address specified by -f
5289 encountered a router with an errors_to setting. This could be provoked only
5290 by a command such as
5294 where an empty string was supplied; "<>" did not hit this bug.
5296 11. Installed PCRE release 4.5.
5298 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
5299 remained set. It is now erased.
5301 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
5302 times from message ids (which are base 36 rather than the normal 62).
5304 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
5305 were being counted as actual protocol errors, and logged if the log
5306 selector +smtp_protocol_error was set. One cannot be perfect in this test,
5307 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
5308 and DATA following a set of rejected RCPTs do not count as protocol errors.
5309 In other words, Exim assumes they were pipelined, though this may not
5310 actually be the case. Of course, in all cases the client gets an
5311 appropriate error code.
5313 15. If a lookup fails in an ACL condition, a message about the failure may
5314 be available; it is used if testing the ACL cannot continue, because most
5315 such messages specify what the cause of the deferral is. However, some
5316 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
5317 that caused an old message to be retained and used if a later statement
5318 caused a defer, replacing the real cause of the deferral.
5320 16. If an IP address had so many PTR records that the DNS lookup buffer
5321 was not large enough to hold them, Exim could crash while trying to process
5322 the truncated data. It now detects and logs this case.
5324 17. Further to 4.21/58, another change has been made: if (and only if) the
5325 first line of a message (the first header line) ends with CRLF, a bare LF
5326 in a subsequent header line has a space inserted after it, so as not to
5327 terminate the header.
5329 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
5330 unnecessarily, used atoi() instead of strtol(), and didn't check the
5331 termination when getting file sizes from file names by regex.
5333 19. Completely re-implemented the support for maildirsize files, in the light
5334 of a number of problems with the previous contributed implementation
5335 (4.30/29). In particular:
5337 . If the quota is zero, the maildirsize file is maintained, but no quota is
5340 . If the maildir directory does not exist, it is created before any attempt
5341 to write a maildirsize file.
5343 . The quota value in the file is just a cache; if the quota is changed in
5344 the transport, the new value overrides.
5346 . A regular expression is available for excluding directories from the
5349 20. The autoreply transport checks the characters in options that define the
5350 message's headers; it allows continued headers, but it was checking with
5351 isspace() after an embedded newline instead of explicitly looking for a
5354 21. If all the "regular" hosts to which an address was routed had passed their
5355 expiry times, and had not reached their retry times, the address was
5356 bounced, even if fallback hosts were defined. Now Exim should go on to try
5359 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
5360 equivalent code in the SMTP transport. Some hosts send humungous responses
5361 to HELO/EHLO, more than 1024 it seems.
5363 23. Refactoring: code in filter.c used (void *) for "any old type" but this
5364 gives compiler warnings in some environments. I've now done it "properly",
5367 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
5368 (because of problems with the built-in one) was declared to return uschar *
5369 instead of char *, causing compiler failure.
5371 25. Fixed a file descriptor leak when processing alias/forward files.
5373 26. Fixed a minor format string issue in dbfn.c.
5375 27. Typo in exim.c: ("dmbnz" for "dbmnz").
5377 28. If a filter file refered to $h_xxx or $message_headers, and the headers
5378 contained RFC 2047 "words", Exim's memory could, under certain conditions,
5381 29. When a sender address is verified, it is cached, to save repeating the test
5382 when there is more than one recipient in a message. However, when the
5383 verification involves a callout, it is possible for different callout
5384 options to be set for different recipients. It is too complicated to keep
5385 track of this in the cache, so now Exim always runs a verification when a
5386 callout is required, relying on the callout cache for the optimization.
5387 The overhead is duplication of the address routing, but this should not be
5390 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
5391 to be verified with callout=postmaster, and the main callout worked but the
5392 postmaster check failed, the verification correctly failed. However, if a
5393 subsequent RCPT command asked for sender verification *without* the
5394 postmaster check, incorrect caching caused this verification also to fail,
5397 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
5398 it was not caching the DNS options (qualify_single, search_parents) that
5399 were used when the lookup failed. A subsequent lookup with different
5400 options therefore always gave the same answer, though there were cases
5401 where it should not have. (Example: a "domains = !$mx_any" option on a
5402 dnslookup router: the "domains" option is always processed without any
5403 widening, but the router might have qualify_single set.) Now Exim uses the
5404 cached value only when the same options are set.
5406 32. Added John Jetmore's "exipick" utility to the distribution.
5408 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
5409 than a timeout (e.g. a certificate is required, and is not provided), an
5410 Exim server now closes the connection immediately. Previously it waited for
5411 the client to close - but if the client is SSL, it seems that they each
5412 wait for each other, leading to a delay before one of them times out.
5414 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
5415 maintained 0.8.x compatibility because I don't think many are using it, and
5416 it is clearly obsolete.
5418 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
5421 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
5422 client certificate was expired. A simple patch fixes this, though I don't
5423 understand the full logic of why the verify callback is called multiple
5426 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
5427 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
5428 which causes problems with some clients (such as the Certicom SSL Plus
5429 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
5430 disables the coutermeasure allowing Eudora to connect."
5432 38. Exim was not checking that a write() to a log file succeeded. This could
5433 lead to Bad Things if a log got too big, in particular if it hit a file
5434 size limit. Exim now panics and dies if it cannot write to a log file, just
5435 as it does if it cannot open a log file.
5437 39. Modified OS/Makefile-Linux so that it now contains
5439 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
5441 The two -D definitions ensure that Exim is compiled with large file
5442 support, which makes it possible to handle log files that are bigger than
5445 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
5446 instance) a domain was checked against a named list that involved a lookup,
5447 causing $domain_data to be set, then another domain was checked against the
5448 same list, then the first domain was re-checked, the value of $domain_data
5449 after the final check could be wrong. In particular, if the second check
5450 failed, it could be set empty. This bug probably also applied to
5453 41. The strip_trailing_dot option was not being applied to the address given
5454 with the -f command-line option.
5456 42. The code for reading a message's header from the spool was incrementing
5457 $received_count, but never initializing it. This meant that the value was
5458 incorrect (doubled) while delivering a message in the same process in which
5459 it was received. In the most common configuration of Exim, this never
5460 happens - a fresh exec is done - but it can happen when
5461 deliver_drop_privilege is set.
5463 43. When Exim logs an SMTP synchronization error - client data sent too soon -
5464 it now includes up to 150 characters of the unexpected data in the log
5467 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
5468 and building data strings. The size of both of these buffers was 10 000
5469 bytes - far larger than anybody would *ever* want, thought I. Needless to
5470 say, somebody hit the limit. I have increased the maximum line length to
5471 20 000 and the maximum data length of concatenated lines to 100 000. I have
5472 also fixed two bugs, because there was no checking on these buffers. Tsk,
5473 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
5474 buffer is too small.
5476 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
5477 lsearch lookups. Now it does.
5479 46. When parsing a route_list item in a manualroute router, a fixed-length
5480 buffer was used for the list of hosts. I made this 1024 bytes long,
5481 thinking that nobody would ever have a list of hosts that long. Wrong.
5482 Somebody had a whole pile of complicated expansion conditions, and the
5483 string was silently truncated, leading to an expansion error. It turns out
5484 that it is easier to change to an unlimited length (owing to other changes
5485 that have happened since this code was originally written) than to build
5486 structure for giving a limitation error. The length of the item that
5487 expands into the list of hosts is now unlimited.
5489 47. The lsearch lookup could not handle data where the length of text line was
5490 more than 4095 characters. Such lines were truncated, leading to shortened
5491 data being returned. It should now handle lines of any length.
5493 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
5494 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
5497 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
5498 replacing strings such as BIN_DIRECTORY in a source file: the replacement
5499 no longer happens in comment lines. A list of replacements is now placed
5500 at the head of all of the source files, except those whose only change is
5501 to replace PERL_COMMAND in the very first #! line.
5503 50. Replaced the slow insertion sort in queue.c, for sorting the list of
5504 messages on the queue, with a bottom-up merge sort, using code contributed
5505 by Michael Haardt. This should make operations like -bp somewhat faster on
5506 large queues. It won't affect queue runners, except when queue_run_in_order
5509 51. Installed eximstats 1.31 in the distribution.
5511 52. Added support for SRV lookups to the dnslookup router.
5513 53. If an ACL referred to $message_body or $message_body_end, the value was not
5514 reset for any messages that followed in the same SMTP session.
5516 54. The store-handling optimization for building very long strings was not
5517 differentiating between the different store pools. I don't think this
5518 actually made any difference in practice, but I've tidied it.
5520 55. While running the routers to verify a sender address, $sender_address
5521 was still set to the sender address. This is wrong, because when routing to
5522 send a bounce to the sender, it would be empty. Therefore, I have changed
5523 it so that, while verifying a sender address, $sender_address is set to <>.
5524 (There is no change to what happens when verifying a recipient address.)
5526 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
5527 target A or AAAA records (if not already returned) without resetting the
5528 qualify_single or search_parents options of the DNS resolver. These are
5529 inappropriate in this case because the targets of MX and SRV records must
5530 be FQDNs. A broken DNS record could cause trouble if it happened to have a
5531 target that, when qualified, matched something in the local domain. These
5532 two options are now turned off when doing these lookups.
5534 57. It seems that at least some releases of Reiserfs (which does not have the
5535 concept of a fixed number of inodes) returns zero and not -1 for the
5536 number of available inodes. This interacted badly with check_spool_inodes,
5537 which assumed that -1 was the "no such thing" setting. What I have done is
5538 to check that the total number of inodes is greater than zero before doing
5539 the test of how many are available.
5541 58. When a "warn" ACL statement has a log_message modifier, the message is
5542 remembered, and not repeated. This is to avoid a lot of repetition when a
5543 message has many recipients that cause the same warning to be written.
5544 Howewer, Exim was preserving the list of already written lines for an
5545 entire SMTP session, which doesn't seem right. The memory is now reset if a
5546 new message is started.
5548 59. The "rewrite" debugging flag was not showing the result of rewriting in the
5549 debugging output unless log_rewrite was also set.
5551 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
5552 of (int)(handle) when we know that handle contains (void *)(-1).
5554 61. The Exim daemon panic-logs an error return when it closes the incoming
5555 connection. However "connection reset by peer" seems to be common, and
5556 isn't really an error worthy of noting specially, so that particular error
5559 62. When Exim is trying to find all the local interfaces, it used to panic and
5560 die if the ioctl to get the interface flags failed. However, it seems that
5561 on at least one OS (Solaris 9) it is possible to have an interface that is
5562 included in the list of interfaces, but for which you get a failure error
5563 for this call. This happens when the interface is not "plumbed" into a
5564 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
5565 failure of the "get flags" call assumes that the interface is down.
5567 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
5568 makes life easier for people who are doing arithmetic on fields extracted
5569 from dates, where you often get leading zeros that should not be
5570 interpreted as octal.
5572 64. Added qualify_domain to the redirect router, to override the global
5575 65. If a pathologically long header line contained very many addresses (the
5576 report of this problem mentioned 10 000) and each of them was rewritten,
5577 Exim could use up a very large amount of memory. (It kept on making new
5578 copies of the header line as it rewrote, and never released the old ones.)
5579 At the expense of a bit more processing, the header rewriting function has
5580 been changed so that it no longer eats memory in this way.
5582 66. The generation of the Received: header has been moved from the time that a
5583 message starts to be received, to the time that it finishes. The timestamp
5584 in the Received: header should now be very close to that of the <= log
5585 line. There are two side-effects of this change:
5587 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
5588 logged header lines no longer include the local Received: line, because
5589 it has not yet been created. The same applies to a copy of the message
5590 that is returned to a non-SMTP sender when a message is rejected.
5592 (b) When a filter file is tested using -bf, no additional Received: header
5593 is added to the test message. After some thought, I decided that this
5596 This change does not affect the value of $received_for. It is still set
5597 after address rewriting, but before local_scan() is called.
5599 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
5601 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
5602 gave an unhelpful panic error message, and a defer error. I have managed to
5603 change this behaviour so that it now rejects any supplied certificate,
5604 which seems right, as the list of acceptable certificates is empty.
5606 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
5607 gave an unhelpful defer error. I have not managed to make this reject any
5608 supplied certificates, but the error message it gives is "no certificate
5609 supplied", which is not helpful.
5611 70. exigrep's output now also includes lines that are not associated with any
5612 message, but which match the given pattern. Implemented by a patch from
5613 Martin Sluka, which also tidied up the Perl a bit.
5615 71. Recipient callout verification, like sender verification, was using <> in
5616 the MAIL FROM command. This isn't really the right thing, since the actual
5617 sender may affect whether the remote host accepts the recipient or not. I
5618 have changed it to use the actual sender in the callout; this means that
5619 the cache record is now keyed on a recipient/sender pair, not just the
5620 recipient address. There doesn't seem to be a real danger of callout loops,
5621 since a callout by the remote host to check the sender would use <>.
5622 [SEE ABOVE: changed after hitting problems.]
5624 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
5625 temporary errors. However, in the case of such a code being given after
5626 the end of a data transmission (i.e. after ".") Exim was failing to write
5627 a retry record for the message. (Yes, there was some broken host that was
5628 actually sending 8xx at this point.)
5630 73. An unknown lookup type in a host list could cause Exim to panic-die when
5631 the list was checked. (An example that provoked this was putting <; in the
5632 middle of a list instead of at the start.) If this happened during a DATA
5633 ACL check, a -D file could be left lying around. This kind of configuration
5634 error no longer causes Exim to die; instead it causes a defer errror. The
5635 incident is still logged to the main and panic logs.
5637 74. Buglet left over from Exim 3 conversion. The message "too many messages
5638 in one connection" was written to the rejectlog but not the mainlog, except
5639 when address rewriting (yes!) was being logged.
5641 75. Added write_rejectlog option.
5643 76. When a system filter was run not as root (that is, when system_filter_user
5644 was set), the values of the $n variables were not being returned to the
5645 main process; thus, they were not subsequently available in the $sn
5648 77. Added +return_path_on_delivery log selector.
5650 78. A connection timeout was being treated differently from recipients deferred
5651 when testing hosts_max_try with a message that was older than the host's
5652 retry timeout. (The host should not be counted, thus allowing all hosts to
5653 be tried at least once before bouncing.) This may have been the cause of an
5654 occasionally reported bug whereby a message would remain on the queue
5655 longer than the retry timeout, but would be bounced if a delivery was
5656 forced. I say "may" because I never totally pinned down the problem;
5657 setting up timeout/retry tests is difficult. See also the next item.
5659 79. The ultimate address timeout was not being applied to errors that involved
5660 a combination of host plus message (for example, a timeout on a MAIL
5661 command). When an address resolved to a number of possible hosts, and they
5662 were not all tried for each delivery (e.g. because of hosts_max_try), a
5663 message could remain on the queue longer than the retry timeout.
5665 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
5668 81. Fixed an obscure SMTP outgoing bug which required at least the following
5669 conditions: (a) there was another message waiting for the same server;
5670 (b) the server returned 5xx to all RCPT commands in the first message so
5671 that the message was not completed; (c) the server dropped the connection
5672 or gave a negative response to the RSET that Exim sends to abort the
5673 transaction. The observed case was a dropped connection after DATA that had
5674 been sent in pipelining mode. That is, the server had advertised PIPELINING
5675 but was not implementing it correctly. The effect of the bug was incorrect
5676 behaviour, such as trying another host, and this could lead to a crash.
5682 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
5683 and daemon.c were passed as pointers to ints; they should have been
5684 pointers to socklen_t variables (which are typically unsigned ints).
5686 2. Some signed/unsigned type warnings in the os.c file for Linux have been
5689 3. Fixed a really odd bug that affected only the testing scheme; patching a
5690 certain fixed string in the binary changed the value of another string that
5691 happened to be identical to the end of the original first string.
5693 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
5694 name", it returns that address as the IP address. On some operating
5695 systems (e.g. Solaris), it also passes back the IP address string as the
5696 "host name". However, on others (e.g. Linux), it passes back an empty
5697 string. Exim wasn't checking for this, and was changing the host name to an
5698 empty string, assuming it had been canonicized.
5700 5. Although rare, it is permitted to have more than one PTR record for a given
5701 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
5702 all the names associated with an address, because they do in Solaris.
5703 However, it seems that they do not in Linux for data that comes from the
5704 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
5705 I found this out when I moved to a new Linux workstation and tried to run
5706 the Exim test suite.
5708 To get round this problem I have changed the code so that it now does its
5709 own call to the DNS to look up PTR records when searching for a host name.
5710 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
5711 addresses that are only in /etc/hosts are still found.
5713 This behaviour is, however, controlled by an option called host_lookup_
5714 order, which defaults to "bydns:byaddr". If people want to use the other
5715 order, or indeed, just use one or the other means of lookup, they can
5716 specify it in this variable.
5718 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
5719 some operating systems, this comes back from gethostbyaddr() as an empty
5720 string, and this is what Exim used to test for. However, it seems that in
5721 other systems, "." is yielded. Exim now tests for this case too.
5723 7. The values of check_spool_space and check_log_space are now held internally
5724 as a number of kilobytes instead of an absolute number of bytes. If a
5725 numbers is specified without 'K' or 'M', it is rounded up to the nearest
5726 kilobyte. This means that much larger values can be stored.
5728 8. Exim monitor: an attempt to get the action menu when not actually pointing
5729 at a message produces an empty menu entitled "No message selected". This
5730 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
5731 no entries in it ("Shell widget menu has zero width and/or height"). So I
5732 have added a single, blank menu entry in this case.
5734 9. Added ${quote_local_part.
5736 10. MIME decoding is now applied to the contents of Subject: header lines when
5739 11. Now that a reference to $sender_host_address automatically causes a reverse
5740 lookup to occur if necessary (4.13/18), there is no need to arrange for a
5741 host lookup before query-style lookups in lists that might use this
5742 variable. This has therefore been abolished, and the "net-" prefix is no
5743 longer necessary for query-style lookups.
5745 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
5746 have been a typo for LFLAGS, so it has been changed.
5748 13. The install script calls Exim with "-C /dev/null" in order to find the
5749 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
5750 to be output. Howeve, since Exim outputs its version number before the
5751 error, it didn't break the script. It just looked ugly. I fixed this by
5752 always allowing "-C /dev/null" if the caller is root.
5754 14. Ignore overlarge ACL variable number when reading spool file - insurance
5755 against a later release with more variables having written the file.
5757 15. The standard form for an IPv6 address literal was being rejected by EHLO.
5758 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5759 this, as well as the form without the "IPv6" on the front.
5761 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
5762 OS/Makefile-Darwin file.
5764 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
5765 only to LDAP libraries that do not have LDAP_OPT_DEREF.
5767 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
5768 variable. However, some OS use ints for inodes. I've added cast to long int
5769 to get rid of the compiler warning.
5771 19. I had forgotten to lock out "/../" in configuration file names when
5772 ALT_CONFIG_PREFIX was set.
5774 20. Routers used for verification do not need to specify transports. However,
5775 if such a router generated a host list, and callout was configured, Exim
5776 crashed, because it could not find a port number from the (non-existent)
5777 transport. It now assumes port 25 in this circumstance.
5779 21. Added the -t option to exigrep.
5781 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
5782 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
5783 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
5784 them is set, LOOKUP_LSEARCH is forced.
5786 23. "exim -bV" now outputs a list of lookups that are included in the binary.
5788 24. Added sender and host information to the "rejected by local_scan()" log
5789 line; previously there was no indication of these.
5791 25. Added .include_if_exists.
5793 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
5794 Linux. It turns out that not all file systems support this. Apparently some
5795 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
5796 do it.) To cope with this, the error EINVAL, which means that sync-ing is
5797 not supported on the file descriptor, is now ignored when Exim is trying to
5798 sync a directory. This applies only to Linux.
5800 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
5802 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
5803 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
5806 29. Added support for maildirsize files from supplied patch (modified a bit).
5808 30. The use of :fail: followed by an empty string could lead Exim to respond to
5809 sender verification failures with (e.g.):
5811 550 Verification failed for <xxx>
5812 550 Sender verify failed
5814 where the first response line was missing the '-' that indicates it is not
5815 the final line of the response.
5817 31. The loop for finding the name of the user that called Exim had a hardwired
5818 limit of 10; it now uses the value of finduser_retries, which is used for
5819 all other user lookups.
5821 32. Added $received_count variable, available in data and not_smtp ACLs, and at
5824 33. Exim was neglecting to zero errno before one call of strtol() when
5825 expanding a string and expecting an integer value. On some systems this
5826 resulted in spurious "integer overflow" errors. Also, it was casting the
5827 result into an int without checking.
5829 34. Testing for a connection timeout using "timeout_connect" in the retry rules
5830 did not work. The code looks as if it has *never* worked, though it appears
5831 to have been documented since at least releast 1.62. I have made it work.
5833 35. The "timeout_DNS" error in retry rules, also documented since at least
5834 1.62, also never worked. As it isn't clear exactly what this means, and
5835 clearly it isn't a major issue, I have abolished the feature by treating it
5836 as "timeout", and writing a warning to the main and panic logs.
5838 36. The display of retry rules for -brt wasn't always showing the error code
5841 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
5842 timeout_connect_A, timeout_connect_MX.
5844 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
5845 to the empty sender.
5847 39. The daemon was not analysing the content of -oX till after it had closed
5848 stderr and disconnected from the controlling terminal. This meant that any
5849 syntax errors were only noted on the panic log, and the return code from
5850 the command was 0. By re-arranging the code a little, I've made the
5851 decoding happen first, so such errors now appear on stderr, and the return
5852 code is 1. However, the actual setting up of the sockets still happens in
5853 the disconnected process, so errors there are still only recorded on the
5856 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
5857 connections (as happens on some IP stacks) was logged at start up time as
5858 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
5859 it from "IPv6 and IPv4", which means that two separate sockets are being
5862 41. The debug output for gethostbyname2() or getipnodebyname() failures now
5863 says whether AF_INET or AF_INET6 was passed as an argument.
5865 42. Exiwhat output was messed up when time zones were included in log
5868 43. Exiwhat now gives more information about the daemon's listening ports,
5869 and whether -tls-on-connect was used.
5871 44. The "port" option of the smtp transport is now expanded.
5873 45. A "message" modifier in a "warn" statement in a non-message ACL was being
5874 silently ignored. Now an error message is written to the main and panic
5877 46. There's a new ACL modifier called "logwrite" which writes to a log file
5878 as soon as it is encountered.
5880 47. Added $local_user_uid and $local_user_gid at routing time.
5882 48. Exim crashed when trying to verify a sender address that was being
5885 49. Exim was recognizing only a space character after ".include". It now also
5886 recognizes a tab character.
5888 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
5889 extracting the relevant information from the specification. The man page no
5890 longer contains scrambled data for the -d option, and I've added a section
5891 at the front about calling Exim under different names.
5893 51. Added "extra_headers" argument to the "mail" command in filter files.
5895 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
5898 53. Installed eximstats 1.29.
5900 54. Added transport_filter_timeout as a generic transport option.
5902 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
5903 Cc: header lines. This was required by RFC 822, but it not required by RFC
5906 56. Exim used to add From:, Date:, and Message-Id: header lines to any
5907 incoming messages that did not have them. Now it does so only if the
5908 message originates locally, that is, if there is no associated remote host
5909 address. When Resent- header lines are present, this applies to the Resent-
5910 lines rather than the non-Resent- lines.
5912 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
5913 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
5915 58. Messages for configuration errors now include the name of the main
5916 configuration file - useful now that there may be more than one file in a
5917 list (.included file names were always shown).
5919 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
5920 for those rare installations that do not start the daemon as root or run it
5921 setuid root. I've cut out the call to initgroups() if the daemon is not
5924 60. The Exim user and group can now be bound into the binary as text strings
5925 that are looked up at the start of Exim's processing.
5927 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
5929 62. Added $mailstore_basename variable.
5931 63. Installed patch to sieve.c from Michael Haardt.
5933 64. When Exim failed to open the panic log after failing to open the main log,
5934 the original message it was trying to log was written to stderr and debug
5935 output, but if they were not available (the usual case in production), it
5936 was lost. Now it is written to syslog before the two lines that record the
5937 failures to open the logs.
5939 65. Users' Exim filters run in subprocesses under the user's uid. It is
5940 possible for a "deliver" command or an alias in a "personal" command to
5941 provoke an address rewrite. If logging of address rewriting is configured,
5942 this fails because the process is not running as root or exim. There may be
5943 a better way of dealing with this, but for the moment (because 4.30 needs
5944 to be released), I have disabled address rewrite logging when running a
5945 filter in a non-root, non-exim process.
5951 1. The buildconfig auxiliary program wasn't quoting the value set for
5952 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
5953 not defined. This bug was masked in 4.22 by the effect that was fixed in
5956 2. Some messages that were rejected after a message id was allocated were
5957 shown as "incomplete" by exigrep. It no longer does this for messages that
5958 are rejected by local_scan() or the DATA or non-SMTP ACLs.
5960 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
5961 have allow_domain_literals set, the ID did not get logged in the <= line.
5962 Domain literals are now always recognized in Message-ID: header lines.
5964 4. The first argument for a ${extract expansion item is the key name or field
5965 number. Leading and trailing spaces in this item were not being ignored,
5966 causing some misleading effects.
5968 5. When deliver_drop_privilege was set, single queue runner processes started
5969 manually (i.e. by the command "exim -q") or by the daemon (which uses the
5970 same command in the process it spins off) were not dropping privilege.
5972 6. When the daemon running as "exim" started a queue runner, it always
5973 re-executed Exim in the spun-off process. This is a waste of effort when
5974 deliver_drop_privilege is set. The new process now just calls the
5975 queue-runner function directly.
5981 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
5984 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
5985 ignored. Though the use of -oP was forcing the writing of a pid file, it
5986 was always written to the default place.
5988 3. If the message "no IP address found for host xxxx" is generated during
5989 incoming verification, it is now followed by identification of the incoming
5990 connection (so you can more easily find what provoked it).
5992 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
5994 5. Added some features to "harden" Exim a bit more against certain attacks:
5996 (a) There is now a build-time option called FIXED_NEVER_USERS that can
5997 be put in Local/Makefile. This is like the never_users runtime option,
5998 but it cannot be overridden. The default setting is "root".
6000 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
6001 prefix string with which any file named in a -C command line option
6004 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
6005 is retained for -C and -D only if the caller of Exim is root. Without
6006 it, the exim user may also use -C and -D and retain privilege.
6008 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
6009 command line option is disabled.
6011 6. Macro names set by the -D option must start with an upper case letter, just
6012 like macro names defined in the configuration file.
6014 7. Added "dereference=" facility to LDAP.
6016 8. Two instances of the typo "uknown" in the source files are fixed.
6018 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
6019 the Configure-Makefile script screwed up while processing it.
6021 10. Incorporated PCRE 4.4.
6023 11. The SMTP synchronization check was not operating right at the start of an
6024 SMTP session. For example, it could not catch a HELO sent before the client
6025 waited for the greeting. There is now a check for outstanding input at the
6026 point when the greeting is written. Because of the duplex, asynchronous
6027 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
6028 way, but not yet received, when the check is performed.
6030 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
6031 on TCP/IP sockets, because this apparently causes some broken clients to
6034 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
6035 unchanged) from the Cygwin maintainer.
6037 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
6038 was supported instead of testing for mbx. Effectively a typo.
6040 15. The spa authenticator server code was not checking that the input it
6041 received was valid base64.
6043 16. The debug output line for the "set" modifier in ACLs was not showing the
6044 name of the variable that was being set.
6046 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
6048 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
6049 lookup on its own. Something else was needed to trigger the lookup. For
6050 example, a match in host_lookup or the need for a host name in a host list.
6051 Now, if $sender_host_name is referenced and the host name has not yet been
6052 looked up, a lookup is performed. If the lookup fails, the variable remains
6053 empty, and $host_lookup_failed is set to "1".
6055 19. Added "eqi" as a case-independent comparison operator.
6057 20. The saslauthd authentication condition could segfault if neither service
6058 nor realm was specified.
6060 21. If an overflowing value such as "2048M" was set for message_size_limit, the
6061 error message that was logged was misleading, and incoming SMTP
6062 connections were dropped. The message is now more accurate, and temporary
6063 errors are given to SMTP connections.
6065 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
6066 (except RSET) with a 421 error, until QUIT is received. However, it was
6067 failing to send a response to QUIT.
6069 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
6070 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
6071 ACL is now run after the helo_try_verify_hosts code.
6073 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
6074 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
6075 case-independent, so other case variants are also recognized). Apparently
6076 some systems use these upper case variants.
6078 25. If more than two messages were waiting for the same host, and a transport
6079 filter was specified for the transport, Exim sent two messages over the
6080 same TCP/IP connection, and then failed with "socket operation on non-
6081 socket" when it tried to send the third.
6083 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
6085 27. The extern definition of crypt16() in expand.c was not being excluded when
6086 the OS had its own crypt16() function.
6088 28. Added bounce_return_body as a new option, and bounce_return_size_limit
6089 as a preferred synonym for return_size_limit, both as an option and as an
6092 29. Added LIBS=-liconv to OS/Makefile-OSF1.
6094 30. Changed the default configuration ACL to relax the local part checking rule
6095 for addresses that are not in any local domains. For these addresses,
6096 slashes and pipe symbols are allowed within local parts, but the sequence
6097 /../ is explicitly forbidden.
6099 31. SPA server authentication was not clearing the challenge buffer before
6102 32. log_message in a "warn" ACL statement was writing to the reject log as
6103 well as to the main log, which contradicts the documentation and doesn't
6104 seem right (because no rejection is happening). So I have stopped it.
6106 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
6107 However, I am unable to do any testing of this.
6109 34. Fixed an infelicity in the appendfile transport. When checking directories
6110 for a mailbox, to see if any needed to be created, it was accidentally
6111 using path names with one or more superfluous leading slashes; tracing
6112 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
6114 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
6115 contents are added to the log line that is written for every discarded
6116 recipient. (Previously a log_message setting was ignored.)
6118 36. The ${quote: operator now quotes the string if it is empty.
6120 37. The install script runs exim in order to find its version number. If for
6121 some reason other than non-existence or emptiness, which it checks, it
6122 could not run './exim', it was installing it with an empty version number,
6123 i.e. as "exim-". This error state is now caught, and the installation is
6126 38. An argument was missing from the function that creates an error message
6127 when Exim fails to connect to the socket for saslauthd authentication.
6128 This could cause Exim to crash, or give a corrupted message.
6130 39. Added isip, isip4, and isip6 to ${if conditions.
6132 40. The ACL variables $acl_xx are now saved with the message, and can be
6133 accessed later in routers, transports, and filters.
6135 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
6136 strings in the file are not string-expanded.
6138 42. If a MAIL command specified a SIZE value that was too large to fit into an
6139 int variable, the check against message_size_limit failed. Such values are
6140 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
6141 day this will have to be increased, but I don't think I want to be around
6142 when emails are that large.
6149 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
6150 iconv() is not standard in FreeBSD.
6152 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
6153 IPv6 enabled. The observed symptom was a segmentation fault on return from
6154 the function os_common_find_running_interfaces() in src/os.c.
6156 3. In the check_special_case() function in daemon.c I had used "errno" as an
6157 argument name, which causes warnings on some systems. This was basically a
6158 typo, since it was named "eno" in the comments!
6160 4. The code that waits for the clock to tick (at a resolution of some fraction
6161 of a second) so as to ensure message-id uniqueness was always waiting for
6162 at least one whole tick, when it could have waited for less. [This is
6163 almost certainly not relevant at current processor speeds, where it is
6164 unlikely to ever wait at all. But we try to future-proof.]
6166 5. The function that sleeps for a time interval that includes fractions of a
6167 second contained a race. It did not block SIGALRM between setting the
6168 timer, and suspending (a couple of lines later). If the interval was short
6169 and the sigsuspend() was delayed until after it had expired, the suspension
6170 never ended. On busy systems this could lead to processes getting stuck for
6173 6. Some uncommon configurations may cause a lookup to happen in a queue runner
6174 process, before it forks any delivery processes. The open lookup caching
6175 mechanism meant that the open file or database connection was passed into
6176 the delivery process. The problem was that delivery processes always tidy
6177 up cached lookup data. This could cause a problem for the next delivery
6178 process started by the queue runner, because the external queue runner
6179 process does not know about the closure. So the next delivery process
6180 still has data in the lookup cache. In the case of a file lookup, there was
6181 no problem because closing a file descriptor in a subprocess doesn't affect
6182 the parent. However, if the lookup was caching a connection to a database,
6183 the connection was closed, and the second delivery process was likely to
6184 see errors such as "PGSQL: query failed: server closed the connection
6185 unexpectedly". The problem has been fixed by closing all cached lookups
6186 in a queue runner before running a delivery process.
6188 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
6189 seem to have the "const" qualifier which it has on other OS. I've
6192 8. Change 4.21/2 was too strict. It is only if there are two authenticators
6193 *of the same type* (client or server) with the same public name that an
6194 error should be diagnosed.
6196 9. When Exim looked up a host name for an IP address, but failed to find the
6197 original IP address when looking up the host name (a safety check), it
6198 output the message "<ip address> does not match any IP for NULL", which was
6199 confusing, to say the least. The bug was that the host name should have
6200 appeared instead of "NULL".
6202 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
6203 user that is built into the binary, and the -C or -D options is used, root
6204 privilege is dropped before the configuration file is read. In addition,
6205 logging is switched to stderr instead of the normal log files. If the
6206 configuration then re-defines the Exim user, the unprivileged environment
6207 is probably not what is expected, so Exim logs a panic warning message (but
6210 However, if deliver_drop_privilege is set, the unprivileged state may well
6211 be exactly what is intended, so the warning has been cut out in that case,
6212 and Exim is allowed to try to write to its normal log files.
6218 1. smtp_return_error_details was not giving details for temporary sender
6219 or receiver verification errors.
6221 2. Diagnose a configuration error if two authenticators have the same public
6224 3. Exim used not to create the message log file for a message until the first
6225 delivery attempt. This could be confusing when incoming messages were held
6226 for policy or load reasons. The message log file is now created at the time
6227 the message is received, and an initial "Received" line is written to it.
6229 4. The automatically generated man page for command line options had a minor
6230 bug that caused no ill effects; however, a more serious problem was that
6231 the procedure for building the man page automatically didn't always
6232 operate. Consequently, release 4.20 contains an out-of-date version. This
6233 shouldn't happen again.
6235 5. When building Exim with embedded Perl support, the script that builds the
6236 Makefile was calling 'perl' to find its compile-time parameters, ignoring
6237 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
6239 6. The freeze_tell option was not being used for messages that were frozen on
6240 arrival, either by an ACL or by local_scan().
6242 7. Added the smtp_incomplete_transaction log selector.
6244 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
6245 was accepting AUTH without a new EHLO.
6247 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
6248 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
6251 10. Logging of TCP/IP connections (when configured) now happens in the main
6252 daemon process instead of the child process, so that the TCP/IP connection
6253 count is more accurate (but it can never be perfect).
6255 11. The use of "drop" in a nested ACL was not being handled correctly in the
6256 outer ACL. Now, if condition failure induced by the nested "drop" causes
6257 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
6258 or "require"), the connection is dropped.
6260 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
6261 that yield "discard" can now be used with an "accept" or a "discard" verb,
6262 but an error is generated for any others (because I can't see a useful way
6263 to define what should happen).
6265 13. When an ACL is read dynamically from a file (or anywhere else), the lines
6266 are now processed in the same way as lines in the Exim configuration file.
6267 In particular, continuation lines are supported.
6269 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
6271 15. Added -ti meaning -t -i.
6273 16. Check for letters, digits, hyphens, and dots in the names of dnslist
6274 domains, and warn by logging if others are found.
6276 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
6277 returned from GIFCONF when Exim is trying to find the list of interfaces on
6278 a host. The code in os.c has been modified to copy each ifreq to an aligned
6279 structure in all cases.
6281 Also, in some cases, the returned ifreq's were being copied to a 'struct
6282 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
6283 means the last couple of bytes of an IPv6 address could be chopped if the
6284 ifreq contained only a normal sockaddr (14 bytes storage).
6286 18. Named domain lists were not supported in the hosts_treat_as_local option.
6287 An entry such as +xxxx was not recognized, and was treated as a literal
6290 19. Ensure that header lines added by a DATA ACL are included in the reject log
6291 if the ACL subsequently rejects the message.
6293 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
6294 MD5 (which is deprecated).
6296 21. When testing a filter file using -bf, Exim was writing a message when it
6297 took the sender from a "From " line in the message, but it was not doing so
6298 when it took $return_path from a Return-Path: header line. It now does.
6300 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
6301 with a valid header line field name (a series of printing characters
6302 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
6304 23. Changed "disc" in the source to "disk" to conform to the documentation and
6305 the book and for uniformity.
6307 24. Ignore Sendmail's -Ooption=value command line item.
6309 25. When execve() failed while trying to run a command in a pipe transport,
6310 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
6311 could be confused with a return value of 69 from the command itself. This
6312 has been changed to 127, the value the shell returns if it is asked to run
6313 a non-existent command. The wording for the related log line suggests a
6314 non-existent command as the problem.
6316 26. If received_header_text expands to an empty string, do not add a Received:
6317 header line to the message. (Well, it adds a token one on the spool, but
6318 marks it "old" so that it doesn't get used or transmitted.)
6320 27. Installed eximstats 1.28 (addition of -nt option).
6322 28. There was no check for failure on the call to getsockname() in the daemon
6323 code. This can fail if there is a shortage of resources on the system, with
6324 ENOMEM, for example. A temporary error is now given on failure.
6326 29. Contrary to the C standard, it seems that in some environments, the
6327 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
6328 program. Exim now does this explicitly; it affects the formatting of
6329 timestamps using strftime().
6331 30. If exiqsumm was given junk data, it threw up some uninitialized variable
6332 complaints. I've now initialized all the variables, to avoid this.
6334 32. Header lines added by a system filter were not being "seen" during
6335 transport-time rewrites.
6337 33. The info_callback() function passed to OpenSSL is set up with type void
6338 (*)(SSL *, int, int), as described somewhere. However, when calling the
6339 function (actually a macro) that sets it up, the type void(*)() is
6340 expected. I've put in a cast to prevent warnings from picky compilers.
6342 34. If a DNS black list lookup found a CNAME record, but there were no A
6343 records associated with the domain it pointed at, Exim crashed.
6345 35. If a DNS black list lookup returned more than one A record, Exim ignored
6346 all but the first. It now scans all returned addresses if a particular IP
6347 value is being sought. In this situation, the contents of the
6348 $dnslist_value variable are a list of all the addresses, separated by a
6351 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
6352 to accept a host name and all its aliases if the forward lookup for any of
6353 them yielded the IP address of the incoming connection. Now it accepts only
6354 those names whose forward lookup yields the correct IP address. Any other
6355 names are discarded. This closes a loophole whereby a rogue DNS
6356 administrator could create reverse DNS records to break through a
6357 wildcarded host restriction in an ACL.
6359 37. If a user filter or a system filter that ran in a subprocess used any of
6360 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
6361 the wrong values were passed to the pipe command ($thisaddress had the
6362 value of $0, $0 had the value of $1, etc). This bug was introduced by
6363 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
6365 38. Improved the line breaking for long SMTP error messages from ACLs.
6366 Previously, if there was no break point between 40 and 75 characters, Exim
6367 left the rest of the message alone. Two changes have been made: (a) I've
6368 reduced the minimum length to 35 characters; (b) if it can't find a break
6369 point between 35 and 75 characters, it looks ahead and uses the first one
6370 that it finds. This may give the occasional overlong line, but at least the
6371 remaining text gets split now.
6373 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
6374 file descriptors might be low, and that setting 1000 would always raise it.
6375 It turns out that in some environments, the limit is already over 1000 and
6376 that lowering it causes trouble. So now Exim takes care not to decrease it.
6378 40. When delivering a message, the value of $return_path is set to $sender_
6379 address at the start of routing (routers may change the value). By an
6380 oversight, this default was not being set up when an address was tested by
6381 -bt or -bv, which affected the outcome if any router or filter referred to
6384 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
6385 log only when "message" or "log_message" are set. However, if one of the
6386 conditions was an address verification, or a call to a nested ACL, the
6387 messages generated by the underlying test were being passed through. This
6388 no longer happens. The underlying message is available in $acl_verify_
6389 message for both "message" and "log_message" expansions, so it can be
6390 passed through if needed.
6392 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
6393 new expansion $bh_ to give the encoded byte string without charset
6394 translation. Translation happens only if iconv() is available; HAVE_ICONV
6395 indicates this at build time. HEADERS_CHARSET gives the charset to
6396 translate to; headers_charset can change it in the configuration, and
6397 "headers charset" can change it in an individual filter file.
6399 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
6400 that creates RFC 2047 encoded "words" labels them as that charset instead
6401 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
6402 expansion operator; (ii) when Exim creates a From: line for a local
6403 message; (iii) when a header line is rewritten to include a "phrase" part.
6405 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
6406 buggy, causing it to skip the first lines of messages whose message ID
6407 ended in 'D'. This would not have bitten before Exim release 4.14, because
6408 message IDs were unlikely to end in 'D' before then. The effect was to have
6409 incorrect size information for certain domains.
6411 45. #include "config.h" was missing at the start of the crypt16.c module. This
6412 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
6415 46. If there was a timeout during a "random" callout check, Exim treated it as
6416 a failure of the random address, and carried on sending RSET and the real
6417 address. If the delay was just some slowness somewhere, the response to the
6418 original RCPT would be taken as a response to RSET and so on, causing
6419 mayhem of various kinds.
6421 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
6422 when I implemented it. It didn't allow for the fact that some option values
6423 may legitimatetly be negative (e.g. size_addition), and it didn't even do
6424 the right test for positive values.
6426 48. Domain names in DNS records are case-independent. Exim always looks them up
6427 in lower case. Some resolvers return domain names in exactly the case they
6428 appear in the zone file, that is, they may contain uppercase letters. Not
6429 all resolvers do this - some return always lower case. Exim was treating a
6430 change of case by a resolver as a change of domain, similar to a widening
6431 of a domain abbreviation. This triggered its re-routing code and so it was
6432 trying to route what was effectively the same domain again. This normally
6433 caused routing to fail (because the router wouldn't handle the domain
6434 twice). Now Exim checks for this case specially, and just changes the
6435 casing of the domain that it ultimately uses when it transmits the message
6438 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
6441 50. If a filter generated a file delivery with a non-absolute name (possible if
6442 no home directory exists for the router), the forbid_file option was not
6445 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
6446 the existing equality matching.
6448 52. Exim was using ints instead of ino_t variables in some places where it was
6449 dealing with inode numbers.
6451 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
6452 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
6453 called TMPDIR, and if it finds it is different, it changes its value.
6455 54. The smtp_printf() function is now made available to local_scan() so
6456 additional output lines can be written before returning. There is also an
6457 smtp_fflush() function to enable the detection of a dropped connection.
6458 The variables smtp_input and smtp_batched_input are exported to
6461 55. Changed the default runtime configuration: the message "Unknown user"
6462 has been removed from the ACL, and instead placed on the localuser router,
6463 using the cannot_route_message feature. This means that any verification
6464 failures that generate their own messages won't get overridden. Similarly,
6465 the "Unrouteable address" message that was in the ACL for unverifiable
6466 relay addresses has also been removed.
6468 56. Added hosts_avoid_esmtp to the smtp transport.
6470 57. The exicyclog script was not checking for the esoteric option
6471 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
6472 will work only if exicyclog is run under the appropriate euid.
6474 58. Following a discussion on the list, the rules by which Exim recognises line
6475 endings on incoming messages have been changed. The -dropcr and drop_cr
6476 options are now no-ops, retained only for backwards compatibility. The
6477 following line terminators are recognized: LF CRLF CR. However, special
6478 processing applies to CR:
6480 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
6481 nor a local message in the state where . is a terminator.
6483 (ii) If a bare CR is encountered in a header line, an extra space is added
6484 after the line terminator so as not to end the header. The reasoning
6485 behind this is that bare CRs in header lines are most likely either
6486 to be mistakes, or people trying to play silly games.
6488 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
6489 was being incorrectly given as 18 bytes larger than it should have been.
6490 This is a VOB (very old bug).
6492 60. This may never have affected anything current, but just in case it has:
6493 When the local host is found other than at the start of a list of hosts,
6494 the local host, those with the same MX, and any that follow, are discarded.
6495 When the list in question was part of a longer list of hosts, the following
6496 hosts (not currently being processed) were also being discarded. This no
6497 longer happens. I'm not sure if this situation could ever has previously
6500 61. Added the "/MX" feature to lists of hosts in the manualroute and query
6503 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
6504 header. This is something that is recommended in a new Internet Draft, and
6505 is something that is documented as being done by Sendmail. There are two
6506 possible values. For messages generated by the autoreply transport, Exim
6509 Auto-Submitted: auto-replied
6511 whereas for all other generated messages (e.g. bounces) it adds
6513 Auto-Submitted: auto-generated
6515 63. The "personal" condition in filters now includes a test for the
6516 Auto-Submitted: header. If it contains the string "auto-" the message it
6517 not considered personal.
6519 64. Added rcpt_include_affixes as a generic transport option.
6521 65. Added queue_only_override (default true).
6523 66. Added the syslog_duplication option.
6525 67. If what should have been the first header line of a message consisted of
6526 a space followed by a colon, Exim was mis-interpreting it as a header line.
6527 It isn't of course - it is syntactically invalid and should therefore be
6528 treated as the start of the message body. The misbehaviour could have
6529 caused a number of strange effects, including loss of data in subsequent
6530 header lines, and spool format errors.
6532 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
6533 client host had authenticated. This control can now be exercised by an ACL
6534 for more flexibility.
6536 69. By default, callouts do not happen when testing with -bh. There is now a
6537 variant, -bhc, which does actually run the callout code, including
6538 consulting and updating the callout cache.
6540 70. Added support for saslauthd authentication, courtesy of Alexander
6543 71. If statvfs() failed on the spool or log directories while checking their
6544 size for availability, Exim confusingly gave the error "space shortage".
6545 Furthermore, in debugging mode it crashed with a floating point exception.
6546 These checks are done if check_{spool,log}_{space,inodes} are set, and when
6547 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
6548 serious problem, Exim now writes to the main and panic logs when this
6549 happens, with details of the failure. It then refuses to accept the
6550 incoming message, giving the message "spool directory problem" or "log
6551 directory problem" with a 421 code for SMTP messages.
6553 72. When Exim is about to re-exec itself, it ensures that the file descriptors
6554 0, 1, and 2 exist, because some OS complain for execs without them (see
6555 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
6556 descriptors. However, the code omitted to check that the open succeeded,
6557 causing mysterious errors if for some reason the permissions on /dev/null
6558 got screwed. Now Exim writes a message to the main and panic logs, and
6559 bombs out if it can't open /dev/null.
6561 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
6562 interact so that it is all more flexible. It is supposed to remain
6563 backwards compatible. Also added extra_local_interfaces.
6565 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
6566 to bomb out with an assertion failure - to the client this appears as a
6567 connection drop. This problem occurs in the part of the code that was taken
6568 from the Samba project. Fortunately, the assertion is in a very simple
6569 function, so I have fixed this by reproducing the function inline in the
6570 one place where it is called, and arranging for authentication to fail
6571 instead of killing the process with assert().
6573 75. The SPA client code was not working when the server requested OEM rather
6574 than Unicode encoding.
6576 76. Added code to make require_files with a specific uid setting more usable in
6577 the case where statting the file as root fails - usually a non-root-mounted
6578 NFS file system. When this happens and the failure is EACCES, Exim now
6579 forks a subprocess and does the per-uid checking as the relevant uid.
6581 77. Added process_log_path.
6583 78. If log_file_path was not explicitly set, a setting of check_log_space or
6584 check_log_inodes was ignored.
6586 79. If a space check for the spool or log partitions fails, the incident is now
6587 logged. Of course, in the latter case the data may get lost...
6589 80. Added the %p formatting code to string_format() so that it can be used to
6590 print addresses in debug_print(). Adjusted all the address printing in the
6591 debugging in store.c to use %p rather than %d.
6593 81. There was a concern that a line of code in smtp_in.c could overflow a
6594 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
6595 initially expressed, the concern was not well-founded, because trailing
6596 spaces are removed early. However, if the trailing spaces were followed by
6597 a NULL, they did not get removed, so the overflow was possible. Two fixes
6600 (a) I re-wrote the offending code in a cleaner fashion.
6601 (b) If an incoming SMTP command contains a NULL character, it is rejected
6604 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
6605 runs initgroups(), so that if the Exim user is in any additional groups,
6606 they will be used during message reception.
6612 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob