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.
111 JH/29 Fix the connection_reject log selector to apply to the connect ACL.
112 Previously it only applied to the main-section connection policy
115 JH/30 Bug 1897: fix callouts connection fallback from TLS to cleartext.
117 PP/01 Changed default Diffie-Hellman parameters to be Exim-specific, created
118 by me. Added RFC7919 DH primes as an alternative.
120 PP/02 Unbreak build via pkg-config with new hash support when crypto headers
121 are not in the system include path.
126 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
127 and 3.4.4 - once the server is enabled to respond to an OCSP request
128 it does even when not requested, resulting in a stapling non-aware
129 client dropping the TLS connection.
131 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
132 support variable-length bit vectors. No functional change.
134 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
135 The I= interface field on outgoing lines is now after the H= remote
136 host field, same as incoming lines. There is a separate
137 outgoing_interface log selector which allows you to disable the
140 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
141 If not running log_selector +smtp_connection the mainlog would be held
142 open indefinitely after a "too many connections" event, including to a
143 deleted file after a log rotate. Leave the per net connection logging
144 leaving it open for efficiency as that will be quickly detected by the
145 check on the next write.
147 HS/01 Bug 1671: Fix post transport crash.
148 Processing the wait-<transport> messages could crash the delivery
149 process if the message IDs didn't exist for some reason. When
150 using 'split_spool_directory=yes' the construction of the spool
151 file name failed already, exposing the same netto behaviour.
153 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
154 mime_regex ACL conditions.
156 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
157 to DSN fail messages (bounces): remote IP, remote greeting, remote response
158 to HELO, local diagnostic string.
160 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
161 log line to debug. Even when configured with a tls authenticator many
162 client connections are expected to not authenticate in this way, so
163 an authenticate fail is not an error.
165 HS/02 Add the Exim version string to the process info. This way exiwhat
166 gives some more detail about the running daemon.
168 JH/06 Bug 1395: time-limit cacheing of DNS lookups, to the TTL value. This may
169 matter for fast-change records such as DNSBLs.
171 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
172 retry record, even if constant. There may be multiple transports with
173 different interface settings and the retry behaviour needs to be kept
176 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
178 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
180 JH/10 Bug 840: fix log_defer_output option of pipe transport
182 JH/11 Bug 830: use same host for all RCPTS of a message, even under
183 hosts_randomize. This matters a lot when combined with mua_wrapper.
185 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
186 ${quote_pgsql:<string>} operator.
188 JH/13 Bug 1708: avoid misaligned access in cached lookup.
190 JH/14 Change header file name for freeradius-client. Relevant if compiling
191 with Radius support; from the Gentoo tree and checked under Fedora.
193 JH/15 Bug 1712: Introduce $prdr_requested flag variable
195 JH/16 Bug 1714: Permit an empty string as expansion result for transport
196 option transport_filter, meaning no filtering.
198 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
200 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
201 defaults to "*" (all hosts). The variable is now available when not built
202 with TLS, default unset, mainly to enable keeping the testuite sane.
203 If a server certificate is not supplied (via tls_certificate) an error is
204 logged, and clients will find TLS connections fail on startup. Presumably
205 they will retry in-clear.
206 Packagers of Exim are strongly encouraged to create a server certificate
207 at installation time.
209 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
210 with the $config_file variable.
212 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
213 in transport context, after the attempt, and per-recipient. The latter type
214 is per host attempted. The event data is the error message, and the errno
215 information encodes the lookup type (A vs. MX) used for the (first) host,
216 and the trailing two digits of the smtp 4xx reponse.
218 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
219 to write to mainlog (or rejectlog, paniclog) in the window between file
220 creation and permissions/ownership being changed. Particularly affects
221 installations where exicyclog is run as root, rather than exim user;
222 result is that the running daemon panics and dies.
224 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
226 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
227 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
228 "pri" and "weight". Note that the previous implicit priority given by the
229 list order is no longer honoured.
231 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalisation
234 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
235 by defining SUPPORT_SOCKS.
237 JH/26 Move PROXY support from Experimental to mainline, enabled for a build
238 by defining SUPPORT_PROXY. Note that the proxy_required_hosts option
239 is renamed to hosts_proxy, and the proxy_{host,target}_{address,port}.
240 variables are renamed to proxy_{local,external}_{address,port}.
242 JH/27 Move Internationalisation support from Experimental to mainline, enabled
243 for a build by defining SUPPORT_I18N
245 JH/28 Bug 1745: Fix redis lookups to handle (quoted) spaces embedded in parts
246 of the query string, and make ${quote_redis:} do that quoting.
248 JH/29 Move Events support from Experimental to mainline, enabled by default
249 and removable for a build by defining DISABLE_EVENT.
251 JH/30 Updated DANE implementation code to current from Viktor Dukhovni.
253 JH/31 Fix bug with hosts_connection_nolog and named-lists which were wrongly
254 cached by the daemon.
256 JH/32 Move Redis support from Experimental to mainline, enabled for a build
257 by defining LOOKUP_REDIS. The libhiredis library is required.
259 JH/33 Bug 1748: Permit ACL dnslists= condition in non-smtp ACLs if explicit
260 keys are given for lookup.
262 JH/34 Bug 1192: replace the embedded copy of PolarSSL RSA routines in the DKIM
263 support, by using OpenSSL or GnuTLS library ones. This means DKIM is
264 only supported when built with TLS support. The PolarSSL SHA routines
265 are still used when the TLS library is too old for convenient support.
267 JH/35 Require SINGLE_DH_USE by default in OpenSSL (main config option
268 openssl_options), for security. OpenSSL forces this from version 1.1.0
269 server-side so match that on older versions.
271 JH/36 Bug 1778: longstanding bug in memory use by the ${run } expansion: A fresh
272 allocation for $value could be released as the expansion processing
273 concluded, but leaving the global pointer active for it.
275 JH/37 Bug 1769: Permit a VRFY ACL to override the default 252 response,
276 and to use the domains and local_parts ACL conditions.
278 JH/38 Fix cutthrough bug with body lines having a single dot. The dot was
279 incorrectly not doubled on cutthrough transmission, hence seen as a
280 body-termination at the receiving system - resulting in truncated mails.
281 Commonly the sender saw a TCP-level error, and retransmitted the message
282 via the normal store-and-forward channel. This could result in duplicates
283 received - but deduplicating mailstores were liable to retain only the
284 initial truncated version.
286 JH/39 Bug 1781: Fix use of DKIM private-keys having trailing '=' in the base-64.
288 JH/40 Fix crash in queryprogram router when compiled with EXPERIMENTAL_SRS.
290 JH/41 Bug 1792: Fix selection of headers to sign for DKIM: bottom-up. While
291 we're in there, support oversigning also; bug 1309.
293 JH/42 Bug 1796: Fix error logged on a malware scanner connection failure.
295 HS/04 Add support for keep_environment and add_environment options.
297 JH/43 Tidy coding issues detected by gcc --fsanitize=undefined. Some remain;
298 either intentional arithmetic overflow during PRNG, or testing config-
301 JH/44 Bug 1800: The combination of a -bhc commandline option and cutthrough
302 delivery resulted in actual delivery. Cancel cutthrough before DATA
305 JH/45 Fix cutthrough, when connection not opened by verify and target hard-
306 rejects a recipient: pass the reject to the originator.
308 JH/46 Multiple issues raised by Coverity. Some were obvious or plausible bugs.
309 Many were false-positives and ignorable, but it's worth fixing the
312 JH/47 Fix build on HP-UX and older Solaris, which need (un)setenv now also
313 for the new environment-manipulation done at startup. Move the routines
314 from being local to tls.c to being global via the os.c file.
316 JH/48 Bug 1807: Fix ${extract } for the numeric/3-string case. While preparsing
317 an extract embedded as result-arg for a map, the first arg for extract
318 is unavailable so we cannot tell if this is a numbered or keyed
319 extraction. Accept either.
325 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
328 JH/02 The smtp transport option "multi_domain" is now expanded.
330 JH/03 The smtp transport now requests PRDR by default, if the server offers
333 JH/04 Certificate name checking on server certificates, when exim is a client,
334 is now done by default. The transport option tls_verify_cert_hostnames
335 can be used to disable this per-host. The build option
336 EXPERIMENTAL_CERTNAMES is withdrawn.
338 JH/05 The value of the tls_verify_certificates smtp transport and main options
339 default to the word "system" to access the system default CA bundle.
340 For GnuTLS, only version 3.0.20 or later.
342 JH/06 Verification of the server certificate for a TLS connection is now tried
343 (but not required) by default. The verification status is now logged by
344 default, for both outbound TLS and client-certificate supplying inbound
347 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
350 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
351 Status Notification (bounce) messages are now MIME format per RFC 3464.
352 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
353 under the control of the dsn_advertise_hosts option, and routers may
354 have a dsn_lasthop option.
356 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
357 default, modifiable by a malware= option. The list separator for
358 the options can now be changed in the usual way. Bug 68.
360 JH/10 The smtp_receive_timeout main option is now expanded before use.
362 JH/11 The incoming_interface log option now also enables logging of the
363 local interface on delivery outgoing connections.
365 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
366 if the interface and destination host and port all match.
368 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
371 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
372 Patch from Andrew Lewis.
374 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
375 now supports optional time-restrictions, weighting, and priority
376 modifiers per server. Patch originally by <rommer@active.by>.
378 JH/16 The spamd_address main option now supports a mixed list of local
379 and remote servers. Remote servers can be IPv6 addresses, and
380 specify a port-range.
382 JH/17 Bug 68: The spamd_address main option now supports an optional
383 timeout value per server.
385 JH/18 Bug 1581: Router and transport options headers_add/remove can
386 now have the list separator specified.
388 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
391 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
394 JH/21 Support for the A6 type of dns record is withdrawn.
396 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
397 rather than the verbs used.
399 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
400 from 255 to 1024 chars.
402 JH/24 Verification callouts now attempt to use TLS by default.
404 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
405 are generic router options now. The defaults didn't change.
407 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
408 Original patch from Alexander Shikoff, worked over by JH.
410 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
411 files if ZCAT_COMMAND is not executable.
413 JH/26 Bug 1539: Add timout/retry options on dnsdb lookups.
415 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
417 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
418 Normally benign, it bites when the pair was led to by a CNAME;
419 modern usage is to not canoicalize the domain to a CNAME target
420 (and we were inconsistent anyway for A-only vs AAAA+A).
422 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
424 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
425 when evaluating $sender_host_dnssec.
427 JH/31 Check the HELO verification lookup for DNSSEC, adding new
428 $sender_helo_dnssec variable.
430 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
432 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
434 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
436 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
437 documented as working, but never had. Support all but $spam_report.
439 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
440 added for tls authenticator.
442 HS/03 Add perl_taintmode main config option
447 TL/01 When running the test suite, the README says that variables such as
448 no_msglog_check are global and can be placed anywhere in a specific
449 test's script, however it was observed that placement needed to be near
450 the beginning for it to behave that way. Changed the runtest perl
451 script to read through the entire script once to detect and set these
452 variables, reset to the beginning of the script, and then run through
453 the script parsing/test process like normal.
455 TL/02 The BSD's have an arc4random API. One of the functions to induce
456 adding randomness was arc4random_stir(), but it has been removed in
457 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
458 function when detected.
460 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
461 cause callback expansion.
463 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
464 syntax errors in an expansion can be treated as a string instead of
465 logging or causing an error, due to the internal use of bool_lax
466 instead of bool when processing it.
468 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
469 server certificates when making smtp deliveries.
471 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
473 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
475 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
477 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
478 Merged patch from Sebastian Wiedenroth.
480 JH/05 Fix results-pipe from transport process. Several recipients, combined
481 with certificate use, exposed issues where response data items split
482 over buffer boundaries were not parsed properly. This eventually
483 resulted in duplicates being sent. This issue only became common enough
484 to notice due to the introduction of conection certificate information,
485 the item size being so much larger. Found and fixed by Wolfgang Breyha.
487 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
488 size buffer was used, resulting in syntax errors when an expansion
491 JH/07 Add support for directories of certificates when compiled with a GnuTLS
492 version 3.3.6 or later.
494 JH/08 Rename the TPDA expermimental facility to Event Actions. The #ifdef
495 is EXPERIMENTAL_EVENT, the main-configuration and transport options
496 both become "event_action", the variables become $event_name, $event_data
497 and $event_defer_errno. There is a new variable $verify_mode, usable in
498 routers, transports and related events. The tls:cert event is now also
499 raised for inbound connections, if the main configuration event_action
502 TL/06 In test suite, disable OCSP for old versions of openssl which contained
503 early OCSP support, but no stapling (appears to be less than 1.0.0).
505 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
506 server certificate names available under the smtp transport option
507 "tls_verify_cert_hostname" now do not permit multi-component wildcard
510 JH/10 Time-related extraction expansions from certificates now use the main
511 option "timezone" setting for output formatting, and are consistent
512 between OpenSSL and GnuTLS compilations. Bug 1541.
514 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
515 encoded parameter in the incoming message. Bug 1558.
517 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
518 include certificate info, eximon was claiming there were spoolfile
521 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
523 JH/14 Log delivery-related information more consistently, using the sequence
524 "H=<name> [<ip>]" wherever possible.
526 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
527 are problematic for Debian distribution, omit them from the release
530 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
532 JH/16 Fix string representation of time values on 64bit time_t anchitectures.
535 JH/17 Fix a null-indirection in certextract expansions when a nondefault
536 output list separator was used.
541 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
542 checkers that were complaining about end of non-void function with no
545 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
546 This was a regression intruduced in 4.83 by another bugfix.
548 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
550 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
551 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
557 TF/01 Correctly close the server side of TLS when forking for delivery.
559 When a message was received over SMTP with TLS, Exim failed to clear up
560 the incoming connection properly after forking off the child process to
561 deliver the message. In some situations the subsequent outgoing
562 delivery connection happened to have the same fd number as the incoming
563 connection previously had. Exim would try to use TLS and fail, logging
564 a "Bad file descriptor" error.
566 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
567 utilities have not been installed.
569 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
570 temporary space as the ACL may create new global variables.
572 TL/01 LDAP support uses per connection or global context settings, depending
573 upon the detected version of the libraries at build time.
575 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
576 to extract and use the src ip:port in logging and expansions as if it
577 were a direct connection from the outside internet. PPv2 support was
578 updated based on HAProxy spec change in May 2014.
580 JH/02 Add ${listextract {number}{list}{success}{fail}}.
582 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
583 Properly escape header and check for NULL return.
585 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
588 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
590 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
591 characters in header names, implemented as a verify condition.
592 Contributed by Michael Fischer v. Mollard.
594 TL/05 Rename SPF condition results err_perm and err_temp to standardized
595 results permerror and temperror. Previous values are deprecated but
596 still accepted. In a future release, err_perm and err_temp will be
597 completely removed, which will be a backward incompatibility if the
598 ACL tests for either of these two old results. Patch contributed by
599 user bes-internal on the mailing list.
601 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
603 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
604 selectors, in both main and reject logs.
606 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
609 JH/07 Add malware type "sock" for talking to simple daemon.
611 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
613 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
614 routers/transports under cutthrough routing.
616 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
617 numbers. Touch up "bool" conditional to keep the same definition.
619 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
621 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
623 JH/12 Expand items in router/transport headers_add or headers_remove lists
624 individually rather than the list as a whole. Bug 1452.
626 Required for reasonable handling of multiple headers_ options when
627 they may be empty; requires that headers_remove items with embedded
628 colons must have them doubled (or the list-separator changed).
630 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
631 view the policy declared in the DMARC record. Currently, $dmarc_status
632 is a combined value of both the record presence and the result of the
635 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
637 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
638 dnslookup router and the smtp transport (applying to the forward
641 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
642 of ldap servers used for a specific lookup. Patch provided by Heiko
645 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
646 New variable $lookup_dnssec_authenticated for observability.
648 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
649 Patch submitted by Lars Timman.
651 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
653 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
654 Requires trusted mode and valid format message id, aborts otherwise.
655 Patch contributed by Heiko Schlichting.
657 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
658 certextract with support for various fields. Bug 1358.
660 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
661 is requested by default, modifiable by smtp transport option
664 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
665 operate on certificate variables to give certificate fingerprints
666 Also new ${sha256:cert_variable}.
668 JH/23 The PRDR feature is moved from being Experimental into the mainline.
670 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
671 Christian Aistleitner.
673 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
675 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
676 file. Patch from Wolfgang Breyha.
678 JH/25 Expand the coverage of the delivery $host and $host_address to
679 client authenticators run in verify callout. Bug 1476.
681 JH/26 Port service names are now accepted for tls_on_connect_ports, to
682 align with daemon_smtp_ports. Bug 72.
684 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
685 support and error reporting did not work properly.
687 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
688 and is readable. Patch from Andrew Colin Kissa.
690 TL/14 Enhance documentation of ${run expansion and how it parses the
691 commandline after expansion, particularly in the case when an
692 unquoted variable expansion results in an empty value.
694 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
696 PP/02 Fix internal collision of T_APL on systems which support RFC3123
697 by renaming away from it. Addresses GH issue 15, reported by
700 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
702 TL/15 SECURITY: prevent double expansion in math comparison functions
703 (can expand unsanitized data). Not remotely exploitable.
710 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
712 PP/02 Make -n do something, by making it not do something.
713 When combined with -bP, the name of an option is not output.
715 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
718 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
719 $sender_host_name and config options to manage this, and basic check
722 PP/05 DSCP support for outbound connections and control modifier for inbound.
724 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
725 (Only plugin which currently uses this is kerberos4, which nobody should
726 be using, but we should make it available and other future plugins might
727 conceivably use it, even though it would break NAT; stuff *should* be
728 using channel bindings instead).
730 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
731 name; added for Sendmail compatibility; requires admin caller.
732 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
733 just ignore it); requires trusted caller.
734 Also parse but ignore: -Ac -Am -X<logfile>
737 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
739 TL/02 Add +smtp_confirmation as a default logging option.
741 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
742 Patch by Magnus Holmgren from 2007-02-20.
744 TL/04 Bugzilla 1281 - Spec typo.
745 Bugzilla 1283 - Spec typo.
746 Bugzilla 1290 - Spec grammar fixes.
748 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
750 TL/06 Add Experimental DMARC support using libopendmarc libraries.
752 TL/07 Fix an out of order global option causing a segfault. Reported to dev
753 mailing list by by Dmitry Isaikin.
755 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
757 JH/02 Support "G" suffix to numbers in ${if comparisons.
759 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
761 NM/01 Bugzilla 1197 - Spec typo
762 Bugzilla 1196 - Spec examples corrections
764 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
766 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
767 gnutls_enable_pkcs11, but renamed to more accurately indicate its
770 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
771 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
773 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
774 "acl {{name}{arg}...}", and optional args on acl condition
777 JH/05 Permit multiple router/transport headers_add/remove lines.
779 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
781 JH/07 Avoid using a waiting database for a single-message-only transport.
782 Performance patch from Paul Fisher. Bugzilla 1262.
784 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
787 JH/09 Add $headers_added variable, with content from use of ACL modifier
788 add_header (but not yet added to the message). Bugzilla 199.
790 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
791 Pulled from Bugzilla 817 by Wolfgang Breyha.
793 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
795 (nb: this is the same fix as in Exim 4.80.1)
797 JH/11 Add A= logging on delivery lines, and a client_set_id option on
800 JH/12 Add optional authenticated_sender logging to A= and a log_selector
803 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
805 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
806 advertise SMTP AUTH mechanism to us, instead of a generic
807 protocol violation error. Also, make Exim more robust to bad
808 data from the Dovecot auth socket.
810 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
812 When a queue runner is handling a message, Exim first routes the
813 recipient addresses, during which it prunes them based on the retry
814 hints database. After that it attempts to deliver the message to
815 any remaining recipients. It then updates the hints database using
818 So if a recipient address works intermittently, it can get repeatedly
819 deferred at routing time. The retry hints record remains fresh so the
820 address never reaches the final cutoff time.
822 This is a fairly common occurrence when a user is bumping up against
823 their storage quota. Exim had some logic in its local delivery code
824 to deal with this. However it did not apply to per-recipient defers
825 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
827 This change adds a proper retry rule check during routing so that the
828 final cutoff time is checked against the message's age. We only do
829 this check if there is an address retry record and there is not a
830 domain retry record; this implies that previous attempts to handle
831 the address had the retry_use_local_parts option turned on. We use
832 this as an approximation for the destination being like a local
833 delivery, as in LMTP.
835 I suspect this new check makes the old local delivery cutoff check
836 redundant, but I have not verified this so I left the code in place.
838 TF/02 Correct gecos expansion when From: is a prefix of the username.
840 Test 0254 submits a message to Exim with the header
844 When I ran the test suite under the user fanf2, Exim expanded
845 the header to contain my full name, whereas it should have added
846 a Resent-Sender: header. It erroneously treats any prefix of the
847 username as equal to the username.
849 This change corrects that bug.
851 GF/01 DCC debug and logging tidyup
852 Error conditions log to paniclog rather than rejectlog.
853 Debug lines prefixed by "DCC: " to remove any ambiguity.
855 TF/03 Avoid unnecessary rebuilds of lookup-related code.
857 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
858 Bug spotted by Jeremy Harris; was flawed since initial commit.
859 Would have resulted in OCSP responses post-SNI triggering an Exim
860 NULL dereference and crash.
862 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
864 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
865 Bug detection, analysis and fix by Samuel Thibault.
866 Bugzilla 1331, Debian bug #698092.
868 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
870 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
871 Server implementation by Todd Lyons, client by JH.
872 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
873 config variable "prdr_enable" controls whether the server
874 advertises the facility. If the client requests PRDR a new
875 acl_data_smtp_prdr ACL is called once for each recipient, after
876 the body content is received and before the acl_smtp_data ACL.
877 The client is controlled by bolth of: a hosts_try_prdr option
878 on the smtp transport, and the server advertisement.
879 Default client logging of deliveries and rejections involving
880 PRDR are flagged with the string "PRDR".
882 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
883 fclose(). Diagnosis by Todd Lyons.
885 PP/17 Update configure.default to handle IPv6 localhost better.
886 Patch by Alain Williams (plus minor tweaks).
889 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
890 This is now consistent with GnuTLS, and is now documented: the
891 previous undocumented portable approach to treating the option as
892 unset was to force an expansion failure. That still works, and
893 an empty string is now equivalent.
895 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
896 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
897 not performing validation itself.
899 PP/20 Added force_command boolean option to pipe transport.
900 Patch from Nick Koston, of cPanel Inc.
902 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
905 TF/04 Added udpsend ACL modifer and hexquote expansion operator
907 PP/21 Fix eximon continuous updating with timestamped log-files.
908 Broken in a format-string cleanup in 4.80, missed when I repaired the
909 other false fix of the same issue.
910 Report and fix from Heiko Schlichting.
913 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
914 Report from Prashanth Katuri.
916 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
917 It's SecureTransport, so affects any MacOS clients which use the
918 system-integrated TLS libraries, including email clients.
920 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
921 using a MIME ACL for non-SMTP local injection.
922 Report and assistance in diagnosis by Warren Baker.
924 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
926 JH/16 Fix comparisons for 64b. Bugzilla 1385.
928 TL/09 Add expansion variable $authenticated_fail_id to keep track of
929 last id that failed so it may be referenced in subsequent ACL's.
931 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
934 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
935 ldap library initialization, allowing self-signed CA's to be
936 used. Also properly sets require_cert option later in code by
937 using NULL (global ldap config) instead of ldap handle (per
938 session). Bug diagnosis and testing by alxgomz.
940 TL/12 Enhanced documentation in the ratelimit.pl script provided in
941 the src/util/ subdirectory.
943 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
944 renamed to Transport Post Delivery Action by Jeremy Harris, as
947 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
948 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
949 redis_servers = needs to be configured which will be used by the redis
950 lookup. Patch from Warren Baker, of The Packet Hub.
952 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
954 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
955 hostname or reverse DNS when processing a host list. Used suggestions
956 from multiple comments on this bug.
958 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
960 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
961 Missed a few lines, added it to make the runtest require no keyboard
964 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
965 contains upper case chars. Make router use caseful_local_part.
967 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
968 support when GnuTLS has been built with p11-kit.
974 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
976 This, or similar/improved, will also be change PP/11 of 4.82.
982 PP/01 Handle short writes when writing local log-files.
983 In practice, only affects FreeBSD (8 onwards).
984 Bugzilla 1053, with thanks to Dmitry Isaikin.
986 NM/01 Bugzilla 949 - Documentation tweak
988 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
991 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
993 PP/02 Implemented gsasl authenticator.
995 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
997 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
998 `pkg-config foo` for cflags/libs.
1000 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
1001 with rest of GSASL and with heimdal_gssapi.
1003 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
1004 `pkg-config foo` for cflags/libs for the TLS implementation.
1006 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
1007 properties get this fed in as external SSF. A number of robustness
1008 and debugging improvements to the cyrus_sasl authenticator.
1010 PP/08 cyrus_sasl server now expands the server_realm option.
1012 PP/09 Bugzilla 1214 - Log authentication information in reject log.
1013 Patch by Jeremy Harris.
1015 PP/10 Added dbmjz lookup type.
1017 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
1019 PP/12 MAIL args handles TAB as well as SP, for better interop with
1020 non-compliant senders.
1021 Analysis and variant patch by Todd Lyons.
1023 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
1024 Bug report from Lars Müller <lars@samba.org> (via SUSE),
1025 Patch from Dirk Mueller <dmueller@suse.com>
1027 PP/13 tls_peerdn now print-escaped for spool files.
1028 Observed some $tls_peerdn in wild which contained \n, which resulted
1029 in spool file corruption.
1031 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
1032 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
1033 or write after TLS renegotiation, which otherwise led to messages
1036 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
1037 as a tracking header (ie: a signed header comes before the signature).
1038 Patch from Wolfgang Breyha.
1040 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
1041 comma-sep list; embedded commas doubled.
1043 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
1045 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
1047 Report and patch from Dmitry Banschikov.
1049 PP/16 Removed "dont_insert_empty_fragments" fron "openssl_options".
1050 Removed SSL_clear() after SSL_new() which led to protocol negotiation
1051 failures. We appear to now support TLS1.1+ with Exim.
1053 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
1054 lets Exim select keys and certificates based upon TLS SNI from client.
1055 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
1056 before an outbound SMTP session. New log_selector, +tls_sni.
1058 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
1059 NULL dereference. Report and patch from Alun Jones.
1061 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
1062 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
1063 Not seeing resolver debug output on NetBSD, but suspect this is a
1064 resolver implementation change.
1066 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
1067 Left warnings. Added "eximon gdb" invocation mode.
1069 PP/21 Defaulting "accept_8bitmime" to true, not false.
1071 PP/22 Added -bw for inetd wait mode support.
1073 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
1074 locate the relevant includes and libraries. Made this the default.
1076 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
1077 Bugzilla 1246, report and most of solution from Tomasz Kusy.
1079 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
1080 This may cause build issues on older platforms.
1082 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
1083 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
1084 gnutls_require_mac & gnutls_require_protocols (no longer supported).
1085 Added SNI support via GnuTLS too.
1086 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
1088 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
1090 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
1091 Applied second patch from Janne, implementing suggestion to default
1092 multiple-strings-in-record handling to match SPF spec.
1094 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
1096 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
1097 read-only, out of scope).
1098 Patch from Wolfgang Breyha, report from Stuart Northfield.
1100 PP/29 Fix three issues highlighted by clang analyser static analysis.
1101 Only crash-plausible issue would require the Cambridge-specific
1102 iplookup router and a misconfiguration.
1103 Report from Marcin Mirosław.
1105 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
1107 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
1108 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
1109 As part of this, removing so much warning spew let me fix some minor
1110 real issues in debug logging.
1112 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
1113 assignment on my part. Fixed.
1115 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
1116 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
1117 Janne Snabb (who went above and beyond: thank you).
1119 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
1120 string otherwise requires a connection and a bunch more work and it's
1121 relatively easy to get wrong. Should also expose TLS library linkage
1124 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
1125 64-bit ${eval} (JH/03).
1127 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
1128 GNU libc to support some of the 64-bit stuff, should not lead to
1129 conflicts. Defined before os.h is pulled in, so if a given platform
1130 needs to override this, it can.
1132 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
1133 protection layer was required, which is not implemented.
1134 Bugzilla 1254, patch from Wolfgang Breyha.
1136 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
1137 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
1138 tls_dhparam take prime identifiers. Also unbreak combination of
1139 OpenSSL+DH_params+TLSSNI.
1141 PP/39 Disable SSLv2 by default in OpenSSL support.
1147 PP/01 Solaris build fix for Oracle's LDAP libraries.
1148 Bugzilla 1109, patch from Stephen Usher.
1150 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
1152 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
1155 TF/02 Fix a couple more cases where we did not log the error message
1156 when unlink() failed. See also change 4.74-TF/03.
1158 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
1159 lock up or crash if it happened to be inside a call to libc when it
1160 got a SIGUSR1 from exiwhat.
1162 The SIGUSR1 handler appends the current process status to the process
1163 log which is later printed by exiwhat. It used to use the general
1164 purpose logging code to do this, but several functions it calls are
1165 not safe for signals.
1167 The new output code in the SIGUSR1 handler is specific to the process
1168 log, and simple enough that it's easy to inspect for signal safety.
1169 Removing some special cases also simplifies the general logging code.
1170 Removing the spurious timestamps from the process log simplifies
1173 TF/04 Improved ratelimit ACL condition.
1175 The /noupdate option has been deprecated in favour of /readonly which
1176 has clearer semantics. The /leaky, /strict, and /readonly update modes
1177 are mutually exclusive. The update mode is no longer included in the
1178 database key; it just determines when the database is updated. (This
1179 means that when you upgrde Exim will forget old rate measurements.)
1181 Exim now checks that the per_* options are used with an update mode that
1182 makes sense for the current ACL. For example, when Exim is processing a
1183 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
1184 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
1185 must specify per_mail/readonly. If you omit the update mode it defaults to
1186 /leaky where that makes sense (as before) or /readonly where required.
1188 The /noupdate option is now undocumented but still supported for
1189 backwards compatibility. It is equivalent to /readonly except that in
1190 ACLs where /readonly is required you may specify /leaky/noupdate or
1191 /strict/noupdate which are treated the same as /readonly.
1193 A useful new feature is the /count= option. This is a generalization
1194 of the per_byte option, so that you can measure the throughput of other
1195 aggregate values. For example, the per_byte option is now equivalent
1196 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
1198 The per_rcpt option has been generalized using the /count= mechanism
1199 (though it's more complicated than the per_byte equivalence). When it is
1200 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
1201 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
1202 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
1203 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
1204 Note that using per_rcpt with a non-readonly update mode in more than
1205 one ACL will cause the recipients to be double-counted. (The per_mail
1206 and per_byte options don't have this problem.)
1208 The handling of very low rates has changed slightly. If the computed rate
1209 is less than the event's count (usually one) then this event is the first
1210 after a long gap. In this case the rate is set to the same as this event's
1211 count, so that the first message of a spam run is counted properly.
1213 The major new feature is a mechanism for counting the rate of unique
1214 events. The new per_addr option counts the number of different
1215 recipients that someone has sent messages to in the last time period. It
1216 behaves like per_rcpt if all the recipient addresses are different, but
1217 duplicate recipient addresses do not increase the measured rate. Like
1218 the /count= option this is a general mechanism, so the per_addr option
1219 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
1220 example, measure the rate that a client uses different sender addresses
1221 with the options per_mail/unique=$sender_address. There are further
1222 details in the main documentation.
1224 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
1226 TF/06 Removed a few PCRE remnants.
1228 TF/07 Automatically extract Exim's version number from tags in the git
1229 repository when doing development or release builds.
1231 PP/02 Raise smtp_cmd_buffer_size to 16kB.
1232 Bugzilla 879. Patch from Paul Fisher.
1234 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
1235 Heavily based on revision 40f9a89a from Simon Arlott's tree.
1238 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
1240 PP/05 Variable $av_failed, true if the AV scanner deferred.
1241 Bugzilla 1078. Patch from John Horne.
1243 PP/06 Stop make process more reliably on build failure.
1244 Bugzilla 1087. Patch from Heiko Schlittermann.
1246 PP/07 Make maildir_use_size_file an _expandable_ boolean.
1247 Bugzilla 1089. Patch from Heiko Schlittermann.
1249 PP/08 Handle ${run} returning more data than OS pipe buffer size.
1250 Bugzilla 1131. Patch from Holger Weiß.
1252 PP/09 Handle IPv6 addresses with SPF.
1253 Bugzilla 860. Patch from Wolfgang Breyha.
1255 PP/10 GnuTLS: support TLS 1.2 & 1.1.
1257 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
1260 PP/11 match_* no longer expand right-hand-side by default.
1261 New compile-time build option, EXPAND_LISTMATCH_RHS.
1262 New expansion conditions, "inlist", "inlisti".
1264 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
1266 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
1268 PP/14 fix log_write() format string regression from TF/03.
1269 Bugzilla 1152. Patch from Dmitry Isaikin.
1275 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
1277 PP/02 Harmonised TLS library version reporting; only show if debugging.
1278 Layout now matches that introduced for other libraries in 4.74 PP/03.
1280 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
1282 PP/04 New "dns_use_edns0" global option.
1284 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
1287 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
1288 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
1290 TK/01 Updated PolarSSL code to 0.14.2.
1291 Bugzilla 1097. Patch from Andreas Metzler.
1293 PP/07 Catch divide-by-zero in ${eval:...}.
1294 Fixes bugzilla 1102.
1296 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
1299 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
1300 format-string attack -- SECURITY: remote arbitrary code execution.
1302 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
1303 time unintentionally subject to list matching rules, letting the header
1304 cause arbitrary Exim lookups (of items which can occur in lists, *not*
1305 arbitrary string expansion). This allowed for information disclosure.
1307 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
1308 INT_MIN/-1 -- value coerced to INT_MAX.
1314 NM/01 Workround for PCRE version dependency in version reporting
1317 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
1318 This fixes portability to compilers other than gcc, notably
1319 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
1321 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
1322 makefiles for portability to HP-UX and POSIX correctness.
1324 PP/01 Permit LOOKUP_foo enabling on the make command-line.
1325 Also via indented variable definition in the Makefile.
1326 (Debugging by Oliver Heesakkers).
1328 PP/02 Restore caching of spamd results with expanded spamd_address.
1329 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
1331 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
1332 Improves build reliability. Fix from: Frank Elsner
1334 NM/02 Fix wide character breakage in the rfc2047 coding
1335 Fixes bug 1064. Patch from Andrey N. Oktyabrski
1337 NM/03 Allow underscore in dnslist lookups
1338 Fixes bug 1026. Patch from Graeme Fowler
1340 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
1341 Code patches from Adam Ciarcinski of NetBSD.
1343 NM/04 Fixed exiqgrep to cope with mailq missing size issue
1346 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
1347 is logged, to avoid truncation. Patch from John Horne.
1349 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
1350 Patch from Jakob Hirsch.
1352 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
1353 SQL string expansion failure details.
1354 Patch from Andrey Oktyabrski.
1356 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
1357 Patch from Simon Arlott.
1359 PP/09 New lookups functionality failed to compile on old gcc which rejects
1360 extern declarations in function scope.
1361 Patch from Oliver Fleischmann
1363 PP/10 Use sig_atomic_t for flags set from signal handlers.
1364 Check getgroups() return and improve debugging.
1365 Fixed developed for diagnosis in bug 927 (which turned out to be
1368 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
1369 Patch from Mark Zealey.
1371 PP/12 Bugzilla 1056: Improved spamd server selection.
1372 Patch from Mark Zealey.
1374 PP/13 Bugzilla 1086: Deal with maildir quota file races.
1375 Based on patch from Heiko Schlittermann.
1377 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
1378 Patch from Uwe Doering, sign-off by Michael Haardt.
1380 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
1381 variable declaration deep within a block. Bug and patch from
1384 PP/15 lookups-Makefile IRIX compatibilty coercion.
1386 PP/16 Make DISABLE_DKIM build knob functional.
1388 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
1389 Patch by Simon Arlott
1391 TF/03 Fix valgrind.h portability to C89 compilers that do not support
1392 variable argument macros. Our copy now differs from upstream.
1398 TF/01 Failure to get a lock on a hints database can have serious
1399 consequences so log it to the panic log.
1401 TF/02 Log LMTP confirmation messages in the same way as SMTP,
1402 controlled using the smtp_confirmation log selector.
1404 TF/03 Include the error message when we fail to unlink a spool file.
1406 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
1407 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
1408 for maintaining out-of-tree patches for some time.
1410 PP/01 Bugzilla 139: Documentation and portability issues.
1411 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
1412 Handle per-OS dynamic-module compilation flags.
1414 PP/02 Let /dev/null have normal permissions.
1415 The 4.73 fixes were a little too stringent and complained about the
1416 permissions on /dev/null. Exempt it from some checks.
1417 Reported by Andreas M. Kirchwitz.
1419 PP/03 Report version information for many libraries, including
1420 Exim version information for dynamically loaded libraries. Created
1421 version.h, now support a version extension string for distributors
1422 who patch heavily. Dynamic module ABI change.
1424 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
1425 privilege escalation vulnerability whereby the Exim run-time user
1426 can cause root to append content of the attacker's choosing to
1429 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
1432 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
1433 If dropping privileges for untrusted macros, we disabled normal logging
1434 on the basis that it would fail; for the Exim run-time user, this is not
1435 the case, and it resulted in successful deliveries going unlogged.
1436 Fixed. Reported by Andreas Metzler.
1442 PP/01 Date: & Message-Id: revert to normally being appended to a message,
1443 only prepend for the Resent-* case. Fixes regression introduced in
1444 Exim 4.70 by NM/22 for Bugzilla 607.
1446 PP/02 Include check_rfc2047_length in configure.default because we're seeing
1447 increasing numbers of administrators be bitten by this.
1449 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
1451 PP/03 Bugzilla 994: added openssl_options main configuration option.
1453 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
1455 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
1457 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
1459 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
1460 without a peer certificate, leading to a segfault because of an
1461 assumption that peers always have certificates. Be a little more
1462 paranoid. Problem reported by Martin Tscholak.
1464 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
1465 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
1466 NB: ClamAV planning to remove STREAM in "middle of 2010".
1467 CL also introduces -bmalware, various -d+acl logging additions and
1468 more caution in buffer sizes.
1470 PP/09 Implemented reverse_ip expansion operator.
1472 PP/10 Bugzilla 937: provide a "debug" ACL control.
1474 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
1476 PP/12 Bugzilla 973: Implement --version.
1478 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
1480 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
1482 PP/15 Bugzilla 816: support multiple condition rules on Routers.
1484 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
1485 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
1486 ignore trailing whitespace.
1488 JJ/02 prevent non-panic DKIM error from being sent to paniclog
1490 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
1493 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
1494 Notification from Dr Andrew Aitchison.
1496 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
1497 ExtendedDetectionInfo response format.
1498 Notification from John Horne.
1500 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
1503 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
1504 XSL and documented dependency on system catalogs, with examples of how
1507 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
1510 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
1511 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
1512 configuration file which is writeable by the Exim user or group.
1514 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
1515 of configuration files to cover files specified with the -C option if
1516 they are going to be used with root privileges, not just the default
1519 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
1520 option (effectively making it always true).
1522 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
1523 files to be used while preserving root privileges.
1525 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
1526 that rogue child processes cannot use them.
1528 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
1529 run-time user, instead of root.
1531 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
1532 Exim run-time user without dropping privileges.
1534 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
1535 result string, instead of calling string_vformat() twice with the same
1538 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
1539 for other users. Others should always drop root privileges if they use
1540 -C on the command line, even for a whitelisted configure file.
1542 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
1544 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
1550 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
1551 $data_path, and $header_path variables; fixed documentation bugs and
1554 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
1555 exipick to access non-standard spools, including the "frozen" queue
1558 NM/01 Bugzilla 965: Support mysql stored procedures.
1559 Patch from Alain Williams
1561 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
1563 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
1564 Patch from Andreas Metzler
1566 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
1567 Patch from Kirill Miazine
1569 NM/05 Bugzilla 671: Added umask to procmail example.
1571 JJ/03 installed exipick 20100323.0, fixing doc bug
1573 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
1574 directory. Notification and patch from Dan Rosenberg.
1576 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
1578 TK/02 Improve log output when DKIM signing operation fails.
1580 MH/01 Treat the transport option dkim_domain as a colon separated
1581 list, not as a single string, and sign the message with each element,
1582 omitting multiple occurences of the same signer.
1584 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
1585 Bugzilla 985, 986. Patch by Simon Arlott
1587 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
1588 Patch by Simon Arlott
1590 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
1591 MBX locking. Notification from Dan Rosenberg.
1597 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
1599 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
1601 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
1603 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
1605 NM/04 Bugzilla 915: Flag broken perl installation during build.
1611 TK/01 Added patch by Johannes Berg that expands the main option
1612 "spamd_address" if it starts with a dollar sign.
1614 TK/02 Write list of recipients to X-Envelope-Sender header when building
1615 the mbox-format spool file for content scanning (suggested by Jakob
1618 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
1619 (http://www.dcc-servers.net/) support via dccifd. Activated by
1620 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
1622 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
1623 by Mark Daniel Reidel <mr@df.eu>.
1625 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
1626 When building exim an external PCRE library is now needed -
1627 PCRE is a system library on the majority of modern systems.
1628 See entry on PCRE_LIBS in EDITME file.
1630 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
1631 conversation. Added nologin parameter to request.
1632 Patch contributed by Kirill Miazine.
1634 TF/01 Do not log submission mode rewrites if they do not change the address.
1636 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
1638 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
1639 log files in place. Contributed by Roberto Lima.
1641 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
1643 TF/03 Bugzilla 615: When checking the local_parts router precondition
1644 after a local_part_suffix or local_part_prefix option, Exim now
1645 does not use the address's named list lookup cache, since this
1646 contains cached lookups for the whole local part.
1648 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
1649 Robert Millan. Documentation is in experimental-spec.txt.
1651 TF/04 Bugzilla 668: Fix parallel build (make -j).
1653 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
1655 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
1656 Patch provided by Jan Srzednicki.
1658 TF/05 Leading white space used to be stripped from $spam_report which
1659 wrecked the formatting. Now it is preserved.
1661 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
1662 that they are available at delivery time.
1664 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
1666 TF/08 TLS error reporting now respects the incoming_interface and
1667 incoming_port log selectors.
1669 TF/09 Produce a more useful error message if an SMTP transport's hosts
1670 setting expands to an empty string.
1672 NM/06 Bugzilla 744: EXPN did not work under TLS.
1673 Patch provided by Phil Pennock.
1675 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
1676 Patch provided by Richard Godbee.
1678 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
1679 acl_smtp_notquit, added index entry.
1681 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
1682 Patch provided by Eugene Bujak.
1684 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
1685 accept(). Patch provided by Maxim Dounin.
1687 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
1688 Patch provided by Phil Pennock.
1690 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
1692 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
1693 Patch provided by Brad "anomie" Jorsch.
1695 NM/14 Bugzilla 622: Added timeout setting to transport filter.
1696 Patch provided by Dean Brooks.
1698 TK/05 Add native DKIM support (does not depend on external libraries).
1700 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
1701 Patch provided by Graeme Fowler.
1703 NM/16 Bugzilla 851: Documentation example syntax fix.
1705 NM/17 Changed NOTICE file to remove references to embedded PCRE.
1707 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
1710 NM/19 Bugzilla 745: TLS version reporting.
1711 Patch provided by Phil Pennock.
1713 NM/20 Bugzilla 167: bool: condition support.
1714 Patch provided by Phil Pennock.
1716 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
1717 clients. Patch provided by Phil Pennock.
1719 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
1720 Patch provided by Brad "anomie" Jorsch.
1722 NM/23 Bugzilla 687: Fix misparses in eximstats.
1723 Patch provided by Heiko Schlittermann.
1725 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
1726 Patch provided by Heiko Schlittermann.
1728 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
1729 plus update to original patch.
1731 NM/26 Bugzilla 799: Documentation correction for ratelimit.
1733 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
1734 Patch provided by David Brownlee.
1736 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
1738 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
1740 NM/30 Bugzilla 888: TLS documentation bugfixes.
1742 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
1744 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
1745 Unlike the original bugzilla I have changed all shell scripts in src tree.
1747 NM/33 Bugzilla 898: Transport filter timeout fix.
1748 Patch by Todd Rinaldo.
1750 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
1751 Patch by Serge Demonchaux.
1753 NM/35 Bugzilla 39: Base64 decode bug fixes.
1754 Patch by Jakob Hirsch.
1756 NM/36 Bugzilla 909: Correct connect() call in dcc code.
1758 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
1760 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
1762 NM/39 Bugzilla 911: Fixed MakeLinks build script.
1768 TK/01 Add preliminary DKIM support. Currently requires a forked version of
1769 ALT-N's libdkim that I have put here:
1770 http://duncanthrax.net/exim-experimental/
1772 Note to Michael Haardt: I had to rename some vars in sieve.c. They
1773 were called 'true' and it seems that C99 defines that as a reserved
1774 keyword to be used with 'bool' variable types. That means you could
1775 not include C99-style headers which use bools without triggering
1776 build errors in sieve.c.
1778 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
1779 as mailq or other aliases. Changed the --help handling significantly
1780 to do whats expected. exim_usage() emits usage/help information.
1782 SC/01 Added the -bylocaldomain option to eximstats.
1784 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
1786 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
1788 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
1794 PH/01 Another patch from the Sieve maintainer.
1796 PH/02 When an IPv6 address is converted to a string for single-key lookup
1797 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
1798 dots are used instead of colons so that keys in lsearch files need not
1799 contain colons. This was done some time before quoting was made available
1800 in lsearch files. However, iplsearch files do require colons in IPv6 keys
1801 (notated using the quote facility) so as to distinguish them from IPv4
1802 keys. This meant that lookups for IP addresses in host lists did not work
1803 for iplsearch lookups.
1805 This has been fixed by arranging for IPv6 addresses to be expressed with
1806 colons if the lookup type is iplsearch. This is not incompatible, because
1807 previously such lookups could never work.
1809 The situation is now rather anomolous, since one *can* have colons in
1810 ordinary lsearch keys. However, making the change in all cases is
1811 incompatible and would probably break a number of configurations.
1813 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
1816 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
1817 conversion specification without a maximum field width, thereby enabling
1818 a rogue spamd server to cause a buffer overflow. While nobody in their
1819 right mind would setup Exim to query an untrusted spamd server, an
1820 attacker that gains access to a server running spamd could potentially
1821 exploit this vulnerability to run arbitrary code as the Exim user.
1823 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
1824 $primary_hostname instead of what libspf2 thinks the hosts name is.
1826 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
1827 a directory entry by the name of the lookup key. Previously, if a
1828 symlink pointed to a non-existing file or a file in a directory that
1829 Exim lacked permissions to read, a lookup for a key matching that
1830 symlink would fail. Now it is enough that a matching directory entry
1831 exists, symlink or not. (Bugzilla 503.)
1833 PH/03 The body_linecount and body_zerocount variables are now exported in the
1836 PH/04 Added the $dnslist_matched variable.
1838 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
1839 This means they are set thereafter only if the connection becomes
1842 PH/06 Added the client_condition to authenticators so that some can be skipped
1843 by clients under certain conditions.
1845 PH/07 The error message for a badly-placed control=no_multiline_responses left
1846 "_responses" off the end of the name.
1848 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
1850 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
1851 (without spaces) instead of just copying the configuration text.
1853 PH/10 Added the /noupdate option to the ratelimit ACL condition.
1855 PH/11 Added $max_received_linelength.
1857 PH/12 Added +ignore_defer and +include_defer to host lists.
1859 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
1860 way in which PCRE > 7.0 is built.
1862 PH/14 Implemented queue_only_load_latch.
1864 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
1865 MAIL command. The effect was to mangle the value on 64-bit systems.
1867 PH/16 Another patch from the Sieve maintainer.
1869 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
1871 PH/18 If a system quota error occurred while trying to create the file for
1872 a maildir delivery, the message "Mailbox is full" was not appended to the
1873 bounce if the delivery eventually timed out. Change 4.67/27 below applied
1874 only to a quota excession during the actual writing of the file.
1876 PH/19 It seems that peer DN values may contain newlines (and other non-printing
1877 characters?) which causes problems in log lines. The DN values are now
1878 passed through string_printing() before being added to log lines.
1880 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
1881 and InterBase are left for another time.)
1883 PH/21 Added message_body_newlines option.
1885 PH/22 Guard against possible overflow in moan_check_errorcopy().
1887 PH/23 POSIX allows open() to be a macro; guard against that.
1889 PH/24 If the recipient of an error message contained an @ in the local part
1890 (suitably quoted, of course), incorrect values were put in $domain and
1891 $local_part during the evaluation of errors_copy.
1897 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
1898 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
1901 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
1902 issue a MAIL command.
1904 PH/02 In an ACL statement such as
1906 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
1908 if a client was not listed at all, or was listed with a value other than
1909 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
1910 the condition was not true (as it should be), so access was not denied.
1911 The bug was that the ! inversion was incorrectly passed on to the second
1912 item. This has been fixed.
1914 PH/03 Added additional dnslists conditions == and =& which are different from
1915 = and & when the dns lookup returns more than one IP address.
1917 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
1918 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
1920 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
1921 FSYNC, which compiles an option called disable_fsync that allows for
1922 bypassing fsync(). The documentation is heavily laced with warnings.
1924 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
1926 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
1927 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
1928 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
1929 including adding "make clean"; (3) Added -fPIC when compiling the test
1930 dynamically loaded module, to get rid of a warning.
1932 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
1933 message fails, move_frozen_messages = true and ignore_bounce_errors_after
1934 = 0s. The bug is otherwise harmless.
1936 PH/07 There was a bug in the dovecot authenticator such that the value of
1937 $auth1 could be overwritten, and so not correctly preserved, after a
1938 successful authentication. This usually meant that the value preserved by
1939 the server_setid option was incorrect.
1941 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
1943 PH/09 Installed PCRE release 7.0.
1945 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
1946 run for batched SMTP input. It is now run at the start of every message
1947 in the batch. While fixing this I discovered that the process information
1948 (output by running exiwhat) was not always getting set for -bs and -bS
1949 input. This is fixed, and it now also says "batched" for BSMTP.
1951 PH/11 Added control=no_pipelining.
1953 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
1954 patch, slightly modified), and move the expansion of helo_data till after
1955 the connection is made in the smtp transport (so it can use these
1958 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
1960 PH/14 Added log_selector = +pid.
1962 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
1964 PH/16 Add ${if forany and ${if forall.
1966 PH/17 Added dsn_from option to vary the From: line in DSNs.
1968 PH/18 Flush SMTP output before performing a callout, unless control =
1969 no_callout_flush is set.
1971 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
1972 was true (the default) a successful delivery failed to delete the retry
1973 item, thus causing premature timeout of the address. The bug is now
1976 PH/20 Added hosts_avoid_pipelining to the smtp transport.
1978 PH/21 Long custom messages for fakedefer and fakereject are now split up
1979 into multiline reponses in the same way that messages for "deny" and
1980 other ACL rejections are.
1982 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
1983 with slight modification.
1985 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
1986 draft, changing the syntax and factoring some duplicate code".
1988 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
1989 for deliveries of the second and subsequent messages over the same SMTP
1992 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
1993 ${reduce, with only minor "tidies".
1995 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
1997 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
1998 expansion side effects.
2000 PH/27 When a message times out after an over-quota error from an Exim-imposed
2001 quota, the bounce message says "mailbox is full". This message was not
2002 being given when it was a system quota that was exceeded. It now should
2005 MH/03 Made $recipients available in local_scan(). local_scan() already has
2006 better access to the recipient list through recipients_list[], but
2007 $recipients can be useful in postmaster-provided expansion strings.
2009 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
2010 in the case of a MAIL command with additional options following the
2011 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
2012 were accidentally chopped off.
2014 PH/29 SMTP synchronization checks are implemented when a command is read -
2015 there is a check that no more input is waiting when there shouldn't be
2016 any. However, for some commands, a delay in an ACL can mean that it is
2017 some time before the response is written. In this time, more input might
2018 arrive, invalidly. So now there are extra checks after an ACL has run for
2019 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
2020 pipelining has not been advertised.
2022 PH/30 MH's patch to allow iscntrl() characters to be list separators.
2024 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
2025 returned in the SMTP response when smtp_return_error_details was false.
2026 This has been fixed.
2028 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
2029 instead of the C I/O that was originally supplied, because problems were
2030 reported on Solaris.
2032 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
2033 Exim which did not show up earlier: it was assuming that a call to
2034 SSL_CTX_set_info_callback() might give an error value. In fact, there is
2035 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
2036 was a macro that became an assignment, so it seemed to work. This has
2037 changed to a proper function call with a void return, hence the compile
2038 error. Exim's code has been fixed.
2040 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
2043 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
2045 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
2047 PH/37 If a message is not accepted after it has had an id assigned (e.g.
2048 because it turns out to be too big or there is a timeout) there is no
2049 "Completed" line in the log. When some messages of this type were
2050 selected by exigrep, they were listed as "not completed". Others were
2051 picked up by some special patterns. I have improved the selection
2052 criteria to be more general.
2054 PH/38 The host_find_failed option in the manualroute router can now be set
2055 to "ignore", to completely ignore a host whose IP address cannot be
2056 found. If all hosts are ignored, the behaviour is controlled by the new
2057 host_all_ignored option.
2059 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
2060 homing or because of multiple MX records with /mx) generated more than
2061 one IP address, and the following item turned out to be the local host,
2062 all the secondary addresses of the first item were incorrectly removed
2063 from the list, along with the local host and any following hosts (which
2064 is what is supposed to happen).
2066 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
2067 whoever called Exim into the -H file. In the case of the daemon it was
2068 behaving confusingly. When first started, it used values for whoever
2069 started the daemon, but after a SIGHUP it used the Exim user (because it
2070 calls itself on a restart). I have changed the code so that it now always
2073 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
2074 message are rejected with the same error (e.g. no authentication or bad
2075 sender address), and a DATA command is nevertheless sent (as can happen
2076 with PIPELINING or a stupid MUA), the error message that was given to the
2077 RCPT commands is included in the rejection of the DATA command. This is
2078 intended to be helpful for MUAs that show only the final error to their
2081 PH/42 Another patch from the Sieve maintainer.
2083 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
2084 Eximstats - Fixed some broken HTML links and added missing column headers
2086 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
2087 columns for Rejects, Temp Rejects, Ham, and Spam rows.
2089 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
2091 PH/43 Yet another patch from the Sieve maintainer.
2093 PH/44 I found a way to check for a TCP/IP connection going away before sending
2094 the response to the final '.' that terminates a message, but only in the
2095 case where the client has not sent further data following the '.'
2096 (unfortunately, this is allowed). However, in many cases there won't be
2097 any further data because there won't be any more messages to send. A call
2098 to select() can be used: if it shows that the input is "ready", there is
2099 either input waiting, or the socket has been closed. An attempt to read
2100 the next input character can distinguish the two cases. Previously, Exim
2101 would have sent an OK response which the client would never have see.
2102 This could lead to message repetition. This fix should cure that, at
2103 least in a lot of common cases.
2105 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
2106 advertised in response to EHLO.
2112 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
2113 fixed by 4.65/MH/01 (is this a record?) are fixed:
2115 (i) An empty string was always treated as zero by the numeric comparison
2116 operators. This behaviour has been restored.
2118 (ii) It is documented that the numeric comparison operators always treat
2119 their arguments as decimal numbers. This was broken in that numbers
2120 starting with 0 were being interpreted as octal.
2122 While fixing these problems I realized that there was another issue that
2123 hadn't been noticed. Values of message_size_limit (both the global option
2124 and the transport option) were treated as octal if they started with 0.
2125 The documentation was vague. These values are now always treated as
2126 decimal, and I will make that clear in the documentation.
2132 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
2133 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
2136 MH/01 Don't check that the operands of numeric comparison operators are
2137 integers when their expansion is in "skipping" mode (fixes bug
2138 introduced by 4.64-PH/07).
2140 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
2141 child addresses, Exim now panics and dies. Previously, because the count
2142 is held in a short int, deliveries were likely to be lost. As such a
2143 large number of recipients for a single message is ridiculous
2144 (performance will be very, very poor), I have chosen to impose a limit
2145 rather than extend the field.
2151 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
2152 leftover -K file (the existence of which was triggered by #402).
2153 While we were at it, introduced process PID as part of the -K
2154 filename. This should rule out race conditions when creating
2157 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
2158 processing considerably. Previous code took too long for large mails,
2159 triggering a timeout which in turn triggers #401.
2161 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
2162 in the DK code in transports.c. sendfile() is not really portable,
2163 hence the _LINUX specificness.
2165 TF/01 In the add_headers option to the mail command in an Exim filter,
2166 there was a bug that Exim would claim a syntax error in any
2167 header after the first one which had an odd number of characters
2170 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
2171 callout verification, Exim cached a "reject" for the entire domain. This
2172 is correct for most verifications, but it is not correct for a recipient
2173 verification with use_sender or use_postmaster set, because in that case
2174 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
2175 case of MAIL FROM:<> rejection from other early rejections (e.g.
2176 rejection of HELO). When verifying a recipient using a non-null MAIL
2177 address, the cache is ignored if it shows MAIL FROM:<> rejection.
2178 Whatever the result of the callout, the value of the domain cache is
2179 left unchanged (for any other kind of callout, getting as far as trying
2180 RCPT means that the domain itself is ok).
2182 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
2185 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
2186 manifest itself as EPIPE rather than ECONNECT. When tidying away a
2187 session, the daemon ignores ECONNECT errors and logs others; it now
2188 ignores EPIPE as well.
2190 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
2191 (quoted-printable decoding).
2193 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
2194 later the small subsequent patch to fix an introduced bug.
2196 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
2198 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
2200 PH/08 An error is now given if message_size_limit is specified negative.
2202 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
2203 to be given (somewhat) arbitrary names.
2205 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
2208 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
2209 miscellaneous code fixes
2211 PH/10 Added the log_reject_target ACL modifier to specify where to log
2214 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
2215 hostname. This is wrong, because it relates to the incoming message (and
2216 probably the interface on which it is arriving) and not to the outgoing
2217 callout (which could be using a different interface). This has been
2218 changed to use the value of the helo_data option from the smtp transport
2219 instead - this is what is used when a message is actually being sent. If
2220 there is no remote transport (possible with a router that sets up host
2221 addresses), $smtp_active_hostname is used.
2223 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
2224 tweaks were necessary in order to get it to work (see also 21 below):
2225 (a) The code assumed that strncpy() returns a negative number on buffer
2226 overflow, which isn't the case. Replaced with Exim's string_format()
2228 (b) There were several signed/unsigned issues. I just did the minimum
2229 hacking in of casts. There is scope for a larger refactoring.
2230 (c) The code used strcasecmp() which is not a standard C function.
2231 Replaced with Exim's strcmpic() function.
2232 (d) The code set only $1; it now sets $auth1 as well.
2233 (e) A simple test gave the error "authentication client didn't specify
2234 service in request". It would seem that Dovecot has changed its
2235 interface. Fortunately there's a specification; I followed it and
2236 changed what the client sends and it appears to be working now.
2238 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
2241 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
2242 address A is aliased to B and C, where B exists and C does not. Without
2243 -v the output is "A verified" because verification stops after a
2244 successful redirection if more than one address is generated. However,
2245 with -v the child addresses are also verified. Exim was outputting "A
2246 failed to verify" and then showing the successful verification for C,
2247 with its parentage. It now outputs "B failed to verify", showing B's
2248 parentage before showing the successful verification of C.
2250 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
2251 look up a TXT record in a specific list after matching in a combined
2254 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
2255 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
2256 they consult the DNS. I had assumed they would set it the way they
2257 wanted; and indeed my experiments on Linux seem to show that in some
2258 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
2259 To be on the safe side, however, I have now made the interface to
2260 host_find_byname() similar to host_find_bydns(), with an argument
2261 containing the DNS resolver options. The host_find_byname() function now
2262 sets these options at its start, just as host_find_bydns() does. The smtp
2263 transport options dns_qualify_single and dns_search_parents are passed to
2264 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
2265 of host_find_byname() use the default settings of RES_DEFNAMES
2266 (qualify_single) but not RES_DNSRCH (search_parents).
2268 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
2269 spool_read_header() do less string testing, by means of a preliminary
2270 switch on the second character of optional "-foo" lines. (This is
2271 overdue, caused by the large number of possibilities that now exist.
2272 Originally there were few.) While I was there, I also converted the
2273 str(n)cmp tests so they don't re-test the leading "-" and the first
2274 character, in the hope this might squeeze out yet more improvement.
2276 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
2277 flag allowing group syntax was set by the header_syntax check but not
2278 turned off, possible causing trouble later; (2) The flag was not being
2279 set at all for the header_verify test, causing "group"-style headers to
2280 be rejected. I have now set it in this case, and also caused header_
2281 verify to ignore an empty address taken from a group. While doing this, I
2282 came across some other cases where the code for allowing group syntax
2283 while scanning a header line wasn't quite right (mostly, not resetting
2284 the flag correctly in the right place). These bugs could have caused
2285 trouble for malformed header lines. I hope it is now all correct.
2287 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
2288 with the "reply" argument non-NULL. The code, however (which originally
2289 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
2290 but it didn't always do it. This confused somebody who was copying the
2291 code for some other use. I have removed all the tests.
2293 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
2294 feature that was used to support insecure browsers during the U.S. crypto
2295 embargo. It requires special client support, and Exim is probably the
2296 only MTA that supported it -- and would never use it because real RSA is
2297 always available. This code has been removed, because it had the bad
2298 effect of slowing Exim down by computing (never used) parameters for the
2299 RSA_EXPORT functionality.
2301 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
2302 authenticator to fail if there's a tab character in the incoming data
2303 (there should never be unless someone is messing about, as it's supposed
2304 to be base64-encoded). Also added, on Timo's advice, the "secured" option
2305 if the connection is using TLS or if the remote IP is the same as the
2306 local IP, and the "valid-client-cert option" if a client certificate has
2309 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
2310 authenticators. This can be used for authorization after authentication
2311 succeeds. (In the case of plaintext, it servers for both authentication
2314 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
2315 if any retry times were supplied.
2317 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
2318 connection, where there is no client IP address to check. In this
2319 situation, the verify now always succeeds.
2321 PH/25 Applied John Jetmore's -Mset patch.
2323 PH/26 Added -bem to be like -Mset, but loading a message from a file.
2325 PH/27 In a string expansion for a processed (not raw) header when multiple
2326 headers of the same name were present, leading whitespace was being
2327 removed from all of them, but trailing whitespace was being removed only
2328 from the last one. Now trailing whitespace is removed from each header
2329 before concatenation. Completely empty headers in a concatenation (as
2330 before) are ignored.
2332 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
2333 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
2335 PH/29 [Removed. This was a change that I later backed out, and forgot to
2336 correct the ChangeLog entry (that I had efficiently created) before
2337 committing the later change.]
2339 PH/30 Exim was sometimes attempting to deliver messages that had suffered
2340 address errors (4xx response to RCPT) over the same connection as other
2341 messages routed to the same hosts. Such deliveries are always "forced",
2342 so retry times are not inspected. This resulted in far too many retries
2343 for the affected addresses. The effect occurred only when there were more
2344 hosts than the hosts_max_try setting in the smtp transport when it had
2345 the 4xx errors. Those hosts that it had tried were not added to the list
2346 of hosts for which the message was waiting, so if all were tried, there
2347 was no problem. Two fixes have been applied:
2349 (i) If there are any address or message errors in an SMTP delivery, none
2350 of the hosts (tried or untried) are now added to the list of hosts
2351 for which the message is waiting, so the message should not be a
2352 candidate for sending over the same connection that was used for a
2353 successful delivery of some other message. This seems entirely
2354 reasonable: after all the message is NOT "waiting for some host".
2355 This is so "obvious" that I'm not sure why it wasn't done
2356 previously. Hope I haven't missed anything, but it can't do any
2357 harm, as the worst effect is to miss an optimization.
2359 (ii) If, despite (i), such a delivery is accidentally attempted, the
2360 routing retry time is respected, so at least it doesn't keep
2361 hammering the server.
2363 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
2364 in ${readsocket because some servers need this prod.
2366 PH/32 Added some extra debug output when updating a wait-xxx database.
2368 PH/33 The hint "could be header name not terminated by colon", which has been
2369 given for certain expansion errors for a long time, was not being given
2370 for the ${if def:h_colon_omitted{... case.
2372 PH/34 The spec says: "With one important exception, whenever a domain list is
2373 being scanned, $domain contains the subject domain." There was at least
2374 one case where this was not true.
2376 PH/35 The error "getsockname() failed: connection reset by peer" was being
2377 written to the panic log as well as the main log, but it isn't really
2378 panic-worthy as it just means the connection died rather early on. I have
2379 removed the panic log writing for the ECONNRESET error when getsockname()
2382 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
2383 runs only) independently of the message's sender address. This meant
2384 that, if the 4xx error was in fact related to the sender, a different
2385 message to the same recipient with a different sender could confuse
2386 things. In particualar, this can happen when sending to a greylisting
2387 server, but other circumstances could also provoke similar problems.
2388 I have changed the default so that the retry time for these errors is now
2389 based a combination of the sender and recipient addresses. This change
2390 can be overridden by setting address_retry_include_sender=false in the
2393 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
2394 remote server are returned as part of bounce messages. This was not
2395 happening for LMTP over a pipe (the lmtp transport), but now it is the
2396 same for both kinds of LMTP.
2398 PH/38 Despite being documented as not happening, Exim was rewriting addresses
2399 in header lines that were in fact CNAMEs. This is no longer the case.
2401 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
2402 and queue runs started by the daemon processed all messages. This has
2403 been fixed so that -R and -S can now usefully be given with -q<time>.
2405 PH/40 Import PCRE release 6.7 (fixes some bugs).
2407 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
2409 PH/42 Give an error if -q is specified more than once.
2411 PH/43 Renamed the variables $interface_address and $interface_port as
2412 $received_ip_address and $received_port, to make it clear that these
2413 values apply to message reception, and not to the outgoing interface when
2414 a message is delivered. (The old names remain recognized, of course.)
2416 PH/44 There was no timeout on the connect() call when using a Unix domain
2417 socket in the ${readsocket expansion. There now is.
2419 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
2420 be meaningful with "accept".
2422 SC/01 Eximstats V1.43
2423 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
2425 SC/02 Eximstats V1.44
2426 Use a glob alias rather than an array ref in the generated
2427 parser. This improves both readability and performance.
2429 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
2430 Collect SpamAssassin and rejection statistics.
2431 Don't display local sender or destination tables unless
2432 there is data to show.
2433 Added average volumes into the top table text output.
2435 SC/04 Eximstats V1.46
2436 Collect data on the number of addresses (recipients)
2437 as well as the number of messages.
2439 SC/05 Eximstats V1.47
2440 Added 'Message too big' to the list of mail rejection
2441 reasons (thanks to Marco Gaiarin).
2443 SC/06 Eximstats V1.48
2444 Mainlog lines which have GMT offsets and are too short to
2445 have a flag are now skipped.
2447 SC/07 Eximstats V1.49 (Alain Williams)
2448 Added the -emptyok flag.
2450 SC/08 Eximstats V1.50
2451 Fixes for obtaining the IP address from reject messages.
2453 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
2454 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
2455 whitesspace changes from 4.64-PH/27
2457 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
2460 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
2461 are found, allow negative numbers in numeric criteria)
2463 JJ/06 exipick.20061117.2, added new $message_body_missing variable
2465 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
2466 to match changes made in 4.64-PH/43
2468 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
2470 PH/47 Put in an explicit test for a DNS lookup of an address record where the
2471 "domain" is actually an IP address, and force a failure. This locks out
2472 those revolvers/nameservers that support "A-for-A" lookups, in
2473 contravention of the specifications.
2475 PH/48 When a host name was looked up from an IP address, and the subsequent
2476 forward lookup of the name timed out, the host name was left in
2477 $sender_host_name, contrary to the specification.
2479 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
2480 restricted to single-key lookups, Exim was not diagnosing an error if
2481 * or *@ was used with a query-style lookup.
2483 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
2485 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
2486 long ago, but noone interested enough thought of it. Let's just say that
2487 the "1.1" means that there are some new functions that weren't there at
2488 some point in the past.
2490 PH/51 Error processing for expansion failure of helo_data from an smtp
2491 transport during callout processing was broken.
2493 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
2494 tested/used via the -bh/-bhc/-bs options.
2496 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
2497 bug, fixed in subsequent PCRE releases).
2499 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
2500 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
2502 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
2508 SC/01 Use a glob alias rather than an array ref in eximstats generated
2509 parser. This improves both readability and performance.
2511 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
2512 Don't display local sender or destination tables in eximstats unless
2513 there is data to show.
2514 Added average volumes into the eximstats top table text output.
2516 SC/03 Collect data on the number of addresses (recipients) as well
2517 as the number of messages in eximstats.
2519 TF/01 Correct an error in the documentation for the redirect router. Exim
2520 does (usually) call initgroups() when daemonizing.
2522 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
2523 with consistent privilege compared to when running as a daemon.
2525 TF/03 Note in the spec that $authenticated_id is not set for local
2526 submissions from trusted users.
2528 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
2529 Thanks to Dean Brooks <dean@iglou.com> for the patch.
2531 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
2532 by adding some example configuration directives to the default
2533 configuration file. A little bit of work is required to uncomment the
2534 directives and define how usernames and passwords are checked, but
2535 there is now a framework to start from.
2537 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
2538 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
2539 without this. I don't know how relevant this is to other LDAP libraries.
2541 PH/02 Add the verb name to the "unknown ACL verb" error.
2543 PH/03 Magnus Holmgren's patch for filter_prepend_home.
2545 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
2547 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
2548 directory not expanded when it should be if an expanded home directory
2549 was set for the address (which is overridden by the transport).
2551 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
2554 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
2555 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
2556 because it is too late at that time, and has no effect.
2558 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
2559 security issue with \' (bugzilla #107). I could not use the
2560 PQescapeStringConn() function, because it needs a PGconn value as one of
2563 PH/08 When testing addresses using -bt, indicate those final addresses that
2564 are duplicates that would not cause an additional delivery. At least one
2565 person was confused, thinking that -bt output corresponded to deliveries.
2566 (Suppressing duplicates isn't a good idea as you lose the information
2567 about possibly different redirections that led to the duplicates.)
2569 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
2570 systems where poll() doesn't work, in particular OS X.
2572 PH/10 Added more information to debugging output for retry time not reached.
2574 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
2575 operations in malware.c.
2577 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
2580 PH/13 If write_rejectlog was set false when logging was sent to syslog with
2581 syslog_duplication set false, log lines that would normally be written
2582 both the the main log and to the reject log were not written to syslog at
2585 PH/14 In the default configuration, change the use of "message" in ACL warn
2586 statements to "add_header".
2588 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
2589 not followed by a command (e.g. "seen endif").
2591 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
2592 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
2595 PH/17 Added extra conditions to the default value of delay_warning_condition
2599 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
2600 { match{$h_precedence:}{(?i)bulk|list|junk} } \
2601 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
2604 The Auto-Submitted: and various List- headers are standardised, whereas I
2605 don't think Precedence: ever was.
2607 PH/18 Refactored debugging code in route_finduser() to show more information,
2608 in particular, the error code if getpwnam() issues one.
2610 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
2611 This is apparently needed in addition to the PH/07 change above to avoid
2612 any possible encoding problems.
2614 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
2615 but not after initializing Perl.
2617 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
2618 output them only if debugging. By default they are written stderr,
2619 apparently, which is not desirable.
2621 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
2624 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
2627 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
2629 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
2630 authenticated or an ident call has been made. Suppress the default
2631 values for $authenticated_id and $authenticated_sender (but permit -oMai
2632 and -oMas) when testing with -bh.
2634 PH/24 Re-jigged the order of the tests in the default configuration so that the
2635 tests for valid domains and recipients precede the DNS black list and CSA
2636 tests, on the grounds that those ones are more expensive.
2638 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
2639 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
2640 This bug exists in every version of Exim that I still have, right back to
2643 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
2644 However, an attempt to turn on case-sensitivity in a regex key by
2645 including (?-i) didn't work because the subject string was already
2646 lowercased, and the effects were non-intuitive. It turns out that a
2647 one-line patch can be used to allow (?-i) to work as expected.
2653 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
2654 other effects) broke the use of negated acl sub-conditions.
2656 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
2659 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
2660 "Deny" causes Exim to reject the incoming connection with a 554 error.
2661 Unfortunately, if there is a major crisis, such as a disk failure,
2662 tcp-wrappers gives "deny", whereas what one would like would be some
2663 kind of temporary error. A kludge has been added to help with this.
2664 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
2665 554 error is used if errno is still zero or contains ENOENT (which occurs
2666 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
2669 PH/03 Add -lutil to the default FreeBSD LIBS setting.
2671 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
2672 errors. Otherwise a message that provokes a temporary error (when other
2673 messages do not) can cause a whole host to time out.
2675 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
2676 addresses were routed directly to files or pipes from a redirect router.
2677 File deliveries just didn't batch; pipe deliveries might have suffered
2680 PH/06 A failure to get a lock for a hints database would erroneously always say
2681 "Failed to get write lock", even when it was really a read lock.
2683 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
2684 of 0600. This has been changed to use the value of the lockfile_mode
2685 option (which defaults to 0600).
2687 PH/08 Applied small patch from the Sieve maintainer.
2689 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
2690 folder from quota calculations, a direct delivery into this folder messed
2691 up the contents of the maildirsize file. This was because the regex was
2692 used only to exclude .Trash (or whatever) when the size of the mailbox
2693 was calculated. There was no check that a delivery was happening into an
2694 excluded directory. This bug has been fixed by ignoring all quota
2695 processing for deliveries into excluded directories.
2697 PH/10 Added the maildirfolder_create_regex option to appendfile.
2703 PH/01 The code for finding all the local interface addresses on a FreeBSD
2704 system running IPv6 was broken. This may well have applied to all BSD
2705 systems, as well as to others that have similar system calls. The broken
2706 code found IPv4 interfaces correctly, but gave incorrect values for the
2707 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
2708 that it would not match correctly against @[] and not recognize the IPv6
2711 PH/02 The ipliteral router was not recognizing addresses of the form user@
2712 [ipv6:....] because it didn't know about the "ipv6:" prefix.
2714 PH/03 Added disable_ipv6.
2716 PH/04 Changed $reply_address to use the raw form of the headers instead of the
2717 decoded form, because it is most often used to construct To: headers
2718 lines in autoreplies, and the decoded form may well be syntactically
2719 invalid. However, $reply_address has leading white space removed, and all
2720 newlines turned into spaces so that the autoreply transport does not
2723 PH/05 If group was specified without a user on a router, and no group or user
2724 was specified on a transport, the group from the router was ignored.
2726 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
2727 for visible compile-time settings that can be used to change these
2728 numbers, for those that want even more. Backwards compatibility with old
2729 spool files has been maintained. However, going back to a previous Exim
2730 release will lost any variables that are in spool files.
2732 PH/07 Two small changes when running in the test harness: increase delay when
2733 passing a TCP/IP connection to a new process, in case the original
2734 process has to generate a bounce, and remove special handling of
2735 127.0.0.2 (sic), which is no longer necessary.
2737 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
2738 be the same on different OS.
2740 PH/09 Moved a debug statement in filter processing to avoid a race problem when
2743 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
2744 whether --show-vars was specified or not
2746 JJ/02 exipick: Added support for new ACL variable spool format introduced
2749 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
2750 syntactically invalid From: or Reply-to: line, and a filter used this to
2751 generate an autoreply, and therefore failed to obtain an address for the
2752 autoreply, Exim could try to deliver to a non-existent relative file
2753 name, causing unrelated and misleading errors. What now happens is that
2754 it logs this as a hard delivery error, but does not attempt to create a
2757 PH/11 The exinext utility has a -C option for testing purposes, but although
2758 the given file was scanned by exinext itself; it wasn't being passed on
2759 when Exim was called.
2761 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
2762 an end-of-file indication when reading a command response.
2764 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
2765 compiled. In many other places in Exim, IPv6 addresses are always
2766 recognized, so I have changed this. It also means that IPv4 domain
2767 literals of the form [IPV4:n.n.n.n] are now always recognized.
2769 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
2770 used if the router is not running as root, for example, when verifying at
2771 ACL time, or when using -bh. The debugging output from this situation was
2772 non-existent - all you got was a failure to exec. I have made two
2775 (a) Failures to set uid/gid, the current directory, or a process leader
2776 in a subprocess such as that created by queryprogram now generate
2777 suitable debugging ouput when -d is set.
2779 (b) The queryprogram router detects when it is not running as root,
2780 outputs suitable debugging information if -d is set, and then runs
2781 the subprocess without attempting to change uid/gid.
2783 PH/15 Minor change to Makefile for building test_host (undocumented testing
2786 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
2787 additional section of a DNS packet that returns MX or SRV records.
2788 Instead, it always explicitly searches for A/AAAA records. This avoids
2789 major problems that occur when a DNS server includes only records of one
2790 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
2791 fixed another bug: if SRV records were looked up and the corresponding
2792 address records were *not* found in the additional section, the port
2793 values from the SRV records were lost.
2795 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
2796 using the correct key (the original address) when searching the retry
2797 rules in order to find which one to use for generating the retry hint.
2799 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
2800 adding the default one. Similarly, if it contains a Reply-To: header, the
2801 errors_reply_to option, if set, is not used.
2803 PH/19 When calculating a retry time, Exim used to measure the "time since
2804 failure" by looking at the "first failed" field in the retry record. Now
2805 it does not use this if it is later than than the arrival time of the
2806 message. Instead it uses the arrival time. This makes for better
2807 behaviour in cases where some deliveries succeed, thus re-setting the
2808 "first failed" field. An example is a quota failure for a huge message
2809 when small messages continue to be delivered. Without this change, the
2810 "time since failure" will always be short, possible causing more frequent
2811 delivery attempts for the huge message than are intended.
2812 [Note: This change was subsequently modified - see PH/04 for 4.62.]
2814 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
2815 $1, $2, $3) because the numerical variables can be reset during some
2816 expansion items (e.g. "match"), thereby losing the authentication data.
2818 PH/21 Make -bV show the size of off_t variables so that the test suite can
2819 decide whether to run tests for quotas > 2G.
2821 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
2822 mailbox_size, and mailbox_filecount in the appendfile transport. If a
2823 filecount value is greater than 2G or if a quota value is greater than 2G
2824 on a system where the size of off_t is not greater than 4, a panic error
2827 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
2828 never match. The debug and -bh output now contains an explicit error
2829 message indicating a malformed IPv4 address or mask.
2831 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
2832 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
2833 PH/23 above applies.
2835 PH/25 Do not write to syslog when running in the test harness. The only
2836 occasion when this arises is a failure to open the main or panic logs
2837 (for which there is an explicit test).
2839 PH/26 Added the /no_tell option to "control=freeze".
2841 PH/27 If a host name lookup failed very early in a connection, for example, if
2842 the IP address matched host_lookup and the reverse lookup yielded a name
2843 that did not have a forward lookup, an error message of the form "no IP
2844 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
2845 could be logged. Now it outputs the IP address instead of "NULL".
2847 PH/28 An enabling patch from MH: add new function child_open_exim2() which
2848 allows the sender and the authenticated sender to be set when
2849 submitting a message from within Exim. Since child_open_exim() is
2850 documented for local_scan(), the new function should be too.
2852 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
2853 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
2854 results in an empty string is now treated as unset.
2856 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
2858 PH/31 Added sender_verify_fail logging option.
2860 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
2861 needed by RFC 822 but not by RFC 2822 was commented out. I have now
2862 tidied the source and removed it altogether.
2864 PH/33 When a queue run was abandoned because the load average was too high, a
2865 log line was always written; now it is written only if the queue_run log
2866 selector is set. In addition, the log line for abandonment now contains
2867 information about the queue run such as the pid. This is always present
2868 in "start" and "stop" lines but was omitted from the "abandon" line.
2870 PH/34 Omit spaces between a header name and the colon in the error message that
2871 is given when verify = headers_syntax fails (if there are lots of them,
2872 the message gets confusing).
2874 PH/35 Change the default for dns_check_names_pattern to allow slashes within
2875 names, as there are now some PTR records that contain slashes. This check
2876 is only to protect against broken name servers that fall over on strange
2877 characters, so the fact that it applies to all lookups doesn't matter.
2879 PH/36 Now that the new test suite is complete, we can remove some of the
2880 special code in Exim that was needed for the old test suite. For example,
2881 sorting DNS records because real resolvers return them in an arbitrary
2882 order. The new test suite's fake resolver always returns records in the
2885 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
2886 bounces) except when queue_only is set, to avoid logging races between
2887 the different processes.
2889 PH/38 Panic-die if .include specifies a non-absolute path.
2891 PH/39 A tweak to the "H" retry rule from its user.
2893 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
2894 a label. They prevented compilation on older perls.
2896 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
2897 a warning to be raised on newish perls.
2899 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
2900 on queue. Changes to match documented behaviour of showing count of
2901 messages matching specified criteria.
2903 PH/40 Changed the default ident timeout from 30s to 5s.
2905 PH/41 Added support for the use of login_cap features, on those BSD systems
2906 that have them, for controlling the resources used by pipe deliveries.
2908 PH/42 The content-scanning code uses fopen() to create files in which to put
2909 message data. Previously it was not paying any attention to the mode of
2910 the files. Exim runs with umask(0) because the rest of the code creates
2911 files with open(), and sets the required mode explicitly. Thus, these
2912 files were ending up world-writeable. This was not a big issue, because,
2913 being within the spool directory, they were not world-accessible. I have
2914 created a function called modefopen, which takes an additional mode
2915 argument. It sets umask(777), creates the file, chmods it to the required
2916 mode, then resets the umask. All the relevant calls to fopen() in the
2917 content scanning code have been changed to use this function.
2919 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
2920 to 24 hours. This avoids potential overflow problems when processing G
2921 and H retry rules. I suspect nobody ever tinkers with this value.
2923 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
2925 PH/45 When the plaintext authenticator is running as a client, the server's
2926 challenges are checked to ensure they are valid base64 strings. By
2927 default, the authentication attempt is cancelled if an invalid string is
2928 received. Setting client_ignore_invalid_base64 true ignores these errors.
2929 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
2930 they are received. Thus, the responses can be made to depend on the
2931 challenges. If an invalid string is ignored, an empty string is placed in
2934 PH/46 Messages that are created by the autoreply transport now contains a
2935 References: header, in accordance with RFCs 2822 and 3834.
2937 PH/47 Added authenticated_sender_force to the smtp transport.
2939 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
2941 PH/49 Installed latest patch from the Sieve maintainer.
2943 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
2944 was also set, the appendfile transport was unnecessarily scanning a
2945 directory of message files (e.g. for maildir delivery) to find the count
2946 of files (along with the size), even though it did not need this
2947 information. It now does the scan only if it needs to find either the
2948 size of the count of files.
2950 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
2952 PH/52 Two bugs concerned with error handling when the smtp transport is
2955 (i) Exim was not creating retry information for temporary errors given
2956 for individual recipients after the DATA command when the smtp transport
2957 was used in LMTP mode. This meant that they could be retried too
2958 frequently, and not timed out correctly.
2960 (ii) Exim was setting the flag that allows error details to be returned
2961 for LMTP errors on RCPT commands, but not for LMTP errors for individual
2962 recipients that were returned after the DATA command.
2964 PH/53 This is related to PH/52, but is more general: for any failing address,
2965 when detailed error information was permitted to be returned to the
2966 sender, but the error was temporary, then after the final timeout, only
2967 "retry timeout exceeded" was returned. Now it returns the full error as
2968 well as "retry timeout exceeded".
2970 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
2971 do this, and (what is worse) MTAs that accept it.
2973 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
2974 will now be deprecated.
2976 PH/56 New os.c-cygwin from the Cygwin maintainer.
2978 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
2979 formats (previously only available in exim formats via -bpr, -bpru,
2980 and -bpra. Now also available in native and exiqgrep formats)
2982 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
2983 with very large, slow to parse queues
2985 JJ/08 exipick: added ! as generic prefix to negate any criteria format
2987 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
2989 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
2990 responses to authentication challenges, though it was showing the
2991 challenges; (ii) I've removed the CR characters from the debug output for
2994 PH/58 Allow for the insertion of a newline as well as a space when a string
2995 is turned into more than one encoded-word during RFC 2047 encoding. The
2996 Sieve code now uses this.
2998 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
2999 data_4xx, lost_connection, tls_required.
3001 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
3002 message was being sent as an SMTP response.
3004 PH/61 Add -l and -k options to exicyclog.
3006 PH/62 When verifying, if an address was redirected to one new address, so that
3007 verification continued, and the new address failed or deferred after
3008 having set something in $address_data, the value of $address_data was not
3009 passed back to the ACL. This was different to the case when no
3010 redirection occurred. The value is now passed back in both cases.
3012 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
3013 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
3014 use login_cap.h, so on its own it isn't the distinguishing feature. The
3015 new name refers directly to the setclassresources() function.
3017 PH/65 Added configuration files for NetBSD3.
3019 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
3021 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
3022 is preferred over IPv4.
3024 PH/68 The bounce_return_message and bounce_return_body options were not being
3025 honoured for bounces generated during the reception of non-SMTP messages.
3026 In particular, this applied to messages rejected by the ACL. This bug has
3027 been fixed. However, if bounce_return_message is true and bounce_return_
3028 body is false, the headers that are returned for a non-SMTP message
3029 include only those that have been read before the error was detected.
3030 (In the case of an ACL rejection, they have all been read.)
3032 PH/69 The HTML version of the specification is now built in a directory called
3033 spec_html instead of spec.html, because the latter looks like a path with
3034 a MIME-type, and this confuses some software.
3036 PH/70 Catch two compiler warnings in sieve.c.
3038 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
3039 function verify_get_ident() calls ip_connect() to connect a socket, but
3040 if the "connect()" function timed out, ip_connect() used to close the
3041 socket. However, verify_get_ident() also closes the socket later, and in
3042 between Exim writes to the log, which may get opened at this point. When
3043 the socket was closed in ip_connect(), the log could get the same file
3044 descriptor number as the socket. This naturally causes chaos. The fix is
3045 not to close the socket in ip_connect(); the socket should be closed by
3046 the function that creates it. There was only one place in the code where
3047 this was missing, in the iplookup router, which I don't think anybody now
3048 uses, but I've fixed it anyway.
3050 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
3051 well as to direct DNS lookups. Otherwise the handling of names in host
3052 lists is inconsistent and therefore confusing.
3058 PH/01 Two changes to the default runtime configuration:
3060 (1) Move the checks for relay_from_hosts and authenticated clients from
3061 after to before the (commented out) DNS black list checks.
3063 (2) Add control=submission to the relay_from_hosts and authenticated
3064 clients checks, on the grounds that messages accepted by these
3065 statements are most likely to be submissions.
3067 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
3069 (1) Generate an error if the third argument for the ${prvs expansion is
3072 (2) Treat a missing third argument of ${prvscheck as if it were an empty
3075 (3) Reset the variables that are obtained from the first argument of
3076 ${prvscheck and used in the second argument before leaving the code,
3077 because their memory is reclaimed, so using them afterwards may do
3080 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
3081 one (it's much easier than Tom thought :-).
3083 (5) Because of (4), we can now allow for the use of $prvscheck_result
3084 inside the third argument.
3086 PH/03 For some reason, the default setting of PATH when running a command from
3087 a pipe transport was just "/usr/bin". I have changed it to
3090 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
3091 anything to be listed in the output from -bV.
3093 PH/05 When a filter generated an autoreply, the entire To: header line was
3094 quoted in the delivery log line, like this:
3096 => >A.N.Other <ano@some.domain> <original@ddress> ...
3098 This has been changed so that it extracts the operative address. There
3099 may be more than one such address. If so, they are comma-separated, like
3102 => >ano@some.domain,ona@other.domain <original@ddress> ...
3104 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
3105 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
3106 not being looked up in the rDNS to get a host name, Exim was showing the
3107 IP address twice in Received: lines, even though the IP addresses were
3108 identical. For example:
3110 Received: from [1.2.3.4] (helo=[1.2.3.4])
3112 However, if the real host name was known, it was omitting the HELO data
3113 if it matched the actual IP address. This has been tidied up so that it
3114 doesn't show the same IP address twice.
3116 PH/07 When both +timestamp and +memory debugging was on, the value given by
3117 $tod_xxx expansions could be wrong, because the tod_stamp() function was
3118 called by the debug printing, thereby overwriting the timestamp buffer.
3119 Debugging no longer uses the tod_stamp() function when +timestamp is set.
3121 PH/08 When the original message was included in an autoreply transport, it
3122 always said "this is a copy of the message, including all the headers",
3123 even if body_only or headers_only was set. It now gives an appropriate
3126 PH/09 Applied a patch from the Sieve maintainer which:
3128 o fixes some comments
3129 o adds the (disabled) notify extension core
3130 o adds some debug output for the result of if/elsif tests
3131 o points to the current vacation draft in the documentation
3132 and documents the missing references header update
3136 o fixes a bug in processing the envelope test (when testing
3137 multiple envelope elements, the last element determinted the
3140 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
3141 Electronic Mail") by including:
3143 Auto-submitted: auto-generated
3145 in the messages that it generates (bounce messages and others, such as
3146 warnings). In the case of bounce messages for non-SMTP mesages, there was
3147 also a typo: it was using "Auto_submitted" (underscore instead of
3148 hyphen). Since every message generated by Exim is necessarily in response
3149 to another message, thes have all been changed to:
3151 Auto-Submitted: auto-replied
3153 in accordance with these statements in the RFC:
3155 The auto-replied keyword:
3157 - SHOULD be used on messages sent in direct response to another
3158 message by an automatic process,
3160 - MUST NOT be used on manually-generated messages,
3162 - MAY be used on Delivery Status Notifications (DSNs) and Message
3163 Disposition Notifications (MDNs),
3165 - MUST NOT be used on messages generated by automatic or periodic
3166 processes, except for messages which are automatic responses to
3169 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
3170 to the default Received: header definition.
3172 PH/12 Added log selector acl_warn_skipped (default on).
3174 PH/13 After a successful wildlsearch lookup, discard the values of numeric
3175 variables because (a) they are in the wrong storage pool and (b) even if
3176 they were copied, it wouldn't work properly because of the caching.
3178 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
3179 checking when decoding. Apparently there are clients that generate
3180 overlong encoded strings. Why am I not surprised?
3182 PH/15 If the first argument of "${if match_address" was not empty, but did not
3183 contain an "@" character, Exim crashed. Now it writes a panic log message
3184 and treats the condition as false.
3186 PH/16 In autoreply, treat an empty string for "once" the same as unset.
3188 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
3189 extension "envelope-auth". The code is finished and in agreement with
3190 other implementations, but there is no documentation so far and in fact,
3191 nobody wrote the draft yet. This extension is currently #undef'ed, thus
3192 not changing the active code.
3194 Print executed "if" and "elsif" statements when debugging is used. This
3195 helps a great deal to understand what a filter does.
3197 Document more things not specified clearly in RFC3028. I had all this
3198 sorted out, when out of a sudden new issues came to my mind. Oops."
3200 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
3203 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
3204 canonical form (as documented). However, after a host name lookup from
3205 the IP address, check_host() was doing a simple string comparison with
3206 addresses acquired from the DNS when checking that the found name did
3207 have the original IP as one of its addresses. Since any found IPv6
3208 addresses are likely to be in abbreviated form, the comparison could
3209 fail. Luckily, there already exists a function for doing the comparison
3210 by converting both addresses to binary, so now that is used instead of
3211 the text comparison.
3213 PH/20 There was another similar case to PH/19, when a complete host name was
3214 given in a host list; looking up its IP address could give an abbreviated
3215 form, whereas the current host's name might or might not be abbreviated.
3216 The same fix has been applied.
3222 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
3223 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
3226 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
3227 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
3229 PH/03 Typo: missing ".o" in src/pcre/Makefile.
3231 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
3232 header line, restrict the check to what is listed in RFCs 2369 and 2929.
3233 Also, for "Auto-Submitted", treat anything other than "no" as
3234 non-personal, in accordance with RFC 3834. (Previously it treated
3235 anything starting "auto-" as non-personal.)
3237 TF/01 The control=submission/name=... option had a problem with syntax
3238 errors if the name included a slash character. The /name= option
3239 now slurps the rest of the string, so it can include any characters
3240 but it must come last in the list of options (after /sender_retain
3243 PH/05 Some modifications to the interface to the fake nameserver for the new
3251 TK/01 Added the "success_on_redirect" address verification option. See
3252 NewStuff for rationale and an example.
3254 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
3256 PH/02 Patch to exigrep to allow it to work on syslog lines.
3258 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
3259 fread() to skip over the body file's header line, because in Cygwin the
3260 header line is locked and is inaccessible.
3262 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
3263 co-exist for some time) to make it clear that it is the Exim ID that is
3264 referenced, not the Message-ID: header line.
3266 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
3267 string_format() function, because snprintf() does not exist on all
3270 PH/06 The use of forbid_filter_existstest now also locks out the use of the
3271 ${stat: expansion item.
3273 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
3274 protocol synchronization error", to keep the pedants happy.
3276 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
3277 well as for IRIX systems, when gcc is being used. See the host.c source
3280 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
3282 PH/10 Named domain lists were not working if used in a queue_smtp_domains
3285 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
3286 transport and to the smtp transport in LMTP mode.
3288 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
3290 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
3291 run a filter in a subprocess. This could lead to confusion in subsequent
3292 lookups in the parent process. There should also be a search_tidyup() at
3293 the end of the subprocess.
3295 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
3296 only if the host matched helo_try_verify_hosts, which caused the
3297 verification to occur when the EHLO/HELO command was issued. The ACL just
3298 tested the remembered result. Now, if a previous verification attempt has
3299 not happened, "verify = helo" does it there and then.
3301 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
3303 TK/03 Fix log output including CR from clamd.
3305 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
3306 exist provoked a memory error which could cause a segfault.
3308 PH/15 Installed PCRE 6.2
3310 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
3312 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
3313 of the problem. Specifically, suggested +O2 rather than +O1 for the
3316 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
3318 PH/20 If a delivery was routed to a non-standard port by means of an SRV
3319 record, the port was not correctly logged when the outgoing_port log
3320 selector was set (it logged the transort's default port).
3322 PH/21 Added support for host-specific ports to manualroute, queryprogram,
3323 fallback_hosts, and "hosts" in the smtp transport.
3325 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
3326 host errors such as "Connection refused".
3328 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
3329 authentication with radiusclient 0.4.9:
3331 - Error returned from rc_read_config was caught wrongly
3332 - Username/password not passed on to radius server due to wrong length.
3334 The presumption is that some radiusclient API changes for 4.51/PH/17
3335 were not taken care of correctly. The code is still untested by me (my
3336 Linux distribution still has 0.3.2 of radiusclient), but it was
3337 contributed by a Radius user.
3339 PH/24 When doing a callout, the value of $domain wasn't set correctly when
3340 expanding the "port" option of the smtp transport.
3342 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
3343 while reading a MIME header. Thanks to Tom Hughes for a patch.
3345 PH/24 Include config.h inside local_scan.h so that configuration settings are
3348 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
3349 that in an ACL for RCPT (for example), you can examine exactly what was
3352 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
3353 commands, but it was not correctly comparing the address with the actual
3354 client host address. Thus, it would show the EHLO address in Received:
3355 header lines when this was not necessary.
3357 PH/27 Added the % operator to ${eval:}.
3359 PH/28 Exim tries to create and chdir to its spool directory when it starts;
3360 it should be ignoring failures (because with -C, for example, it has lost
3361 privilege). It wasn't ignoring creation failures other than "already
3364 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
3367 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
3368 because an input line was too long, either on its own, or by virtue of
3369 too many continuations, the temporary file was not being removed, and the
3370 return code was incorrect.
3372 PH/31 Missing "BOOL" in function definition in filtertest.c.
3374 PH/32 Applied Sieve patches from the maintainer.
3376 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
3378 PH/33 Added "verify = not_blind".
3380 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
3381 Local/Makefile (with some defaults set). These are used in built scripts
3382 such as exicyclog, but they have never been used in the exim_install
3383 script (though there are many overriding facilities there). I have
3384 arranged that the exim_install script now takes note of these two
3387 PH/35 Installed configuration files for Dragonfly.
3389 PH/36 When a locally submitted message by a trusted user did not contain a
3390 From: header, and the sender address was obtained from -f or from an SMTP
3391 MAIL command, and the trusted user did not use -F to supply a sender
3392 name, $originator_name was incorrectly used when constructing a From:
3393 header. Furthermore, $originator_name was used for submission mode
3394 messages from external hosts without From: headers in a similar way,
3395 which is clearly wrong.
3397 PH/37 Added control=suppress_local_fixups.
3399 PH/38 When log_selector = +received_sender was set, and the addition of the
3400 sender made the log line's construction buffer exactly full, or one byte
3401 less than full, an overflow happened when the terminating "\n" was
3404 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
3405 when the result of a list match is failure because a DNS lookup failed.
3407 PH/40 RM_COMMAND is now used in the building process.
3409 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
3410 the "build-* directories that it finds.
3412 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
3413 address in a domain literal was a prefix of an interface address.
3415 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
3416 when verifying a sender address, unless rewrite_headers is false.
3418 PH/44 Wrote a long comment about why errors_to addresses are verified as
3419 recipients, not senders.
3421 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
3422 the ratelimit ACL was added.
3424 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
3426 PH/46 Added extra information about PostgreSQL errors to the error string.
3428 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
3429 avoiding the need to install special zones in a real server. This is
3430 backwards compatible; if it can't find the fake resolver, it drops back.
3431 Thus, both old and new test suites can be run.
3433 TF/02 Added util/ratelimit.pl
3435 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
3436 clock is set back in time.
3438 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
3439 Candler <B.Candler@pobox.com>.
3441 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
3442 OK for addresses that are the result of redirections.
3444 PH/48 A number of further additions for the benefit of the new test suite,
3445 including a fake gethostbyname() that interfaces to the fake DNS resolver
3448 TF/06 The fix for widen_domains has also been applied to qualify_single and
3449 search_parents which are the other dnslookup options that can cause
3452 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
3455 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
3457 TF/07 Exim produced the error message "an SRV record indicated no SMTP
3458 service" if it encountered an MX record with an empty target hostname.
3459 The message is now "an MX or SRV record indicated no SMTP service".
3461 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
3462 if the DNS of the sending site is misconfigured. This is quite a
3463 common situation. This change restores the behaviour of treating a
3464 helo verification defer as a failure.
3466 PH/51 If self=fail was set on a router, the bounce message did not include the
3467 actual error message.
3473 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
3475 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
3476 command itself ended in error, the underlying message about the transport
3477 filter timeout was being overwritten with the pipe command error. Now the
3478 underlying error message should be appended to the second error message.
3480 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
3482 PH/02 Reduce the amount of output that "make" produces by default. Full output
3483 can still be requested.
3485 PH/03 The warning log line about a condition test deferring for a "warn" verb
3486 was being output only once per connection, rather than after each
3487 occurrence (because it was using the same function as for successful
3488 "warn" verbs). This seems wrong, so I have changed it.
3490 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
3491 it should not have, which might have caused a crash in the right
3492 circumstances, but probably never did.
3494 PH/04 Installed a modified version of Tony Finch's patch to make submission
3495 mode fix the return path as well as the Sender: header line, and to
3496 add a /name= option so that you can make the user's friendly name appear
3499 TF/03 Added the control = fakedefer ACL modifier.
3501 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
3502 Mark Lowes for thorough testing.
3504 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
3506 TK/03 Merged latest SRS patch from Miles Wilton.
3508 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
3509 with the definition in sysexits.h (which is #included earlier).
3510 Fortunately, Exim does not actually use EX_OK. The code used to try to
3511 preserve the sysexits.h value, by assumimg that macro definitions were
3512 scanned for macro replacements. I have been disabused of this notion,
3513 so now the code just undefines EX_OK before #including unistd.h.
3515 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
3516 in the smtp transport. When a block could not be written in a single
3517 write() function, the timeout was being re-applied to each part-write.
3518 This seems wrong - if the receiver was accepting one byte at a time it
3519 would take for ever. The timeout is now adjusted when this happens. It
3520 doesn't have to be particularly precise.
3522 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
3523 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
3525 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
3526 without a domain if the check to <postmaster@domain> fails.
3528 SC/01 Eximstats: added -xls and the ability to specify output files
3529 (patch written by Frank Heydlauf).
3531 SC/02 Eximstats: use FileHandles for outputing results.
3533 SC/03 Eximstats: allow any combination of xls, txt, and html output.
3535 SC/04 Eximstats: fixed display of large numbers with -nvr option
3537 SC/05 Eximstats: fixed merging of reports with empty tables.
3539 SC/06 Eximstats: added the -include_original_destination flag
3541 SC/07 Eximstats: removed tabs and trailing whitespace.
3543 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
3545 TK/06 MBOX spool code: Add real "From " MBOX separator line
3546 so the .eml file is really in mbox format (even though
3547 most programs do not really care). Patch from Alex Miller.
3549 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
3550 The latter is generated from $received_to and is only set if the
3551 message has one envelope recipient. SA can use these headers,
3552 obviously out-of-the-box. Patch from Alex Miller.
3554 PH/08 The ${def test on a variable was returning false if the variable's
3555 value was "0", contrary to what the specification has always said!
3556 The result should be true unless the variable is empty.
3558 PH/09 The syntax error of a character other than { following "${if
3559 def:variable_name" (after optional whitespace) was not being diagnosed.
3560 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
3561 accidental colon was present, for example, could give incorrect results.
3563 PH/10 Tidied the code in a number of places where the st_size field of a stat()
3564 result is used (not including appendfile, where other changes are about
3567 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
3568 This involved changing a lot of size variables from int to off_t. It
3569 should work with maildirs and everything.
3571 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
3572 spamd dying while we are connected to it.
3574 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
3577 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
3578 function for BDB 4.3.
3580 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
3582 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
3583 the sources for PCRE have been re-arranged and more files are now
3586 PH/15 The code I had for printing potentially long long variables in PH/11
3587 above was not the best (it lost precision). The length of off_t variables
3588 is now inspected at build time, and an appropriate printing format (%ld
3589 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
3590 to be "long long int" or "long int". This is needed for the internal
3591 formatting function string_vformat().
3593 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
3594 the configuration file to be ":syslog", then the script "guesses" where
3595 the logs files are, rather than using the compiled in default. In our
3596 case the guess is not the same as the compiled default, so the script
3597 suddenly stopped working when I started to use syslog. The patch checks
3598 to see if log_file_path is "". If so, it attempts to read it from exim
3599 with no configuration file to get the compiled in version, before it
3600 falls back to the previous guessing code."
3602 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
3603 implementing BATV in an Exim configuration. See NewStuff for the gory
3606 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
3607 Makefile that are specific to HP-UX.
3609 PH/18 If the "use_postmaster" option was set for a recipient callout together
3610 with the "random" option, the postmaster address was used as the MAIL
3611 FROM address for the random test, but not for the subsequent recipient
3612 test. It is now used for both.
3614 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
3615 patch removes a few documentation additions to RFC 3028, because the
3616 latest draft now contains them. It adds the new en;ascii-case comparator
3617 and a new error check for 8bit text in MIME parts. Comparator and
3618 require names are now matched exactly. I enabled the subaddress
3619 extension, but it is not well tested yet (read: it works for me)."
3621 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
3622 rework some of the code of TK/09 above to avoid the hardwired use of
3623 "%lld" and "long long". Replaced the call to snprintf() with a call to
3626 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
3627 records point to non-existent hosts", "retry timeout exceeded", and
3628 "retry time not reached for any host after a long failure period".
3630 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
3631 experimental DomainKeys support:
3633 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
3634 (2) On an error such as an illegally used "control", the wrong name for
3635 the control was given.
3637 These problems did NOT occur unless DomainKeys support was compiled.
3639 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
3641 PH/24 Added ${if match_ip condition.
3643 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
3644 databases so that it will be absolutely obvious if a crash occurs in the
3645 DB library. This is a regular occurrence (often caused by mis-matched
3648 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
3649 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
3650 picked up on a user's system that detects such things. There doesn't seem
3651 to be a gcc warning option for this - only an attribute that has to be
3652 put on the function's prototype. It seems that in Fedora Core 4 they have
3653 set this on a number of new functions. No doubt there will be more in due
3656 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
3657 specify a transport. However, if an address that was verified by such a
3658 router was the subject of a callout, Exim crashed because it tried to
3659 read the rcpt_include_affixes from the non-existent transport. Now it
3660 just assumes that the setting of that option is false. This bug was
3661 introduced by 4.51/PH/31.
3663 PH/28 Changed -d+all to exclude +memory, because that information is very
3664 rarely of interest, but it makes the output a lot bigger. People tend to
3665 do -d+all out of habit.
3667 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
3668 code in os-type was giving problems when libc.so lives in lib64, like on
3671 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
3672 aren't the modern standard, and it seems that some systems' include files
3673 don't always have them. Exim was already checking for some of the newer
3674 ones like T_AAAA, and defining it itself. I've added checks for all the
3675 record types that Exim uses.
3677 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
3678 not automatically generating a new one, as it is supposed to. This
3679 prevented TLS from working. If the file did exist, but contained invalid
3680 data, a new version was generated, as expected. It was only the case of a
3681 non-existent file that was broken.
3683 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
3684 with a change in libdomainkeys > 0.64.
3686 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
3687 from DNS. If the selector record carries the flag, it now has
3688 precedence over the domain-wide flag.
3690 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
3692 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
3693 the use of a port name that isn't defined in /etc/services) occurred, the
3694 message was deferred as in a normal delivery, and thus remained on the
3695 spool, instead of being failed because of the mua_wrapper setting. This
3696 is now fixed, and I tidied up some of the mua_wrapper messages at the
3699 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
3700 the messages in a hash of arrays rather than using individual hashes.
3701 This is a bit cleaner and results in dramatic memory savings, albeit
3702 at a slight CPU cost.
3704 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
3705 as requested by Marc Sherman.
3707 SC/10 Eximstats: added histograms for user specified patterns as requested
3710 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
3712 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
3713 fopen() in the content-scanning modules that did not already have it.
3719 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
3720 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
3722 TK/02 Fix ACL "control" statement not being available in MIME ACL.
3724 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
3726 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
3727 to test Sieve filters that use "vacation".
3729 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
3730 that changes the way the GnuTLS parameters are stored in the cache file.
3731 The new format can be generated externally. For backward compatibility,
3732 if the data in the cache doesn't make sense, Exim assumes it has read an
3733 old-format file, and it generates new data and writes a new file. This
3734 means that you can't go back to an older release without removing the
3737 PH/03 A redirect router that has both "unseen" and "one_time" set does not
3738 work if there are any delivery delays because "one_time" forces the
3739 parent to be marked "delivered", so its unseen clone is never tried
3740 again. For this reason, Exim now forbids the simultaneous setting of
3743 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
3744 redirected to themselves ("homonym" addresses). Read the long ChangeLog
3745 entry if you want to know the details. The fix, however, neglected to
3746 consider the case when local delivery batching is involved. The test for
3747 "previously delivered" was not happening when checking to see if an
3748 address could be batched with a previous (undelivered) one; under
3749 certain circumstances this could lead to multiple deliveries to the same
3752 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
3753 in its include files, and this causes problems building Exim.
3755 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
3756 header_syntax) but Exim was just ignoring anything given after a slash.
3757 In particular, this caused confusion with an attempt to use "verify =
3758 reverse_host_lookup/defer_ok". An error is now given when options are
3759 supplied for verify items that do not have them. (Maybe reverse_host_
3760 lookup should have a defer_ok option, but that's a different point.)
3762 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
3763 defined by RFC 821) to 2048, because there were problems with some AUTH
3764 commands, and RFC 1869 says the size should be increased for extended
3765 SMTP commands that take arguments.
3767 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
3770 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
3771 "unknown" error; now it says that the functionality isn't in the binary.
3773 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
3774 an address' error message when a string expansion fails (syntax or
3775 whatever). Otherwise the password may appear in the log. Following change
3776 PH/42 below, there is no longer a chance of it appearing in a bounce
3779 PH/11 Installed exipick version 20050225.0 from John Jetmore.
3781 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
3782 of its addresses was ever tried. (Bugzilla bug #2.)
3784 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
3785 the result incorrectly in the debug output. (It correctly added a newline
3786 to what was transported.)
3788 TF/01 Added $received_time.
3790 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
3791 commented out examples of how to interface to a virus scanner and to
3792 SpamAssassin. Also added commented examples of av_scanner and
3793 spamd_address settings.
3795 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
3796 and controls are allowed in which ACLs. There were a couple of minor
3797 errors. Some of the entries in the conditions table (which is a table of
3798 where they are NOT allowed) were getting very unwieldy; rewrote them as a
3799 negation of where the condition IS allowed.
3801 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
3803 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
3804 header file does not have a version number, so I've had to invent a new
3805 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
3806 API. The code is untested by me (my Linux distribution still has 0.3.2 of
3807 radiusclient), but it was contributed by a Radius user.
3809 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
3810 files or directories, for OpenSSL.
3812 PH/19 When an Exim process that is running as root has to create an Exim log
3813 file, it does so in a subprocess that runs as exim:exim so as to get the
3814 ownership right at creation (otherwise, other Exim processes might see
3815 the file with the wrong ownership). There was no test for failure of this
3816 fork() call, which would lead to the process getting stuck as it waited
3817 for a non-existent subprocess. Forks do occasionally fail when resources
3818 run out. I reviewed all the other calls to fork(); they all seem to check
3821 PH/20 When checking for unexpected SMTP input at connect time (before writing
3822 the banner), Exim was not dealing correctly with a non-positive return
3823 from the read() function. If the client had disconnected by this time,
3824 the result was a log entry for a synchronization error with an empty
3825 string after "input=" when read() returned zero. If read() returned -1
3826 (an event I could not check), uninitialized data bytes were printed.
3827 There were reports of junk text (parts of files, etc) appearing after
3830 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
3832 PH/22 Added support for macro redefinition, and (re)definition in between
3833 driver and ACL definitions.
3835 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
3836 forgetting to use the resulting value; it was using the unexpanded value.
3838 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
3839 hadn't been configured. The fix is from Juergen Kreileder, who
3840 understands it better than I do:
3842 "Here's what I see happening with three configured cyrus_sasl
3843 authenticators configured (plain, login, cram-md5):
3845 On startup auth_cyrus_sasl_init() gets called for each of these.
3846 This means three calls to sasl_listmech() without a specified mech_list.
3847 => SASL tests which mechs of all available mechs actually work
3848 => three warnings about OTP not working
3849 => the returned list contains: plain, login, cram-md5, digest-md5, ...
3851 With the patch, sasl_listmech() also gets called three times. But now
3852 SASL's mech_list option is set to the server_mech specified in the the
3853 authenticator. Or in other words, the answer from sasl_listmech()
3854 gets limited to just the mech you're testing for (which is different
3856 => the return list contains just 'plain' or 'login', 'cram-md5' or
3857 nothing depending on the value of ob->server_mech.
3859 I've just tested the patch: Authentication still works fine,
3860 unavailable mechs specified in the exim configuration are still
3861 caught, and the auth.log warnings about OTP are gone."
3863 PH/25 When debugging is enabled, the contents of the command line are added
3864 to the debugging output, even when log_selector=+arguments is not
3867 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
3868 answer is "GNU", and only if the return is "GNU/something" is the answer
3871 PH/27 $acl_verify_message is now set immediately after the failure of a
3872 verification in an ACL, and so is available in subsequent modifiers. In
3873 particular, the message can be preserved by coding like this:
3875 warn !verify = sender
3876 set acl_m0 = $acl_verify_message
3878 Previously, $acl_verify_message was set only while expanding "message"
3879 and "log_message" when a very denied access.
3881 PH/28 Modified OS/os.c-Linux with
3883 -#ifndef OS_LOAD_AVERAGE
3884 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
3886 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
3887 nomenclature these days.)
3889 PH/29 Installed patch from the Sieve maintainer that adds the options
3890 sieve_useraddress and sieve_subaddress to the redirect router.
3892 PH/30 In these circumstances:
3893 . Two addresses routed to the same list of hosts;
3894 . First host does not offer TLS;
3895 . First host accepts first address;
3896 . First host gives temporary error to second address;
3897 . Second host offers TLS and a TLS session is established;
3898 . Second host accepts second address.
3899 Exim incorrectly logged both deliveries with the TLS parameters (cipher
3900 and peerdn, if requested) that were in fact used only for the second
3903 PH/31 When doing a callout as part of verifying an address, Exim was not paying
3904 attention to any local part prefix or suffix that was matched by the
3905 router that accepted the address. It now behaves in the same way as it
3906 does for delivery: the affixes are removed from the local part unless
3907 rcpt_include_affixes is set on the transport.
3909 PH/32 Add the sender address, as F=<...>, to the log line when logging a
3910 timeout during the DATA phase of an incoming message.
3912 PH/33 Sieve envelope tests were broken for match types other than :is. I have
3913 applied a patch sanctioned by the Sieve maintainer.
3915 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
3916 the uid or gid is negative. A case of a negative gid caused this to be
3917 noticed. The fix allows for either to be negative.
3919 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
3920 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
3923 PH/36 The change PH/12 above was broken. Fixed it.
3925 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
3926 the grounds that routing the same address twice would always produce the
3927 same answer. This might have been true once, but it is certainly no
3928 longer true now. Routing a child address may depend on the previous
3929 routing that produced that child. Some complicated redirection strategies
3930 went wrong when messages had multiple recipients, and made Exim's
3931 behaviour dependent on the order in which the addresses were given.
3933 I have moved the duplicate checking until after the routing is complete.
3934 Exim scans the addresses that are assigned to local and remote
3935 transports, and removes any duplicates. This means that more work will be
3936 done, as duplicates will always all be routed, but duplicates are
3937 presumably rare, so I don't expect this is of any significance.
3939 For deliveries to pipes, files, and autoreplies, the duplicate checking
3940 still happens during the routing process, since they are not going to be
3943 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
3944 It corrects a timeout issue with spamd. This is Ian's comment: "The
3945 background is that sometimes spamd either never reads data from a
3946 connection it has accepted, or it never writes response data. The exiscan
3947 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
3948 blindly assumes that writes won't block so it may never time out."
3950 PH/39 Allow G after quota size as well as K and M.
3952 PH/40 The value set for $authenticated_id in an authenticator may not contain
3953 binary zeroes or newlines because the value is written to log lines and
3954 to spool files. There was no check on this. Now the value is run through
3955 the string_printing() function so that such characters are converted to
3956 printable escape sequences.
3958 PH/41 $message_linecount is a new variable that contains the total number of
3959 lines in the message. Compare $body_linecount, which is the count for the
3962 PH/42 Exim no longer gives details of delivery errors for specific addresses in
3963 bounce and delay warning messages, except in certain special cases, which
3966 (a) An SMTP error message from a remote host;
3967 (b) A message specified in a :fail: redirection;
3968 (c) A message specified in a "fail" command in a system filter;
3969 (d) A message specified in a FAIL return from the queryprogram router;
3970 (e) A message specified by the cannot_route_message router option.
3972 In these cases only, Exim does include the error details in bounce and
3973 warning messages. There are also a few cases where bland messages such
3974 as "unrouteable address" or "local delivery error" are given.
3976 PH/43 $value is now also set for the "else" part of a ${run expansion.
3978 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
3979 being worked on, but at least Exim now implements the latest version to
3982 PH/45 In a pipe transport, although a timeout while waiting for the pipe
3983 process to complete was treated as a delivery failure, a timeout while
3984 writing the message to the pipe was logged, but erroneously treated as a
3985 successful delivery. Such timeouts include transport filter timeouts. For
3986 consistency with the overall process timeout, these timeouts are now
3987 treated as errors, giving rise to delivery failures by default. However,
3988 there is now a new Boolean option for the pipe transport called
3989 timeout_defer, which, if set TRUE, converts the failures into defers for
3990 both kinds of timeout. A transport filter timeout is now identified in
3993 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
3994 systems where "make" and "gmake" are different, calling "gmake" at top
3995 level broke things. I've arranged for the value of $(MAKE) to be passed
3996 from the Makefile to this script so that it can call the same version of
4000 A note about Exim versions 4.44 and 4.50
4001 ----------------------------------------
4003 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
4004 changes of various kinds. As a consequence, a big documentation update was
4005 needed. This delayed the release for rather longer than seemed good, especially
4006 in the light of a couple of (minor) security issues. Therefore, the changes
4007 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
4008 release. So 4.44 and 4.50 are in effect two different branches that both start
4011 I have left the 4.50 change log unchanged; it contains all the changes since
4012 4.43. The change log for 4.44 is below; many of its items are identical to
4013 those for 4.50. This seems to be the most sensible way to preserve the
4014 historical information.
4020 1. Minor wording change to the doc/README.SIEVE file.
4022 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
4023 computation of the current number of files was incorrect.
4025 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4026 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4029 4. Give more explanation in the error message when the command for a transport
4030 filter fails to execute.
4032 5. There are several places where Exim runs a non-Exim command in a
4033 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4034 was being done only for the command run by the queryprogram router. It is
4035 now done for all such subprocesses. The other cases are: ${run, transport
4036 filters, and the commands run by the lmtp and pipe transports.
4038 6. Added CONFIGURE_GROUP build-time option.
4040 7. Some older OS have a limit of 256 on the maximum number of file
4041 descriptors. Exim was using setrlimit() to set 1000 as a large value
4042 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4043 systems. I've change it so that if it can't get 1000, it tries for 256.
4045 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4046 was an oversight, and furthermore, ever since the addition of extra
4047 controls (e.g. 4.43/32), the checks on when to allow different forms of
4048 "control" were broken. There should now be diagnostics for all cases when a
4049 control that does not make sense is encountered.
4051 9. Added the /retain_sender option to "control=submission".
4053 10. $recipients is now available in the predata ACL (oversight).
4055 11. Tidy the search cache before the fork to do a delivery from a message
4056 received from the command line. Otherwise the child will trigger a lookup
4057 failure and thereby defer the delivery if it tries to use (for example) a
4058 cached ldap connection that the parent has called unbind on.
4060 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4061 of $address_data from the recipient verification was clobbered by the
4062 sender verification.
4064 13. The value of address_data from a sender verification is now available in
4065 $sender_address_data in subsequent conditions in the ACL statement.
4067 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
4069 15. Added a new option "connect=<time>" to callout options, to set a different
4072 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4073 was its contents. (It was OK if the option was not defined at all.)
4075 17. A "Completed" log line is now written for messages that are removed from
4076 the spool by the -Mrm option.
4078 18. New variables $sender_verify_failure and $recipient_verify_failure contain
4079 information about exactly what failed.
4081 19. Added -dd to debug only the daemon process.
4083 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
4084 handles timeouts, both on the server side and network timeouts. Renamed the
4085 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
4087 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
4088 It is now set to "smtps".
4090 22. $host_address is now set to the target address during the checking of
4091 ignore_target_hosts.
4093 23. When checking ignore_target_hosts for an ipliteral router, no host name was
4094 being passed; this would have caused $sender_host_name to have been used if
4095 matching the list had actually called for a host name (not very likely,
4096 since this list is usually IP addresses). A host name is now passed as
4099 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4100 code that specifies a non-restarting handler (typically sigaction() in
4101 modern systems) in an attempt to fix a rare and obscure crash bug.
4103 25. Narrowed the window for a race in the daemon that could cause it to ignore
4104 SIGCHLD signals. This is not a major problem, because they are used only to
4105 wake it up if nothing else does.
4107 26. A malformed maildirsize file could cause Exim to calculate negative values
4108 for the mailbox size or file count. Odd effects could occur as a result.
4109 The maildirsize information is now recalculated if the size or filecount
4112 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4113 support for a long time. Removed HAVE_SYS_VFS_H.
4115 28. Installed the latest version of exipick from John Jetmore.
4117 29. In an address list, if the pattern was not a regular expression, an empty
4118 subject address (from a bounce message) matched only if the pattern was an
4119 empty string. Non-empty patterns were not even tested. This was the wrong
4120 because it is perfectly reasonable to use an empty address as part of a
4121 database query. An empty address is now tested by patterns that are
4122 lookups. However, all the other forms of pattern expect the subject to
4123 contain a local part and a domain, and therefore, for them, an empty
4124 address still always fails if the pattern is not itself empty.
4126 30. Exim went into a mad DNS loop when attempting to do a callout where the
4127 host was specified on an smtp transport, and looking it up yielded more
4128 than one IP address.
4130 31. Re-factored the code for checking spool and log partition space into a
4131 function that finds that data and another that does the check. The former
4132 is then used to implement four new variables: $spool_space, $log_space,
4133 $spool_inodes, and $log_inodes.
4135 32. The RFC2047 encoding function was originally intended for short strings
4136 such as real names; it was not keeping to the 75-character limit for
4137 encoded words that the RFC imposes. It now respects the limit, and
4138 generates multiple encoded words if necessary. To be on the safe side, I
4139 have increased the buffer size for the ${rfc2047: expansion operator from
4142 33. It is now permitted to omit both strings after an "if" condition; if the
4143 condition is true, the result is "true". As before, when the second string
4144 is omitted, a false condition yields an empty string. This makes it less
4145 cumbersome to write custom ACL and router conditions.
4147 34. Failure to deliver a bounce message always caused it to be frozen, even if
4148 there was an errors_to setting on the router. The errors_to setting is now
4151 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4152 canonical form (fully expanded) before being placed in
4153 $sender_host_address.
4155 36. The table in the code that translates DNS record types into text (T_A to
4156 "A" for instance) was missing entries for NS and CNAME. It is just possible
4157 that this could have caused confusion if both these types were looked up
4158 for the same domain, because the text type is used as part of Exim's
4159 per-process caching. But the chance of anyone hitting this buglet seems
4162 37. The dnsdb lookup has been extended in a number of ways.
4164 (1) There is a new type, "zns", which walks up the domain tree until it
4165 finds some nameserver records. It should be used with care.
4167 (2) There is a new type, "mxh", which is like "mx" except that it returns
4168 just the host names, not the priorities.
4170 (3) It is now possible to give a list of domains (or IP addresses) to be
4171 looked up. The behaviour when one of the lookups defers can be
4172 controlled by a keyword.
4174 (4) It is now possible to specify the separator character for use when
4175 multiple records are returned.
4177 38. The dnslists ACL condition has been extended: it is now possible to supply
4178 a list of IP addresses and/or domains to be looked up in a particular DNS
4181 39. Added log_selector=+queue_time_overall.
4183 40. When running the queue in the test harness, wait just a tad after forking a
4184 delivery process, to get repeatability of debugging output.
4186 41. Include certificate and key file names in error message when GnuTLS fails
4187 to set them up, because the GnuTLS error message doesn't include the name
4188 of the failing file when there is a problem reading it.
4190 42. Allow both -bf and -bF in the same test run.
4192 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
4194 44. The "Exiscan patch" is now merged into the mainline Exim source.
4196 45. Sometimes the final signoff response after QUIT could fail to get
4197 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4198 < 0 before doing a fflush(). This bug looks as though it goes back to the
4199 introduction of TLS in release 3.20, but "sometimes" must have been rare
4200 because the tests only now provoked it.
4202 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
4203 (this can affect the format of dates).
4205 47. exim_tidydb, when checking for the continued existence of a message for
4206 which it has found a message-specific retry record, was not finding
4207 messages that were in split spool directories. Consequently, it was
4208 deleting retry records that should have stayed in existence.
4210 48. Steve fixed some bugs in eximstats.
4212 49. The SPA authentication driver was not abandoning authentication and moving
4213 on to the next authenticator when an expansion was forced to fail,
4214 contradicting the general specification for all authenticators. Instead it
4215 was generating a temporary error. It now behaves as specified.
4217 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
4218 (the order specifies the preference for clients). The order is now AES256,
4219 AES128, 3DES, ARCFOUR128.
4221 51. Small patch to Sieve code - explicitly set From: when generating an
4224 52. Exim crashed if a remote delivery caused a very long error message to be
4225 recorded - for instance if somebody sent an entire SpamAssassin report back
4226 as a large number of 550 error lines. This bug was coincidentally fixed by
4227 increasing the size of one of Exim's internal buffers (big_buffer) that
4228 happened as part of the Exiscan merge. However, to be on the safe side, I
4229 have made the code more robust (and fixed the comments that describe what
4232 53. Now that there can be additional text after "Completed" in log lines (if
4233 the queue_time_overall log selector is set), a one-byte patch to exigrep
4234 was needed to allow it to recognize "Completed" as not the last thing in
4237 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
4238 patch that reportedly fixes this has been added. I am not expert enough to
4239 create a test for it. This is what the patch creator wrote:
4241 "I found a little strange behaviour of ldap code when working with
4242 Windows 2003 AD Domain, where users was placed in more than one
4243 Organization Units. When I tried to give exim partial DN, the exit code
4244 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
4245 But simultaneously result of request was absolutely normal ldap result,
4246 so I produce this patch..."
4248 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
4249 REFERENCE, so I have modified the code to exclude the patch when that macro
4252 55. Some experimental protocols are using DNS PTR records for new purposes. The
4253 keys for these records are domain names, not reversed IP addresses. The
4254 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4255 leaves it alone. Component reversal etc. now happens only for IP addresses.
4258 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
4260 57. Double the size of the debug message buffer (to 2048) so that more of very
4261 long debug lines gets shown.
4263 58. The exicyclog utility now does better if the number of log files to keep
4264 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
4266 59. Two changes related to the smtp_active_hostname option:
4268 (1) $smtp_active_hostname is now available as a variable.
4269 (2) The default for smtp_banner uses $smtp_active_hostname instead
4270 of $primary_hostname.
4272 60. The host_aton() function is supposed to be passed a string that is known
4273 to be a valid IP address. However, in the case of IPv6 addresses, it was
4274 not checking this. This is a hostage to fortune. Exim now panics and dies
4275 if the condition is not met. A case was found where this could be provoked
4276 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4277 components; fortuitously, this particular loophole had already been fixed
4278 by change 4.50/55 above.
4280 If there are any other similar loopholes, the new check in host_aton()
4281 itself should stop them being exploited. The report I received stated that
4282 data on the command line could provoke the exploit when Exim was running as
4283 exim, but did not say which command line option was involved. All I could
4284 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4285 running as the user.
4288 61. There was a buffer overflow vulnerability in the SPA authentication code
4289 (which came originally from the Samba project). I have added a test to the
4290 spa_base64_to_bits() function which I hope fixes it.
4293 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
4294 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
4296 63. The daemon start-up calls getloadavg() while still root for those OS that
4297 need the first call to be done as root, but it missed one case: when
4298 deliver_queue_load_max is set with deliver_drop_privilege. This is
4299 necessary for the benefit of the queue runner, because there is no re-exec
4300 when deliver_drop_privilege is set.
4302 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
4303 This has been fixed.
4305 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4306 was in use, was not putting the data itself into the right store pool;
4307 consequently, it could be overwritten for a subsequent message in the same
4308 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4311 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
4313 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
4314 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
4315 the function were treating the return as a boolean value, which happened to
4316 work because 0=false and not-0=true, but is not correct code.
4318 68. The host_aton() function was not handling scoped IPv6 addresses (those
4319 with, for example, "%eth0" on the end) correctly.
4321 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
4322 negated items (that is, ~something) in unsigned ints. Some compilers
4323 apparently mutter when there is no cast.
4325 70. If an address verification called from an ACL failed, and did not produce a
4326 user-specific message (i.e. there was only a "system" message), nothing was
4327 put in $acl_verify_message. In this situation, it now puts the system
4330 71. Change 4.23/11 added synchronization checking at the start of an SMTP
4331 session; change 4.31/43 added the unwanted input to the log line - except
4332 that it did not do this in the start of session case. It now does.
4334 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
4335 This is wrong and can cause the other end to generate a synchronization
4336 error if it is another Exim or anything else that does the synchronization
4337 check. A QUIT command is no longer sent after a timeout.
4339 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
4340 during host lookups.
4342 74. The defer_ok option of callout verification was not working if it was used
4343 when verifying addresses in header lines, that is, for this case:
4345 verify = header_sender/callout=defer_ok
4347 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
4348 those file descriptors could be used for SMTP connections. If anything
4349 wrote to stderr (the example that came up was "warn" in embedded Perl), it
4350 could be sent to the SMTP client, causing chaos. The daemon now opens
4351 stdin, stdout, and stderr to /dev/null when it puts itself into the
4354 76. Arrange for output from Perl's "warn" command to be written to Exim's main
4355 log by default. The user can override this with suitable Perl magic.
4357 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
4358 the log message when discard triggers, was not working for the DATA ACL or
4359 for the non-SMTP ACL.
4361 78. Error message wording change in sieve.c.
4363 79. If smtp_accept_max_per_host was set, the number of connections could be
4364 restricted to fewer than expected, because the daemon was trying to set up
4365 a new connection before checking whether the processes handling previous
4366 connections had finished. The check for completed processes is now done
4367 earlier. On busy systems, this bug wouldn't be noticed because something
4368 else would have woken the daemon, and it would have reaped the completed
4371 80. If a message was submitted locally by a user whose login name contained one
4372 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
4373 It caused a spool format error. I have fixed the spool reading code. A
4374 related problem was that the "from" clause in the Received: line became
4375 illegal because of the space(s). It is now covered by ${quote_local_part.
4377 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
4380 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
4382 83. Patch from Sieve maintainer to fix unterminated string problem in
4383 "vacation" handling.
4385 84. Some minor changes to the Linux configuration files to help with other
4386 OS variants using glibc.
4388 85. One more patch for Sieve to update vacation handling to latest spec.
4391 ----------------------------------------------------
4392 See the note above about the 4.44 and 4.50 releases.
4393 ----------------------------------------------------
4399 1. Change 4.43/35 introduced a bug that caused file counts to be
4400 incorrectly computed when quota_filecount was set in an appendfile
4403 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4404 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4407 3. Give more explanation in the error message when the command for a transport
4408 filter fails to execute.
4410 4. There are several places where Exim runs a non-Exim command in a
4411 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4412 was being done only for the command run by the queryprogram router. It is
4413 now done for all such subprocesses. The other cases are: ${run, transport
4414 filters, and the commands run by the lmtp and pipe transports.
4416 5. Some older OS have a limit of 256 on the maximum number of file
4417 descriptors. Exim was using setrlimit() to set 1000 as a large value
4418 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4419 systems. I've change it so that if it can't get 1000, it tries for 256.
4421 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4422 was an oversight, and furthermore, ever since the addition of extra
4423 controls (e.g. 4.43/32), the checks on when to allow different forms of
4424 "control" were broken. There should now be diagnostics for all cases when a
4425 control that does not make sense is encountered.
4427 7. $recipients is now available in the predata ACL (oversight).
4429 8. Tidy the search cache before the fork to do a delivery from a message
4430 received from the command line. Otherwise the child will trigger a lookup
4431 failure and thereby defer the delivery if it tries to use (for example) a
4432 cached ldap connection that the parent has called unbind on.
4434 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4435 of $address_data from the recipient verification was clobbered by the
4436 sender verification.
4438 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4439 was its contents. (It was OK if the option was not defined at all.)
4441 11. A "Completed" log line is now written for messages that are removed from
4442 the spool by the -Mrm option.
4444 12. $host_address is now set to the target address during the checking of
4445 ignore_target_hosts.
4447 13. When checking ignore_target_hosts for an ipliteral router, no host name was
4448 being passed; this would have caused $sender_host_name to have been used if
4449 matching the list had actually called for a host name (not very likely,
4450 since this list is usually IP addresses). A host name is now passed as
4453 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4454 code that specifies a non-restarting handler (typically sigaction() in
4455 modern systems) in an attempt to fix a rare and obscure crash bug.
4457 15. Narrowed the window for a race in the daemon that could cause it to ignore
4458 SIGCHLD signals. This is not a major problem, because they are used only to
4459 wake it up if nothing else does.
4461 16. A malformed maildirsize file could cause Exim to calculate negative values
4462 for the mailbox size or file count. Odd effects could occur as a result.
4463 The maildirsize information is now recalculated if the size or filecount
4466 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4467 support for a long time. Removed HAVE_SYS_VFS_H.
4469 18. Updated exipick to current release from John Jetmore.
4471 19. Allow an empty sender to be matched against a lookup in an address list.
4472 Previously the only cases considered were a regular expression, or an
4475 20. Exim went into a mad DNS lookup loop when doing a callout where the
4476 host was specified on the transport, if the DNS lookup yielded more than
4479 21. The RFC2047 encoding function was originally intended for short strings
4480 such as real names; it was not keeping to the 75-character limit for
4481 encoded words that the RFC imposes. It now respects the limit, and
4482 generates multiple encoded words if necessary. To be on the safe side, I
4483 have increased the buffer size for the ${rfc2047: expansion operator from
4486 22. Failure to deliver a bounce message always caused it to be frozen, even if
4487 there was an errors_to setting on the router. The errors_to setting is now
4490 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4491 canonical form (fully expanded) before being placed in
4492 $sender_host_address.
4494 24. Updated eximstats to version 1.33
4496 25. Include certificate and key file names in error message when GnuTLS fails
4497 to set them up, because the GnuTLS error message doesn't include the name
4498 of the failing file when there is a problem reading it.
4500 26. Expand error message when OpenSSL has problems setting up cert/key files.
4503 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
4504 (this can affect the format of dates).
4506 28. exim_tidydb, when checking for the continued existence of a message for
4507 which it has found a message-specific retry record, was not finding
4508 messages that were in split spool directories. Consequently, it was
4509 deleting retry records that should have stayed in existence.
4511 29. eximstats updated to version 1.35
4512 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
4513 1.35 - bugfix such that pie charts by volume are generated correctly
4515 30. The SPA authentication driver was not abandoning authentication and moving
4516 on to the next authenticator when an expansion was forced to fail,
4517 contradicting the general specification for all authenticators. Instead it
4518 was generating a temporary error. It now behaves as specified.
4520 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
4521 (the order specifies the preference for clients). The order is now AES256,
4522 AES128, 3DES, ARCFOUR128.
4524 31. Small patch to Sieve code - explicitly set From: when generating an
4527 32. Exim crashed if a remote delivery caused a very long error message to be
4528 recorded - for instance if somebody sent an entire SpamAssassin report back
4529 as a large number of 550 error lines. This bug was coincidentally fixed by
4530 increasing the size of one of Exim's internal buffers (big_buffer) that
4531 happened as part of the Exiscan merge. However, to be on the safe side, I
4532 have made the code more robust (and fixed the comments that describe what
4535 33. Some experimental protocols are using DNS PTR records for new purposes. The
4536 keys for these records are domain names, not reversed IP addresses. The
4537 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4538 leaves it alone. Component reversal etc. now happens only for IP addresses.
4541 34. The host_aton() function is supposed to be passed a string that is known
4542 to be a valid IP address. However, in the case of IPv6 addresses, it was
4543 not checking this. This is a hostage to fortune. Exim now panics and dies
4544 if the condition is not met. A case was found where this could be provoked
4545 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4546 components; fortuitously, this particular loophole had already been fixed
4547 by change 4.50/55 or 4.44/33 above.
4549 If there are any other similar loopholes, the new check in host_aton()
4550 itself should stop them being exploited. The report I received stated that
4551 data on the command line could provoke the exploit when Exim was running as
4552 exim, but did not say which command line option was involved. All I could
4553 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4554 running as the user.
4557 35. There was a buffer overflow vulnerability in the SPA authentication code
4558 (which came originally from the Samba project). I have added a test to the
4559 spa_base64_to_bits() function which I hope fixes it.
4562 36. The daemon start-up calls getloadavg() while still root for those OS that
4563 need the first call to be done as root, but it missed one case: when
4564 deliver_queue_load_max is set with deliver_drop_privilege. This is
4565 necessary for the benefit of the queue runner, because there is no re-exec
4566 when deliver_drop_privilege is set.
4568 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4569 was in use, was not putting the data itself into the right store pool;
4570 consequently, it could be overwritten for a subsequent message in the same
4571 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4574 38. Sometimes the final signoff response after QUIT could fail to get
4575 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4576 < 0 before doing a fflush(). This bug looks as though it goes back to the
4577 introduction of TLS in release 3.20, but "sometimes" must have been rare
4578 because the tests only now provoked it.
4584 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
4585 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
4586 Now it returns an integer. A number of calls were still expecting a T/F
4587 return. Fortuitously, in all cases, the tests worked in OK situations,
4588 which is the norm. However, things would have gone wrong on any write
4589 failures on the smtp file descriptor. This function is used when sending
4590 messages over SMTP and also when doing verify callouts.
4592 2. When Exim is called to do synchronous delivery of a locally submitted
4593 message (the -odf or -odi options), it no longer closes stderr before doing
4596 3. Implemented the mua_wrapper option.
4598 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
4600 5. Implemented the functions header_remove(), header_testname(),
4601 header_add_at_position(), and receive_remove_recipient(), and exported them
4604 6. If an ACL "warn" statement specified the addition of headers, Exim already
4605 inserted X-ACL-Warn: at the start if there was no header name. However, it
4606 was not making this test for the second and subsequent header lines if
4607 there were newlines in the string. This meant that an invalid header could
4608 be inserted if Exim was badly configured.
4610 7. Allow an ACL "warn" statement to add header lines at the start or after all
4611 the Received: headers, as well as at the end.
4613 8. Added the rcpt_4xx retry error code.
4615 9. Added postmaster_mailfrom=xxx to callout verification option.
4617 10. Added mailfrom=xxxx to the callout verification option, for verify=
4620 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
4621 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
4623 12. Inserted some casts to stop certain compilers warning when using pointer
4624 differences as field lengths or precisions in printf-type calls (mostly
4625 affecting debugging statements).
4627 13. Added optional readline() support for -be (dynamically loaded).
4629 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
4630 same clock tick as a message's arrival, so that its received time was the
4631 same as the "first fail" time on the retry record, and that message
4632 remained on the queue past the ultimate address timeout, every queue runner
4633 would try a delivery (because it was past the ultimate address timeout) but
4634 after another failure, the ultimate address timeout, which should have then
4635 bounced the address, did not kick in. This was a "< instead of <=" error;
4636 in most cases the first failure would have been in the next clock tick
4637 after the received time, and all would be well.
4639 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
4640 being recognized when the domain list was tested by the match_domain
4641 condition in an expansion string.
4643 16. Added the ${str2b64: operator.
4645 17. Exim was always calling setrlimit() to set a large limit for the number of
4646 processes, without checking whether the existing limit was already
4647 adequate. (It did check for the limit on file descriptors.) Furthermore,
4648 errors from getrlimit() and setrlimit() were being ignored. Now they are
4649 logged to the main and panic logs, but Exim does carry on, to try to do its
4650 job under whatever limits there are.
4652 18. Imported PCRE 5.0.
4654 19. Trivial typo in log message " temporarily refused connection" (the leading
4657 20. If the log selector return_path_on_delivery was set and an address was
4658 redirected to /dev/null, the delivery process crashed because it assumed
4659 that a return path would always be set for a "successful" delivery. In this
4660 case, the whole delivery is bypassed as an optimization, and therefore no
4663 21. Internal re-arrangement: the function for sending a challenge and reading
4664 a response while authentication was assuming a zero-terminated challenge
4665 string. It's now changed to take a pointer and a length, to allow for
4666 binary data in such strings.
4668 22. Added the cyrus_sasl authenticator (code supplied by MBM).
4670 23. Exim was not respecting finduser_retries when seeking the login of the
4671 uid under which it was called; it was always trying 10 times. (The default
4672 setting of finduser_retries is zero.) Also, it was sleeping after the final
4673 failure, which is pointless.
4675 24. Implemented tls_on_connect_ports.
4677 25. Implemented acl_smtp_predata.
4679 26. If the domain in control=submission is set empty, Exim assumes that the
4680 authenticated id is a complete email address when it generates From: or
4681 Sender: header lines.
4683 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
4684 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
4685 chown and chgrp in /bin and hostname in /usr/bin.
4687 28. Exim was keeping the "process log" file open after each use, just as it
4688 does for the main log. This opens the possibility of it remaining open for
4689 long periods when the USR1 signal hits a daemon. Occasional processlog
4690 errors were reported, that could have been caused by this. Anyway, it seems
4691 much more sensible not to leave this file open at all, so that is what now
4694 29. The long-running daemon process does not normally write to the log once it
4695 has entered its main loop, and it closes the log before doing so. This is
4696 so that log files can straightforwardly be renamed and moved. However,
4697 there are a couple of unusual error situations where the daemon does write
4698 log entries, and I had neglected to close the log afterwards.
4700 30. The text of an SMTP error response that was received during a remote
4701 delivery was being truncated at 512 bytes. This is too short for some of
4702 the long messages that one sometimes sees. I've increased the limit to
4705 31. It is now possible to make retry rules that apply only when a message has a
4706 specific sender, in particular, an empty sender.
4708 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
4709 it possible to be selective about when SMTP synchronization is enforced.
4711 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
4713 32. Implemented hosts_connection_nolog.
4715 33. Added an ACL for QUIT.
4717 34. Setting "delay_warning=" to disable warnings was not working; it gave a
4720 35. Added mailbox_size and mailbox_filecount to appendfile.
4722 36. Added control = no_multiline_responses to ACLs.
4724 37. There was a bug in the logic of the code that waits for the clock to tick
4725 in the case where the clock went backwards by a substantial amount such
4726 that the microsecond fraction of "now" was more than the microsecond
4727 fraction of "then" (but the whole seconds number was less).
4729 38. Added support for the libradius Radius client library this is found on
4730 FreeBSD (previously only the radiusclient library was supported).
4736 1. When certain lookups returned multiple values in the form name=value, the
4737 quoting of the values was not always being done properly. Specifically:
4738 (a) If the value started with a double quote, but contained no whitespace,
4740 (b) If the value contained whitespace other than a space character (i.e.
4741 tabs or newlines or carriage returns) it was not quoted.
4742 This fix has been applied to the mysql and pgsql lookups by writing a
4743 separate quoting function and calling it from the lookup code. The fix
4744 should probably also be applied to nisplus, ibase and oracle lookups, but
4745 since I cannot test any of those, I have not disturbed their existing code.
4747 2. A hit in the callout cache for a specific address caused a log line with no
4748 reason for rejecting RCPT. Now it says "Previous (cached) callout
4749 verification failure".
4751 3. There was an off-by-one bug in the queryprogram router. An over-long
4752 return line was truncated at 256 instead of 255 characters, thereby
4753 overflowing its buffer with the terminating zero. As well as fixing this, I
4754 have increased the buffer size to 1024 (and made a note to document this).
4756 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
4757 when Exim is waiting for an SMTP response from a remote server, Exim
4758 restarts its select() call on the socket, thereby resetting its timeout.
4759 This is not a problem when such interrupts are rare. Somebody set up a cron
4760 job to run exiwhat every 2 minutes, which is less than the normal select()
4761 timeout (5 or 10 minutes). This meant that the select() timeout never
4762 kicked in because it was always reset. I have fixed this by comparing the
4763 time when an interrupt arrives with the time at the start of the first call
4764 to select(). If more time than the timeout has elapsed, the interrupt is
4765 treated as a timeout.
4767 5. Some internal re-factoring in preparation for the addition of Sieve
4768 extensions (by MH). In particular, the "personal" test is moved to a
4769 separate function, and given an option for scanning Cc: and Bcc: (which is
4770 not set for Exim filters).
4772 6. When Exim created an email address using the login of the caller as the
4773 local part (e.g. when creating a From: or Sender: header line), it was not
4774 quoting the local part when it contained special characters such as @.
4776 7. Installed new OpenBSD configuration files.
4778 8. Reworded some messages for syntax errors in "and" and "or" conditions to
4779 try to make them clearer.
4781 9. Callout options, other than the timeout value, were being ignored when
4782 verifying sender addresses in header lines. For example, when using
4784 verify = header_sender/callout=no_cache
4786 the cache was (incorrectly) being used.
4788 10. Added a missing instance of ${EXE} to the exim_install script; this affects
4789 only the Cygwin environment.
4791 11. When return_path_on_delivery was set as a log selector, if different remote
4792 addresses in the same message used different return paths and parallel
4793 remote delivery occurred, the wrong values would sometimes be logged.
4794 (Whenever a remote delivery process finished, the return path value from
4795 the most recently started remote delivery process was logged.)
4797 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
4798 lines when AUTH and/or TLS are in use. This is the "received protocol"
4799 field. Exim used to use "asmtp" for authenticated SMTP, without any
4800 indication (in the protocol name) for TLS use. Now it follows the RFC and
4801 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
4802 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
4803 names appear in log lines as well as in Received: header lines.
4805 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
4806 extensions, and comparison tests, and to fix some bugs.
4808 14. Changes to the "personal" filter test:
4810 (1) The test was buggy in that it was just doing the equivalent of
4811 "contains" tests on header lines. For example, if a user's address was
4812 anne@some.where, the "personal" test would incorrectly be true for
4814 To: susanne@some.where
4816 This test is now done by extracting each address from the header in turn,
4817 and checking the entire address. Other tests that are part of "personal"
4818 are now done using regular expressions (for example, to check local parts
4819 of addresses in From: header lines).
4821 (2) The list of non-personal local parts in From: addresses has been
4822 extended to include "listserv", "majordomo", "*-request", and "owner-*",
4823 taken from the Sieve specification recommendations.
4825 (3) If the message contains any header line starting with "List-" it is
4826 treated as non-personal.
4828 (4) The test for "circular" in the Subject: header line has been removed
4829 because it now seems ill-conceived.
4831 15. Minor typos in src/EDITME comments corrected.
4833 16. Installed latest exipick from John Jetmore.
4835 17. If headers_add on a router specified a text string that was too long for
4836 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
4837 of string_sprintf() is now avoided.
4839 18. $message_body_size was not set (it was always zero) when running the DATA
4840 ACL and the local_scan() function.
4842 19. For the "mail" command in an Exim filter, no default was being set for
4843 the once_repeat time, causing a random time value to be used if "once" was
4844 specified. (If the value happened to be <= 0, no repeat happened.) The
4845 default is now 0s, meaning "never repeat". The "vacation" command was OK
4846 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
4847 (I found it when inspecting the code).
4849 20. There is now an overall timeout for performing a callout verification. It
4850 defaults to 4 times the callout timeout, which applies to individual SMTP
4851 commands during the callout. The overall timeout applies when there is more
4852 than one host that can be tried. The timeout is checked before trying the
4853 next host. This prevents very long delays if there are a large number of
4854 hosts and all are timing out (e.g. when the network connections are timing
4855 out). The value of the overall timeout can be changed by specifying an
4856 additional sub-option for "callout", called "maxwait". For example:
4858 verify = sender/callout=5s,maxwait=20s
4860 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
4861 to the end before writing, but this should make it even safer).
4863 22. Exim was forgetting that it had advertised PIPELINING for the second and
4864 subsequent messages on an SMTP connection. It was also not resetting its
4865 memory on STARTTLS and an internal HELO.
4867 23. When Exim logs an SMTP synchronization error within a session, it now
4868 records whether PIPELINING has been advertised or not.
4870 24. Added 3 instances of "(long int)" casts to time_t variables that were being
4871 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
4872 rather than long int.
4874 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
4876 26. Added the never_mail option to autoreply.
4882 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
4883 crash if the getsockname() call failed; this can happen if a connection is
4884 closed very soon after it is established. The problem was simply in the
4885 order in which certain operations were done, causing Exim to try to write
4886 to the SMTP stream before it had set up the file descriptor. The bug has
4887 been fixed by making things happen in the correct order.
4893 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
4894 before the connection was closed, thus losing the rejection response.
4896 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
4897 some early Solaris releases, but causes trouble in current releases where
4898 socklen_t is defined.
4900 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
4903 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
4906 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
4907 that releasing the top of it at the end releases what was used for sub-
4908 expansions (unless the block got too big). However, discard this block if
4909 the first thing is a variable or header, so that we can use its block when
4910 it is dynamic (useful for very large $message_headers, for example).
4912 6. Lookups now cache *every* query, not just the most recent. A new, separate
4913 store pool is used for this. It can be recovered when all lookup caches are
4914 flushed. Lookups now release memory at the end of their result strings.
4915 This has involved some general refactoring of the lookup sources.
4917 7. Some code has been added to the store_xxx() functions to reduce the amount
4918 of flapping under certain conditions.
4920 8. log_incoming_interface used to affect only the <= reception log lines. Now
4921 it causes the local interface and port to be added to several more SMTP log
4922 lines, for example "SMTP connection from", and rejection lines.
4924 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
4926 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
4928 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
4929 could be overwritten at the end of the current message (or the start of a
4930 new message if it was set in a HELO ACL). The value is now preserved for
4931 the duration of the SMTP connection.
4933 12. If a transport had a headers_rewrite setting, and a matching header line
4934 contained an unqualified address, that address was qualified, even if it
4935 did not match any rewriting rules. The underlying bug was that the values
4936 of the flags that permit the existence of unqualified sender and recipient
4937 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
4938 non-local messages, and by -bnq for local messages) were not being
4939 preserved with the message after it was received.
4941 13. When Exim was logging an SMTP synchronization error, it could sometimes log
4942 "next input=" as part of the text comprising the host identity instead of
4943 the correct text. The code was using the same buffer for two different
4944 strings. However, depending on which order the printing function evaluated
4945 its arguments, the bug did not always show up. Under Linux, for example, my
4946 test suite worked just fine.
4948 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
4949 doesn't work with some older versions of Perl. It has been changed to "my",
4950 which in any case is probably the better facility to use.
4952 15. A really picky compiler found some instances of statements for creating
4953 error messages that either had too many or two few arguments for the format
4956 16. The size of the buffer for calls to the DNS resolver has been increased
4957 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
4958 for addresses that have a lot of PTR records. This alleviates a problem; it
4959 does not fully solve it.
4961 17. A dnsdb lookup for PTR records that receives more data than will fit in the
4962 buffer now truncates the list and logs the incident, which is the same
4963 action as happens when Exim is looking up a host name and its aliases.
4964 Previously in this situation something unpredictable would happen;
4965 sometimes it was "internal error: store_reset failed".
4967 18. If a server dropped the connection unexpectedly when an Exim client was
4968 using GnuTLS and trying to read a response, the client delivery process
4969 crashed while trying to generate an error log message.
4971 19. If a "warn" verb in an ACL added multiple headers to a message in a single
4972 string, for example:
4974 warn message = H1: something\nH2: something
4976 the text was added as a single header line from Exim's point of view
4977 though it ended up OK in the delivered message. However, searching for the
4978 second and subsequent header lines using $h_h2: did not work. This has been
4979 fixed. Similarly, if a system filter added multiple headers in this way,
4980 the routers could not see them.
4982 20. Expanded the error message when iplsearch is called with an invalid key to
4983 suggest using net-iplsearch in a host list.
4985 21. When running tests using -bh, any delays imposed by "delay" modifiers in
4986 ACLs are no longer actually imposed (and a message to that effect is
4989 22. If a "gecos" field in a passwd entry contained escaped characters, in
4990 particular, if it contained a \" sequence, Exim got it wrong when building
4991 a From: or a Sender: header from that name. A second bug also caused
4992 incorrect handling when an unquoted " was present following a character
4993 that needed quoting.
4995 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
4996 was not being matched caselessly.
4998 24. Arranged for all hyphens in the exim.8 source to be escaped with
5001 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
5002 properly. Recipient callout cache records were still being keyed to include
5003 the sender, even when use_sender was set false. This led to far more
5004 callouts that were necessary. The sender is no longer included in the key
5005 when use_sender is false.
5007 26. Added "control = submission" modifier to ACLs.
5009 27. Added the ${base62d: operator to decode base 62 numbers.
5011 28. dnsdb lookups can now access SRV records.
5013 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
5014 the configuration file.
5016 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
5017 (-v) mode. This makes the output for a verbose queue run more intelligible.
5019 31. Added a use_postmaster feature to recipient callouts.
5021 32. Added the $body_zerocount variable, containing the number of binary zero
5022 bytes in the message body.
5024 33. The time of last modification of the "new" subdirectory is now used as the
5025 "mailbox time last read" when there is a quota error for a maildir
5028 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
5030 35. Added +ignore_unknown as a special item in host lists.
5032 36. Code for decoding IPv6 addresses in host lists is now included, even if
5033 IPv6 support is not being compiled. This fixes a bug in which an IPv6
5034 address was recognized as an IP address, but was then not correctly decoded
5035 into binary, causing unexpected and incorrect effects when compared with
5042 1. Very minor rewording of debugging text in manualroute to say "list of
5043 hosts" instead of "hostlist".
5045 2. If verify=header_syntax was set, and a header line with an unqualified
5046 address (no domain) and a large number of spaces between the end of the
5047 name and the colon was received, the reception process suffered a buffer
5048 overflow, and (when I tested it) crashed. This was caused by some obsolete
5049 code that should have been removed. The fix is to remove it!
5051 3. When running in the test harness, delay a bit after writing a bounce
5052 message to get a bit more predictability in the log output.
5054 4. Added a call to search_tidyup() just before forking a reception process. In
5055 theory, someone could use a lookup in the expansion of smtp_accept_max_
5056 per_host which, without the tidyup, could leave open a database connection.
5058 5. Added the variables $recipient_data and $sender_data which get set from a
5059 lookup success in an ACL "recipients" or "senders" condition, or a router
5060 "senders" option, similar to $domain_data and $local_part_data.
5062 6. Moved the writing of debug_print from before to after the "senders" test
5065 7. Change 4.31/66 (moving the time when the Received: is generated) caused
5066 problems for message scanning, either using a data ACL, or using
5067 local_scan() because the Received: header was not generated till after they
5068 were called (in order to set the time as the time of reception completion).
5069 I have revised the way this works. The header is now generated after the
5070 body is received, but before the ACL or local_scan() are called. After they
5071 are run, the timestamp in the header is updated.
5077 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
5078 before starting a queue runner without re-exec. This happened only when
5079 deliver_drop_privilege was set or when the Exim user was set to root. The
5080 effect of the bug was that timeouts during subsequent deliveries caused
5081 crashes instead of being properly handled. The handler is now left at its
5082 default (and expected) setting.
5084 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
5085 message, again when deliver_drop_privilege is set or Exim is run as root.
5086 The bug described in (1) was not present in this case, but the tidying up
5087 of the other signals was missing. I have made the two cases consistent.
5089 3. The ignore_target_hosts setting on a manualroute router was being ignored
5090 for hosts that were looked up using the /MX notation.
5092 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
5095 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
5096 operated on the sender address. After changing the $sender_address to <>
5097 for the sender address verify, Exim was re-instated it as the original
5098 (before rewriting) address, but remembering that it had rewritten it, so it
5099 wasn't rewriting it again. This bug also had the effect of breaking the
5100 sender address verification caching when the sender address was rewritten.
5102 6. The ignore_target_hosts option was being ignored by the ipliteral router.
5103 This has been changed so that if the ip literal address matches
5104 ignore_target_hosts, the router declines.
5106 7. Added expansion conditions match_domain, match_address, and match_local_
5107 part (NOT match_host).
5109 8. The placeholder for the Received: header didn't have a length field set.
5111 9. Added code to Exim itself and to exim_lock to test for a specific race
5112 condition that could lead to file corruption when using MBX delivery. The
5113 issue is with the lockfile that is created in /tmp. If this file is removed
5114 after a process has opened it but before that process has acquired a lock,
5115 there is the potential for a second process to recreate the file and also
5116 acquire a lock. This could lead to two Exim processes writing to the file
5117 at the same time. The added code performs the same test as UW imapd; it
5118 checks after acquiring the lock that its file descriptor still refers to
5119 the same named file.
5121 10. The buffer for building added header lines was of fixed size, 8192 bytes.
5122 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
5125 11. Added the smtp_active_hostname option. If used, this will typically be made
5126 to depend on the incoming interface address. Because $interface_address is
5127 not set up until the daemon has forked a reception process, error responses
5128 that can happen earlier (such as "too many connections") no longer contain
5131 12. If an expansion in a condition on a "warn" statement fails because a lookup
5132 defers, the "warn" statement is abandoned, and the next ACL statement is
5133 processed. Previously this caused the whole ACL to be aborted.
5135 13. Added the iplsearch lookup type.
5137 14. Added ident_timeout as a log selector.
5139 15. Added tls_certificate_verified as a log selector.
5141 16. Added a global option tls_require_ciphers (compare the smtp transport
5142 option of the same name). This controls incoming TLS connections.
5144 17. I finally figured out how to make tls_require_ciphers do a similar thing
5145 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
5146 before starting the TLS session.
5148 18. Tabs are now shown as \t in -bP output.
5150 19. If the log selector return_path_on_delivery was set, Exim crashed when
5151 bouncing a message because it had too many Received: header lines.
5153 20. If two routers both had headers_remove settings, and the first one included
5154 a superfluous trailing colon, the final name in the first list and the
5155 first name in the second list were incorrectly joined into one item (with a
5156 colon in the middle).
5162 1. Added -C and -D options to the exinext utility, mainly to make it easier
5163 to include in the automated testing, but these could be helpful when
5164 multiple configurations are in use.
5166 2. The exinext utility was not formatting the output nicely when there was
5167 an alternate port involved in the retry record key, nor when there was a
5168 message id as well (for retries that were specific to a specific message
5169 and a specific host). It was also confused by IPv6 addresses, because of
5170 the additional colons they contain. I have fixed the IPv4 problem, and
5171 patched it up to do a reasonable job for IPv6.
5173 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
5174 delivery, the log line now contains "pipelined" if PIPELINING was used.
5176 4. An SMTP transport process used to panic and die if the bind() call to set
5177 an explicit outgoing interface failed. This has been changed; it is now
5178 treated in the same way as a connect() failure.
5180 5. A reference to $sender_host_name in the part of a conditional expansion
5181 that was being skipped was still causing a DNS lookup. This no longer
5184 6. The def: expansion condition was not recognizing references to header lines
5185 that used bh_ and bheader_.
5187 7. Added the _cache feature to named lists.
5189 8. The code for checking quota_filecount in the appendfile transport was
5190 allowing one more file than it should have been.
5192 9. For compatibility with Sendmail, the command line option
5200 and sets the incoming protocol and host name (for trusted callers). The
5201 host name and its colon can be omitted when only the protocol is to be set.
5202 Note the Exim already has two private options, -pd and -ps, that refer to
5203 embedded Perl. It is therefore impossible to set a protocol value of "d" or
5204 "s", but I don't think that's a major issue.
5206 10. A number of refactoring changes to the code, none of which should affect
5209 (a) The number of logging options was getting close to filling up the
5210 32-bit word that was used as a bit map. I have split them into two classes:
5211 those that are passed in the argument to log_write(), and those that are
5212 only ever tested independently outside of that function. These are now in
5213 separate 32-bit words, so there is plenty of room for expansion again.
5214 There is no change in the user interface or the logging behaviour.
5216 (b) When building, for example, log lines, the code previously used a
5217 macro that called string_cat() twice, in order to add two strings. This is
5218 not really sufficiently general. Furthermore, there was one instance where
5219 it was actually wrong because one of the argument was used twice, and in
5220 one call a function was used. (As it happened, calling the function twice
5221 did not affect the overall behaviour.) The macro has been replaced by a
5222 function that can join an arbitrary number of extra strings onto a growing
5225 (c) The code for expansion conditions now uses a table and a binary chop
5226 instead of a serial search (which was left over from when there were very
5227 few conditions). Also, it now recognizes conditions like "pam" even when
5228 the relevant support is not compiled in: a suitably worded error message is
5229 given if an attempt is made to use such a condition.
5231 11. Added ${time_interval:xxxxx}.
5233 12. A bug was causing one of the ddress fields not to be passed back correctly
5234 from remote delivery subprocesses. The field in question was not being
5235 subsequently used, so this caused to problems in practice.
5237 13. Added new log selectors queue_time and deliver_time.
5239 14. Might have fixed a bug in maildirsizefile handling that threw up
5240 "unexpected character" debug warnings, and recalculated the data
5241 unnecessarily. In any case, I expanded the warning message to give more
5244 15. Added the message "Restricted characters in address" to the statements in
5245 the default ACL that block characters like @ and % in local parts.
5247 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
5248 Three changes have been made:
5250 (a) There was a serious bug; a negative response to MAIL caused the whole
5251 recipient domain to be cached as invalid, thereby blocking all messages
5252 to all local parts at the same domain, from all senders. This bug has
5253 been fixed. The domain is no longer cached after a negative response to
5254 MAIL if the sender used is not empty.
5256 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
5259 (c) A new callout option, "use_sender" has been added for people who want
5260 the modified behaviour.
5266 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
5269 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
5270 indeed breaks things for older releases.
5272 3. Added additional logging to the case where there is a problem reading data
5273 from a filter that is running in a subprocess using a pipe, in order to
5274 try to track down a specific problem.
5276 4. Testing facility fudge: when running in the test harness and attempting
5277 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
5278 getting "No route to host". Convert this to a timeout.
5280 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
5283 6. Some OS don't have socklen_t but use size_t instead. This affects the
5284 fifth argument of getsockopt() amongst other things. This is now
5285 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
5286 can be set for individual OS. I have set it for SunOS5, OSF1, and
5287 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
5288 some earlier ones do not.
5290 7. Change 4.30/15 was not doing the test caselessly.
5292 8. The standard form for an IPv6 address literal was being rejected by address
5293 parsing in, for example, MAIL and RCPT commands. An example of this kind of
5294 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5295 this, as well as the form without the "IPv6" on the front (but only when
5296 address literals are enabled, of course).
5298 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
5300 10. Exim crashed if a message with an empty sender address specified by -f
5301 encountered a router with an errors_to setting. This could be provoked only
5302 by a command such as
5306 where an empty string was supplied; "<>" did not hit this bug.
5308 11. Installed PCRE release 4.5.
5310 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
5311 remained set. It is now erased.
5313 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
5314 times from message ids (which are base 36 rather than the normal 62).
5316 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
5317 were being counted as actual protocol errors, and logged if the log
5318 selector +smtp_protocol_error was set. One cannot be perfect in this test,
5319 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
5320 and DATA following a set of rejected RCPTs do not count as protocol errors.
5321 In other words, Exim assumes they were pipelined, though this may not
5322 actually be the case. Of course, in all cases the client gets an
5323 appropriate error code.
5325 15. If a lookup fails in an ACL condition, a message about the failure may
5326 be available; it is used if testing the ACL cannot continue, because most
5327 such messages specify what the cause of the deferral is. However, some
5328 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
5329 that caused an old message to be retained and used if a later statement
5330 caused a defer, replacing the real cause of the deferral.
5332 16. If an IP address had so many PTR records that the DNS lookup buffer
5333 was not large enough to hold them, Exim could crash while trying to process
5334 the truncated data. It now detects and logs this case.
5336 17. Further to 4.21/58, another change has been made: if (and only if) the
5337 first line of a message (the first header line) ends with CRLF, a bare LF
5338 in a subsequent header line has a space inserted after it, so as not to
5339 terminate the header.
5341 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
5342 unnecessarily, used atoi() instead of strtol(), and didn't check the
5343 termination when getting file sizes from file names by regex.
5345 19. Completely re-implemented the support for maildirsize files, in the light
5346 of a number of problems with the previous contributed implementation
5347 (4.30/29). In particular:
5349 . If the quota is zero, the maildirsize file is maintained, but no quota is
5352 . If the maildir directory does not exist, it is created before any attempt
5353 to write a maildirsize file.
5355 . The quota value in the file is just a cache; if the quota is changed in
5356 the transport, the new value overrides.
5358 . A regular expression is available for excluding directories from the
5361 20. The autoreply transport checks the characters in options that define the
5362 message's headers; it allows continued headers, but it was checking with
5363 isspace() after an embedded newline instead of explicitly looking for a
5366 21. If all the "regular" hosts to which an address was routed had passed their
5367 expiry times, and had not reached their retry times, the address was
5368 bounced, even if fallback hosts were defined. Now Exim should go on to try
5371 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
5372 equivalent code in the SMTP transport. Some hosts send humungous responses
5373 to HELO/EHLO, more than 1024 it seems.
5375 23. Refactoring: code in filter.c used (void *) for "any old type" but this
5376 gives compiler warnings in some environments. I've now done it "properly",
5379 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
5380 (because of problems with the built-in one) was declared to return uschar *
5381 instead of char *, causing compiler failure.
5383 25. Fixed a file descriptor leak when processing alias/forward files.
5385 26. Fixed a minor format string issue in dbfn.c.
5387 27. Typo in exim.c: ("dmbnz" for "dbmnz").
5389 28. If a filter file refered to $h_xxx or $message_headers, and the headers
5390 contained RFC 2047 "words", Exim's memory could, under certain conditions,
5393 29. When a sender address is verified, it is cached, to save repeating the test
5394 when there is more than one recipient in a message. However, when the
5395 verification involves a callout, it is possible for different callout
5396 options to be set for different recipients. It is too complicated to keep
5397 track of this in the cache, so now Exim always runs a verification when a
5398 callout is required, relying on the callout cache for the optimization.
5399 The overhead is duplication of the address routing, but this should not be
5402 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
5403 to be verified with callout=postmaster, and the main callout worked but the
5404 postmaster check failed, the verification correctly failed. However, if a
5405 subsequent RCPT command asked for sender verification *without* the
5406 postmaster check, incorrect caching caused this verification also to fail,
5409 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
5410 it was not caching the DNS options (qualify_single, search_parents) that
5411 were used when the lookup failed. A subsequent lookup with different
5412 options therefore always gave the same answer, though there were cases
5413 where it should not have. (Example: a "domains = !$mx_any" option on a
5414 dnslookup router: the "domains" option is always processed without any
5415 widening, but the router might have qualify_single set.) Now Exim uses the
5416 cached value only when the same options are set.
5418 32. Added John Jetmore's "exipick" utility to the distribution.
5420 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
5421 than a timeout (e.g. a certificate is required, and is not provided), an
5422 Exim server now closes the connection immediately. Previously it waited for
5423 the client to close - but if the client is SSL, it seems that they each
5424 wait for each other, leading to a delay before one of them times out.
5426 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
5427 maintained 0.8.x compatibility because I don't think many are using it, and
5428 it is clearly obsolete.
5430 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
5433 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
5434 client certificate was expired. A simple patch fixes this, though I don't
5435 understand the full logic of why the verify callback is called multiple
5438 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
5439 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
5440 which causes problems with some clients (such as the Certicom SSL Plus
5441 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
5442 disables the coutermeasure allowing Eudora to connect."
5444 38. Exim was not checking that a write() to a log file succeeded. This could
5445 lead to Bad Things if a log got too big, in particular if it hit a file
5446 size limit. Exim now panics and dies if it cannot write to a log file, just
5447 as it does if it cannot open a log file.
5449 39. Modified OS/Makefile-Linux so that it now contains
5451 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
5453 The two -D definitions ensure that Exim is compiled with large file
5454 support, which makes it possible to handle log files that are bigger than
5457 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
5458 instance) a domain was checked against a named list that involved a lookup,
5459 causing $domain_data to be set, then another domain was checked against the
5460 same list, then the first domain was re-checked, the value of $domain_data
5461 after the final check could be wrong. In particular, if the second check
5462 failed, it could be set empty. This bug probably also applied to
5465 41. The strip_trailing_dot option was not being applied to the address given
5466 with the -f command-line option.
5468 42. The code for reading a message's header from the spool was incrementing
5469 $received_count, but never initializing it. This meant that the value was
5470 incorrect (doubled) while delivering a message in the same process in which
5471 it was received. In the most common configuration of Exim, this never
5472 happens - a fresh exec is done - but it can happen when
5473 deliver_drop_privilege is set.
5475 43. When Exim logs an SMTP synchronization error - client data sent too soon -
5476 it now includes up to 150 characters of the unexpected data in the log
5479 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
5480 and building data strings. The size of both of these buffers was 10 000
5481 bytes - far larger than anybody would *ever* want, thought I. Needless to
5482 say, somebody hit the limit. I have increased the maximum line length to
5483 20 000 and the maximum data length of concatenated lines to 100 000. I have
5484 also fixed two bugs, because there was no checking on these buffers. Tsk,
5485 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
5486 buffer is too small.
5488 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
5489 lsearch lookups. Now it does.
5491 46. When parsing a route_list item in a manualroute router, a fixed-length
5492 buffer was used for the list of hosts. I made this 1024 bytes long,
5493 thinking that nobody would ever have a list of hosts that long. Wrong.
5494 Somebody had a whole pile of complicated expansion conditions, and the
5495 string was silently truncated, leading to an expansion error. It turns out
5496 that it is easier to change to an unlimited length (owing to other changes
5497 that have happened since this code was originally written) than to build
5498 structure for giving a limitation error. The length of the item that
5499 expands into the list of hosts is now unlimited.
5501 47. The lsearch lookup could not handle data where the length of text line was
5502 more than 4095 characters. Such lines were truncated, leading to shortened
5503 data being returned. It should now handle lines of any length.
5505 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
5506 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
5509 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
5510 replacing strings such as BIN_DIRECTORY in a source file: the replacement
5511 no longer happens in comment lines. A list of replacements is now placed
5512 at the head of all of the source files, except those whose only change is
5513 to replace PERL_COMMAND in the very first #! line.
5515 50. Replaced the slow insertion sort in queue.c, for sorting the list of
5516 messages on the queue, with a bottom-up merge sort, using code contributed
5517 by Michael Haardt. This should make operations like -bp somewhat faster on
5518 large queues. It won't affect queue runners, except when queue_run_in_order
5521 51. Installed eximstats 1.31 in the distribution.
5523 52. Added support for SRV lookups to the dnslookup router.
5525 53. If an ACL referred to $message_body or $message_body_end, the value was not
5526 reset for any messages that followed in the same SMTP session.
5528 54. The store-handling optimization for building very long strings was not
5529 differentiating between the different store pools. I don't think this
5530 actually made any difference in practice, but I've tidied it.
5532 55. While running the routers to verify a sender address, $sender_address
5533 was still set to the sender address. This is wrong, because when routing to
5534 send a bounce to the sender, it would be empty. Therefore, I have changed
5535 it so that, while verifying a sender address, $sender_address is set to <>.
5536 (There is no change to what happens when verifying a recipient address.)
5538 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
5539 target A or AAAA records (if not already returned) without resetting the
5540 qualify_single or search_parents options of the DNS resolver. These are
5541 inappropriate in this case because the targets of MX and SRV records must
5542 be FQDNs. A broken DNS record could cause trouble if it happened to have a
5543 target that, when qualified, matched something in the local domain. These
5544 two options are now turned off when doing these lookups.
5546 57. It seems that at least some releases of Reiserfs (which does not have the
5547 concept of a fixed number of inodes) returns zero and not -1 for the
5548 number of available inodes. This interacted badly with check_spool_inodes,
5549 which assumed that -1 was the "no such thing" setting. What I have done is
5550 to check that the total number of inodes is greater than zero before doing
5551 the test of how many are available.
5553 58. When a "warn" ACL statement has a log_message modifier, the message is
5554 remembered, and not repeated. This is to avoid a lot of repetition when a
5555 message has many recipients that cause the same warning to be written.
5556 Howewer, Exim was preserving the list of already written lines for an
5557 entire SMTP session, which doesn't seem right. The memory is now reset if a
5558 new message is started.
5560 59. The "rewrite" debugging flag was not showing the result of rewriting in the
5561 debugging output unless log_rewrite was also set.
5563 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
5564 of (int)(handle) when we know that handle contains (void *)(-1).
5566 61. The Exim daemon panic-logs an error return when it closes the incoming
5567 connection. However "connection reset by peer" seems to be common, and
5568 isn't really an error worthy of noting specially, so that particular error
5571 62. When Exim is trying to find all the local interfaces, it used to panic and
5572 die if the ioctl to get the interface flags failed. However, it seems that
5573 on at least one OS (Solaris 9) it is possible to have an interface that is
5574 included in the list of interfaces, but for which you get a failure error
5575 for this call. This happens when the interface is not "plumbed" into a
5576 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
5577 failure of the "get flags" call assumes that the interface is down.
5579 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
5580 makes life easier for people who are doing arithmetic on fields extracted
5581 from dates, where you often get leading zeros that should not be
5582 interpreted as octal.
5584 64. Added qualify_domain to the redirect router, to override the global
5587 65. If a pathologically long header line contained very many addresses (the
5588 report of this problem mentioned 10 000) and each of them was rewritten,
5589 Exim could use up a very large amount of memory. (It kept on making new
5590 copies of the header line as it rewrote, and never released the old ones.)
5591 At the expense of a bit more processing, the header rewriting function has
5592 been changed so that it no longer eats memory in this way.
5594 66. The generation of the Received: header has been moved from the time that a
5595 message starts to be received, to the time that it finishes. The timestamp
5596 in the Received: header should now be very close to that of the <= log
5597 line. There are two side-effects of this change:
5599 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
5600 logged header lines no longer include the local Received: line, because
5601 it has not yet been created. The same applies to a copy of the message
5602 that is returned to a non-SMTP sender when a message is rejected.
5604 (b) When a filter file is tested using -bf, no additional Received: header
5605 is added to the test message. After some thought, I decided that this
5608 This change does not affect the value of $received_for. It is still set
5609 after address rewriting, but before local_scan() is called.
5611 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
5613 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
5614 gave an unhelpful panic error message, and a defer error. I have managed to
5615 change this behaviour so that it now rejects any supplied certificate,
5616 which seems right, as the list of acceptable certificates is empty.
5618 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
5619 gave an unhelpful defer error. I have not managed to make this reject any
5620 supplied certificates, but the error message it gives is "no certificate
5621 supplied", which is not helpful.
5623 70. exigrep's output now also includes lines that are not associated with any
5624 message, but which match the given pattern. Implemented by a patch from
5625 Martin Sluka, which also tidied up the Perl a bit.
5627 71. Recipient callout verification, like sender verification, was using <> in
5628 the MAIL FROM command. This isn't really the right thing, since the actual
5629 sender may affect whether the remote host accepts the recipient or not. I
5630 have changed it to use the actual sender in the callout; this means that
5631 the cache record is now keyed on a recipient/sender pair, not just the
5632 recipient address. There doesn't seem to be a real danger of callout loops,
5633 since a callout by the remote host to check the sender would use <>.
5634 [SEE ABOVE: changed after hitting problems.]
5636 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
5637 temporary errors. However, in the case of such a code being given after
5638 the end of a data transmission (i.e. after ".") Exim was failing to write
5639 a retry record for the message. (Yes, there was some broken host that was
5640 actually sending 8xx at this point.)
5642 73. An unknown lookup type in a host list could cause Exim to panic-die when
5643 the list was checked. (An example that provoked this was putting <; in the
5644 middle of a list instead of at the start.) If this happened during a DATA
5645 ACL check, a -D file could be left lying around. This kind of configuration
5646 error no longer causes Exim to die; instead it causes a defer errror. The
5647 incident is still logged to the main and panic logs.
5649 74. Buglet left over from Exim 3 conversion. The message "too many messages
5650 in one connection" was written to the rejectlog but not the mainlog, except
5651 when address rewriting (yes!) was being logged.
5653 75. Added write_rejectlog option.
5655 76. When a system filter was run not as root (that is, when system_filter_user
5656 was set), the values of the $n variables were not being returned to the
5657 main process; thus, they were not subsequently available in the $sn
5660 77. Added +return_path_on_delivery log selector.
5662 78. A connection timeout was being treated differently from recipients deferred
5663 when testing hosts_max_try with a message that was older than the host's
5664 retry timeout. (The host should not be counted, thus allowing all hosts to
5665 be tried at least once before bouncing.) This may have been the cause of an
5666 occasionally reported bug whereby a message would remain on the queue
5667 longer than the retry timeout, but would be bounced if a delivery was
5668 forced. I say "may" because I never totally pinned down the problem;
5669 setting up timeout/retry tests is difficult. See also the next item.
5671 79. The ultimate address timeout was not being applied to errors that involved
5672 a combination of host plus message (for example, a timeout on a MAIL
5673 command). When an address resolved to a number of possible hosts, and they
5674 were not all tried for each delivery (e.g. because of hosts_max_try), a
5675 message could remain on the queue longer than the retry timeout.
5677 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
5680 81. Fixed an obscure SMTP outgoing bug which required at least the following
5681 conditions: (a) there was another message waiting for the same server;
5682 (b) the server returned 5xx to all RCPT commands in the first message so
5683 that the message was not completed; (c) the server dropped the connection
5684 or gave a negative response to the RSET that Exim sends to abort the
5685 transaction. The observed case was a dropped connection after DATA that had
5686 been sent in pipelining mode. That is, the server had advertised PIPELINING
5687 but was not implementing it correctly. The effect of the bug was incorrect
5688 behaviour, such as trying another host, and this could lead to a crash.
5694 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
5695 and daemon.c were passed as pointers to ints; they should have been
5696 pointers to socklen_t variables (which are typically unsigned ints).
5698 2. Some signed/unsigned type warnings in the os.c file for Linux have been
5701 3. Fixed a really odd bug that affected only the testing scheme; patching a
5702 certain fixed string in the binary changed the value of another string that
5703 happened to be identical to the end of the original first string.
5705 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
5706 name", it returns that address as the IP address. On some operating
5707 systems (e.g. Solaris), it also passes back the IP address string as the
5708 "host name". However, on others (e.g. Linux), it passes back an empty
5709 string. Exim wasn't checking for this, and was changing the host name to an
5710 empty string, assuming it had been canonicized.
5712 5. Although rare, it is permitted to have more than one PTR record for a given
5713 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
5714 all the names associated with an address, because they do in Solaris.
5715 However, it seems that they do not in Linux for data that comes from the
5716 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
5717 I found this out when I moved to a new Linux workstation and tried to run
5718 the Exim test suite.
5720 To get round this problem I have changed the code so that it now does its
5721 own call to the DNS to look up PTR records when searching for a host name.
5722 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
5723 addresses that are only in /etc/hosts are still found.
5725 This behaviour is, however, controlled by an option called host_lookup_
5726 order, which defaults to "bydns:byaddr". If people want to use the other
5727 order, or indeed, just use one or the other means of lookup, they can
5728 specify it in this variable.
5730 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
5731 some operating systems, this comes back from gethostbyaddr() as an empty
5732 string, and this is what Exim used to test for. However, it seems that in
5733 other systems, "." is yielded. Exim now tests for this case too.
5735 7. The values of check_spool_space and check_log_space are now held internally
5736 as a number of kilobytes instead of an absolute number of bytes. If a
5737 numbers is specified without 'K' or 'M', it is rounded up to the nearest
5738 kilobyte. This means that much larger values can be stored.
5740 8. Exim monitor: an attempt to get the action menu when not actually pointing
5741 at a message produces an empty menu entitled "No message selected". This
5742 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
5743 no entries in it ("Shell widget menu has zero width and/or height"). So I
5744 have added a single, blank menu entry in this case.
5746 9. Added ${quote_local_part.
5748 10. MIME decoding is now applied to the contents of Subject: header lines when
5751 11. Now that a reference to $sender_host_address automatically causes a reverse
5752 lookup to occur if necessary (4.13/18), there is no need to arrange for a
5753 host lookup before query-style lookups in lists that might use this
5754 variable. This has therefore been abolished, and the "net-" prefix is no
5755 longer necessary for query-style lookups.
5757 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
5758 have been a typo for LFLAGS, so it has been changed.
5760 13. The install script calls Exim with "-C /dev/null" in order to find the
5761 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
5762 to be output. Howeve, since Exim outputs its version number before the
5763 error, it didn't break the script. It just looked ugly. I fixed this by
5764 always allowing "-C /dev/null" if the caller is root.
5766 14. Ignore overlarge ACL variable number when reading spool file - insurance
5767 against a later release with more variables having written the file.
5769 15. The standard form for an IPv6 address literal was being rejected by EHLO.
5770 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5771 this, as well as the form without the "IPv6" on the front.
5773 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
5774 OS/Makefile-Darwin file.
5776 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
5777 only to LDAP libraries that do not have LDAP_OPT_DEREF.
5779 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
5780 variable. However, some OS use ints for inodes. I've added cast to long int
5781 to get rid of the compiler warning.
5783 19. I had forgotten to lock out "/../" in configuration file names when
5784 ALT_CONFIG_PREFIX was set.
5786 20. Routers used for verification do not need to specify transports. However,
5787 if such a router generated a host list, and callout was configured, Exim
5788 crashed, because it could not find a port number from the (non-existent)
5789 transport. It now assumes port 25 in this circumstance.
5791 21. Added the -t option to exigrep.
5793 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
5794 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
5795 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
5796 them is set, LOOKUP_LSEARCH is forced.
5798 23. "exim -bV" now outputs a list of lookups that are included in the binary.
5800 24. Added sender and host information to the "rejected by local_scan()" log
5801 line; previously there was no indication of these.
5803 25. Added .include_if_exists.
5805 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
5806 Linux. It turns out that not all file systems support this. Apparently some
5807 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
5808 do it.) To cope with this, the error EINVAL, which means that sync-ing is
5809 not supported on the file descriptor, is now ignored when Exim is trying to
5810 sync a directory. This applies only to Linux.
5812 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
5814 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
5815 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
5818 29. Added support for maildirsize files from supplied patch (modified a bit).
5820 30. The use of :fail: followed by an empty string could lead Exim to respond to
5821 sender verification failures with (e.g.):
5823 550 Verification failed for <xxx>
5824 550 Sender verify failed
5826 where the first response line was missing the '-' that indicates it is not
5827 the final line of the response.
5829 31. The loop for finding the name of the user that called Exim had a hardwired
5830 limit of 10; it now uses the value of finduser_retries, which is used for
5831 all other user lookups.
5833 32. Added $received_count variable, available in data and not_smtp ACLs, and at
5836 33. Exim was neglecting to zero errno before one call of strtol() when
5837 expanding a string and expecting an integer value. On some systems this
5838 resulted in spurious "integer overflow" errors. Also, it was casting the
5839 result into an int without checking.
5841 34. Testing for a connection timeout using "timeout_connect" in the retry rules
5842 did not work. The code looks as if it has *never* worked, though it appears
5843 to have been documented since at least releast 1.62. I have made it work.
5845 35. The "timeout_DNS" error in retry rules, also documented since at least
5846 1.62, also never worked. As it isn't clear exactly what this means, and
5847 clearly it isn't a major issue, I have abolished the feature by treating it
5848 as "timeout", and writing a warning to the main and panic logs.
5850 36. The display of retry rules for -brt wasn't always showing the error code
5853 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
5854 timeout_connect_A, timeout_connect_MX.
5856 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
5857 to the empty sender.
5859 39. The daemon was not analysing the content of -oX till after it had closed
5860 stderr and disconnected from the controlling terminal. This meant that any
5861 syntax errors were only noted on the panic log, and the return code from
5862 the command was 0. By re-arranging the code a little, I've made the
5863 decoding happen first, so such errors now appear on stderr, and the return
5864 code is 1. However, the actual setting up of the sockets still happens in
5865 the disconnected process, so errors there are still only recorded on the
5868 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
5869 connections (as happens on some IP stacks) was logged at start up time as
5870 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
5871 it from "IPv6 and IPv4", which means that two separate sockets are being
5874 41. The debug output for gethostbyname2() or getipnodebyname() failures now
5875 says whether AF_INET or AF_INET6 was passed as an argument.
5877 42. Exiwhat output was messed up when time zones were included in log
5880 43. Exiwhat now gives more information about the daemon's listening ports,
5881 and whether -tls-on-connect was used.
5883 44. The "port" option of the smtp transport is now expanded.
5885 45. A "message" modifier in a "warn" statement in a non-message ACL was being
5886 silently ignored. Now an error message is written to the main and panic
5889 46. There's a new ACL modifier called "logwrite" which writes to a log file
5890 as soon as it is encountered.
5892 47. Added $local_user_uid and $local_user_gid at routing time.
5894 48. Exim crashed when trying to verify a sender address that was being
5897 49. Exim was recognizing only a space character after ".include". It now also
5898 recognizes a tab character.
5900 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
5901 extracting the relevant information from the specification. The man page no
5902 longer contains scrambled data for the -d option, and I've added a section
5903 at the front about calling Exim under different names.
5905 51. Added "extra_headers" argument to the "mail" command in filter files.
5907 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
5910 53. Installed eximstats 1.29.
5912 54. Added transport_filter_timeout as a generic transport option.
5914 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
5915 Cc: header lines. This was required by RFC 822, but it not required by RFC
5918 56. Exim used to add From:, Date:, and Message-Id: header lines to any
5919 incoming messages that did not have them. Now it does so only if the
5920 message originates locally, that is, if there is no associated remote host
5921 address. When Resent- header lines are present, this applies to the Resent-
5922 lines rather than the non-Resent- lines.
5924 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
5925 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
5927 58. Messages for configuration errors now include the name of the main
5928 configuration file - useful now that there may be more than one file in a
5929 list (.included file names were always shown).
5931 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
5932 for those rare installations that do not start the daemon as root or run it
5933 setuid root. I've cut out the call to initgroups() if the daemon is not
5936 60. The Exim user and group can now be bound into the binary as text strings
5937 that are looked up at the start of Exim's processing.
5939 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
5941 62. Added $mailstore_basename variable.
5943 63. Installed patch to sieve.c from Michael Haardt.
5945 64. When Exim failed to open the panic log after failing to open the main log,
5946 the original message it was trying to log was written to stderr and debug
5947 output, but if they were not available (the usual case in production), it
5948 was lost. Now it is written to syslog before the two lines that record the
5949 failures to open the logs.
5951 65. Users' Exim filters run in subprocesses under the user's uid. It is
5952 possible for a "deliver" command or an alias in a "personal" command to
5953 provoke an address rewrite. If logging of address rewriting is configured,
5954 this fails because the process is not running as root or exim. There may be
5955 a better way of dealing with this, but for the moment (because 4.30 needs
5956 to be released), I have disabled address rewrite logging when running a
5957 filter in a non-root, non-exim process.
5963 1. The buildconfig auxiliary program wasn't quoting the value set for
5964 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
5965 not defined. This bug was masked in 4.22 by the effect that was fixed in
5968 2. Some messages that were rejected after a message id was allocated were
5969 shown as "incomplete" by exigrep. It no longer does this for messages that
5970 are rejected by local_scan() or the DATA or non-SMTP ACLs.
5972 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
5973 have allow_domain_literals set, the ID did not get logged in the <= line.
5974 Domain literals are now always recognized in Message-ID: header lines.
5976 4. The first argument for a ${extract expansion item is the key name or field
5977 number. Leading and trailing spaces in this item were not being ignored,
5978 causing some misleading effects.
5980 5. When deliver_drop_privilege was set, single queue runner processes started
5981 manually (i.e. by the command "exim -q") or by the daemon (which uses the
5982 same command in the process it spins off) were not dropping privilege.
5984 6. When the daemon running as "exim" started a queue runner, it always
5985 re-executed Exim in the spun-off process. This is a waste of effort when
5986 deliver_drop_privilege is set. The new process now just calls the
5987 queue-runner function directly.
5993 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
5996 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
5997 ignored. Though the use of -oP was forcing the writing of a pid file, it
5998 was always written to the default place.
6000 3. If the message "no IP address found for host xxxx" is generated during
6001 incoming verification, it is now followed by identification of the incoming
6002 connection (so you can more easily find what provoked it).
6004 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
6006 5. Added some features to "harden" Exim a bit more against certain attacks:
6008 (a) There is now a build-time option called FIXED_NEVER_USERS that can
6009 be put in Local/Makefile. This is like the never_users runtime option,
6010 but it cannot be overridden. The default setting is "root".
6012 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
6013 prefix string with which any file named in a -C command line option
6016 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
6017 is retained for -C and -D only if the caller of Exim is root. Without
6018 it, the exim user may also use -C and -D and retain privilege.
6020 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
6021 command line option is disabled.
6023 6. Macro names set by the -D option must start with an upper case letter, just
6024 like macro names defined in the configuration file.
6026 7. Added "dereference=" facility to LDAP.
6028 8. Two instances of the typo "uknown" in the source files are fixed.
6030 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
6031 the Configure-Makefile script screwed up while processing it.
6033 10. Incorporated PCRE 4.4.
6035 11. The SMTP synchronization check was not operating right at the start of an
6036 SMTP session. For example, it could not catch a HELO sent before the client
6037 waited for the greeting. There is now a check for outstanding input at the
6038 point when the greeting is written. Because of the duplex, asynchronous
6039 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
6040 way, but not yet received, when the check is performed.
6042 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
6043 on TCP/IP sockets, because this apparently causes some broken clients to
6046 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
6047 unchanged) from the Cygwin maintainer.
6049 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
6050 was supported instead of testing for mbx. Effectively a typo.
6052 15. The spa authenticator server code was not checking that the input it
6053 received was valid base64.
6055 16. The debug output line for the "set" modifier in ACLs was not showing the
6056 name of the variable that was being set.
6058 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
6060 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
6061 lookup on its own. Something else was needed to trigger the lookup. For
6062 example, a match in host_lookup or the need for a host name in a host list.
6063 Now, if $sender_host_name is referenced and the host name has not yet been
6064 looked up, a lookup is performed. If the lookup fails, the variable remains
6065 empty, and $host_lookup_failed is set to "1".
6067 19. Added "eqi" as a case-independent comparison operator.
6069 20. The saslauthd authentication condition could segfault if neither service
6070 nor realm was specified.
6072 21. If an overflowing value such as "2048M" was set for message_size_limit, the
6073 error message that was logged was misleading, and incoming SMTP
6074 connections were dropped. The message is now more accurate, and temporary
6075 errors are given to SMTP connections.
6077 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
6078 (except RSET) with a 421 error, until QUIT is received. However, it was
6079 failing to send a response to QUIT.
6081 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
6082 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
6083 ACL is now run after the helo_try_verify_hosts code.
6085 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
6086 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
6087 case-independent, so other case variants are also recognized). Apparently
6088 some systems use these upper case variants.
6090 25. If more than two messages were waiting for the same host, and a transport
6091 filter was specified for the transport, Exim sent two messages over the
6092 same TCP/IP connection, and then failed with "socket operation on non-
6093 socket" when it tried to send the third.
6095 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
6097 27. The extern definition of crypt16() in expand.c was not being excluded when
6098 the OS had its own crypt16() function.
6100 28. Added bounce_return_body as a new option, and bounce_return_size_limit
6101 as a preferred synonym for return_size_limit, both as an option and as an
6104 29. Added LIBS=-liconv to OS/Makefile-OSF1.
6106 30. Changed the default configuration ACL to relax the local part checking rule
6107 for addresses that are not in any local domains. For these addresses,
6108 slashes and pipe symbols are allowed within local parts, but the sequence
6109 /../ is explicitly forbidden.
6111 31. SPA server authentication was not clearing the challenge buffer before
6114 32. log_message in a "warn" ACL statement was writing to the reject log as
6115 well as to the main log, which contradicts the documentation and doesn't
6116 seem right (because no rejection is happening). So I have stopped it.
6118 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
6119 However, I am unable to do any testing of this.
6121 34. Fixed an infelicity in the appendfile transport. When checking directories
6122 for a mailbox, to see if any needed to be created, it was accidentally
6123 using path names with one or more superfluous leading slashes; tracing
6124 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
6126 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
6127 contents are added to the log line that is written for every discarded
6128 recipient. (Previously a log_message setting was ignored.)
6130 36. The ${quote: operator now quotes the string if it is empty.
6132 37. The install script runs exim in order to find its version number. If for
6133 some reason other than non-existence or emptiness, which it checks, it
6134 could not run './exim', it was installing it with an empty version number,
6135 i.e. as "exim-". This error state is now caught, and the installation is
6138 38. An argument was missing from the function that creates an error message
6139 when Exim fails to connect to the socket for saslauthd authentication.
6140 This could cause Exim to crash, or give a corrupted message.
6142 39. Added isip, isip4, and isip6 to ${if conditions.
6144 40. The ACL variables $acl_xx are now saved with the message, and can be
6145 accessed later in routers, transports, and filters.
6147 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
6148 strings in the file are not string-expanded.
6150 42. If a MAIL command specified a SIZE value that was too large to fit into an
6151 int variable, the check against message_size_limit failed. Such values are
6152 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
6153 day this will have to be increased, but I don't think I want to be around
6154 when emails are that large.
6161 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
6162 iconv() is not standard in FreeBSD.
6164 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
6165 IPv6 enabled. The observed symptom was a segmentation fault on return from
6166 the function os_common_find_running_interfaces() in src/os.c.
6168 3. In the check_special_case() function in daemon.c I had used "errno" as an
6169 argument name, which causes warnings on some systems. This was basically a
6170 typo, since it was named "eno" in the comments!
6172 4. The code that waits for the clock to tick (at a resolution of some fraction
6173 of a second) so as to ensure message-id uniqueness was always waiting for
6174 at least one whole tick, when it could have waited for less. [This is
6175 almost certainly not relevant at current processor speeds, where it is
6176 unlikely to ever wait at all. But we try to future-proof.]
6178 5. The function that sleeps for a time interval that includes fractions of a
6179 second contained a race. It did not block SIGALRM between setting the
6180 timer, and suspending (a couple of lines later). If the interval was short
6181 and the sigsuspend() was delayed until after it had expired, the suspension
6182 never ended. On busy systems this could lead to processes getting stuck for
6185 6. Some uncommon configurations may cause a lookup to happen in a queue runner
6186 process, before it forks any delivery processes. The open lookup caching
6187 mechanism meant that the open file or database connection was passed into
6188 the delivery process. The problem was that delivery processes always tidy
6189 up cached lookup data. This could cause a problem for the next delivery
6190 process started by the queue runner, because the external queue runner
6191 process does not know about the closure. So the next delivery process
6192 still has data in the lookup cache. In the case of a file lookup, there was
6193 no problem because closing a file descriptor in a subprocess doesn't affect
6194 the parent. However, if the lookup was caching a connection to a database,
6195 the connection was closed, and the second delivery process was likely to
6196 see errors such as "PGSQL: query failed: server closed the connection
6197 unexpectedly". The problem has been fixed by closing all cached lookups
6198 in a queue runner before running a delivery process.
6200 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
6201 seem to have the "const" qualifier which it has on other OS. I've
6204 8. Change 4.21/2 was too strict. It is only if there are two authenticators
6205 *of the same type* (client or server) with the same public name that an
6206 error should be diagnosed.
6208 9. When Exim looked up a host name for an IP address, but failed to find the
6209 original IP address when looking up the host name (a safety check), it
6210 output the message "<ip address> does not match any IP for NULL", which was
6211 confusing, to say the least. The bug was that the host name should have
6212 appeared instead of "NULL".
6214 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
6215 user that is built into the binary, and the -C or -D options is used, root
6216 privilege is dropped before the configuration file is read. In addition,
6217 logging is switched to stderr instead of the normal log files. If the
6218 configuration then re-defines the Exim user, the unprivileged environment
6219 is probably not what is expected, so Exim logs a panic warning message (but
6222 However, if deliver_drop_privilege is set, the unprivileged state may well
6223 be exactly what is intended, so the warning has been cut out in that case,
6224 and Exim is allowed to try to write to its normal log files.
6230 1. smtp_return_error_details was not giving details for temporary sender
6231 or receiver verification errors.
6233 2. Diagnose a configuration error if two authenticators have the same public
6236 3. Exim used not to create the message log file for a message until the first
6237 delivery attempt. This could be confusing when incoming messages were held
6238 for policy or load reasons. The message log file is now created at the time
6239 the message is received, and an initial "Received" line is written to it.
6241 4. The automatically generated man page for command line options had a minor
6242 bug that caused no ill effects; however, a more serious problem was that
6243 the procedure for building the man page automatically didn't always
6244 operate. Consequently, release 4.20 contains an out-of-date version. This
6245 shouldn't happen again.
6247 5. When building Exim with embedded Perl support, the script that builds the
6248 Makefile was calling 'perl' to find its compile-time parameters, ignoring
6249 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
6251 6. The freeze_tell option was not being used for messages that were frozen on
6252 arrival, either by an ACL or by local_scan().
6254 7. Added the smtp_incomplete_transaction log selector.
6256 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
6257 was accepting AUTH without a new EHLO.
6259 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
6260 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
6263 10. Logging of TCP/IP connections (when configured) now happens in the main
6264 daemon process instead of the child process, so that the TCP/IP connection
6265 count is more accurate (but it can never be perfect).
6267 11. The use of "drop" in a nested ACL was not being handled correctly in the
6268 outer ACL. Now, if condition failure induced by the nested "drop" causes
6269 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
6270 or "require"), the connection is dropped.
6272 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
6273 that yield "discard" can now be used with an "accept" or a "discard" verb,
6274 but an error is generated for any others (because I can't see a useful way
6275 to define what should happen).
6277 13. When an ACL is read dynamically from a file (or anywhere else), the lines
6278 are now processed in the same way as lines in the Exim configuration file.
6279 In particular, continuation lines are supported.
6281 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
6283 15. Added -ti meaning -t -i.
6285 16. Check for letters, digits, hyphens, and dots in the names of dnslist
6286 domains, and warn by logging if others are found.
6288 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
6289 returned from GIFCONF when Exim is trying to find the list of interfaces on
6290 a host. The code in os.c has been modified to copy each ifreq to an aligned
6291 structure in all cases.
6293 Also, in some cases, the returned ifreq's were being copied to a 'struct
6294 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
6295 means the last couple of bytes of an IPv6 address could be chopped if the
6296 ifreq contained only a normal sockaddr (14 bytes storage).
6298 18. Named domain lists were not supported in the hosts_treat_as_local option.
6299 An entry such as +xxxx was not recognized, and was treated as a literal
6302 19. Ensure that header lines added by a DATA ACL are included in the reject log
6303 if the ACL subsequently rejects the message.
6305 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
6306 MD5 (which is deprecated).
6308 21. When testing a filter file using -bf, Exim was writing a message when it
6309 took the sender from a "From " line in the message, but it was not doing so
6310 when it took $return_path from a Return-Path: header line. It now does.
6312 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
6313 with a valid header line field name (a series of printing characters
6314 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
6316 23. Changed "disc" in the source to "disk" to conform to the documentation and
6317 the book and for uniformity.
6319 24. Ignore Sendmail's -Ooption=value command line item.
6321 25. When execve() failed while trying to run a command in a pipe transport,
6322 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
6323 could be confused with a return value of 69 from the command itself. This
6324 has been changed to 127, the value the shell returns if it is asked to run
6325 a non-existent command. The wording for the related log line suggests a
6326 non-existent command as the problem.
6328 26. If received_header_text expands to an empty string, do not add a Received:
6329 header line to the message. (Well, it adds a token one on the spool, but
6330 marks it "old" so that it doesn't get used or transmitted.)
6332 27. Installed eximstats 1.28 (addition of -nt option).
6334 28. There was no check for failure on the call to getsockname() in the daemon
6335 code. This can fail if there is a shortage of resources on the system, with
6336 ENOMEM, for example. A temporary error is now given on failure.
6338 29. Contrary to the C standard, it seems that in some environments, the
6339 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
6340 program. Exim now does this explicitly; it affects the formatting of
6341 timestamps using strftime().
6343 30. If exiqsumm was given junk data, it threw up some uninitialized variable
6344 complaints. I've now initialized all the variables, to avoid this.
6346 32. Header lines added by a system filter were not being "seen" during
6347 transport-time rewrites.
6349 33. The info_callback() function passed to OpenSSL is set up with type void
6350 (*)(SSL *, int, int), as described somewhere. However, when calling the
6351 function (actually a macro) that sets it up, the type void(*)() is
6352 expected. I've put in a cast to prevent warnings from picky compilers.
6354 34. If a DNS black list lookup found a CNAME record, but there were no A
6355 records associated with the domain it pointed at, Exim crashed.
6357 35. If a DNS black list lookup returned more than one A record, Exim ignored
6358 all but the first. It now scans all returned addresses if a particular IP
6359 value is being sought. In this situation, the contents of the
6360 $dnslist_value variable are a list of all the addresses, separated by a
6363 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
6364 to accept a host name and all its aliases if the forward lookup for any of
6365 them yielded the IP address of the incoming connection. Now it accepts only
6366 those names whose forward lookup yields the correct IP address. Any other
6367 names are discarded. This closes a loophole whereby a rogue DNS
6368 administrator could create reverse DNS records to break through a
6369 wildcarded host restriction in an ACL.
6371 37. If a user filter or a system filter that ran in a subprocess used any of
6372 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
6373 the wrong values were passed to the pipe command ($thisaddress had the
6374 value of $0, $0 had the value of $1, etc). This bug was introduced by
6375 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
6377 38. Improved the line breaking for long SMTP error messages from ACLs.
6378 Previously, if there was no break point between 40 and 75 characters, Exim
6379 left the rest of the message alone. Two changes have been made: (a) I've
6380 reduced the minimum length to 35 characters; (b) if it can't find a break
6381 point between 35 and 75 characters, it looks ahead and uses the first one
6382 that it finds. This may give the occasional overlong line, but at least the
6383 remaining text gets split now.
6385 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
6386 file descriptors might be low, and that setting 1000 would always raise it.
6387 It turns out that in some environments, the limit is already over 1000 and
6388 that lowering it causes trouble. So now Exim takes care not to decrease it.
6390 40. When delivering a message, the value of $return_path is set to $sender_
6391 address at the start of routing (routers may change the value). By an
6392 oversight, this default was not being set up when an address was tested by
6393 -bt or -bv, which affected the outcome if any router or filter referred to
6396 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
6397 log only when "message" or "log_message" are set. However, if one of the
6398 conditions was an address verification, or a call to a nested ACL, the
6399 messages generated by the underlying test were being passed through. This
6400 no longer happens. The underlying message is available in $acl_verify_
6401 message for both "message" and "log_message" expansions, so it can be
6402 passed through if needed.
6404 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
6405 new expansion $bh_ to give the encoded byte string without charset
6406 translation. Translation happens only if iconv() is available; HAVE_ICONV
6407 indicates this at build time. HEADERS_CHARSET gives the charset to
6408 translate to; headers_charset can change it in the configuration, and
6409 "headers charset" can change it in an individual filter file.
6411 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
6412 that creates RFC 2047 encoded "words" labels them as that charset instead
6413 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
6414 expansion operator; (ii) when Exim creates a From: line for a local
6415 message; (iii) when a header line is rewritten to include a "phrase" part.
6417 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
6418 buggy, causing it to skip the first lines of messages whose message ID
6419 ended in 'D'. This would not have bitten before Exim release 4.14, because
6420 message IDs were unlikely to end in 'D' before then. The effect was to have
6421 incorrect size information for certain domains.
6423 45. #include "config.h" was missing at the start of the crypt16.c module. This
6424 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
6427 46. If there was a timeout during a "random" callout check, Exim treated it as
6428 a failure of the random address, and carried on sending RSET and the real
6429 address. If the delay was just some slowness somewhere, the response to the
6430 original RCPT would be taken as a response to RSET and so on, causing
6431 mayhem of various kinds.
6433 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
6434 when I implemented it. It didn't allow for the fact that some option values
6435 may legitimatetly be negative (e.g. size_addition), and it didn't even do
6436 the right test for positive values.
6438 48. Domain names in DNS records are case-independent. Exim always looks them up
6439 in lower case. Some resolvers return domain names in exactly the case they
6440 appear in the zone file, that is, they may contain uppercase letters. Not
6441 all resolvers do this - some return always lower case. Exim was treating a
6442 change of case by a resolver as a change of domain, similar to a widening
6443 of a domain abbreviation. This triggered its re-routing code and so it was
6444 trying to route what was effectively the same domain again. This normally
6445 caused routing to fail (because the router wouldn't handle the domain
6446 twice). Now Exim checks for this case specially, and just changes the
6447 casing of the domain that it ultimately uses when it transmits the message
6450 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
6453 50. If a filter generated a file delivery with a non-absolute name (possible if
6454 no home directory exists for the router), the forbid_file option was not
6457 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
6458 the existing equality matching.
6460 52. Exim was using ints instead of ino_t variables in some places where it was
6461 dealing with inode numbers.
6463 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
6464 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
6465 called TMPDIR, and if it finds it is different, it changes its value.
6467 54. The smtp_printf() function is now made available to local_scan() so
6468 additional output lines can be written before returning. There is also an
6469 smtp_fflush() function to enable the detection of a dropped connection.
6470 The variables smtp_input and smtp_batched_input are exported to
6473 55. Changed the default runtime configuration: the message "Unknown user"
6474 has been removed from the ACL, and instead placed on the localuser router,
6475 using the cannot_route_message feature. This means that any verification
6476 failures that generate their own messages won't get overridden. Similarly,
6477 the "Unrouteable address" message that was in the ACL for unverifiable
6478 relay addresses has also been removed.
6480 56. Added hosts_avoid_esmtp to the smtp transport.
6482 57. The exicyclog script was not checking for the esoteric option
6483 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
6484 will work only if exicyclog is run under the appropriate euid.
6486 58. Following a discussion on the list, the rules by which Exim recognises line
6487 endings on incoming messages have been changed. The -dropcr and drop_cr
6488 options are now no-ops, retained only for backwards compatibility. The
6489 following line terminators are recognized: LF CRLF CR. However, special
6490 processing applies to CR:
6492 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
6493 nor a local message in the state where . is a terminator.
6495 (ii) If a bare CR is encountered in a header line, an extra space is added
6496 after the line terminator so as not to end the header. The reasoning
6497 behind this is that bare CRs in header lines are most likely either
6498 to be mistakes, or people trying to play silly games.
6500 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
6501 was being incorrectly given as 18 bytes larger than it should have been.
6502 This is a VOB (very old bug).
6504 60. This may never have affected anything current, but just in case it has:
6505 When the local host is found other than at the start of a list of hosts,
6506 the local host, those with the same MX, and any that follow, are discarded.
6507 When the list in question was part of a longer list of hosts, the following
6508 hosts (not currently being processed) were also being discarded. This no
6509 longer happens. I'm not sure if this situation could ever has previously
6512 61. Added the "/MX" feature to lists of hosts in the manualroute and query
6515 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
6516 header. This is something that is recommended in a new Internet Draft, and
6517 is something that is documented as being done by Sendmail. There are two
6518 possible values. For messages generated by the autoreply transport, Exim
6521 Auto-Submitted: auto-replied
6523 whereas for all other generated messages (e.g. bounces) it adds
6525 Auto-Submitted: auto-generated
6527 63. The "personal" condition in filters now includes a test for the
6528 Auto-Submitted: header. If it contains the string "auto-" the message it
6529 not considered personal.
6531 64. Added rcpt_include_affixes as a generic transport option.
6533 65. Added queue_only_override (default true).
6535 66. Added the syslog_duplication option.
6537 67. If what should have been the first header line of a message consisted of
6538 a space followed by a colon, Exim was mis-interpreting it as a header line.
6539 It isn't of course - it is syntactically invalid and should therefore be
6540 treated as the start of the message body. The misbehaviour could have
6541 caused a number of strange effects, including loss of data in subsequent
6542 header lines, and spool format errors.
6544 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
6545 client host had authenticated. This control can now be exercised by an ACL
6546 for more flexibility.
6548 69. By default, callouts do not happen when testing with -bh. There is now a
6549 variant, -bhc, which does actually run the callout code, including
6550 consulting and updating the callout cache.
6552 70. Added support for saslauthd authentication, courtesy of Alexander
6555 71. If statvfs() failed on the spool or log directories while checking their
6556 size for availability, Exim confusingly gave the error "space shortage".
6557 Furthermore, in debugging mode it crashed with a floating point exception.
6558 These checks are done if check_{spool,log}_{space,inodes} are set, and when
6559 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
6560 serious problem, Exim now writes to the main and panic logs when this
6561 happens, with details of the failure. It then refuses to accept the
6562 incoming message, giving the message "spool directory problem" or "log
6563 directory problem" with a 421 code for SMTP messages.
6565 72. When Exim is about to re-exec itself, it ensures that the file descriptors
6566 0, 1, and 2 exist, because some OS complain for execs without them (see
6567 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
6568 descriptors. However, the code omitted to check that the open succeeded,
6569 causing mysterious errors if for some reason the permissions on /dev/null
6570 got screwed. Now Exim writes a message to the main and panic logs, and
6571 bombs out if it can't open /dev/null.
6573 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
6574 interact so that it is all more flexible. It is supposed to remain
6575 backwards compatible. Also added extra_local_interfaces.
6577 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
6578 to bomb out with an assertion failure - to the client this appears as a
6579 connection drop. This problem occurs in the part of the code that was taken
6580 from the Samba project. Fortunately, the assertion is in a very simple
6581 function, so I have fixed this by reproducing the function inline in the
6582 one place where it is called, and arranging for authentication to fail
6583 instead of killing the process with assert().
6585 75. The SPA client code was not working when the server requested OEM rather
6586 than Unicode encoding.
6588 76. Added code to make require_files with a specific uid setting more usable in
6589 the case where statting the file as root fails - usually a non-root-mounted
6590 NFS file system. When this happens and the failure is EACCES, Exim now
6591 forks a subprocess and does the per-uid checking as the relevant uid.
6593 77. Added process_log_path.
6595 78. If log_file_path was not explicitly set, a setting of check_log_space or
6596 check_log_inodes was ignored.
6598 79. If a space check for the spool or log partitions fails, the incident is now
6599 logged. Of course, in the latter case the data may get lost...
6601 80. Added the %p formatting code to string_format() so that it can be used to
6602 print addresses in debug_print(). Adjusted all the address printing in the
6603 debugging in store.c to use %p rather than %d.
6605 81. There was a concern that a line of code in smtp_in.c could overflow a
6606 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
6607 initially expressed, the concern was not well-founded, because trailing
6608 spaces are removed early. However, if the trailing spaces were followed by
6609 a NULL, they did not get removed, so the overflow was possible. Two fixes
6612 (a) I re-wrote the offending code in a cleaner fashion.
6613 (b) If an incoming SMTP command contains a NULL character, it is rejected
6616 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
6617 runs initgroups(), so that if the Exim user is in any additional groups,
6618 they will be used during message reception.
6624 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob