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.
123 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
124 and 3.4.4 - once the server is enabled to respond to an OCSP request
125 it does even when not requested, resulting in a stapling non-aware
126 client dropping the TLS connection.
128 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
129 support variable-length bit vectors. No functional change.
131 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
132 The I= interface field on outgoing lines is now after the H= remote
133 host field, same as incoming lines. There is a separate
134 outgoing_interface log selector which allows you to disable the
137 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
138 If not running log_selector +smtp_connection the mainlog would be held
139 open indefinitely after a "too many connections" event, including to a
140 deleted file after a log rotate. Leave the per net connection logging
141 leaving it open for efficiency as that will be quickly detected by the
142 check on the next write.
144 HS/01 Bug 1671: Fix post transport crash.
145 Processing the wait-<transport> messages could crash the delivery
146 process if the message IDs didn't exist for some reason. When
147 using 'split_spool_directory=yes' the construction of the spool
148 file name failed already, exposing the same netto behaviour.
150 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
151 mime_regex ACL conditions.
153 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
154 to DSN fail messages (bounces): remote IP, remote greeting, remote response
155 to HELO, local diagnostic string.
157 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
158 log line to debug. Even when configured with a tls authenticator many
159 client connections are expected to not authenticate in this way, so
160 an authenticate fail is not an error.
162 HS/02 Add the Exim version string to the process info. This way exiwhat
163 gives some more detail about the running daemon.
165 JH/06 Bug 1395: time-limit cacheing of DNS lookups, to the TTL value. This may
166 matter for fast-change records such as DNSBLs.
168 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
169 retry record, even if constant. There may be multiple transports with
170 different interface settings and the retry behaviour needs to be kept
173 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
175 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
177 JH/10 Bug 840: fix log_defer_output option of pipe transport
179 JH/11 Bug 830: use same host for all RCPTS of a message, even under
180 hosts_randomize. This matters a lot when combined with mua_wrapper.
182 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
183 ${quote_pgsql:<string>} operator.
185 JH/13 Bug 1708: avoid misaligned access in cached lookup.
187 JH/14 Change header file name for freeradius-client. Relevant if compiling
188 with Radius support; from the Gentoo tree and checked under Fedora.
190 JH/15 Bug 1712: Introduce $prdr_requested flag variable
192 JH/16 Bug 1714: Permit an empty string as expansion result for transport
193 option transport_filter, meaning no filtering.
195 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
197 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
198 defaults to "*" (all hosts). The variable is now available when not built
199 with TLS, default unset, mainly to enable keeping the testuite sane.
200 If a server certificate is not supplied (via tls_certificate) an error is
201 logged, and clients will find TLS connections fail on startup. Presumably
202 they will retry in-clear.
203 Packagers of Exim are strongly encouraged to create a server certificate
204 at installation time.
206 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
207 with the $config_file variable.
209 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
210 in transport context, after the attempt, and per-recipient. The latter type
211 is per host attempted. The event data is the error message, and the errno
212 information encodes the lookup type (A vs. MX) used for the (first) host,
213 and the trailing two digits of the smtp 4xx reponse.
215 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
216 to write to mainlog (or rejectlog, paniclog) in the window between file
217 creation and permissions/ownership being changed. Particularly affects
218 installations where exicyclog is run as root, rather than exim user;
219 result is that the running daemon panics and dies.
221 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
223 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
224 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
225 "pri" and "weight". Note that the previous implicit priority given by the
226 list order is no longer honoured.
228 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalisation
231 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
232 by defining SUPPORT_SOCKS.
234 JH/26 Move PROXY support from Experimental to mainline, enabled for a build
235 by defining SUPPORT_PROXY. Note that the proxy_required_hosts option
236 is renamed to hosts_proxy, and the proxy_{host,target}_{address,port}.
237 variables are renamed to proxy_{local,external}_{address,port}.
239 JH/27 Move Internationalisation support from Experimental to mainline, enabled
240 for a build by defining SUPPORT_I18N
242 JH/28 Bug 1745: Fix redis lookups to handle (quoted) spaces embedded in parts
243 of the query string, and make ${quote_redis:} do that quoting.
245 JH/29 Move Events support from Experimental to mainline, enabled by default
246 and removable for a build by defining DISABLE_EVENT.
248 JH/30 Updated DANE implementation code to current from Viktor Dukhovni.
250 JH/31 Fix bug with hosts_connection_nolog and named-lists which were wrongly
251 cached by the daemon.
253 JH/32 Move Redis support from Experimental to mainline, enabled for a build
254 by defining LOOKUP_REDIS. The libhiredis library is required.
256 JH/33 Bug 1748: Permit ACL dnslists= condition in non-smtp ACLs if explicit
257 keys are given for lookup.
259 JH/34 Bug 1192: replace the embedded copy of PolarSSL RSA routines in the DKIM
260 support, by using OpenSSL or GnuTLS library ones. This means DKIM is
261 only supported when built with TLS support. The PolarSSL SHA routines
262 are still used when the TLS library is too old for convenient support.
264 JH/35 Require SINGLE_DH_USE by default in OpenSSL (main config option
265 openssl_options), for security. OpenSSL forces this from version 1.1.0
266 server-side so match that on older versions.
268 JH/36 Bug 1778: longstanding bug in memory use by the ${run } expansion: A fresh
269 allocation for $value could be released as the expansion processing
270 concluded, but leaving the global pointer active for it.
272 JH/37 Bug 1769: Permit a VRFY ACL to override the default 252 response,
273 and to use the domains and local_parts ACL conditions.
275 JH/38 Fix cutthrough bug with body lines having a single dot. The dot was
276 incorrectly not doubled on cutthrough transmission, hence seen as a
277 body-termination at the receiving system - resulting in truncated mails.
278 Commonly the sender saw a TCP-level error, and retransmitted the message
279 via the normal store-and-forward channel. This could result in duplicates
280 received - but deduplicating mailstores were liable to retain only the
281 initial truncated version.
283 JH/39 Bug 1781: Fix use of DKIM private-keys having trailing '=' in the base-64.
285 JH/40 Fix crash in queryprogram router when compiled with EXPERIMENTAL_SRS.
287 JH/41 Bug 1792: Fix selection of headers to sign for DKIM: bottom-up. While
288 we're in there, support oversigning also; bug 1309.
290 JH/42 Bug 1796: Fix error logged on a malware scanner connection failure.
292 HS/04 Add support for keep_environment and add_environment options.
294 JH/43 Tidy coding issues detected by gcc --fsanitize=undefined. Some remain;
295 either intentional arithmetic overflow during PRNG, or testing config-
298 JH/44 Bug 1800: The combination of a -bhc commandline option and cutthrough
299 delivery resulted in actual delivery. Cancel cutthrough before DATA
302 JH/45 Fix cutthrough, when connection not opened by verify and target hard-
303 rejects a recipient: pass the reject to the originator.
305 JH/46 Multiple issues raised by Coverity. Some were obvious or plausible bugs.
306 Many were false-positives and ignorable, but it's worth fixing the
309 JH/47 Fix build on HP-UX and older Solaris, which need (un)setenv now also
310 for the new environment-manipulation done at startup. Move the routines
311 from being local to tls.c to being global via the os.c file.
313 JH/48 Bug 1807: Fix ${extract } for the numeric/3-string case. While preparsing
314 an extract embedded as result-arg for a map, the first arg for extract
315 is unavailable so we cannot tell if this is a numbered or keyed
316 extraction. Accept either.
322 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
325 JH/02 The smtp transport option "multi_domain" is now expanded.
327 JH/03 The smtp transport now requests PRDR by default, if the server offers
330 JH/04 Certificate name checking on server certificates, when exim is a client,
331 is now done by default. The transport option tls_verify_cert_hostnames
332 can be used to disable this per-host. The build option
333 EXPERIMENTAL_CERTNAMES is withdrawn.
335 JH/05 The value of the tls_verify_certificates smtp transport and main options
336 default to the word "system" to access the system default CA bundle.
337 For GnuTLS, only version 3.0.20 or later.
339 JH/06 Verification of the server certificate for a TLS connection is now tried
340 (but not required) by default. The verification status is now logged by
341 default, for both outbound TLS and client-certificate supplying inbound
344 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
347 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
348 Status Notification (bounce) messages are now MIME format per RFC 3464.
349 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
350 under the control of the dsn_advertise_hosts option, and routers may
351 have a dsn_lasthop option.
353 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
354 default, modifiable by a malware= option. The list separator for
355 the options can now be changed in the usual way. Bug 68.
357 JH/10 The smtp_receive_timeout main option is now expanded before use.
359 JH/11 The incoming_interface log option now also enables logging of the
360 local interface on delivery outgoing connections.
362 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
363 if the interface and destination host and port all match.
365 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
368 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
369 Patch from Andrew Lewis.
371 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
372 now supports optional time-restrictions, weighting, and priority
373 modifiers per server. Patch originally by <rommer@active.by>.
375 JH/16 The spamd_address main option now supports a mixed list of local
376 and remote servers. Remote servers can be IPv6 addresses, and
377 specify a port-range.
379 JH/17 Bug 68: The spamd_address main option now supports an optional
380 timeout value per server.
382 JH/18 Bug 1581: Router and transport options headers_add/remove can
383 now have the list separator specified.
385 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
388 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
391 JH/21 Support for the A6 type of dns record is withdrawn.
393 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
394 rather than the verbs used.
396 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
397 from 255 to 1024 chars.
399 JH/24 Verification callouts now attempt to use TLS by default.
401 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
402 are generic router options now. The defaults didn't change.
404 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
405 Original patch from Alexander Shikoff, worked over by JH.
407 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
408 files if ZCAT_COMMAND is not executable.
410 JH/26 Bug 1539: Add timout/retry options on dnsdb lookups.
412 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
414 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
415 Normally benign, it bites when the pair was led to by a CNAME;
416 modern usage is to not canoicalize the domain to a CNAME target
417 (and we were inconsistent anyway for A-only vs AAAA+A).
419 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
421 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
422 when evaluating $sender_host_dnssec.
424 JH/31 Check the HELO verification lookup for DNSSEC, adding new
425 $sender_helo_dnssec variable.
427 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
429 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
431 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
433 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
434 documented as working, but never had. Support all but $spam_report.
436 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
437 added for tls authenticator.
439 HS/03 Add perl_taintmode main config option
444 TL/01 When running the test suite, the README says that variables such as
445 no_msglog_check are global and can be placed anywhere in a specific
446 test's script, however it was observed that placement needed to be near
447 the beginning for it to behave that way. Changed the runtest perl
448 script to read through the entire script once to detect and set these
449 variables, reset to the beginning of the script, and then run through
450 the script parsing/test process like normal.
452 TL/02 The BSD's have an arc4random API. One of the functions to induce
453 adding randomness was arc4random_stir(), but it has been removed in
454 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
455 function when detected.
457 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
458 cause callback expansion.
460 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
461 syntax errors in an expansion can be treated as a string instead of
462 logging or causing an error, due to the internal use of bool_lax
463 instead of bool when processing it.
465 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
466 server certificates when making smtp deliveries.
468 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
470 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
472 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
474 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
475 Merged patch from Sebastian Wiedenroth.
477 JH/05 Fix results-pipe from transport process. Several recipients, combined
478 with certificate use, exposed issues where response data items split
479 over buffer boundaries were not parsed properly. This eventually
480 resulted in duplicates being sent. This issue only became common enough
481 to notice due to the introduction of conection certificate information,
482 the item size being so much larger. Found and fixed by Wolfgang Breyha.
484 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
485 size buffer was used, resulting in syntax errors when an expansion
488 JH/07 Add support for directories of certificates when compiled with a GnuTLS
489 version 3.3.6 or later.
491 JH/08 Rename the TPDA expermimental facility to Event Actions. The #ifdef
492 is EXPERIMENTAL_EVENT, the main-configuration and transport options
493 both become "event_action", the variables become $event_name, $event_data
494 and $event_defer_errno. There is a new variable $verify_mode, usable in
495 routers, transports and related events. The tls:cert event is now also
496 raised for inbound connections, if the main configuration event_action
499 TL/06 In test suite, disable OCSP for old versions of openssl which contained
500 early OCSP support, but no stapling (appears to be less than 1.0.0).
502 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
503 server certificate names available under the smtp transport option
504 "tls_verify_cert_hostname" now do not permit multi-component wildcard
507 JH/10 Time-related extraction expansions from certificates now use the main
508 option "timezone" setting for output formatting, and are consistent
509 between OpenSSL and GnuTLS compilations. Bug 1541.
511 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
512 encoded parameter in the incoming message. Bug 1558.
514 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
515 include certificate info, eximon was claiming there were spoolfile
518 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
520 JH/14 Log delivery-related information more consistently, using the sequence
521 "H=<name> [<ip>]" wherever possible.
523 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
524 are problematic for Debian distribution, omit them from the release
527 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
529 JH/16 Fix string representation of time values on 64bit time_t anchitectures.
532 JH/17 Fix a null-indirection in certextract expansions when a nondefault
533 output list separator was used.
538 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
539 checkers that were complaining about end of non-void function with no
542 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
543 This was a regression intruduced in 4.83 by another bugfix.
545 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
547 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
548 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
554 TF/01 Correctly close the server side of TLS when forking for delivery.
556 When a message was received over SMTP with TLS, Exim failed to clear up
557 the incoming connection properly after forking off the child process to
558 deliver the message. In some situations the subsequent outgoing
559 delivery connection happened to have the same fd number as the incoming
560 connection previously had. Exim would try to use TLS and fail, logging
561 a "Bad file descriptor" error.
563 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
564 utilities have not been installed.
566 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
567 temporary space as the ACL may create new global variables.
569 TL/01 LDAP support uses per connection or global context settings, depending
570 upon the detected version of the libraries at build time.
572 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
573 to extract and use the src ip:port in logging and expansions as if it
574 were a direct connection from the outside internet. PPv2 support was
575 updated based on HAProxy spec change in May 2014.
577 JH/02 Add ${listextract {number}{list}{success}{fail}}.
579 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
580 Properly escape header and check for NULL return.
582 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
585 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
587 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
588 characters in header names, implemented as a verify condition.
589 Contributed by Michael Fischer v. Mollard.
591 TL/05 Rename SPF condition results err_perm and err_temp to standardized
592 results permerror and temperror. Previous values are deprecated but
593 still accepted. In a future release, err_perm and err_temp will be
594 completely removed, which will be a backward incompatibility if the
595 ACL tests for either of these two old results. Patch contributed by
596 user bes-internal on the mailing list.
598 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
600 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
601 selectors, in both main and reject logs.
603 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
606 JH/07 Add malware type "sock" for talking to simple daemon.
608 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
610 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
611 routers/transports under cutthrough routing.
613 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
614 numbers. Touch up "bool" conditional to keep the same definition.
616 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
618 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
620 JH/12 Expand items in router/transport headers_add or headers_remove lists
621 individually rather than the list as a whole. Bug 1452.
623 Required for reasonable handling of multiple headers_ options when
624 they may be empty; requires that headers_remove items with embedded
625 colons must have them doubled (or the list-separator changed).
627 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
628 view the policy declared in the DMARC record. Currently, $dmarc_status
629 is a combined value of both the record presence and the result of the
632 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
634 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
635 dnslookup router and the smtp transport (applying to the forward
638 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
639 of ldap servers used for a specific lookup. Patch provided by Heiko
642 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
643 New variable $lookup_dnssec_authenticated for observability.
645 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
646 Patch submitted by Lars Timman.
648 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
650 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
651 Requires trusted mode and valid format message id, aborts otherwise.
652 Patch contributed by Heiko Schlichting.
654 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
655 certextract with support for various fields. Bug 1358.
657 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
658 is requested by default, modifiable by smtp transport option
661 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
662 operate on certificate variables to give certificate fingerprints
663 Also new ${sha256:cert_variable}.
665 JH/23 The PRDR feature is moved from being Experimental into the mainline.
667 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
668 Christian Aistleitner.
670 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
672 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
673 file. Patch from Wolfgang Breyha.
675 JH/25 Expand the coverage of the delivery $host and $host_address to
676 client authenticators run in verify callout. Bug 1476.
678 JH/26 Port service names are now accepted for tls_on_connect_ports, to
679 align with daemon_smtp_ports. Bug 72.
681 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
682 support and error reporting did not work properly.
684 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
685 and is readable. Patch from Andrew Colin Kissa.
687 TL/14 Enhance documentation of ${run expansion and how it parses the
688 commandline after expansion, particularly in the case when an
689 unquoted variable expansion results in an empty value.
691 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
693 PP/02 Fix internal collision of T_APL on systems which support RFC3123
694 by renaming away from it. Addresses GH issue 15, reported by
697 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
699 TL/15 SECURITY: prevent double expansion in math comparison functions
700 (can expand unsanitized data). Not remotely exploitable.
707 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
709 PP/02 Make -n do something, by making it not do something.
710 When combined with -bP, the name of an option is not output.
712 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
715 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
716 $sender_host_name and config options to manage this, and basic check
719 PP/05 DSCP support for outbound connections and control modifier for inbound.
721 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
722 (Only plugin which currently uses this is kerberos4, which nobody should
723 be using, but we should make it available and other future plugins might
724 conceivably use it, even though it would break NAT; stuff *should* be
725 using channel bindings instead).
727 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
728 name; added for Sendmail compatibility; requires admin caller.
729 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
730 just ignore it); requires trusted caller.
731 Also parse but ignore: -Ac -Am -X<logfile>
734 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
736 TL/02 Add +smtp_confirmation as a default logging option.
738 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
739 Patch by Magnus Holmgren from 2007-02-20.
741 TL/04 Bugzilla 1281 - Spec typo.
742 Bugzilla 1283 - Spec typo.
743 Bugzilla 1290 - Spec grammar fixes.
745 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
747 TL/06 Add Experimental DMARC support using libopendmarc libraries.
749 TL/07 Fix an out of order global option causing a segfault. Reported to dev
750 mailing list by by Dmitry Isaikin.
752 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
754 JH/02 Support "G" suffix to numbers in ${if comparisons.
756 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
758 NM/01 Bugzilla 1197 - Spec typo
759 Bugzilla 1196 - Spec examples corrections
761 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
763 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
764 gnutls_enable_pkcs11, but renamed to more accurately indicate its
767 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
768 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
770 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
771 "acl {{name}{arg}...}", and optional args on acl condition
774 JH/05 Permit multiple router/transport headers_add/remove lines.
776 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
778 JH/07 Avoid using a waiting database for a single-message-only transport.
779 Performance patch from Paul Fisher. Bugzilla 1262.
781 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
784 JH/09 Add $headers_added variable, with content from use of ACL modifier
785 add_header (but not yet added to the message). Bugzilla 199.
787 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
788 Pulled from Bugzilla 817 by Wolfgang Breyha.
790 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
792 (nb: this is the same fix as in Exim 4.80.1)
794 JH/11 Add A= logging on delivery lines, and a client_set_id option on
797 JH/12 Add optional authenticated_sender logging to A= and a log_selector
800 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
802 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
803 advertise SMTP AUTH mechanism to us, instead of a generic
804 protocol violation error. Also, make Exim more robust to bad
805 data from the Dovecot auth socket.
807 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
809 When a queue runner is handling a message, Exim first routes the
810 recipient addresses, during which it prunes them based on the retry
811 hints database. After that it attempts to deliver the message to
812 any remaining recipients. It then updates the hints database using
815 So if a recipient address works intermittently, it can get repeatedly
816 deferred at routing time. The retry hints record remains fresh so the
817 address never reaches the final cutoff time.
819 This is a fairly common occurrence when a user is bumping up against
820 their storage quota. Exim had some logic in its local delivery code
821 to deal with this. However it did not apply to per-recipient defers
822 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
824 This change adds a proper retry rule check during routing so that the
825 final cutoff time is checked against the message's age. We only do
826 this check if there is an address retry record and there is not a
827 domain retry record; this implies that previous attempts to handle
828 the address had the retry_use_local_parts option turned on. We use
829 this as an approximation for the destination being like a local
830 delivery, as in LMTP.
832 I suspect this new check makes the old local delivery cutoff check
833 redundant, but I have not verified this so I left the code in place.
835 TF/02 Correct gecos expansion when From: is a prefix of the username.
837 Test 0254 submits a message to Exim with the header
841 When I ran the test suite under the user fanf2, Exim expanded
842 the header to contain my full name, whereas it should have added
843 a Resent-Sender: header. It erroneously treats any prefix of the
844 username as equal to the username.
846 This change corrects that bug.
848 GF/01 DCC debug and logging tidyup
849 Error conditions log to paniclog rather than rejectlog.
850 Debug lines prefixed by "DCC: " to remove any ambiguity.
852 TF/03 Avoid unnecessary rebuilds of lookup-related code.
854 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
855 Bug spotted by Jeremy Harris; was flawed since initial commit.
856 Would have resulted in OCSP responses post-SNI triggering an Exim
857 NULL dereference and crash.
859 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
861 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
862 Bug detection, analysis and fix by Samuel Thibault.
863 Bugzilla 1331, Debian bug #698092.
865 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
867 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
868 Server implementation by Todd Lyons, client by JH.
869 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
870 config variable "prdr_enable" controls whether the server
871 advertises the facility. If the client requests PRDR a new
872 acl_data_smtp_prdr ACL is called once for each recipient, after
873 the body content is received and before the acl_smtp_data ACL.
874 The client is controlled by bolth of: a hosts_try_prdr option
875 on the smtp transport, and the server advertisement.
876 Default client logging of deliveries and rejections involving
877 PRDR are flagged with the string "PRDR".
879 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
880 fclose(). Diagnosis by Todd Lyons.
882 PP/17 Update configure.default to handle IPv6 localhost better.
883 Patch by Alain Williams (plus minor tweaks).
886 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
887 This is now consistent with GnuTLS, and is now documented: the
888 previous undocumented portable approach to treating the option as
889 unset was to force an expansion failure. That still works, and
890 an empty string is now equivalent.
892 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
893 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
894 not performing validation itself.
896 PP/20 Added force_command boolean option to pipe transport.
897 Patch from Nick Koston, of cPanel Inc.
899 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
902 TF/04 Added udpsend ACL modifer and hexquote expansion operator
904 PP/21 Fix eximon continuous updating with timestamped log-files.
905 Broken in a format-string cleanup in 4.80, missed when I repaired the
906 other false fix of the same issue.
907 Report and fix from Heiko Schlichting.
910 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
911 Report from Prashanth Katuri.
913 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
914 It's SecureTransport, so affects any MacOS clients which use the
915 system-integrated TLS libraries, including email clients.
917 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
918 using a MIME ACL for non-SMTP local injection.
919 Report and assistance in diagnosis by Warren Baker.
921 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
923 JH/16 Fix comparisons for 64b. Bugzilla 1385.
925 TL/09 Add expansion variable $authenticated_fail_id to keep track of
926 last id that failed so it may be referenced in subsequent ACL's.
928 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
931 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
932 ldap library initialization, allowing self-signed CA's to be
933 used. Also properly sets require_cert option later in code by
934 using NULL (global ldap config) instead of ldap handle (per
935 session). Bug diagnosis and testing by alxgomz.
937 TL/12 Enhanced documentation in the ratelimit.pl script provided in
938 the src/util/ subdirectory.
940 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
941 renamed to Transport Post Delivery Action by Jeremy Harris, as
944 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
945 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
946 redis_servers = needs to be configured which will be used by the redis
947 lookup. Patch from Warren Baker, of The Packet Hub.
949 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
951 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
952 hostname or reverse DNS when processing a host list. Used suggestions
953 from multiple comments on this bug.
955 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
957 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
958 Missed a few lines, added it to make the runtest require no keyboard
961 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
962 contains upper case chars. Make router use caseful_local_part.
964 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
965 support when GnuTLS has been built with p11-kit.
971 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
973 This, or similar/improved, will also be change PP/11 of 4.82.
979 PP/01 Handle short writes when writing local log-files.
980 In practice, only affects FreeBSD (8 onwards).
981 Bugzilla 1053, with thanks to Dmitry Isaikin.
983 NM/01 Bugzilla 949 - Documentation tweak
985 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
988 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
990 PP/02 Implemented gsasl authenticator.
992 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
994 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
995 `pkg-config foo` for cflags/libs.
997 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
998 with rest of GSASL and with heimdal_gssapi.
1000 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
1001 `pkg-config foo` for cflags/libs for the TLS implementation.
1003 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
1004 properties get this fed in as external SSF. A number of robustness
1005 and debugging improvements to the cyrus_sasl authenticator.
1007 PP/08 cyrus_sasl server now expands the server_realm option.
1009 PP/09 Bugzilla 1214 - Log authentication information in reject log.
1010 Patch by Jeremy Harris.
1012 PP/10 Added dbmjz lookup type.
1014 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
1016 PP/12 MAIL args handles TAB as well as SP, for better interop with
1017 non-compliant senders.
1018 Analysis and variant patch by Todd Lyons.
1020 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
1021 Bug report from Lars Müller <lars@samba.org> (via SUSE),
1022 Patch from Dirk Mueller <dmueller@suse.com>
1024 PP/13 tls_peerdn now print-escaped for spool files.
1025 Observed some $tls_peerdn in wild which contained \n, which resulted
1026 in spool file corruption.
1028 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
1029 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
1030 or write after TLS renegotiation, which otherwise led to messages
1033 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
1034 as a tracking header (ie: a signed header comes before the signature).
1035 Patch from Wolfgang Breyha.
1037 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
1038 comma-sep list; embedded commas doubled.
1040 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
1042 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
1044 Report and patch from Dmitry Banschikov.
1046 PP/16 Removed "dont_insert_empty_fragments" fron "openssl_options".
1047 Removed SSL_clear() after SSL_new() which led to protocol negotiation
1048 failures. We appear to now support TLS1.1+ with Exim.
1050 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
1051 lets Exim select keys and certificates based upon TLS SNI from client.
1052 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
1053 before an outbound SMTP session. New log_selector, +tls_sni.
1055 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
1056 NULL dereference. Report and patch from Alun Jones.
1058 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
1059 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
1060 Not seeing resolver debug output on NetBSD, but suspect this is a
1061 resolver implementation change.
1063 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
1064 Left warnings. Added "eximon gdb" invocation mode.
1066 PP/21 Defaulting "accept_8bitmime" to true, not false.
1068 PP/22 Added -bw for inetd wait mode support.
1070 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
1071 locate the relevant includes and libraries. Made this the default.
1073 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
1074 Bugzilla 1246, report and most of solution from Tomasz Kusy.
1076 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
1077 This may cause build issues on older platforms.
1079 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
1080 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
1081 gnutls_require_mac & gnutls_require_protocols (no longer supported).
1082 Added SNI support via GnuTLS too.
1083 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
1085 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
1087 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
1088 Applied second patch from Janne, implementing suggestion to default
1089 multiple-strings-in-record handling to match SPF spec.
1091 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
1093 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
1094 read-only, out of scope).
1095 Patch from Wolfgang Breyha, report from Stuart Northfield.
1097 PP/29 Fix three issues highlighted by clang analyser static analysis.
1098 Only crash-plausible issue would require the Cambridge-specific
1099 iplookup router and a misconfiguration.
1100 Report from Marcin Mirosław.
1102 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
1104 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
1105 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
1106 As part of this, removing so much warning spew let me fix some minor
1107 real issues in debug logging.
1109 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
1110 assignment on my part. Fixed.
1112 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
1113 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
1114 Janne Snabb (who went above and beyond: thank you).
1116 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
1117 string otherwise requires a connection and a bunch more work and it's
1118 relatively easy to get wrong. Should also expose TLS library linkage
1121 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
1122 64-bit ${eval} (JH/03).
1124 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
1125 GNU libc to support some of the 64-bit stuff, should not lead to
1126 conflicts. Defined before os.h is pulled in, so if a given platform
1127 needs to override this, it can.
1129 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
1130 protection layer was required, which is not implemented.
1131 Bugzilla 1254, patch from Wolfgang Breyha.
1133 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
1134 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
1135 tls_dhparam take prime identifiers. Also unbreak combination of
1136 OpenSSL+DH_params+TLSSNI.
1138 PP/39 Disable SSLv2 by default in OpenSSL support.
1144 PP/01 Solaris build fix for Oracle's LDAP libraries.
1145 Bugzilla 1109, patch from Stephen Usher.
1147 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
1149 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
1152 TF/02 Fix a couple more cases where we did not log the error message
1153 when unlink() failed. See also change 4.74-TF/03.
1155 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
1156 lock up or crash if it happened to be inside a call to libc when it
1157 got a SIGUSR1 from exiwhat.
1159 The SIGUSR1 handler appends the current process status to the process
1160 log which is later printed by exiwhat. It used to use the general
1161 purpose logging code to do this, but several functions it calls are
1162 not safe for signals.
1164 The new output code in the SIGUSR1 handler is specific to the process
1165 log, and simple enough that it's easy to inspect for signal safety.
1166 Removing some special cases also simplifies the general logging code.
1167 Removing the spurious timestamps from the process log simplifies
1170 TF/04 Improved ratelimit ACL condition.
1172 The /noupdate option has been deprecated in favour of /readonly which
1173 has clearer semantics. The /leaky, /strict, and /readonly update modes
1174 are mutually exclusive. The update mode is no longer included in the
1175 database key; it just determines when the database is updated. (This
1176 means that when you upgrde Exim will forget old rate measurements.)
1178 Exim now checks that the per_* options are used with an update mode that
1179 makes sense for the current ACL. For example, when Exim is processing a
1180 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
1181 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
1182 must specify per_mail/readonly. If you omit the update mode it defaults to
1183 /leaky where that makes sense (as before) or /readonly where required.
1185 The /noupdate option is now undocumented but still supported for
1186 backwards compatibility. It is equivalent to /readonly except that in
1187 ACLs where /readonly is required you may specify /leaky/noupdate or
1188 /strict/noupdate which are treated the same as /readonly.
1190 A useful new feature is the /count= option. This is a generalization
1191 of the per_byte option, so that you can measure the throughput of other
1192 aggregate values. For example, the per_byte option is now equivalent
1193 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
1195 The per_rcpt option has been generalized using the /count= mechanism
1196 (though it's more complicated than the per_byte equivalence). When it is
1197 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
1198 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
1199 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
1200 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
1201 Note that using per_rcpt with a non-readonly update mode in more than
1202 one ACL will cause the recipients to be double-counted. (The per_mail
1203 and per_byte options don't have this problem.)
1205 The handling of very low rates has changed slightly. If the computed rate
1206 is less than the event's count (usually one) then this event is the first
1207 after a long gap. In this case the rate is set to the same as this event's
1208 count, so that the first message of a spam run is counted properly.
1210 The major new feature is a mechanism for counting the rate of unique
1211 events. The new per_addr option counts the number of different
1212 recipients that someone has sent messages to in the last time period. It
1213 behaves like per_rcpt if all the recipient addresses are different, but
1214 duplicate recipient addresses do not increase the measured rate. Like
1215 the /count= option this is a general mechanism, so the per_addr option
1216 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
1217 example, measure the rate that a client uses different sender addresses
1218 with the options per_mail/unique=$sender_address. There are further
1219 details in the main documentation.
1221 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
1223 TF/06 Removed a few PCRE remnants.
1225 TF/07 Automatically extract Exim's version number from tags in the git
1226 repository when doing development or release builds.
1228 PP/02 Raise smtp_cmd_buffer_size to 16kB.
1229 Bugzilla 879. Patch from Paul Fisher.
1231 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
1232 Heavily based on revision 40f9a89a from Simon Arlott's tree.
1235 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
1237 PP/05 Variable $av_failed, true if the AV scanner deferred.
1238 Bugzilla 1078. Patch from John Horne.
1240 PP/06 Stop make process more reliably on build failure.
1241 Bugzilla 1087. Patch from Heiko Schlittermann.
1243 PP/07 Make maildir_use_size_file an _expandable_ boolean.
1244 Bugzilla 1089. Patch from Heiko Schlittermann.
1246 PP/08 Handle ${run} returning more data than OS pipe buffer size.
1247 Bugzilla 1131. Patch from Holger Weiß.
1249 PP/09 Handle IPv6 addresses with SPF.
1250 Bugzilla 860. Patch from Wolfgang Breyha.
1252 PP/10 GnuTLS: support TLS 1.2 & 1.1.
1254 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
1257 PP/11 match_* no longer expand right-hand-side by default.
1258 New compile-time build option, EXPAND_LISTMATCH_RHS.
1259 New expansion conditions, "inlist", "inlisti".
1261 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
1263 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
1265 PP/14 fix log_write() format string regression from TF/03.
1266 Bugzilla 1152. Patch from Dmitry Isaikin.
1272 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
1274 PP/02 Harmonised TLS library version reporting; only show if debugging.
1275 Layout now matches that introduced for other libraries in 4.74 PP/03.
1277 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
1279 PP/04 New "dns_use_edns0" global option.
1281 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
1284 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
1285 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
1287 TK/01 Updated PolarSSL code to 0.14.2.
1288 Bugzilla 1097. Patch from Andreas Metzler.
1290 PP/07 Catch divide-by-zero in ${eval:...}.
1291 Fixes bugzilla 1102.
1293 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
1296 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
1297 format-string attack -- SECURITY: remote arbitrary code execution.
1299 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
1300 time unintentionally subject to list matching rules, letting the header
1301 cause arbitrary Exim lookups (of items which can occur in lists, *not*
1302 arbitrary string expansion). This allowed for information disclosure.
1304 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
1305 INT_MIN/-1 -- value coerced to INT_MAX.
1311 NM/01 Workround for PCRE version dependency in version reporting
1314 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
1315 This fixes portability to compilers other than gcc, notably
1316 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
1318 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
1319 makefiles for portability to HP-UX and POSIX correctness.
1321 PP/01 Permit LOOKUP_foo enabling on the make command-line.
1322 Also via indented variable definition in the Makefile.
1323 (Debugging by Oliver Heesakkers).
1325 PP/02 Restore caching of spamd results with expanded spamd_address.
1326 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
1328 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
1329 Improves build reliability. Fix from: Frank Elsner
1331 NM/02 Fix wide character breakage in the rfc2047 coding
1332 Fixes bug 1064. Patch from Andrey N. Oktyabrski
1334 NM/03 Allow underscore in dnslist lookups
1335 Fixes bug 1026. Patch from Graeme Fowler
1337 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
1338 Code patches from Adam Ciarcinski of NetBSD.
1340 NM/04 Fixed exiqgrep to cope with mailq missing size issue
1343 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
1344 is logged, to avoid truncation. Patch from John Horne.
1346 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
1347 Patch from Jakob Hirsch.
1349 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
1350 SQL string expansion failure details.
1351 Patch from Andrey Oktyabrski.
1353 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
1354 Patch from Simon Arlott.
1356 PP/09 New lookups functionality failed to compile on old gcc which rejects
1357 extern declarations in function scope.
1358 Patch from Oliver Fleischmann
1360 PP/10 Use sig_atomic_t for flags set from signal handlers.
1361 Check getgroups() return and improve debugging.
1362 Fixed developed for diagnosis in bug 927 (which turned out to be
1365 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
1366 Patch from Mark Zealey.
1368 PP/12 Bugzilla 1056: Improved spamd server selection.
1369 Patch from Mark Zealey.
1371 PP/13 Bugzilla 1086: Deal with maildir quota file races.
1372 Based on patch from Heiko Schlittermann.
1374 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
1375 Patch from Uwe Doering, sign-off by Michael Haardt.
1377 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
1378 variable declaration deep within a block. Bug and patch from
1381 PP/15 lookups-Makefile IRIX compatibilty coercion.
1383 PP/16 Make DISABLE_DKIM build knob functional.
1385 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
1386 Patch by Simon Arlott
1388 TF/03 Fix valgrind.h portability to C89 compilers that do not support
1389 variable argument macros. Our copy now differs from upstream.
1395 TF/01 Failure to get a lock on a hints database can have serious
1396 consequences so log it to the panic log.
1398 TF/02 Log LMTP confirmation messages in the same way as SMTP,
1399 controlled using the smtp_confirmation log selector.
1401 TF/03 Include the error message when we fail to unlink a spool file.
1403 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
1404 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
1405 for maintaining out-of-tree patches for some time.
1407 PP/01 Bugzilla 139: Documentation and portability issues.
1408 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
1409 Handle per-OS dynamic-module compilation flags.
1411 PP/02 Let /dev/null have normal permissions.
1412 The 4.73 fixes were a little too stringent and complained about the
1413 permissions on /dev/null. Exempt it from some checks.
1414 Reported by Andreas M. Kirchwitz.
1416 PP/03 Report version information for many libraries, including
1417 Exim version information for dynamically loaded libraries. Created
1418 version.h, now support a version extension string for distributors
1419 who patch heavily. Dynamic module ABI change.
1421 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
1422 privilege escalation vulnerability whereby the Exim run-time user
1423 can cause root to append content of the attacker's choosing to
1426 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
1429 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
1430 If dropping privileges for untrusted macros, we disabled normal logging
1431 on the basis that it would fail; for the Exim run-time user, this is not
1432 the case, and it resulted in successful deliveries going unlogged.
1433 Fixed. Reported by Andreas Metzler.
1439 PP/01 Date: & Message-Id: revert to normally being appended to a message,
1440 only prepend for the Resent-* case. Fixes regression introduced in
1441 Exim 4.70 by NM/22 for Bugzilla 607.
1443 PP/02 Include check_rfc2047_length in configure.default because we're seeing
1444 increasing numbers of administrators be bitten by this.
1446 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
1448 PP/03 Bugzilla 994: added openssl_options main configuration option.
1450 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
1452 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
1454 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
1456 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
1457 without a peer certificate, leading to a segfault because of an
1458 assumption that peers always have certificates. Be a little more
1459 paranoid. Problem reported by Martin Tscholak.
1461 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
1462 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
1463 NB: ClamAV planning to remove STREAM in "middle of 2010".
1464 CL also introduces -bmalware, various -d+acl logging additions and
1465 more caution in buffer sizes.
1467 PP/09 Implemented reverse_ip expansion operator.
1469 PP/10 Bugzilla 937: provide a "debug" ACL control.
1471 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
1473 PP/12 Bugzilla 973: Implement --version.
1475 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
1477 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
1479 PP/15 Bugzilla 816: support multiple condition rules on Routers.
1481 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
1482 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
1483 ignore trailing whitespace.
1485 JJ/02 prevent non-panic DKIM error from being sent to paniclog
1487 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
1490 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
1491 Notification from Dr Andrew Aitchison.
1493 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
1494 ExtendedDetectionInfo response format.
1495 Notification from John Horne.
1497 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
1500 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
1501 XSL and documented dependency on system catalogs, with examples of how
1504 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
1507 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
1508 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
1509 configuration file which is writeable by the Exim user or group.
1511 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
1512 of configuration files to cover files specified with the -C option if
1513 they are going to be used with root privileges, not just the default
1516 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
1517 option (effectively making it always true).
1519 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
1520 files to be used while preserving root privileges.
1522 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
1523 that rogue child processes cannot use them.
1525 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
1526 run-time user, instead of root.
1528 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
1529 Exim run-time user without dropping privileges.
1531 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
1532 result string, instead of calling string_vformat() twice with the same
1535 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
1536 for other users. Others should always drop root privileges if they use
1537 -C on the command line, even for a whitelisted configure file.
1539 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
1541 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
1547 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
1548 $data_path, and $header_path variables; fixed documentation bugs and
1551 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
1552 exipick to access non-standard spools, including the "frozen" queue
1555 NM/01 Bugzilla 965: Support mysql stored procedures.
1556 Patch from Alain Williams
1558 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
1560 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
1561 Patch from Andreas Metzler
1563 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
1564 Patch from Kirill Miazine
1566 NM/05 Bugzilla 671: Added umask to procmail example.
1568 JJ/03 installed exipick 20100323.0, fixing doc bug
1570 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
1571 directory. Notification and patch from Dan Rosenberg.
1573 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
1575 TK/02 Improve log output when DKIM signing operation fails.
1577 MH/01 Treat the transport option dkim_domain as a colon separated
1578 list, not as a single string, and sign the message with each element,
1579 omitting multiple occurences of the same signer.
1581 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
1582 Bugzilla 985, 986. Patch by Simon Arlott
1584 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
1585 Patch by Simon Arlott
1587 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
1588 MBX locking. Notification from Dan Rosenberg.
1594 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
1596 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
1598 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
1600 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
1602 NM/04 Bugzilla 915: Flag broken perl installation during build.
1608 TK/01 Added patch by Johannes Berg that expands the main option
1609 "spamd_address" if it starts with a dollar sign.
1611 TK/02 Write list of recipients to X-Envelope-Sender header when building
1612 the mbox-format spool file for content scanning (suggested by Jakob
1615 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
1616 (http://www.dcc-servers.net/) support via dccifd. Activated by
1617 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
1619 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
1620 by Mark Daniel Reidel <mr@df.eu>.
1622 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
1623 When building exim an external PCRE library is now needed -
1624 PCRE is a system library on the majority of modern systems.
1625 See entry on PCRE_LIBS in EDITME file.
1627 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
1628 conversation. Added nologin parameter to request.
1629 Patch contributed by Kirill Miazine.
1631 TF/01 Do not log submission mode rewrites if they do not change the address.
1633 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
1635 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
1636 log files in place. Contributed by Roberto Lima.
1638 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
1640 TF/03 Bugzilla 615: When checking the local_parts router precondition
1641 after a local_part_suffix or local_part_prefix option, Exim now
1642 does not use the address's named list lookup cache, since this
1643 contains cached lookups for the whole local part.
1645 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
1646 Robert Millan. Documentation is in experimental-spec.txt.
1648 TF/04 Bugzilla 668: Fix parallel build (make -j).
1650 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
1652 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
1653 Patch provided by Jan Srzednicki.
1655 TF/05 Leading white space used to be stripped from $spam_report which
1656 wrecked the formatting. Now it is preserved.
1658 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
1659 that they are available at delivery time.
1661 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
1663 TF/08 TLS error reporting now respects the incoming_interface and
1664 incoming_port log selectors.
1666 TF/09 Produce a more useful error message if an SMTP transport's hosts
1667 setting expands to an empty string.
1669 NM/06 Bugzilla 744: EXPN did not work under TLS.
1670 Patch provided by Phil Pennock.
1672 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
1673 Patch provided by Richard Godbee.
1675 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
1676 acl_smtp_notquit, added index entry.
1678 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
1679 Patch provided by Eugene Bujak.
1681 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
1682 accept(). Patch provided by Maxim Dounin.
1684 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
1685 Patch provided by Phil Pennock.
1687 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
1689 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
1690 Patch provided by Brad "anomie" Jorsch.
1692 NM/14 Bugzilla 622: Added timeout setting to transport filter.
1693 Patch provided by Dean Brooks.
1695 TK/05 Add native DKIM support (does not depend on external libraries).
1697 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
1698 Patch provided by Graeme Fowler.
1700 NM/16 Bugzilla 851: Documentation example syntax fix.
1702 NM/17 Changed NOTICE file to remove references to embedded PCRE.
1704 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
1707 NM/19 Bugzilla 745: TLS version reporting.
1708 Patch provided by Phil Pennock.
1710 NM/20 Bugzilla 167: bool: condition support.
1711 Patch provided by Phil Pennock.
1713 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
1714 clients. Patch provided by Phil Pennock.
1716 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
1717 Patch provided by Brad "anomie" Jorsch.
1719 NM/23 Bugzilla 687: Fix misparses in eximstats.
1720 Patch provided by Heiko Schlittermann.
1722 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
1723 Patch provided by Heiko Schlittermann.
1725 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
1726 plus update to original patch.
1728 NM/26 Bugzilla 799: Documentation correction for ratelimit.
1730 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
1731 Patch provided by David Brownlee.
1733 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
1735 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
1737 NM/30 Bugzilla 888: TLS documentation bugfixes.
1739 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
1741 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
1742 Unlike the original bugzilla I have changed all shell scripts in src tree.
1744 NM/33 Bugzilla 898: Transport filter timeout fix.
1745 Patch by Todd Rinaldo.
1747 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
1748 Patch by Serge Demonchaux.
1750 NM/35 Bugzilla 39: Base64 decode bug fixes.
1751 Patch by Jakob Hirsch.
1753 NM/36 Bugzilla 909: Correct connect() call in dcc code.
1755 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
1757 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
1759 NM/39 Bugzilla 911: Fixed MakeLinks build script.
1765 TK/01 Add preliminary DKIM support. Currently requires a forked version of
1766 ALT-N's libdkim that I have put here:
1767 http://duncanthrax.net/exim-experimental/
1769 Note to Michael Haardt: I had to rename some vars in sieve.c. They
1770 were called 'true' and it seems that C99 defines that as a reserved
1771 keyword to be used with 'bool' variable types. That means you could
1772 not include C99-style headers which use bools without triggering
1773 build errors in sieve.c.
1775 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
1776 as mailq or other aliases. Changed the --help handling significantly
1777 to do whats expected. exim_usage() emits usage/help information.
1779 SC/01 Added the -bylocaldomain option to eximstats.
1781 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
1783 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
1785 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
1791 PH/01 Another patch from the Sieve maintainer.
1793 PH/02 When an IPv6 address is converted to a string for single-key lookup
1794 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
1795 dots are used instead of colons so that keys in lsearch files need not
1796 contain colons. This was done some time before quoting was made available
1797 in lsearch files. However, iplsearch files do require colons in IPv6 keys
1798 (notated using the quote facility) so as to distinguish them from IPv4
1799 keys. This meant that lookups for IP addresses in host lists did not work
1800 for iplsearch lookups.
1802 This has been fixed by arranging for IPv6 addresses to be expressed with
1803 colons if the lookup type is iplsearch. This is not incompatible, because
1804 previously such lookups could never work.
1806 The situation is now rather anomolous, since one *can* have colons in
1807 ordinary lsearch keys. However, making the change in all cases is
1808 incompatible and would probably break a number of configurations.
1810 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
1813 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
1814 conversion specification without a maximum field width, thereby enabling
1815 a rogue spamd server to cause a buffer overflow. While nobody in their
1816 right mind would setup Exim to query an untrusted spamd server, an
1817 attacker that gains access to a server running spamd could potentially
1818 exploit this vulnerability to run arbitrary code as the Exim user.
1820 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
1821 $primary_hostname instead of what libspf2 thinks the hosts name is.
1823 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
1824 a directory entry by the name of the lookup key. Previously, if a
1825 symlink pointed to a non-existing file or a file in a directory that
1826 Exim lacked permissions to read, a lookup for a key matching that
1827 symlink would fail. Now it is enough that a matching directory entry
1828 exists, symlink or not. (Bugzilla 503.)
1830 PH/03 The body_linecount and body_zerocount variables are now exported in the
1833 PH/04 Added the $dnslist_matched variable.
1835 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
1836 This means they are set thereafter only if the connection becomes
1839 PH/06 Added the client_condition to authenticators so that some can be skipped
1840 by clients under certain conditions.
1842 PH/07 The error message for a badly-placed control=no_multiline_responses left
1843 "_responses" off the end of the name.
1845 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
1847 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
1848 (without spaces) instead of just copying the configuration text.
1850 PH/10 Added the /noupdate option to the ratelimit ACL condition.
1852 PH/11 Added $max_received_linelength.
1854 PH/12 Added +ignore_defer and +include_defer to host lists.
1856 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
1857 way in which PCRE > 7.0 is built.
1859 PH/14 Implemented queue_only_load_latch.
1861 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
1862 MAIL command. The effect was to mangle the value on 64-bit systems.
1864 PH/16 Another patch from the Sieve maintainer.
1866 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
1868 PH/18 If a system quota error occurred while trying to create the file for
1869 a maildir delivery, the message "Mailbox is full" was not appended to the
1870 bounce if the delivery eventually timed out. Change 4.67/27 below applied
1871 only to a quota excession during the actual writing of the file.
1873 PH/19 It seems that peer DN values may contain newlines (and other non-printing
1874 characters?) which causes problems in log lines. The DN values are now
1875 passed through string_printing() before being added to log lines.
1877 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
1878 and InterBase are left for another time.)
1880 PH/21 Added message_body_newlines option.
1882 PH/22 Guard against possible overflow in moan_check_errorcopy().
1884 PH/23 POSIX allows open() to be a macro; guard against that.
1886 PH/24 If the recipient of an error message contained an @ in the local part
1887 (suitably quoted, of course), incorrect values were put in $domain and
1888 $local_part during the evaluation of errors_copy.
1894 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
1895 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
1898 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
1899 issue a MAIL command.
1901 PH/02 In an ACL statement such as
1903 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
1905 if a client was not listed at all, or was listed with a value other than
1906 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
1907 the condition was not true (as it should be), so access was not denied.
1908 The bug was that the ! inversion was incorrectly passed on to the second
1909 item. This has been fixed.
1911 PH/03 Added additional dnslists conditions == and =& which are different from
1912 = and & when the dns lookup returns more than one IP address.
1914 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
1915 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
1917 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
1918 FSYNC, which compiles an option called disable_fsync that allows for
1919 bypassing fsync(). The documentation is heavily laced with warnings.
1921 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
1923 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
1924 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
1925 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
1926 including adding "make clean"; (3) Added -fPIC when compiling the test
1927 dynamically loaded module, to get rid of a warning.
1929 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
1930 message fails, move_frozen_messages = true and ignore_bounce_errors_after
1931 = 0s. The bug is otherwise harmless.
1933 PH/07 There was a bug in the dovecot authenticator such that the value of
1934 $auth1 could be overwritten, and so not correctly preserved, after a
1935 successful authentication. This usually meant that the value preserved by
1936 the server_setid option was incorrect.
1938 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
1940 PH/09 Installed PCRE release 7.0.
1942 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
1943 run for batched SMTP input. It is now run at the start of every message
1944 in the batch. While fixing this I discovered that the process information
1945 (output by running exiwhat) was not always getting set for -bs and -bS
1946 input. This is fixed, and it now also says "batched" for BSMTP.
1948 PH/11 Added control=no_pipelining.
1950 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
1951 patch, slightly modified), and move the expansion of helo_data till after
1952 the connection is made in the smtp transport (so it can use these
1955 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
1957 PH/14 Added log_selector = +pid.
1959 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
1961 PH/16 Add ${if forany and ${if forall.
1963 PH/17 Added dsn_from option to vary the From: line in DSNs.
1965 PH/18 Flush SMTP output before performing a callout, unless control =
1966 no_callout_flush is set.
1968 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
1969 was true (the default) a successful delivery failed to delete the retry
1970 item, thus causing premature timeout of the address. The bug is now
1973 PH/20 Added hosts_avoid_pipelining to the smtp transport.
1975 PH/21 Long custom messages for fakedefer and fakereject are now split up
1976 into multiline reponses in the same way that messages for "deny" and
1977 other ACL rejections are.
1979 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
1980 with slight modification.
1982 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
1983 draft, changing the syntax and factoring some duplicate code".
1985 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
1986 for deliveries of the second and subsequent messages over the same SMTP
1989 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
1990 ${reduce, with only minor "tidies".
1992 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
1994 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
1995 expansion side effects.
1997 PH/27 When a message times out after an over-quota error from an Exim-imposed
1998 quota, the bounce message says "mailbox is full". This message was not
1999 being given when it was a system quota that was exceeded. It now should
2002 MH/03 Made $recipients available in local_scan(). local_scan() already has
2003 better access to the recipient list through recipients_list[], but
2004 $recipients can be useful in postmaster-provided expansion strings.
2006 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
2007 in the case of a MAIL command with additional options following the
2008 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
2009 were accidentally chopped off.
2011 PH/29 SMTP synchronization checks are implemented when a command is read -
2012 there is a check that no more input is waiting when there shouldn't be
2013 any. However, for some commands, a delay in an ACL can mean that it is
2014 some time before the response is written. In this time, more input might
2015 arrive, invalidly. So now there are extra checks after an ACL has run for
2016 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
2017 pipelining has not been advertised.
2019 PH/30 MH's patch to allow iscntrl() characters to be list separators.
2021 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
2022 returned in the SMTP response when smtp_return_error_details was false.
2023 This has been fixed.
2025 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
2026 instead of the C I/O that was originally supplied, because problems were
2027 reported on Solaris.
2029 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
2030 Exim which did not show up earlier: it was assuming that a call to
2031 SSL_CTX_set_info_callback() might give an error value. In fact, there is
2032 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
2033 was a macro that became an assignment, so it seemed to work. This has
2034 changed to a proper function call with a void return, hence the compile
2035 error. Exim's code has been fixed.
2037 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
2040 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
2042 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
2044 PH/37 If a message is not accepted after it has had an id assigned (e.g.
2045 because it turns out to be too big or there is a timeout) there is no
2046 "Completed" line in the log. When some messages of this type were
2047 selected by exigrep, they were listed as "not completed". Others were
2048 picked up by some special patterns. I have improved the selection
2049 criteria to be more general.
2051 PH/38 The host_find_failed option in the manualroute router can now be set
2052 to "ignore", to completely ignore a host whose IP address cannot be
2053 found. If all hosts are ignored, the behaviour is controlled by the new
2054 host_all_ignored option.
2056 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
2057 homing or because of multiple MX records with /mx) generated more than
2058 one IP address, and the following item turned out to be the local host,
2059 all the secondary addresses of the first item were incorrectly removed
2060 from the list, along with the local host and any following hosts (which
2061 is what is supposed to happen).
2063 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
2064 whoever called Exim into the -H file. In the case of the daemon it was
2065 behaving confusingly. When first started, it used values for whoever
2066 started the daemon, but after a SIGHUP it used the Exim user (because it
2067 calls itself on a restart). I have changed the code so that it now always
2070 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
2071 message are rejected with the same error (e.g. no authentication or bad
2072 sender address), and a DATA command is nevertheless sent (as can happen
2073 with PIPELINING or a stupid MUA), the error message that was given to the
2074 RCPT commands is included in the rejection of the DATA command. This is
2075 intended to be helpful for MUAs that show only the final error to their
2078 PH/42 Another patch from the Sieve maintainer.
2080 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
2081 Eximstats - Fixed some broken HTML links and added missing column headers
2083 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
2084 columns for Rejects, Temp Rejects, Ham, and Spam rows.
2086 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
2088 PH/43 Yet another patch from the Sieve maintainer.
2090 PH/44 I found a way to check for a TCP/IP connection going away before sending
2091 the response to the final '.' that terminates a message, but only in the
2092 case where the client has not sent further data following the '.'
2093 (unfortunately, this is allowed). However, in many cases there won't be
2094 any further data because there won't be any more messages to send. A call
2095 to select() can be used: if it shows that the input is "ready", there is
2096 either input waiting, or the socket has been closed. An attempt to read
2097 the next input character can distinguish the two cases. Previously, Exim
2098 would have sent an OK response which the client would never have see.
2099 This could lead to message repetition. This fix should cure that, at
2100 least in a lot of common cases.
2102 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
2103 advertised in response to EHLO.
2109 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
2110 fixed by 4.65/MH/01 (is this a record?) are fixed:
2112 (i) An empty string was always treated as zero by the numeric comparison
2113 operators. This behaviour has been restored.
2115 (ii) It is documented that the numeric comparison operators always treat
2116 their arguments as decimal numbers. This was broken in that numbers
2117 starting with 0 were being interpreted as octal.
2119 While fixing these problems I realized that there was another issue that
2120 hadn't been noticed. Values of message_size_limit (both the global option
2121 and the transport option) were treated as octal if they started with 0.
2122 The documentation was vague. These values are now always treated as
2123 decimal, and I will make that clear in the documentation.
2129 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
2130 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
2133 MH/01 Don't check that the operands of numeric comparison operators are
2134 integers when their expansion is in "skipping" mode (fixes bug
2135 introduced by 4.64-PH/07).
2137 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
2138 child addresses, Exim now panics and dies. Previously, because the count
2139 is held in a short int, deliveries were likely to be lost. As such a
2140 large number of recipients for a single message is ridiculous
2141 (performance will be very, very poor), I have chosen to impose a limit
2142 rather than extend the field.
2148 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
2149 leftover -K file (the existence of which was triggered by #402).
2150 While we were at it, introduced process PID as part of the -K
2151 filename. This should rule out race conditions when creating
2154 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
2155 processing considerably. Previous code took too long for large mails,
2156 triggering a timeout which in turn triggers #401.
2158 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
2159 in the DK code in transports.c. sendfile() is not really portable,
2160 hence the _LINUX specificness.
2162 TF/01 In the add_headers option to the mail command in an Exim filter,
2163 there was a bug that Exim would claim a syntax error in any
2164 header after the first one which had an odd number of characters
2167 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
2168 callout verification, Exim cached a "reject" for the entire domain. This
2169 is correct for most verifications, but it is not correct for a recipient
2170 verification with use_sender or use_postmaster set, because in that case
2171 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
2172 case of MAIL FROM:<> rejection from other early rejections (e.g.
2173 rejection of HELO). When verifying a recipient using a non-null MAIL
2174 address, the cache is ignored if it shows MAIL FROM:<> rejection.
2175 Whatever the result of the callout, the value of the domain cache is
2176 left unchanged (for any other kind of callout, getting as far as trying
2177 RCPT means that the domain itself is ok).
2179 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
2182 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
2183 manifest itself as EPIPE rather than ECONNECT. When tidying away a
2184 session, the daemon ignores ECONNECT errors and logs others; it now
2185 ignores EPIPE as well.
2187 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
2188 (quoted-printable decoding).
2190 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
2191 later the small subsequent patch to fix an introduced bug.
2193 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
2195 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
2197 PH/08 An error is now given if message_size_limit is specified negative.
2199 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
2200 to be given (somewhat) arbitrary names.
2202 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
2205 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
2206 miscellaneous code fixes
2208 PH/10 Added the log_reject_target ACL modifier to specify where to log
2211 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
2212 hostname. This is wrong, because it relates to the incoming message (and
2213 probably the interface on which it is arriving) and not to the outgoing
2214 callout (which could be using a different interface). This has been
2215 changed to use the value of the helo_data option from the smtp transport
2216 instead - this is what is used when a message is actually being sent. If
2217 there is no remote transport (possible with a router that sets up host
2218 addresses), $smtp_active_hostname is used.
2220 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
2221 tweaks were necessary in order to get it to work (see also 21 below):
2222 (a) The code assumed that strncpy() returns a negative number on buffer
2223 overflow, which isn't the case. Replaced with Exim's string_format()
2225 (b) There were several signed/unsigned issues. I just did the minimum
2226 hacking in of casts. There is scope for a larger refactoring.
2227 (c) The code used strcasecmp() which is not a standard C function.
2228 Replaced with Exim's strcmpic() function.
2229 (d) The code set only $1; it now sets $auth1 as well.
2230 (e) A simple test gave the error "authentication client didn't specify
2231 service in request". It would seem that Dovecot has changed its
2232 interface. Fortunately there's a specification; I followed it and
2233 changed what the client sends and it appears to be working now.
2235 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
2238 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
2239 address A is aliased to B and C, where B exists and C does not. Without
2240 -v the output is "A verified" because verification stops after a
2241 successful redirection if more than one address is generated. However,
2242 with -v the child addresses are also verified. Exim was outputting "A
2243 failed to verify" and then showing the successful verification for C,
2244 with its parentage. It now outputs "B failed to verify", showing B's
2245 parentage before showing the successful verification of C.
2247 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
2248 look up a TXT record in a specific list after matching in a combined
2251 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
2252 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
2253 they consult the DNS. I had assumed they would set it the way they
2254 wanted; and indeed my experiments on Linux seem to show that in some
2255 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
2256 To be on the safe side, however, I have now made the interface to
2257 host_find_byname() similar to host_find_bydns(), with an argument
2258 containing the DNS resolver options. The host_find_byname() function now
2259 sets these options at its start, just as host_find_bydns() does. The smtp
2260 transport options dns_qualify_single and dns_search_parents are passed to
2261 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
2262 of host_find_byname() use the default settings of RES_DEFNAMES
2263 (qualify_single) but not RES_DNSRCH (search_parents).
2265 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
2266 spool_read_header() do less string testing, by means of a preliminary
2267 switch on the second character of optional "-foo" lines. (This is
2268 overdue, caused by the large number of possibilities that now exist.
2269 Originally there were few.) While I was there, I also converted the
2270 str(n)cmp tests so they don't re-test the leading "-" and the first
2271 character, in the hope this might squeeze out yet more improvement.
2273 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
2274 flag allowing group syntax was set by the header_syntax check but not
2275 turned off, possible causing trouble later; (2) The flag was not being
2276 set at all for the header_verify test, causing "group"-style headers to
2277 be rejected. I have now set it in this case, and also caused header_
2278 verify to ignore an empty address taken from a group. While doing this, I
2279 came across some other cases where the code for allowing group syntax
2280 while scanning a header line wasn't quite right (mostly, not resetting
2281 the flag correctly in the right place). These bugs could have caused
2282 trouble for malformed header lines. I hope it is now all correct.
2284 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
2285 with the "reply" argument non-NULL. The code, however (which originally
2286 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
2287 but it didn't always do it. This confused somebody who was copying the
2288 code for some other use. I have removed all the tests.
2290 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
2291 feature that was used to support insecure browsers during the U.S. crypto
2292 embargo. It requires special client support, and Exim is probably the
2293 only MTA that supported it -- and would never use it because real RSA is
2294 always available. This code has been removed, because it had the bad
2295 effect of slowing Exim down by computing (never used) parameters for the
2296 RSA_EXPORT functionality.
2298 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
2299 authenticator to fail if there's a tab character in the incoming data
2300 (there should never be unless someone is messing about, as it's supposed
2301 to be base64-encoded). Also added, on Timo's advice, the "secured" option
2302 if the connection is using TLS or if the remote IP is the same as the
2303 local IP, and the "valid-client-cert option" if a client certificate has
2306 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
2307 authenticators. This can be used for authorization after authentication
2308 succeeds. (In the case of plaintext, it servers for both authentication
2311 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
2312 if any retry times were supplied.
2314 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
2315 connection, where there is no client IP address to check. In this
2316 situation, the verify now always succeeds.
2318 PH/25 Applied John Jetmore's -Mset patch.
2320 PH/26 Added -bem to be like -Mset, but loading a message from a file.
2322 PH/27 In a string expansion for a processed (not raw) header when multiple
2323 headers of the same name were present, leading whitespace was being
2324 removed from all of them, but trailing whitespace was being removed only
2325 from the last one. Now trailing whitespace is removed from each header
2326 before concatenation. Completely empty headers in a concatenation (as
2327 before) are ignored.
2329 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
2330 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
2332 PH/29 [Removed. This was a change that I later backed out, and forgot to
2333 correct the ChangeLog entry (that I had efficiently created) before
2334 committing the later change.]
2336 PH/30 Exim was sometimes attempting to deliver messages that had suffered
2337 address errors (4xx response to RCPT) over the same connection as other
2338 messages routed to the same hosts. Such deliveries are always "forced",
2339 so retry times are not inspected. This resulted in far too many retries
2340 for the affected addresses. The effect occurred only when there were more
2341 hosts than the hosts_max_try setting in the smtp transport when it had
2342 the 4xx errors. Those hosts that it had tried were not added to the list
2343 of hosts for which the message was waiting, so if all were tried, there
2344 was no problem. Two fixes have been applied:
2346 (i) If there are any address or message errors in an SMTP delivery, none
2347 of the hosts (tried or untried) are now added to the list of hosts
2348 for which the message is waiting, so the message should not be a
2349 candidate for sending over the same connection that was used for a
2350 successful delivery of some other message. This seems entirely
2351 reasonable: after all the message is NOT "waiting for some host".
2352 This is so "obvious" that I'm not sure why it wasn't done
2353 previously. Hope I haven't missed anything, but it can't do any
2354 harm, as the worst effect is to miss an optimization.
2356 (ii) If, despite (i), such a delivery is accidentally attempted, the
2357 routing retry time is respected, so at least it doesn't keep
2358 hammering the server.
2360 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
2361 in ${readsocket because some servers need this prod.
2363 PH/32 Added some extra debug output when updating a wait-xxx database.
2365 PH/33 The hint "could be header name not terminated by colon", which has been
2366 given for certain expansion errors for a long time, was not being given
2367 for the ${if def:h_colon_omitted{... case.
2369 PH/34 The spec says: "With one important exception, whenever a domain list is
2370 being scanned, $domain contains the subject domain." There was at least
2371 one case where this was not true.
2373 PH/35 The error "getsockname() failed: connection reset by peer" was being
2374 written to the panic log as well as the main log, but it isn't really
2375 panic-worthy as it just means the connection died rather early on. I have
2376 removed the panic log writing for the ECONNRESET error when getsockname()
2379 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
2380 runs only) independently of the message's sender address. This meant
2381 that, if the 4xx error was in fact related to the sender, a different
2382 message to the same recipient with a different sender could confuse
2383 things. In particualar, this can happen when sending to a greylisting
2384 server, but other circumstances could also provoke similar problems.
2385 I have changed the default so that the retry time for these errors is now
2386 based a combination of the sender and recipient addresses. This change
2387 can be overridden by setting address_retry_include_sender=false in the
2390 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
2391 remote server are returned as part of bounce messages. This was not
2392 happening for LMTP over a pipe (the lmtp transport), but now it is the
2393 same for both kinds of LMTP.
2395 PH/38 Despite being documented as not happening, Exim was rewriting addresses
2396 in header lines that were in fact CNAMEs. This is no longer the case.
2398 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
2399 and queue runs started by the daemon processed all messages. This has
2400 been fixed so that -R and -S can now usefully be given with -q<time>.
2402 PH/40 Import PCRE release 6.7 (fixes some bugs).
2404 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
2406 PH/42 Give an error if -q is specified more than once.
2408 PH/43 Renamed the variables $interface_address and $interface_port as
2409 $received_ip_address and $received_port, to make it clear that these
2410 values apply to message reception, and not to the outgoing interface when
2411 a message is delivered. (The old names remain recognized, of course.)
2413 PH/44 There was no timeout on the connect() call when using a Unix domain
2414 socket in the ${readsocket expansion. There now is.
2416 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
2417 be meaningful with "accept".
2419 SC/01 Eximstats V1.43
2420 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
2422 SC/02 Eximstats V1.44
2423 Use a glob alias rather than an array ref in the generated
2424 parser. This improves both readability and performance.
2426 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
2427 Collect SpamAssassin and rejection statistics.
2428 Don't display local sender or destination tables unless
2429 there is data to show.
2430 Added average volumes into the top table text output.
2432 SC/04 Eximstats V1.46
2433 Collect data on the number of addresses (recipients)
2434 as well as the number of messages.
2436 SC/05 Eximstats V1.47
2437 Added 'Message too big' to the list of mail rejection
2438 reasons (thanks to Marco Gaiarin).
2440 SC/06 Eximstats V1.48
2441 Mainlog lines which have GMT offsets and are too short to
2442 have a flag are now skipped.
2444 SC/07 Eximstats V1.49 (Alain Williams)
2445 Added the -emptyok flag.
2447 SC/08 Eximstats V1.50
2448 Fixes for obtaining the IP address from reject messages.
2450 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
2451 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
2452 whitesspace changes from 4.64-PH/27
2454 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
2457 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
2458 are found, allow negative numbers in numeric criteria)
2460 JJ/06 exipick.20061117.2, added new $message_body_missing variable
2462 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
2463 to match changes made in 4.64-PH/43
2465 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
2467 PH/47 Put in an explicit test for a DNS lookup of an address record where the
2468 "domain" is actually an IP address, and force a failure. This locks out
2469 those revolvers/nameservers that support "A-for-A" lookups, in
2470 contravention of the specifications.
2472 PH/48 When a host name was looked up from an IP address, and the subsequent
2473 forward lookup of the name timed out, the host name was left in
2474 $sender_host_name, contrary to the specification.
2476 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
2477 restricted to single-key lookups, Exim was not diagnosing an error if
2478 * or *@ was used with a query-style lookup.
2480 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
2482 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
2483 long ago, but noone interested enough thought of it. Let's just say that
2484 the "1.1" means that there are some new functions that weren't there at
2485 some point in the past.
2487 PH/51 Error processing for expansion failure of helo_data from an smtp
2488 transport during callout processing was broken.
2490 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
2491 tested/used via the -bh/-bhc/-bs options.
2493 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
2494 bug, fixed in subsequent PCRE releases).
2496 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
2497 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
2499 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
2505 SC/01 Use a glob alias rather than an array ref in eximstats generated
2506 parser. This improves both readability and performance.
2508 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
2509 Don't display local sender or destination tables in eximstats unless
2510 there is data to show.
2511 Added average volumes into the eximstats top table text output.
2513 SC/03 Collect data on the number of addresses (recipients) as well
2514 as the number of messages in eximstats.
2516 TF/01 Correct an error in the documentation for the redirect router. Exim
2517 does (usually) call initgroups() when daemonizing.
2519 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
2520 with consistent privilege compared to when running as a daemon.
2522 TF/03 Note in the spec that $authenticated_id is not set for local
2523 submissions from trusted users.
2525 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
2526 Thanks to Dean Brooks <dean@iglou.com> for the patch.
2528 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
2529 by adding some example configuration directives to the default
2530 configuration file. A little bit of work is required to uncomment the
2531 directives and define how usernames and passwords are checked, but
2532 there is now a framework to start from.
2534 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
2535 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
2536 without this. I don't know how relevant this is to other LDAP libraries.
2538 PH/02 Add the verb name to the "unknown ACL verb" error.
2540 PH/03 Magnus Holmgren's patch for filter_prepend_home.
2542 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
2544 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
2545 directory not expanded when it should be if an expanded home directory
2546 was set for the address (which is overridden by the transport).
2548 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
2551 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
2552 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
2553 because it is too late at that time, and has no effect.
2555 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
2556 security issue with \' (bugzilla #107). I could not use the
2557 PQescapeStringConn() function, because it needs a PGconn value as one of
2560 PH/08 When testing addresses using -bt, indicate those final addresses that
2561 are duplicates that would not cause an additional delivery. At least one
2562 person was confused, thinking that -bt output corresponded to deliveries.
2563 (Suppressing duplicates isn't a good idea as you lose the information
2564 about possibly different redirections that led to the duplicates.)
2566 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
2567 systems where poll() doesn't work, in particular OS X.
2569 PH/10 Added more information to debugging output for retry time not reached.
2571 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
2572 operations in malware.c.
2574 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
2577 PH/13 If write_rejectlog was set false when logging was sent to syslog with
2578 syslog_duplication set false, log lines that would normally be written
2579 both the the main log and to the reject log were not written to syslog at
2582 PH/14 In the default configuration, change the use of "message" in ACL warn
2583 statements to "add_header".
2585 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
2586 not followed by a command (e.g. "seen endif").
2588 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
2589 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
2592 PH/17 Added extra conditions to the default value of delay_warning_condition
2596 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
2597 { match{$h_precedence:}{(?i)bulk|list|junk} } \
2598 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
2601 The Auto-Submitted: and various List- headers are standardised, whereas I
2602 don't think Precedence: ever was.
2604 PH/18 Refactored debugging code in route_finduser() to show more information,
2605 in particular, the error code if getpwnam() issues one.
2607 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
2608 This is apparently needed in addition to the PH/07 change above to avoid
2609 any possible encoding problems.
2611 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
2612 but not after initializing Perl.
2614 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
2615 output them only if debugging. By default they are written stderr,
2616 apparently, which is not desirable.
2618 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
2621 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
2624 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
2626 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
2627 authenticated or an ident call has been made. Suppress the default
2628 values for $authenticated_id and $authenticated_sender (but permit -oMai
2629 and -oMas) when testing with -bh.
2631 PH/24 Re-jigged the order of the tests in the default configuration so that the
2632 tests for valid domains and recipients precede the DNS black list and CSA
2633 tests, on the grounds that those ones are more expensive.
2635 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
2636 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
2637 This bug exists in every version of Exim that I still have, right back to
2640 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
2641 However, an attempt to turn on case-sensitivity in a regex key by
2642 including (?-i) didn't work because the subject string was already
2643 lowercased, and the effects were non-intuitive. It turns out that a
2644 one-line patch can be used to allow (?-i) to work as expected.
2650 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
2651 other effects) broke the use of negated acl sub-conditions.
2653 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
2656 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
2657 "Deny" causes Exim to reject the incoming connection with a 554 error.
2658 Unfortunately, if there is a major crisis, such as a disk failure,
2659 tcp-wrappers gives "deny", whereas what one would like would be some
2660 kind of temporary error. A kludge has been added to help with this.
2661 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
2662 554 error is used if errno is still zero or contains ENOENT (which occurs
2663 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
2666 PH/03 Add -lutil to the default FreeBSD LIBS setting.
2668 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
2669 errors. Otherwise a message that provokes a temporary error (when other
2670 messages do not) can cause a whole host to time out.
2672 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
2673 addresses were routed directly to files or pipes from a redirect router.
2674 File deliveries just didn't batch; pipe deliveries might have suffered
2677 PH/06 A failure to get a lock for a hints database would erroneously always say
2678 "Failed to get write lock", even when it was really a read lock.
2680 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
2681 of 0600. This has been changed to use the value of the lockfile_mode
2682 option (which defaults to 0600).
2684 PH/08 Applied small patch from the Sieve maintainer.
2686 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
2687 folder from quota calculations, a direct delivery into this folder messed
2688 up the contents of the maildirsize file. This was because the regex was
2689 used only to exclude .Trash (or whatever) when the size of the mailbox
2690 was calculated. There was no check that a delivery was happening into an
2691 excluded directory. This bug has been fixed by ignoring all quota
2692 processing for deliveries into excluded directories.
2694 PH/10 Added the maildirfolder_create_regex option to appendfile.
2700 PH/01 The code for finding all the local interface addresses on a FreeBSD
2701 system running IPv6 was broken. This may well have applied to all BSD
2702 systems, as well as to others that have similar system calls. The broken
2703 code found IPv4 interfaces correctly, but gave incorrect values for the
2704 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
2705 that it would not match correctly against @[] and not recognize the IPv6
2708 PH/02 The ipliteral router was not recognizing addresses of the form user@
2709 [ipv6:....] because it didn't know about the "ipv6:" prefix.
2711 PH/03 Added disable_ipv6.
2713 PH/04 Changed $reply_address to use the raw form of the headers instead of the
2714 decoded form, because it is most often used to construct To: headers
2715 lines in autoreplies, and the decoded form may well be syntactically
2716 invalid. However, $reply_address has leading white space removed, and all
2717 newlines turned into spaces so that the autoreply transport does not
2720 PH/05 If group was specified without a user on a router, and no group or user
2721 was specified on a transport, the group from the router was ignored.
2723 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
2724 for visible compile-time settings that can be used to change these
2725 numbers, for those that want even more. Backwards compatibility with old
2726 spool files has been maintained. However, going back to a previous Exim
2727 release will lost any variables that are in spool files.
2729 PH/07 Two small changes when running in the test harness: increase delay when
2730 passing a TCP/IP connection to a new process, in case the original
2731 process has to generate a bounce, and remove special handling of
2732 127.0.0.2 (sic), which is no longer necessary.
2734 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
2735 be the same on different OS.
2737 PH/09 Moved a debug statement in filter processing to avoid a race problem when
2740 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
2741 whether --show-vars was specified or not
2743 JJ/02 exipick: Added support for new ACL variable spool format introduced
2746 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
2747 syntactically invalid From: or Reply-to: line, and a filter used this to
2748 generate an autoreply, and therefore failed to obtain an address for the
2749 autoreply, Exim could try to deliver to a non-existent relative file
2750 name, causing unrelated and misleading errors. What now happens is that
2751 it logs this as a hard delivery error, but does not attempt to create a
2754 PH/11 The exinext utility has a -C option for testing purposes, but although
2755 the given file was scanned by exinext itself; it wasn't being passed on
2756 when Exim was called.
2758 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
2759 an end-of-file indication when reading a command response.
2761 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
2762 compiled. In many other places in Exim, IPv6 addresses are always
2763 recognized, so I have changed this. It also means that IPv4 domain
2764 literals of the form [IPV4:n.n.n.n] are now always recognized.
2766 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
2767 used if the router is not running as root, for example, when verifying at
2768 ACL time, or when using -bh. The debugging output from this situation was
2769 non-existent - all you got was a failure to exec. I have made two
2772 (a) Failures to set uid/gid, the current directory, or a process leader
2773 in a subprocess such as that created by queryprogram now generate
2774 suitable debugging ouput when -d is set.
2776 (b) The queryprogram router detects when it is not running as root,
2777 outputs suitable debugging information if -d is set, and then runs
2778 the subprocess without attempting to change uid/gid.
2780 PH/15 Minor change to Makefile for building test_host (undocumented testing
2783 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
2784 additional section of a DNS packet that returns MX or SRV records.
2785 Instead, it always explicitly searches for A/AAAA records. This avoids
2786 major problems that occur when a DNS server includes only records of one
2787 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
2788 fixed another bug: if SRV records were looked up and the corresponding
2789 address records were *not* found in the additional section, the port
2790 values from the SRV records were lost.
2792 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
2793 using the correct key (the original address) when searching the retry
2794 rules in order to find which one to use for generating the retry hint.
2796 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
2797 adding the default one. Similarly, if it contains a Reply-To: header, the
2798 errors_reply_to option, if set, is not used.
2800 PH/19 When calculating a retry time, Exim used to measure the "time since
2801 failure" by looking at the "first failed" field in the retry record. Now
2802 it does not use this if it is later than than the arrival time of the
2803 message. Instead it uses the arrival time. This makes for better
2804 behaviour in cases where some deliveries succeed, thus re-setting the
2805 "first failed" field. An example is a quota failure for a huge message
2806 when small messages continue to be delivered. Without this change, the
2807 "time since failure" will always be short, possible causing more frequent
2808 delivery attempts for the huge message than are intended.
2809 [Note: This change was subsequently modified - see PH/04 for 4.62.]
2811 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
2812 $1, $2, $3) because the numerical variables can be reset during some
2813 expansion items (e.g. "match"), thereby losing the authentication data.
2815 PH/21 Make -bV show the size of off_t variables so that the test suite can
2816 decide whether to run tests for quotas > 2G.
2818 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
2819 mailbox_size, and mailbox_filecount in the appendfile transport. If a
2820 filecount value is greater than 2G or if a quota value is greater than 2G
2821 on a system where the size of off_t is not greater than 4, a panic error
2824 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
2825 never match. The debug and -bh output now contains an explicit error
2826 message indicating a malformed IPv4 address or mask.
2828 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
2829 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
2830 PH/23 above applies.
2832 PH/25 Do not write to syslog when running in the test harness. The only
2833 occasion when this arises is a failure to open the main or panic logs
2834 (for which there is an explicit test).
2836 PH/26 Added the /no_tell option to "control=freeze".
2838 PH/27 If a host name lookup failed very early in a connection, for example, if
2839 the IP address matched host_lookup and the reverse lookup yielded a name
2840 that did not have a forward lookup, an error message of the form "no IP
2841 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
2842 could be logged. Now it outputs the IP address instead of "NULL".
2844 PH/28 An enabling patch from MH: add new function child_open_exim2() which
2845 allows the sender and the authenticated sender to be set when
2846 submitting a message from within Exim. Since child_open_exim() is
2847 documented for local_scan(), the new function should be too.
2849 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
2850 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
2851 results in an empty string is now treated as unset.
2853 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
2855 PH/31 Added sender_verify_fail logging option.
2857 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
2858 needed by RFC 822 but not by RFC 2822 was commented out. I have now
2859 tidied the source and removed it altogether.
2861 PH/33 When a queue run was abandoned because the load average was too high, a
2862 log line was always written; now it is written only if the queue_run log
2863 selector is set. In addition, the log line for abandonment now contains
2864 information about the queue run such as the pid. This is always present
2865 in "start" and "stop" lines but was omitted from the "abandon" line.
2867 PH/34 Omit spaces between a header name and the colon in the error message that
2868 is given when verify = headers_syntax fails (if there are lots of them,
2869 the message gets confusing).
2871 PH/35 Change the default for dns_check_names_pattern to allow slashes within
2872 names, as there are now some PTR records that contain slashes. This check
2873 is only to protect against broken name servers that fall over on strange
2874 characters, so the fact that it applies to all lookups doesn't matter.
2876 PH/36 Now that the new test suite is complete, we can remove some of the
2877 special code in Exim that was needed for the old test suite. For example,
2878 sorting DNS records because real resolvers return them in an arbitrary
2879 order. The new test suite's fake resolver always returns records in the
2882 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
2883 bounces) except when queue_only is set, to avoid logging races between
2884 the different processes.
2886 PH/38 Panic-die if .include specifies a non-absolute path.
2888 PH/39 A tweak to the "H" retry rule from its user.
2890 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
2891 a label. They prevented compilation on older perls.
2893 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
2894 a warning to be raised on newish perls.
2896 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
2897 on queue. Changes to match documented behaviour of showing count of
2898 messages matching specified criteria.
2900 PH/40 Changed the default ident timeout from 30s to 5s.
2902 PH/41 Added support for the use of login_cap features, on those BSD systems
2903 that have them, for controlling the resources used by pipe deliveries.
2905 PH/42 The content-scanning code uses fopen() to create files in which to put
2906 message data. Previously it was not paying any attention to the mode of
2907 the files. Exim runs with umask(0) because the rest of the code creates
2908 files with open(), and sets the required mode explicitly. Thus, these
2909 files were ending up world-writeable. This was not a big issue, because,
2910 being within the spool directory, they were not world-accessible. I have
2911 created a function called modefopen, which takes an additional mode
2912 argument. It sets umask(777), creates the file, chmods it to the required
2913 mode, then resets the umask. All the relevant calls to fopen() in the
2914 content scanning code have been changed to use this function.
2916 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
2917 to 24 hours. This avoids potential overflow problems when processing G
2918 and H retry rules. I suspect nobody ever tinkers with this value.
2920 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
2922 PH/45 When the plaintext authenticator is running as a client, the server's
2923 challenges are checked to ensure they are valid base64 strings. By
2924 default, the authentication attempt is cancelled if an invalid string is
2925 received. Setting client_ignore_invalid_base64 true ignores these errors.
2926 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
2927 they are received. Thus, the responses can be made to depend on the
2928 challenges. If an invalid string is ignored, an empty string is placed in
2931 PH/46 Messages that are created by the autoreply transport now contains a
2932 References: header, in accordance with RFCs 2822 and 3834.
2934 PH/47 Added authenticated_sender_force to the smtp transport.
2936 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
2938 PH/49 Installed latest patch from the Sieve maintainer.
2940 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
2941 was also set, the appendfile transport was unnecessarily scanning a
2942 directory of message files (e.g. for maildir delivery) to find the count
2943 of files (along with the size), even though it did not need this
2944 information. It now does the scan only if it needs to find either the
2945 size of the count of files.
2947 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
2949 PH/52 Two bugs concerned with error handling when the smtp transport is
2952 (i) Exim was not creating retry information for temporary errors given
2953 for individual recipients after the DATA command when the smtp transport
2954 was used in LMTP mode. This meant that they could be retried too
2955 frequently, and not timed out correctly.
2957 (ii) Exim was setting the flag that allows error details to be returned
2958 for LMTP errors on RCPT commands, but not for LMTP errors for individual
2959 recipients that were returned after the DATA command.
2961 PH/53 This is related to PH/52, but is more general: for any failing address,
2962 when detailed error information was permitted to be returned to the
2963 sender, but the error was temporary, then after the final timeout, only
2964 "retry timeout exceeded" was returned. Now it returns the full error as
2965 well as "retry timeout exceeded".
2967 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
2968 do this, and (what is worse) MTAs that accept it.
2970 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
2971 will now be deprecated.
2973 PH/56 New os.c-cygwin from the Cygwin maintainer.
2975 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
2976 formats (previously only available in exim formats via -bpr, -bpru,
2977 and -bpra. Now also available in native and exiqgrep formats)
2979 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
2980 with very large, slow to parse queues
2982 JJ/08 exipick: added ! as generic prefix to negate any criteria format
2984 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
2986 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
2987 responses to authentication challenges, though it was showing the
2988 challenges; (ii) I've removed the CR characters from the debug output for
2991 PH/58 Allow for the insertion of a newline as well as a space when a string
2992 is turned into more than one encoded-word during RFC 2047 encoding. The
2993 Sieve code now uses this.
2995 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
2996 data_4xx, lost_connection, tls_required.
2998 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
2999 message was being sent as an SMTP response.
3001 PH/61 Add -l and -k options to exicyclog.
3003 PH/62 When verifying, if an address was redirected to one new address, so that
3004 verification continued, and the new address failed or deferred after
3005 having set something in $address_data, the value of $address_data was not
3006 passed back to the ACL. This was different to the case when no
3007 redirection occurred. The value is now passed back in both cases.
3009 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
3010 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
3011 use login_cap.h, so on its own it isn't the distinguishing feature. The
3012 new name refers directly to the setclassresources() function.
3014 PH/65 Added configuration files for NetBSD3.
3016 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
3018 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
3019 is preferred over IPv4.
3021 PH/68 The bounce_return_message and bounce_return_body options were not being
3022 honoured for bounces generated during the reception of non-SMTP messages.
3023 In particular, this applied to messages rejected by the ACL. This bug has
3024 been fixed. However, if bounce_return_message is true and bounce_return_
3025 body is false, the headers that are returned for a non-SMTP message
3026 include only those that have been read before the error was detected.
3027 (In the case of an ACL rejection, they have all been read.)
3029 PH/69 The HTML version of the specification is now built in a directory called
3030 spec_html instead of spec.html, because the latter looks like a path with
3031 a MIME-type, and this confuses some software.
3033 PH/70 Catch two compiler warnings in sieve.c.
3035 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
3036 function verify_get_ident() calls ip_connect() to connect a socket, but
3037 if the "connect()" function timed out, ip_connect() used to close the
3038 socket. However, verify_get_ident() also closes the socket later, and in
3039 between Exim writes to the log, which may get opened at this point. When
3040 the socket was closed in ip_connect(), the log could get the same file
3041 descriptor number as the socket. This naturally causes chaos. The fix is
3042 not to close the socket in ip_connect(); the socket should be closed by
3043 the function that creates it. There was only one place in the code where
3044 this was missing, in the iplookup router, which I don't think anybody now
3045 uses, but I've fixed it anyway.
3047 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
3048 well as to direct DNS lookups. Otherwise the handling of names in host
3049 lists is inconsistent and therefore confusing.
3055 PH/01 Two changes to the default runtime configuration:
3057 (1) Move the checks for relay_from_hosts and authenticated clients from
3058 after to before the (commented out) DNS black list checks.
3060 (2) Add control=submission to the relay_from_hosts and authenticated
3061 clients checks, on the grounds that messages accepted by these
3062 statements are most likely to be submissions.
3064 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
3066 (1) Generate an error if the third argument for the ${prvs expansion is
3069 (2) Treat a missing third argument of ${prvscheck as if it were an empty
3072 (3) Reset the variables that are obtained from the first argument of
3073 ${prvscheck and used in the second argument before leaving the code,
3074 because their memory is reclaimed, so using them afterwards may do
3077 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
3078 one (it's much easier than Tom thought :-).
3080 (5) Because of (4), we can now allow for the use of $prvscheck_result
3081 inside the third argument.
3083 PH/03 For some reason, the default setting of PATH when running a command from
3084 a pipe transport was just "/usr/bin". I have changed it to
3087 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
3088 anything to be listed in the output from -bV.
3090 PH/05 When a filter generated an autoreply, the entire To: header line was
3091 quoted in the delivery log line, like this:
3093 => >A.N.Other <ano@some.domain> <original@ddress> ...
3095 This has been changed so that it extracts the operative address. There
3096 may be more than one such address. If so, they are comma-separated, like
3099 => >ano@some.domain,ona@other.domain <original@ddress> ...
3101 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
3102 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
3103 not being looked up in the rDNS to get a host name, Exim was showing the
3104 IP address twice in Received: lines, even though the IP addresses were
3105 identical. For example:
3107 Received: from [1.2.3.4] (helo=[1.2.3.4])
3109 However, if the real host name was known, it was omitting the HELO data
3110 if it matched the actual IP address. This has been tidied up so that it
3111 doesn't show the same IP address twice.
3113 PH/07 When both +timestamp and +memory debugging was on, the value given by
3114 $tod_xxx expansions could be wrong, because the tod_stamp() function was
3115 called by the debug printing, thereby overwriting the timestamp buffer.
3116 Debugging no longer uses the tod_stamp() function when +timestamp is set.
3118 PH/08 When the original message was included in an autoreply transport, it
3119 always said "this is a copy of the message, including all the headers",
3120 even if body_only or headers_only was set. It now gives an appropriate
3123 PH/09 Applied a patch from the Sieve maintainer which:
3125 o fixes some comments
3126 o adds the (disabled) notify extension core
3127 o adds some debug output for the result of if/elsif tests
3128 o points to the current vacation draft in the documentation
3129 and documents the missing references header update
3133 o fixes a bug in processing the envelope test (when testing
3134 multiple envelope elements, the last element determinted the
3137 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
3138 Electronic Mail") by including:
3140 Auto-submitted: auto-generated
3142 in the messages that it generates (bounce messages and others, such as
3143 warnings). In the case of bounce messages for non-SMTP mesages, there was
3144 also a typo: it was using "Auto_submitted" (underscore instead of
3145 hyphen). Since every message generated by Exim is necessarily in response
3146 to another message, thes have all been changed to:
3148 Auto-Submitted: auto-replied
3150 in accordance with these statements in the RFC:
3152 The auto-replied keyword:
3154 - SHOULD be used on messages sent in direct response to another
3155 message by an automatic process,
3157 - MUST NOT be used on manually-generated messages,
3159 - MAY be used on Delivery Status Notifications (DSNs) and Message
3160 Disposition Notifications (MDNs),
3162 - MUST NOT be used on messages generated by automatic or periodic
3163 processes, except for messages which are automatic responses to
3166 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
3167 to the default Received: header definition.
3169 PH/12 Added log selector acl_warn_skipped (default on).
3171 PH/13 After a successful wildlsearch lookup, discard the values of numeric
3172 variables because (a) they are in the wrong storage pool and (b) even if
3173 they were copied, it wouldn't work properly because of the caching.
3175 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
3176 checking when decoding. Apparently there are clients that generate
3177 overlong encoded strings. Why am I not surprised?
3179 PH/15 If the first argument of "${if match_address" was not empty, but did not
3180 contain an "@" character, Exim crashed. Now it writes a panic log message
3181 and treats the condition as false.
3183 PH/16 In autoreply, treat an empty string for "once" the same as unset.
3185 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
3186 extension "envelope-auth". The code is finished and in agreement with
3187 other implementations, but there is no documentation so far and in fact,
3188 nobody wrote the draft yet. This extension is currently #undef'ed, thus
3189 not changing the active code.
3191 Print executed "if" and "elsif" statements when debugging is used. This
3192 helps a great deal to understand what a filter does.
3194 Document more things not specified clearly in RFC3028. I had all this
3195 sorted out, when out of a sudden new issues came to my mind. Oops."
3197 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
3200 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
3201 canonical form (as documented). However, after a host name lookup from
3202 the IP address, check_host() was doing a simple string comparison with
3203 addresses acquired from the DNS when checking that the found name did
3204 have the original IP as one of its addresses. Since any found IPv6
3205 addresses are likely to be in abbreviated form, the comparison could
3206 fail. Luckily, there already exists a function for doing the comparison
3207 by converting both addresses to binary, so now that is used instead of
3208 the text comparison.
3210 PH/20 There was another similar case to PH/19, when a complete host name was
3211 given in a host list; looking up its IP address could give an abbreviated
3212 form, whereas the current host's name might or might not be abbreviated.
3213 The same fix has been applied.
3219 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
3220 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
3223 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
3224 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
3226 PH/03 Typo: missing ".o" in src/pcre/Makefile.
3228 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
3229 header line, restrict the check to what is listed in RFCs 2369 and 2929.
3230 Also, for "Auto-Submitted", treat anything other than "no" as
3231 non-personal, in accordance with RFC 3834. (Previously it treated
3232 anything starting "auto-" as non-personal.)
3234 TF/01 The control=submission/name=... option had a problem with syntax
3235 errors if the name included a slash character. The /name= option
3236 now slurps the rest of the string, so it can include any characters
3237 but it must come last in the list of options (after /sender_retain
3240 PH/05 Some modifications to the interface to the fake nameserver for the new
3248 TK/01 Added the "success_on_redirect" address verification option. See
3249 NewStuff for rationale and an example.
3251 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
3253 PH/02 Patch to exigrep to allow it to work on syslog lines.
3255 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
3256 fread() to skip over the body file's header line, because in Cygwin the
3257 header line is locked and is inaccessible.
3259 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
3260 co-exist for some time) to make it clear that it is the Exim ID that is
3261 referenced, not the Message-ID: header line.
3263 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
3264 string_format() function, because snprintf() does not exist on all
3267 PH/06 The use of forbid_filter_existstest now also locks out the use of the
3268 ${stat: expansion item.
3270 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
3271 protocol synchronization error", to keep the pedants happy.
3273 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
3274 well as for IRIX systems, when gcc is being used. See the host.c source
3277 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
3279 PH/10 Named domain lists were not working if used in a queue_smtp_domains
3282 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
3283 transport and to the smtp transport in LMTP mode.
3285 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
3287 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
3288 run a filter in a subprocess. This could lead to confusion in subsequent
3289 lookups in the parent process. There should also be a search_tidyup() at
3290 the end of the subprocess.
3292 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
3293 only if the host matched helo_try_verify_hosts, which caused the
3294 verification to occur when the EHLO/HELO command was issued. The ACL just
3295 tested the remembered result. Now, if a previous verification attempt has
3296 not happened, "verify = helo" does it there and then.
3298 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
3300 TK/03 Fix log output including CR from clamd.
3302 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
3303 exist provoked a memory error which could cause a segfault.
3305 PH/15 Installed PCRE 6.2
3307 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
3309 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
3310 of the problem. Specifically, suggested +O2 rather than +O1 for the
3313 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
3315 PH/20 If a delivery was routed to a non-standard port by means of an SRV
3316 record, the port was not correctly logged when the outgoing_port log
3317 selector was set (it logged the transort's default port).
3319 PH/21 Added support for host-specific ports to manualroute, queryprogram,
3320 fallback_hosts, and "hosts" in the smtp transport.
3322 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
3323 host errors such as "Connection refused".
3325 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
3326 authentication with radiusclient 0.4.9:
3328 - Error returned from rc_read_config was caught wrongly
3329 - Username/password not passed on to radius server due to wrong length.
3331 The presumption is that some radiusclient API changes for 4.51/PH/17
3332 were not taken care of correctly. The code is still untested by me (my
3333 Linux distribution still has 0.3.2 of radiusclient), but it was
3334 contributed by a Radius user.
3336 PH/24 When doing a callout, the value of $domain wasn't set correctly when
3337 expanding the "port" option of the smtp transport.
3339 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
3340 while reading a MIME header. Thanks to Tom Hughes for a patch.
3342 PH/24 Include config.h inside local_scan.h so that configuration settings are
3345 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
3346 that in an ACL for RCPT (for example), you can examine exactly what was
3349 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
3350 commands, but it was not correctly comparing the address with the actual
3351 client host address. Thus, it would show the EHLO address in Received:
3352 header lines when this was not necessary.
3354 PH/27 Added the % operator to ${eval:}.
3356 PH/28 Exim tries to create and chdir to its spool directory when it starts;
3357 it should be ignoring failures (because with -C, for example, it has lost
3358 privilege). It wasn't ignoring creation failures other than "already
3361 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
3364 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
3365 because an input line was too long, either on its own, or by virtue of
3366 too many continuations, the temporary file was not being removed, and the
3367 return code was incorrect.
3369 PH/31 Missing "BOOL" in function definition in filtertest.c.
3371 PH/32 Applied Sieve patches from the maintainer.
3373 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
3375 PH/33 Added "verify = not_blind".
3377 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
3378 Local/Makefile (with some defaults set). These are used in built scripts
3379 such as exicyclog, but they have never been used in the exim_install
3380 script (though there are many overriding facilities there). I have
3381 arranged that the exim_install script now takes note of these two
3384 PH/35 Installed configuration files for Dragonfly.
3386 PH/36 When a locally submitted message by a trusted user did not contain a
3387 From: header, and the sender address was obtained from -f or from an SMTP
3388 MAIL command, and the trusted user did not use -F to supply a sender
3389 name, $originator_name was incorrectly used when constructing a From:
3390 header. Furthermore, $originator_name was used for submission mode
3391 messages from external hosts without From: headers in a similar way,
3392 which is clearly wrong.
3394 PH/37 Added control=suppress_local_fixups.
3396 PH/38 When log_selector = +received_sender was set, and the addition of the
3397 sender made the log line's construction buffer exactly full, or one byte
3398 less than full, an overflow happened when the terminating "\n" was
3401 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
3402 when the result of a list match is failure because a DNS lookup failed.
3404 PH/40 RM_COMMAND is now used in the building process.
3406 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
3407 the "build-* directories that it finds.
3409 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
3410 address in a domain literal was a prefix of an interface address.
3412 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
3413 when verifying a sender address, unless rewrite_headers is false.
3415 PH/44 Wrote a long comment about why errors_to addresses are verified as
3416 recipients, not senders.
3418 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
3419 the ratelimit ACL was added.
3421 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
3423 PH/46 Added extra information about PostgreSQL errors to the error string.
3425 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
3426 avoiding the need to install special zones in a real server. This is
3427 backwards compatible; if it can't find the fake resolver, it drops back.
3428 Thus, both old and new test suites can be run.
3430 TF/02 Added util/ratelimit.pl
3432 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
3433 clock is set back in time.
3435 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
3436 Candler <B.Candler@pobox.com>.
3438 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
3439 OK for addresses that are the result of redirections.
3441 PH/48 A number of further additions for the benefit of the new test suite,
3442 including a fake gethostbyname() that interfaces to the fake DNS resolver
3445 TF/06 The fix for widen_domains has also been applied to qualify_single and
3446 search_parents which are the other dnslookup options that can cause
3449 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
3452 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
3454 TF/07 Exim produced the error message "an SRV record indicated no SMTP
3455 service" if it encountered an MX record with an empty target hostname.
3456 The message is now "an MX or SRV record indicated no SMTP service".
3458 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
3459 if the DNS of the sending site is misconfigured. This is quite a
3460 common situation. This change restores the behaviour of treating a
3461 helo verification defer as a failure.
3463 PH/51 If self=fail was set on a router, the bounce message did not include the
3464 actual error message.
3470 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
3472 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
3473 command itself ended in error, the underlying message about the transport
3474 filter timeout was being overwritten with the pipe command error. Now the
3475 underlying error message should be appended to the second error message.
3477 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
3479 PH/02 Reduce the amount of output that "make" produces by default. Full output
3480 can still be requested.
3482 PH/03 The warning log line about a condition test deferring for a "warn" verb
3483 was being output only once per connection, rather than after each
3484 occurrence (because it was using the same function as for successful
3485 "warn" verbs). This seems wrong, so I have changed it.
3487 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
3488 it should not have, which might have caused a crash in the right
3489 circumstances, but probably never did.
3491 PH/04 Installed a modified version of Tony Finch's patch to make submission
3492 mode fix the return path as well as the Sender: header line, and to
3493 add a /name= option so that you can make the user's friendly name appear
3496 TF/03 Added the control = fakedefer ACL modifier.
3498 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
3499 Mark Lowes for thorough testing.
3501 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
3503 TK/03 Merged latest SRS patch from Miles Wilton.
3505 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
3506 with the definition in sysexits.h (which is #included earlier).
3507 Fortunately, Exim does not actually use EX_OK. The code used to try to
3508 preserve the sysexits.h value, by assumimg that macro definitions were
3509 scanned for macro replacements. I have been disabused of this notion,
3510 so now the code just undefines EX_OK before #including unistd.h.
3512 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
3513 in the smtp transport. When a block could not be written in a single
3514 write() function, the timeout was being re-applied to each part-write.
3515 This seems wrong - if the receiver was accepting one byte at a time it
3516 would take for ever. The timeout is now adjusted when this happens. It
3517 doesn't have to be particularly precise.
3519 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
3520 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
3522 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
3523 without a domain if the check to <postmaster@domain> fails.
3525 SC/01 Eximstats: added -xls and the ability to specify output files
3526 (patch written by Frank Heydlauf).
3528 SC/02 Eximstats: use FileHandles for outputing results.
3530 SC/03 Eximstats: allow any combination of xls, txt, and html output.
3532 SC/04 Eximstats: fixed display of large numbers with -nvr option
3534 SC/05 Eximstats: fixed merging of reports with empty tables.
3536 SC/06 Eximstats: added the -include_original_destination flag
3538 SC/07 Eximstats: removed tabs and trailing whitespace.
3540 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
3542 TK/06 MBOX spool code: Add real "From " MBOX separator line
3543 so the .eml file is really in mbox format (even though
3544 most programs do not really care). Patch from Alex Miller.
3546 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
3547 The latter is generated from $received_to and is only set if the
3548 message has one envelope recipient. SA can use these headers,
3549 obviously out-of-the-box. Patch from Alex Miller.
3551 PH/08 The ${def test on a variable was returning false if the variable's
3552 value was "0", contrary to what the specification has always said!
3553 The result should be true unless the variable is empty.
3555 PH/09 The syntax error of a character other than { following "${if
3556 def:variable_name" (after optional whitespace) was not being diagnosed.
3557 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
3558 accidental colon was present, for example, could give incorrect results.
3560 PH/10 Tidied the code in a number of places where the st_size field of a stat()
3561 result is used (not including appendfile, where other changes are about
3564 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
3565 This involved changing a lot of size variables from int to off_t. It
3566 should work with maildirs and everything.
3568 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
3569 spamd dying while we are connected to it.
3571 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
3574 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
3575 function for BDB 4.3.
3577 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
3579 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
3580 the sources for PCRE have been re-arranged and more files are now
3583 PH/15 The code I had for printing potentially long long variables in PH/11
3584 above was not the best (it lost precision). The length of off_t variables
3585 is now inspected at build time, and an appropriate printing format (%ld
3586 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
3587 to be "long long int" or "long int". This is needed for the internal
3588 formatting function string_vformat().
3590 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
3591 the configuration file to be ":syslog", then the script "guesses" where
3592 the logs files are, rather than using the compiled in default. In our
3593 case the guess is not the same as the compiled default, so the script
3594 suddenly stopped working when I started to use syslog. The patch checks
3595 to see if log_file_path is "". If so, it attempts to read it from exim
3596 with no configuration file to get the compiled in version, before it
3597 falls back to the previous guessing code."
3599 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
3600 implementing BATV in an Exim configuration. See NewStuff for the gory
3603 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
3604 Makefile that are specific to HP-UX.
3606 PH/18 If the "use_postmaster" option was set for a recipient callout together
3607 with the "random" option, the postmaster address was used as the MAIL
3608 FROM address for the random test, but not for the subsequent recipient
3609 test. It is now used for both.
3611 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
3612 patch removes a few documentation additions to RFC 3028, because the
3613 latest draft now contains them. It adds the new en;ascii-case comparator
3614 and a new error check for 8bit text in MIME parts. Comparator and
3615 require names are now matched exactly. I enabled the subaddress
3616 extension, but it is not well tested yet (read: it works for me)."
3618 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
3619 rework some of the code of TK/09 above to avoid the hardwired use of
3620 "%lld" and "long long". Replaced the call to snprintf() with a call to
3623 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
3624 records point to non-existent hosts", "retry timeout exceeded", and
3625 "retry time not reached for any host after a long failure period".
3627 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
3628 experimental DomainKeys support:
3630 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
3631 (2) On an error such as an illegally used "control", the wrong name for
3632 the control was given.
3634 These problems did NOT occur unless DomainKeys support was compiled.
3636 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
3638 PH/24 Added ${if match_ip condition.
3640 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
3641 databases so that it will be absolutely obvious if a crash occurs in the
3642 DB library. This is a regular occurrence (often caused by mis-matched
3645 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
3646 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
3647 picked up on a user's system that detects such things. There doesn't seem
3648 to be a gcc warning option for this - only an attribute that has to be
3649 put on the function's prototype. It seems that in Fedora Core 4 they have
3650 set this on a number of new functions. No doubt there will be more in due
3653 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
3654 specify a transport. However, if an address that was verified by such a
3655 router was the subject of a callout, Exim crashed because it tried to
3656 read the rcpt_include_affixes from the non-existent transport. Now it
3657 just assumes that the setting of that option is false. This bug was
3658 introduced by 4.51/PH/31.
3660 PH/28 Changed -d+all to exclude +memory, because that information is very
3661 rarely of interest, but it makes the output a lot bigger. People tend to
3662 do -d+all out of habit.
3664 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
3665 code in os-type was giving problems when libc.so lives in lib64, like on
3668 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
3669 aren't the modern standard, and it seems that some systems' include files
3670 don't always have them. Exim was already checking for some of the newer
3671 ones like T_AAAA, and defining it itself. I've added checks for all the
3672 record types that Exim uses.
3674 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
3675 not automatically generating a new one, as it is supposed to. This
3676 prevented TLS from working. If the file did exist, but contained invalid
3677 data, a new version was generated, as expected. It was only the case of a
3678 non-existent file that was broken.
3680 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
3681 with a change in libdomainkeys > 0.64.
3683 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
3684 from DNS. If the selector record carries the flag, it now has
3685 precedence over the domain-wide flag.
3687 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
3689 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
3690 the use of a port name that isn't defined in /etc/services) occurred, the
3691 message was deferred as in a normal delivery, and thus remained on the
3692 spool, instead of being failed because of the mua_wrapper setting. This
3693 is now fixed, and I tidied up some of the mua_wrapper messages at the
3696 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
3697 the messages in a hash of arrays rather than using individual hashes.
3698 This is a bit cleaner and results in dramatic memory savings, albeit
3699 at a slight CPU cost.
3701 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
3702 as requested by Marc Sherman.
3704 SC/10 Eximstats: added histograms for user specified patterns as requested
3707 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
3709 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
3710 fopen() in the content-scanning modules that did not already have it.
3716 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
3717 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
3719 TK/02 Fix ACL "control" statement not being available in MIME ACL.
3721 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
3723 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
3724 to test Sieve filters that use "vacation".
3726 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
3727 that changes the way the GnuTLS parameters are stored in the cache file.
3728 The new format can be generated externally. For backward compatibility,
3729 if the data in the cache doesn't make sense, Exim assumes it has read an
3730 old-format file, and it generates new data and writes a new file. This
3731 means that you can't go back to an older release without removing the
3734 PH/03 A redirect router that has both "unseen" and "one_time" set does not
3735 work if there are any delivery delays because "one_time" forces the
3736 parent to be marked "delivered", so its unseen clone is never tried
3737 again. For this reason, Exim now forbids the simultaneous setting of
3740 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
3741 redirected to themselves ("homonym" addresses). Read the long ChangeLog
3742 entry if you want to know the details. The fix, however, neglected to
3743 consider the case when local delivery batching is involved. The test for
3744 "previously delivered" was not happening when checking to see if an
3745 address could be batched with a previous (undelivered) one; under
3746 certain circumstances this could lead to multiple deliveries to the same
3749 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
3750 in its include files, and this causes problems building Exim.
3752 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
3753 header_syntax) but Exim was just ignoring anything given after a slash.
3754 In particular, this caused confusion with an attempt to use "verify =
3755 reverse_host_lookup/defer_ok". An error is now given when options are
3756 supplied for verify items that do not have them. (Maybe reverse_host_
3757 lookup should have a defer_ok option, but that's a different point.)
3759 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
3760 defined by RFC 821) to 2048, because there were problems with some AUTH
3761 commands, and RFC 1869 says the size should be increased for extended
3762 SMTP commands that take arguments.
3764 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
3767 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
3768 "unknown" error; now it says that the functionality isn't in the binary.
3770 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
3771 an address' error message when a string expansion fails (syntax or
3772 whatever). Otherwise the password may appear in the log. Following change
3773 PH/42 below, there is no longer a chance of it appearing in a bounce
3776 PH/11 Installed exipick version 20050225.0 from John Jetmore.
3778 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
3779 of its addresses was ever tried. (Bugzilla bug #2.)
3781 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
3782 the result incorrectly in the debug output. (It correctly added a newline
3783 to what was transported.)
3785 TF/01 Added $received_time.
3787 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
3788 commented out examples of how to interface to a virus scanner and to
3789 SpamAssassin. Also added commented examples of av_scanner and
3790 spamd_address settings.
3792 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
3793 and controls are allowed in which ACLs. There were a couple of minor
3794 errors. Some of the entries in the conditions table (which is a table of
3795 where they are NOT allowed) were getting very unwieldy; rewrote them as a
3796 negation of where the condition IS allowed.
3798 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
3800 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
3801 header file does not have a version number, so I've had to invent a new
3802 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
3803 API. The code is untested by me (my Linux distribution still has 0.3.2 of
3804 radiusclient), but it was contributed by a Radius user.
3806 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
3807 files or directories, for OpenSSL.
3809 PH/19 When an Exim process that is running as root has to create an Exim log
3810 file, it does so in a subprocess that runs as exim:exim so as to get the
3811 ownership right at creation (otherwise, other Exim processes might see
3812 the file with the wrong ownership). There was no test for failure of this
3813 fork() call, which would lead to the process getting stuck as it waited
3814 for a non-existent subprocess. Forks do occasionally fail when resources
3815 run out. I reviewed all the other calls to fork(); they all seem to check
3818 PH/20 When checking for unexpected SMTP input at connect time (before writing
3819 the banner), Exim was not dealing correctly with a non-positive return
3820 from the read() function. If the client had disconnected by this time,
3821 the result was a log entry for a synchronization error with an empty
3822 string after "input=" when read() returned zero. If read() returned -1
3823 (an event I could not check), uninitialized data bytes were printed.
3824 There were reports of junk text (parts of files, etc) appearing after
3827 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
3829 PH/22 Added support for macro redefinition, and (re)definition in between
3830 driver and ACL definitions.
3832 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
3833 forgetting to use the resulting value; it was using the unexpanded value.
3835 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
3836 hadn't been configured. The fix is from Juergen Kreileder, who
3837 understands it better than I do:
3839 "Here's what I see happening with three configured cyrus_sasl
3840 authenticators configured (plain, login, cram-md5):
3842 On startup auth_cyrus_sasl_init() gets called for each of these.
3843 This means three calls to sasl_listmech() without a specified mech_list.
3844 => SASL tests which mechs of all available mechs actually work
3845 => three warnings about OTP not working
3846 => the returned list contains: plain, login, cram-md5, digest-md5, ...
3848 With the patch, sasl_listmech() also gets called three times. But now
3849 SASL's mech_list option is set to the server_mech specified in the the
3850 authenticator. Or in other words, the answer from sasl_listmech()
3851 gets limited to just the mech you're testing for (which is different
3853 => the return list contains just 'plain' or 'login', 'cram-md5' or
3854 nothing depending on the value of ob->server_mech.
3856 I've just tested the patch: Authentication still works fine,
3857 unavailable mechs specified in the exim configuration are still
3858 caught, and the auth.log warnings about OTP are gone."
3860 PH/25 When debugging is enabled, the contents of the command line are added
3861 to the debugging output, even when log_selector=+arguments is not
3864 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
3865 answer is "GNU", and only if the return is "GNU/something" is the answer
3868 PH/27 $acl_verify_message is now set immediately after the failure of a
3869 verification in an ACL, and so is available in subsequent modifiers. In
3870 particular, the message can be preserved by coding like this:
3872 warn !verify = sender
3873 set acl_m0 = $acl_verify_message
3875 Previously, $acl_verify_message was set only while expanding "message"
3876 and "log_message" when a very denied access.
3878 PH/28 Modified OS/os.c-Linux with
3880 -#ifndef OS_LOAD_AVERAGE
3881 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
3883 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
3884 nomenclature these days.)
3886 PH/29 Installed patch from the Sieve maintainer that adds the options
3887 sieve_useraddress and sieve_subaddress to the redirect router.
3889 PH/30 In these circumstances:
3890 . Two addresses routed to the same list of hosts;
3891 . First host does not offer TLS;
3892 . First host accepts first address;
3893 . First host gives temporary error to second address;
3894 . Second host offers TLS and a TLS session is established;
3895 . Second host accepts second address.
3896 Exim incorrectly logged both deliveries with the TLS parameters (cipher
3897 and peerdn, if requested) that were in fact used only for the second
3900 PH/31 When doing a callout as part of verifying an address, Exim was not paying
3901 attention to any local part prefix or suffix that was matched by the
3902 router that accepted the address. It now behaves in the same way as it
3903 does for delivery: the affixes are removed from the local part unless
3904 rcpt_include_affixes is set on the transport.
3906 PH/32 Add the sender address, as F=<...>, to the log line when logging a
3907 timeout during the DATA phase of an incoming message.
3909 PH/33 Sieve envelope tests were broken for match types other than :is. I have
3910 applied a patch sanctioned by the Sieve maintainer.
3912 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
3913 the uid or gid is negative. A case of a negative gid caused this to be
3914 noticed. The fix allows for either to be negative.
3916 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
3917 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
3920 PH/36 The change PH/12 above was broken. Fixed it.
3922 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
3923 the grounds that routing the same address twice would always produce the
3924 same answer. This might have been true once, but it is certainly no
3925 longer true now. Routing a child address may depend on the previous
3926 routing that produced that child. Some complicated redirection strategies
3927 went wrong when messages had multiple recipients, and made Exim's
3928 behaviour dependent on the order in which the addresses were given.
3930 I have moved the duplicate checking until after the routing is complete.
3931 Exim scans the addresses that are assigned to local and remote
3932 transports, and removes any duplicates. This means that more work will be
3933 done, as duplicates will always all be routed, but duplicates are
3934 presumably rare, so I don't expect this is of any significance.
3936 For deliveries to pipes, files, and autoreplies, the duplicate checking
3937 still happens during the routing process, since they are not going to be
3940 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
3941 It corrects a timeout issue with spamd. This is Ian's comment: "The
3942 background is that sometimes spamd either never reads data from a
3943 connection it has accepted, or it never writes response data. The exiscan
3944 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
3945 blindly assumes that writes won't block so it may never time out."
3947 PH/39 Allow G after quota size as well as K and M.
3949 PH/40 The value set for $authenticated_id in an authenticator may not contain
3950 binary zeroes or newlines because the value is written to log lines and
3951 to spool files. There was no check on this. Now the value is run through
3952 the string_printing() function so that such characters are converted to
3953 printable escape sequences.
3955 PH/41 $message_linecount is a new variable that contains the total number of
3956 lines in the message. Compare $body_linecount, which is the count for the
3959 PH/42 Exim no longer gives details of delivery errors for specific addresses in
3960 bounce and delay warning messages, except in certain special cases, which
3963 (a) An SMTP error message from a remote host;
3964 (b) A message specified in a :fail: redirection;
3965 (c) A message specified in a "fail" command in a system filter;
3966 (d) A message specified in a FAIL return from the queryprogram router;
3967 (e) A message specified by the cannot_route_message router option.
3969 In these cases only, Exim does include the error details in bounce and
3970 warning messages. There are also a few cases where bland messages such
3971 as "unrouteable address" or "local delivery error" are given.
3973 PH/43 $value is now also set for the "else" part of a ${run expansion.
3975 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
3976 being worked on, but at least Exim now implements the latest version to
3979 PH/45 In a pipe transport, although a timeout while waiting for the pipe
3980 process to complete was treated as a delivery failure, a timeout while
3981 writing the message to the pipe was logged, but erroneously treated as a
3982 successful delivery. Such timeouts include transport filter timeouts. For
3983 consistency with the overall process timeout, these timeouts are now
3984 treated as errors, giving rise to delivery failures by default. However,
3985 there is now a new Boolean option for the pipe transport called
3986 timeout_defer, which, if set TRUE, converts the failures into defers for
3987 both kinds of timeout. A transport filter timeout is now identified in
3990 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
3991 systems where "make" and "gmake" are different, calling "gmake" at top
3992 level broke things. I've arranged for the value of $(MAKE) to be passed
3993 from the Makefile to this script so that it can call the same version of
3997 A note about Exim versions 4.44 and 4.50
3998 ----------------------------------------
4000 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
4001 changes of various kinds. As a consequence, a big documentation update was
4002 needed. This delayed the release for rather longer than seemed good, especially
4003 in the light of a couple of (minor) security issues. Therefore, the changes
4004 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
4005 release. So 4.44 and 4.50 are in effect two different branches that both start
4008 I have left the 4.50 change log unchanged; it contains all the changes since
4009 4.43. The change log for 4.44 is below; many of its items are identical to
4010 those for 4.50. This seems to be the most sensible way to preserve the
4011 historical information.
4017 1. Minor wording change to the doc/README.SIEVE file.
4019 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
4020 computation of the current number of files was incorrect.
4022 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4023 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4026 4. Give more explanation in the error message when the command for a transport
4027 filter fails to execute.
4029 5. There are several places where Exim runs a non-Exim command in a
4030 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4031 was being done only for the command run by the queryprogram router. It is
4032 now done for all such subprocesses. The other cases are: ${run, transport
4033 filters, and the commands run by the lmtp and pipe transports.
4035 6. Added CONFIGURE_GROUP build-time option.
4037 7. Some older OS have a limit of 256 on the maximum number of file
4038 descriptors. Exim was using setrlimit() to set 1000 as a large value
4039 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4040 systems. I've change it so that if it can't get 1000, it tries for 256.
4042 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4043 was an oversight, and furthermore, ever since the addition of extra
4044 controls (e.g. 4.43/32), the checks on when to allow different forms of
4045 "control" were broken. There should now be diagnostics for all cases when a
4046 control that does not make sense is encountered.
4048 9. Added the /retain_sender option to "control=submission".
4050 10. $recipients is now available in the predata ACL (oversight).
4052 11. Tidy the search cache before the fork to do a delivery from a message
4053 received from the command line. Otherwise the child will trigger a lookup
4054 failure and thereby defer the delivery if it tries to use (for example) a
4055 cached ldap connection that the parent has called unbind on.
4057 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4058 of $address_data from the recipient verification was clobbered by the
4059 sender verification.
4061 13. The value of address_data from a sender verification is now available in
4062 $sender_address_data in subsequent conditions in the ACL statement.
4064 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
4066 15. Added a new option "connect=<time>" to callout options, to set a different
4069 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4070 was its contents. (It was OK if the option was not defined at all.)
4072 17. A "Completed" log line is now written for messages that are removed from
4073 the spool by the -Mrm option.
4075 18. New variables $sender_verify_failure and $recipient_verify_failure contain
4076 information about exactly what failed.
4078 19. Added -dd to debug only the daemon process.
4080 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
4081 handles timeouts, both on the server side and network timeouts. Renamed the
4082 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
4084 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
4085 It is now set to "smtps".
4087 22. $host_address is now set to the target address during the checking of
4088 ignore_target_hosts.
4090 23. When checking ignore_target_hosts for an ipliteral router, no host name was
4091 being passed; this would have caused $sender_host_name to have been used if
4092 matching the list had actually called for a host name (not very likely,
4093 since this list is usually IP addresses). A host name is now passed as
4096 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4097 code that specifies a non-restarting handler (typically sigaction() in
4098 modern systems) in an attempt to fix a rare and obscure crash bug.
4100 25. Narrowed the window for a race in the daemon that could cause it to ignore
4101 SIGCHLD signals. This is not a major problem, because they are used only to
4102 wake it up if nothing else does.
4104 26. A malformed maildirsize file could cause Exim to calculate negative values
4105 for the mailbox size or file count. Odd effects could occur as a result.
4106 The maildirsize information is now recalculated if the size or filecount
4109 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4110 support for a long time. Removed HAVE_SYS_VFS_H.
4112 28. Installed the latest version of exipick from John Jetmore.
4114 29. In an address list, if the pattern was not a regular expression, an empty
4115 subject address (from a bounce message) matched only if the pattern was an
4116 empty string. Non-empty patterns were not even tested. This was the wrong
4117 because it is perfectly reasonable to use an empty address as part of a
4118 database query. An empty address is now tested by patterns that are
4119 lookups. However, all the other forms of pattern expect the subject to
4120 contain a local part and a domain, and therefore, for them, an empty
4121 address still always fails if the pattern is not itself empty.
4123 30. Exim went into a mad DNS loop when attempting to do a callout where the
4124 host was specified on an smtp transport, and looking it up yielded more
4125 than one IP address.
4127 31. Re-factored the code for checking spool and log partition space into a
4128 function that finds that data and another that does the check. The former
4129 is then used to implement four new variables: $spool_space, $log_space,
4130 $spool_inodes, and $log_inodes.
4132 32. The RFC2047 encoding function was originally intended for short strings
4133 such as real names; it was not keeping to the 75-character limit for
4134 encoded words that the RFC imposes. It now respects the limit, and
4135 generates multiple encoded words if necessary. To be on the safe side, I
4136 have increased the buffer size for the ${rfc2047: expansion operator from
4139 33. It is now permitted to omit both strings after an "if" condition; if the
4140 condition is true, the result is "true". As before, when the second string
4141 is omitted, a false condition yields an empty string. This makes it less
4142 cumbersome to write custom ACL and router conditions.
4144 34. Failure to deliver a bounce message always caused it to be frozen, even if
4145 there was an errors_to setting on the router. The errors_to setting is now
4148 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4149 canonical form (fully expanded) before being placed in
4150 $sender_host_address.
4152 36. The table in the code that translates DNS record types into text (T_A to
4153 "A" for instance) was missing entries for NS and CNAME. It is just possible
4154 that this could have caused confusion if both these types were looked up
4155 for the same domain, because the text type is used as part of Exim's
4156 per-process caching. But the chance of anyone hitting this buglet seems
4159 37. The dnsdb lookup has been extended in a number of ways.
4161 (1) There is a new type, "zns", which walks up the domain tree until it
4162 finds some nameserver records. It should be used with care.
4164 (2) There is a new type, "mxh", which is like "mx" except that it returns
4165 just the host names, not the priorities.
4167 (3) It is now possible to give a list of domains (or IP addresses) to be
4168 looked up. The behaviour when one of the lookups defers can be
4169 controlled by a keyword.
4171 (4) It is now possible to specify the separator character for use when
4172 multiple records are returned.
4174 38. The dnslists ACL condition has been extended: it is now possible to supply
4175 a list of IP addresses and/or domains to be looked up in a particular DNS
4178 39. Added log_selector=+queue_time_overall.
4180 40. When running the queue in the test harness, wait just a tad after forking a
4181 delivery process, to get repeatability of debugging output.
4183 41. Include certificate and key file names in error message when GnuTLS fails
4184 to set them up, because the GnuTLS error message doesn't include the name
4185 of the failing file when there is a problem reading it.
4187 42. Allow both -bf and -bF in the same test run.
4189 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
4191 44. The "Exiscan patch" is now merged into the mainline Exim source.
4193 45. Sometimes the final signoff response after QUIT could fail to get
4194 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4195 < 0 before doing a fflush(). This bug looks as though it goes back to the
4196 introduction of TLS in release 3.20, but "sometimes" must have been rare
4197 because the tests only now provoked it.
4199 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
4200 (this can affect the format of dates).
4202 47. exim_tidydb, when checking for the continued existence of a message for
4203 which it has found a message-specific retry record, was not finding
4204 messages that were in split spool directories. Consequently, it was
4205 deleting retry records that should have stayed in existence.
4207 48. Steve fixed some bugs in eximstats.
4209 49. The SPA authentication driver was not abandoning authentication and moving
4210 on to the next authenticator when an expansion was forced to fail,
4211 contradicting the general specification for all authenticators. Instead it
4212 was generating a temporary error. It now behaves as specified.
4214 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
4215 (the order specifies the preference for clients). The order is now AES256,
4216 AES128, 3DES, ARCFOUR128.
4218 51. Small patch to Sieve code - explicitly set From: when generating an
4221 52. Exim crashed if a remote delivery caused a very long error message to be
4222 recorded - for instance if somebody sent an entire SpamAssassin report back
4223 as a large number of 550 error lines. This bug was coincidentally fixed by
4224 increasing the size of one of Exim's internal buffers (big_buffer) that
4225 happened as part of the Exiscan merge. However, to be on the safe side, I
4226 have made the code more robust (and fixed the comments that describe what
4229 53. Now that there can be additional text after "Completed" in log lines (if
4230 the queue_time_overall log selector is set), a one-byte patch to exigrep
4231 was needed to allow it to recognize "Completed" as not the last thing in
4234 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
4235 patch that reportedly fixes this has been added. I am not expert enough to
4236 create a test for it. This is what the patch creator wrote:
4238 "I found a little strange behaviour of ldap code when working with
4239 Windows 2003 AD Domain, where users was placed in more than one
4240 Organization Units. When I tried to give exim partial DN, the exit code
4241 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
4242 But simultaneously result of request was absolutely normal ldap result,
4243 so I produce this patch..."
4245 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
4246 REFERENCE, so I have modified the code to exclude the patch when that macro
4249 55. Some experimental protocols are using DNS PTR records for new purposes. The
4250 keys for these records are domain names, not reversed IP addresses. The
4251 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4252 leaves it alone. Component reversal etc. now happens only for IP addresses.
4255 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
4257 57. Double the size of the debug message buffer (to 2048) so that more of very
4258 long debug lines gets shown.
4260 58. The exicyclog utility now does better if the number of log files to keep
4261 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
4263 59. Two changes related to the smtp_active_hostname option:
4265 (1) $smtp_active_hostname is now available as a variable.
4266 (2) The default for smtp_banner uses $smtp_active_hostname instead
4267 of $primary_hostname.
4269 60. The host_aton() function is supposed to be passed a string that is known
4270 to be a valid IP address. However, in the case of IPv6 addresses, it was
4271 not checking this. This is a hostage to fortune. Exim now panics and dies
4272 if the condition is not met. A case was found where this could be provoked
4273 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4274 components; fortuitously, this particular loophole had already been fixed
4275 by change 4.50/55 above.
4277 If there are any other similar loopholes, the new check in host_aton()
4278 itself should stop them being exploited. The report I received stated that
4279 data on the command line could provoke the exploit when Exim was running as
4280 exim, but did not say which command line option was involved. All I could
4281 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4282 running as the user.
4285 61. There was a buffer overflow vulnerability in the SPA authentication code
4286 (which came originally from the Samba project). I have added a test to the
4287 spa_base64_to_bits() function which I hope fixes it.
4290 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
4291 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
4293 63. The daemon start-up calls getloadavg() while still root for those OS that
4294 need the first call to be done as root, but it missed one case: when
4295 deliver_queue_load_max is set with deliver_drop_privilege. This is
4296 necessary for the benefit of the queue runner, because there is no re-exec
4297 when deliver_drop_privilege is set.
4299 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
4300 This has been fixed.
4302 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4303 was in use, was not putting the data itself into the right store pool;
4304 consequently, it could be overwritten for a subsequent message in the same
4305 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4308 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
4310 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
4311 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
4312 the function were treating the return as a boolean value, which happened to
4313 work because 0=false and not-0=true, but is not correct code.
4315 68. The host_aton() function was not handling scoped IPv6 addresses (those
4316 with, for example, "%eth0" on the end) correctly.
4318 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
4319 negated items (that is, ~something) in unsigned ints. Some compilers
4320 apparently mutter when there is no cast.
4322 70. If an address verification called from an ACL failed, and did not produce a
4323 user-specific message (i.e. there was only a "system" message), nothing was
4324 put in $acl_verify_message. In this situation, it now puts the system
4327 71. Change 4.23/11 added synchronization checking at the start of an SMTP
4328 session; change 4.31/43 added the unwanted input to the log line - except
4329 that it did not do this in the start of session case. It now does.
4331 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
4332 This is wrong and can cause the other end to generate a synchronization
4333 error if it is another Exim or anything else that does the synchronization
4334 check. A QUIT command is no longer sent after a timeout.
4336 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
4337 during host lookups.
4339 74. The defer_ok option of callout verification was not working if it was used
4340 when verifying addresses in header lines, that is, for this case:
4342 verify = header_sender/callout=defer_ok
4344 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
4345 those file descriptors could be used for SMTP connections. If anything
4346 wrote to stderr (the example that came up was "warn" in embedded Perl), it
4347 could be sent to the SMTP client, causing chaos. The daemon now opens
4348 stdin, stdout, and stderr to /dev/null when it puts itself into the
4351 76. Arrange for output from Perl's "warn" command to be written to Exim's main
4352 log by default. The user can override this with suitable Perl magic.
4354 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
4355 the log message when discard triggers, was not working for the DATA ACL or
4356 for the non-SMTP ACL.
4358 78. Error message wording change in sieve.c.
4360 79. If smtp_accept_max_per_host was set, the number of connections could be
4361 restricted to fewer than expected, because the daemon was trying to set up
4362 a new connection before checking whether the processes handling previous
4363 connections had finished. The check for completed processes is now done
4364 earlier. On busy systems, this bug wouldn't be noticed because something
4365 else would have woken the daemon, and it would have reaped the completed
4368 80. If a message was submitted locally by a user whose login name contained one
4369 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
4370 It caused a spool format error. I have fixed the spool reading code. A
4371 related problem was that the "from" clause in the Received: line became
4372 illegal because of the space(s). It is now covered by ${quote_local_part.
4374 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
4377 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
4379 83. Patch from Sieve maintainer to fix unterminated string problem in
4380 "vacation" handling.
4382 84. Some minor changes to the Linux configuration files to help with other
4383 OS variants using glibc.
4385 85. One more patch for Sieve to update vacation handling to latest spec.
4388 ----------------------------------------------------
4389 See the note above about the 4.44 and 4.50 releases.
4390 ----------------------------------------------------
4396 1. Change 4.43/35 introduced a bug that caused file counts to be
4397 incorrectly computed when quota_filecount was set in an appendfile
4400 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4401 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4404 3. Give more explanation in the error message when the command for a transport
4405 filter fails to execute.
4407 4. There are several places where Exim runs a non-Exim command in a
4408 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4409 was being done only for the command run by the queryprogram router. It is
4410 now done for all such subprocesses. The other cases are: ${run, transport
4411 filters, and the commands run by the lmtp and pipe transports.
4413 5. Some older OS have a limit of 256 on the maximum number of file
4414 descriptors. Exim was using setrlimit() to set 1000 as a large value
4415 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4416 systems. I've change it so that if it can't get 1000, it tries for 256.
4418 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4419 was an oversight, and furthermore, ever since the addition of extra
4420 controls (e.g. 4.43/32), the checks on when to allow different forms of
4421 "control" were broken. There should now be diagnostics for all cases when a
4422 control that does not make sense is encountered.
4424 7. $recipients is now available in the predata ACL (oversight).
4426 8. Tidy the search cache before the fork to do a delivery from a message
4427 received from the command line. Otherwise the child will trigger a lookup
4428 failure and thereby defer the delivery if it tries to use (for example) a
4429 cached ldap connection that the parent has called unbind on.
4431 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4432 of $address_data from the recipient verification was clobbered by the
4433 sender verification.
4435 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4436 was its contents. (It was OK if the option was not defined at all.)
4438 11. A "Completed" log line is now written for messages that are removed from
4439 the spool by the -Mrm option.
4441 12. $host_address is now set to the target address during the checking of
4442 ignore_target_hosts.
4444 13. When checking ignore_target_hosts for an ipliteral router, no host name was
4445 being passed; this would have caused $sender_host_name to have been used if
4446 matching the list had actually called for a host name (not very likely,
4447 since this list is usually IP addresses). A host name is now passed as
4450 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4451 code that specifies a non-restarting handler (typically sigaction() in
4452 modern systems) in an attempt to fix a rare and obscure crash bug.
4454 15. Narrowed the window for a race in the daemon that could cause it to ignore
4455 SIGCHLD signals. This is not a major problem, because they are used only to
4456 wake it up if nothing else does.
4458 16. A malformed maildirsize file could cause Exim to calculate negative values
4459 for the mailbox size or file count. Odd effects could occur as a result.
4460 The maildirsize information is now recalculated if the size or filecount
4463 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4464 support for a long time. Removed HAVE_SYS_VFS_H.
4466 18. Updated exipick to current release from John Jetmore.
4468 19. Allow an empty sender to be matched against a lookup in an address list.
4469 Previously the only cases considered were a regular expression, or an
4472 20. Exim went into a mad DNS lookup loop when doing a callout where the
4473 host was specified on the transport, if the DNS lookup yielded more than
4476 21. The RFC2047 encoding function was originally intended for short strings
4477 such as real names; it was not keeping to the 75-character limit for
4478 encoded words that the RFC imposes. It now respects the limit, and
4479 generates multiple encoded words if necessary. To be on the safe side, I
4480 have increased the buffer size for the ${rfc2047: expansion operator from
4483 22. Failure to deliver a bounce message always caused it to be frozen, even if
4484 there was an errors_to setting on the router. The errors_to setting is now
4487 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4488 canonical form (fully expanded) before being placed in
4489 $sender_host_address.
4491 24. Updated eximstats to version 1.33
4493 25. Include certificate and key file names in error message when GnuTLS fails
4494 to set them up, because the GnuTLS error message doesn't include the name
4495 of the failing file when there is a problem reading it.
4497 26. Expand error message when OpenSSL has problems setting up cert/key files.
4500 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
4501 (this can affect the format of dates).
4503 28. exim_tidydb, when checking for the continued existence of a message for
4504 which it has found a message-specific retry record, was not finding
4505 messages that were in split spool directories. Consequently, it was
4506 deleting retry records that should have stayed in existence.
4508 29. eximstats updated to version 1.35
4509 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
4510 1.35 - bugfix such that pie charts by volume are generated correctly
4512 30. The SPA authentication driver was not abandoning authentication and moving
4513 on to the next authenticator when an expansion was forced to fail,
4514 contradicting the general specification for all authenticators. Instead it
4515 was generating a temporary error. It now behaves as specified.
4517 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
4518 (the order specifies the preference for clients). The order is now AES256,
4519 AES128, 3DES, ARCFOUR128.
4521 31. Small patch to Sieve code - explicitly set From: when generating an
4524 32. Exim crashed if a remote delivery caused a very long error message to be
4525 recorded - for instance if somebody sent an entire SpamAssassin report back
4526 as a large number of 550 error lines. This bug was coincidentally fixed by
4527 increasing the size of one of Exim's internal buffers (big_buffer) that
4528 happened as part of the Exiscan merge. However, to be on the safe side, I
4529 have made the code more robust (and fixed the comments that describe what
4532 33. Some experimental protocols are using DNS PTR records for new purposes. The
4533 keys for these records are domain names, not reversed IP addresses. The
4534 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4535 leaves it alone. Component reversal etc. now happens only for IP addresses.
4538 34. The host_aton() function is supposed to be passed a string that is known
4539 to be a valid IP address. However, in the case of IPv6 addresses, it was
4540 not checking this. This is a hostage to fortune. Exim now panics and dies
4541 if the condition is not met. A case was found where this could be provoked
4542 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4543 components; fortuitously, this particular loophole had already been fixed
4544 by change 4.50/55 or 4.44/33 above.
4546 If there are any other similar loopholes, the new check in host_aton()
4547 itself should stop them being exploited. The report I received stated that
4548 data on the command line could provoke the exploit when Exim was running as
4549 exim, but did not say which command line option was involved. All I could
4550 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4551 running as the user.
4554 35. There was a buffer overflow vulnerability in the SPA authentication code
4555 (which came originally from the Samba project). I have added a test to the
4556 spa_base64_to_bits() function which I hope fixes it.
4559 36. The daemon start-up calls getloadavg() while still root for those OS that
4560 need the first call to be done as root, but it missed one case: when
4561 deliver_queue_load_max is set with deliver_drop_privilege. This is
4562 necessary for the benefit of the queue runner, because there is no re-exec
4563 when deliver_drop_privilege is set.
4565 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4566 was in use, was not putting the data itself into the right store pool;
4567 consequently, it could be overwritten for a subsequent message in the same
4568 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4571 38. Sometimes the final signoff response after QUIT could fail to get
4572 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4573 < 0 before doing a fflush(). This bug looks as though it goes back to the
4574 introduction of TLS in release 3.20, but "sometimes" must have been rare
4575 because the tests only now provoked it.
4581 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
4582 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
4583 Now it returns an integer. A number of calls were still expecting a T/F
4584 return. Fortuitously, in all cases, the tests worked in OK situations,
4585 which is the norm. However, things would have gone wrong on any write
4586 failures on the smtp file descriptor. This function is used when sending
4587 messages over SMTP and also when doing verify callouts.
4589 2. When Exim is called to do synchronous delivery of a locally submitted
4590 message (the -odf or -odi options), it no longer closes stderr before doing
4593 3. Implemented the mua_wrapper option.
4595 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
4597 5. Implemented the functions header_remove(), header_testname(),
4598 header_add_at_position(), and receive_remove_recipient(), and exported them
4601 6. If an ACL "warn" statement specified the addition of headers, Exim already
4602 inserted X-ACL-Warn: at the start if there was no header name. However, it
4603 was not making this test for the second and subsequent header lines if
4604 there were newlines in the string. This meant that an invalid header could
4605 be inserted if Exim was badly configured.
4607 7. Allow an ACL "warn" statement to add header lines at the start or after all
4608 the Received: headers, as well as at the end.
4610 8. Added the rcpt_4xx retry error code.
4612 9. Added postmaster_mailfrom=xxx to callout verification option.
4614 10. Added mailfrom=xxxx to the callout verification option, for verify=
4617 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
4618 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
4620 12. Inserted some casts to stop certain compilers warning when using pointer
4621 differences as field lengths or precisions in printf-type calls (mostly
4622 affecting debugging statements).
4624 13. Added optional readline() support for -be (dynamically loaded).
4626 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
4627 same clock tick as a message's arrival, so that its received time was the
4628 same as the "first fail" time on the retry record, and that message
4629 remained on the queue past the ultimate address timeout, every queue runner
4630 would try a delivery (because it was past the ultimate address timeout) but
4631 after another failure, the ultimate address timeout, which should have then
4632 bounced the address, did not kick in. This was a "< instead of <=" error;
4633 in most cases the first failure would have been in the next clock tick
4634 after the received time, and all would be well.
4636 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
4637 being recognized when the domain list was tested by the match_domain
4638 condition in an expansion string.
4640 16. Added the ${str2b64: operator.
4642 17. Exim was always calling setrlimit() to set a large limit for the number of
4643 processes, without checking whether the existing limit was already
4644 adequate. (It did check for the limit on file descriptors.) Furthermore,
4645 errors from getrlimit() and setrlimit() were being ignored. Now they are
4646 logged to the main and panic logs, but Exim does carry on, to try to do its
4647 job under whatever limits there are.
4649 18. Imported PCRE 5.0.
4651 19. Trivial typo in log message " temporarily refused connection" (the leading
4654 20. If the log selector return_path_on_delivery was set and an address was
4655 redirected to /dev/null, the delivery process crashed because it assumed
4656 that a return path would always be set for a "successful" delivery. In this
4657 case, the whole delivery is bypassed as an optimization, and therefore no
4660 21. Internal re-arrangement: the function for sending a challenge and reading
4661 a response while authentication was assuming a zero-terminated challenge
4662 string. It's now changed to take a pointer and a length, to allow for
4663 binary data in such strings.
4665 22. Added the cyrus_sasl authenticator (code supplied by MBM).
4667 23. Exim was not respecting finduser_retries when seeking the login of the
4668 uid under which it was called; it was always trying 10 times. (The default
4669 setting of finduser_retries is zero.) Also, it was sleeping after the final
4670 failure, which is pointless.
4672 24. Implemented tls_on_connect_ports.
4674 25. Implemented acl_smtp_predata.
4676 26. If the domain in control=submission is set empty, Exim assumes that the
4677 authenticated id is a complete email address when it generates From: or
4678 Sender: header lines.
4680 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
4681 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
4682 chown and chgrp in /bin and hostname in /usr/bin.
4684 28. Exim was keeping the "process log" file open after each use, just as it
4685 does for the main log. This opens the possibility of it remaining open for
4686 long periods when the USR1 signal hits a daemon. Occasional processlog
4687 errors were reported, that could have been caused by this. Anyway, it seems
4688 much more sensible not to leave this file open at all, so that is what now
4691 29. The long-running daemon process does not normally write to the log once it
4692 has entered its main loop, and it closes the log before doing so. This is
4693 so that log files can straightforwardly be renamed and moved. However,
4694 there are a couple of unusual error situations where the daemon does write
4695 log entries, and I had neglected to close the log afterwards.
4697 30. The text of an SMTP error response that was received during a remote
4698 delivery was being truncated at 512 bytes. This is too short for some of
4699 the long messages that one sometimes sees. I've increased the limit to
4702 31. It is now possible to make retry rules that apply only when a message has a
4703 specific sender, in particular, an empty sender.
4705 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
4706 it possible to be selective about when SMTP synchronization is enforced.
4708 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
4710 32. Implemented hosts_connection_nolog.
4712 33. Added an ACL for QUIT.
4714 34. Setting "delay_warning=" to disable warnings was not working; it gave a
4717 35. Added mailbox_size and mailbox_filecount to appendfile.
4719 36. Added control = no_multiline_responses to ACLs.
4721 37. There was a bug in the logic of the code that waits for the clock to tick
4722 in the case where the clock went backwards by a substantial amount such
4723 that the microsecond fraction of "now" was more than the microsecond
4724 fraction of "then" (but the whole seconds number was less).
4726 38. Added support for the libradius Radius client library this is found on
4727 FreeBSD (previously only the radiusclient library was supported).
4733 1. When certain lookups returned multiple values in the form name=value, the
4734 quoting of the values was not always being done properly. Specifically:
4735 (a) If the value started with a double quote, but contained no whitespace,
4737 (b) If the value contained whitespace other than a space character (i.e.
4738 tabs or newlines or carriage returns) it was not quoted.
4739 This fix has been applied to the mysql and pgsql lookups by writing a
4740 separate quoting function and calling it from the lookup code. The fix
4741 should probably also be applied to nisplus, ibase and oracle lookups, but
4742 since I cannot test any of those, I have not disturbed their existing code.
4744 2. A hit in the callout cache for a specific address caused a log line with no
4745 reason for rejecting RCPT. Now it says "Previous (cached) callout
4746 verification failure".
4748 3. There was an off-by-one bug in the queryprogram router. An over-long
4749 return line was truncated at 256 instead of 255 characters, thereby
4750 overflowing its buffer with the terminating zero. As well as fixing this, I
4751 have increased the buffer size to 1024 (and made a note to document this).
4753 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
4754 when Exim is waiting for an SMTP response from a remote server, Exim
4755 restarts its select() call on the socket, thereby resetting its timeout.
4756 This is not a problem when such interrupts are rare. Somebody set up a cron
4757 job to run exiwhat every 2 minutes, which is less than the normal select()
4758 timeout (5 or 10 minutes). This meant that the select() timeout never
4759 kicked in because it was always reset. I have fixed this by comparing the
4760 time when an interrupt arrives with the time at the start of the first call
4761 to select(). If more time than the timeout has elapsed, the interrupt is
4762 treated as a timeout.
4764 5. Some internal re-factoring in preparation for the addition of Sieve
4765 extensions (by MH). In particular, the "personal" test is moved to a
4766 separate function, and given an option for scanning Cc: and Bcc: (which is
4767 not set for Exim filters).
4769 6. When Exim created an email address using the login of the caller as the
4770 local part (e.g. when creating a From: or Sender: header line), it was not
4771 quoting the local part when it contained special characters such as @.
4773 7. Installed new OpenBSD configuration files.
4775 8. Reworded some messages for syntax errors in "and" and "or" conditions to
4776 try to make them clearer.
4778 9. Callout options, other than the timeout value, were being ignored when
4779 verifying sender addresses in header lines. For example, when using
4781 verify = header_sender/callout=no_cache
4783 the cache was (incorrectly) being used.
4785 10. Added a missing instance of ${EXE} to the exim_install script; this affects
4786 only the Cygwin environment.
4788 11. When return_path_on_delivery was set as a log selector, if different remote
4789 addresses in the same message used different return paths and parallel
4790 remote delivery occurred, the wrong values would sometimes be logged.
4791 (Whenever a remote delivery process finished, the return path value from
4792 the most recently started remote delivery process was logged.)
4794 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
4795 lines when AUTH and/or TLS are in use. This is the "received protocol"
4796 field. Exim used to use "asmtp" for authenticated SMTP, without any
4797 indication (in the protocol name) for TLS use. Now it follows the RFC and
4798 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
4799 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
4800 names appear in log lines as well as in Received: header lines.
4802 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
4803 extensions, and comparison tests, and to fix some bugs.
4805 14. Changes to the "personal" filter test:
4807 (1) The test was buggy in that it was just doing the equivalent of
4808 "contains" tests on header lines. For example, if a user's address was
4809 anne@some.where, the "personal" test would incorrectly be true for
4811 To: susanne@some.where
4813 This test is now done by extracting each address from the header in turn,
4814 and checking the entire address. Other tests that are part of "personal"
4815 are now done using regular expressions (for example, to check local parts
4816 of addresses in From: header lines).
4818 (2) The list of non-personal local parts in From: addresses has been
4819 extended to include "listserv", "majordomo", "*-request", and "owner-*",
4820 taken from the Sieve specification recommendations.
4822 (3) If the message contains any header line starting with "List-" it is
4823 treated as non-personal.
4825 (4) The test for "circular" in the Subject: header line has been removed
4826 because it now seems ill-conceived.
4828 15. Minor typos in src/EDITME comments corrected.
4830 16. Installed latest exipick from John Jetmore.
4832 17. If headers_add on a router specified a text string that was too long for
4833 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
4834 of string_sprintf() is now avoided.
4836 18. $message_body_size was not set (it was always zero) when running the DATA
4837 ACL and the local_scan() function.
4839 19. For the "mail" command in an Exim filter, no default was being set for
4840 the once_repeat time, causing a random time value to be used if "once" was
4841 specified. (If the value happened to be <= 0, no repeat happened.) The
4842 default is now 0s, meaning "never repeat". The "vacation" command was OK
4843 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
4844 (I found it when inspecting the code).
4846 20. There is now an overall timeout for performing a callout verification. It
4847 defaults to 4 times the callout timeout, which applies to individual SMTP
4848 commands during the callout. The overall timeout applies when there is more
4849 than one host that can be tried. The timeout is checked before trying the
4850 next host. This prevents very long delays if there are a large number of
4851 hosts and all are timing out (e.g. when the network connections are timing
4852 out). The value of the overall timeout can be changed by specifying an
4853 additional sub-option for "callout", called "maxwait". For example:
4855 verify = sender/callout=5s,maxwait=20s
4857 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
4858 to the end before writing, but this should make it even safer).
4860 22. Exim was forgetting that it had advertised PIPELINING for the second and
4861 subsequent messages on an SMTP connection. It was also not resetting its
4862 memory on STARTTLS and an internal HELO.
4864 23. When Exim logs an SMTP synchronization error within a session, it now
4865 records whether PIPELINING has been advertised or not.
4867 24. Added 3 instances of "(long int)" casts to time_t variables that were being
4868 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
4869 rather than long int.
4871 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
4873 26. Added the never_mail option to autoreply.
4879 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
4880 crash if the getsockname() call failed; this can happen if a connection is
4881 closed very soon after it is established. The problem was simply in the
4882 order in which certain operations were done, causing Exim to try to write
4883 to the SMTP stream before it had set up the file descriptor. The bug has
4884 been fixed by making things happen in the correct order.
4890 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
4891 before the connection was closed, thus losing the rejection response.
4893 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
4894 some early Solaris releases, but causes trouble in current releases where
4895 socklen_t is defined.
4897 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
4900 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
4903 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
4904 that releasing the top of it at the end releases what was used for sub-
4905 expansions (unless the block got too big). However, discard this block if
4906 the first thing is a variable or header, so that we can use its block when
4907 it is dynamic (useful for very large $message_headers, for example).
4909 6. Lookups now cache *every* query, not just the most recent. A new, separate
4910 store pool is used for this. It can be recovered when all lookup caches are
4911 flushed. Lookups now release memory at the end of their result strings.
4912 This has involved some general refactoring of the lookup sources.
4914 7. Some code has been added to the store_xxx() functions to reduce the amount
4915 of flapping under certain conditions.
4917 8. log_incoming_interface used to affect only the <= reception log lines. Now
4918 it causes the local interface and port to be added to several more SMTP log
4919 lines, for example "SMTP connection from", and rejection lines.
4921 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
4923 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
4925 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
4926 could be overwritten at the end of the current message (or the start of a
4927 new message if it was set in a HELO ACL). The value is now preserved for
4928 the duration of the SMTP connection.
4930 12. If a transport had a headers_rewrite setting, and a matching header line
4931 contained an unqualified address, that address was qualified, even if it
4932 did not match any rewriting rules. The underlying bug was that the values
4933 of the flags that permit the existence of unqualified sender and recipient
4934 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
4935 non-local messages, and by -bnq for local messages) were not being
4936 preserved with the message after it was received.
4938 13. When Exim was logging an SMTP synchronization error, it could sometimes log
4939 "next input=" as part of the text comprising the host identity instead of
4940 the correct text. The code was using the same buffer for two different
4941 strings. However, depending on which order the printing function evaluated
4942 its arguments, the bug did not always show up. Under Linux, for example, my
4943 test suite worked just fine.
4945 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
4946 doesn't work with some older versions of Perl. It has been changed to "my",
4947 which in any case is probably the better facility to use.
4949 15. A really picky compiler found some instances of statements for creating
4950 error messages that either had too many or two few arguments for the format
4953 16. The size of the buffer for calls to the DNS resolver has been increased
4954 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
4955 for addresses that have a lot of PTR records. This alleviates a problem; it
4956 does not fully solve it.
4958 17. A dnsdb lookup for PTR records that receives more data than will fit in the
4959 buffer now truncates the list and logs the incident, which is the same
4960 action as happens when Exim is looking up a host name and its aliases.
4961 Previously in this situation something unpredictable would happen;
4962 sometimes it was "internal error: store_reset failed".
4964 18. If a server dropped the connection unexpectedly when an Exim client was
4965 using GnuTLS and trying to read a response, the client delivery process
4966 crashed while trying to generate an error log message.
4968 19. If a "warn" verb in an ACL added multiple headers to a message in a single
4969 string, for example:
4971 warn message = H1: something\nH2: something
4973 the text was added as a single header line from Exim's point of view
4974 though it ended up OK in the delivered message. However, searching for the
4975 second and subsequent header lines using $h_h2: did not work. This has been
4976 fixed. Similarly, if a system filter added multiple headers in this way,
4977 the routers could not see them.
4979 20. Expanded the error message when iplsearch is called with an invalid key to
4980 suggest using net-iplsearch in a host list.
4982 21. When running tests using -bh, any delays imposed by "delay" modifiers in
4983 ACLs are no longer actually imposed (and a message to that effect is
4986 22. If a "gecos" field in a passwd entry contained escaped characters, in
4987 particular, if it contained a \" sequence, Exim got it wrong when building
4988 a From: or a Sender: header from that name. A second bug also caused
4989 incorrect handling when an unquoted " was present following a character
4990 that needed quoting.
4992 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
4993 was not being matched caselessly.
4995 24. Arranged for all hyphens in the exim.8 source to be escaped with
4998 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
4999 properly. Recipient callout cache records were still being keyed to include
5000 the sender, even when use_sender was set false. This led to far more
5001 callouts that were necessary. The sender is no longer included in the key
5002 when use_sender is false.
5004 26. Added "control = submission" modifier to ACLs.
5006 27. Added the ${base62d: operator to decode base 62 numbers.
5008 28. dnsdb lookups can now access SRV records.
5010 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
5011 the configuration file.
5013 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
5014 (-v) mode. This makes the output for a verbose queue run more intelligible.
5016 31. Added a use_postmaster feature to recipient callouts.
5018 32. Added the $body_zerocount variable, containing the number of binary zero
5019 bytes in the message body.
5021 33. The time of last modification of the "new" subdirectory is now used as the
5022 "mailbox time last read" when there is a quota error for a maildir
5025 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
5027 35. Added +ignore_unknown as a special item in host lists.
5029 36. Code for decoding IPv6 addresses in host lists is now included, even if
5030 IPv6 support is not being compiled. This fixes a bug in which an IPv6
5031 address was recognized as an IP address, but was then not correctly decoded
5032 into binary, causing unexpected and incorrect effects when compared with
5039 1. Very minor rewording of debugging text in manualroute to say "list of
5040 hosts" instead of "hostlist".
5042 2. If verify=header_syntax was set, and a header line with an unqualified
5043 address (no domain) and a large number of spaces between the end of the
5044 name and the colon was received, the reception process suffered a buffer
5045 overflow, and (when I tested it) crashed. This was caused by some obsolete
5046 code that should have been removed. The fix is to remove it!
5048 3. When running in the test harness, delay a bit after writing a bounce
5049 message to get a bit more predictability in the log output.
5051 4. Added a call to search_tidyup() just before forking a reception process. In
5052 theory, someone could use a lookup in the expansion of smtp_accept_max_
5053 per_host which, without the tidyup, could leave open a database connection.
5055 5. Added the variables $recipient_data and $sender_data which get set from a
5056 lookup success in an ACL "recipients" or "senders" condition, or a router
5057 "senders" option, similar to $domain_data and $local_part_data.
5059 6. Moved the writing of debug_print from before to after the "senders" test
5062 7. Change 4.31/66 (moving the time when the Received: is generated) caused
5063 problems for message scanning, either using a data ACL, or using
5064 local_scan() because the Received: header was not generated till after they
5065 were called (in order to set the time as the time of reception completion).
5066 I have revised the way this works. The header is now generated after the
5067 body is received, but before the ACL or local_scan() are called. After they
5068 are run, the timestamp in the header is updated.
5074 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
5075 before starting a queue runner without re-exec. This happened only when
5076 deliver_drop_privilege was set or when the Exim user was set to root. The
5077 effect of the bug was that timeouts during subsequent deliveries caused
5078 crashes instead of being properly handled. The handler is now left at its
5079 default (and expected) setting.
5081 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
5082 message, again when deliver_drop_privilege is set or Exim is run as root.
5083 The bug described in (1) was not present in this case, but the tidying up
5084 of the other signals was missing. I have made the two cases consistent.
5086 3. The ignore_target_hosts setting on a manualroute router was being ignored
5087 for hosts that were looked up using the /MX notation.
5089 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
5092 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
5093 operated on the sender address. After changing the $sender_address to <>
5094 for the sender address verify, Exim was re-instated it as the original
5095 (before rewriting) address, but remembering that it had rewritten it, so it
5096 wasn't rewriting it again. This bug also had the effect of breaking the
5097 sender address verification caching when the sender address was rewritten.
5099 6. The ignore_target_hosts option was being ignored by the ipliteral router.
5100 This has been changed so that if the ip literal address matches
5101 ignore_target_hosts, the router declines.
5103 7. Added expansion conditions match_domain, match_address, and match_local_
5104 part (NOT match_host).
5106 8. The placeholder for the Received: header didn't have a length field set.
5108 9. Added code to Exim itself and to exim_lock to test for a specific race
5109 condition that could lead to file corruption when using MBX delivery. The
5110 issue is with the lockfile that is created in /tmp. If this file is removed
5111 after a process has opened it but before that process has acquired a lock,
5112 there is the potential for a second process to recreate the file and also
5113 acquire a lock. This could lead to two Exim processes writing to the file
5114 at the same time. The added code performs the same test as UW imapd; it
5115 checks after acquiring the lock that its file descriptor still refers to
5116 the same named file.
5118 10. The buffer for building added header lines was of fixed size, 8192 bytes.
5119 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
5122 11. Added the smtp_active_hostname option. If used, this will typically be made
5123 to depend on the incoming interface address. Because $interface_address is
5124 not set up until the daemon has forked a reception process, error responses
5125 that can happen earlier (such as "too many connections") no longer contain
5128 12. If an expansion in a condition on a "warn" statement fails because a lookup
5129 defers, the "warn" statement is abandoned, and the next ACL statement is
5130 processed. Previously this caused the whole ACL to be aborted.
5132 13. Added the iplsearch lookup type.
5134 14. Added ident_timeout as a log selector.
5136 15. Added tls_certificate_verified as a log selector.
5138 16. Added a global option tls_require_ciphers (compare the smtp transport
5139 option of the same name). This controls incoming TLS connections.
5141 17. I finally figured out how to make tls_require_ciphers do a similar thing
5142 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
5143 before starting the TLS session.
5145 18. Tabs are now shown as \t in -bP output.
5147 19. If the log selector return_path_on_delivery was set, Exim crashed when
5148 bouncing a message because it had too many Received: header lines.
5150 20. If two routers both had headers_remove settings, and the first one included
5151 a superfluous trailing colon, the final name in the first list and the
5152 first name in the second list were incorrectly joined into one item (with a
5153 colon in the middle).
5159 1. Added -C and -D options to the exinext utility, mainly to make it easier
5160 to include in the automated testing, but these could be helpful when
5161 multiple configurations are in use.
5163 2. The exinext utility was not formatting the output nicely when there was
5164 an alternate port involved in the retry record key, nor when there was a
5165 message id as well (for retries that were specific to a specific message
5166 and a specific host). It was also confused by IPv6 addresses, because of
5167 the additional colons they contain. I have fixed the IPv4 problem, and
5168 patched it up to do a reasonable job for IPv6.
5170 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
5171 delivery, the log line now contains "pipelined" if PIPELINING was used.
5173 4. An SMTP transport process used to panic and die if the bind() call to set
5174 an explicit outgoing interface failed. This has been changed; it is now
5175 treated in the same way as a connect() failure.
5177 5. A reference to $sender_host_name in the part of a conditional expansion
5178 that was being skipped was still causing a DNS lookup. This no longer
5181 6. The def: expansion condition was not recognizing references to header lines
5182 that used bh_ and bheader_.
5184 7. Added the _cache feature to named lists.
5186 8. The code for checking quota_filecount in the appendfile transport was
5187 allowing one more file than it should have been.
5189 9. For compatibility with Sendmail, the command line option
5197 and sets the incoming protocol and host name (for trusted callers). The
5198 host name and its colon can be omitted when only the protocol is to be set.
5199 Note the Exim already has two private options, -pd and -ps, that refer to
5200 embedded Perl. It is therefore impossible to set a protocol value of "d" or
5201 "s", but I don't think that's a major issue.
5203 10. A number of refactoring changes to the code, none of which should affect
5206 (a) The number of logging options was getting close to filling up the
5207 32-bit word that was used as a bit map. I have split them into two classes:
5208 those that are passed in the argument to log_write(), and those that are
5209 only ever tested independently outside of that function. These are now in
5210 separate 32-bit words, so there is plenty of room for expansion again.
5211 There is no change in the user interface or the logging behaviour.
5213 (b) When building, for example, log lines, the code previously used a
5214 macro that called string_cat() twice, in order to add two strings. This is
5215 not really sufficiently general. Furthermore, there was one instance where
5216 it was actually wrong because one of the argument was used twice, and in
5217 one call a function was used. (As it happened, calling the function twice
5218 did not affect the overall behaviour.) The macro has been replaced by a
5219 function that can join an arbitrary number of extra strings onto a growing
5222 (c) The code for expansion conditions now uses a table and a binary chop
5223 instead of a serial search (which was left over from when there were very
5224 few conditions). Also, it now recognizes conditions like "pam" even when
5225 the relevant support is not compiled in: a suitably worded error message is
5226 given if an attempt is made to use such a condition.
5228 11. Added ${time_interval:xxxxx}.
5230 12. A bug was causing one of the ddress fields not to be passed back correctly
5231 from remote delivery subprocesses. The field in question was not being
5232 subsequently used, so this caused to problems in practice.
5234 13. Added new log selectors queue_time and deliver_time.
5236 14. Might have fixed a bug in maildirsizefile handling that threw up
5237 "unexpected character" debug warnings, and recalculated the data
5238 unnecessarily. In any case, I expanded the warning message to give more
5241 15. Added the message "Restricted characters in address" to the statements in
5242 the default ACL that block characters like @ and % in local parts.
5244 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
5245 Three changes have been made:
5247 (a) There was a serious bug; a negative response to MAIL caused the whole
5248 recipient domain to be cached as invalid, thereby blocking all messages
5249 to all local parts at the same domain, from all senders. This bug has
5250 been fixed. The domain is no longer cached after a negative response to
5251 MAIL if the sender used is not empty.
5253 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
5256 (c) A new callout option, "use_sender" has been added for people who want
5257 the modified behaviour.
5263 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
5266 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
5267 indeed breaks things for older releases.
5269 3. Added additional logging to the case where there is a problem reading data
5270 from a filter that is running in a subprocess using a pipe, in order to
5271 try to track down a specific problem.
5273 4. Testing facility fudge: when running in the test harness and attempting
5274 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
5275 getting "No route to host". Convert this to a timeout.
5277 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
5280 6. Some OS don't have socklen_t but use size_t instead. This affects the
5281 fifth argument of getsockopt() amongst other things. This is now
5282 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
5283 can be set for individual OS. I have set it for SunOS5, OSF1, and
5284 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
5285 some earlier ones do not.
5287 7. Change 4.30/15 was not doing the test caselessly.
5289 8. The standard form for an IPv6 address literal was being rejected by address
5290 parsing in, for example, MAIL and RCPT commands. An example of this kind of
5291 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5292 this, as well as the form without the "IPv6" on the front (but only when
5293 address literals are enabled, of course).
5295 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
5297 10. Exim crashed if a message with an empty sender address specified by -f
5298 encountered a router with an errors_to setting. This could be provoked only
5299 by a command such as
5303 where an empty string was supplied; "<>" did not hit this bug.
5305 11. Installed PCRE release 4.5.
5307 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
5308 remained set. It is now erased.
5310 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
5311 times from message ids (which are base 36 rather than the normal 62).
5313 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
5314 were being counted as actual protocol errors, and logged if the log
5315 selector +smtp_protocol_error was set. One cannot be perfect in this test,
5316 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
5317 and DATA following a set of rejected RCPTs do not count as protocol errors.
5318 In other words, Exim assumes they were pipelined, though this may not
5319 actually be the case. Of course, in all cases the client gets an
5320 appropriate error code.
5322 15. If a lookup fails in an ACL condition, a message about the failure may
5323 be available; it is used if testing the ACL cannot continue, because most
5324 such messages specify what the cause of the deferral is. However, some
5325 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
5326 that caused an old message to be retained and used if a later statement
5327 caused a defer, replacing the real cause of the deferral.
5329 16. If an IP address had so many PTR records that the DNS lookup buffer
5330 was not large enough to hold them, Exim could crash while trying to process
5331 the truncated data. It now detects and logs this case.
5333 17. Further to 4.21/58, another change has been made: if (and only if) the
5334 first line of a message (the first header line) ends with CRLF, a bare LF
5335 in a subsequent header line has a space inserted after it, so as not to
5336 terminate the header.
5338 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
5339 unnecessarily, used atoi() instead of strtol(), and didn't check the
5340 termination when getting file sizes from file names by regex.
5342 19. Completely re-implemented the support for maildirsize files, in the light
5343 of a number of problems with the previous contributed implementation
5344 (4.30/29). In particular:
5346 . If the quota is zero, the maildirsize file is maintained, but no quota is
5349 . If the maildir directory does not exist, it is created before any attempt
5350 to write a maildirsize file.
5352 . The quota value in the file is just a cache; if the quota is changed in
5353 the transport, the new value overrides.
5355 . A regular expression is available for excluding directories from the
5358 20. The autoreply transport checks the characters in options that define the
5359 message's headers; it allows continued headers, but it was checking with
5360 isspace() after an embedded newline instead of explicitly looking for a
5363 21. If all the "regular" hosts to which an address was routed had passed their
5364 expiry times, and had not reached their retry times, the address was
5365 bounced, even if fallback hosts were defined. Now Exim should go on to try
5368 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
5369 equivalent code in the SMTP transport. Some hosts send humungous responses
5370 to HELO/EHLO, more than 1024 it seems.
5372 23. Refactoring: code in filter.c used (void *) for "any old type" but this
5373 gives compiler warnings in some environments. I've now done it "properly",
5376 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
5377 (because of problems with the built-in one) was declared to return uschar *
5378 instead of char *, causing compiler failure.
5380 25. Fixed a file descriptor leak when processing alias/forward files.
5382 26. Fixed a minor format string issue in dbfn.c.
5384 27. Typo in exim.c: ("dmbnz" for "dbmnz").
5386 28. If a filter file refered to $h_xxx or $message_headers, and the headers
5387 contained RFC 2047 "words", Exim's memory could, under certain conditions,
5390 29. When a sender address is verified, it is cached, to save repeating the test
5391 when there is more than one recipient in a message. However, when the
5392 verification involves a callout, it is possible for different callout
5393 options to be set for different recipients. It is too complicated to keep
5394 track of this in the cache, so now Exim always runs a verification when a
5395 callout is required, relying on the callout cache for the optimization.
5396 The overhead is duplication of the address routing, but this should not be
5399 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
5400 to be verified with callout=postmaster, and the main callout worked but the
5401 postmaster check failed, the verification correctly failed. However, if a
5402 subsequent RCPT command asked for sender verification *without* the
5403 postmaster check, incorrect caching caused this verification also to fail,
5406 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
5407 it was not caching the DNS options (qualify_single, search_parents) that
5408 were used when the lookup failed. A subsequent lookup with different
5409 options therefore always gave the same answer, though there were cases
5410 where it should not have. (Example: a "domains = !$mx_any" option on a
5411 dnslookup router: the "domains" option is always processed without any
5412 widening, but the router might have qualify_single set.) Now Exim uses the
5413 cached value only when the same options are set.
5415 32. Added John Jetmore's "exipick" utility to the distribution.
5417 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
5418 than a timeout (e.g. a certificate is required, and is not provided), an
5419 Exim server now closes the connection immediately. Previously it waited for
5420 the client to close - but if the client is SSL, it seems that they each
5421 wait for each other, leading to a delay before one of them times out.
5423 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
5424 maintained 0.8.x compatibility because I don't think many are using it, and
5425 it is clearly obsolete.
5427 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
5430 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
5431 client certificate was expired. A simple patch fixes this, though I don't
5432 understand the full logic of why the verify callback is called multiple
5435 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
5436 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
5437 which causes problems with some clients (such as the Certicom SSL Plus
5438 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
5439 disables the coutermeasure allowing Eudora to connect."
5441 38. Exim was not checking that a write() to a log file succeeded. This could
5442 lead to Bad Things if a log got too big, in particular if it hit a file
5443 size limit. Exim now panics and dies if it cannot write to a log file, just
5444 as it does if it cannot open a log file.
5446 39. Modified OS/Makefile-Linux so that it now contains
5448 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
5450 The two -D definitions ensure that Exim is compiled with large file
5451 support, which makes it possible to handle log files that are bigger than
5454 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
5455 instance) a domain was checked against a named list that involved a lookup,
5456 causing $domain_data to be set, then another domain was checked against the
5457 same list, then the first domain was re-checked, the value of $domain_data
5458 after the final check could be wrong. In particular, if the second check
5459 failed, it could be set empty. This bug probably also applied to
5462 41. The strip_trailing_dot option was not being applied to the address given
5463 with the -f command-line option.
5465 42. The code for reading a message's header from the spool was incrementing
5466 $received_count, but never initializing it. This meant that the value was
5467 incorrect (doubled) while delivering a message in the same process in which
5468 it was received. In the most common configuration of Exim, this never
5469 happens - a fresh exec is done - but it can happen when
5470 deliver_drop_privilege is set.
5472 43. When Exim logs an SMTP synchronization error - client data sent too soon -
5473 it now includes up to 150 characters of the unexpected data in the log
5476 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
5477 and building data strings. The size of both of these buffers was 10 000
5478 bytes - far larger than anybody would *ever* want, thought I. Needless to
5479 say, somebody hit the limit. I have increased the maximum line length to
5480 20 000 and the maximum data length of concatenated lines to 100 000. I have
5481 also fixed two bugs, because there was no checking on these buffers. Tsk,
5482 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
5483 buffer is too small.
5485 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
5486 lsearch lookups. Now it does.
5488 46. When parsing a route_list item in a manualroute router, a fixed-length
5489 buffer was used for the list of hosts. I made this 1024 bytes long,
5490 thinking that nobody would ever have a list of hosts that long. Wrong.
5491 Somebody had a whole pile of complicated expansion conditions, and the
5492 string was silently truncated, leading to an expansion error. It turns out
5493 that it is easier to change to an unlimited length (owing to other changes
5494 that have happened since this code was originally written) than to build
5495 structure for giving a limitation error. The length of the item that
5496 expands into the list of hosts is now unlimited.
5498 47. The lsearch lookup could not handle data where the length of text line was
5499 more than 4095 characters. Such lines were truncated, leading to shortened
5500 data being returned. It should now handle lines of any length.
5502 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
5503 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
5506 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
5507 replacing strings such as BIN_DIRECTORY in a source file: the replacement
5508 no longer happens in comment lines. A list of replacements is now placed
5509 at the head of all of the source files, except those whose only change is
5510 to replace PERL_COMMAND in the very first #! line.
5512 50. Replaced the slow insertion sort in queue.c, for sorting the list of
5513 messages on the queue, with a bottom-up merge sort, using code contributed
5514 by Michael Haardt. This should make operations like -bp somewhat faster on
5515 large queues. It won't affect queue runners, except when queue_run_in_order
5518 51. Installed eximstats 1.31 in the distribution.
5520 52. Added support for SRV lookups to the dnslookup router.
5522 53. If an ACL referred to $message_body or $message_body_end, the value was not
5523 reset for any messages that followed in the same SMTP session.
5525 54. The store-handling optimization for building very long strings was not
5526 differentiating between the different store pools. I don't think this
5527 actually made any difference in practice, but I've tidied it.
5529 55. While running the routers to verify a sender address, $sender_address
5530 was still set to the sender address. This is wrong, because when routing to
5531 send a bounce to the sender, it would be empty. Therefore, I have changed
5532 it so that, while verifying a sender address, $sender_address is set to <>.
5533 (There is no change to what happens when verifying a recipient address.)
5535 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
5536 target A or AAAA records (if not already returned) without resetting the
5537 qualify_single or search_parents options of the DNS resolver. These are
5538 inappropriate in this case because the targets of MX and SRV records must
5539 be FQDNs. A broken DNS record could cause trouble if it happened to have a
5540 target that, when qualified, matched something in the local domain. These
5541 two options are now turned off when doing these lookups.
5543 57. It seems that at least some releases of Reiserfs (which does not have the
5544 concept of a fixed number of inodes) returns zero and not -1 for the
5545 number of available inodes. This interacted badly with check_spool_inodes,
5546 which assumed that -1 was the "no such thing" setting. What I have done is
5547 to check that the total number of inodes is greater than zero before doing
5548 the test of how many are available.
5550 58. When a "warn" ACL statement has a log_message modifier, the message is
5551 remembered, and not repeated. This is to avoid a lot of repetition when a
5552 message has many recipients that cause the same warning to be written.
5553 Howewer, Exim was preserving the list of already written lines for an
5554 entire SMTP session, which doesn't seem right. The memory is now reset if a
5555 new message is started.
5557 59. The "rewrite" debugging flag was not showing the result of rewriting in the
5558 debugging output unless log_rewrite was also set.
5560 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
5561 of (int)(handle) when we know that handle contains (void *)(-1).
5563 61. The Exim daemon panic-logs an error return when it closes the incoming
5564 connection. However "connection reset by peer" seems to be common, and
5565 isn't really an error worthy of noting specially, so that particular error
5568 62. When Exim is trying to find all the local interfaces, it used to panic and
5569 die if the ioctl to get the interface flags failed. However, it seems that
5570 on at least one OS (Solaris 9) it is possible to have an interface that is
5571 included in the list of interfaces, but for which you get a failure error
5572 for this call. This happens when the interface is not "plumbed" into a
5573 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
5574 failure of the "get flags" call assumes that the interface is down.
5576 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
5577 makes life easier for people who are doing arithmetic on fields extracted
5578 from dates, where you often get leading zeros that should not be
5579 interpreted as octal.
5581 64. Added qualify_domain to the redirect router, to override the global
5584 65. If a pathologically long header line contained very many addresses (the
5585 report of this problem mentioned 10 000) and each of them was rewritten,
5586 Exim could use up a very large amount of memory. (It kept on making new
5587 copies of the header line as it rewrote, and never released the old ones.)
5588 At the expense of a bit more processing, the header rewriting function has
5589 been changed so that it no longer eats memory in this way.
5591 66. The generation of the Received: header has been moved from the time that a
5592 message starts to be received, to the time that it finishes. The timestamp
5593 in the Received: header should now be very close to that of the <= log
5594 line. There are two side-effects of this change:
5596 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
5597 logged header lines no longer include the local Received: line, because
5598 it has not yet been created. The same applies to a copy of the message
5599 that is returned to a non-SMTP sender when a message is rejected.
5601 (b) When a filter file is tested using -bf, no additional Received: header
5602 is added to the test message. After some thought, I decided that this
5605 This change does not affect the value of $received_for. It is still set
5606 after address rewriting, but before local_scan() is called.
5608 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
5610 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
5611 gave an unhelpful panic error message, and a defer error. I have managed to
5612 change this behaviour so that it now rejects any supplied certificate,
5613 which seems right, as the list of acceptable certificates is empty.
5615 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
5616 gave an unhelpful defer error. I have not managed to make this reject any
5617 supplied certificates, but the error message it gives is "no certificate
5618 supplied", which is not helpful.
5620 70. exigrep's output now also includes lines that are not associated with any
5621 message, but which match the given pattern. Implemented by a patch from
5622 Martin Sluka, which also tidied up the Perl a bit.
5624 71. Recipient callout verification, like sender verification, was using <> in
5625 the MAIL FROM command. This isn't really the right thing, since the actual
5626 sender may affect whether the remote host accepts the recipient or not. I
5627 have changed it to use the actual sender in the callout; this means that
5628 the cache record is now keyed on a recipient/sender pair, not just the
5629 recipient address. There doesn't seem to be a real danger of callout loops,
5630 since a callout by the remote host to check the sender would use <>.
5631 [SEE ABOVE: changed after hitting problems.]
5633 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
5634 temporary errors. However, in the case of such a code being given after
5635 the end of a data transmission (i.e. after ".") Exim was failing to write
5636 a retry record for the message. (Yes, there was some broken host that was
5637 actually sending 8xx at this point.)
5639 73. An unknown lookup type in a host list could cause Exim to panic-die when
5640 the list was checked. (An example that provoked this was putting <; in the
5641 middle of a list instead of at the start.) If this happened during a DATA
5642 ACL check, a -D file could be left lying around. This kind of configuration
5643 error no longer causes Exim to die; instead it causes a defer errror. The
5644 incident is still logged to the main and panic logs.
5646 74. Buglet left over from Exim 3 conversion. The message "too many messages
5647 in one connection" was written to the rejectlog but not the mainlog, except
5648 when address rewriting (yes!) was being logged.
5650 75. Added write_rejectlog option.
5652 76. When a system filter was run not as root (that is, when system_filter_user
5653 was set), the values of the $n variables were not being returned to the
5654 main process; thus, they were not subsequently available in the $sn
5657 77. Added +return_path_on_delivery log selector.
5659 78. A connection timeout was being treated differently from recipients deferred
5660 when testing hosts_max_try with a message that was older than the host's
5661 retry timeout. (The host should not be counted, thus allowing all hosts to
5662 be tried at least once before bouncing.) This may have been the cause of an
5663 occasionally reported bug whereby a message would remain on the queue
5664 longer than the retry timeout, but would be bounced if a delivery was
5665 forced. I say "may" because I never totally pinned down the problem;
5666 setting up timeout/retry tests is difficult. See also the next item.
5668 79. The ultimate address timeout was not being applied to errors that involved
5669 a combination of host plus message (for example, a timeout on a MAIL
5670 command). When an address resolved to a number of possible hosts, and they
5671 were not all tried for each delivery (e.g. because of hosts_max_try), a
5672 message could remain on the queue longer than the retry timeout.
5674 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
5677 81. Fixed an obscure SMTP outgoing bug which required at least the following
5678 conditions: (a) there was another message waiting for the same server;
5679 (b) the server returned 5xx to all RCPT commands in the first message so
5680 that the message was not completed; (c) the server dropped the connection
5681 or gave a negative response to the RSET that Exim sends to abort the
5682 transaction. The observed case was a dropped connection after DATA that had
5683 been sent in pipelining mode. That is, the server had advertised PIPELINING
5684 but was not implementing it correctly. The effect of the bug was incorrect
5685 behaviour, such as trying another host, and this could lead to a crash.
5691 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
5692 and daemon.c were passed as pointers to ints; they should have been
5693 pointers to socklen_t variables (which are typically unsigned ints).
5695 2. Some signed/unsigned type warnings in the os.c file for Linux have been
5698 3. Fixed a really odd bug that affected only the testing scheme; patching a
5699 certain fixed string in the binary changed the value of another string that
5700 happened to be identical to the end of the original first string.
5702 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
5703 name", it returns that address as the IP address. On some operating
5704 systems (e.g. Solaris), it also passes back the IP address string as the
5705 "host name". However, on others (e.g. Linux), it passes back an empty
5706 string. Exim wasn't checking for this, and was changing the host name to an
5707 empty string, assuming it had been canonicized.
5709 5. Although rare, it is permitted to have more than one PTR record for a given
5710 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
5711 all the names associated with an address, because they do in Solaris.
5712 However, it seems that they do not in Linux for data that comes from the
5713 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
5714 I found this out when I moved to a new Linux workstation and tried to run
5715 the Exim test suite.
5717 To get round this problem I have changed the code so that it now does its
5718 own call to the DNS to look up PTR records when searching for a host name.
5719 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
5720 addresses that are only in /etc/hosts are still found.
5722 This behaviour is, however, controlled by an option called host_lookup_
5723 order, which defaults to "bydns:byaddr". If people want to use the other
5724 order, or indeed, just use one or the other means of lookup, they can
5725 specify it in this variable.
5727 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
5728 some operating systems, this comes back from gethostbyaddr() as an empty
5729 string, and this is what Exim used to test for. However, it seems that in
5730 other systems, "." is yielded. Exim now tests for this case too.
5732 7. The values of check_spool_space and check_log_space are now held internally
5733 as a number of kilobytes instead of an absolute number of bytes. If a
5734 numbers is specified without 'K' or 'M', it is rounded up to the nearest
5735 kilobyte. This means that much larger values can be stored.
5737 8. Exim monitor: an attempt to get the action menu when not actually pointing
5738 at a message produces an empty menu entitled "No message selected". This
5739 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
5740 no entries in it ("Shell widget menu has zero width and/or height"). So I
5741 have added a single, blank menu entry in this case.
5743 9. Added ${quote_local_part.
5745 10. MIME decoding is now applied to the contents of Subject: header lines when
5748 11. Now that a reference to $sender_host_address automatically causes a reverse
5749 lookup to occur if necessary (4.13/18), there is no need to arrange for a
5750 host lookup before query-style lookups in lists that might use this
5751 variable. This has therefore been abolished, and the "net-" prefix is no
5752 longer necessary for query-style lookups.
5754 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
5755 have been a typo for LFLAGS, so it has been changed.
5757 13. The install script calls Exim with "-C /dev/null" in order to find the
5758 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
5759 to be output. Howeve, since Exim outputs its version number before the
5760 error, it didn't break the script. It just looked ugly. I fixed this by
5761 always allowing "-C /dev/null" if the caller is root.
5763 14. Ignore overlarge ACL variable number when reading spool file - insurance
5764 against a later release with more variables having written the file.
5766 15. The standard form for an IPv6 address literal was being rejected by EHLO.
5767 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5768 this, as well as the form without the "IPv6" on the front.
5770 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
5771 OS/Makefile-Darwin file.
5773 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
5774 only to LDAP libraries that do not have LDAP_OPT_DEREF.
5776 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
5777 variable. However, some OS use ints for inodes. I've added cast to long int
5778 to get rid of the compiler warning.
5780 19. I had forgotten to lock out "/../" in configuration file names when
5781 ALT_CONFIG_PREFIX was set.
5783 20. Routers used for verification do not need to specify transports. However,
5784 if such a router generated a host list, and callout was configured, Exim
5785 crashed, because it could not find a port number from the (non-existent)
5786 transport. It now assumes port 25 in this circumstance.
5788 21. Added the -t option to exigrep.
5790 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
5791 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
5792 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
5793 them is set, LOOKUP_LSEARCH is forced.
5795 23. "exim -bV" now outputs a list of lookups that are included in the binary.
5797 24. Added sender and host information to the "rejected by local_scan()" log
5798 line; previously there was no indication of these.
5800 25. Added .include_if_exists.
5802 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
5803 Linux. It turns out that not all file systems support this. Apparently some
5804 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
5805 do it.) To cope with this, the error EINVAL, which means that sync-ing is
5806 not supported on the file descriptor, is now ignored when Exim is trying to
5807 sync a directory. This applies only to Linux.
5809 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
5811 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
5812 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
5815 29. Added support for maildirsize files from supplied patch (modified a bit).
5817 30. The use of :fail: followed by an empty string could lead Exim to respond to
5818 sender verification failures with (e.g.):
5820 550 Verification failed for <xxx>
5821 550 Sender verify failed
5823 where the first response line was missing the '-' that indicates it is not
5824 the final line of the response.
5826 31. The loop for finding the name of the user that called Exim had a hardwired
5827 limit of 10; it now uses the value of finduser_retries, which is used for
5828 all other user lookups.
5830 32. Added $received_count variable, available in data and not_smtp ACLs, and at
5833 33. Exim was neglecting to zero errno before one call of strtol() when
5834 expanding a string and expecting an integer value. On some systems this
5835 resulted in spurious "integer overflow" errors. Also, it was casting the
5836 result into an int without checking.
5838 34. Testing for a connection timeout using "timeout_connect" in the retry rules
5839 did not work. The code looks as if it has *never* worked, though it appears
5840 to have been documented since at least releast 1.62. I have made it work.
5842 35. The "timeout_DNS" error in retry rules, also documented since at least
5843 1.62, also never worked. As it isn't clear exactly what this means, and
5844 clearly it isn't a major issue, I have abolished the feature by treating it
5845 as "timeout", and writing a warning to the main and panic logs.
5847 36. The display of retry rules for -brt wasn't always showing the error code
5850 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
5851 timeout_connect_A, timeout_connect_MX.
5853 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
5854 to the empty sender.
5856 39. The daemon was not analysing the content of -oX till after it had closed
5857 stderr and disconnected from the controlling terminal. This meant that any
5858 syntax errors were only noted on the panic log, and the return code from
5859 the command was 0. By re-arranging the code a little, I've made the
5860 decoding happen first, so such errors now appear on stderr, and the return
5861 code is 1. However, the actual setting up of the sockets still happens in
5862 the disconnected process, so errors there are still only recorded on the
5865 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
5866 connections (as happens on some IP stacks) was logged at start up time as
5867 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
5868 it from "IPv6 and IPv4", which means that two separate sockets are being
5871 41. The debug output for gethostbyname2() or getipnodebyname() failures now
5872 says whether AF_INET or AF_INET6 was passed as an argument.
5874 42. Exiwhat output was messed up when time zones were included in log
5877 43. Exiwhat now gives more information about the daemon's listening ports,
5878 and whether -tls-on-connect was used.
5880 44. The "port" option of the smtp transport is now expanded.
5882 45. A "message" modifier in a "warn" statement in a non-message ACL was being
5883 silently ignored. Now an error message is written to the main and panic
5886 46. There's a new ACL modifier called "logwrite" which writes to a log file
5887 as soon as it is encountered.
5889 47. Added $local_user_uid and $local_user_gid at routing time.
5891 48. Exim crashed when trying to verify a sender address that was being
5894 49. Exim was recognizing only a space character after ".include". It now also
5895 recognizes a tab character.
5897 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
5898 extracting the relevant information from the specification. The man page no
5899 longer contains scrambled data for the -d option, and I've added a section
5900 at the front about calling Exim under different names.
5902 51. Added "extra_headers" argument to the "mail" command in filter files.
5904 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
5907 53. Installed eximstats 1.29.
5909 54. Added transport_filter_timeout as a generic transport option.
5911 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
5912 Cc: header lines. This was required by RFC 822, but it not required by RFC
5915 56. Exim used to add From:, Date:, and Message-Id: header lines to any
5916 incoming messages that did not have them. Now it does so only if the
5917 message originates locally, that is, if there is no associated remote host
5918 address. When Resent- header lines are present, this applies to the Resent-
5919 lines rather than the non-Resent- lines.
5921 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
5922 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
5924 58. Messages for configuration errors now include the name of the main
5925 configuration file - useful now that there may be more than one file in a
5926 list (.included file names were always shown).
5928 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
5929 for those rare installations that do not start the daemon as root or run it
5930 setuid root. I've cut out the call to initgroups() if the daemon is not
5933 60. The Exim user and group can now be bound into the binary as text strings
5934 that are looked up at the start of Exim's processing.
5936 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
5938 62. Added $mailstore_basename variable.
5940 63. Installed patch to sieve.c from Michael Haardt.
5942 64. When Exim failed to open the panic log after failing to open the main log,
5943 the original message it was trying to log was written to stderr and debug
5944 output, but if they were not available (the usual case in production), it
5945 was lost. Now it is written to syslog before the two lines that record the
5946 failures to open the logs.
5948 65. Users' Exim filters run in subprocesses under the user's uid. It is
5949 possible for a "deliver" command or an alias in a "personal" command to
5950 provoke an address rewrite. If logging of address rewriting is configured,
5951 this fails because the process is not running as root or exim. There may be
5952 a better way of dealing with this, but for the moment (because 4.30 needs
5953 to be released), I have disabled address rewrite logging when running a
5954 filter in a non-root, non-exim process.
5960 1. The buildconfig auxiliary program wasn't quoting the value set for
5961 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
5962 not defined. This bug was masked in 4.22 by the effect that was fixed in
5965 2. Some messages that were rejected after a message id was allocated were
5966 shown as "incomplete" by exigrep. It no longer does this for messages that
5967 are rejected by local_scan() or the DATA or non-SMTP ACLs.
5969 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
5970 have allow_domain_literals set, the ID did not get logged in the <= line.
5971 Domain literals are now always recognized in Message-ID: header lines.
5973 4. The first argument for a ${extract expansion item is the key name or field
5974 number. Leading and trailing spaces in this item were not being ignored,
5975 causing some misleading effects.
5977 5. When deliver_drop_privilege was set, single queue runner processes started
5978 manually (i.e. by the command "exim -q") or by the daemon (which uses the
5979 same command in the process it spins off) were not dropping privilege.
5981 6. When the daemon running as "exim" started a queue runner, it always
5982 re-executed Exim in the spun-off process. This is a waste of effort when
5983 deliver_drop_privilege is set. The new process now just calls the
5984 queue-runner function directly.
5990 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
5993 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
5994 ignored. Though the use of -oP was forcing the writing of a pid file, it
5995 was always written to the default place.
5997 3. If the message "no IP address found for host xxxx" is generated during
5998 incoming verification, it is now followed by identification of the incoming
5999 connection (so you can more easily find what provoked it).
6001 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
6003 5. Added some features to "harden" Exim a bit more against certain attacks:
6005 (a) There is now a build-time option called FIXED_NEVER_USERS that can
6006 be put in Local/Makefile. This is like the never_users runtime option,
6007 but it cannot be overridden. The default setting is "root".
6009 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
6010 prefix string with which any file named in a -C command line option
6013 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
6014 is retained for -C and -D only if the caller of Exim is root. Without
6015 it, the exim user may also use -C and -D and retain privilege.
6017 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
6018 command line option is disabled.
6020 6. Macro names set by the -D option must start with an upper case letter, just
6021 like macro names defined in the configuration file.
6023 7. Added "dereference=" facility to LDAP.
6025 8. Two instances of the typo "uknown" in the source files are fixed.
6027 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
6028 the Configure-Makefile script screwed up while processing it.
6030 10. Incorporated PCRE 4.4.
6032 11. The SMTP synchronization check was not operating right at the start of an
6033 SMTP session. For example, it could not catch a HELO sent before the client
6034 waited for the greeting. There is now a check for outstanding input at the
6035 point when the greeting is written. Because of the duplex, asynchronous
6036 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
6037 way, but not yet received, when the check is performed.
6039 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
6040 on TCP/IP sockets, because this apparently causes some broken clients to
6043 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
6044 unchanged) from the Cygwin maintainer.
6046 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
6047 was supported instead of testing for mbx. Effectively a typo.
6049 15. The spa authenticator server code was not checking that the input it
6050 received was valid base64.
6052 16. The debug output line for the "set" modifier in ACLs was not showing the
6053 name of the variable that was being set.
6055 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
6057 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
6058 lookup on its own. Something else was needed to trigger the lookup. For
6059 example, a match in host_lookup or the need for a host name in a host list.
6060 Now, if $sender_host_name is referenced and the host name has not yet been
6061 looked up, a lookup is performed. If the lookup fails, the variable remains
6062 empty, and $host_lookup_failed is set to "1".
6064 19. Added "eqi" as a case-independent comparison operator.
6066 20. The saslauthd authentication condition could segfault if neither service
6067 nor realm was specified.
6069 21. If an overflowing value such as "2048M" was set for message_size_limit, the
6070 error message that was logged was misleading, and incoming SMTP
6071 connections were dropped. The message is now more accurate, and temporary
6072 errors are given to SMTP connections.
6074 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
6075 (except RSET) with a 421 error, until QUIT is received. However, it was
6076 failing to send a response to QUIT.
6078 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
6079 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
6080 ACL is now run after the helo_try_verify_hosts code.
6082 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
6083 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
6084 case-independent, so other case variants are also recognized). Apparently
6085 some systems use these upper case variants.
6087 25. If more than two messages were waiting for the same host, and a transport
6088 filter was specified for the transport, Exim sent two messages over the
6089 same TCP/IP connection, and then failed with "socket operation on non-
6090 socket" when it tried to send the third.
6092 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
6094 27. The extern definition of crypt16() in expand.c was not being excluded when
6095 the OS had its own crypt16() function.
6097 28. Added bounce_return_body as a new option, and bounce_return_size_limit
6098 as a preferred synonym for return_size_limit, both as an option and as an
6101 29. Added LIBS=-liconv to OS/Makefile-OSF1.
6103 30. Changed the default configuration ACL to relax the local part checking rule
6104 for addresses that are not in any local domains. For these addresses,
6105 slashes and pipe symbols are allowed within local parts, but the sequence
6106 /../ is explicitly forbidden.
6108 31. SPA server authentication was not clearing the challenge buffer before
6111 32. log_message in a "warn" ACL statement was writing to the reject log as
6112 well as to the main log, which contradicts the documentation and doesn't
6113 seem right (because no rejection is happening). So I have stopped it.
6115 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
6116 However, I am unable to do any testing of this.
6118 34. Fixed an infelicity in the appendfile transport. When checking directories
6119 for a mailbox, to see if any needed to be created, it was accidentally
6120 using path names with one or more superfluous leading slashes; tracing
6121 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
6123 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
6124 contents are added to the log line that is written for every discarded
6125 recipient. (Previously a log_message setting was ignored.)
6127 36. The ${quote: operator now quotes the string if it is empty.
6129 37. The install script runs exim in order to find its version number. If for
6130 some reason other than non-existence or emptiness, which it checks, it
6131 could not run './exim', it was installing it with an empty version number,
6132 i.e. as "exim-". This error state is now caught, and the installation is
6135 38. An argument was missing from the function that creates an error message
6136 when Exim fails to connect to the socket for saslauthd authentication.
6137 This could cause Exim to crash, or give a corrupted message.
6139 39. Added isip, isip4, and isip6 to ${if conditions.
6141 40. The ACL variables $acl_xx are now saved with the message, and can be
6142 accessed later in routers, transports, and filters.
6144 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
6145 strings in the file are not string-expanded.
6147 42. If a MAIL command specified a SIZE value that was too large to fit into an
6148 int variable, the check against message_size_limit failed. Such values are
6149 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
6150 day this will have to be increased, but I don't think I want to be around
6151 when emails are that large.
6158 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
6159 iconv() is not standard in FreeBSD.
6161 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
6162 IPv6 enabled. The observed symptom was a segmentation fault on return from
6163 the function os_common_find_running_interfaces() in src/os.c.
6165 3. In the check_special_case() function in daemon.c I had used "errno" as an
6166 argument name, which causes warnings on some systems. This was basically a
6167 typo, since it was named "eno" in the comments!
6169 4. The code that waits for the clock to tick (at a resolution of some fraction
6170 of a second) so as to ensure message-id uniqueness was always waiting for
6171 at least one whole tick, when it could have waited for less. [This is
6172 almost certainly not relevant at current processor speeds, where it is
6173 unlikely to ever wait at all. But we try to future-proof.]
6175 5. The function that sleeps for a time interval that includes fractions of a
6176 second contained a race. It did not block SIGALRM between setting the
6177 timer, and suspending (a couple of lines later). If the interval was short
6178 and the sigsuspend() was delayed until after it had expired, the suspension
6179 never ended. On busy systems this could lead to processes getting stuck for
6182 6. Some uncommon configurations may cause a lookup to happen in a queue runner
6183 process, before it forks any delivery processes. The open lookup caching
6184 mechanism meant that the open file or database connection was passed into
6185 the delivery process. The problem was that delivery processes always tidy
6186 up cached lookup data. This could cause a problem for the next delivery
6187 process started by the queue runner, because the external queue runner
6188 process does not know about the closure. So the next delivery process
6189 still has data in the lookup cache. In the case of a file lookup, there was
6190 no problem because closing a file descriptor in a subprocess doesn't affect
6191 the parent. However, if the lookup was caching a connection to a database,
6192 the connection was closed, and the second delivery process was likely to
6193 see errors such as "PGSQL: query failed: server closed the connection
6194 unexpectedly". The problem has been fixed by closing all cached lookups
6195 in a queue runner before running a delivery process.
6197 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
6198 seem to have the "const" qualifier which it has on other OS. I've
6201 8. Change 4.21/2 was too strict. It is only if there are two authenticators
6202 *of the same type* (client or server) with the same public name that an
6203 error should be diagnosed.
6205 9. When Exim looked up a host name for an IP address, but failed to find the
6206 original IP address when looking up the host name (a safety check), it
6207 output the message "<ip address> does not match any IP for NULL", which was
6208 confusing, to say the least. The bug was that the host name should have
6209 appeared instead of "NULL".
6211 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
6212 user that is built into the binary, and the -C or -D options is used, root
6213 privilege is dropped before the configuration file is read. In addition,
6214 logging is switched to stderr instead of the normal log files. If the
6215 configuration then re-defines the Exim user, the unprivileged environment
6216 is probably not what is expected, so Exim logs a panic warning message (but
6219 However, if deliver_drop_privilege is set, the unprivileged state may well
6220 be exactly what is intended, so the warning has been cut out in that case,
6221 and Exim is allowed to try to write to its normal log files.
6227 1. smtp_return_error_details was not giving details for temporary sender
6228 or receiver verification errors.
6230 2. Diagnose a configuration error if two authenticators have the same public
6233 3. Exim used not to create the message log file for a message until the first
6234 delivery attempt. This could be confusing when incoming messages were held
6235 for policy or load reasons. The message log file is now created at the time
6236 the message is received, and an initial "Received" line is written to it.
6238 4. The automatically generated man page for command line options had a minor
6239 bug that caused no ill effects; however, a more serious problem was that
6240 the procedure for building the man page automatically didn't always
6241 operate. Consequently, release 4.20 contains an out-of-date version. This
6242 shouldn't happen again.
6244 5. When building Exim with embedded Perl support, the script that builds the
6245 Makefile was calling 'perl' to find its compile-time parameters, ignoring
6246 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
6248 6. The freeze_tell option was not being used for messages that were frozen on
6249 arrival, either by an ACL or by local_scan().
6251 7. Added the smtp_incomplete_transaction log selector.
6253 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
6254 was accepting AUTH without a new EHLO.
6256 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
6257 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
6260 10. Logging of TCP/IP connections (when configured) now happens in the main
6261 daemon process instead of the child process, so that the TCP/IP connection
6262 count is more accurate (but it can never be perfect).
6264 11. The use of "drop" in a nested ACL was not being handled correctly in the
6265 outer ACL. Now, if condition failure induced by the nested "drop" causes
6266 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
6267 or "require"), the connection is dropped.
6269 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
6270 that yield "discard" can now be used with an "accept" or a "discard" verb,
6271 but an error is generated for any others (because I can't see a useful way
6272 to define what should happen).
6274 13. When an ACL is read dynamically from a file (or anywhere else), the lines
6275 are now processed in the same way as lines in the Exim configuration file.
6276 In particular, continuation lines are supported.
6278 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
6280 15. Added -ti meaning -t -i.
6282 16. Check for letters, digits, hyphens, and dots in the names of dnslist
6283 domains, and warn by logging if others are found.
6285 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
6286 returned from GIFCONF when Exim is trying to find the list of interfaces on
6287 a host. The code in os.c has been modified to copy each ifreq to an aligned
6288 structure in all cases.
6290 Also, in some cases, the returned ifreq's were being copied to a 'struct
6291 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
6292 means the last couple of bytes of an IPv6 address could be chopped if the
6293 ifreq contained only a normal sockaddr (14 bytes storage).
6295 18. Named domain lists were not supported in the hosts_treat_as_local option.
6296 An entry such as +xxxx was not recognized, and was treated as a literal
6299 19. Ensure that header lines added by a DATA ACL are included in the reject log
6300 if the ACL subsequently rejects the message.
6302 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
6303 MD5 (which is deprecated).
6305 21. When testing a filter file using -bf, Exim was writing a message when it
6306 took the sender from a "From " line in the message, but it was not doing so
6307 when it took $return_path from a Return-Path: header line. It now does.
6309 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
6310 with a valid header line field name (a series of printing characters
6311 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
6313 23. Changed "disc" in the source to "disk" to conform to the documentation and
6314 the book and for uniformity.
6316 24. Ignore Sendmail's -Ooption=value command line item.
6318 25. When execve() failed while trying to run a command in a pipe transport,
6319 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
6320 could be confused with a return value of 69 from the command itself. This
6321 has been changed to 127, the value the shell returns if it is asked to run
6322 a non-existent command. The wording for the related log line suggests a
6323 non-existent command as the problem.
6325 26. If received_header_text expands to an empty string, do not add a Received:
6326 header line to the message. (Well, it adds a token one on the spool, but
6327 marks it "old" so that it doesn't get used or transmitted.)
6329 27. Installed eximstats 1.28 (addition of -nt option).
6331 28. There was no check for failure on the call to getsockname() in the daemon
6332 code. This can fail if there is a shortage of resources on the system, with
6333 ENOMEM, for example. A temporary error is now given on failure.
6335 29. Contrary to the C standard, it seems that in some environments, the
6336 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
6337 program. Exim now does this explicitly; it affects the formatting of
6338 timestamps using strftime().
6340 30. If exiqsumm was given junk data, it threw up some uninitialized variable
6341 complaints. I've now initialized all the variables, to avoid this.
6343 32. Header lines added by a system filter were not being "seen" during
6344 transport-time rewrites.
6346 33. The info_callback() function passed to OpenSSL is set up with type void
6347 (*)(SSL *, int, int), as described somewhere. However, when calling the
6348 function (actually a macro) that sets it up, the type void(*)() is
6349 expected. I've put in a cast to prevent warnings from picky compilers.
6351 34. If a DNS black list lookup found a CNAME record, but there were no A
6352 records associated with the domain it pointed at, Exim crashed.
6354 35. If a DNS black list lookup returned more than one A record, Exim ignored
6355 all but the first. It now scans all returned addresses if a particular IP
6356 value is being sought. In this situation, the contents of the
6357 $dnslist_value variable are a list of all the addresses, separated by a
6360 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
6361 to accept a host name and all its aliases if the forward lookup for any of
6362 them yielded the IP address of the incoming connection. Now it accepts only
6363 those names whose forward lookup yields the correct IP address. Any other
6364 names are discarded. This closes a loophole whereby a rogue DNS
6365 administrator could create reverse DNS records to break through a
6366 wildcarded host restriction in an ACL.
6368 37. If a user filter or a system filter that ran in a subprocess used any of
6369 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
6370 the wrong values were passed to the pipe command ($thisaddress had the
6371 value of $0, $0 had the value of $1, etc). This bug was introduced by
6372 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
6374 38. Improved the line breaking for long SMTP error messages from ACLs.
6375 Previously, if there was no break point between 40 and 75 characters, Exim
6376 left the rest of the message alone. Two changes have been made: (a) I've
6377 reduced the minimum length to 35 characters; (b) if it can't find a break
6378 point between 35 and 75 characters, it looks ahead and uses the first one
6379 that it finds. This may give the occasional overlong line, but at least the
6380 remaining text gets split now.
6382 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
6383 file descriptors might be low, and that setting 1000 would always raise it.
6384 It turns out that in some environments, the limit is already over 1000 and
6385 that lowering it causes trouble. So now Exim takes care not to decrease it.
6387 40. When delivering a message, the value of $return_path is set to $sender_
6388 address at the start of routing (routers may change the value). By an
6389 oversight, this default was not being set up when an address was tested by
6390 -bt or -bv, which affected the outcome if any router or filter referred to
6393 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
6394 log only when "message" or "log_message" are set. However, if one of the
6395 conditions was an address verification, or a call to a nested ACL, the
6396 messages generated by the underlying test were being passed through. This
6397 no longer happens. The underlying message is available in $acl_verify_
6398 message for both "message" and "log_message" expansions, so it can be
6399 passed through if needed.
6401 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
6402 new expansion $bh_ to give the encoded byte string without charset
6403 translation. Translation happens only if iconv() is available; HAVE_ICONV
6404 indicates this at build time. HEADERS_CHARSET gives the charset to
6405 translate to; headers_charset can change it in the configuration, and
6406 "headers charset" can change it in an individual filter file.
6408 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
6409 that creates RFC 2047 encoded "words" labels them as that charset instead
6410 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
6411 expansion operator; (ii) when Exim creates a From: line for a local
6412 message; (iii) when a header line is rewritten to include a "phrase" part.
6414 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
6415 buggy, causing it to skip the first lines of messages whose message ID
6416 ended in 'D'. This would not have bitten before Exim release 4.14, because
6417 message IDs were unlikely to end in 'D' before then. The effect was to have
6418 incorrect size information for certain domains.
6420 45. #include "config.h" was missing at the start of the crypt16.c module. This
6421 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
6424 46. If there was a timeout during a "random" callout check, Exim treated it as
6425 a failure of the random address, and carried on sending RSET and the real
6426 address. If the delay was just some slowness somewhere, the response to the
6427 original RCPT would be taken as a response to RSET and so on, causing
6428 mayhem of various kinds.
6430 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
6431 when I implemented it. It didn't allow for the fact that some option values
6432 may legitimatetly be negative (e.g. size_addition), and it didn't even do
6433 the right test for positive values.
6435 48. Domain names in DNS records are case-independent. Exim always looks them up
6436 in lower case. Some resolvers return domain names in exactly the case they
6437 appear in the zone file, that is, they may contain uppercase letters. Not
6438 all resolvers do this - some return always lower case. Exim was treating a
6439 change of case by a resolver as a change of domain, similar to a widening
6440 of a domain abbreviation. This triggered its re-routing code and so it was
6441 trying to route what was effectively the same domain again. This normally
6442 caused routing to fail (because the router wouldn't handle the domain
6443 twice). Now Exim checks for this case specially, and just changes the
6444 casing of the domain that it ultimately uses when it transmits the message
6447 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
6450 50. If a filter generated a file delivery with a non-absolute name (possible if
6451 no home directory exists for the router), the forbid_file option was not
6454 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
6455 the existing equality matching.
6457 52. Exim was using ints instead of ino_t variables in some places where it was
6458 dealing with inode numbers.
6460 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
6461 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
6462 called TMPDIR, and if it finds it is different, it changes its value.
6464 54. The smtp_printf() function is now made available to local_scan() so
6465 additional output lines can be written before returning. There is also an
6466 smtp_fflush() function to enable the detection of a dropped connection.
6467 The variables smtp_input and smtp_batched_input are exported to
6470 55. Changed the default runtime configuration: the message "Unknown user"
6471 has been removed from the ACL, and instead placed on the localuser router,
6472 using the cannot_route_message feature. This means that any verification
6473 failures that generate their own messages won't get overridden. Similarly,
6474 the "Unrouteable address" message that was in the ACL for unverifiable
6475 relay addresses has also been removed.
6477 56. Added hosts_avoid_esmtp to the smtp transport.
6479 57. The exicyclog script was not checking for the esoteric option
6480 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
6481 will work only if exicyclog is run under the appropriate euid.
6483 58. Following a discussion on the list, the rules by which Exim recognises line
6484 endings on incoming messages have been changed. The -dropcr and drop_cr
6485 options are now no-ops, retained only for backwards compatibility. The
6486 following line terminators are recognized: LF CRLF CR. However, special
6487 processing applies to CR:
6489 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
6490 nor a local message in the state where . is a terminator.
6492 (ii) If a bare CR is encountered in a header line, an extra space is added
6493 after the line terminator so as not to end the header. The reasoning
6494 behind this is that bare CRs in header lines are most likely either
6495 to be mistakes, or people trying to play silly games.
6497 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
6498 was being incorrectly given as 18 bytes larger than it should have been.
6499 This is a VOB (very old bug).
6501 60. This may never have affected anything current, but just in case it has:
6502 When the local host is found other than at the start of a list of hosts,
6503 the local host, those with the same MX, and any that follow, are discarded.
6504 When the list in question was part of a longer list of hosts, the following
6505 hosts (not currently being processed) were also being discarded. This no
6506 longer happens. I'm not sure if this situation could ever has previously
6509 61. Added the "/MX" feature to lists of hosts in the manualroute and query
6512 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
6513 header. This is something that is recommended in a new Internet Draft, and
6514 is something that is documented as being done by Sendmail. There are two
6515 possible values. For messages generated by the autoreply transport, Exim
6518 Auto-Submitted: auto-replied
6520 whereas for all other generated messages (e.g. bounces) it adds
6522 Auto-Submitted: auto-generated
6524 63. The "personal" condition in filters now includes a test for the
6525 Auto-Submitted: header. If it contains the string "auto-" the message it
6526 not considered personal.
6528 64. Added rcpt_include_affixes as a generic transport option.
6530 65. Added queue_only_override (default true).
6532 66. Added the syslog_duplication option.
6534 67. If what should have been the first header line of a message consisted of
6535 a space followed by a colon, Exim was mis-interpreting it as a header line.
6536 It isn't of course - it is syntactically invalid and should therefore be
6537 treated as the start of the message body. The misbehaviour could have
6538 caused a number of strange effects, including loss of data in subsequent
6539 header lines, and spool format errors.
6541 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
6542 client host had authenticated. This control can now be exercised by an ACL
6543 for more flexibility.
6545 69. By default, callouts do not happen when testing with -bh. There is now a
6546 variant, -bhc, which does actually run the callout code, including
6547 consulting and updating the callout cache.
6549 70. Added support for saslauthd authentication, courtesy of Alexander
6552 71. If statvfs() failed on the spool or log directories while checking their
6553 size for availability, Exim confusingly gave the error "space shortage".
6554 Furthermore, in debugging mode it crashed with a floating point exception.
6555 These checks are done if check_{spool,log}_{space,inodes} are set, and when
6556 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
6557 serious problem, Exim now writes to the main and panic logs when this
6558 happens, with details of the failure. It then refuses to accept the
6559 incoming message, giving the message "spool directory problem" or "log
6560 directory problem" with a 421 code for SMTP messages.
6562 72. When Exim is about to re-exec itself, it ensures that the file descriptors
6563 0, 1, and 2 exist, because some OS complain for execs without them (see
6564 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
6565 descriptors. However, the code omitted to check that the open succeeded,
6566 causing mysterious errors if for some reason the permissions on /dev/null
6567 got screwed. Now Exim writes a message to the main and panic logs, and
6568 bombs out if it can't open /dev/null.
6570 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
6571 interact so that it is all more flexible. It is supposed to remain
6572 backwards compatible. Also added extra_local_interfaces.
6574 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
6575 to bomb out with an assertion failure - to the client this appears as a
6576 connection drop. This problem occurs in the part of the code that was taken
6577 from the Samba project. Fortunately, the assertion is in a very simple
6578 function, so I have fixed this by reproducing the function inline in the
6579 one place where it is called, and arranging for authentication to fail
6580 instead of killing the process with assert().
6582 75. The SPA client code was not working when the server requested OEM rather
6583 than Unicode encoding.
6585 76. Added code to make require_files with a specific uid setting more usable in
6586 the case where statting the file as root fails - usually a non-root-mounted
6587 NFS file system. When this happens and the failure is EACCES, Exim now
6588 forks a subprocess and does the per-uid checking as the relevant uid.
6590 77. Added process_log_path.
6592 78. If log_file_path was not explicitly set, a setting of check_log_space or
6593 check_log_inodes was ignored.
6595 79. If a space check for the spool or log partitions fails, the incident is now
6596 logged. Of course, in the latter case the data may get lost...
6598 80. Added the %p formatting code to string_format() so that it can be used to
6599 print addresses in debug_print(). Adjusted all the address printing in the
6600 debugging in store.c to use %p rather than %d.
6602 81. There was a concern that a line of code in smtp_in.c could overflow a
6603 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
6604 initially expressed, the concern was not well-founded, because trailing
6605 spaces are removed early. However, if the trailing spaces were followed by
6606 a NULL, they did not get removed, so the overflow was possible. Two fixes
6609 (a) I re-wrote the offending code in a cleaner fashion.
6610 (b) If an incoming SMTP command contains a NULL character, it is rejected
6613 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
6614 runs initgroups(), so that if the Exim user is in any additional groups,
6615 they will be used during message reception.
6621 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / users / heiko / exim.git / blob