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.
94 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
95 and 3.4.4 - once the server is enabled to respond to an OCSP request
96 it does even when not requested, resulting in a stapling non-aware
97 client dropping the TLS connection.
99 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
100 support variable-length bit vectors. No functional change.
102 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
103 The I= interface field on outgoing lines is now after the H= remote
104 host field, same as incoming lines. There is a separate
105 outgoing_interface log selector which allows you to disable the
108 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
109 If not running log_selector +smtp_connection the mainlog would be held
110 open indefinitely after a "too many connections" event, including to a
111 deleted file after a log rotate. Leave the per net connection logging
112 leaving it open for efficiency as that will be quickly detected by the
113 check on the next write.
115 HS/01 Bug 1671: Fix post transport crash.
116 Processing the wait-<transport> messages could crash the delivery
117 process if the message IDs didn't exist for some reason. When
118 using 'split_spool_directory=yes' the construction of the spool
119 file name failed already, exposing the same netto behaviour.
121 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
122 mime_regex ACL conditions.
124 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
125 to DSN fail messages (bounces): remote IP, remote greeting, remote response
126 to HELO, local diagnostic string.
128 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
129 log line to debug. Even when configured with a tls authenticator many
130 client connections are expected to not authenticate in this way, so
131 an authenticate fail is not an error.
133 HS/02 Add the Exim version string to the process info. This way exiwhat
134 gives some more detail about the running daemon.
136 JH/06 Bug 1395: time-limit cacheing of DNS lookups, to the TTL value. This may
137 matter for fast-change records such as DNSBLs.
139 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
140 retry record, even if constant. There may be multiple transports with
141 different interface settings and the retry behaviour needs to be kept
144 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
146 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
148 JH/10 Bug 840: fix log_defer_output option of pipe transport
150 JH/11 Bug 830: use same host for all RCPTS of a message, even under
151 hosts_randomize. This matters a lot when combined with mua_wrapper.
153 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
154 ${quote_pgsql:<string>} operator.
156 JH/13 Bug 1708: avoid misaligned access in cached lookup.
158 JH/14 Change header file name for freeradius-client. Relevant if compiling
159 with Radius support; from the Gentoo tree and checked under Fedora.
161 JH/15 Bug 1712: Introduce $prdr_requested flag variable
163 JH/16 Bug 1714: Permit an empty string as expansion result for transport
164 option transport_filter, meaning no filtering.
166 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
168 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
169 defaults to "*" (all hosts). The variable is now available when not built
170 with TLS, default unset, mainly to enable keeping the testuite sane.
171 If a server certificate is not supplied (via tls_certificate) an error is
172 logged, and clients will find TLS connections fail on startup. Presumably
173 they will retry in-clear.
174 Packagers of Exim are strongly encouraged to create a server certificate
175 at installation time.
177 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
178 with the $config_file variable.
180 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
181 in transport context, after the attempt, and per-recipient. The latter type
182 is per host attempted. The event data is the error message, and the errno
183 information encodes the lookup type (A vs. MX) used for the (first) host,
184 and the trailing two digits of the smtp 4xx reponse.
186 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
187 to write to mainlog (or rejectlog, paniclog) in the window between file
188 creation and permissions/ownership being changed. Particularly affects
189 installations where exicyclog is run as root, rather than exim user;
190 result is that the running daemon panics and dies.
192 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
194 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
195 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
196 "pri" and "weight". Note that the previous implicit priority given by the
197 list order is no longer honoured.
199 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalisation
202 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
203 by defining SUPPORT_SOCKS.
205 JH/26 Move PROXY support from Experimental to mainline, enabled for a build
206 by defining SUPPORT_PROXY. Note that the proxy_required_hosts option
207 is renamed to hosts_proxy, and the proxy_{host,target}_{address,port}.
208 variables are renamed to proxy_{local,external}_{address,port}.
210 JH/27 Move Internationalisation support from Experimental to mainline, enabled
211 for a build by defining SUPPORT_I18N
213 JH/28 Bug 1745: Fix redis lookups to handle (quoted) spaces embedded in parts
214 of the query string, and make ${quote_redis:} do that quoting.
216 JH/29 Move Events support from Experimental to mainline, enabled by default
217 and removable for a build by defining DISABLE_EVENT.
219 JH/30 Updated DANE implementation code to current from Viktor Dukhovni.
221 JH/31 Fix bug with hosts_connection_nolog and named-lists which were wrongly
222 cached by the daemon.
224 JH/32 Move Redis support from Experimental to mainline, enabled for a build
225 by defining LOOKUP_REDIS. The libhiredis library is required.
227 JH/33 Bug 1748: Permit ACL dnslists= condition in non-smtp ACLs if explicit
228 keys are given for lookup.
230 JH/34 Bug 1192: replace the embedded copy of PolarSSL RSA routines in the DKIM
231 support, by using OpenSSL or GnuTLS library ones. This means DKIM is
232 only supported when built with TLS support. The PolarSSL SHA routines
233 are still used when the TLS library is too old for convenient support.
235 JH/35 Require SINGLE_DH_USE by default in OpenSSL (main config option
236 openssl_options), for security. OpenSSL forces this from version 1.1.0
237 server-side so match that on older versions.
239 JH/36 Bug 1778: longstanding bug in memory use by the ${run } expansion: A fresh
240 allocation for $value could be released as the expansion processing
241 concluded, but leaving the global pointer active for it.
243 JH/37 Bug 1769: Permit a VRFY ACL to override the default 252 response,
244 and to use the domains and local_parts ACL conditions.
246 JH/38 Fix cutthrough bug with body lines having a single dot. The dot was
247 incorrectly not doubled on cutthrough transmission, hence seen as a
248 body-termination at the receiving system - resulting in truncated mails.
249 Commonly the sender saw a TCP-level error, and retransmitted the message
250 via the normal store-and-forward channel. This could result in duplicates
251 received - but deduplicating mailstores were liable to retain only the
252 initial truncated version.
254 JH/39 Bug 1781: Fix use of DKIM private-keys having trailing '=' in the base-64.
256 JH/40 Fix crash in queryprogram router when compiled with EXPERIMENTAL_SRS.
258 JH/41 Bug 1792: Fix selection of headers to sign for DKIM: bottom-up. While
259 we're in there, support oversigning also; bug 1309.
261 JH/42 Bug 1796: Fix error logged on a malware scanner connection failure.
263 HS/04 Add support for keep_environment and add_environment options.
265 JH/43 Tidy coding issues detected by gcc --fsanitize=undefined. Some remain;
266 either intentional arithmetic overflow during PRNG, or testing config-
269 JH/44 Bug 1800: The combination of a -bhc commandline option and cutthrough
270 delivery resulted in actual delivery. Cancel cutthrough before DATA
273 JH/45 Fix cutthrough, when connection not opened by verify and target hard-
274 rejects a recipient: pass the reject to the originator.
276 JH/46 Multiple issues raised by Coverity. Some were obvious or plausible bugs.
277 Many were false-positives and ignorable, but it's worth fixing the
280 JH/47 Fix build on HP-UX and older Solaris, which need (un)setenv now also
281 for the new environment-manipulation done at startup. Move the routines
282 from being local to tls.c to being global via the os.c file.
284 JH/48 Bug 1807: Fix ${extract } for the numeric/3-string case. While preparsing
285 an extract embedded as result-arg for a map, the first arg for extract
286 is unavailable so we cannot tell if this is a numbered or keyed
287 extraction. Accept either.
293 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
296 JH/02 The smtp transport option "multi_domain" is now expanded.
298 JH/03 The smtp transport now requests PRDR by default, if the server offers
301 JH/04 Certificate name checking on server certificates, when exim is a client,
302 is now done by default. The transport option tls_verify_cert_hostnames
303 can be used to disable this per-host. The build option
304 EXPERIMENTAL_CERTNAMES is withdrawn.
306 JH/05 The value of the tls_verify_certificates smtp transport and main options
307 default to the word "system" to access the system default CA bundle.
308 For GnuTLS, only version 3.0.20 or later.
310 JH/06 Verification of the server certificate for a TLS connection is now tried
311 (but not required) by default. The verification status is now logged by
312 default, for both outbound TLS and client-certificate supplying inbound
315 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
318 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
319 Status Notification (bounce) messages are now MIME format per RFC 3464.
320 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
321 under the control of the dsn_advertise_hosts option, and routers may
322 have a dsn_lasthop option.
324 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
325 default, modifiable by a malware= option. The list separator for
326 the options can now be changed in the usual way. Bug 68.
328 JH/10 The smtp_receive_timeout main option is now expanded before use.
330 JH/11 The incoming_interface log option now also enables logging of the
331 local interface on delivery outgoing connections.
333 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
334 if the interface and destination host and port all match.
336 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
339 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
340 Patch from Andrew Lewis.
342 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
343 now supports optional time-restrictions, weighting, and priority
344 modifiers per server. Patch originally by <rommer@active.by>.
346 JH/16 The spamd_address main option now supports a mixed list of local
347 and remote servers. Remote servers can be IPv6 addresses, and
348 specify a port-range.
350 JH/17 Bug 68: The spamd_address main option now supports an optional
351 timeout value per server.
353 JH/18 Bug 1581: Router and transport options headers_add/remove can
354 now have the list separator specified.
356 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
359 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
362 JH/21 Support for the A6 type of dns record is withdrawn.
364 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
365 rather than the verbs used.
367 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
368 from 255 to 1024 chars.
370 JH/24 Verification callouts now attempt to use TLS by default.
372 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
373 are generic router options now. The defaults didn't change.
375 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
376 Original patch from Alexander Shikoff, worked over by JH.
378 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
379 files if ZCAT_COMMAND is not executable.
381 JH/26 Bug 1539: Add timout/retry options on dnsdb lookups.
383 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
385 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
386 Normally benign, it bites when the pair was led to by a CNAME;
387 modern usage is to not canoicalize the domain to a CNAME target
388 (and we were inconsistent anyway for A-only vs AAAA+A).
390 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
392 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
393 when evaluating $sender_host_dnssec.
395 JH/31 Check the HELO verification lookup for DNSSEC, adding new
396 $sender_helo_dnssec variable.
398 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
400 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
402 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
404 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
405 documented as working, but never had. Support all but $spam_report.
407 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
408 added for tls authenticator.
410 HS/03 Add perl_taintmode main config option
415 TL/01 When running the test suite, the README says that variables such as
416 no_msglog_check are global and can be placed anywhere in a specific
417 test's script, however it was observed that placement needed to be near
418 the beginning for it to behave that way. Changed the runtest perl
419 script to read through the entire script once to detect and set these
420 variables, reset to the beginning of the script, and then run through
421 the script parsing/test process like normal.
423 TL/02 The BSD's have an arc4random API. One of the functions to induce
424 adding randomness was arc4random_stir(), but it has been removed in
425 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
426 function when detected.
428 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
429 cause callback expansion.
431 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
432 syntax errors in an expansion can be treated as a string instead of
433 logging or causing an error, due to the internal use of bool_lax
434 instead of bool when processing it.
436 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
437 server certificates when making smtp deliveries.
439 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
441 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
443 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
445 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
446 Merged patch from Sebastian Wiedenroth.
448 JH/05 Fix results-pipe from transport process. Several recipients, combined
449 with certificate use, exposed issues where response data items split
450 over buffer boundaries were not parsed properly. This eventually
451 resulted in duplicates being sent. This issue only became common enough
452 to notice due to the introduction of conection certificate information,
453 the item size being so much larger. Found and fixed by Wolfgang Breyha.
455 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
456 size buffer was used, resulting in syntax errors when an expansion
459 JH/07 Add support for directories of certificates when compiled with a GnuTLS
460 version 3.3.6 or later.
462 JH/08 Rename the TPDA expermimental facility to Event Actions. The #ifdef
463 is EXPERIMENTAL_EVENT, the main-configuration and transport options
464 both become "event_action", the variables become $event_name, $event_data
465 and $event_defer_errno. There is a new variable $verify_mode, usable in
466 routers, transports and related events. The tls:cert event is now also
467 raised for inbound connections, if the main configuration event_action
470 TL/06 In test suite, disable OCSP for old versions of openssl which contained
471 early OCSP support, but no stapling (appears to be less than 1.0.0).
473 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
474 server certificate names available under the smtp transport option
475 "tls_verify_cert_hostname" now do not permit multi-component wildcard
478 JH/10 Time-related extraction expansions from certificates now use the main
479 option "timezone" setting for output formatting, and are consistent
480 between OpenSSL and GnuTLS compilations. Bug 1541.
482 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
483 encoded parameter in the incoming message. Bug 1558.
485 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
486 include certificate info, eximon was claiming there were spoolfile
489 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
491 JH/14 Log delivery-related information more consistently, using the sequence
492 "H=<name> [<ip>]" wherever possible.
494 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
495 are problematic for Debian distribution, omit them from the release
498 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
500 JH/16 Fix string representation of time values on 64bit time_t anchitectures.
503 JH/17 Fix a null-indirection in certextract expansions when a nondefault
504 output list separator was used.
509 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
510 checkers that were complaining about end of non-void function with no
513 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
514 This was a regression intruduced in 4.83 by another bugfix.
516 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
518 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
519 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
525 TF/01 Correctly close the server side of TLS when forking for delivery.
527 When a message was received over SMTP with TLS, Exim failed to clear up
528 the incoming connection properly after forking off the child process to
529 deliver the message. In some situations the subsequent outgoing
530 delivery connection happened to have the same fd number as the incoming
531 connection previously had. Exim would try to use TLS and fail, logging
532 a "Bad file descriptor" error.
534 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
535 utilities have not been installed.
537 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
538 temporary space as the ACL may create new global variables.
540 TL/01 LDAP support uses per connection or global context settings, depending
541 upon the detected version of the libraries at build time.
543 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
544 to extract and use the src ip:port in logging and expansions as if it
545 were a direct connection from the outside internet. PPv2 support was
546 updated based on HAProxy spec change in May 2014.
548 JH/02 Add ${listextract {number}{list}{success}{fail}}.
550 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
551 Properly escape header and check for NULL return.
553 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
556 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
558 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
559 characters in header names, implemented as a verify condition.
560 Contributed by Michael Fischer v. Mollard.
562 TL/05 Rename SPF condition results err_perm and err_temp to standardized
563 results permerror and temperror. Previous values are deprecated but
564 still accepted. In a future release, err_perm and err_temp will be
565 completely removed, which will be a backward incompatibility if the
566 ACL tests for either of these two old results. Patch contributed by
567 user bes-internal on the mailing list.
569 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
571 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
572 selectors, in both main and reject logs.
574 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
577 JH/07 Add malware type "sock" for talking to simple daemon.
579 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
581 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
582 routers/transports under cutthrough routing.
584 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
585 numbers. Touch up "bool" conditional to keep the same definition.
587 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
589 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
591 JH/12 Expand items in router/transport headers_add or headers_remove lists
592 individually rather than the list as a whole. Bug 1452.
594 Required for reasonable handling of multiple headers_ options when
595 they may be empty; requires that headers_remove items with embedded
596 colons must have them doubled (or the list-separator changed).
598 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
599 view the policy declared in the DMARC record. Currently, $dmarc_status
600 is a combined value of both the record presence and the result of the
603 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
605 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
606 dnslookup router and the smtp transport (applying to the forward
609 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
610 of ldap servers used for a specific lookup. Patch provided by Heiko
613 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
614 New variable $lookup_dnssec_authenticated for observability.
616 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
617 Patch submitted by Lars Timman.
619 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
621 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
622 Requires trusted mode and valid format message id, aborts otherwise.
623 Patch contributed by Heiko Schlichting.
625 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
626 certextract with support for various fields. Bug 1358.
628 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
629 is requested by default, modifiable by smtp transport option
632 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
633 operate on certificate variables to give certificate fingerprints
634 Also new ${sha256:cert_variable}.
636 JH/23 The PRDR feature is moved from being Experimental into the mainline.
638 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
639 Christian Aistleitner.
641 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
643 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
644 file. Patch from Wolfgang Breyha.
646 JH/25 Expand the coverage of the delivery $host and $host_address to
647 client authenticators run in verify callout. Bug 1476.
649 JH/26 Port service names are now accepted for tls_on_connect_ports, to
650 align with daemon_smtp_ports. Bug 72.
652 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
653 support and error reporting did not work properly.
655 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
656 and is readable. Patch from Andrew Colin Kissa.
658 TL/14 Enhance documentation of ${run expansion and how it parses the
659 commandline after expansion, particularly in the case when an
660 unquoted variable expansion results in an empty value.
662 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
664 PP/02 Fix internal collision of T_APL on systems which support RFC3123
665 by renaming away from it. Addresses GH issue 15, reported by
668 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
670 TL/15 SECURITY: prevent double expansion in math comparison functions
671 (can expand unsanitized data). Not remotely exploitable.
678 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
680 PP/02 Make -n do something, by making it not do something.
681 When combined with -bP, the name of an option is not output.
683 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
686 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
687 $sender_host_name and config options to manage this, and basic check
690 PP/05 DSCP support for outbound connections and control modifier for inbound.
692 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
693 (Only plugin which currently uses this is kerberos4, which nobody should
694 be using, but we should make it available and other future plugins might
695 conceivably use it, even though it would break NAT; stuff *should* be
696 using channel bindings instead).
698 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
699 name; added for Sendmail compatibility; requires admin caller.
700 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
701 just ignore it); requires trusted caller.
702 Also parse but ignore: -Ac -Am -X<logfile>
705 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
707 TL/02 Add +smtp_confirmation as a default logging option.
709 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
710 Patch by Magnus Holmgren from 2007-02-20.
712 TL/04 Bugzilla 1281 - Spec typo.
713 Bugzilla 1283 - Spec typo.
714 Bugzilla 1290 - Spec grammar fixes.
716 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
718 TL/06 Add Experimental DMARC support using libopendmarc libraries.
720 TL/07 Fix an out of order global option causing a segfault. Reported to dev
721 mailing list by by Dmitry Isaikin.
723 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
725 JH/02 Support "G" suffix to numbers in ${if comparisons.
727 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
729 NM/01 Bugzilla 1197 - Spec typo
730 Bugzilla 1196 - Spec examples corrections
732 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
734 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
735 gnutls_enable_pkcs11, but renamed to more accurately indicate its
738 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
739 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
741 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
742 "acl {{name}{arg}...}", and optional args on acl condition
745 JH/05 Permit multiple router/transport headers_add/remove lines.
747 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
749 JH/07 Avoid using a waiting database for a single-message-only transport.
750 Performance patch from Paul Fisher. Bugzilla 1262.
752 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
755 JH/09 Add $headers_added variable, with content from use of ACL modifier
756 add_header (but not yet added to the message). Bugzilla 199.
758 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
759 Pulled from Bugzilla 817 by Wolfgang Breyha.
761 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
763 (nb: this is the same fix as in Exim 4.80.1)
765 JH/11 Add A= logging on delivery lines, and a client_set_id option on
768 JH/12 Add optional authenticated_sender logging to A= and a log_selector
771 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
773 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
774 advertise SMTP AUTH mechanism to us, instead of a generic
775 protocol violation error. Also, make Exim more robust to bad
776 data from the Dovecot auth socket.
778 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
780 When a queue runner is handling a message, Exim first routes the
781 recipient addresses, during which it prunes them based on the retry
782 hints database. After that it attempts to deliver the message to
783 any remaining recipients. It then updates the hints database using
786 So if a recipient address works intermittently, it can get repeatedly
787 deferred at routing time. The retry hints record remains fresh so the
788 address never reaches the final cutoff time.
790 This is a fairly common occurrence when a user is bumping up against
791 their storage quota. Exim had some logic in its local delivery code
792 to deal with this. However it did not apply to per-recipient defers
793 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
795 This change adds a proper retry rule check during routing so that the
796 final cutoff time is checked against the message's age. We only do
797 this check if there is an address retry record and there is not a
798 domain retry record; this implies that previous attempts to handle
799 the address had the retry_use_local_parts option turned on. We use
800 this as an approximation for the destination being like a local
801 delivery, as in LMTP.
803 I suspect this new check makes the old local delivery cutoff check
804 redundant, but I have not verified this so I left the code in place.
806 TF/02 Correct gecos expansion when From: is a prefix of the username.
808 Test 0254 submits a message to Exim with the header
812 When I ran the test suite under the user fanf2, Exim expanded
813 the header to contain my full name, whereas it should have added
814 a Resent-Sender: header. It erroneously treats any prefix of the
815 username as equal to the username.
817 This change corrects that bug.
819 GF/01 DCC debug and logging tidyup
820 Error conditions log to paniclog rather than rejectlog.
821 Debug lines prefixed by "DCC: " to remove any ambiguity.
823 TF/03 Avoid unnecessary rebuilds of lookup-related code.
825 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
826 Bug spotted by Jeremy Harris; was flawed since initial commit.
827 Would have resulted in OCSP responses post-SNI triggering an Exim
828 NULL dereference and crash.
830 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
832 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
833 Bug detection, analysis and fix by Samuel Thibault.
834 Bugzilla 1331, Debian bug #698092.
836 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
838 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
839 Server implementation by Todd Lyons, client by JH.
840 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
841 config variable "prdr_enable" controls whether the server
842 advertises the facility. If the client requests PRDR a new
843 acl_data_smtp_prdr ACL is called once for each recipient, after
844 the body content is received and before the acl_smtp_data ACL.
845 The client is controlled by bolth of: a hosts_try_prdr option
846 on the smtp transport, and the server advertisement.
847 Default client logging of deliveries and rejections involving
848 PRDR are flagged with the string "PRDR".
850 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
851 fclose(). Diagnosis by Todd Lyons.
853 PP/17 Update configure.default to handle IPv6 localhost better.
854 Patch by Alain Williams (plus minor tweaks).
857 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
858 This is now consistent with GnuTLS, and is now documented: the
859 previous undocumented portable approach to treating the option as
860 unset was to force an expansion failure. That still works, and
861 an empty string is now equivalent.
863 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
864 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
865 not performing validation itself.
867 PP/20 Added force_command boolean option to pipe transport.
868 Patch from Nick Koston, of cPanel Inc.
870 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
873 TF/04 Added udpsend ACL modifer and hexquote expansion operator
875 PP/21 Fix eximon continuous updating with timestamped log-files.
876 Broken in a format-string cleanup in 4.80, missed when I repaired the
877 other false fix of the same issue.
878 Report and fix from Heiko Schlichting.
881 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
882 Report from Prashanth Katuri.
884 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
885 It's SecureTransport, so affects any MacOS clients which use the
886 system-integrated TLS libraries, including email clients.
888 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
889 using a MIME ACL for non-SMTP local injection.
890 Report and assistance in diagnosis by Warren Baker.
892 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
894 JH/16 Fix comparisons for 64b. Bugzilla 1385.
896 TL/09 Add expansion variable $authenticated_fail_id to keep track of
897 last id that failed so it may be referenced in subsequent ACL's.
899 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
902 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
903 ldap library initialization, allowing self-signed CA's to be
904 used. Also properly sets require_cert option later in code by
905 using NULL (global ldap config) instead of ldap handle (per
906 session). Bug diagnosis and testing by alxgomz.
908 TL/12 Enhanced documentation in the ratelimit.pl script provided in
909 the src/util/ subdirectory.
911 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
912 renamed to Transport Post Delivery Action by Jeremy Harris, as
915 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
916 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
917 redis_servers = needs to be configured which will be used by the redis
918 lookup. Patch from Warren Baker, of The Packet Hub.
920 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
922 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
923 hostname or reverse DNS when processing a host list. Used suggestions
924 from multiple comments on this bug.
926 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
928 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
929 Missed a few lines, added it to make the runtest require no keyboard
932 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
933 contains upper case chars. Make router use caseful_local_part.
935 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
936 support when GnuTLS has been built with p11-kit.
942 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
944 This, or similar/improved, will also be change PP/11 of 4.82.
950 PP/01 Handle short writes when writing local log-files.
951 In practice, only affects FreeBSD (8 onwards).
952 Bugzilla 1053, with thanks to Dmitry Isaikin.
954 NM/01 Bugzilla 949 - Documentation tweak
956 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
959 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
961 PP/02 Implemented gsasl authenticator.
963 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
965 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
966 `pkg-config foo` for cflags/libs.
968 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
969 with rest of GSASL and with heimdal_gssapi.
971 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
972 `pkg-config foo` for cflags/libs for the TLS implementation.
974 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
975 properties get this fed in as external SSF. A number of robustness
976 and debugging improvements to the cyrus_sasl authenticator.
978 PP/08 cyrus_sasl server now expands the server_realm option.
980 PP/09 Bugzilla 1214 - Log authentication information in reject log.
981 Patch by Jeremy Harris.
983 PP/10 Added dbmjz lookup type.
985 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
987 PP/12 MAIL args handles TAB as well as SP, for better interop with
988 non-compliant senders.
989 Analysis and variant patch by Todd Lyons.
991 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
992 Bug report from Lars Müller <lars@samba.org> (via SUSE),
993 Patch from Dirk Mueller <dmueller@suse.com>
995 PP/13 tls_peerdn now print-escaped for spool files.
996 Observed some $tls_peerdn in wild which contained \n, which resulted
997 in spool file corruption.
999 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
1000 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
1001 or write after TLS renegotiation, which otherwise led to messages
1004 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
1005 as a tracking header (ie: a signed header comes before the signature).
1006 Patch from Wolfgang Breyha.
1008 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
1009 comma-sep list; embedded commas doubled.
1011 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
1013 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
1015 Report and patch from Dmitry Banschikov.
1017 PP/16 Removed "dont_insert_empty_fragments" fron "openssl_options".
1018 Removed SSL_clear() after SSL_new() which led to protocol negotiation
1019 failures. We appear to now support TLS1.1+ with Exim.
1021 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
1022 lets Exim select keys and certificates based upon TLS SNI from client.
1023 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
1024 before an outbound SMTP session. New log_selector, +tls_sni.
1026 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
1027 NULL dereference. Report and patch from Alun Jones.
1029 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
1030 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
1031 Not seeing resolver debug output on NetBSD, but suspect this is a
1032 resolver implementation change.
1034 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
1035 Left warnings. Added "eximon gdb" invocation mode.
1037 PP/21 Defaulting "accept_8bitmime" to true, not false.
1039 PP/22 Added -bw for inetd wait mode support.
1041 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
1042 locate the relevant includes and libraries. Made this the default.
1044 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
1045 Bugzilla 1246, report and most of solution from Tomasz Kusy.
1047 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
1048 This may cause build issues on older platforms.
1050 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
1051 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
1052 gnutls_require_mac & gnutls_require_protocols (no longer supported).
1053 Added SNI support via GnuTLS too.
1054 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
1056 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
1058 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
1059 Applied second patch from Janne, implementing suggestion to default
1060 multiple-strings-in-record handling to match SPF spec.
1062 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
1064 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
1065 read-only, out of scope).
1066 Patch from Wolfgang Breyha, report from Stuart Northfield.
1068 PP/29 Fix three issues highlighted by clang analyser static analysis.
1069 Only crash-plausible issue would require the Cambridge-specific
1070 iplookup router and a misconfiguration.
1071 Report from Marcin Mirosław.
1073 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
1075 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
1076 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
1077 As part of this, removing so much warning spew let me fix some minor
1078 real issues in debug logging.
1080 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
1081 assignment on my part. Fixed.
1083 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
1084 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
1085 Janne Snabb (who went above and beyond: thank you).
1087 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
1088 string otherwise requires a connection and a bunch more work and it's
1089 relatively easy to get wrong. Should also expose TLS library linkage
1092 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
1093 64-bit ${eval} (JH/03).
1095 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
1096 GNU libc to support some of the 64-bit stuff, should not lead to
1097 conflicts. Defined before os.h is pulled in, so if a given platform
1098 needs to override this, it can.
1100 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
1101 protection layer was required, which is not implemented.
1102 Bugzilla 1254, patch from Wolfgang Breyha.
1104 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
1105 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
1106 tls_dhparam take prime identifiers. Also unbreak combination of
1107 OpenSSL+DH_params+TLSSNI.
1109 PP/39 Disable SSLv2 by default in OpenSSL support.
1115 PP/01 Solaris build fix for Oracle's LDAP libraries.
1116 Bugzilla 1109, patch from Stephen Usher.
1118 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
1120 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
1123 TF/02 Fix a couple more cases where we did not log the error message
1124 when unlink() failed. See also change 4.74-TF/03.
1126 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
1127 lock up or crash if it happened to be inside a call to libc when it
1128 got a SIGUSR1 from exiwhat.
1130 The SIGUSR1 handler appends the current process status to the process
1131 log which is later printed by exiwhat. It used to use the general
1132 purpose logging code to do this, but several functions it calls are
1133 not safe for signals.
1135 The new output code in the SIGUSR1 handler is specific to the process
1136 log, and simple enough that it's easy to inspect for signal safety.
1137 Removing some special cases also simplifies the general logging code.
1138 Removing the spurious timestamps from the process log simplifies
1141 TF/04 Improved ratelimit ACL condition.
1143 The /noupdate option has been deprecated in favour of /readonly which
1144 has clearer semantics. The /leaky, /strict, and /readonly update modes
1145 are mutually exclusive. The update mode is no longer included in the
1146 database key; it just determines when the database is updated. (This
1147 means that when you upgrde Exim will forget old rate measurements.)
1149 Exim now checks that the per_* options are used with an update mode that
1150 makes sense for the current ACL. For example, when Exim is processing a
1151 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
1152 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
1153 must specify per_mail/readonly. If you omit the update mode it defaults to
1154 /leaky where that makes sense (as before) or /readonly where required.
1156 The /noupdate option is now undocumented but still supported for
1157 backwards compatibility. It is equivalent to /readonly except that in
1158 ACLs where /readonly is required you may specify /leaky/noupdate or
1159 /strict/noupdate which are treated the same as /readonly.
1161 A useful new feature is the /count= option. This is a generalization
1162 of the per_byte option, so that you can measure the throughput of other
1163 aggregate values. For example, the per_byte option is now equivalent
1164 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
1166 The per_rcpt option has been generalized using the /count= mechanism
1167 (though it's more complicated than the per_byte equivalence). When it is
1168 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
1169 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
1170 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
1171 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
1172 Note that using per_rcpt with a non-readonly update mode in more than
1173 one ACL will cause the recipients to be double-counted. (The per_mail
1174 and per_byte options don't have this problem.)
1176 The handling of very low rates has changed slightly. If the computed rate
1177 is less than the event's count (usually one) then this event is the first
1178 after a long gap. In this case the rate is set to the same as this event's
1179 count, so that the first message of a spam run is counted properly.
1181 The major new feature is a mechanism for counting the rate of unique
1182 events. The new per_addr option counts the number of different
1183 recipients that someone has sent messages to in the last time period. It
1184 behaves like per_rcpt if all the recipient addresses are different, but
1185 duplicate recipient addresses do not increase the measured rate. Like
1186 the /count= option this is a general mechanism, so the per_addr option
1187 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
1188 example, measure the rate that a client uses different sender addresses
1189 with the options per_mail/unique=$sender_address. There are further
1190 details in the main documentation.
1192 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
1194 TF/06 Removed a few PCRE remnants.
1196 TF/07 Automatically extract Exim's version number from tags in the git
1197 repository when doing development or release builds.
1199 PP/02 Raise smtp_cmd_buffer_size to 16kB.
1200 Bugzilla 879. Patch from Paul Fisher.
1202 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
1203 Heavily based on revision 40f9a89a from Simon Arlott's tree.
1206 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
1208 PP/05 Variable $av_failed, true if the AV scanner deferred.
1209 Bugzilla 1078. Patch from John Horne.
1211 PP/06 Stop make process more reliably on build failure.
1212 Bugzilla 1087. Patch from Heiko Schlittermann.
1214 PP/07 Make maildir_use_size_file an _expandable_ boolean.
1215 Bugzilla 1089. Patch from Heiko Schlittermann.
1217 PP/08 Handle ${run} returning more data than OS pipe buffer size.
1218 Bugzilla 1131. Patch from Holger Weiß.
1220 PP/09 Handle IPv6 addresses with SPF.
1221 Bugzilla 860. Patch from Wolfgang Breyha.
1223 PP/10 GnuTLS: support TLS 1.2 & 1.1.
1225 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
1228 PP/11 match_* no longer expand right-hand-side by default.
1229 New compile-time build option, EXPAND_LISTMATCH_RHS.
1230 New expansion conditions, "inlist", "inlisti".
1232 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
1234 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
1236 PP/14 fix log_write() format string regression from TF/03.
1237 Bugzilla 1152. Patch from Dmitry Isaikin.
1243 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
1245 PP/02 Harmonised TLS library version reporting; only show if debugging.
1246 Layout now matches that introduced for other libraries in 4.74 PP/03.
1248 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
1250 PP/04 New "dns_use_edns0" global option.
1252 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
1255 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
1256 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
1258 TK/01 Updated PolarSSL code to 0.14.2.
1259 Bugzilla 1097. Patch from Andreas Metzler.
1261 PP/07 Catch divide-by-zero in ${eval:...}.
1262 Fixes bugzilla 1102.
1264 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
1267 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
1268 format-string attack -- SECURITY: remote arbitrary code execution.
1270 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
1271 time unintentionally subject to list matching rules, letting the header
1272 cause arbitrary Exim lookups (of items which can occur in lists, *not*
1273 arbitrary string expansion). This allowed for information disclosure.
1275 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
1276 INT_MIN/-1 -- value coerced to INT_MAX.
1282 NM/01 Workround for PCRE version dependency in version reporting
1285 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
1286 This fixes portability to compilers other than gcc, notably
1287 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
1289 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
1290 makefiles for portability to HP-UX and POSIX correctness.
1292 PP/01 Permit LOOKUP_foo enabling on the make command-line.
1293 Also via indented variable definition in the Makefile.
1294 (Debugging by Oliver Heesakkers).
1296 PP/02 Restore caching of spamd results with expanded spamd_address.
1297 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
1299 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
1300 Improves build reliability. Fix from: Frank Elsner
1302 NM/02 Fix wide character breakage in the rfc2047 coding
1303 Fixes bug 1064. Patch from Andrey N. Oktyabrski
1305 NM/03 Allow underscore in dnslist lookups
1306 Fixes bug 1026. Patch from Graeme Fowler
1308 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
1309 Code patches from Adam Ciarcinski of NetBSD.
1311 NM/04 Fixed exiqgrep to cope with mailq missing size issue
1314 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
1315 is logged, to avoid truncation. Patch from John Horne.
1317 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
1318 Patch from Jakob Hirsch.
1320 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
1321 SQL string expansion failure details.
1322 Patch from Andrey Oktyabrski.
1324 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
1325 Patch from Simon Arlott.
1327 PP/09 New lookups functionality failed to compile on old gcc which rejects
1328 extern declarations in function scope.
1329 Patch from Oliver Fleischmann
1331 PP/10 Use sig_atomic_t for flags set from signal handlers.
1332 Check getgroups() return and improve debugging.
1333 Fixed developed for diagnosis in bug 927 (which turned out to be
1336 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
1337 Patch from Mark Zealey.
1339 PP/12 Bugzilla 1056: Improved spamd server selection.
1340 Patch from Mark Zealey.
1342 PP/13 Bugzilla 1086: Deal with maildir quota file races.
1343 Based on patch from Heiko Schlittermann.
1345 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
1346 Patch from Uwe Doering, sign-off by Michael Haardt.
1348 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
1349 variable declaration deep within a block. Bug and patch from
1352 PP/15 lookups-Makefile IRIX compatibilty coercion.
1354 PP/16 Make DISABLE_DKIM build knob functional.
1356 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
1357 Patch by Simon Arlott
1359 TF/03 Fix valgrind.h portability to C89 compilers that do not support
1360 variable argument macros. Our copy now differs from upstream.
1366 TF/01 Failure to get a lock on a hints database can have serious
1367 consequences so log it to the panic log.
1369 TF/02 Log LMTP confirmation messages in the same way as SMTP,
1370 controlled using the smtp_confirmation log selector.
1372 TF/03 Include the error message when we fail to unlink a spool file.
1374 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
1375 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
1376 for maintaining out-of-tree patches for some time.
1378 PP/01 Bugzilla 139: Documentation and portability issues.
1379 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
1380 Handle per-OS dynamic-module compilation flags.
1382 PP/02 Let /dev/null have normal permissions.
1383 The 4.73 fixes were a little too stringent and complained about the
1384 permissions on /dev/null. Exempt it from some checks.
1385 Reported by Andreas M. Kirchwitz.
1387 PP/03 Report version information for many libraries, including
1388 Exim version information for dynamically loaded libraries. Created
1389 version.h, now support a version extension string for distributors
1390 who patch heavily. Dynamic module ABI change.
1392 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
1393 privilege escalation vulnerability whereby the Exim run-time user
1394 can cause root to append content of the attacker's choosing to
1397 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
1400 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
1401 If dropping privileges for untrusted macros, we disabled normal logging
1402 on the basis that it would fail; for the Exim run-time user, this is not
1403 the case, and it resulted in successful deliveries going unlogged.
1404 Fixed. Reported by Andreas Metzler.
1410 PP/01 Date: & Message-Id: revert to normally being appended to a message,
1411 only prepend for the Resent-* case. Fixes regression introduced in
1412 Exim 4.70 by NM/22 for Bugzilla 607.
1414 PP/02 Include check_rfc2047_length in configure.default because we're seeing
1415 increasing numbers of administrators be bitten by this.
1417 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
1419 PP/03 Bugzilla 994: added openssl_options main configuration option.
1421 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
1423 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
1425 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
1427 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
1428 without a peer certificate, leading to a segfault because of an
1429 assumption that peers always have certificates. Be a little more
1430 paranoid. Problem reported by Martin Tscholak.
1432 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
1433 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
1434 NB: ClamAV planning to remove STREAM in "middle of 2010".
1435 CL also introduces -bmalware, various -d+acl logging additions and
1436 more caution in buffer sizes.
1438 PP/09 Implemented reverse_ip expansion operator.
1440 PP/10 Bugzilla 937: provide a "debug" ACL control.
1442 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
1444 PP/12 Bugzilla 973: Implement --version.
1446 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
1448 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
1450 PP/15 Bugzilla 816: support multiple condition rules on Routers.
1452 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
1453 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
1454 ignore trailing whitespace.
1456 JJ/02 prevent non-panic DKIM error from being sent to paniclog
1458 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
1461 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
1462 Notification from Dr Andrew Aitchison.
1464 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
1465 ExtendedDetectionInfo response format.
1466 Notification from John Horne.
1468 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
1471 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
1472 XSL and documented dependency on system catalogs, with examples of how
1475 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
1478 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
1479 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
1480 configuration file which is writeable by the Exim user or group.
1482 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
1483 of configuration files to cover files specified with the -C option if
1484 they are going to be used with root privileges, not just the default
1487 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
1488 option (effectively making it always true).
1490 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
1491 files to be used while preserving root privileges.
1493 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
1494 that rogue child processes cannot use them.
1496 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
1497 run-time user, instead of root.
1499 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
1500 Exim run-time user without dropping privileges.
1502 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
1503 result string, instead of calling string_vformat() twice with the same
1506 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
1507 for other users. Others should always drop root privileges if they use
1508 -C on the command line, even for a whitelisted configure file.
1510 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
1512 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
1518 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
1519 $data_path, and $header_path variables; fixed documentation bugs and
1522 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
1523 exipick to access non-standard spools, including the "frozen" queue
1526 NM/01 Bugzilla 965: Support mysql stored procedures.
1527 Patch from Alain Williams
1529 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
1531 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
1532 Patch from Andreas Metzler
1534 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
1535 Patch from Kirill Miazine
1537 NM/05 Bugzilla 671: Added umask to procmail example.
1539 JJ/03 installed exipick 20100323.0, fixing doc bug
1541 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
1542 directory. Notification and patch from Dan Rosenberg.
1544 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
1546 TK/02 Improve log output when DKIM signing operation fails.
1548 MH/01 Treat the transport option dkim_domain as a colon separated
1549 list, not as a single string, and sign the message with each element,
1550 omitting multiple occurences of the same signer.
1552 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
1553 Bugzilla 985, 986. Patch by Simon Arlott
1555 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
1556 Patch by Simon Arlott
1558 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
1559 MBX locking. Notification from Dan Rosenberg.
1565 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
1567 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
1569 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
1571 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
1573 NM/04 Bugzilla 915: Flag broken perl installation during build.
1579 TK/01 Added patch by Johannes Berg that expands the main option
1580 "spamd_address" if it starts with a dollar sign.
1582 TK/02 Write list of recipients to X-Envelope-Sender header when building
1583 the mbox-format spool file for content scanning (suggested by Jakob
1586 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
1587 (http://www.dcc-servers.net/) support via dccifd. Activated by
1588 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
1590 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
1591 by Mark Daniel Reidel <mr@df.eu>.
1593 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
1594 When building exim an external PCRE library is now needed -
1595 PCRE is a system library on the majority of modern systems.
1596 See entry on PCRE_LIBS in EDITME file.
1598 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
1599 conversation. Added nologin parameter to request.
1600 Patch contributed by Kirill Miazine.
1602 TF/01 Do not log submission mode rewrites if they do not change the address.
1604 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
1606 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
1607 log files in place. Contributed by Roberto Lima.
1609 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
1611 TF/03 Bugzilla 615: When checking the local_parts router precondition
1612 after a local_part_suffix or local_part_prefix option, Exim now
1613 does not use the address's named list lookup cache, since this
1614 contains cached lookups for the whole local part.
1616 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
1617 Robert Millan. Documentation is in experimental-spec.txt.
1619 TF/04 Bugzilla 668: Fix parallel build (make -j).
1621 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
1623 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
1624 Patch provided by Jan Srzednicki.
1626 TF/05 Leading white space used to be stripped from $spam_report which
1627 wrecked the formatting. Now it is preserved.
1629 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
1630 that they are available at delivery time.
1632 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
1634 TF/08 TLS error reporting now respects the incoming_interface and
1635 incoming_port log selectors.
1637 TF/09 Produce a more useful error message if an SMTP transport's hosts
1638 setting expands to an empty string.
1640 NM/06 Bugzilla 744: EXPN did not work under TLS.
1641 Patch provided by Phil Pennock.
1643 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
1644 Patch provided by Richard Godbee.
1646 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
1647 acl_smtp_notquit, added index entry.
1649 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
1650 Patch provided by Eugene Bujak.
1652 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
1653 accept(). Patch provided by Maxim Dounin.
1655 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
1656 Patch provided by Phil Pennock.
1658 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
1660 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
1661 Patch provided by Brad "anomie" Jorsch.
1663 NM/14 Bugzilla 622: Added timeout setting to transport filter.
1664 Patch provided by Dean Brooks.
1666 TK/05 Add native DKIM support (does not depend on external libraries).
1668 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
1669 Patch provided by Graeme Fowler.
1671 NM/16 Bugzilla 851: Documentation example syntax fix.
1673 NM/17 Changed NOTICE file to remove references to embedded PCRE.
1675 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
1678 NM/19 Bugzilla 745: TLS version reporting.
1679 Patch provided by Phil Pennock.
1681 NM/20 Bugzilla 167: bool: condition support.
1682 Patch provided by Phil Pennock.
1684 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
1685 clients. Patch provided by Phil Pennock.
1687 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
1688 Patch provided by Brad "anomie" Jorsch.
1690 NM/23 Bugzilla 687: Fix misparses in eximstats.
1691 Patch provided by Heiko Schlittermann.
1693 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
1694 Patch provided by Heiko Schlittermann.
1696 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
1697 plus update to original patch.
1699 NM/26 Bugzilla 799: Documentation correction for ratelimit.
1701 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
1702 Patch provided by David Brownlee.
1704 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
1706 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
1708 NM/30 Bugzilla 888: TLS documentation bugfixes.
1710 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
1712 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
1713 Unlike the original bugzilla I have changed all shell scripts in src tree.
1715 NM/33 Bugzilla 898: Transport filter timeout fix.
1716 Patch by Todd Rinaldo.
1718 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
1719 Patch by Serge Demonchaux.
1721 NM/35 Bugzilla 39: Base64 decode bug fixes.
1722 Patch by Jakob Hirsch.
1724 NM/36 Bugzilla 909: Correct connect() call in dcc code.
1726 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
1728 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
1730 NM/39 Bugzilla 911: Fixed MakeLinks build script.
1736 TK/01 Add preliminary DKIM support. Currently requires a forked version of
1737 ALT-N's libdkim that I have put here:
1738 http://duncanthrax.net/exim-experimental/
1740 Note to Michael Haardt: I had to rename some vars in sieve.c. They
1741 were called 'true' and it seems that C99 defines that as a reserved
1742 keyword to be used with 'bool' variable types. That means you could
1743 not include C99-style headers which use bools without triggering
1744 build errors in sieve.c.
1746 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
1747 as mailq or other aliases. Changed the --help handling significantly
1748 to do whats expected. exim_usage() emits usage/help information.
1750 SC/01 Added the -bylocaldomain option to eximstats.
1752 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
1754 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
1756 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
1762 PH/01 Another patch from the Sieve maintainer.
1764 PH/02 When an IPv6 address is converted to a string for single-key lookup
1765 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
1766 dots are used instead of colons so that keys in lsearch files need not
1767 contain colons. This was done some time before quoting was made available
1768 in lsearch files. However, iplsearch files do require colons in IPv6 keys
1769 (notated using the quote facility) so as to distinguish them from IPv4
1770 keys. This meant that lookups for IP addresses in host lists did not work
1771 for iplsearch lookups.
1773 This has been fixed by arranging for IPv6 addresses to be expressed with
1774 colons if the lookup type is iplsearch. This is not incompatible, because
1775 previously such lookups could never work.
1777 The situation is now rather anomolous, since one *can* have colons in
1778 ordinary lsearch keys. However, making the change in all cases is
1779 incompatible and would probably break a number of configurations.
1781 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
1784 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
1785 conversion specification without a maximum field width, thereby enabling
1786 a rogue spamd server to cause a buffer overflow. While nobody in their
1787 right mind would setup Exim to query an untrusted spamd server, an
1788 attacker that gains access to a server running spamd could potentially
1789 exploit this vulnerability to run arbitrary code as the Exim user.
1791 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
1792 $primary_hostname instead of what libspf2 thinks the hosts name is.
1794 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
1795 a directory entry by the name of the lookup key. Previously, if a
1796 symlink pointed to a non-existing file or a file in a directory that
1797 Exim lacked permissions to read, a lookup for a key matching that
1798 symlink would fail. Now it is enough that a matching directory entry
1799 exists, symlink or not. (Bugzilla 503.)
1801 PH/03 The body_linecount and body_zerocount variables are now exported in the
1804 PH/04 Added the $dnslist_matched variable.
1806 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
1807 This means they are set thereafter only if the connection becomes
1810 PH/06 Added the client_condition to authenticators so that some can be skipped
1811 by clients under certain conditions.
1813 PH/07 The error message for a badly-placed control=no_multiline_responses left
1814 "_responses" off the end of the name.
1816 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
1818 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
1819 (without spaces) instead of just copying the configuration text.
1821 PH/10 Added the /noupdate option to the ratelimit ACL condition.
1823 PH/11 Added $max_received_linelength.
1825 PH/12 Added +ignore_defer and +include_defer to host lists.
1827 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
1828 way in which PCRE > 7.0 is built.
1830 PH/14 Implemented queue_only_load_latch.
1832 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
1833 MAIL command. The effect was to mangle the value on 64-bit systems.
1835 PH/16 Another patch from the Sieve maintainer.
1837 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
1839 PH/18 If a system quota error occurred while trying to create the file for
1840 a maildir delivery, the message "Mailbox is full" was not appended to the
1841 bounce if the delivery eventually timed out. Change 4.67/27 below applied
1842 only to a quota excession during the actual writing of the file.
1844 PH/19 It seems that peer DN values may contain newlines (and other non-printing
1845 characters?) which causes problems in log lines. The DN values are now
1846 passed through string_printing() before being added to log lines.
1848 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
1849 and InterBase are left for another time.)
1851 PH/21 Added message_body_newlines option.
1853 PH/22 Guard against possible overflow in moan_check_errorcopy().
1855 PH/23 POSIX allows open() to be a macro; guard against that.
1857 PH/24 If the recipient of an error message contained an @ in the local part
1858 (suitably quoted, of course), incorrect values were put in $domain and
1859 $local_part during the evaluation of errors_copy.
1865 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
1866 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
1869 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
1870 issue a MAIL command.
1872 PH/02 In an ACL statement such as
1874 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
1876 if a client was not listed at all, or was listed with a value other than
1877 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
1878 the condition was not true (as it should be), so access was not denied.
1879 The bug was that the ! inversion was incorrectly passed on to the second
1880 item. This has been fixed.
1882 PH/03 Added additional dnslists conditions == and =& which are different from
1883 = and & when the dns lookup returns more than one IP address.
1885 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
1886 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
1888 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
1889 FSYNC, which compiles an option called disable_fsync that allows for
1890 bypassing fsync(). The documentation is heavily laced with warnings.
1892 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
1894 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
1895 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
1896 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
1897 including adding "make clean"; (3) Added -fPIC when compiling the test
1898 dynamically loaded module, to get rid of a warning.
1900 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
1901 message fails, move_frozen_messages = true and ignore_bounce_errors_after
1902 = 0s. The bug is otherwise harmless.
1904 PH/07 There was a bug in the dovecot authenticator such that the value of
1905 $auth1 could be overwritten, and so not correctly preserved, after a
1906 successful authentication. This usually meant that the value preserved by
1907 the server_setid option was incorrect.
1909 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
1911 PH/09 Installed PCRE release 7.0.
1913 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
1914 run for batched SMTP input. It is now run at the start of every message
1915 in the batch. While fixing this I discovered that the process information
1916 (output by running exiwhat) was not always getting set for -bs and -bS
1917 input. This is fixed, and it now also says "batched" for BSMTP.
1919 PH/11 Added control=no_pipelining.
1921 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
1922 patch, slightly modified), and move the expansion of helo_data till after
1923 the connection is made in the smtp transport (so it can use these
1926 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
1928 PH/14 Added log_selector = +pid.
1930 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
1932 PH/16 Add ${if forany and ${if forall.
1934 PH/17 Added dsn_from option to vary the From: line in DSNs.
1936 PH/18 Flush SMTP output before performing a callout, unless control =
1937 no_callout_flush is set.
1939 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
1940 was true (the default) a successful delivery failed to delete the retry
1941 item, thus causing premature timeout of the address. The bug is now
1944 PH/20 Added hosts_avoid_pipelining to the smtp transport.
1946 PH/21 Long custom messages for fakedefer and fakereject are now split up
1947 into multiline reponses in the same way that messages for "deny" and
1948 other ACL rejections are.
1950 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
1951 with slight modification.
1953 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
1954 draft, changing the syntax and factoring some duplicate code".
1956 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
1957 for deliveries of the second and subsequent messages over the same SMTP
1960 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
1961 ${reduce, with only minor "tidies".
1963 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
1965 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
1966 expansion side effects.
1968 PH/27 When a message times out after an over-quota error from an Exim-imposed
1969 quota, the bounce message says "mailbox is full". This message was not
1970 being given when it was a system quota that was exceeded. It now should
1973 MH/03 Made $recipients available in local_scan(). local_scan() already has
1974 better access to the recipient list through recipients_list[], but
1975 $recipients can be useful in postmaster-provided expansion strings.
1977 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
1978 in the case of a MAIL command with additional options following the
1979 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
1980 were accidentally chopped off.
1982 PH/29 SMTP synchronization checks are implemented when a command is read -
1983 there is a check that no more input is waiting when there shouldn't be
1984 any. However, for some commands, a delay in an ACL can mean that it is
1985 some time before the response is written. In this time, more input might
1986 arrive, invalidly. So now there are extra checks after an ACL has run for
1987 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
1988 pipelining has not been advertised.
1990 PH/30 MH's patch to allow iscntrl() characters to be list separators.
1992 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
1993 returned in the SMTP response when smtp_return_error_details was false.
1994 This has been fixed.
1996 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
1997 instead of the C I/O that was originally supplied, because problems were
1998 reported on Solaris.
2000 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
2001 Exim which did not show up earlier: it was assuming that a call to
2002 SSL_CTX_set_info_callback() might give an error value. In fact, there is
2003 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
2004 was a macro that became an assignment, so it seemed to work. This has
2005 changed to a proper function call with a void return, hence the compile
2006 error. Exim's code has been fixed.
2008 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
2011 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
2013 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
2015 PH/37 If a message is not accepted after it has had an id assigned (e.g.
2016 because it turns out to be too big or there is a timeout) there is no
2017 "Completed" line in the log. When some messages of this type were
2018 selected by exigrep, they were listed as "not completed". Others were
2019 picked up by some special patterns. I have improved the selection
2020 criteria to be more general.
2022 PH/38 The host_find_failed option in the manualroute router can now be set
2023 to "ignore", to completely ignore a host whose IP address cannot be
2024 found. If all hosts are ignored, the behaviour is controlled by the new
2025 host_all_ignored option.
2027 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
2028 homing or because of multiple MX records with /mx) generated more than
2029 one IP address, and the following item turned out to be the local host,
2030 all the secondary addresses of the first item were incorrectly removed
2031 from the list, along with the local host and any following hosts (which
2032 is what is supposed to happen).
2034 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
2035 whoever called Exim into the -H file. In the case of the daemon it was
2036 behaving confusingly. When first started, it used values for whoever
2037 started the daemon, but after a SIGHUP it used the Exim user (because it
2038 calls itself on a restart). I have changed the code so that it now always
2041 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
2042 message are rejected with the same error (e.g. no authentication or bad
2043 sender address), and a DATA command is nevertheless sent (as can happen
2044 with PIPELINING or a stupid MUA), the error message that was given to the
2045 RCPT commands is included in the rejection of the DATA command. This is
2046 intended to be helpful for MUAs that show only the final error to their
2049 PH/42 Another patch from the Sieve maintainer.
2051 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
2052 Eximstats - Fixed some broken HTML links and added missing column headers
2054 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
2055 columns for Rejects, Temp Rejects, Ham, and Spam rows.
2057 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
2059 PH/43 Yet another patch from the Sieve maintainer.
2061 PH/44 I found a way to check for a TCP/IP connection going away before sending
2062 the response to the final '.' that terminates a message, but only in the
2063 case where the client has not sent further data following the '.'
2064 (unfortunately, this is allowed). However, in many cases there won't be
2065 any further data because there won't be any more messages to send. A call
2066 to select() can be used: if it shows that the input is "ready", there is
2067 either input waiting, or the socket has been closed. An attempt to read
2068 the next input character can distinguish the two cases. Previously, Exim
2069 would have sent an OK response which the client would never have see.
2070 This could lead to message repetition. This fix should cure that, at
2071 least in a lot of common cases.
2073 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
2074 advertised in response to EHLO.
2080 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
2081 fixed by 4.65/MH/01 (is this a record?) are fixed:
2083 (i) An empty string was always treated as zero by the numeric comparison
2084 operators. This behaviour has been restored.
2086 (ii) It is documented that the numeric comparison operators always treat
2087 their arguments as decimal numbers. This was broken in that numbers
2088 starting with 0 were being interpreted as octal.
2090 While fixing these problems I realized that there was another issue that
2091 hadn't been noticed. Values of message_size_limit (both the global option
2092 and the transport option) were treated as octal if they started with 0.
2093 The documentation was vague. These values are now always treated as
2094 decimal, and I will make that clear in the documentation.
2100 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
2101 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
2104 MH/01 Don't check that the operands of numeric comparison operators are
2105 integers when their expansion is in "skipping" mode (fixes bug
2106 introduced by 4.64-PH/07).
2108 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
2109 child addresses, Exim now panics and dies. Previously, because the count
2110 is held in a short int, deliveries were likely to be lost. As such a
2111 large number of recipients for a single message is ridiculous
2112 (performance will be very, very poor), I have chosen to impose a limit
2113 rather than extend the field.
2119 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
2120 leftover -K file (the existence of which was triggered by #402).
2121 While we were at it, introduced process PID as part of the -K
2122 filename. This should rule out race conditions when creating
2125 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
2126 processing considerably. Previous code took too long for large mails,
2127 triggering a timeout which in turn triggers #401.
2129 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
2130 in the DK code in transports.c. sendfile() is not really portable,
2131 hence the _LINUX specificness.
2133 TF/01 In the add_headers option to the mail command in an Exim filter,
2134 there was a bug that Exim would claim a syntax error in any
2135 header after the first one which had an odd number of characters
2138 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
2139 callout verification, Exim cached a "reject" for the entire domain. This
2140 is correct for most verifications, but it is not correct for a recipient
2141 verification with use_sender or use_postmaster set, because in that case
2142 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
2143 case of MAIL FROM:<> rejection from other early rejections (e.g.
2144 rejection of HELO). When verifying a recipient using a non-null MAIL
2145 address, the cache is ignored if it shows MAIL FROM:<> rejection.
2146 Whatever the result of the callout, the value of the domain cache is
2147 left unchanged (for any other kind of callout, getting as far as trying
2148 RCPT means that the domain itself is ok).
2150 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
2153 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
2154 manifest itself as EPIPE rather than ECONNECT. When tidying away a
2155 session, the daemon ignores ECONNECT errors and logs others; it now
2156 ignores EPIPE as well.
2158 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
2159 (quoted-printable decoding).
2161 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
2162 later the small subsequent patch to fix an introduced bug.
2164 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
2166 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
2168 PH/08 An error is now given if message_size_limit is specified negative.
2170 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
2171 to be given (somewhat) arbitrary names.
2173 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
2176 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
2177 miscellaneous code fixes
2179 PH/10 Added the log_reject_target ACL modifier to specify where to log
2182 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
2183 hostname. This is wrong, because it relates to the incoming message (and
2184 probably the interface on which it is arriving) and not to the outgoing
2185 callout (which could be using a different interface). This has been
2186 changed to use the value of the helo_data option from the smtp transport
2187 instead - this is what is used when a message is actually being sent. If
2188 there is no remote transport (possible with a router that sets up host
2189 addresses), $smtp_active_hostname is used.
2191 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
2192 tweaks were necessary in order to get it to work (see also 21 below):
2193 (a) The code assumed that strncpy() returns a negative number on buffer
2194 overflow, which isn't the case. Replaced with Exim's string_format()
2196 (b) There were several signed/unsigned issues. I just did the minimum
2197 hacking in of casts. There is scope for a larger refactoring.
2198 (c) The code used strcasecmp() which is not a standard C function.
2199 Replaced with Exim's strcmpic() function.
2200 (d) The code set only $1; it now sets $auth1 as well.
2201 (e) A simple test gave the error "authentication client didn't specify
2202 service in request". It would seem that Dovecot has changed its
2203 interface. Fortunately there's a specification; I followed it and
2204 changed what the client sends and it appears to be working now.
2206 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
2209 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
2210 address A is aliased to B and C, where B exists and C does not. Without
2211 -v the output is "A verified" because verification stops after a
2212 successful redirection if more than one address is generated. However,
2213 with -v the child addresses are also verified. Exim was outputting "A
2214 failed to verify" and then showing the successful verification for C,
2215 with its parentage. It now outputs "B failed to verify", showing B's
2216 parentage before showing the successful verification of C.
2218 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
2219 look up a TXT record in a specific list after matching in a combined
2222 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
2223 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
2224 they consult the DNS. I had assumed they would set it the way they
2225 wanted; and indeed my experiments on Linux seem to show that in some
2226 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
2227 To be on the safe side, however, I have now made the interface to
2228 host_find_byname() similar to host_find_bydns(), with an argument
2229 containing the DNS resolver options. The host_find_byname() function now
2230 sets these options at its start, just as host_find_bydns() does. The smtp
2231 transport options dns_qualify_single and dns_search_parents are passed to
2232 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
2233 of host_find_byname() use the default settings of RES_DEFNAMES
2234 (qualify_single) but not RES_DNSRCH (search_parents).
2236 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
2237 spool_read_header() do less string testing, by means of a preliminary
2238 switch on the second character of optional "-foo" lines. (This is
2239 overdue, caused by the large number of possibilities that now exist.
2240 Originally there were few.) While I was there, I also converted the
2241 str(n)cmp tests so they don't re-test the leading "-" and the first
2242 character, in the hope this might squeeze out yet more improvement.
2244 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
2245 flag allowing group syntax was set by the header_syntax check but not
2246 turned off, possible causing trouble later; (2) The flag was not being
2247 set at all for the header_verify test, causing "group"-style headers to
2248 be rejected. I have now set it in this case, and also caused header_
2249 verify to ignore an empty address taken from a group. While doing this, I
2250 came across some other cases where the code for allowing group syntax
2251 while scanning a header line wasn't quite right (mostly, not resetting
2252 the flag correctly in the right place). These bugs could have caused
2253 trouble for malformed header lines. I hope it is now all correct.
2255 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
2256 with the "reply" argument non-NULL. The code, however (which originally
2257 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
2258 but it didn't always do it. This confused somebody who was copying the
2259 code for some other use. I have removed all the tests.
2261 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
2262 feature that was used to support insecure browsers during the U.S. crypto
2263 embargo. It requires special client support, and Exim is probably the
2264 only MTA that supported it -- and would never use it because real RSA is
2265 always available. This code has been removed, because it had the bad
2266 effect of slowing Exim down by computing (never used) parameters for the
2267 RSA_EXPORT functionality.
2269 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
2270 authenticator to fail if there's a tab character in the incoming data
2271 (there should never be unless someone is messing about, as it's supposed
2272 to be base64-encoded). Also added, on Timo's advice, the "secured" option
2273 if the connection is using TLS or if the remote IP is the same as the
2274 local IP, and the "valid-client-cert option" if a client certificate has
2277 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
2278 authenticators. This can be used for authorization after authentication
2279 succeeds. (In the case of plaintext, it servers for both authentication
2282 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
2283 if any retry times were supplied.
2285 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
2286 connection, where there is no client IP address to check. In this
2287 situation, the verify now always succeeds.
2289 PH/25 Applied John Jetmore's -Mset patch.
2291 PH/26 Added -bem to be like -Mset, but loading a message from a file.
2293 PH/27 In a string expansion for a processed (not raw) header when multiple
2294 headers of the same name were present, leading whitespace was being
2295 removed from all of them, but trailing whitespace was being removed only
2296 from the last one. Now trailing whitespace is removed from each header
2297 before concatenation. Completely empty headers in a concatenation (as
2298 before) are ignored.
2300 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
2301 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
2303 PH/29 [Removed. This was a change that I later backed out, and forgot to
2304 correct the ChangeLog entry (that I had efficiently created) before
2305 committing the later change.]
2307 PH/30 Exim was sometimes attempting to deliver messages that had suffered
2308 address errors (4xx response to RCPT) over the same connection as other
2309 messages routed to the same hosts. Such deliveries are always "forced",
2310 so retry times are not inspected. This resulted in far too many retries
2311 for the affected addresses. The effect occurred only when there were more
2312 hosts than the hosts_max_try setting in the smtp transport when it had
2313 the 4xx errors. Those hosts that it had tried were not added to the list
2314 of hosts for which the message was waiting, so if all were tried, there
2315 was no problem. Two fixes have been applied:
2317 (i) If there are any address or message errors in an SMTP delivery, none
2318 of the hosts (tried or untried) are now added to the list of hosts
2319 for which the message is waiting, so the message should not be a
2320 candidate for sending over the same connection that was used for a
2321 successful delivery of some other message. This seems entirely
2322 reasonable: after all the message is NOT "waiting for some host".
2323 This is so "obvious" that I'm not sure why it wasn't done
2324 previously. Hope I haven't missed anything, but it can't do any
2325 harm, as the worst effect is to miss an optimization.
2327 (ii) If, despite (i), such a delivery is accidentally attempted, the
2328 routing retry time is respected, so at least it doesn't keep
2329 hammering the server.
2331 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
2332 in ${readsocket because some servers need this prod.
2334 PH/32 Added some extra debug output when updating a wait-xxx database.
2336 PH/33 The hint "could be header name not terminated by colon", which has been
2337 given for certain expansion errors for a long time, was not being given
2338 for the ${if def:h_colon_omitted{... case.
2340 PH/34 The spec says: "With one important exception, whenever a domain list is
2341 being scanned, $domain contains the subject domain." There was at least
2342 one case where this was not true.
2344 PH/35 The error "getsockname() failed: connection reset by peer" was being
2345 written to the panic log as well as the main log, but it isn't really
2346 panic-worthy as it just means the connection died rather early on. I have
2347 removed the panic log writing for the ECONNRESET error when getsockname()
2350 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
2351 runs only) independently of the message's sender address. This meant
2352 that, if the 4xx error was in fact related to the sender, a different
2353 message to the same recipient with a different sender could confuse
2354 things. In particualar, this can happen when sending to a greylisting
2355 server, but other circumstances could also provoke similar problems.
2356 I have changed the default so that the retry time for these errors is now
2357 based a combination of the sender and recipient addresses. This change
2358 can be overridden by setting address_retry_include_sender=false in the
2361 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
2362 remote server are returned as part of bounce messages. This was not
2363 happening for LMTP over a pipe (the lmtp transport), but now it is the
2364 same for both kinds of LMTP.
2366 PH/38 Despite being documented as not happening, Exim was rewriting addresses
2367 in header lines that were in fact CNAMEs. This is no longer the case.
2369 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
2370 and queue runs started by the daemon processed all messages. This has
2371 been fixed so that -R and -S can now usefully be given with -q<time>.
2373 PH/40 Import PCRE release 6.7 (fixes some bugs).
2375 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
2377 PH/42 Give an error if -q is specified more than once.
2379 PH/43 Renamed the variables $interface_address and $interface_port as
2380 $received_ip_address and $received_port, to make it clear that these
2381 values apply to message reception, and not to the outgoing interface when
2382 a message is delivered. (The old names remain recognized, of course.)
2384 PH/44 There was no timeout on the connect() call when using a Unix domain
2385 socket in the ${readsocket expansion. There now is.
2387 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
2388 be meaningful with "accept".
2390 SC/01 Eximstats V1.43
2391 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
2393 SC/02 Eximstats V1.44
2394 Use a glob alias rather than an array ref in the generated
2395 parser. This improves both readability and performance.
2397 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
2398 Collect SpamAssassin and rejection statistics.
2399 Don't display local sender or destination tables unless
2400 there is data to show.
2401 Added average volumes into the top table text output.
2403 SC/04 Eximstats V1.46
2404 Collect data on the number of addresses (recipients)
2405 as well as the number of messages.
2407 SC/05 Eximstats V1.47
2408 Added 'Message too big' to the list of mail rejection
2409 reasons (thanks to Marco Gaiarin).
2411 SC/06 Eximstats V1.48
2412 Mainlog lines which have GMT offsets and are too short to
2413 have a flag are now skipped.
2415 SC/07 Eximstats V1.49 (Alain Williams)
2416 Added the -emptyok flag.
2418 SC/08 Eximstats V1.50
2419 Fixes for obtaining the IP address from reject messages.
2421 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
2422 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
2423 whitesspace changes from 4.64-PH/27
2425 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
2428 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
2429 are found, allow negative numbers in numeric criteria)
2431 JJ/06 exipick.20061117.2, added new $message_body_missing variable
2433 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
2434 to match changes made in 4.64-PH/43
2436 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
2438 PH/47 Put in an explicit test for a DNS lookup of an address record where the
2439 "domain" is actually an IP address, and force a failure. This locks out
2440 those revolvers/nameservers that support "A-for-A" lookups, in
2441 contravention of the specifications.
2443 PH/48 When a host name was looked up from an IP address, and the subsequent
2444 forward lookup of the name timed out, the host name was left in
2445 $sender_host_name, contrary to the specification.
2447 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
2448 restricted to single-key lookups, Exim was not diagnosing an error if
2449 * or *@ was used with a query-style lookup.
2451 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
2453 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
2454 long ago, but noone interested enough thought of it. Let's just say that
2455 the "1.1" means that there are some new functions that weren't there at
2456 some point in the past.
2458 PH/51 Error processing for expansion failure of helo_data from an smtp
2459 transport during callout processing was broken.
2461 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
2462 tested/used via the -bh/-bhc/-bs options.
2464 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
2465 bug, fixed in subsequent PCRE releases).
2467 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
2468 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
2470 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
2476 SC/01 Use a glob alias rather than an array ref in eximstats generated
2477 parser. This improves both readability and performance.
2479 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
2480 Don't display local sender or destination tables in eximstats unless
2481 there is data to show.
2482 Added average volumes into the eximstats top table text output.
2484 SC/03 Collect data on the number of addresses (recipients) as well
2485 as the number of messages in eximstats.
2487 TF/01 Correct an error in the documentation for the redirect router. Exim
2488 does (usually) call initgroups() when daemonizing.
2490 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
2491 with consistent privilege compared to when running as a daemon.
2493 TF/03 Note in the spec that $authenticated_id is not set for local
2494 submissions from trusted users.
2496 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
2497 Thanks to Dean Brooks <dean@iglou.com> for the patch.
2499 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
2500 by adding some example configuration directives to the default
2501 configuration file. A little bit of work is required to uncomment the
2502 directives and define how usernames and passwords are checked, but
2503 there is now a framework to start from.
2505 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
2506 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
2507 without this. I don't know how relevant this is to other LDAP libraries.
2509 PH/02 Add the verb name to the "unknown ACL verb" error.
2511 PH/03 Magnus Holmgren's patch for filter_prepend_home.
2513 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
2515 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
2516 directory not expanded when it should be if an expanded home directory
2517 was set for the address (which is overridden by the transport).
2519 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
2522 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
2523 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
2524 because it is too late at that time, and has no effect.
2526 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
2527 security issue with \' (bugzilla #107). I could not use the
2528 PQescapeStringConn() function, because it needs a PGconn value as one of
2531 PH/08 When testing addresses using -bt, indicate those final addresses that
2532 are duplicates that would not cause an additional delivery. At least one
2533 person was confused, thinking that -bt output corresponded to deliveries.
2534 (Suppressing duplicates isn't a good idea as you lose the information
2535 about possibly different redirections that led to the duplicates.)
2537 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
2538 systems where poll() doesn't work, in particular OS X.
2540 PH/10 Added more information to debugging output for retry time not reached.
2542 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
2543 operations in malware.c.
2545 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
2548 PH/13 If write_rejectlog was set false when logging was sent to syslog with
2549 syslog_duplication set false, log lines that would normally be written
2550 both the the main log and to the reject log were not written to syslog at
2553 PH/14 In the default configuration, change the use of "message" in ACL warn
2554 statements to "add_header".
2556 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
2557 not followed by a command (e.g. "seen endif").
2559 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
2560 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
2563 PH/17 Added extra conditions to the default value of delay_warning_condition
2567 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
2568 { match{$h_precedence:}{(?i)bulk|list|junk} } \
2569 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
2572 The Auto-Submitted: and various List- headers are standardised, whereas I
2573 don't think Precedence: ever was.
2575 PH/18 Refactored debugging code in route_finduser() to show more information,
2576 in particular, the error code if getpwnam() issues one.
2578 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
2579 This is apparently needed in addition to the PH/07 change above to avoid
2580 any possible encoding problems.
2582 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
2583 but not after initializing Perl.
2585 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
2586 output them only if debugging. By default they are written stderr,
2587 apparently, which is not desirable.
2589 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
2592 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
2595 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
2597 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
2598 authenticated or an ident call has been made. Suppress the default
2599 values for $authenticated_id and $authenticated_sender (but permit -oMai
2600 and -oMas) when testing with -bh.
2602 PH/24 Re-jigged the order of the tests in the default configuration so that the
2603 tests for valid domains and recipients precede the DNS black list and CSA
2604 tests, on the grounds that those ones are more expensive.
2606 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
2607 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
2608 This bug exists in every version of Exim that I still have, right back to
2611 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
2612 However, an attempt to turn on case-sensitivity in a regex key by
2613 including (?-i) didn't work because the subject string was already
2614 lowercased, and the effects were non-intuitive. It turns out that a
2615 one-line patch can be used to allow (?-i) to work as expected.
2621 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
2622 other effects) broke the use of negated acl sub-conditions.
2624 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
2627 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
2628 "Deny" causes Exim to reject the incoming connection with a 554 error.
2629 Unfortunately, if there is a major crisis, such as a disk failure,
2630 tcp-wrappers gives "deny", whereas what one would like would be some
2631 kind of temporary error. A kludge has been added to help with this.
2632 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
2633 554 error is used if errno is still zero or contains ENOENT (which occurs
2634 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
2637 PH/03 Add -lutil to the default FreeBSD LIBS setting.
2639 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
2640 errors. Otherwise a message that provokes a temporary error (when other
2641 messages do not) can cause a whole host to time out.
2643 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
2644 addresses were routed directly to files or pipes from a redirect router.
2645 File deliveries just didn't batch; pipe deliveries might have suffered
2648 PH/06 A failure to get a lock for a hints database would erroneously always say
2649 "Failed to get write lock", even when it was really a read lock.
2651 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
2652 of 0600. This has been changed to use the value of the lockfile_mode
2653 option (which defaults to 0600).
2655 PH/08 Applied small patch from the Sieve maintainer.
2657 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
2658 folder from quota calculations, a direct delivery into this folder messed
2659 up the contents of the maildirsize file. This was because the regex was
2660 used only to exclude .Trash (or whatever) when the size of the mailbox
2661 was calculated. There was no check that a delivery was happening into an
2662 excluded directory. This bug has been fixed by ignoring all quota
2663 processing for deliveries into excluded directories.
2665 PH/10 Added the maildirfolder_create_regex option to appendfile.
2671 PH/01 The code for finding all the local interface addresses on a FreeBSD
2672 system running IPv6 was broken. This may well have applied to all BSD
2673 systems, as well as to others that have similar system calls. The broken
2674 code found IPv4 interfaces correctly, but gave incorrect values for the
2675 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
2676 that it would not match correctly against @[] and not recognize the IPv6
2679 PH/02 The ipliteral router was not recognizing addresses of the form user@
2680 [ipv6:....] because it didn't know about the "ipv6:" prefix.
2682 PH/03 Added disable_ipv6.
2684 PH/04 Changed $reply_address to use the raw form of the headers instead of the
2685 decoded form, because it is most often used to construct To: headers
2686 lines in autoreplies, and the decoded form may well be syntactically
2687 invalid. However, $reply_address has leading white space removed, and all
2688 newlines turned into spaces so that the autoreply transport does not
2691 PH/05 If group was specified without a user on a router, and no group or user
2692 was specified on a transport, the group from the router was ignored.
2694 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
2695 for visible compile-time settings that can be used to change these
2696 numbers, for those that want even more. Backwards compatibility with old
2697 spool files has been maintained. However, going back to a previous Exim
2698 release will lost any variables that are in spool files.
2700 PH/07 Two small changes when running in the test harness: increase delay when
2701 passing a TCP/IP connection to a new process, in case the original
2702 process has to generate a bounce, and remove special handling of
2703 127.0.0.2 (sic), which is no longer necessary.
2705 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
2706 be the same on different OS.
2708 PH/09 Moved a debug statement in filter processing to avoid a race problem when
2711 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
2712 whether --show-vars was specified or not
2714 JJ/02 exipick: Added support for new ACL variable spool format introduced
2717 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
2718 syntactically invalid From: or Reply-to: line, and a filter used this to
2719 generate an autoreply, and therefore failed to obtain an address for the
2720 autoreply, Exim could try to deliver to a non-existent relative file
2721 name, causing unrelated and misleading errors. What now happens is that
2722 it logs this as a hard delivery error, but does not attempt to create a
2725 PH/11 The exinext utility has a -C option for testing purposes, but although
2726 the given file was scanned by exinext itself; it wasn't being passed on
2727 when Exim was called.
2729 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
2730 an end-of-file indication when reading a command response.
2732 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
2733 compiled. In many other places in Exim, IPv6 addresses are always
2734 recognized, so I have changed this. It also means that IPv4 domain
2735 literals of the form [IPV4:n.n.n.n] are now always recognized.
2737 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
2738 used if the router is not running as root, for example, when verifying at
2739 ACL time, or when using -bh. The debugging output from this situation was
2740 non-existent - all you got was a failure to exec. I have made two
2743 (a) Failures to set uid/gid, the current directory, or a process leader
2744 in a subprocess such as that created by queryprogram now generate
2745 suitable debugging ouput when -d is set.
2747 (b) The queryprogram router detects when it is not running as root,
2748 outputs suitable debugging information if -d is set, and then runs
2749 the subprocess without attempting to change uid/gid.
2751 PH/15 Minor change to Makefile for building test_host (undocumented testing
2754 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
2755 additional section of a DNS packet that returns MX or SRV records.
2756 Instead, it always explicitly searches for A/AAAA records. This avoids
2757 major problems that occur when a DNS server includes only records of one
2758 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
2759 fixed another bug: if SRV records were looked up and the corresponding
2760 address records were *not* found in the additional section, the port
2761 values from the SRV records were lost.
2763 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
2764 using the correct key (the original address) when searching the retry
2765 rules in order to find which one to use for generating the retry hint.
2767 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
2768 adding the default one. Similarly, if it contains a Reply-To: header, the
2769 errors_reply_to option, if set, is not used.
2771 PH/19 When calculating a retry time, Exim used to measure the "time since
2772 failure" by looking at the "first failed" field in the retry record. Now
2773 it does not use this if it is later than than the arrival time of the
2774 message. Instead it uses the arrival time. This makes for better
2775 behaviour in cases where some deliveries succeed, thus re-setting the
2776 "first failed" field. An example is a quota failure for a huge message
2777 when small messages continue to be delivered. Without this change, the
2778 "time since failure" will always be short, possible causing more frequent
2779 delivery attempts for the huge message than are intended.
2780 [Note: This change was subsequently modified - see PH/04 for 4.62.]
2782 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
2783 $1, $2, $3) because the numerical variables can be reset during some
2784 expansion items (e.g. "match"), thereby losing the authentication data.
2786 PH/21 Make -bV show the size of off_t variables so that the test suite can
2787 decide whether to run tests for quotas > 2G.
2789 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
2790 mailbox_size, and mailbox_filecount in the appendfile transport. If a
2791 filecount value is greater than 2G or if a quota value is greater than 2G
2792 on a system where the size of off_t is not greater than 4, a panic error
2795 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
2796 never match. The debug and -bh output now contains an explicit error
2797 message indicating a malformed IPv4 address or mask.
2799 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
2800 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
2801 PH/23 above applies.
2803 PH/25 Do not write to syslog when running in the test harness. The only
2804 occasion when this arises is a failure to open the main or panic logs
2805 (for which there is an explicit test).
2807 PH/26 Added the /no_tell option to "control=freeze".
2809 PH/27 If a host name lookup failed very early in a connection, for example, if
2810 the IP address matched host_lookup and the reverse lookup yielded a name
2811 that did not have a forward lookup, an error message of the form "no IP
2812 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
2813 could be logged. Now it outputs the IP address instead of "NULL".
2815 PH/28 An enabling patch from MH: add new function child_open_exim2() which
2816 allows the sender and the authenticated sender to be set when
2817 submitting a message from within Exim. Since child_open_exim() is
2818 documented for local_scan(), the new function should be too.
2820 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
2821 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
2822 results in an empty string is now treated as unset.
2824 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
2826 PH/31 Added sender_verify_fail logging option.
2828 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
2829 needed by RFC 822 but not by RFC 2822 was commented out. I have now
2830 tidied the source and removed it altogether.
2832 PH/33 When a queue run was abandoned because the load average was too high, a
2833 log line was always written; now it is written only if the queue_run log
2834 selector is set. In addition, the log line for abandonment now contains
2835 information about the queue run such as the pid. This is always present
2836 in "start" and "stop" lines but was omitted from the "abandon" line.
2838 PH/34 Omit spaces between a header name and the colon in the error message that
2839 is given when verify = headers_syntax fails (if there are lots of them,
2840 the message gets confusing).
2842 PH/35 Change the default for dns_check_names_pattern to allow slashes within
2843 names, as there are now some PTR records that contain slashes. This check
2844 is only to protect against broken name servers that fall over on strange
2845 characters, so the fact that it applies to all lookups doesn't matter.
2847 PH/36 Now that the new test suite is complete, we can remove some of the
2848 special code in Exim that was needed for the old test suite. For example,
2849 sorting DNS records because real resolvers return them in an arbitrary
2850 order. The new test suite's fake resolver always returns records in the
2853 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
2854 bounces) except when queue_only is set, to avoid logging races between
2855 the different processes.
2857 PH/38 Panic-die if .include specifies a non-absolute path.
2859 PH/39 A tweak to the "H" retry rule from its user.
2861 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
2862 a label. They prevented compilation on older perls.
2864 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
2865 a warning to be raised on newish perls.
2867 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
2868 on queue. Changes to match documented behaviour of showing count of
2869 messages matching specified criteria.
2871 PH/40 Changed the default ident timeout from 30s to 5s.
2873 PH/41 Added support for the use of login_cap features, on those BSD systems
2874 that have them, for controlling the resources used by pipe deliveries.
2876 PH/42 The content-scanning code uses fopen() to create files in which to put
2877 message data. Previously it was not paying any attention to the mode of
2878 the files. Exim runs with umask(0) because the rest of the code creates
2879 files with open(), and sets the required mode explicitly. Thus, these
2880 files were ending up world-writeable. This was not a big issue, because,
2881 being within the spool directory, they were not world-accessible. I have
2882 created a function called modefopen, which takes an additional mode
2883 argument. It sets umask(777), creates the file, chmods it to the required
2884 mode, then resets the umask. All the relevant calls to fopen() in the
2885 content scanning code have been changed to use this function.
2887 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
2888 to 24 hours. This avoids potential overflow problems when processing G
2889 and H retry rules. I suspect nobody ever tinkers with this value.
2891 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
2893 PH/45 When the plaintext authenticator is running as a client, the server's
2894 challenges are checked to ensure they are valid base64 strings. By
2895 default, the authentication attempt is cancelled if an invalid string is
2896 received. Setting client_ignore_invalid_base64 true ignores these errors.
2897 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
2898 they are received. Thus, the responses can be made to depend on the
2899 challenges. If an invalid string is ignored, an empty string is placed in
2902 PH/46 Messages that are created by the autoreply transport now contains a
2903 References: header, in accordance with RFCs 2822 and 3834.
2905 PH/47 Added authenticated_sender_force to the smtp transport.
2907 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
2909 PH/49 Installed latest patch from the Sieve maintainer.
2911 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
2912 was also set, the appendfile transport was unnecessarily scanning a
2913 directory of message files (e.g. for maildir delivery) to find the count
2914 of files (along with the size), even though it did not need this
2915 information. It now does the scan only if it needs to find either the
2916 size of the count of files.
2918 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
2920 PH/52 Two bugs concerned with error handling when the smtp transport is
2923 (i) Exim was not creating retry information for temporary errors given
2924 for individual recipients after the DATA command when the smtp transport
2925 was used in LMTP mode. This meant that they could be retried too
2926 frequently, and not timed out correctly.
2928 (ii) Exim was setting the flag that allows error details to be returned
2929 for LMTP errors on RCPT commands, but not for LMTP errors for individual
2930 recipients that were returned after the DATA command.
2932 PH/53 This is related to PH/52, but is more general: for any failing address,
2933 when detailed error information was permitted to be returned to the
2934 sender, but the error was temporary, then after the final timeout, only
2935 "retry timeout exceeded" was returned. Now it returns the full error as
2936 well as "retry timeout exceeded".
2938 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
2939 do this, and (what is worse) MTAs that accept it.
2941 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
2942 will now be deprecated.
2944 PH/56 New os.c-cygwin from the Cygwin maintainer.
2946 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
2947 formats (previously only available in exim formats via -bpr, -bpru,
2948 and -bpra. Now also available in native and exiqgrep formats)
2950 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
2951 with very large, slow to parse queues
2953 JJ/08 exipick: added ! as generic prefix to negate any criteria format
2955 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
2957 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
2958 responses to authentication challenges, though it was showing the
2959 challenges; (ii) I've removed the CR characters from the debug output for
2962 PH/58 Allow for the insertion of a newline as well as a space when a string
2963 is turned into more than one encoded-word during RFC 2047 encoding. The
2964 Sieve code now uses this.
2966 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
2967 data_4xx, lost_connection, tls_required.
2969 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
2970 message was being sent as an SMTP response.
2972 PH/61 Add -l and -k options to exicyclog.
2974 PH/62 When verifying, if an address was redirected to one new address, so that
2975 verification continued, and the new address failed or deferred after
2976 having set something in $address_data, the value of $address_data was not
2977 passed back to the ACL. This was different to the case when no
2978 redirection occurred. The value is now passed back in both cases.
2980 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
2981 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
2982 use login_cap.h, so on its own it isn't the distinguishing feature. The
2983 new name refers directly to the setclassresources() function.
2985 PH/65 Added configuration files for NetBSD3.
2987 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
2989 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
2990 is preferred over IPv4.
2992 PH/68 The bounce_return_message and bounce_return_body options were not being
2993 honoured for bounces generated during the reception of non-SMTP messages.
2994 In particular, this applied to messages rejected by the ACL. This bug has
2995 been fixed. However, if bounce_return_message is true and bounce_return_
2996 body is false, the headers that are returned for a non-SMTP message
2997 include only those that have been read before the error was detected.
2998 (In the case of an ACL rejection, they have all been read.)
3000 PH/69 The HTML version of the specification is now built in a directory called
3001 spec_html instead of spec.html, because the latter looks like a path with
3002 a MIME-type, and this confuses some software.
3004 PH/70 Catch two compiler warnings in sieve.c.
3006 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
3007 function verify_get_ident() calls ip_connect() to connect a socket, but
3008 if the "connect()" function timed out, ip_connect() used to close the
3009 socket. However, verify_get_ident() also closes the socket later, and in
3010 between Exim writes to the log, which may get opened at this point. When
3011 the socket was closed in ip_connect(), the log could get the same file
3012 descriptor number as the socket. This naturally causes chaos. The fix is
3013 not to close the socket in ip_connect(); the socket should be closed by
3014 the function that creates it. There was only one place in the code where
3015 this was missing, in the iplookup router, which I don't think anybody now
3016 uses, but I've fixed it anyway.
3018 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
3019 well as to direct DNS lookups. Otherwise the handling of names in host
3020 lists is inconsistent and therefore confusing.
3026 PH/01 Two changes to the default runtime configuration:
3028 (1) Move the checks for relay_from_hosts and authenticated clients from
3029 after to before the (commented out) DNS black list checks.
3031 (2) Add control=submission to the relay_from_hosts and authenticated
3032 clients checks, on the grounds that messages accepted by these
3033 statements are most likely to be submissions.
3035 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
3037 (1) Generate an error if the third argument for the ${prvs expansion is
3040 (2) Treat a missing third argument of ${prvscheck as if it were an empty
3043 (3) Reset the variables that are obtained from the first argument of
3044 ${prvscheck and used in the second argument before leaving the code,
3045 because their memory is reclaimed, so using them afterwards may do
3048 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
3049 one (it's much easier than Tom thought :-).
3051 (5) Because of (4), we can now allow for the use of $prvscheck_result
3052 inside the third argument.
3054 PH/03 For some reason, the default setting of PATH when running a command from
3055 a pipe transport was just "/usr/bin". I have changed it to
3058 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
3059 anything to be listed in the output from -bV.
3061 PH/05 When a filter generated an autoreply, the entire To: header line was
3062 quoted in the delivery log line, like this:
3064 => >A.N.Other <ano@some.domain> <original@ddress> ...
3066 This has been changed so that it extracts the operative address. There
3067 may be more than one such address. If so, they are comma-separated, like
3070 => >ano@some.domain,ona@other.domain <original@ddress> ...
3072 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
3073 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
3074 not being looked up in the rDNS to get a host name, Exim was showing the
3075 IP address twice in Received: lines, even though the IP addresses were
3076 identical. For example:
3078 Received: from [1.2.3.4] (helo=[1.2.3.4])
3080 However, if the real host name was known, it was omitting the HELO data
3081 if it matched the actual IP address. This has been tidied up so that it
3082 doesn't show the same IP address twice.
3084 PH/07 When both +timestamp and +memory debugging was on, the value given by
3085 $tod_xxx expansions could be wrong, because the tod_stamp() function was
3086 called by the debug printing, thereby overwriting the timestamp buffer.
3087 Debugging no longer uses the tod_stamp() function when +timestamp is set.
3089 PH/08 When the original message was included in an autoreply transport, it
3090 always said "this is a copy of the message, including all the headers",
3091 even if body_only or headers_only was set. It now gives an appropriate
3094 PH/09 Applied a patch from the Sieve maintainer which:
3096 o fixes some comments
3097 o adds the (disabled) notify extension core
3098 o adds some debug output for the result of if/elsif tests
3099 o points to the current vacation draft in the documentation
3100 and documents the missing references header update
3104 o fixes a bug in processing the envelope test (when testing
3105 multiple envelope elements, the last element determinted the
3108 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
3109 Electronic Mail") by including:
3111 Auto-submitted: auto-generated
3113 in the messages that it generates (bounce messages and others, such as
3114 warnings). In the case of bounce messages for non-SMTP mesages, there was
3115 also a typo: it was using "Auto_submitted" (underscore instead of
3116 hyphen). Since every message generated by Exim is necessarily in response
3117 to another message, thes have all been changed to:
3119 Auto-Submitted: auto-replied
3121 in accordance with these statements in the RFC:
3123 The auto-replied keyword:
3125 - SHOULD be used on messages sent in direct response to another
3126 message by an automatic process,
3128 - MUST NOT be used on manually-generated messages,
3130 - MAY be used on Delivery Status Notifications (DSNs) and Message
3131 Disposition Notifications (MDNs),
3133 - MUST NOT be used on messages generated by automatic or periodic
3134 processes, except for messages which are automatic responses to
3137 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
3138 to the default Received: header definition.
3140 PH/12 Added log selector acl_warn_skipped (default on).
3142 PH/13 After a successful wildlsearch lookup, discard the values of numeric
3143 variables because (a) they are in the wrong storage pool and (b) even if
3144 they were copied, it wouldn't work properly because of the caching.
3146 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
3147 checking when decoding. Apparently there are clients that generate
3148 overlong encoded strings. Why am I not surprised?
3150 PH/15 If the first argument of "${if match_address" was not empty, but did not
3151 contain an "@" character, Exim crashed. Now it writes a panic log message
3152 and treats the condition as false.
3154 PH/16 In autoreply, treat an empty string for "once" the same as unset.
3156 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
3157 extension "envelope-auth". The code is finished and in agreement with
3158 other implementations, but there is no documentation so far and in fact,
3159 nobody wrote the draft yet. This extension is currently #undef'ed, thus
3160 not changing the active code.
3162 Print executed "if" and "elsif" statements when debugging is used. This
3163 helps a great deal to understand what a filter does.
3165 Document more things not specified clearly in RFC3028. I had all this
3166 sorted out, when out of a sudden new issues came to my mind. Oops."
3168 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
3171 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
3172 canonical form (as documented). However, after a host name lookup from
3173 the IP address, check_host() was doing a simple string comparison with
3174 addresses acquired from the DNS when checking that the found name did
3175 have the original IP as one of its addresses. Since any found IPv6
3176 addresses are likely to be in abbreviated form, the comparison could
3177 fail. Luckily, there already exists a function for doing the comparison
3178 by converting both addresses to binary, so now that is used instead of
3179 the text comparison.
3181 PH/20 There was another similar case to PH/19, when a complete host name was
3182 given in a host list; looking up its IP address could give an abbreviated
3183 form, whereas the current host's name might or might not be abbreviated.
3184 The same fix has been applied.
3190 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
3191 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
3194 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
3195 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
3197 PH/03 Typo: missing ".o" in src/pcre/Makefile.
3199 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
3200 header line, restrict the check to what is listed in RFCs 2369 and 2929.
3201 Also, for "Auto-Submitted", treat anything other than "no" as
3202 non-personal, in accordance with RFC 3834. (Previously it treated
3203 anything starting "auto-" as non-personal.)
3205 TF/01 The control=submission/name=... option had a problem with syntax
3206 errors if the name included a slash character. The /name= option
3207 now slurps the rest of the string, so it can include any characters
3208 but it must come last in the list of options (after /sender_retain
3211 PH/05 Some modifications to the interface to the fake nameserver for the new
3219 TK/01 Added the "success_on_redirect" address verification option. See
3220 NewStuff for rationale and an example.
3222 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
3224 PH/02 Patch to exigrep to allow it to work on syslog lines.
3226 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
3227 fread() to skip over the body file's header line, because in Cygwin the
3228 header line is locked and is inaccessible.
3230 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
3231 co-exist for some time) to make it clear that it is the Exim ID that is
3232 referenced, not the Message-ID: header line.
3234 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
3235 string_format() function, because snprintf() does not exist on all
3238 PH/06 The use of forbid_filter_existstest now also locks out the use of the
3239 ${stat: expansion item.
3241 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
3242 protocol synchronization error", to keep the pedants happy.
3244 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
3245 well as for IRIX systems, when gcc is being used. See the host.c source
3248 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
3250 PH/10 Named domain lists were not working if used in a queue_smtp_domains
3253 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
3254 transport and to the smtp transport in LMTP mode.
3256 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
3258 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
3259 run a filter in a subprocess. This could lead to confusion in subsequent
3260 lookups in the parent process. There should also be a search_tidyup() at
3261 the end of the subprocess.
3263 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
3264 only if the host matched helo_try_verify_hosts, which caused the
3265 verification to occur when the EHLO/HELO command was issued. The ACL just
3266 tested the remembered result. Now, if a previous verification attempt has
3267 not happened, "verify = helo" does it there and then.
3269 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
3271 TK/03 Fix log output including CR from clamd.
3273 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
3274 exist provoked a memory error which could cause a segfault.
3276 PH/15 Installed PCRE 6.2
3278 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
3280 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
3281 of the problem. Specifically, suggested +O2 rather than +O1 for the
3284 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
3286 PH/20 If a delivery was routed to a non-standard port by means of an SRV
3287 record, the port was not correctly logged when the outgoing_port log
3288 selector was set (it logged the transort's default port).
3290 PH/21 Added support for host-specific ports to manualroute, queryprogram,
3291 fallback_hosts, and "hosts" in the smtp transport.
3293 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
3294 host errors such as "Connection refused".
3296 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
3297 authentication with radiusclient 0.4.9:
3299 - Error returned from rc_read_config was caught wrongly
3300 - Username/password not passed on to radius server due to wrong length.
3302 The presumption is that some radiusclient API changes for 4.51/PH/17
3303 were not taken care of correctly. The code is still untested by me (my
3304 Linux distribution still has 0.3.2 of radiusclient), but it was
3305 contributed by a Radius user.
3307 PH/24 When doing a callout, the value of $domain wasn't set correctly when
3308 expanding the "port" option of the smtp transport.
3310 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
3311 while reading a MIME header. Thanks to Tom Hughes for a patch.
3313 PH/24 Include config.h inside local_scan.h so that configuration settings are
3316 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
3317 that in an ACL for RCPT (for example), you can examine exactly what was
3320 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
3321 commands, but it was not correctly comparing the address with the actual
3322 client host address. Thus, it would show the EHLO address in Received:
3323 header lines when this was not necessary.
3325 PH/27 Added the % operator to ${eval:}.
3327 PH/28 Exim tries to create and chdir to its spool directory when it starts;
3328 it should be ignoring failures (because with -C, for example, it has lost
3329 privilege). It wasn't ignoring creation failures other than "already
3332 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
3335 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
3336 because an input line was too long, either on its own, or by virtue of
3337 too many continuations, the temporary file was not being removed, and the
3338 return code was incorrect.
3340 PH/31 Missing "BOOL" in function definition in filtertest.c.
3342 PH/32 Applied Sieve patches from the maintainer.
3344 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
3346 PH/33 Added "verify = not_blind".
3348 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
3349 Local/Makefile (with some defaults set). These are used in built scripts
3350 such as exicyclog, but they have never been used in the exim_install
3351 script (though there are many overriding facilities there). I have
3352 arranged that the exim_install script now takes note of these two
3355 PH/35 Installed configuration files for Dragonfly.
3357 PH/36 When a locally submitted message by a trusted user did not contain a
3358 From: header, and the sender address was obtained from -f or from an SMTP
3359 MAIL command, and the trusted user did not use -F to supply a sender
3360 name, $originator_name was incorrectly used when constructing a From:
3361 header. Furthermore, $originator_name was used for submission mode
3362 messages from external hosts without From: headers in a similar way,
3363 which is clearly wrong.
3365 PH/37 Added control=suppress_local_fixups.
3367 PH/38 When log_selector = +received_sender was set, and the addition of the
3368 sender made the log line's construction buffer exactly full, or one byte
3369 less than full, an overflow happened when the terminating "\n" was
3372 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
3373 when the result of a list match is failure because a DNS lookup failed.
3375 PH/40 RM_COMMAND is now used in the building process.
3377 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
3378 the "build-* directories that it finds.
3380 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
3381 address in a domain literal was a prefix of an interface address.
3383 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
3384 when verifying a sender address, unless rewrite_headers is false.
3386 PH/44 Wrote a long comment about why errors_to addresses are verified as
3387 recipients, not senders.
3389 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
3390 the ratelimit ACL was added.
3392 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
3394 PH/46 Added extra information about PostgreSQL errors to the error string.
3396 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
3397 avoiding the need to install special zones in a real server. This is
3398 backwards compatible; if it can't find the fake resolver, it drops back.
3399 Thus, both old and new test suites can be run.
3401 TF/02 Added util/ratelimit.pl
3403 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
3404 clock is set back in time.
3406 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
3407 Candler <B.Candler@pobox.com>.
3409 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
3410 OK for addresses that are the result of redirections.
3412 PH/48 A number of further additions for the benefit of the new test suite,
3413 including a fake gethostbyname() that interfaces to the fake DNS resolver
3416 TF/06 The fix for widen_domains has also been applied to qualify_single and
3417 search_parents which are the other dnslookup options that can cause
3420 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
3423 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
3425 TF/07 Exim produced the error message "an SRV record indicated no SMTP
3426 service" if it encountered an MX record with an empty target hostname.
3427 The message is now "an MX or SRV record indicated no SMTP service".
3429 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
3430 if the DNS of the sending site is misconfigured. This is quite a
3431 common situation. This change restores the behaviour of treating a
3432 helo verification defer as a failure.
3434 PH/51 If self=fail was set on a router, the bounce message did not include the
3435 actual error message.
3441 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
3443 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
3444 command itself ended in error, the underlying message about the transport
3445 filter timeout was being overwritten with the pipe command error. Now the
3446 underlying error message should be appended to the second error message.
3448 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
3450 PH/02 Reduce the amount of output that "make" produces by default. Full output
3451 can still be requested.
3453 PH/03 The warning log line about a condition test deferring for a "warn" verb
3454 was being output only once per connection, rather than after each
3455 occurrence (because it was using the same function as for successful
3456 "warn" verbs). This seems wrong, so I have changed it.
3458 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
3459 it should not have, which might have caused a crash in the right
3460 circumstances, but probably never did.
3462 PH/04 Installed a modified version of Tony Finch's patch to make submission
3463 mode fix the return path as well as the Sender: header line, and to
3464 add a /name= option so that you can make the user's friendly name appear
3467 TF/03 Added the control = fakedefer ACL modifier.
3469 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
3470 Mark Lowes for thorough testing.
3472 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
3474 TK/03 Merged latest SRS patch from Miles Wilton.
3476 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
3477 with the definition in sysexits.h (which is #included earlier).
3478 Fortunately, Exim does not actually use EX_OK. The code used to try to
3479 preserve the sysexits.h value, by assumimg that macro definitions were
3480 scanned for macro replacements. I have been disabused of this notion,
3481 so now the code just undefines EX_OK before #including unistd.h.
3483 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
3484 in the smtp transport. When a block could not be written in a single
3485 write() function, the timeout was being re-applied to each part-write.
3486 This seems wrong - if the receiver was accepting one byte at a time it
3487 would take for ever. The timeout is now adjusted when this happens. It
3488 doesn't have to be particularly precise.
3490 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
3491 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
3493 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
3494 without a domain if the check to <postmaster@domain> fails.
3496 SC/01 Eximstats: added -xls and the ability to specify output files
3497 (patch written by Frank Heydlauf).
3499 SC/02 Eximstats: use FileHandles for outputing results.
3501 SC/03 Eximstats: allow any combination of xls, txt, and html output.
3503 SC/04 Eximstats: fixed display of large numbers with -nvr option
3505 SC/05 Eximstats: fixed merging of reports with empty tables.
3507 SC/06 Eximstats: added the -include_original_destination flag
3509 SC/07 Eximstats: removed tabs and trailing whitespace.
3511 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
3513 TK/06 MBOX spool code: Add real "From " MBOX separator line
3514 so the .eml file is really in mbox format (even though
3515 most programs do not really care). Patch from Alex Miller.
3517 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
3518 The latter is generated from $received_to and is only set if the
3519 message has one envelope recipient. SA can use these headers,
3520 obviously out-of-the-box. Patch from Alex Miller.
3522 PH/08 The ${def test on a variable was returning false if the variable's
3523 value was "0", contrary to what the specification has always said!
3524 The result should be true unless the variable is empty.
3526 PH/09 The syntax error of a character other than { following "${if
3527 def:variable_name" (after optional whitespace) was not being diagnosed.
3528 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
3529 accidental colon was present, for example, could give incorrect results.
3531 PH/10 Tidied the code in a number of places where the st_size field of a stat()
3532 result is used (not including appendfile, where other changes are about
3535 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
3536 This involved changing a lot of size variables from int to off_t. It
3537 should work with maildirs and everything.
3539 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
3540 spamd dying while we are connected to it.
3542 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
3545 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
3546 function for BDB 4.3.
3548 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
3550 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
3551 the sources for PCRE have been re-arranged and more files are now
3554 PH/15 The code I had for printing potentially long long variables in PH/11
3555 above was not the best (it lost precision). The length of off_t variables
3556 is now inspected at build time, and an appropriate printing format (%ld
3557 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
3558 to be "long long int" or "long int". This is needed for the internal
3559 formatting function string_vformat().
3561 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
3562 the configuration file to be ":syslog", then the script "guesses" where
3563 the logs files are, rather than using the compiled in default. In our
3564 case the guess is not the same as the compiled default, so the script
3565 suddenly stopped working when I started to use syslog. The patch checks
3566 to see if log_file_path is "". If so, it attempts to read it from exim
3567 with no configuration file to get the compiled in version, before it
3568 falls back to the previous guessing code."
3570 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
3571 implementing BATV in an Exim configuration. See NewStuff for the gory
3574 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
3575 Makefile that are specific to HP-UX.
3577 PH/18 If the "use_postmaster" option was set for a recipient callout together
3578 with the "random" option, the postmaster address was used as the MAIL
3579 FROM address for the random test, but not for the subsequent recipient
3580 test. It is now used for both.
3582 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
3583 patch removes a few documentation additions to RFC 3028, because the
3584 latest draft now contains them. It adds the new en;ascii-case comparator
3585 and a new error check for 8bit text in MIME parts. Comparator and
3586 require names are now matched exactly. I enabled the subaddress
3587 extension, but it is not well tested yet (read: it works for me)."
3589 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
3590 rework some of the code of TK/09 above to avoid the hardwired use of
3591 "%lld" and "long long". Replaced the call to snprintf() with a call to
3594 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
3595 records point to non-existent hosts", "retry timeout exceeded", and
3596 "retry time not reached for any host after a long failure period".
3598 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
3599 experimental DomainKeys support:
3601 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
3602 (2) On an error such as an illegally used "control", the wrong name for
3603 the control was given.
3605 These problems did NOT occur unless DomainKeys support was compiled.
3607 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
3609 PH/24 Added ${if match_ip condition.
3611 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
3612 databases so that it will be absolutely obvious if a crash occurs in the
3613 DB library. This is a regular occurrence (often caused by mis-matched
3616 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
3617 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
3618 picked up on a user's system that detects such things. There doesn't seem
3619 to be a gcc warning option for this - only an attribute that has to be
3620 put on the function's prototype. It seems that in Fedora Core 4 they have
3621 set this on a number of new functions. No doubt there will be more in due
3624 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
3625 specify a transport. However, if an address that was verified by such a
3626 router was the subject of a callout, Exim crashed because it tried to
3627 read the rcpt_include_affixes from the non-existent transport. Now it
3628 just assumes that the setting of that option is false. This bug was
3629 introduced by 4.51/PH/31.
3631 PH/28 Changed -d+all to exclude +memory, because that information is very
3632 rarely of interest, but it makes the output a lot bigger. People tend to
3633 do -d+all out of habit.
3635 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
3636 code in os-type was giving problems when libc.so lives in lib64, like on
3639 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
3640 aren't the modern standard, and it seems that some systems' include files
3641 don't always have them. Exim was already checking for some of the newer
3642 ones like T_AAAA, and defining it itself. I've added checks for all the
3643 record types that Exim uses.
3645 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
3646 not automatically generating a new one, as it is supposed to. This
3647 prevented TLS from working. If the file did exist, but contained invalid
3648 data, a new version was generated, as expected. It was only the case of a
3649 non-existent file that was broken.
3651 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
3652 with a change in libdomainkeys > 0.64.
3654 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
3655 from DNS. If the selector record carries the flag, it now has
3656 precedence over the domain-wide flag.
3658 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
3660 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
3661 the use of a port name that isn't defined in /etc/services) occurred, the
3662 message was deferred as in a normal delivery, and thus remained on the
3663 spool, instead of being failed because of the mua_wrapper setting. This
3664 is now fixed, and I tidied up some of the mua_wrapper messages at the
3667 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
3668 the messages in a hash of arrays rather than using individual hashes.
3669 This is a bit cleaner and results in dramatic memory savings, albeit
3670 at a slight CPU cost.
3672 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
3673 as requested by Marc Sherman.
3675 SC/10 Eximstats: added histograms for user specified patterns as requested
3678 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
3680 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
3681 fopen() in the content-scanning modules that did not already have it.
3687 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
3688 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
3690 TK/02 Fix ACL "control" statement not being available in MIME ACL.
3692 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
3694 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
3695 to test Sieve filters that use "vacation".
3697 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
3698 that changes the way the GnuTLS parameters are stored in the cache file.
3699 The new format can be generated externally. For backward compatibility,
3700 if the data in the cache doesn't make sense, Exim assumes it has read an
3701 old-format file, and it generates new data and writes a new file. This
3702 means that you can't go back to an older release without removing the
3705 PH/03 A redirect router that has both "unseen" and "one_time" set does not
3706 work if there are any delivery delays because "one_time" forces the
3707 parent to be marked "delivered", so its unseen clone is never tried
3708 again. For this reason, Exim now forbids the simultaneous setting of
3711 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
3712 redirected to themselves ("homonym" addresses). Read the long ChangeLog
3713 entry if you want to know the details. The fix, however, neglected to
3714 consider the case when local delivery batching is involved. The test for
3715 "previously delivered" was not happening when checking to see if an
3716 address could be batched with a previous (undelivered) one; under
3717 certain circumstances this could lead to multiple deliveries to the same
3720 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
3721 in its include files, and this causes problems building Exim.
3723 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
3724 header_syntax) but Exim was just ignoring anything given after a slash.
3725 In particular, this caused confusion with an attempt to use "verify =
3726 reverse_host_lookup/defer_ok". An error is now given when options are
3727 supplied for verify items that do not have them. (Maybe reverse_host_
3728 lookup should have a defer_ok option, but that's a different point.)
3730 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
3731 defined by RFC 821) to 2048, because there were problems with some AUTH
3732 commands, and RFC 1869 says the size should be increased for extended
3733 SMTP commands that take arguments.
3735 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
3738 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
3739 "unknown" error; now it says that the functionality isn't in the binary.
3741 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
3742 an address' error message when a string expansion fails (syntax or
3743 whatever). Otherwise the password may appear in the log. Following change
3744 PH/42 below, there is no longer a chance of it appearing in a bounce
3747 PH/11 Installed exipick version 20050225.0 from John Jetmore.
3749 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
3750 of its addresses was ever tried. (Bugzilla bug #2.)
3752 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
3753 the result incorrectly in the debug output. (It correctly added a newline
3754 to what was transported.)
3756 TF/01 Added $received_time.
3758 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
3759 commented out examples of how to interface to a virus scanner and to
3760 SpamAssassin. Also added commented examples of av_scanner and
3761 spamd_address settings.
3763 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
3764 and controls are allowed in which ACLs. There were a couple of minor
3765 errors. Some of the entries in the conditions table (which is a table of
3766 where they are NOT allowed) were getting very unwieldy; rewrote them as a
3767 negation of where the condition IS allowed.
3769 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
3771 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
3772 header file does not have a version number, so I've had to invent a new
3773 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
3774 API. The code is untested by me (my Linux distribution still has 0.3.2 of
3775 radiusclient), but it was contributed by a Radius user.
3777 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
3778 files or directories, for OpenSSL.
3780 PH/19 When an Exim process that is running as root has to create an Exim log
3781 file, it does so in a subprocess that runs as exim:exim so as to get the
3782 ownership right at creation (otherwise, other Exim processes might see
3783 the file with the wrong ownership). There was no test for failure of this
3784 fork() call, which would lead to the process getting stuck as it waited
3785 for a non-existent subprocess. Forks do occasionally fail when resources
3786 run out. I reviewed all the other calls to fork(); they all seem to check
3789 PH/20 When checking for unexpected SMTP input at connect time (before writing
3790 the banner), Exim was not dealing correctly with a non-positive return
3791 from the read() function. If the client had disconnected by this time,
3792 the result was a log entry for a synchronization error with an empty
3793 string after "input=" when read() returned zero. If read() returned -1
3794 (an event I could not check), uninitialized data bytes were printed.
3795 There were reports of junk text (parts of files, etc) appearing after
3798 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
3800 PH/22 Added support for macro redefinition, and (re)definition in between
3801 driver and ACL definitions.
3803 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
3804 forgetting to use the resulting value; it was using the unexpanded value.
3806 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
3807 hadn't been configured. The fix is from Juergen Kreileder, who
3808 understands it better than I do:
3810 "Here's what I see happening with three configured cyrus_sasl
3811 authenticators configured (plain, login, cram-md5):
3813 On startup auth_cyrus_sasl_init() gets called for each of these.
3814 This means three calls to sasl_listmech() without a specified mech_list.
3815 => SASL tests which mechs of all available mechs actually work
3816 => three warnings about OTP not working
3817 => the returned list contains: plain, login, cram-md5, digest-md5, ...
3819 With the patch, sasl_listmech() also gets called three times. But now
3820 SASL's mech_list option is set to the server_mech specified in the the
3821 authenticator. Or in other words, the answer from sasl_listmech()
3822 gets limited to just the mech you're testing for (which is different
3824 => the return list contains just 'plain' or 'login', 'cram-md5' or
3825 nothing depending on the value of ob->server_mech.
3827 I've just tested the patch: Authentication still works fine,
3828 unavailable mechs specified in the exim configuration are still
3829 caught, and the auth.log warnings about OTP are gone."
3831 PH/25 When debugging is enabled, the contents of the command line are added
3832 to the debugging output, even when log_selector=+arguments is not
3835 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
3836 answer is "GNU", and only if the return is "GNU/something" is the answer
3839 PH/27 $acl_verify_message is now set immediately after the failure of a
3840 verification in an ACL, and so is available in subsequent modifiers. In
3841 particular, the message can be preserved by coding like this:
3843 warn !verify = sender
3844 set acl_m0 = $acl_verify_message
3846 Previously, $acl_verify_message was set only while expanding "message"
3847 and "log_message" when a very denied access.
3849 PH/28 Modified OS/os.c-Linux with
3851 -#ifndef OS_LOAD_AVERAGE
3852 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
3854 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
3855 nomenclature these days.)
3857 PH/29 Installed patch from the Sieve maintainer that adds the options
3858 sieve_useraddress and sieve_subaddress to the redirect router.
3860 PH/30 In these circumstances:
3861 . Two addresses routed to the same list of hosts;
3862 . First host does not offer TLS;
3863 . First host accepts first address;
3864 . First host gives temporary error to second address;
3865 . Second host offers TLS and a TLS session is established;
3866 . Second host accepts second address.
3867 Exim incorrectly logged both deliveries with the TLS parameters (cipher
3868 and peerdn, if requested) that were in fact used only for the second
3871 PH/31 When doing a callout as part of verifying an address, Exim was not paying
3872 attention to any local part prefix or suffix that was matched by the
3873 router that accepted the address. It now behaves in the same way as it
3874 does for delivery: the affixes are removed from the local part unless
3875 rcpt_include_affixes is set on the transport.
3877 PH/32 Add the sender address, as F=<...>, to the log line when logging a
3878 timeout during the DATA phase of an incoming message.
3880 PH/33 Sieve envelope tests were broken for match types other than :is. I have
3881 applied a patch sanctioned by the Sieve maintainer.
3883 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
3884 the uid or gid is negative. A case of a negative gid caused this to be
3885 noticed. The fix allows for either to be negative.
3887 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
3888 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
3891 PH/36 The change PH/12 above was broken. Fixed it.
3893 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
3894 the grounds that routing the same address twice would always produce the
3895 same answer. This might have been true once, but it is certainly no
3896 longer true now. Routing a child address may depend on the previous
3897 routing that produced that child. Some complicated redirection strategies
3898 went wrong when messages had multiple recipients, and made Exim's
3899 behaviour dependent on the order in which the addresses were given.
3901 I have moved the duplicate checking until after the routing is complete.
3902 Exim scans the addresses that are assigned to local and remote
3903 transports, and removes any duplicates. This means that more work will be
3904 done, as duplicates will always all be routed, but duplicates are
3905 presumably rare, so I don't expect this is of any significance.
3907 For deliveries to pipes, files, and autoreplies, the duplicate checking
3908 still happens during the routing process, since they are not going to be
3911 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
3912 It corrects a timeout issue with spamd. This is Ian's comment: "The
3913 background is that sometimes spamd either never reads data from a
3914 connection it has accepted, or it never writes response data. The exiscan
3915 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
3916 blindly assumes that writes won't block so it may never time out."
3918 PH/39 Allow G after quota size as well as K and M.
3920 PH/40 The value set for $authenticated_id in an authenticator may not contain
3921 binary zeroes or newlines because the value is written to log lines and
3922 to spool files. There was no check on this. Now the value is run through
3923 the string_printing() function so that such characters are converted to
3924 printable escape sequences.
3926 PH/41 $message_linecount is a new variable that contains the total number of
3927 lines in the message. Compare $body_linecount, which is the count for the
3930 PH/42 Exim no longer gives details of delivery errors for specific addresses in
3931 bounce and delay warning messages, except in certain special cases, which
3934 (a) An SMTP error message from a remote host;
3935 (b) A message specified in a :fail: redirection;
3936 (c) A message specified in a "fail" command in a system filter;
3937 (d) A message specified in a FAIL return from the queryprogram router;
3938 (e) A message specified by the cannot_route_message router option.
3940 In these cases only, Exim does include the error details in bounce and
3941 warning messages. There are also a few cases where bland messages such
3942 as "unrouteable address" or "local delivery error" are given.
3944 PH/43 $value is now also set for the "else" part of a ${run expansion.
3946 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
3947 being worked on, but at least Exim now implements the latest version to
3950 PH/45 In a pipe transport, although a timeout while waiting for the pipe
3951 process to complete was treated as a delivery failure, a timeout while
3952 writing the message to the pipe was logged, but erroneously treated as a
3953 successful delivery. Such timeouts include transport filter timeouts. For
3954 consistency with the overall process timeout, these timeouts are now
3955 treated as errors, giving rise to delivery failures by default. However,
3956 there is now a new Boolean option for the pipe transport called
3957 timeout_defer, which, if set TRUE, converts the failures into defers for
3958 both kinds of timeout. A transport filter timeout is now identified in
3961 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
3962 systems where "make" and "gmake" are different, calling "gmake" at top
3963 level broke things. I've arranged for the value of $(MAKE) to be passed
3964 from the Makefile to this script so that it can call the same version of
3968 A note about Exim versions 4.44 and 4.50
3969 ----------------------------------------
3971 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
3972 changes of various kinds. As a consequence, a big documentation update was
3973 needed. This delayed the release for rather longer than seemed good, especially
3974 in the light of a couple of (minor) security issues. Therefore, the changes
3975 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
3976 release. So 4.44 and 4.50 are in effect two different branches that both start
3979 I have left the 4.50 change log unchanged; it contains all the changes since
3980 4.43. The change log for 4.44 is below; many of its items are identical to
3981 those for 4.50. This seems to be the most sensible way to preserve the
3982 historical information.
3988 1. Minor wording change to the doc/README.SIEVE file.
3990 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
3991 computation of the current number of files was incorrect.
3993 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
3994 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
3997 4. Give more explanation in the error message when the command for a transport
3998 filter fails to execute.
4000 5. There are several places where Exim runs a non-Exim command in a
4001 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4002 was being done only for the command run by the queryprogram router. It is
4003 now done for all such subprocesses. The other cases are: ${run, transport
4004 filters, and the commands run by the lmtp and pipe transports.
4006 6. Added CONFIGURE_GROUP build-time option.
4008 7. Some older OS have a limit of 256 on the maximum number of file
4009 descriptors. Exim was using setrlimit() to set 1000 as a large value
4010 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4011 systems. I've change it so that if it can't get 1000, it tries for 256.
4013 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4014 was an oversight, and furthermore, ever since the addition of extra
4015 controls (e.g. 4.43/32), the checks on when to allow different forms of
4016 "control" were broken. There should now be diagnostics for all cases when a
4017 control that does not make sense is encountered.
4019 9. Added the /retain_sender option to "control=submission".
4021 10. $recipients is now available in the predata ACL (oversight).
4023 11. Tidy the search cache before the fork to do a delivery from a message
4024 received from the command line. Otherwise the child will trigger a lookup
4025 failure and thereby defer the delivery if it tries to use (for example) a
4026 cached ldap connection that the parent has called unbind on.
4028 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4029 of $address_data from the recipient verification was clobbered by the
4030 sender verification.
4032 13. The value of address_data from a sender verification is now available in
4033 $sender_address_data in subsequent conditions in the ACL statement.
4035 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
4037 15. Added a new option "connect=<time>" to callout options, to set a different
4040 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4041 was its contents. (It was OK if the option was not defined at all.)
4043 17. A "Completed" log line is now written for messages that are removed from
4044 the spool by the -Mrm option.
4046 18. New variables $sender_verify_failure and $recipient_verify_failure contain
4047 information about exactly what failed.
4049 19. Added -dd to debug only the daemon process.
4051 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
4052 handles timeouts, both on the server side and network timeouts. Renamed the
4053 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
4055 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
4056 It is now set to "smtps".
4058 22. $host_address is now set to the target address during the checking of
4059 ignore_target_hosts.
4061 23. When checking ignore_target_hosts for an ipliteral router, no host name was
4062 being passed; this would have caused $sender_host_name to have been used if
4063 matching the list had actually called for a host name (not very likely,
4064 since this list is usually IP addresses). A host name is now passed as
4067 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4068 code that specifies a non-restarting handler (typically sigaction() in
4069 modern systems) in an attempt to fix a rare and obscure crash bug.
4071 25. Narrowed the window for a race in the daemon that could cause it to ignore
4072 SIGCHLD signals. This is not a major problem, because they are used only to
4073 wake it up if nothing else does.
4075 26. A malformed maildirsize file could cause Exim to calculate negative values
4076 for the mailbox size or file count. Odd effects could occur as a result.
4077 The maildirsize information is now recalculated if the size or filecount
4080 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4081 support for a long time. Removed HAVE_SYS_VFS_H.
4083 28. Installed the latest version of exipick from John Jetmore.
4085 29. In an address list, if the pattern was not a regular expression, an empty
4086 subject address (from a bounce message) matched only if the pattern was an
4087 empty string. Non-empty patterns were not even tested. This was the wrong
4088 because it is perfectly reasonable to use an empty address as part of a
4089 database query. An empty address is now tested by patterns that are
4090 lookups. However, all the other forms of pattern expect the subject to
4091 contain a local part and a domain, and therefore, for them, an empty
4092 address still always fails if the pattern is not itself empty.
4094 30. Exim went into a mad DNS loop when attempting to do a callout where the
4095 host was specified on an smtp transport, and looking it up yielded more
4096 than one IP address.
4098 31. Re-factored the code for checking spool and log partition space into a
4099 function that finds that data and another that does the check. The former
4100 is then used to implement four new variables: $spool_space, $log_space,
4101 $spool_inodes, and $log_inodes.
4103 32. The RFC2047 encoding function was originally intended for short strings
4104 such as real names; it was not keeping to the 75-character limit for
4105 encoded words that the RFC imposes. It now respects the limit, and
4106 generates multiple encoded words if necessary. To be on the safe side, I
4107 have increased the buffer size for the ${rfc2047: expansion operator from
4110 33. It is now permitted to omit both strings after an "if" condition; if the
4111 condition is true, the result is "true". As before, when the second string
4112 is omitted, a false condition yields an empty string. This makes it less
4113 cumbersome to write custom ACL and router conditions.
4115 34. Failure to deliver a bounce message always caused it to be frozen, even if
4116 there was an errors_to setting on the router. The errors_to setting is now
4119 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4120 canonical form (fully expanded) before being placed in
4121 $sender_host_address.
4123 36. The table in the code that translates DNS record types into text (T_A to
4124 "A" for instance) was missing entries for NS and CNAME. It is just possible
4125 that this could have caused confusion if both these types were looked up
4126 for the same domain, because the text type is used as part of Exim's
4127 per-process caching. But the chance of anyone hitting this buglet seems
4130 37. The dnsdb lookup has been extended in a number of ways.
4132 (1) There is a new type, "zns", which walks up the domain tree until it
4133 finds some nameserver records. It should be used with care.
4135 (2) There is a new type, "mxh", which is like "mx" except that it returns
4136 just the host names, not the priorities.
4138 (3) It is now possible to give a list of domains (or IP addresses) to be
4139 looked up. The behaviour when one of the lookups defers can be
4140 controlled by a keyword.
4142 (4) It is now possible to specify the separator character for use when
4143 multiple records are returned.
4145 38. The dnslists ACL condition has been extended: it is now possible to supply
4146 a list of IP addresses and/or domains to be looked up in a particular DNS
4149 39. Added log_selector=+queue_time_overall.
4151 40. When running the queue in the test harness, wait just a tad after forking a
4152 delivery process, to get repeatability of debugging output.
4154 41. Include certificate and key file names in error message when GnuTLS fails
4155 to set them up, because the GnuTLS error message doesn't include the name
4156 of the failing file when there is a problem reading it.
4158 42. Allow both -bf and -bF in the same test run.
4160 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
4162 44. The "Exiscan patch" is now merged into the mainline Exim source.
4164 45. Sometimes the final signoff response after QUIT could fail to get
4165 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4166 < 0 before doing a fflush(). This bug looks as though it goes back to the
4167 introduction of TLS in release 3.20, but "sometimes" must have been rare
4168 because the tests only now provoked it.
4170 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
4171 (this can affect the format of dates).
4173 47. exim_tidydb, when checking for the continued existence of a message for
4174 which it has found a message-specific retry record, was not finding
4175 messages that were in split spool directories. Consequently, it was
4176 deleting retry records that should have stayed in existence.
4178 48. Steve fixed some bugs in eximstats.
4180 49. The SPA authentication driver was not abandoning authentication and moving
4181 on to the next authenticator when an expansion was forced to fail,
4182 contradicting the general specification for all authenticators. Instead it
4183 was generating a temporary error. It now behaves as specified.
4185 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
4186 (the order specifies the preference for clients). The order is now AES256,
4187 AES128, 3DES, ARCFOUR128.
4189 51. Small patch to Sieve code - explicitly set From: when generating an
4192 52. Exim crashed if a remote delivery caused a very long error message to be
4193 recorded - for instance if somebody sent an entire SpamAssassin report back
4194 as a large number of 550 error lines. This bug was coincidentally fixed by
4195 increasing the size of one of Exim's internal buffers (big_buffer) that
4196 happened as part of the Exiscan merge. However, to be on the safe side, I
4197 have made the code more robust (and fixed the comments that describe what
4200 53. Now that there can be additional text after "Completed" in log lines (if
4201 the queue_time_overall log selector is set), a one-byte patch to exigrep
4202 was needed to allow it to recognize "Completed" as not the last thing in
4205 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
4206 patch that reportedly fixes this has been added. I am not expert enough to
4207 create a test for it. This is what the patch creator wrote:
4209 "I found a little strange behaviour of ldap code when working with
4210 Windows 2003 AD Domain, where users was placed in more than one
4211 Organization Units. When I tried to give exim partial DN, the exit code
4212 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
4213 But simultaneously result of request was absolutely normal ldap result,
4214 so I produce this patch..."
4216 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
4217 REFERENCE, so I have modified the code to exclude the patch when that macro
4220 55. Some experimental protocols are using DNS PTR records for new purposes. The
4221 keys for these records are domain names, not reversed IP addresses. The
4222 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4223 leaves it alone. Component reversal etc. now happens only for IP addresses.
4226 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
4228 57. Double the size of the debug message buffer (to 2048) so that more of very
4229 long debug lines gets shown.
4231 58. The exicyclog utility now does better if the number of log files to keep
4232 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
4234 59. Two changes related to the smtp_active_hostname option:
4236 (1) $smtp_active_hostname is now available as a variable.
4237 (2) The default for smtp_banner uses $smtp_active_hostname instead
4238 of $primary_hostname.
4240 60. The host_aton() function is supposed to be passed a string that is known
4241 to be a valid IP address. However, in the case of IPv6 addresses, it was
4242 not checking this. This is a hostage to fortune. Exim now panics and dies
4243 if the condition is not met. A case was found where this could be provoked
4244 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4245 components; fortuitously, this particular loophole had already been fixed
4246 by change 4.50/55 above.
4248 If there are any other similar loopholes, the new check in host_aton()
4249 itself should stop them being exploited. The report I received stated that
4250 data on the command line could provoke the exploit when Exim was running as
4251 exim, but did not say which command line option was involved. All I could
4252 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4253 running as the user.
4256 61. There was a buffer overflow vulnerability in the SPA authentication code
4257 (which came originally from the Samba project). I have added a test to the
4258 spa_base64_to_bits() function which I hope fixes it.
4261 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
4262 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
4264 63. The daemon start-up calls getloadavg() while still root for those OS that
4265 need the first call to be done as root, but it missed one case: when
4266 deliver_queue_load_max is set with deliver_drop_privilege. This is
4267 necessary for the benefit of the queue runner, because there is no re-exec
4268 when deliver_drop_privilege is set.
4270 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
4271 This has been fixed.
4273 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4274 was in use, was not putting the data itself into the right store pool;
4275 consequently, it could be overwritten for a subsequent message in the same
4276 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4279 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
4281 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
4282 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
4283 the function were treating the return as a boolean value, which happened to
4284 work because 0=false and not-0=true, but is not correct code.
4286 68. The host_aton() function was not handling scoped IPv6 addresses (those
4287 with, for example, "%eth0" on the end) correctly.
4289 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
4290 negated items (that is, ~something) in unsigned ints. Some compilers
4291 apparently mutter when there is no cast.
4293 70. If an address verification called from an ACL failed, and did not produce a
4294 user-specific message (i.e. there was only a "system" message), nothing was
4295 put in $acl_verify_message. In this situation, it now puts the system
4298 71. Change 4.23/11 added synchronization checking at the start of an SMTP
4299 session; change 4.31/43 added the unwanted input to the log line - except
4300 that it did not do this in the start of session case. It now does.
4302 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
4303 This is wrong and can cause the other end to generate a synchronization
4304 error if it is another Exim or anything else that does the synchronization
4305 check. A QUIT command is no longer sent after a timeout.
4307 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
4308 during host lookups.
4310 74. The defer_ok option of callout verification was not working if it was used
4311 when verifying addresses in header lines, that is, for this case:
4313 verify = header_sender/callout=defer_ok
4315 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
4316 those file descriptors could be used for SMTP connections. If anything
4317 wrote to stderr (the example that came up was "warn" in embedded Perl), it
4318 could be sent to the SMTP client, causing chaos. The daemon now opens
4319 stdin, stdout, and stderr to /dev/null when it puts itself into the
4322 76. Arrange for output from Perl's "warn" command to be written to Exim's main
4323 log by default. The user can override this with suitable Perl magic.
4325 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
4326 the log message when discard triggers, was not working for the DATA ACL or
4327 for the non-SMTP ACL.
4329 78. Error message wording change in sieve.c.
4331 79. If smtp_accept_max_per_host was set, the number of connections could be
4332 restricted to fewer than expected, because the daemon was trying to set up
4333 a new connection before checking whether the processes handling previous
4334 connections had finished. The check for completed processes is now done
4335 earlier. On busy systems, this bug wouldn't be noticed because something
4336 else would have woken the daemon, and it would have reaped the completed
4339 80. If a message was submitted locally by a user whose login name contained one
4340 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
4341 It caused a spool format error. I have fixed the spool reading code. A
4342 related problem was that the "from" clause in the Received: line became
4343 illegal because of the space(s). It is now covered by ${quote_local_part.
4345 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
4348 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
4350 83. Patch from Sieve maintainer to fix unterminated string problem in
4351 "vacation" handling.
4353 84. Some minor changes to the Linux configuration files to help with other
4354 OS variants using glibc.
4356 85. One more patch for Sieve to update vacation handling to latest spec.
4359 ----------------------------------------------------
4360 See the note above about the 4.44 and 4.50 releases.
4361 ----------------------------------------------------
4367 1. Change 4.43/35 introduced a bug that caused file counts to be
4368 incorrectly computed when quota_filecount was set in an appendfile
4371 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4372 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4375 3. Give more explanation in the error message when the command for a transport
4376 filter fails to execute.
4378 4. There are several places where Exim runs a non-Exim command in a
4379 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4380 was being done only for the command run by the queryprogram router. It is
4381 now done for all such subprocesses. The other cases are: ${run, transport
4382 filters, and the commands run by the lmtp and pipe transports.
4384 5. Some older OS have a limit of 256 on the maximum number of file
4385 descriptors. Exim was using setrlimit() to set 1000 as a large value
4386 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4387 systems. I've change it so that if it can't get 1000, it tries for 256.
4389 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4390 was an oversight, and furthermore, ever since the addition of extra
4391 controls (e.g. 4.43/32), the checks on when to allow different forms of
4392 "control" were broken. There should now be diagnostics for all cases when a
4393 control that does not make sense is encountered.
4395 7. $recipients is now available in the predata ACL (oversight).
4397 8. Tidy the search cache before the fork to do a delivery from a message
4398 received from the command line. Otherwise the child will trigger a lookup
4399 failure and thereby defer the delivery if it tries to use (for example) a
4400 cached ldap connection that the parent has called unbind on.
4402 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4403 of $address_data from the recipient verification was clobbered by the
4404 sender verification.
4406 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4407 was its contents. (It was OK if the option was not defined at all.)
4409 11. A "Completed" log line is now written for messages that are removed from
4410 the spool by the -Mrm option.
4412 12. $host_address is now set to the target address during the checking of
4413 ignore_target_hosts.
4415 13. When checking ignore_target_hosts for an ipliteral router, no host name was
4416 being passed; this would have caused $sender_host_name to have been used if
4417 matching the list had actually called for a host name (not very likely,
4418 since this list is usually IP addresses). A host name is now passed as
4421 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4422 code that specifies a non-restarting handler (typically sigaction() in
4423 modern systems) in an attempt to fix a rare and obscure crash bug.
4425 15. Narrowed the window for a race in the daemon that could cause it to ignore
4426 SIGCHLD signals. This is not a major problem, because they are used only to
4427 wake it up if nothing else does.
4429 16. A malformed maildirsize file could cause Exim to calculate negative values
4430 for the mailbox size or file count. Odd effects could occur as a result.
4431 The maildirsize information is now recalculated if the size or filecount
4434 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4435 support for a long time. Removed HAVE_SYS_VFS_H.
4437 18. Updated exipick to current release from John Jetmore.
4439 19. Allow an empty sender to be matched against a lookup in an address list.
4440 Previously the only cases considered were a regular expression, or an
4443 20. Exim went into a mad DNS lookup loop when doing a callout where the
4444 host was specified on the transport, if the DNS lookup yielded more than
4447 21. The RFC2047 encoding function was originally intended for short strings
4448 such as real names; it was not keeping to the 75-character limit for
4449 encoded words that the RFC imposes. It now respects the limit, and
4450 generates multiple encoded words if necessary. To be on the safe side, I
4451 have increased the buffer size for the ${rfc2047: expansion operator from
4454 22. Failure to deliver a bounce message always caused it to be frozen, even if
4455 there was an errors_to setting on the router. The errors_to setting is now
4458 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4459 canonical form (fully expanded) before being placed in
4460 $sender_host_address.
4462 24. Updated eximstats to version 1.33
4464 25. Include certificate and key file names in error message when GnuTLS fails
4465 to set them up, because the GnuTLS error message doesn't include the name
4466 of the failing file when there is a problem reading it.
4468 26. Expand error message when OpenSSL has problems setting up cert/key files.
4471 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
4472 (this can affect the format of dates).
4474 28. exim_tidydb, when checking for the continued existence of a message for
4475 which it has found a message-specific retry record, was not finding
4476 messages that were in split spool directories. Consequently, it was
4477 deleting retry records that should have stayed in existence.
4479 29. eximstats updated to version 1.35
4480 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
4481 1.35 - bugfix such that pie charts by volume are generated correctly
4483 30. The SPA authentication driver was not abandoning authentication and moving
4484 on to the next authenticator when an expansion was forced to fail,
4485 contradicting the general specification for all authenticators. Instead it
4486 was generating a temporary error. It now behaves as specified.
4488 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
4489 (the order specifies the preference for clients). The order is now AES256,
4490 AES128, 3DES, ARCFOUR128.
4492 31. Small patch to Sieve code - explicitly set From: when generating an
4495 32. Exim crashed if a remote delivery caused a very long error message to be
4496 recorded - for instance if somebody sent an entire SpamAssassin report back
4497 as a large number of 550 error lines. This bug was coincidentally fixed by
4498 increasing the size of one of Exim's internal buffers (big_buffer) that
4499 happened as part of the Exiscan merge. However, to be on the safe side, I
4500 have made the code more robust (and fixed the comments that describe what
4503 33. Some experimental protocols are using DNS PTR records for new purposes. The
4504 keys for these records are domain names, not reversed IP addresses. The
4505 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4506 leaves it alone. Component reversal etc. now happens only for IP addresses.
4509 34. The host_aton() function is supposed to be passed a string that is known
4510 to be a valid IP address. However, in the case of IPv6 addresses, it was
4511 not checking this. This is a hostage to fortune. Exim now panics and dies
4512 if the condition is not met. A case was found where this could be provoked
4513 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4514 components; fortuitously, this particular loophole had already been fixed
4515 by change 4.50/55 or 4.44/33 above.
4517 If there are any other similar loopholes, the new check in host_aton()
4518 itself should stop them being exploited. The report I received stated that
4519 data on the command line could provoke the exploit when Exim was running as
4520 exim, but did not say which command line option was involved. All I could
4521 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4522 running as the user.
4525 35. There was a buffer overflow vulnerability in the SPA authentication code
4526 (which came originally from the Samba project). I have added a test to the
4527 spa_base64_to_bits() function which I hope fixes it.
4530 36. The daemon start-up calls getloadavg() while still root for those OS that
4531 need the first call to be done as root, but it missed one case: when
4532 deliver_queue_load_max is set with deliver_drop_privilege. This is
4533 necessary for the benefit of the queue runner, because there is no re-exec
4534 when deliver_drop_privilege is set.
4536 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4537 was in use, was not putting the data itself into the right store pool;
4538 consequently, it could be overwritten for a subsequent message in the same
4539 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4542 38. Sometimes the final signoff response after QUIT could fail to get
4543 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4544 < 0 before doing a fflush(). This bug looks as though it goes back to the
4545 introduction of TLS in release 3.20, but "sometimes" must have been rare
4546 because the tests only now provoked it.
4552 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
4553 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
4554 Now it returns an integer. A number of calls were still expecting a T/F
4555 return. Fortuitously, in all cases, the tests worked in OK situations,
4556 which is the norm. However, things would have gone wrong on any write
4557 failures on the smtp file descriptor. This function is used when sending
4558 messages over SMTP and also when doing verify callouts.
4560 2. When Exim is called to do synchronous delivery of a locally submitted
4561 message (the -odf or -odi options), it no longer closes stderr before doing
4564 3. Implemented the mua_wrapper option.
4566 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
4568 5. Implemented the functions header_remove(), header_testname(),
4569 header_add_at_position(), and receive_remove_recipient(), and exported them
4572 6. If an ACL "warn" statement specified the addition of headers, Exim already
4573 inserted X-ACL-Warn: at the start if there was no header name. However, it
4574 was not making this test for the second and subsequent header lines if
4575 there were newlines in the string. This meant that an invalid header could
4576 be inserted if Exim was badly configured.
4578 7. Allow an ACL "warn" statement to add header lines at the start or after all
4579 the Received: headers, as well as at the end.
4581 8. Added the rcpt_4xx retry error code.
4583 9. Added postmaster_mailfrom=xxx to callout verification option.
4585 10. Added mailfrom=xxxx to the callout verification option, for verify=
4588 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
4589 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
4591 12. Inserted some casts to stop certain compilers warning when using pointer
4592 differences as field lengths or precisions in printf-type calls (mostly
4593 affecting debugging statements).
4595 13. Added optional readline() support for -be (dynamically loaded).
4597 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
4598 same clock tick as a message's arrival, so that its received time was the
4599 same as the "first fail" time on the retry record, and that message
4600 remained on the queue past the ultimate address timeout, every queue runner
4601 would try a delivery (because it was past the ultimate address timeout) but
4602 after another failure, the ultimate address timeout, which should have then
4603 bounced the address, did not kick in. This was a "< instead of <=" error;
4604 in most cases the first failure would have been in the next clock tick
4605 after the received time, and all would be well.
4607 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
4608 being recognized when the domain list was tested by the match_domain
4609 condition in an expansion string.
4611 16. Added the ${str2b64: operator.
4613 17. Exim was always calling setrlimit() to set a large limit for the number of
4614 processes, without checking whether the existing limit was already
4615 adequate. (It did check for the limit on file descriptors.) Furthermore,
4616 errors from getrlimit() and setrlimit() were being ignored. Now they are
4617 logged to the main and panic logs, but Exim does carry on, to try to do its
4618 job under whatever limits there are.
4620 18. Imported PCRE 5.0.
4622 19. Trivial typo in log message " temporarily refused connection" (the leading
4625 20. If the log selector return_path_on_delivery was set and an address was
4626 redirected to /dev/null, the delivery process crashed because it assumed
4627 that a return path would always be set for a "successful" delivery. In this
4628 case, the whole delivery is bypassed as an optimization, and therefore no
4631 21. Internal re-arrangement: the function for sending a challenge and reading
4632 a response while authentication was assuming a zero-terminated challenge
4633 string. It's now changed to take a pointer and a length, to allow for
4634 binary data in such strings.
4636 22. Added the cyrus_sasl authenticator (code supplied by MBM).
4638 23. Exim was not respecting finduser_retries when seeking the login of the
4639 uid under which it was called; it was always trying 10 times. (The default
4640 setting of finduser_retries is zero.) Also, it was sleeping after the final
4641 failure, which is pointless.
4643 24. Implemented tls_on_connect_ports.
4645 25. Implemented acl_smtp_predata.
4647 26. If the domain in control=submission is set empty, Exim assumes that the
4648 authenticated id is a complete email address when it generates From: or
4649 Sender: header lines.
4651 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
4652 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
4653 chown and chgrp in /bin and hostname in /usr/bin.
4655 28. Exim was keeping the "process log" file open after each use, just as it
4656 does for the main log. This opens the possibility of it remaining open for
4657 long periods when the USR1 signal hits a daemon. Occasional processlog
4658 errors were reported, that could have been caused by this. Anyway, it seems
4659 much more sensible not to leave this file open at all, so that is what now
4662 29. The long-running daemon process does not normally write to the log once it
4663 has entered its main loop, and it closes the log before doing so. This is
4664 so that log files can straightforwardly be renamed and moved. However,
4665 there are a couple of unusual error situations where the daemon does write
4666 log entries, and I had neglected to close the log afterwards.
4668 30. The text of an SMTP error response that was received during a remote
4669 delivery was being truncated at 512 bytes. This is too short for some of
4670 the long messages that one sometimes sees. I've increased the limit to
4673 31. It is now possible to make retry rules that apply only when a message has a
4674 specific sender, in particular, an empty sender.
4676 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
4677 it possible to be selective about when SMTP synchronization is enforced.
4679 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
4681 32. Implemented hosts_connection_nolog.
4683 33. Added an ACL for QUIT.
4685 34. Setting "delay_warning=" to disable warnings was not working; it gave a
4688 35. Added mailbox_size and mailbox_filecount to appendfile.
4690 36. Added control = no_multiline_responses to ACLs.
4692 37. There was a bug in the logic of the code that waits for the clock to tick
4693 in the case where the clock went backwards by a substantial amount such
4694 that the microsecond fraction of "now" was more than the microsecond
4695 fraction of "then" (but the whole seconds number was less).
4697 38. Added support for the libradius Radius client library this is found on
4698 FreeBSD (previously only the radiusclient library was supported).
4704 1. When certain lookups returned multiple values in the form name=value, the
4705 quoting of the values was not always being done properly. Specifically:
4706 (a) If the value started with a double quote, but contained no whitespace,
4708 (b) If the value contained whitespace other than a space character (i.e.
4709 tabs or newlines or carriage returns) it was not quoted.
4710 This fix has been applied to the mysql and pgsql lookups by writing a
4711 separate quoting function and calling it from the lookup code. The fix
4712 should probably also be applied to nisplus, ibase and oracle lookups, but
4713 since I cannot test any of those, I have not disturbed their existing code.
4715 2. A hit in the callout cache for a specific address caused a log line with no
4716 reason for rejecting RCPT. Now it says "Previous (cached) callout
4717 verification failure".
4719 3. There was an off-by-one bug in the queryprogram router. An over-long
4720 return line was truncated at 256 instead of 255 characters, thereby
4721 overflowing its buffer with the terminating zero. As well as fixing this, I
4722 have increased the buffer size to 1024 (and made a note to document this).
4724 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
4725 when Exim is waiting for an SMTP response from a remote server, Exim
4726 restarts its select() call on the socket, thereby resetting its timeout.
4727 This is not a problem when such interrupts are rare. Somebody set up a cron
4728 job to run exiwhat every 2 minutes, which is less than the normal select()
4729 timeout (5 or 10 minutes). This meant that the select() timeout never
4730 kicked in because it was always reset. I have fixed this by comparing the
4731 time when an interrupt arrives with the time at the start of the first call
4732 to select(). If more time than the timeout has elapsed, the interrupt is
4733 treated as a timeout.
4735 5. Some internal re-factoring in preparation for the addition of Sieve
4736 extensions (by MH). In particular, the "personal" test is moved to a
4737 separate function, and given an option for scanning Cc: and Bcc: (which is
4738 not set for Exim filters).
4740 6. When Exim created an email address using the login of the caller as the
4741 local part (e.g. when creating a From: or Sender: header line), it was not
4742 quoting the local part when it contained special characters such as @.
4744 7. Installed new OpenBSD configuration files.
4746 8. Reworded some messages for syntax errors in "and" and "or" conditions to
4747 try to make them clearer.
4749 9. Callout options, other than the timeout value, were being ignored when
4750 verifying sender addresses in header lines. For example, when using
4752 verify = header_sender/callout=no_cache
4754 the cache was (incorrectly) being used.
4756 10. Added a missing instance of ${EXE} to the exim_install script; this affects
4757 only the Cygwin environment.
4759 11. When return_path_on_delivery was set as a log selector, if different remote
4760 addresses in the same message used different return paths and parallel
4761 remote delivery occurred, the wrong values would sometimes be logged.
4762 (Whenever a remote delivery process finished, the return path value from
4763 the most recently started remote delivery process was logged.)
4765 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
4766 lines when AUTH and/or TLS are in use. This is the "received protocol"
4767 field. Exim used to use "asmtp" for authenticated SMTP, without any
4768 indication (in the protocol name) for TLS use. Now it follows the RFC and
4769 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
4770 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
4771 names appear in log lines as well as in Received: header lines.
4773 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
4774 extensions, and comparison tests, and to fix some bugs.
4776 14. Changes to the "personal" filter test:
4778 (1) The test was buggy in that it was just doing the equivalent of
4779 "contains" tests on header lines. For example, if a user's address was
4780 anne@some.where, the "personal" test would incorrectly be true for
4782 To: susanne@some.where
4784 This test is now done by extracting each address from the header in turn,
4785 and checking the entire address. Other tests that are part of "personal"
4786 are now done using regular expressions (for example, to check local parts
4787 of addresses in From: header lines).
4789 (2) The list of non-personal local parts in From: addresses has been
4790 extended to include "listserv", "majordomo", "*-request", and "owner-*",
4791 taken from the Sieve specification recommendations.
4793 (3) If the message contains any header line starting with "List-" it is
4794 treated as non-personal.
4796 (4) The test for "circular" in the Subject: header line has been removed
4797 because it now seems ill-conceived.
4799 15. Minor typos in src/EDITME comments corrected.
4801 16. Installed latest exipick from John Jetmore.
4803 17. If headers_add on a router specified a text string that was too long for
4804 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
4805 of string_sprintf() is now avoided.
4807 18. $message_body_size was not set (it was always zero) when running the DATA
4808 ACL and the local_scan() function.
4810 19. For the "mail" command in an Exim filter, no default was being set for
4811 the once_repeat time, causing a random time value to be used if "once" was
4812 specified. (If the value happened to be <= 0, no repeat happened.) The
4813 default is now 0s, meaning "never repeat". The "vacation" command was OK
4814 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
4815 (I found it when inspecting the code).
4817 20. There is now an overall timeout for performing a callout verification. It
4818 defaults to 4 times the callout timeout, which applies to individual SMTP
4819 commands during the callout. The overall timeout applies when there is more
4820 than one host that can be tried. The timeout is checked before trying the
4821 next host. This prevents very long delays if there are a large number of
4822 hosts and all are timing out (e.g. when the network connections are timing
4823 out). The value of the overall timeout can be changed by specifying an
4824 additional sub-option for "callout", called "maxwait". For example:
4826 verify = sender/callout=5s,maxwait=20s
4828 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
4829 to the end before writing, but this should make it even safer).
4831 22. Exim was forgetting that it had advertised PIPELINING for the second and
4832 subsequent messages on an SMTP connection. It was also not resetting its
4833 memory on STARTTLS and an internal HELO.
4835 23. When Exim logs an SMTP synchronization error within a session, it now
4836 records whether PIPELINING has been advertised or not.
4838 24. Added 3 instances of "(long int)" casts to time_t variables that were being
4839 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
4840 rather than long int.
4842 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
4844 26. Added the never_mail option to autoreply.
4850 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
4851 crash if the getsockname() call failed; this can happen if a connection is
4852 closed very soon after it is established. The problem was simply in the
4853 order in which certain operations were done, causing Exim to try to write
4854 to the SMTP stream before it had set up the file descriptor. The bug has
4855 been fixed by making things happen in the correct order.
4861 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
4862 before the connection was closed, thus losing the rejection response.
4864 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
4865 some early Solaris releases, but causes trouble in current releases where
4866 socklen_t is defined.
4868 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
4871 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
4874 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
4875 that releasing the top of it at the end releases what was used for sub-
4876 expansions (unless the block got too big). However, discard this block if
4877 the first thing is a variable or header, so that we can use its block when
4878 it is dynamic (useful for very large $message_headers, for example).
4880 6. Lookups now cache *every* query, not just the most recent. A new, separate
4881 store pool is used for this. It can be recovered when all lookup caches are
4882 flushed. Lookups now release memory at the end of their result strings.
4883 This has involved some general refactoring of the lookup sources.
4885 7. Some code has been added to the store_xxx() functions to reduce the amount
4886 of flapping under certain conditions.
4888 8. log_incoming_interface used to affect only the <= reception log lines. Now
4889 it causes the local interface and port to be added to several more SMTP log
4890 lines, for example "SMTP connection from", and rejection lines.
4892 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
4894 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
4896 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
4897 could be overwritten at the end of the current message (or the start of a
4898 new message if it was set in a HELO ACL). The value is now preserved for
4899 the duration of the SMTP connection.
4901 12. If a transport had a headers_rewrite setting, and a matching header line
4902 contained an unqualified address, that address was qualified, even if it
4903 did not match any rewriting rules. The underlying bug was that the values
4904 of the flags that permit the existence of unqualified sender and recipient
4905 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
4906 non-local messages, and by -bnq for local messages) were not being
4907 preserved with the message after it was received.
4909 13. When Exim was logging an SMTP synchronization error, it could sometimes log
4910 "next input=" as part of the text comprising the host identity instead of
4911 the correct text. The code was using the same buffer for two different
4912 strings. However, depending on which order the printing function evaluated
4913 its arguments, the bug did not always show up. Under Linux, for example, my
4914 test suite worked just fine.
4916 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
4917 doesn't work with some older versions of Perl. It has been changed to "my",
4918 which in any case is probably the better facility to use.
4920 15. A really picky compiler found some instances of statements for creating
4921 error messages that either had too many or two few arguments for the format
4924 16. The size of the buffer for calls to the DNS resolver has been increased
4925 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
4926 for addresses that have a lot of PTR records. This alleviates a problem; it
4927 does not fully solve it.
4929 17. A dnsdb lookup for PTR records that receives more data than will fit in the
4930 buffer now truncates the list and logs the incident, which is the same
4931 action as happens when Exim is looking up a host name and its aliases.
4932 Previously in this situation something unpredictable would happen;
4933 sometimes it was "internal error: store_reset failed".
4935 18. If a server dropped the connection unexpectedly when an Exim client was
4936 using GnuTLS and trying to read a response, the client delivery process
4937 crashed while trying to generate an error log message.
4939 19. If a "warn" verb in an ACL added multiple headers to a message in a single
4940 string, for example:
4942 warn message = H1: something\nH2: something
4944 the text was added as a single header line from Exim's point of view
4945 though it ended up OK in the delivered message. However, searching for the
4946 second and subsequent header lines using $h_h2: did not work. This has been
4947 fixed. Similarly, if a system filter added multiple headers in this way,
4948 the routers could not see them.
4950 20. Expanded the error message when iplsearch is called with an invalid key to
4951 suggest using net-iplsearch in a host list.
4953 21. When running tests using -bh, any delays imposed by "delay" modifiers in
4954 ACLs are no longer actually imposed (and a message to that effect is
4957 22. If a "gecos" field in a passwd entry contained escaped characters, in
4958 particular, if it contained a \" sequence, Exim got it wrong when building
4959 a From: or a Sender: header from that name. A second bug also caused
4960 incorrect handling when an unquoted " was present following a character
4961 that needed quoting.
4963 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
4964 was not being matched caselessly.
4966 24. Arranged for all hyphens in the exim.8 source to be escaped with
4969 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
4970 properly. Recipient callout cache records were still being keyed to include
4971 the sender, even when use_sender was set false. This led to far more
4972 callouts that were necessary. The sender is no longer included in the key
4973 when use_sender is false.
4975 26. Added "control = submission" modifier to ACLs.
4977 27. Added the ${base62d: operator to decode base 62 numbers.
4979 28. dnsdb lookups can now access SRV records.
4981 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
4982 the configuration file.
4984 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
4985 (-v) mode. This makes the output for a verbose queue run more intelligible.
4987 31. Added a use_postmaster feature to recipient callouts.
4989 32. Added the $body_zerocount variable, containing the number of binary zero
4990 bytes in the message body.
4992 33. The time of last modification of the "new" subdirectory is now used as the
4993 "mailbox time last read" when there is a quota error for a maildir
4996 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
4998 35. Added +ignore_unknown as a special item in host lists.
5000 36. Code for decoding IPv6 addresses in host lists is now included, even if
5001 IPv6 support is not being compiled. This fixes a bug in which an IPv6
5002 address was recognized as an IP address, but was then not correctly decoded
5003 into binary, causing unexpected and incorrect effects when compared with
5010 1. Very minor rewording of debugging text in manualroute to say "list of
5011 hosts" instead of "hostlist".
5013 2. If verify=header_syntax was set, and a header line with an unqualified
5014 address (no domain) and a large number of spaces between the end of the
5015 name and the colon was received, the reception process suffered a buffer
5016 overflow, and (when I tested it) crashed. This was caused by some obsolete
5017 code that should have been removed. The fix is to remove it!
5019 3. When running in the test harness, delay a bit after writing a bounce
5020 message to get a bit more predictability in the log output.
5022 4. Added a call to search_tidyup() just before forking a reception process. In
5023 theory, someone could use a lookup in the expansion of smtp_accept_max_
5024 per_host which, without the tidyup, could leave open a database connection.
5026 5. Added the variables $recipient_data and $sender_data which get set from a
5027 lookup success in an ACL "recipients" or "senders" condition, or a router
5028 "senders" option, similar to $domain_data and $local_part_data.
5030 6. Moved the writing of debug_print from before to after the "senders" test
5033 7. Change 4.31/66 (moving the time when the Received: is generated) caused
5034 problems for message scanning, either using a data ACL, or using
5035 local_scan() because the Received: header was not generated till after they
5036 were called (in order to set the time as the time of reception completion).
5037 I have revised the way this works. The header is now generated after the
5038 body is received, but before the ACL or local_scan() are called. After they
5039 are run, the timestamp in the header is updated.
5045 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
5046 before starting a queue runner without re-exec. This happened only when
5047 deliver_drop_privilege was set or when the Exim user was set to root. The
5048 effect of the bug was that timeouts during subsequent deliveries caused
5049 crashes instead of being properly handled. The handler is now left at its
5050 default (and expected) setting.
5052 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
5053 message, again when deliver_drop_privilege is set or Exim is run as root.
5054 The bug described in (1) was not present in this case, but the tidying up
5055 of the other signals was missing. I have made the two cases consistent.
5057 3. The ignore_target_hosts setting on a manualroute router was being ignored
5058 for hosts that were looked up using the /MX notation.
5060 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
5063 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
5064 operated on the sender address. After changing the $sender_address to <>
5065 for the sender address verify, Exim was re-instated it as the original
5066 (before rewriting) address, but remembering that it had rewritten it, so it
5067 wasn't rewriting it again. This bug also had the effect of breaking the
5068 sender address verification caching when the sender address was rewritten.
5070 6. The ignore_target_hosts option was being ignored by the ipliteral router.
5071 This has been changed so that if the ip literal address matches
5072 ignore_target_hosts, the router declines.
5074 7. Added expansion conditions match_domain, match_address, and match_local_
5075 part (NOT match_host).
5077 8. The placeholder for the Received: header didn't have a length field set.
5079 9. Added code to Exim itself and to exim_lock to test for a specific race
5080 condition that could lead to file corruption when using MBX delivery. The
5081 issue is with the lockfile that is created in /tmp. If this file is removed
5082 after a process has opened it but before that process has acquired a lock,
5083 there is the potential for a second process to recreate the file and also
5084 acquire a lock. This could lead to two Exim processes writing to the file
5085 at the same time. The added code performs the same test as UW imapd; it
5086 checks after acquiring the lock that its file descriptor still refers to
5087 the same named file.
5089 10. The buffer for building added header lines was of fixed size, 8192 bytes.
5090 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
5093 11. Added the smtp_active_hostname option. If used, this will typically be made
5094 to depend on the incoming interface address. Because $interface_address is
5095 not set up until the daemon has forked a reception process, error responses
5096 that can happen earlier (such as "too many connections") no longer contain
5099 12. If an expansion in a condition on a "warn" statement fails because a lookup
5100 defers, the "warn" statement is abandoned, and the next ACL statement is
5101 processed. Previously this caused the whole ACL to be aborted.
5103 13. Added the iplsearch lookup type.
5105 14. Added ident_timeout as a log selector.
5107 15. Added tls_certificate_verified as a log selector.
5109 16. Added a global option tls_require_ciphers (compare the smtp transport
5110 option of the same name). This controls incoming TLS connections.
5112 17. I finally figured out how to make tls_require_ciphers do a similar thing
5113 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
5114 before starting the TLS session.
5116 18. Tabs are now shown as \t in -bP output.
5118 19. If the log selector return_path_on_delivery was set, Exim crashed when
5119 bouncing a message because it had too many Received: header lines.
5121 20. If two routers both had headers_remove settings, and the first one included
5122 a superfluous trailing colon, the final name in the first list and the
5123 first name in the second list were incorrectly joined into one item (with a
5124 colon in the middle).
5130 1. Added -C and -D options to the exinext utility, mainly to make it easier
5131 to include in the automated testing, but these could be helpful when
5132 multiple configurations are in use.
5134 2. The exinext utility was not formatting the output nicely when there was
5135 an alternate port involved in the retry record key, nor when there was a
5136 message id as well (for retries that were specific to a specific message
5137 and a specific host). It was also confused by IPv6 addresses, because of
5138 the additional colons they contain. I have fixed the IPv4 problem, and
5139 patched it up to do a reasonable job for IPv6.
5141 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
5142 delivery, the log line now contains "pipelined" if PIPELINING was used.
5144 4. An SMTP transport process used to panic and die if the bind() call to set
5145 an explicit outgoing interface failed. This has been changed; it is now
5146 treated in the same way as a connect() failure.
5148 5. A reference to $sender_host_name in the part of a conditional expansion
5149 that was being skipped was still causing a DNS lookup. This no longer
5152 6. The def: expansion condition was not recognizing references to header lines
5153 that used bh_ and bheader_.
5155 7. Added the _cache feature to named lists.
5157 8. The code for checking quota_filecount in the appendfile transport was
5158 allowing one more file than it should have been.
5160 9. For compatibility with Sendmail, the command line option
5168 and sets the incoming protocol and host name (for trusted callers). The
5169 host name and its colon can be omitted when only the protocol is to be set.
5170 Note the Exim already has two private options, -pd and -ps, that refer to
5171 embedded Perl. It is therefore impossible to set a protocol value of "d" or
5172 "s", but I don't think that's a major issue.
5174 10. A number of refactoring changes to the code, none of which should affect
5177 (a) The number of logging options was getting close to filling up the
5178 32-bit word that was used as a bit map. I have split them into two classes:
5179 those that are passed in the argument to log_write(), and those that are
5180 only ever tested independently outside of that function. These are now in
5181 separate 32-bit words, so there is plenty of room for expansion again.
5182 There is no change in the user interface or the logging behaviour.
5184 (b) When building, for example, log lines, the code previously used a
5185 macro that called string_cat() twice, in order to add two strings. This is
5186 not really sufficiently general. Furthermore, there was one instance where
5187 it was actually wrong because one of the argument was used twice, and in
5188 one call a function was used. (As it happened, calling the function twice
5189 did not affect the overall behaviour.) The macro has been replaced by a
5190 function that can join an arbitrary number of extra strings onto a growing
5193 (c) The code for expansion conditions now uses a table and a binary chop
5194 instead of a serial search (which was left over from when there were very
5195 few conditions). Also, it now recognizes conditions like "pam" even when
5196 the relevant support is not compiled in: a suitably worded error message is
5197 given if an attempt is made to use such a condition.
5199 11. Added ${time_interval:xxxxx}.
5201 12. A bug was causing one of the ddress fields not to be passed back correctly
5202 from remote delivery subprocesses. The field in question was not being
5203 subsequently used, so this caused to problems in practice.
5205 13. Added new log selectors queue_time and deliver_time.
5207 14. Might have fixed a bug in maildirsizefile handling that threw up
5208 "unexpected character" debug warnings, and recalculated the data
5209 unnecessarily. In any case, I expanded the warning message to give more
5212 15. Added the message "Restricted characters in address" to the statements in
5213 the default ACL that block characters like @ and % in local parts.
5215 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
5216 Three changes have been made:
5218 (a) There was a serious bug; a negative response to MAIL caused the whole
5219 recipient domain to be cached as invalid, thereby blocking all messages
5220 to all local parts at the same domain, from all senders. This bug has
5221 been fixed. The domain is no longer cached after a negative response to
5222 MAIL if the sender used is not empty.
5224 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
5227 (c) A new callout option, "use_sender" has been added for people who want
5228 the modified behaviour.
5234 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
5237 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
5238 indeed breaks things for older releases.
5240 3. Added additional logging to the case where there is a problem reading data
5241 from a filter that is running in a subprocess using a pipe, in order to
5242 try to track down a specific problem.
5244 4. Testing facility fudge: when running in the test harness and attempting
5245 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
5246 getting "No route to host". Convert this to a timeout.
5248 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
5251 6. Some OS don't have socklen_t but use size_t instead. This affects the
5252 fifth argument of getsockopt() amongst other things. This is now
5253 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
5254 can be set for individual OS. I have set it for SunOS5, OSF1, and
5255 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
5256 some earlier ones do not.
5258 7. Change 4.30/15 was not doing the test caselessly.
5260 8. The standard form for an IPv6 address literal was being rejected by address
5261 parsing in, for example, MAIL and RCPT commands. An example of this kind of
5262 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5263 this, as well as the form without the "IPv6" on the front (but only when
5264 address literals are enabled, of course).
5266 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
5268 10. Exim crashed if a message with an empty sender address specified by -f
5269 encountered a router with an errors_to setting. This could be provoked only
5270 by a command such as
5274 where an empty string was supplied; "<>" did not hit this bug.
5276 11. Installed PCRE release 4.5.
5278 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
5279 remained set. It is now erased.
5281 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
5282 times from message ids (which are base 36 rather than the normal 62).
5284 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
5285 were being counted as actual protocol errors, and logged if the log
5286 selector +smtp_protocol_error was set. One cannot be perfect in this test,
5287 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
5288 and DATA following a set of rejected RCPTs do not count as protocol errors.
5289 In other words, Exim assumes they were pipelined, though this may not
5290 actually be the case. Of course, in all cases the client gets an
5291 appropriate error code.
5293 15. If a lookup fails in an ACL condition, a message about the failure may
5294 be available; it is used if testing the ACL cannot continue, because most
5295 such messages specify what the cause of the deferral is. However, some
5296 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
5297 that caused an old message to be retained and used if a later statement
5298 caused a defer, replacing the real cause of the deferral.
5300 16. If an IP address had so many PTR records that the DNS lookup buffer
5301 was not large enough to hold them, Exim could crash while trying to process
5302 the truncated data. It now detects and logs this case.
5304 17. Further to 4.21/58, another change has been made: if (and only if) the
5305 first line of a message (the first header line) ends with CRLF, a bare LF
5306 in a subsequent header line has a space inserted after it, so as not to
5307 terminate the header.
5309 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
5310 unnecessarily, used atoi() instead of strtol(), and didn't check the
5311 termination when getting file sizes from file names by regex.
5313 19. Completely re-implemented the support for maildirsize files, in the light
5314 of a number of problems with the previous contributed implementation
5315 (4.30/29). In particular:
5317 . If the quota is zero, the maildirsize file is maintained, but no quota is
5320 . If the maildir directory does not exist, it is created before any attempt
5321 to write a maildirsize file.
5323 . The quota value in the file is just a cache; if the quota is changed in
5324 the transport, the new value overrides.
5326 . A regular expression is available for excluding directories from the
5329 20. The autoreply transport checks the characters in options that define the
5330 message's headers; it allows continued headers, but it was checking with
5331 isspace() after an embedded newline instead of explicitly looking for a
5334 21. If all the "regular" hosts to which an address was routed had passed their
5335 expiry times, and had not reached their retry times, the address was
5336 bounced, even if fallback hosts were defined. Now Exim should go on to try
5339 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
5340 equivalent code in the SMTP transport. Some hosts send humungous responses
5341 to HELO/EHLO, more than 1024 it seems.
5343 23. Refactoring: code in filter.c used (void *) for "any old type" but this
5344 gives compiler warnings in some environments. I've now done it "properly",
5347 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
5348 (because of problems with the built-in one) was declared to return uschar *
5349 instead of char *, causing compiler failure.
5351 25. Fixed a file descriptor leak when processing alias/forward files.
5353 26. Fixed a minor format string issue in dbfn.c.
5355 27. Typo in exim.c: ("dmbnz" for "dbmnz").
5357 28. If a filter file refered to $h_xxx or $message_headers, and the headers
5358 contained RFC 2047 "words", Exim's memory could, under certain conditions,
5361 29. When a sender address is verified, it is cached, to save repeating the test
5362 when there is more than one recipient in a message. However, when the
5363 verification involves a callout, it is possible for different callout
5364 options to be set for different recipients. It is too complicated to keep
5365 track of this in the cache, so now Exim always runs a verification when a
5366 callout is required, relying on the callout cache for the optimization.
5367 The overhead is duplication of the address routing, but this should not be
5370 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
5371 to be verified with callout=postmaster, and the main callout worked but the
5372 postmaster check failed, the verification correctly failed. However, if a
5373 subsequent RCPT command asked for sender verification *without* the
5374 postmaster check, incorrect caching caused this verification also to fail,
5377 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
5378 it was not caching the DNS options (qualify_single, search_parents) that
5379 were used when the lookup failed. A subsequent lookup with different
5380 options therefore always gave the same answer, though there were cases
5381 where it should not have. (Example: a "domains = !$mx_any" option on a
5382 dnslookup router: the "domains" option is always processed without any
5383 widening, but the router might have qualify_single set.) Now Exim uses the
5384 cached value only when the same options are set.
5386 32. Added John Jetmore's "exipick" utility to the distribution.
5388 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
5389 than a timeout (e.g. a certificate is required, and is not provided), an
5390 Exim server now closes the connection immediately. Previously it waited for
5391 the client to close - but if the client is SSL, it seems that they each
5392 wait for each other, leading to a delay before one of them times out.
5394 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
5395 maintained 0.8.x compatibility because I don't think many are using it, and
5396 it is clearly obsolete.
5398 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
5401 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
5402 client certificate was expired. A simple patch fixes this, though I don't
5403 understand the full logic of why the verify callback is called multiple
5406 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
5407 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
5408 which causes problems with some clients (such as the Certicom SSL Plus
5409 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
5410 disables the coutermeasure allowing Eudora to connect."
5412 38. Exim was not checking that a write() to a log file succeeded. This could
5413 lead to Bad Things if a log got too big, in particular if it hit a file
5414 size limit. Exim now panics and dies if it cannot write to a log file, just
5415 as it does if it cannot open a log file.
5417 39. Modified OS/Makefile-Linux so that it now contains
5419 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
5421 The two -D definitions ensure that Exim is compiled with large file
5422 support, which makes it possible to handle log files that are bigger than
5425 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
5426 instance) a domain was checked against a named list that involved a lookup,
5427 causing $domain_data to be set, then another domain was checked against the
5428 same list, then the first domain was re-checked, the value of $domain_data
5429 after the final check could be wrong. In particular, if the second check
5430 failed, it could be set empty. This bug probably also applied to
5433 41. The strip_trailing_dot option was not being applied to the address given
5434 with the -f command-line option.
5436 42. The code for reading a message's header from the spool was incrementing
5437 $received_count, but never initializing it. This meant that the value was
5438 incorrect (doubled) while delivering a message in the same process in which
5439 it was received. In the most common configuration of Exim, this never
5440 happens - a fresh exec is done - but it can happen when
5441 deliver_drop_privilege is set.
5443 43. When Exim logs an SMTP synchronization error - client data sent too soon -
5444 it now includes up to 150 characters of the unexpected data in the log
5447 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
5448 and building data strings. The size of both of these buffers was 10 000
5449 bytes - far larger than anybody would *ever* want, thought I. Needless to
5450 say, somebody hit the limit. I have increased the maximum line length to
5451 20 000 and the maximum data length of concatenated lines to 100 000. I have
5452 also fixed two bugs, because there was no checking on these buffers. Tsk,
5453 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
5454 buffer is too small.
5456 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
5457 lsearch lookups. Now it does.
5459 46. When parsing a route_list item in a manualroute router, a fixed-length
5460 buffer was used for the list of hosts. I made this 1024 bytes long,
5461 thinking that nobody would ever have a list of hosts that long. Wrong.
5462 Somebody had a whole pile of complicated expansion conditions, and the
5463 string was silently truncated, leading to an expansion error. It turns out
5464 that it is easier to change to an unlimited length (owing to other changes
5465 that have happened since this code was originally written) than to build
5466 structure for giving a limitation error. The length of the item that
5467 expands into the list of hosts is now unlimited.
5469 47. The lsearch lookup could not handle data where the length of text line was
5470 more than 4095 characters. Such lines were truncated, leading to shortened
5471 data being returned. It should now handle lines of any length.
5473 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
5474 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
5477 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
5478 replacing strings such as BIN_DIRECTORY in a source file: the replacement
5479 no longer happens in comment lines. A list of replacements is now placed
5480 at the head of all of the source files, except those whose only change is
5481 to replace PERL_COMMAND in the very first #! line.
5483 50. Replaced the slow insertion sort in queue.c, for sorting the list of
5484 messages on the queue, with a bottom-up merge sort, using code contributed
5485 by Michael Haardt. This should make operations like -bp somewhat faster on
5486 large queues. It won't affect queue runners, except when queue_run_in_order
5489 51. Installed eximstats 1.31 in the distribution.
5491 52. Added support for SRV lookups to the dnslookup router.
5493 53. If an ACL referred to $message_body or $message_body_end, the value was not
5494 reset for any messages that followed in the same SMTP session.
5496 54. The store-handling optimization for building very long strings was not
5497 differentiating between the different store pools. I don't think this
5498 actually made any difference in practice, but I've tidied it.
5500 55. While running the routers to verify a sender address, $sender_address
5501 was still set to the sender address. This is wrong, because when routing to
5502 send a bounce to the sender, it would be empty. Therefore, I have changed
5503 it so that, while verifying a sender address, $sender_address is set to <>.
5504 (There is no change to what happens when verifying a recipient address.)
5506 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
5507 target A or AAAA records (if not already returned) without resetting the
5508 qualify_single or search_parents options of the DNS resolver. These are
5509 inappropriate in this case because the targets of MX and SRV records must
5510 be FQDNs. A broken DNS record could cause trouble if it happened to have a
5511 target that, when qualified, matched something in the local domain. These
5512 two options are now turned off when doing these lookups.
5514 57. It seems that at least some releases of Reiserfs (which does not have the
5515 concept of a fixed number of inodes) returns zero and not -1 for the
5516 number of available inodes. This interacted badly with check_spool_inodes,
5517 which assumed that -1 was the "no such thing" setting. What I have done is
5518 to check that the total number of inodes is greater than zero before doing
5519 the test of how many are available.
5521 58. When a "warn" ACL statement has a log_message modifier, the message is
5522 remembered, and not repeated. This is to avoid a lot of repetition when a
5523 message has many recipients that cause the same warning to be written.
5524 Howewer, Exim was preserving the list of already written lines for an
5525 entire SMTP session, which doesn't seem right. The memory is now reset if a
5526 new message is started.
5528 59. The "rewrite" debugging flag was not showing the result of rewriting in the
5529 debugging output unless log_rewrite was also set.
5531 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
5532 of (int)(handle) when we know that handle contains (void *)(-1).
5534 61. The Exim daemon panic-logs an error return when it closes the incoming
5535 connection. However "connection reset by peer" seems to be common, and
5536 isn't really an error worthy of noting specially, so that particular error
5539 62. When Exim is trying to find all the local interfaces, it used to panic and
5540 die if the ioctl to get the interface flags failed. However, it seems that
5541 on at least one OS (Solaris 9) it is possible to have an interface that is
5542 included in the list of interfaces, but for which you get a failure error
5543 for this call. This happens when the interface is not "plumbed" into a
5544 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
5545 failure of the "get flags" call assumes that the interface is down.
5547 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
5548 makes life easier for people who are doing arithmetic on fields extracted
5549 from dates, where you often get leading zeros that should not be
5550 interpreted as octal.
5552 64. Added qualify_domain to the redirect router, to override the global
5555 65. If a pathologically long header line contained very many addresses (the
5556 report of this problem mentioned 10 000) and each of them was rewritten,
5557 Exim could use up a very large amount of memory. (It kept on making new
5558 copies of the header line as it rewrote, and never released the old ones.)
5559 At the expense of a bit more processing, the header rewriting function has
5560 been changed so that it no longer eats memory in this way.
5562 66. The generation of the Received: header has been moved from the time that a
5563 message starts to be received, to the time that it finishes. The timestamp
5564 in the Received: header should now be very close to that of the <= log
5565 line. There are two side-effects of this change:
5567 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
5568 logged header lines no longer include the local Received: line, because
5569 it has not yet been created. The same applies to a copy of the message
5570 that is returned to a non-SMTP sender when a message is rejected.
5572 (b) When a filter file is tested using -bf, no additional Received: header
5573 is added to the test message. After some thought, I decided that this
5576 This change does not affect the value of $received_for. It is still set
5577 after address rewriting, but before local_scan() is called.
5579 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
5581 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
5582 gave an unhelpful panic error message, and a defer error. I have managed to
5583 change this behaviour so that it now rejects any supplied certificate,
5584 which seems right, as the list of acceptable certificates is empty.
5586 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
5587 gave an unhelpful defer error. I have not managed to make this reject any
5588 supplied certificates, but the error message it gives is "no certificate
5589 supplied", which is not helpful.
5591 70. exigrep's output now also includes lines that are not associated with any
5592 message, but which match the given pattern. Implemented by a patch from
5593 Martin Sluka, which also tidied up the Perl a bit.
5595 71. Recipient callout verification, like sender verification, was using <> in
5596 the MAIL FROM command. This isn't really the right thing, since the actual
5597 sender may affect whether the remote host accepts the recipient or not. I
5598 have changed it to use the actual sender in the callout; this means that
5599 the cache record is now keyed on a recipient/sender pair, not just the
5600 recipient address. There doesn't seem to be a real danger of callout loops,
5601 since a callout by the remote host to check the sender would use <>.
5602 [SEE ABOVE: changed after hitting problems.]
5604 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
5605 temporary errors. However, in the case of such a code being given after
5606 the end of a data transmission (i.e. after ".") Exim was failing to write
5607 a retry record for the message. (Yes, there was some broken host that was
5608 actually sending 8xx at this point.)
5610 73. An unknown lookup type in a host list could cause Exim to panic-die when
5611 the list was checked. (An example that provoked this was putting <; in the
5612 middle of a list instead of at the start.) If this happened during a DATA
5613 ACL check, a -D file could be left lying around. This kind of configuration
5614 error no longer causes Exim to die; instead it causes a defer errror. The
5615 incident is still logged to the main and panic logs.
5617 74. Buglet left over from Exim 3 conversion. The message "too many messages
5618 in one connection" was written to the rejectlog but not the mainlog, except
5619 when address rewriting (yes!) was being logged.
5621 75. Added write_rejectlog option.
5623 76. When a system filter was run not as root (that is, when system_filter_user
5624 was set), the values of the $n variables were not being returned to the
5625 main process; thus, they were not subsequently available in the $sn
5628 77. Added +return_path_on_delivery log selector.
5630 78. A connection timeout was being treated differently from recipients deferred
5631 when testing hosts_max_try with a message that was older than the host's
5632 retry timeout. (The host should not be counted, thus allowing all hosts to
5633 be tried at least once before bouncing.) This may have been the cause of an
5634 occasionally reported bug whereby a message would remain on the queue
5635 longer than the retry timeout, but would be bounced if a delivery was
5636 forced. I say "may" because I never totally pinned down the problem;
5637 setting up timeout/retry tests is difficult. See also the next item.
5639 79. The ultimate address timeout was not being applied to errors that involved
5640 a combination of host plus message (for example, a timeout on a MAIL
5641 command). When an address resolved to a number of possible hosts, and they
5642 were not all tried for each delivery (e.g. because of hosts_max_try), a
5643 message could remain on the queue longer than the retry timeout.
5645 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
5648 81. Fixed an obscure SMTP outgoing bug which required at least the following
5649 conditions: (a) there was another message waiting for the same server;
5650 (b) the server returned 5xx to all RCPT commands in the first message so
5651 that the message was not completed; (c) the server dropped the connection
5652 or gave a negative response to the RSET that Exim sends to abort the
5653 transaction. The observed case was a dropped connection after DATA that had
5654 been sent in pipelining mode. That is, the server had advertised PIPELINING
5655 but was not implementing it correctly. The effect of the bug was incorrect
5656 behaviour, such as trying another host, and this could lead to a crash.
5662 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
5663 and daemon.c were passed as pointers to ints; they should have been
5664 pointers to socklen_t variables (which are typically unsigned ints).
5666 2. Some signed/unsigned type warnings in the os.c file for Linux have been
5669 3. Fixed a really odd bug that affected only the testing scheme; patching a
5670 certain fixed string in the binary changed the value of another string that
5671 happened to be identical to the end of the original first string.
5673 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
5674 name", it returns that address as the IP address. On some operating
5675 systems (e.g. Solaris), it also passes back the IP address string as the
5676 "host name". However, on others (e.g. Linux), it passes back an empty
5677 string. Exim wasn't checking for this, and was changing the host name to an
5678 empty string, assuming it had been canonicized.
5680 5. Although rare, it is permitted to have more than one PTR record for a given
5681 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
5682 all the names associated with an address, because they do in Solaris.
5683 However, it seems that they do not in Linux for data that comes from the
5684 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
5685 I found this out when I moved to a new Linux workstation and tried to run
5686 the Exim test suite.
5688 To get round this problem I have changed the code so that it now does its
5689 own call to the DNS to look up PTR records when searching for a host name.
5690 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
5691 addresses that are only in /etc/hosts are still found.
5693 This behaviour is, however, controlled by an option called host_lookup_
5694 order, which defaults to "bydns:byaddr". If people want to use the other
5695 order, or indeed, just use one or the other means of lookup, they can
5696 specify it in this variable.
5698 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
5699 some operating systems, this comes back from gethostbyaddr() as an empty
5700 string, and this is what Exim used to test for. However, it seems that in
5701 other systems, "." is yielded. Exim now tests for this case too.
5703 7. The values of check_spool_space and check_log_space are now held internally
5704 as a number of kilobytes instead of an absolute number of bytes. If a
5705 numbers is specified without 'K' or 'M', it is rounded up to the nearest
5706 kilobyte. This means that much larger values can be stored.
5708 8. Exim monitor: an attempt to get the action menu when not actually pointing
5709 at a message produces an empty menu entitled "No message selected". This
5710 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
5711 no entries in it ("Shell widget menu has zero width and/or height"). So I
5712 have added a single, blank menu entry in this case.
5714 9. Added ${quote_local_part.
5716 10. MIME decoding is now applied to the contents of Subject: header lines when
5719 11. Now that a reference to $sender_host_address automatically causes a reverse
5720 lookup to occur if necessary (4.13/18), there is no need to arrange for a
5721 host lookup before query-style lookups in lists that might use this
5722 variable. This has therefore been abolished, and the "net-" prefix is no
5723 longer necessary for query-style lookups.
5725 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
5726 have been a typo for LFLAGS, so it has been changed.
5728 13. The install script calls Exim with "-C /dev/null" in order to find the
5729 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
5730 to be output. Howeve, since Exim outputs its version number before the
5731 error, it didn't break the script. It just looked ugly. I fixed this by
5732 always allowing "-C /dev/null" if the caller is root.
5734 14. Ignore overlarge ACL variable number when reading spool file - insurance
5735 against a later release with more variables having written the file.
5737 15. The standard form for an IPv6 address literal was being rejected by EHLO.
5738 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5739 this, as well as the form without the "IPv6" on the front.
5741 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
5742 OS/Makefile-Darwin file.
5744 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
5745 only to LDAP libraries that do not have LDAP_OPT_DEREF.
5747 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
5748 variable. However, some OS use ints for inodes. I've added cast to long int
5749 to get rid of the compiler warning.
5751 19. I had forgotten to lock out "/../" in configuration file names when
5752 ALT_CONFIG_PREFIX was set.
5754 20. Routers used for verification do not need to specify transports. However,
5755 if such a router generated a host list, and callout was configured, Exim
5756 crashed, because it could not find a port number from the (non-existent)
5757 transport. It now assumes port 25 in this circumstance.
5759 21. Added the -t option to exigrep.
5761 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
5762 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
5763 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
5764 them is set, LOOKUP_LSEARCH is forced.
5766 23. "exim -bV" now outputs a list of lookups that are included in the binary.
5768 24. Added sender and host information to the "rejected by local_scan()" log
5769 line; previously there was no indication of these.
5771 25. Added .include_if_exists.
5773 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
5774 Linux. It turns out that not all file systems support this. Apparently some
5775 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
5776 do it.) To cope with this, the error EINVAL, which means that sync-ing is
5777 not supported on the file descriptor, is now ignored when Exim is trying to
5778 sync a directory. This applies only to Linux.
5780 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
5782 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
5783 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
5786 29. Added support for maildirsize files from supplied patch (modified a bit).
5788 30. The use of :fail: followed by an empty string could lead Exim to respond to
5789 sender verification failures with (e.g.):
5791 550 Verification failed for <xxx>
5792 550 Sender verify failed
5794 where the first response line was missing the '-' that indicates it is not
5795 the final line of the response.
5797 31. The loop for finding the name of the user that called Exim had a hardwired
5798 limit of 10; it now uses the value of finduser_retries, which is used for
5799 all other user lookups.
5801 32. Added $received_count variable, available in data and not_smtp ACLs, and at
5804 33. Exim was neglecting to zero errno before one call of strtol() when
5805 expanding a string and expecting an integer value. On some systems this
5806 resulted in spurious "integer overflow" errors. Also, it was casting the
5807 result into an int without checking.
5809 34. Testing for a connection timeout using "timeout_connect" in the retry rules
5810 did not work. The code looks as if it has *never* worked, though it appears
5811 to have been documented since at least releast 1.62. I have made it work.
5813 35. The "timeout_DNS" error in retry rules, also documented since at least
5814 1.62, also never worked. As it isn't clear exactly what this means, and
5815 clearly it isn't a major issue, I have abolished the feature by treating it
5816 as "timeout", and writing a warning to the main and panic logs.
5818 36. The display of retry rules for -brt wasn't always showing the error code
5821 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
5822 timeout_connect_A, timeout_connect_MX.
5824 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
5825 to the empty sender.
5827 39. The daemon was not analysing the content of -oX till after it had closed
5828 stderr and disconnected from the controlling terminal. This meant that any
5829 syntax errors were only noted on the panic log, and the return code from
5830 the command was 0. By re-arranging the code a little, I've made the
5831 decoding happen first, so such errors now appear on stderr, and the return
5832 code is 1. However, the actual setting up of the sockets still happens in
5833 the disconnected process, so errors there are still only recorded on the
5836 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
5837 connections (as happens on some IP stacks) was logged at start up time as
5838 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
5839 it from "IPv6 and IPv4", which means that two separate sockets are being
5842 41. The debug output for gethostbyname2() or getipnodebyname() failures now
5843 says whether AF_INET or AF_INET6 was passed as an argument.
5845 42. Exiwhat output was messed up when time zones were included in log
5848 43. Exiwhat now gives more information about the daemon's listening ports,
5849 and whether -tls-on-connect was used.
5851 44. The "port" option of the smtp transport is now expanded.
5853 45. A "message" modifier in a "warn" statement in a non-message ACL was being
5854 silently ignored. Now an error message is written to the main and panic
5857 46. There's a new ACL modifier called "logwrite" which writes to a log file
5858 as soon as it is encountered.
5860 47. Added $local_user_uid and $local_user_gid at routing time.
5862 48. Exim crashed when trying to verify a sender address that was being
5865 49. Exim was recognizing only a space character after ".include". It now also
5866 recognizes a tab character.
5868 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
5869 extracting the relevant information from the specification. The man page no
5870 longer contains scrambled data for the -d option, and I've added a section
5871 at the front about calling Exim under different names.
5873 51. Added "extra_headers" argument to the "mail" command in filter files.
5875 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
5878 53. Installed eximstats 1.29.
5880 54. Added transport_filter_timeout as a generic transport option.
5882 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
5883 Cc: header lines. This was required by RFC 822, but it not required by RFC
5886 56. Exim used to add From:, Date:, and Message-Id: header lines to any
5887 incoming messages that did not have them. Now it does so only if the
5888 message originates locally, that is, if there is no associated remote host
5889 address. When Resent- header lines are present, this applies to the Resent-
5890 lines rather than the non-Resent- lines.
5892 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
5893 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
5895 58. Messages for configuration errors now include the name of the main
5896 configuration file - useful now that there may be more than one file in a
5897 list (.included file names were always shown).
5899 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
5900 for those rare installations that do not start the daemon as root or run it
5901 setuid root. I've cut out the call to initgroups() if the daemon is not
5904 60. The Exim user and group can now be bound into the binary as text strings
5905 that are looked up at the start of Exim's processing.
5907 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
5909 62. Added $mailstore_basename variable.
5911 63. Installed patch to sieve.c from Michael Haardt.
5913 64. When Exim failed to open the panic log after failing to open the main log,
5914 the original message it was trying to log was written to stderr and debug
5915 output, but if they were not available (the usual case in production), it
5916 was lost. Now it is written to syslog before the two lines that record the
5917 failures to open the logs.
5919 65. Users' Exim filters run in subprocesses under the user's uid. It is
5920 possible for a "deliver" command or an alias in a "personal" command to
5921 provoke an address rewrite. If logging of address rewriting is configured,
5922 this fails because the process is not running as root or exim. There may be
5923 a better way of dealing with this, but for the moment (because 4.30 needs
5924 to be released), I have disabled address rewrite logging when running a
5925 filter in a non-root, non-exim process.
5931 1. The buildconfig auxiliary program wasn't quoting the value set for
5932 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
5933 not defined. This bug was masked in 4.22 by the effect that was fixed in
5936 2. Some messages that were rejected after a message id was allocated were
5937 shown as "incomplete" by exigrep. It no longer does this for messages that
5938 are rejected by local_scan() or the DATA or non-SMTP ACLs.
5940 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
5941 have allow_domain_literals set, the ID did not get logged in the <= line.
5942 Domain literals are now always recognized in Message-ID: header lines.
5944 4. The first argument for a ${extract expansion item is the key name or field
5945 number. Leading and trailing spaces in this item were not being ignored,
5946 causing some misleading effects.
5948 5. When deliver_drop_privilege was set, single queue runner processes started
5949 manually (i.e. by the command "exim -q") or by the daemon (which uses the
5950 same command in the process it spins off) were not dropping privilege.
5952 6. When the daemon running as "exim" started a queue runner, it always
5953 re-executed Exim in the spun-off process. This is a waste of effort when
5954 deliver_drop_privilege is set. The new process now just calls the
5955 queue-runner function directly.
5961 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
5964 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
5965 ignored. Though the use of -oP was forcing the writing of a pid file, it
5966 was always written to the default place.
5968 3. If the message "no IP address found for host xxxx" is generated during
5969 incoming verification, it is now followed by identification of the incoming
5970 connection (so you can more easily find what provoked it).
5972 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
5974 5. Added some features to "harden" Exim a bit more against certain attacks:
5976 (a) There is now a build-time option called FIXED_NEVER_USERS that can
5977 be put in Local/Makefile. This is like the never_users runtime option,
5978 but it cannot be overridden. The default setting is "root".
5980 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
5981 prefix string with which any file named in a -C command line option
5984 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
5985 is retained for -C and -D only if the caller of Exim is root. Without
5986 it, the exim user may also use -C and -D and retain privilege.
5988 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
5989 command line option is disabled.
5991 6. Macro names set by the -D option must start with an upper case letter, just
5992 like macro names defined in the configuration file.
5994 7. Added "dereference=" facility to LDAP.
5996 8. Two instances of the typo "uknown" in the source files are fixed.
5998 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
5999 the Configure-Makefile script screwed up while processing it.
6001 10. Incorporated PCRE 4.4.
6003 11. The SMTP synchronization check was not operating right at the start of an
6004 SMTP session. For example, it could not catch a HELO sent before the client
6005 waited for the greeting. There is now a check for outstanding input at the
6006 point when the greeting is written. Because of the duplex, asynchronous
6007 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
6008 way, but not yet received, when the check is performed.
6010 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
6011 on TCP/IP sockets, because this apparently causes some broken clients to
6014 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
6015 unchanged) from the Cygwin maintainer.
6017 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
6018 was supported instead of testing for mbx. Effectively a typo.
6020 15. The spa authenticator server code was not checking that the input it
6021 received was valid base64.
6023 16. The debug output line for the "set" modifier in ACLs was not showing the
6024 name of the variable that was being set.
6026 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
6028 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
6029 lookup on its own. Something else was needed to trigger the lookup. For
6030 example, a match in host_lookup or the need for a host name in a host list.
6031 Now, if $sender_host_name is referenced and the host name has not yet been
6032 looked up, a lookup is performed. If the lookup fails, the variable remains
6033 empty, and $host_lookup_failed is set to "1".
6035 19. Added "eqi" as a case-independent comparison operator.
6037 20. The saslauthd authentication condition could segfault if neither service
6038 nor realm was specified.
6040 21. If an overflowing value such as "2048M" was set for message_size_limit, the
6041 error message that was logged was misleading, and incoming SMTP
6042 connections were dropped. The message is now more accurate, and temporary
6043 errors are given to SMTP connections.
6045 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
6046 (except RSET) with a 421 error, until QUIT is received. However, it was
6047 failing to send a response to QUIT.
6049 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
6050 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
6051 ACL is now run after the helo_try_verify_hosts code.
6053 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
6054 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
6055 case-independent, so other case variants are also recognized). Apparently
6056 some systems use these upper case variants.
6058 25. If more than two messages were waiting for the same host, and a transport
6059 filter was specified for the transport, Exim sent two messages over the
6060 same TCP/IP connection, and then failed with "socket operation on non-
6061 socket" when it tried to send the third.
6063 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
6065 27. The extern definition of crypt16() in expand.c was not being excluded when
6066 the OS had its own crypt16() function.
6068 28. Added bounce_return_body as a new option, and bounce_return_size_limit
6069 as a preferred synonym for return_size_limit, both as an option and as an
6072 29. Added LIBS=-liconv to OS/Makefile-OSF1.
6074 30. Changed the default configuration ACL to relax the local part checking rule
6075 for addresses that are not in any local domains. For these addresses,
6076 slashes and pipe symbols are allowed within local parts, but the sequence
6077 /../ is explicitly forbidden.
6079 31. SPA server authentication was not clearing the challenge buffer before
6082 32. log_message in a "warn" ACL statement was writing to the reject log as
6083 well as to the main log, which contradicts the documentation and doesn't
6084 seem right (because no rejection is happening). So I have stopped it.
6086 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
6087 However, I am unable to do any testing of this.
6089 34. Fixed an infelicity in the appendfile transport. When checking directories
6090 for a mailbox, to see if any needed to be created, it was accidentally
6091 using path names with one or more superfluous leading slashes; tracing
6092 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
6094 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
6095 contents are added to the log line that is written for every discarded
6096 recipient. (Previously a log_message setting was ignored.)
6098 36. The ${quote: operator now quotes the string if it is empty.
6100 37. The install script runs exim in order to find its version number. If for
6101 some reason other than non-existence or emptiness, which it checks, it
6102 could not run './exim', it was installing it with an empty version number,
6103 i.e. as "exim-". This error state is now caught, and the installation is
6106 38. An argument was missing from the function that creates an error message
6107 when Exim fails to connect to the socket for saslauthd authentication.
6108 This could cause Exim to crash, or give a corrupted message.
6110 39. Added isip, isip4, and isip6 to ${if conditions.
6112 40. The ACL variables $acl_xx are now saved with the message, and can be
6113 accessed later in routers, transports, and filters.
6115 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
6116 strings in the file are not string-expanded.
6118 42. If a MAIL command specified a SIZE value that was too large to fit into an
6119 int variable, the check against message_size_limit failed. Such values are
6120 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
6121 day this will have to be increased, but I don't think I want to be around
6122 when emails are that large.
6129 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
6130 iconv() is not standard in FreeBSD.
6132 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
6133 IPv6 enabled. The observed symptom was a segmentation fault on return from
6134 the function os_common_find_running_interfaces() in src/os.c.
6136 3. In the check_special_case() function in daemon.c I had used "errno" as an
6137 argument name, which causes warnings on some systems. This was basically a
6138 typo, since it was named "eno" in the comments!
6140 4. The code that waits for the clock to tick (at a resolution of some fraction
6141 of a second) so as to ensure message-id uniqueness was always waiting for
6142 at least one whole tick, when it could have waited for less. [This is
6143 almost certainly not relevant at current processor speeds, where it is
6144 unlikely to ever wait at all. But we try to future-proof.]
6146 5. The function that sleeps for a time interval that includes fractions of a
6147 second contained a race. It did not block SIGALRM between setting the
6148 timer, and suspending (a couple of lines later). If the interval was short
6149 and the sigsuspend() was delayed until after it had expired, the suspension
6150 never ended. On busy systems this could lead to processes getting stuck for
6153 6. Some uncommon configurations may cause a lookup to happen in a queue runner
6154 process, before it forks any delivery processes. The open lookup caching
6155 mechanism meant that the open file or database connection was passed into
6156 the delivery process. The problem was that delivery processes always tidy
6157 up cached lookup data. This could cause a problem for the next delivery
6158 process started by the queue runner, because the external queue runner
6159 process does not know about the closure. So the next delivery process
6160 still has data in the lookup cache. In the case of a file lookup, there was
6161 no problem because closing a file descriptor in a subprocess doesn't affect
6162 the parent. However, if the lookup was caching a connection to a database,
6163 the connection was closed, and the second delivery process was likely to
6164 see errors such as "PGSQL: query failed: server closed the connection
6165 unexpectedly". The problem has been fixed by closing all cached lookups
6166 in a queue runner before running a delivery process.
6168 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
6169 seem to have the "const" qualifier which it has on other OS. I've
6172 8. Change 4.21/2 was too strict. It is only if there are two authenticators
6173 *of the same type* (client or server) with the same public name that an
6174 error should be diagnosed.
6176 9. When Exim looked up a host name for an IP address, but failed to find the
6177 original IP address when looking up the host name (a safety check), it
6178 output the message "<ip address> does not match any IP for NULL", which was
6179 confusing, to say the least. The bug was that the host name should have
6180 appeared instead of "NULL".
6182 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
6183 user that is built into the binary, and the -C or -D options is used, root
6184 privilege is dropped before the configuration file is read. In addition,
6185 logging is switched to stderr instead of the normal log files. If the
6186 configuration then re-defines the Exim user, the unprivileged environment
6187 is probably not what is expected, so Exim logs a panic warning message (but
6190 However, if deliver_drop_privilege is set, the unprivileged state may well
6191 be exactly what is intended, so the warning has been cut out in that case,
6192 and Exim is allowed to try to write to its normal log files.
6198 1. smtp_return_error_details was not giving details for temporary sender
6199 or receiver verification errors.
6201 2. Diagnose a configuration error if two authenticators have the same public
6204 3. Exim used not to create the message log file for a message until the first
6205 delivery attempt. This could be confusing when incoming messages were held
6206 for policy or load reasons. The message log file is now created at the time
6207 the message is received, and an initial "Received" line is written to it.
6209 4. The automatically generated man page for command line options had a minor
6210 bug that caused no ill effects; however, a more serious problem was that
6211 the procedure for building the man page automatically didn't always
6212 operate. Consequently, release 4.20 contains an out-of-date version. This
6213 shouldn't happen again.
6215 5. When building Exim with embedded Perl support, the script that builds the
6216 Makefile was calling 'perl' to find its compile-time parameters, ignoring
6217 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
6219 6. The freeze_tell option was not being used for messages that were frozen on
6220 arrival, either by an ACL or by local_scan().
6222 7. Added the smtp_incomplete_transaction log selector.
6224 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
6225 was accepting AUTH without a new EHLO.
6227 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
6228 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
6231 10. Logging of TCP/IP connections (when configured) now happens in the main
6232 daemon process instead of the child process, so that the TCP/IP connection
6233 count is more accurate (but it can never be perfect).
6235 11. The use of "drop" in a nested ACL was not being handled correctly in the
6236 outer ACL. Now, if condition failure induced by the nested "drop" causes
6237 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
6238 or "require"), the connection is dropped.
6240 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
6241 that yield "discard" can now be used with an "accept" or a "discard" verb,
6242 but an error is generated for any others (because I can't see a useful way
6243 to define what should happen).
6245 13. When an ACL is read dynamically from a file (or anywhere else), the lines
6246 are now processed in the same way as lines in the Exim configuration file.
6247 In particular, continuation lines are supported.
6249 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
6251 15. Added -ti meaning -t -i.
6253 16. Check for letters, digits, hyphens, and dots in the names of dnslist
6254 domains, and warn by logging if others are found.
6256 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
6257 returned from GIFCONF when Exim is trying to find the list of interfaces on
6258 a host. The code in os.c has been modified to copy each ifreq to an aligned
6259 structure in all cases.
6261 Also, in some cases, the returned ifreq's were being copied to a 'struct
6262 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
6263 means the last couple of bytes of an IPv6 address could be chopped if the
6264 ifreq contained only a normal sockaddr (14 bytes storage).
6266 18. Named domain lists were not supported in the hosts_treat_as_local option.
6267 An entry such as +xxxx was not recognized, and was treated as a literal
6270 19. Ensure that header lines added by a DATA ACL are included in the reject log
6271 if the ACL subsequently rejects the message.
6273 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
6274 MD5 (which is deprecated).
6276 21. When testing a filter file using -bf, Exim was writing a message when it
6277 took the sender from a "From " line in the message, but it was not doing so
6278 when it took $return_path from a Return-Path: header line. It now does.
6280 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
6281 with a valid header line field name (a series of printing characters
6282 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
6284 23. Changed "disc" in the source to "disk" to conform to the documentation and
6285 the book and for uniformity.
6287 24. Ignore Sendmail's -Ooption=value command line item.
6289 25. When execve() failed while trying to run a command in a pipe transport,
6290 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
6291 could be confused with a return value of 69 from the command itself. This
6292 has been changed to 127, the value the shell returns if it is asked to run
6293 a non-existent command. The wording for the related log line suggests a
6294 non-existent command as the problem.
6296 26. If received_header_text expands to an empty string, do not add a Received:
6297 header line to the message. (Well, it adds a token one on the spool, but
6298 marks it "old" so that it doesn't get used or transmitted.)
6300 27. Installed eximstats 1.28 (addition of -nt option).
6302 28. There was no check for failure on the call to getsockname() in the daemon
6303 code. This can fail if there is a shortage of resources on the system, with
6304 ENOMEM, for example. A temporary error is now given on failure.
6306 29. Contrary to the C standard, it seems that in some environments, the
6307 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
6308 program. Exim now does this explicitly; it affects the formatting of
6309 timestamps using strftime().
6311 30. If exiqsumm was given junk data, it threw up some uninitialized variable
6312 complaints. I've now initialized all the variables, to avoid this.
6314 32. Header lines added by a system filter were not being "seen" during
6315 transport-time rewrites.
6317 33. The info_callback() function passed to OpenSSL is set up with type void
6318 (*)(SSL *, int, int), as described somewhere. However, when calling the
6319 function (actually a macro) that sets it up, the type void(*)() is
6320 expected. I've put in a cast to prevent warnings from picky compilers.
6322 34. If a DNS black list lookup found a CNAME record, but there were no A
6323 records associated with the domain it pointed at, Exim crashed.
6325 35. If a DNS black list lookup returned more than one A record, Exim ignored
6326 all but the first. It now scans all returned addresses if a particular IP
6327 value is being sought. In this situation, the contents of the
6328 $dnslist_value variable are a list of all the addresses, separated by a
6331 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
6332 to accept a host name and all its aliases if the forward lookup for any of
6333 them yielded the IP address of the incoming connection. Now it accepts only
6334 those names whose forward lookup yields the correct IP address. Any other
6335 names are discarded. This closes a loophole whereby a rogue DNS
6336 administrator could create reverse DNS records to break through a
6337 wildcarded host restriction in an ACL.
6339 37. If a user filter or a system filter that ran in a subprocess used any of
6340 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
6341 the wrong values were passed to the pipe command ($thisaddress had the
6342 value of $0, $0 had the value of $1, etc). This bug was introduced by
6343 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
6345 38. Improved the line breaking for long SMTP error messages from ACLs.
6346 Previously, if there was no break point between 40 and 75 characters, Exim
6347 left the rest of the message alone. Two changes have been made: (a) I've
6348 reduced the minimum length to 35 characters; (b) if it can't find a break
6349 point between 35 and 75 characters, it looks ahead and uses the first one
6350 that it finds. This may give the occasional overlong line, but at least the
6351 remaining text gets split now.
6353 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
6354 file descriptors might be low, and that setting 1000 would always raise it.
6355 It turns out that in some environments, the limit is already over 1000 and
6356 that lowering it causes trouble. So now Exim takes care not to decrease it.
6358 40. When delivering a message, the value of $return_path is set to $sender_
6359 address at the start of routing (routers may change the value). By an
6360 oversight, this default was not being set up when an address was tested by
6361 -bt or -bv, which affected the outcome if any router or filter referred to
6364 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
6365 log only when "message" or "log_message" are set. However, if one of the
6366 conditions was an address verification, or a call to a nested ACL, the
6367 messages generated by the underlying test were being passed through. This
6368 no longer happens. The underlying message is available in $acl_verify_
6369 message for both "message" and "log_message" expansions, so it can be
6370 passed through if needed.
6372 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
6373 new expansion $bh_ to give the encoded byte string without charset
6374 translation. Translation happens only if iconv() is available; HAVE_ICONV
6375 indicates this at build time. HEADERS_CHARSET gives the charset to
6376 translate to; headers_charset can change it in the configuration, and
6377 "headers charset" can change it in an individual filter file.
6379 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
6380 that creates RFC 2047 encoded "words" labels them as that charset instead
6381 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
6382 expansion operator; (ii) when Exim creates a From: line for a local
6383 message; (iii) when a header line is rewritten to include a "phrase" part.
6385 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
6386 buggy, causing it to skip the first lines of messages whose message ID
6387 ended in 'D'. This would not have bitten before Exim release 4.14, because
6388 message IDs were unlikely to end in 'D' before then. The effect was to have
6389 incorrect size information for certain domains.
6391 45. #include "config.h" was missing at the start of the crypt16.c module. This
6392 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
6395 46. If there was a timeout during a "random" callout check, Exim treated it as
6396 a failure of the random address, and carried on sending RSET and the real
6397 address. If the delay was just some slowness somewhere, the response to the
6398 original RCPT would be taken as a response to RSET and so on, causing
6399 mayhem of various kinds.
6401 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
6402 when I implemented it. It didn't allow for the fact that some option values
6403 may legitimatetly be negative (e.g. size_addition), and it didn't even do
6404 the right test for positive values.
6406 48. Domain names in DNS records are case-independent. Exim always looks them up
6407 in lower case. Some resolvers return domain names in exactly the case they
6408 appear in the zone file, that is, they may contain uppercase letters. Not
6409 all resolvers do this - some return always lower case. Exim was treating a
6410 change of case by a resolver as a change of domain, similar to a widening
6411 of a domain abbreviation. This triggered its re-routing code and so it was
6412 trying to route what was effectively the same domain again. This normally
6413 caused routing to fail (because the router wouldn't handle the domain
6414 twice). Now Exim checks for this case specially, and just changes the
6415 casing of the domain that it ultimately uses when it transmits the message
6418 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
6421 50. If a filter generated a file delivery with a non-absolute name (possible if
6422 no home directory exists for the router), the forbid_file option was not
6425 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
6426 the existing equality matching.
6428 52. Exim was using ints instead of ino_t variables in some places where it was
6429 dealing with inode numbers.
6431 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
6432 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
6433 called TMPDIR, and if it finds it is different, it changes its value.
6435 54. The smtp_printf() function is now made available to local_scan() so
6436 additional output lines can be written before returning. There is also an
6437 smtp_fflush() function to enable the detection of a dropped connection.
6438 The variables smtp_input and smtp_batched_input are exported to
6441 55. Changed the default runtime configuration: the message "Unknown user"
6442 has been removed from the ACL, and instead placed on the localuser router,
6443 using the cannot_route_message feature. This means that any verification
6444 failures that generate their own messages won't get overridden. Similarly,
6445 the "Unrouteable address" message that was in the ACL for unverifiable
6446 relay addresses has also been removed.
6448 56. Added hosts_avoid_esmtp to the smtp transport.
6450 57. The exicyclog script was not checking for the esoteric option
6451 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
6452 will work only if exicyclog is run under the appropriate euid.
6454 58. Following a discussion on the list, the rules by which Exim recognises line
6455 endings on incoming messages have been changed. The -dropcr and drop_cr
6456 options are now no-ops, retained only for backwards compatibility. The
6457 following line terminators are recognized: LF CRLF CR. However, special
6458 processing applies to CR:
6460 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
6461 nor a local message in the state where . is a terminator.
6463 (ii) If a bare CR is encountered in a header line, an extra space is added
6464 after the line terminator so as not to end the header. The reasoning
6465 behind this is that bare CRs in header lines are most likely either
6466 to be mistakes, or people trying to play silly games.
6468 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
6469 was being incorrectly given as 18 bytes larger than it should have been.
6470 This is a VOB (very old bug).
6472 60. This may never have affected anything current, but just in case it has:
6473 When the local host is found other than at the start of a list of hosts,
6474 the local host, those with the same MX, and any that follow, are discarded.
6475 When the list in question was part of a longer list of hosts, the following
6476 hosts (not currently being processed) were also being discarded. This no
6477 longer happens. I'm not sure if this situation could ever has previously
6480 61. Added the "/MX" feature to lists of hosts in the manualroute and query
6483 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
6484 header. This is something that is recommended in a new Internet Draft, and
6485 is something that is documented as being done by Sendmail. There are two
6486 possible values. For messages generated by the autoreply transport, Exim
6489 Auto-Submitted: auto-replied
6491 whereas for all other generated messages (e.g. bounces) it adds
6493 Auto-Submitted: auto-generated
6495 63. The "personal" condition in filters now includes a test for the
6496 Auto-Submitted: header. If it contains the string "auto-" the message it
6497 not considered personal.
6499 64. Added rcpt_include_affixes as a generic transport option.
6501 65. Added queue_only_override (default true).
6503 66. Added the syslog_duplication option.
6505 67. If what should have been the first header line of a message consisted of
6506 a space followed by a colon, Exim was mis-interpreting it as a header line.
6507 It isn't of course - it is syntactically invalid and should therefore be
6508 treated as the start of the message body. The misbehaviour could have
6509 caused a number of strange effects, including loss of data in subsequent
6510 header lines, and spool format errors.
6512 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
6513 client host had authenticated. This control can now be exercised by an ACL
6514 for more flexibility.
6516 69. By default, callouts do not happen when testing with -bh. There is now a
6517 variant, -bhc, which does actually run the callout code, including
6518 consulting and updating the callout cache.
6520 70. Added support for saslauthd authentication, courtesy of Alexander
6523 71. If statvfs() failed on the spool or log directories while checking their
6524 size for availability, Exim confusingly gave the error "space shortage".
6525 Furthermore, in debugging mode it crashed with a floating point exception.
6526 These checks are done if check_{spool,log}_{space,inodes} are set, and when
6527 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
6528 serious problem, Exim now writes to the main and panic logs when this
6529 happens, with details of the failure. It then refuses to accept the
6530 incoming message, giving the message "spool directory problem" or "log
6531 directory problem" with a 421 code for SMTP messages.
6533 72. When Exim is about to re-exec itself, it ensures that the file descriptors
6534 0, 1, and 2 exist, because some OS complain for execs without them (see
6535 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
6536 descriptors. However, the code omitted to check that the open succeeded,
6537 causing mysterious errors if for some reason the permissions on /dev/null
6538 got screwed. Now Exim writes a message to the main and panic logs, and
6539 bombs out if it can't open /dev/null.
6541 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
6542 interact so that it is all more flexible. It is supposed to remain
6543 backwards compatible. Also added extra_local_interfaces.
6545 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
6546 to bomb out with an assertion failure - to the client this appears as a
6547 connection drop. This problem occurs in the part of the code that was taken
6548 from the Samba project. Fortunately, the assertion is in a very simple
6549 function, so I have fixed this by reproducing the function inline in the
6550 one place where it is called, and arranging for authentication to fail
6551 instead of killing the process with assert().
6553 75. The SPA client code was not working when the server requested OEM rather
6554 than Unicode encoding.
6556 76. Added code to make require_files with a specific uid setting more usable in
6557 the case where statting the file as root fails - usually a non-root-mounted
6558 NFS file system. When this happens and the failure is EACCES, Exim now
6559 forks a subprocess and does the per-uid checking as the relevant uid.
6561 77. Added process_log_path.
6563 78. If log_file_path was not explicitly set, a setting of check_log_space or
6564 check_log_inodes was ignored.
6566 79. If a space check for the spool or log partitions fails, the incident is now
6567 logged. Of course, in the latter case the data may get lost...
6569 80. Added the %p formatting code to string_format() so that it can be used to
6570 print addresses in debug_print(). Adjusted all the address printing in the
6571 debugging in store.c to use %p rather than %d.
6573 81. There was a concern that a line of code in smtp_in.c could overflow a
6574 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
6575 initially expressed, the concern was not well-founded, because trailing
6576 spaces are removed early. However, if the trailing spaces were followed by
6577 a NULL, they did not get removed, so the overflow was possible. Two fixes
6580 (a) I re-wrote the offending code in a cleaner fashion.
6581 (b) If an incoming SMTP command contains a NULL character, it is rejected
6584 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
6585 runs initgroups(), so that if the Exim user is in any additional groups,
6586 they will be used during message reception.
6592 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob