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.
71 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
72 and 3.4.4 - once the server is enabled to respond to an OCSP request
73 it does even when not requested, resulting in a stapling non-aware
74 client dropping the TLS connection.
76 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
77 support variable-length bit vectors. No functional change.
79 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
80 The I= interface field on outgoing lines is now after the H= remote
81 host field, same as incoming lines. There is a separate
82 outgoing_interface log selector which allows you to disable the
85 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
86 If not running log_selector +smtp_connection the mainlog would be held
87 open indefinitely after a "too many connections" event, including to a
88 deleted file after a log rotate. Leave the per net connection logging
89 leaving it open for efficiency as that will be quickly detected by the
90 check on the next write.
92 HS/01 Bug 1671: Fix post transport crash.
93 Processing the wait-<transport> messages could crash the delivery
94 process if the message IDs didn't exist for some reason. When
95 using 'split_spool_directory=yes' the construction of the spool
96 file name failed already, exposing the same netto behaviour.
98 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
99 mime_regex ACL conditions.
101 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
102 to DSN fail messages (bounces): remote IP, remote greeting, remote response
103 to HELO, local diagnostic string.
105 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
106 log line to debug. Even when configured with a tls authenticator many
107 client connections are expected to not authenticate in this way, so
108 an authenticate fail is not an error.
110 HS/02 Add the Exim version string to the process info. This way exiwhat
111 gives some more detail about the running daemon.
113 JH/06 Bug 1395: time-limit cacheing of DNS lookups, to the TTL value. This may
114 matter for fast-change records such as DNSBLs.
116 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
117 retry record, even if constant. There may be multiple transports with
118 different interface settings and the retry behaviour needs to be kept
121 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
123 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
125 JH/10 Bug 840: fix log_defer_output option of pipe transport
127 JH/11 Bug 830: use same host for all RCPTS of a message, even under
128 hosts_randomize. This matters a lot when combined with mua_wrapper.
130 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
131 ${quote_pgsql:<string>} operator.
133 JH/13 Bug 1708: avoid misaligned access in cached lookup.
135 JH/14 Change header file name for freeradius-client. Relevant if compiling
136 with Radius support; from the Gentoo tree and checked under Fedora.
138 JH/15 Bug 1712: Introduce $prdr_requested flag variable
140 JH/16 Bug 1714: Permit an empty string as expansion result for transport
141 option transport_filter, meaning no filtering.
143 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
145 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
146 defaults to "*" (all hosts). The variable is now available when not built
147 with TLS, default unset, mainly to enable keeping the testuite sane.
148 If a server certificate is not supplied (via tls_certificate) an error is
149 logged, and clients will find TLS connections fail on startup. Presumably
150 they will retry in-clear.
151 Packagers of Exim are strongly encouraged to create a server certificate
152 at installation time.
154 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
155 with the $config_file variable.
157 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
158 in transport context, after the attempt, and per-recipient. The latter type
159 is per host attempted. The event data is the error message, and the errno
160 information encodes the lookup type (A vs. MX) used for the (first) host,
161 and the trailing two digits of the smtp 4xx reponse.
163 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
164 to write to mainlog (or rejectlog, paniclog) in the window between file
165 creation and permissions/ownership being changed. Particularly affects
166 installations where exicyclog is run as root, rather than exim user;
167 result is that the running daemon panics and dies.
169 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
171 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
172 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
173 "pri" and "weight". Note that the previous implicit priority given by the
174 list order is no longer honoured.
176 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalisation
179 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
180 by defining SUPPORT_SOCKS.
182 JH/26 Move PROXY support from Experimental to mainline, enabled for a build
183 by defining SUPPORT_PROXY. Note that the proxy_required_hosts option
184 is renamed to hosts_proxy, and the proxy_{host,target}_{address,port}.
185 variables are renamed to proxy_{local,external}_{address,port}.
187 JH/27 Move Internationalisation support from Experimental to mainline, enabled
188 for a build by defining SUPPORT_I18N
190 JH/28 Bug 1745: Fix redis lookups to handle (quoted) spaces embedded in parts
191 of the query string, and make ${quote_redis:} do that quoting.
193 JH/29 Move Events support from Experimental to mainline, enabled by default
194 and removable for a build by defining DISABLE_EVENT.
196 JH/30 Updated DANE implementation code to current from Viktor Dukhovni.
198 JH/31 Fix bug with hosts_connection_nolog and named-lists which were wrongly
199 cached by the daemon.
201 JH/32 Move Redis support from Experimental to mainline, enabled for a build
202 by defining LOOKUP_REDIS. The libhiredis library is required.
204 JH/33 Bug 1748: Permit ACL dnslists= condition in non-smtp ACLs if explicit
205 keys are given for lookup.
207 JH/34 Bug 1192: replace the embedded copy of PolarSSL RSA routines in the DKIM
208 support, by using OpenSSL or GnuTLS library ones. This means DKIM is
209 only supported when built with TLS support. The PolarSSL SHA routines
210 are still used when the TLS library is too old for convenient support.
212 JH/35 Require SINGLE_DH_USE by default in OpenSSL (main config option
213 openssl_options), for security. OpenSSL forces this from version 1.1.0
214 server-side so match that on older versions.
216 JH/36 Bug 1778: longstanding bug in memory use by the ${run } expansion: A fresh
217 allocation for $value could be released as the expansion processing
218 concluded, but leaving the global pointer active for it.
220 JH/37 Bug 1769: Permit a VRFY ACL to override the default 252 response,
221 and to use the domains and local_parts ACL conditions.
223 JH/38 Fix cutthrough bug with body lines having a single dot. The dot was
224 incorrectly not doubled on cutthrough transmission, hence seen as a
225 body-termination at the receiving system - resulting in truncated mails.
226 Commonly the sender saw a TCP-level error, and retransmitted the message
227 via the normal store-and-forward channel. This could result in duplicates
228 received - but deduplicating mailstores were liable to retain only the
229 initial truncated version.
231 JH/39 Bug 1781: Fix use of DKIM private-keys having trailing '=' in the base-64.
233 JH/40 Fix crash in queryprogram router when compiled with EXPERIMENTAL_SRS.
235 JH/41 Bug 1792: Fix selection of headers to sign for DKIM: bottom-up. While
236 we're in there, support oversigning also; bug 1309.
238 JH/42 Bug 1796: Fix error logged on a malware scanner connection failure.
240 HS/04 Add support for keep_environment and add_environment options.
242 JH/43 Tidy coding issues detected by gcc --fsanitize=undefined. Some remain;
243 either intentional arithmetic overflow during PRNG, or testing config-
246 JH/44 Bug 1800: The combination of a -bhc commandline option and cutthrough
247 delivery resulted in actual delivery. Cancel cutthrough before DATA
250 JH/45 Fix cutthrough, when connection not opened by verify and target hard-
251 rejects a recipient: pass the reject to the originator.
253 JH/46 Multiple issues raised by Coverity. Some were obvious or plausible bugs.
254 Many were false-positives and ignorable, but it's worth fixing the
257 JH/47 Fix build on HP-UX and older Solaris, which need (un)setenv now also
258 for the new environment-manipulation done at startup. Move the routines
259 from being local to tls.c to being global via the os.c file.
261 JH/48 Bug 1807: Fix ${extract } for the numeric/3-string case. While preparsing
262 an extract embedded as result-arg for a map, the first arg for extract
263 is unavailable so we cannot tell if this is a numbered or keyed
264 extraction. Accept either.
270 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
273 JH/02 The smtp transport option "multi_domain" is now expanded.
275 JH/03 The smtp transport now requests PRDR by default, if the server offers
278 JH/04 Certificate name checking on server certificates, when exim is a client,
279 is now done by default. The transport option tls_verify_cert_hostnames
280 can be used to disable this per-host. The build option
281 EXPERIMENTAL_CERTNAMES is withdrawn.
283 JH/05 The value of the tls_verify_certificates smtp transport and main options
284 default to the word "system" to access the system default CA bundle.
285 For GnuTLS, only version 3.0.20 or later.
287 JH/06 Verification of the server certificate for a TLS connection is now tried
288 (but not required) by default. The verification status is now logged by
289 default, for both outbound TLS and client-certificate supplying inbound
292 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
295 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
296 Status Notification (bounce) messages are now MIME format per RFC 3464.
297 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
298 under the control of the dsn_advertise_hosts option, and routers may
299 have a dsn_lasthop option.
301 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
302 default, modifiable by a malware= option. The list separator for
303 the options can now be changed in the usual way. Bug 68.
305 JH/10 The smtp_receive_timeout main option is now expanded before use.
307 JH/11 The incoming_interface log option now also enables logging of the
308 local interface on delivery outgoing connections.
310 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
311 if the interface and destination host and port all match.
313 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
316 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
317 Patch from Andrew Lewis.
319 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
320 now supports optional time-restrictions, weighting, and priority
321 modifiers per server. Patch originally by <rommer@active.by>.
323 JH/16 The spamd_address main option now supports a mixed list of local
324 and remote servers. Remote servers can be IPv6 addresses, and
325 specify a port-range.
327 JH/17 Bug 68: The spamd_address main option now supports an optional
328 timeout value per server.
330 JH/18 Bug 1581: Router and transport options headers_add/remove can
331 now have the list separator specified.
333 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
336 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
339 JH/21 Support for the A6 type of dns record is withdrawn.
341 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
342 rather than the verbs used.
344 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
345 from 255 to 1024 chars.
347 JH/24 Verification callouts now attempt to use TLS by default.
349 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
350 are generic router options now. The defaults didn't change.
352 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
353 Original patch from Alexander Shikoff, worked over by JH.
355 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
356 files if ZCAT_COMMAND is not executable.
358 JH/26 Bug 1539: Add timout/retry options on dnsdb lookups.
360 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
362 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
363 Normally benign, it bites when the pair was led to by a CNAME;
364 modern usage is to not canoicalize the domain to a CNAME target
365 (and we were inconsistent anyway for A-only vs AAAA+A).
367 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
369 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
370 when evaluating $sender_host_dnssec.
372 JH/31 Check the HELO verification lookup for DNSSEC, adding new
373 $sender_helo_dnssec variable.
375 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
377 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
379 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
381 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
382 documented as working, but never had. Support all but $spam_report.
384 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
385 added for tls authenticator.
387 HS/03 Add perl_taintmode main config option
392 TL/01 When running the test suite, the README says that variables such as
393 no_msglog_check are global and can be placed anywhere in a specific
394 test's script, however it was observed that placement needed to be near
395 the beginning for it to behave that way. Changed the runtest perl
396 script to read through the entire script once to detect and set these
397 variables, reset to the beginning of the script, and then run through
398 the script parsing/test process like normal.
400 TL/02 The BSD's have an arc4random API. One of the functions to induce
401 adding randomness was arc4random_stir(), but it has been removed in
402 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
403 function when detected.
405 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
406 cause callback expansion.
408 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
409 syntax errors in an expansion can be treated as a string instead of
410 logging or causing an error, due to the internal use of bool_lax
411 instead of bool when processing it.
413 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
414 server certificates when making smtp deliveries.
416 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
418 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
420 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
422 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
423 Merged patch from Sebastian Wiedenroth.
425 JH/05 Fix results-pipe from transport process. Several recipients, combined
426 with certificate use, exposed issues where response data items split
427 over buffer boundaries were not parsed properly. This eventually
428 resulted in duplicates being sent. This issue only became common enough
429 to notice due to the introduction of conection certificate information,
430 the item size being so much larger. Found and fixed by Wolfgang Breyha.
432 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
433 size buffer was used, resulting in syntax errors when an expansion
436 JH/07 Add support for directories of certificates when compiled with a GnuTLS
437 version 3.3.6 or later.
439 JH/08 Rename the TPDA expermimental facility to Event Actions. The #ifdef
440 is EXPERIMENTAL_EVENT, the main-configuration and transport options
441 both become "event_action", the variables become $event_name, $event_data
442 and $event_defer_errno. There is a new variable $verify_mode, usable in
443 routers, transports and related events. The tls:cert event is now also
444 raised for inbound connections, if the main configuration event_action
447 TL/06 In test suite, disable OCSP for old versions of openssl which contained
448 early OCSP support, but no stapling (appears to be less than 1.0.0).
450 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
451 server certificate names available under the smtp transport option
452 "tls_verify_cert_hostname" now do not permit multi-component wildcard
455 JH/10 Time-related extraction expansions from certificates now use the main
456 option "timezone" setting for output formatting, and are consistent
457 between OpenSSL and GnuTLS compilations. Bug 1541.
459 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
460 encoded parameter in the incoming message. Bug 1558.
462 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
463 include certificate info, eximon was claiming there were spoolfile
466 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
468 JH/14 Log delivery-related information more consistently, using the sequence
469 "H=<name> [<ip>]" wherever possible.
471 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
472 are problematic for Debian distribution, omit them from the release
475 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
477 JH/16 Fix string representation of time values on 64bit time_t anchitectures.
480 JH/17 Fix a null-indirection in certextract expansions when a nondefault
481 output list separator was used.
486 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
487 checkers that were complaining about end of non-void function with no
490 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
491 This was a regression intruduced in 4.83 by another bugfix.
493 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
495 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
496 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
502 TF/01 Correctly close the server side of TLS when forking for delivery.
504 When a message was received over SMTP with TLS, Exim failed to clear up
505 the incoming connection properly after forking off the child process to
506 deliver the message. In some situations the subsequent outgoing
507 delivery connection happened to have the same fd number as the incoming
508 connection previously had. Exim would try to use TLS and fail, logging
509 a "Bad file descriptor" error.
511 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
512 utilities have not been installed.
514 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
515 temporary space as the ACL may create new global variables.
517 TL/01 LDAP support uses per connection or global context settings, depending
518 upon the detected version of the libraries at build time.
520 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
521 to extract and use the src ip:port in logging and expansions as if it
522 were a direct connection from the outside internet. PPv2 support was
523 updated based on HAProxy spec change in May 2014.
525 JH/02 Add ${listextract {number}{list}{success}{fail}}.
527 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
528 Properly escape header and check for NULL return.
530 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
533 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
535 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
536 characters in header names, implemented as a verify condition.
537 Contributed by Michael Fischer v. Mollard.
539 TL/05 Rename SPF condition results err_perm and err_temp to standardized
540 results permerror and temperror. Previous values are deprecated but
541 still accepted. In a future release, err_perm and err_temp will be
542 completely removed, which will be a backward incompatibility if the
543 ACL tests for either of these two old results. Patch contributed by
544 user bes-internal on the mailing list.
546 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
548 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
549 selectors, in both main and reject logs.
551 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
554 JH/07 Add malware type "sock" for talking to simple daemon.
556 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
558 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
559 routers/transports under cutthrough routing.
561 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
562 numbers. Touch up "bool" conditional to keep the same definition.
564 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
566 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
568 JH/12 Expand items in router/transport headers_add or headers_remove lists
569 individually rather than the list as a whole. Bug 1452.
571 Required for reasonable handling of multiple headers_ options when
572 they may be empty; requires that headers_remove items with embedded
573 colons must have them doubled (or the list-separator changed).
575 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
576 view the policy declared in the DMARC record. Currently, $dmarc_status
577 is a combined value of both the record presence and the result of the
580 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
582 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
583 dnslookup router and the smtp transport (applying to the forward
586 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
587 of ldap servers used for a specific lookup. Patch provided by Heiko
590 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
591 New variable $lookup_dnssec_authenticated for observability.
593 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
594 Patch submitted by Lars Timman.
596 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
598 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
599 Requires trusted mode and valid format message id, aborts otherwise.
600 Patch contributed by Heiko Schlichting.
602 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
603 certextract with support for various fields. Bug 1358.
605 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
606 is requested by default, modifiable by smtp transport option
609 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
610 operate on certificate variables to give certificate fingerprints
611 Also new ${sha256:cert_variable}.
613 JH/23 The PRDR feature is moved from being Experimental into the mainline.
615 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
616 Christian Aistleitner.
618 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
620 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
621 file. Patch from Wolfgang Breyha.
623 JH/25 Expand the coverage of the delivery $host and $host_address to
624 client authenticators run in verify callout. Bug 1476.
626 JH/26 Port service names are now accepted for tls_on_connect_ports, to
627 align with daemon_smtp_ports. Bug 72.
629 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
630 support and error reporting did not work properly.
632 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
633 and is readable. Patch from Andrew Colin Kissa.
635 TL/14 Enhance documentation of ${run expansion and how it parses the
636 commandline after expansion, particularly in the case when an
637 unquoted variable expansion results in an empty value.
639 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
641 PP/02 Fix internal collision of T_APL on systems which support RFC3123
642 by renaming away from it. Addresses GH issue 15, reported by
645 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
647 TL/15 SECURITY: prevent double expansion in math comparison functions
648 (can expand unsanitized data). Not remotely exploitable.
655 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
657 PP/02 Make -n do something, by making it not do something.
658 When combined with -bP, the name of an option is not output.
660 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
663 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
664 $sender_host_name and config options to manage this, and basic check
667 PP/05 DSCP support for outbound connections and control modifier for inbound.
669 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
670 (Only plugin which currently uses this is kerberos4, which nobody should
671 be using, but we should make it available and other future plugins might
672 conceivably use it, even though it would break NAT; stuff *should* be
673 using channel bindings instead).
675 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
676 name; added for Sendmail compatibility; requires admin caller.
677 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
678 just ignore it); requires trusted caller.
679 Also parse but ignore: -Ac -Am -X<logfile>
682 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
684 TL/02 Add +smtp_confirmation as a default logging option.
686 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
687 Patch by Magnus Holmgren from 2007-02-20.
689 TL/04 Bugzilla 1281 - Spec typo.
690 Bugzilla 1283 - Spec typo.
691 Bugzilla 1290 - Spec grammar fixes.
693 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
695 TL/06 Add Experimental DMARC support using libopendmarc libraries.
697 TL/07 Fix an out of order global option causing a segfault. Reported to dev
698 mailing list by by Dmitry Isaikin.
700 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
702 JH/02 Support "G" suffix to numbers in ${if comparisons.
704 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
706 NM/01 Bugzilla 1197 - Spec typo
707 Bugzilla 1196 - Spec examples corrections
709 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
711 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
712 gnutls_enable_pkcs11, but renamed to more accurately indicate its
715 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
716 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
718 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
719 "acl {{name}{arg}...}", and optional args on acl condition
722 JH/05 Permit multiple router/transport headers_add/remove lines.
724 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
726 JH/07 Avoid using a waiting database for a single-message-only transport.
727 Performance patch from Paul Fisher. Bugzilla 1262.
729 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
732 JH/09 Add $headers_added variable, with content from use of ACL modifier
733 add_header (but not yet added to the message). Bugzilla 199.
735 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
736 Pulled from Bugzilla 817 by Wolfgang Breyha.
738 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
740 (nb: this is the same fix as in Exim 4.80.1)
742 JH/11 Add A= logging on delivery lines, and a client_set_id option on
745 JH/12 Add optional authenticated_sender logging to A= and a log_selector
748 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
750 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
751 advertise SMTP AUTH mechanism to us, instead of a generic
752 protocol violation error. Also, make Exim more robust to bad
753 data from the Dovecot auth socket.
755 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
757 When a queue runner is handling a message, Exim first routes the
758 recipient addresses, during which it prunes them based on the retry
759 hints database. After that it attempts to deliver the message to
760 any remaining recipients. It then updates the hints database using
763 So if a recipient address works intermittently, it can get repeatedly
764 deferred at routing time. The retry hints record remains fresh so the
765 address never reaches the final cutoff time.
767 This is a fairly common occurrence when a user is bumping up against
768 their storage quota. Exim had some logic in its local delivery code
769 to deal with this. However it did not apply to per-recipient defers
770 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
772 This change adds a proper retry rule check during routing so that the
773 final cutoff time is checked against the message's age. We only do
774 this check if there is an address retry record and there is not a
775 domain retry record; this implies that previous attempts to handle
776 the address had the retry_use_local_parts option turned on. We use
777 this as an approximation for the destination being like a local
778 delivery, as in LMTP.
780 I suspect this new check makes the old local delivery cutoff check
781 redundant, but I have not verified this so I left the code in place.
783 TF/02 Correct gecos expansion when From: is a prefix of the username.
785 Test 0254 submits a message to Exim with the header
789 When I ran the test suite under the user fanf2, Exim expanded
790 the header to contain my full name, whereas it should have added
791 a Resent-Sender: header. It erroneously treats any prefix of the
792 username as equal to the username.
794 This change corrects that bug.
796 GF/01 DCC debug and logging tidyup
797 Error conditions log to paniclog rather than rejectlog.
798 Debug lines prefixed by "DCC: " to remove any ambiguity.
800 TF/03 Avoid unnecessary rebuilds of lookup-related code.
802 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
803 Bug spotted by Jeremy Harris; was flawed since initial commit.
804 Would have resulted in OCSP responses post-SNI triggering an Exim
805 NULL dereference and crash.
807 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
809 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
810 Bug detection, analysis and fix by Samuel Thibault.
811 Bugzilla 1331, Debian bug #698092.
813 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
815 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
816 Server implementation by Todd Lyons, client by JH.
817 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
818 config variable "prdr_enable" controls whether the server
819 advertises the facility. If the client requests PRDR a new
820 acl_data_smtp_prdr ACL is called once for each recipient, after
821 the body content is received and before the acl_smtp_data ACL.
822 The client is controlled by bolth of: a hosts_try_prdr option
823 on the smtp transport, and the server advertisement.
824 Default client logging of deliveries and rejections involving
825 PRDR are flagged with the string "PRDR".
827 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
828 fclose(). Diagnosis by Todd Lyons.
830 PP/17 Update configure.default to handle IPv6 localhost better.
831 Patch by Alain Williams (plus minor tweaks).
834 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
835 This is now consistent with GnuTLS, and is now documented: the
836 previous undocumented portable approach to treating the option as
837 unset was to force an expansion failure. That still works, and
838 an empty string is now equivalent.
840 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
841 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
842 not performing validation itself.
844 PP/20 Added force_command boolean option to pipe transport.
845 Patch from Nick Koston, of cPanel Inc.
847 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
850 TF/04 Added udpsend ACL modifer and hexquote expansion operator
852 PP/21 Fix eximon continuous updating with timestamped log-files.
853 Broken in a format-string cleanup in 4.80, missed when I repaired the
854 other false fix of the same issue.
855 Report and fix from Heiko Schlichting.
858 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
859 Report from Prashanth Katuri.
861 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
862 It's SecureTransport, so affects any MacOS clients which use the
863 system-integrated TLS libraries, including email clients.
865 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
866 using a MIME ACL for non-SMTP local injection.
867 Report and assistance in diagnosis by Warren Baker.
869 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
871 JH/16 Fix comparisons for 64b. Bugzilla 1385.
873 TL/09 Add expansion variable $authenticated_fail_id to keep track of
874 last id that failed so it may be referenced in subsequent ACL's.
876 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
879 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
880 ldap library initialization, allowing self-signed CA's to be
881 used. Also properly sets require_cert option later in code by
882 using NULL (global ldap config) instead of ldap handle (per
883 session). Bug diagnosis and testing by alxgomz.
885 TL/12 Enhanced documentation in the ratelimit.pl script provided in
886 the src/util/ subdirectory.
888 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
889 renamed to Transport Post Delivery Action by Jeremy Harris, as
892 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
893 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
894 redis_servers = needs to be configured which will be used by the redis
895 lookup. Patch from Warren Baker, of The Packet Hub.
897 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
899 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
900 hostname or reverse DNS when processing a host list. Used suggestions
901 from multiple comments on this bug.
903 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
905 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
906 Missed a few lines, added it to make the runtest require no keyboard
909 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
910 contains upper case chars. Make router use caseful_local_part.
912 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
913 support when GnuTLS has been built with p11-kit.
919 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
921 This, or similar/improved, will also be change PP/11 of 4.82.
927 PP/01 Handle short writes when writing local log-files.
928 In practice, only affects FreeBSD (8 onwards).
929 Bugzilla 1053, with thanks to Dmitry Isaikin.
931 NM/01 Bugzilla 949 - Documentation tweak
933 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
936 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
938 PP/02 Implemented gsasl authenticator.
940 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
942 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
943 `pkg-config foo` for cflags/libs.
945 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
946 with rest of GSASL and with heimdal_gssapi.
948 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
949 `pkg-config foo` for cflags/libs for the TLS implementation.
951 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
952 properties get this fed in as external SSF. A number of robustness
953 and debugging improvements to the cyrus_sasl authenticator.
955 PP/08 cyrus_sasl server now expands the server_realm option.
957 PP/09 Bugzilla 1214 - Log authentication information in reject log.
958 Patch by Jeremy Harris.
960 PP/10 Added dbmjz lookup type.
962 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
964 PP/12 MAIL args handles TAB as well as SP, for better interop with
965 non-compliant senders.
966 Analysis and variant patch by Todd Lyons.
968 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
969 Bug report from Lars Müller <lars@samba.org> (via SUSE),
970 Patch from Dirk Mueller <dmueller@suse.com>
972 PP/13 tls_peerdn now print-escaped for spool files.
973 Observed some $tls_peerdn in wild which contained \n, which resulted
974 in spool file corruption.
976 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
977 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
978 or write after TLS renegotiation, which otherwise led to messages
981 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
982 as a tracking header (ie: a signed header comes before the signature).
983 Patch from Wolfgang Breyha.
985 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
986 comma-sep list; embedded commas doubled.
988 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
990 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
992 Report and patch from Dmitry Banschikov.
994 PP/16 Removed "dont_insert_empty_fragments" fron "openssl_options".
995 Removed SSL_clear() after SSL_new() which led to protocol negotiation
996 failures. We appear to now support TLS1.1+ with Exim.
998 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
999 lets Exim select keys and certificates based upon TLS SNI from client.
1000 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
1001 before an outbound SMTP session. New log_selector, +tls_sni.
1003 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
1004 NULL dereference. Report and patch from Alun Jones.
1006 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
1007 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
1008 Not seeing resolver debug output on NetBSD, but suspect this is a
1009 resolver implementation change.
1011 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
1012 Left warnings. Added "eximon gdb" invocation mode.
1014 PP/21 Defaulting "accept_8bitmime" to true, not false.
1016 PP/22 Added -bw for inetd wait mode support.
1018 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
1019 locate the relevant includes and libraries. Made this the default.
1021 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
1022 Bugzilla 1246, report and most of solution from Tomasz Kusy.
1024 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
1025 This may cause build issues on older platforms.
1027 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
1028 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
1029 gnutls_require_mac & gnutls_require_protocols (no longer supported).
1030 Added SNI support via GnuTLS too.
1031 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
1033 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
1035 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
1036 Applied second patch from Janne, implementing suggestion to default
1037 multiple-strings-in-record handling to match SPF spec.
1039 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
1041 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
1042 read-only, out of scope).
1043 Patch from Wolfgang Breyha, report from Stuart Northfield.
1045 PP/29 Fix three issues highlighted by clang analyser static analysis.
1046 Only crash-plausible issue would require the Cambridge-specific
1047 iplookup router and a misconfiguration.
1048 Report from Marcin Mirosław.
1050 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
1052 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
1053 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
1054 As part of this, removing so much warning spew let me fix some minor
1055 real issues in debug logging.
1057 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
1058 assignment on my part. Fixed.
1060 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
1061 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
1062 Janne Snabb (who went above and beyond: thank you).
1064 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
1065 string otherwise requires a connection and a bunch more work and it's
1066 relatively easy to get wrong. Should also expose TLS library linkage
1069 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
1070 64-bit ${eval} (JH/03).
1072 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
1073 GNU libc to support some of the 64-bit stuff, should not lead to
1074 conflicts. Defined before os.h is pulled in, so if a given platform
1075 needs to override this, it can.
1077 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
1078 protection layer was required, which is not implemented.
1079 Bugzilla 1254, patch from Wolfgang Breyha.
1081 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
1082 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
1083 tls_dhparam take prime identifiers. Also unbreak combination of
1084 OpenSSL+DH_params+TLSSNI.
1086 PP/39 Disable SSLv2 by default in OpenSSL support.
1092 PP/01 Solaris build fix for Oracle's LDAP libraries.
1093 Bugzilla 1109, patch from Stephen Usher.
1095 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
1097 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
1100 TF/02 Fix a couple more cases where we did not log the error message
1101 when unlink() failed. See also change 4.74-TF/03.
1103 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
1104 lock up or crash if it happened to be inside a call to libc when it
1105 got a SIGUSR1 from exiwhat.
1107 The SIGUSR1 handler appends the current process status to the process
1108 log which is later printed by exiwhat. It used to use the general
1109 purpose logging code to do this, but several functions it calls are
1110 not safe for signals.
1112 The new output code in the SIGUSR1 handler is specific to the process
1113 log, and simple enough that it's easy to inspect for signal safety.
1114 Removing some special cases also simplifies the general logging code.
1115 Removing the spurious timestamps from the process log simplifies
1118 TF/04 Improved ratelimit ACL condition.
1120 The /noupdate option has been deprecated in favour of /readonly which
1121 has clearer semantics. The /leaky, /strict, and /readonly update modes
1122 are mutually exclusive. The update mode is no longer included in the
1123 database key; it just determines when the database is updated. (This
1124 means that when you upgrde Exim will forget old rate measurements.)
1126 Exim now checks that the per_* options are used with an update mode that
1127 makes sense for the current ACL. For example, when Exim is processing a
1128 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
1129 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
1130 must specify per_mail/readonly. If you omit the update mode it defaults to
1131 /leaky where that makes sense (as before) or /readonly where required.
1133 The /noupdate option is now undocumented but still supported for
1134 backwards compatibility. It is equivalent to /readonly except that in
1135 ACLs where /readonly is required you may specify /leaky/noupdate or
1136 /strict/noupdate which are treated the same as /readonly.
1138 A useful new feature is the /count= option. This is a generalization
1139 of the per_byte option, so that you can measure the throughput of other
1140 aggregate values. For example, the per_byte option is now equivalent
1141 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
1143 The per_rcpt option has been generalized using the /count= mechanism
1144 (though it's more complicated than the per_byte equivalence). When it is
1145 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
1146 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
1147 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
1148 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
1149 Note that using per_rcpt with a non-readonly update mode in more than
1150 one ACL will cause the recipients to be double-counted. (The per_mail
1151 and per_byte options don't have this problem.)
1153 The handling of very low rates has changed slightly. If the computed rate
1154 is less than the event's count (usually one) then this event is the first
1155 after a long gap. In this case the rate is set to the same as this event's
1156 count, so that the first message of a spam run is counted properly.
1158 The major new feature is a mechanism for counting the rate of unique
1159 events. The new per_addr option counts the number of different
1160 recipients that someone has sent messages to in the last time period. It
1161 behaves like per_rcpt if all the recipient addresses are different, but
1162 duplicate recipient addresses do not increase the measured rate. Like
1163 the /count= option this is a general mechanism, so the per_addr option
1164 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
1165 example, measure the rate that a client uses different sender addresses
1166 with the options per_mail/unique=$sender_address. There are further
1167 details in the main documentation.
1169 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
1171 TF/06 Removed a few PCRE remnants.
1173 TF/07 Automatically extract Exim's version number from tags in the git
1174 repository when doing development or release builds.
1176 PP/02 Raise smtp_cmd_buffer_size to 16kB.
1177 Bugzilla 879. Patch from Paul Fisher.
1179 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
1180 Heavily based on revision 40f9a89a from Simon Arlott's tree.
1183 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
1185 PP/05 Variable $av_failed, true if the AV scanner deferred.
1186 Bugzilla 1078. Patch from John Horne.
1188 PP/06 Stop make process more reliably on build failure.
1189 Bugzilla 1087. Patch from Heiko Schlittermann.
1191 PP/07 Make maildir_use_size_file an _expandable_ boolean.
1192 Bugzilla 1089. Patch from Heiko Schlittermann.
1194 PP/08 Handle ${run} returning more data than OS pipe buffer size.
1195 Bugzilla 1131. Patch from Holger Weiß.
1197 PP/09 Handle IPv6 addresses with SPF.
1198 Bugzilla 860. Patch from Wolfgang Breyha.
1200 PP/10 GnuTLS: support TLS 1.2 & 1.1.
1202 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
1205 PP/11 match_* no longer expand right-hand-side by default.
1206 New compile-time build option, EXPAND_LISTMATCH_RHS.
1207 New expansion conditions, "inlist", "inlisti".
1209 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
1211 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
1213 PP/14 fix log_write() format string regression from TF/03.
1214 Bugzilla 1152. Patch from Dmitry Isaikin.
1220 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
1222 PP/02 Harmonised TLS library version reporting; only show if debugging.
1223 Layout now matches that introduced for other libraries in 4.74 PP/03.
1225 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
1227 PP/04 New "dns_use_edns0" global option.
1229 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
1232 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
1233 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
1235 TK/01 Updated PolarSSL code to 0.14.2.
1236 Bugzilla 1097. Patch from Andreas Metzler.
1238 PP/07 Catch divide-by-zero in ${eval:...}.
1239 Fixes bugzilla 1102.
1241 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
1244 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
1245 format-string attack -- SECURITY: remote arbitrary code execution.
1247 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
1248 time unintentionally subject to list matching rules, letting the header
1249 cause arbitrary Exim lookups (of items which can occur in lists, *not*
1250 arbitrary string expansion). This allowed for information disclosure.
1252 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
1253 INT_MIN/-1 -- value coerced to INT_MAX.
1259 NM/01 Workround for PCRE version dependency in version reporting
1262 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
1263 This fixes portability to compilers other than gcc, notably
1264 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
1266 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
1267 makefiles for portability to HP-UX and POSIX correctness.
1269 PP/01 Permit LOOKUP_foo enabling on the make command-line.
1270 Also via indented variable definition in the Makefile.
1271 (Debugging by Oliver Heesakkers).
1273 PP/02 Restore caching of spamd results with expanded spamd_address.
1274 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
1276 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
1277 Improves build reliability. Fix from: Frank Elsner
1279 NM/02 Fix wide character breakage in the rfc2047 coding
1280 Fixes bug 1064. Patch from Andrey N. Oktyabrski
1282 NM/03 Allow underscore in dnslist lookups
1283 Fixes bug 1026. Patch from Graeme Fowler
1285 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
1286 Code patches from Adam Ciarcinski of NetBSD.
1288 NM/04 Fixed exiqgrep to cope with mailq missing size issue
1291 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
1292 is logged, to avoid truncation. Patch from John Horne.
1294 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
1295 Patch from Jakob Hirsch.
1297 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
1298 SQL string expansion failure details.
1299 Patch from Andrey Oktyabrski.
1301 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
1302 Patch from Simon Arlott.
1304 PP/09 New lookups functionality failed to compile on old gcc which rejects
1305 extern declarations in function scope.
1306 Patch from Oliver Fleischmann
1308 PP/10 Use sig_atomic_t for flags set from signal handlers.
1309 Check getgroups() return and improve debugging.
1310 Fixed developed for diagnosis in bug 927 (which turned out to be
1313 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
1314 Patch from Mark Zealey.
1316 PP/12 Bugzilla 1056: Improved spamd server selection.
1317 Patch from Mark Zealey.
1319 PP/13 Bugzilla 1086: Deal with maildir quota file races.
1320 Based on patch from Heiko Schlittermann.
1322 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
1323 Patch from Uwe Doering, sign-off by Michael Haardt.
1325 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
1326 variable declaration deep within a block. Bug and patch from
1329 PP/15 lookups-Makefile IRIX compatibilty coercion.
1331 PP/16 Make DISABLE_DKIM build knob functional.
1333 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
1334 Patch by Simon Arlott
1336 TF/03 Fix valgrind.h portability to C89 compilers that do not support
1337 variable argument macros. Our copy now differs from upstream.
1343 TF/01 Failure to get a lock on a hints database can have serious
1344 consequences so log it to the panic log.
1346 TF/02 Log LMTP confirmation messages in the same way as SMTP,
1347 controlled using the smtp_confirmation log selector.
1349 TF/03 Include the error message when we fail to unlink a spool file.
1351 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
1352 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
1353 for maintaining out-of-tree patches for some time.
1355 PP/01 Bugzilla 139: Documentation and portability issues.
1356 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
1357 Handle per-OS dynamic-module compilation flags.
1359 PP/02 Let /dev/null have normal permissions.
1360 The 4.73 fixes were a little too stringent and complained about the
1361 permissions on /dev/null. Exempt it from some checks.
1362 Reported by Andreas M. Kirchwitz.
1364 PP/03 Report version information for many libraries, including
1365 Exim version information for dynamically loaded libraries. Created
1366 version.h, now support a version extension string for distributors
1367 who patch heavily. Dynamic module ABI change.
1369 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
1370 privilege escalation vulnerability whereby the Exim run-time user
1371 can cause root to append content of the attacker's choosing to
1374 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
1377 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
1378 If dropping privileges for untrusted macros, we disabled normal logging
1379 on the basis that it would fail; for the Exim run-time user, this is not
1380 the case, and it resulted in successful deliveries going unlogged.
1381 Fixed. Reported by Andreas Metzler.
1387 PP/01 Date: & Message-Id: revert to normally being appended to a message,
1388 only prepend for the Resent-* case. Fixes regression introduced in
1389 Exim 4.70 by NM/22 for Bugzilla 607.
1391 PP/02 Include check_rfc2047_length in configure.default because we're seeing
1392 increasing numbers of administrators be bitten by this.
1394 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
1396 PP/03 Bugzilla 994: added openssl_options main configuration option.
1398 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
1400 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
1402 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
1404 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
1405 without a peer certificate, leading to a segfault because of an
1406 assumption that peers always have certificates. Be a little more
1407 paranoid. Problem reported by Martin Tscholak.
1409 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
1410 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
1411 NB: ClamAV planning to remove STREAM in "middle of 2010".
1412 CL also introduces -bmalware, various -d+acl logging additions and
1413 more caution in buffer sizes.
1415 PP/09 Implemented reverse_ip expansion operator.
1417 PP/10 Bugzilla 937: provide a "debug" ACL control.
1419 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
1421 PP/12 Bugzilla 973: Implement --version.
1423 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
1425 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
1427 PP/15 Bugzilla 816: support multiple condition rules on Routers.
1429 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
1430 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
1431 ignore trailing whitespace.
1433 JJ/02 prevent non-panic DKIM error from being sent to paniclog
1435 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
1438 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
1439 Notification from Dr Andrew Aitchison.
1441 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
1442 ExtendedDetectionInfo response format.
1443 Notification from John Horne.
1445 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
1448 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
1449 XSL and documented dependency on system catalogs, with examples of how
1452 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
1455 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
1456 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
1457 configuration file which is writeable by the Exim user or group.
1459 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
1460 of configuration files to cover files specified with the -C option if
1461 they are going to be used with root privileges, not just the default
1464 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
1465 option (effectively making it always true).
1467 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
1468 files to be used while preserving root privileges.
1470 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
1471 that rogue child processes cannot use them.
1473 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
1474 run-time user, instead of root.
1476 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
1477 Exim run-time user without dropping privileges.
1479 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
1480 result string, instead of calling string_vformat() twice with the same
1483 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
1484 for other users. Others should always drop root privileges if they use
1485 -C on the command line, even for a whitelisted configure file.
1487 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
1489 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
1495 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
1496 $data_path, and $header_path variables; fixed documentation bugs and
1499 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
1500 exipick to access non-standard spools, including the "frozen" queue
1503 NM/01 Bugzilla 965: Support mysql stored procedures.
1504 Patch from Alain Williams
1506 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
1508 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
1509 Patch from Andreas Metzler
1511 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
1512 Patch from Kirill Miazine
1514 NM/05 Bugzilla 671: Added umask to procmail example.
1516 JJ/03 installed exipick 20100323.0, fixing doc bug
1518 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
1519 directory. Notification and patch from Dan Rosenberg.
1521 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
1523 TK/02 Improve log output when DKIM signing operation fails.
1525 MH/01 Treat the transport option dkim_domain as a colon separated
1526 list, not as a single string, and sign the message with each element,
1527 omitting multiple occurences of the same signer.
1529 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
1530 Bugzilla 985, 986. Patch by Simon Arlott
1532 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
1533 Patch by Simon Arlott
1535 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
1536 MBX locking. Notification from Dan Rosenberg.
1542 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
1544 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
1546 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
1548 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
1550 NM/04 Bugzilla 915: Flag broken perl installation during build.
1556 TK/01 Added patch by Johannes Berg that expands the main option
1557 "spamd_address" if it starts with a dollar sign.
1559 TK/02 Write list of recipients to X-Envelope-Sender header when building
1560 the mbox-format spool file for content scanning (suggested by Jakob
1563 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
1564 (http://www.dcc-servers.net/) support via dccifd. Activated by
1565 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
1567 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
1568 by Mark Daniel Reidel <mr@df.eu>.
1570 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
1571 When building exim an external PCRE library is now needed -
1572 PCRE is a system library on the majority of modern systems.
1573 See entry on PCRE_LIBS in EDITME file.
1575 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
1576 conversation. Added nologin parameter to request.
1577 Patch contributed by Kirill Miazine.
1579 TF/01 Do not log submission mode rewrites if they do not change the address.
1581 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
1583 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
1584 log files in place. Contributed by Roberto Lima.
1586 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
1588 TF/03 Bugzilla 615: When checking the local_parts router precondition
1589 after a local_part_suffix or local_part_prefix option, Exim now
1590 does not use the address's named list lookup cache, since this
1591 contains cached lookups for the whole local part.
1593 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
1594 Robert Millan. Documentation is in experimental-spec.txt.
1596 TF/04 Bugzilla 668: Fix parallel build (make -j).
1598 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
1600 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
1601 Patch provided by Jan Srzednicki.
1603 TF/05 Leading white space used to be stripped from $spam_report which
1604 wrecked the formatting. Now it is preserved.
1606 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
1607 that they are available at delivery time.
1609 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
1611 TF/08 TLS error reporting now respects the incoming_interface and
1612 incoming_port log selectors.
1614 TF/09 Produce a more useful error message if an SMTP transport's hosts
1615 setting expands to an empty string.
1617 NM/06 Bugzilla 744: EXPN did not work under TLS.
1618 Patch provided by Phil Pennock.
1620 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
1621 Patch provided by Richard Godbee.
1623 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
1624 acl_smtp_notquit, added index entry.
1626 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
1627 Patch provided by Eugene Bujak.
1629 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
1630 accept(). Patch provided by Maxim Dounin.
1632 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
1633 Patch provided by Phil Pennock.
1635 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
1637 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
1638 Patch provided by Brad "anomie" Jorsch.
1640 NM/14 Bugzilla 622: Added timeout setting to transport filter.
1641 Patch provided by Dean Brooks.
1643 TK/05 Add native DKIM support (does not depend on external libraries).
1645 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
1646 Patch provided by Graeme Fowler.
1648 NM/16 Bugzilla 851: Documentation example syntax fix.
1650 NM/17 Changed NOTICE file to remove references to embedded PCRE.
1652 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
1655 NM/19 Bugzilla 745: TLS version reporting.
1656 Patch provided by Phil Pennock.
1658 NM/20 Bugzilla 167: bool: condition support.
1659 Patch provided by Phil Pennock.
1661 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
1662 clients. Patch provided by Phil Pennock.
1664 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
1665 Patch provided by Brad "anomie" Jorsch.
1667 NM/23 Bugzilla 687: Fix misparses in eximstats.
1668 Patch provided by Heiko Schlittermann.
1670 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
1671 Patch provided by Heiko Schlittermann.
1673 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
1674 plus update to original patch.
1676 NM/26 Bugzilla 799: Documentation correction for ratelimit.
1678 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
1679 Patch provided by David Brownlee.
1681 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
1683 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
1685 NM/30 Bugzilla 888: TLS documentation bugfixes.
1687 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
1689 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
1690 Unlike the original bugzilla I have changed all shell scripts in src tree.
1692 NM/33 Bugzilla 898: Transport filter timeout fix.
1693 Patch by Todd Rinaldo.
1695 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
1696 Patch by Serge Demonchaux.
1698 NM/35 Bugzilla 39: Base64 decode bug fixes.
1699 Patch by Jakob Hirsch.
1701 NM/36 Bugzilla 909: Correct connect() call in dcc code.
1703 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
1705 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
1707 NM/39 Bugzilla 911: Fixed MakeLinks build script.
1713 TK/01 Add preliminary DKIM support. Currently requires a forked version of
1714 ALT-N's libdkim that I have put here:
1715 http://duncanthrax.net/exim-experimental/
1717 Note to Michael Haardt: I had to rename some vars in sieve.c. They
1718 were called 'true' and it seems that C99 defines that as a reserved
1719 keyword to be used with 'bool' variable types. That means you could
1720 not include C99-style headers which use bools without triggering
1721 build errors in sieve.c.
1723 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
1724 as mailq or other aliases. Changed the --help handling significantly
1725 to do whats expected. exim_usage() emits usage/help information.
1727 SC/01 Added the -bylocaldomain option to eximstats.
1729 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
1731 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
1733 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
1739 PH/01 Another patch from the Sieve maintainer.
1741 PH/02 When an IPv6 address is converted to a string for single-key lookup
1742 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
1743 dots are used instead of colons so that keys in lsearch files need not
1744 contain colons. This was done some time before quoting was made available
1745 in lsearch files. However, iplsearch files do require colons in IPv6 keys
1746 (notated using the quote facility) so as to distinguish them from IPv4
1747 keys. This meant that lookups for IP addresses in host lists did not work
1748 for iplsearch lookups.
1750 This has been fixed by arranging for IPv6 addresses to be expressed with
1751 colons if the lookup type is iplsearch. This is not incompatible, because
1752 previously such lookups could never work.
1754 The situation is now rather anomolous, since one *can* have colons in
1755 ordinary lsearch keys. However, making the change in all cases is
1756 incompatible and would probably break a number of configurations.
1758 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
1761 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
1762 conversion specification without a maximum field width, thereby enabling
1763 a rogue spamd server to cause a buffer overflow. While nobody in their
1764 right mind would setup Exim to query an untrusted spamd server, an
1765 attacker that gains access to a server running spamd could potentially
1766 exploit this vulnerability to run arbitrary code as the Exim user.
1768 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
1769 $primary_hostname instead of what libspf2 thinks the hosts name is.
1771 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
1772 a directory entry by the name of the lookup key. Previously, if a
1773 symlink pointed to a non-existing file or a file in a directory that
1774 Exim lacked permissions to read, a lookup for a key matching that
1775 symlink would fail. Now it is enough that a matching directory entry
1776 exists, symlink or not. (Bugzilla 503.)
1778 PH/03 The body_linecount and body_zerocount variables are now exported in the
1781 PH/04 Added the $dnslist_matched variable.
1783 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
1784 This means they are set thereafter only if the connection becomes
1787 PH/06 Added the client_condition to authenticators so that some can be skipped
1788 by clients under certain conditions.
1790 PH/07 The error message for a badly-placed control=no_multiline_responses left
1791 "_responses" off the end of the name.
1793 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
1795 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
1796 (without spaces) instead of just copying the configuration text.
1798 PH/10 Added the /noupdate option to the ratelimit ACL condition.
1800 PH/11 Added $max_received_linelength.
1802 PH/12 Added +ignore_defer and +include_defer to host lists.
1804 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
1805 way in which PCRE > 7.0 is built.
1807 PH/14 Implemented queue_only_load_latch.
1809 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
1810 MAIL command. The effect was to mangle the value on 64-bit systems.
1812 PH/16 Another patch from the Sieve maintainer.
1814 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
1816 PH/18 If a system quota error occurred while trying to create the file for
1817 a maildir delivery, the message "Mailbox is full" was not appended to the
1818 bounce if the delivery eventually timed out. Change 4.67/27 below applied
1819 only to a quota excession during the actual writing of the file.
1821 PH/19 It seems that peer DN values may contain newlines (and other non-printing
1822 characters?) which causes problems in log lines. The DN values are now
1823 passed through string_printing() before being added to log lines.
1825 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
1826 and InterBase are left for another time.)
1828 PH/21 Added message_body_newlines option.
1830 PH/22 Guard against possible overflow in moan_check_errorcopy().
1832 PH/23 POSIX allows open() to be a macro; guard against that.
1834 PH/24 If the recipient of an error message contained an @ in the local part
1835 (suitably quoted, of course), incorrect values were put in $domain and
1836 $local_part during the evaluation of errors_copy.
1842 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
1843 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
1846 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
1847 issue a MAIL command.
1849 PH/02 In an ACL statement such as
1851 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
1853 if a client was not listed at all, or was listed with a value other than
1854 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
1855 the condition was not true (as it should be), so access was not denied.
1856 The bug was that the ! inversion was incorrectly passed on to the second
1857 item. This has been fixed.
1859 PH/03 Added additional dnslists conditions == and =& which are different from
1860 = and & when the dns lookup returns more than one IP address.
1862 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
1863 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
1865 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
1866 FSYNC, which compiles an option called disable_fsync that allows for
1867 bypassing fsync(). The documentation is heavily laced with warnings.
1869 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
1871 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
1872 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
1873 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
1874 including adding "make clean"; (3) Added -fPIC when compiling the test
1875 dynamically loaded module, to get rid of a warning.
1877 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
1878 message fails, move_frozen_messages = true and ignore_bounce_errors_after
1879 = 0s. The bug is otherwise harmless.
1881 PH/07 There was a bug in the dovecot authenticator such that the value of
1882 $auth1 could be overwritten, and so not correctly preserved, after a
1883 successful authentication. This usually meant that the value preserved by
1884 the server_setid option was incorrect.
1886 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
1888 PH/09 Installed PCRE release 7.0.
1890 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
1891 run for batched SMTP input. It is now run at the start of every message
1892 in the batch. While fixing this I discovered that the process information
1893 (output by running exiwhat) was not always getting set for -bs and -bS
1894 input. This is fixed, and it now also says "batched" for BSMTP.
1896 PH/11 Added control=no_pipelining.
1898 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
1899 patch, slightly modified), and move the expansion of helo_data till after
1900 the connection is made in the smtp transport (so it can use these
1903 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
1905 PH/14 Added log_selector = +pid.
1907 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
1909 PH/16 Add ${if forany and ${if forall.
1911 PH/17 Added dsn_from option to vary the From: line in DSNs.
1913 PH/18 Flush SMTP output before performing a callout, unless control =
1914 no_callout_flush is set.
1916 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
1917 was true (the default) a successful delivery failed to delete the retry
1918 item, thus causing premature timeout of the address. The bug is now
1921 PH/20 Added hosts_avoid_pipelining to the smtp transport.
1923 PH/21 Long custom messages for fakedefer and fakereject are now split up
1924 into multiline reponses in the same way that messages for "deny" and
1925 other ACL rejections are.
1927 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
1928 with slight modification.
1930 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
1931 draft, changing the syntax and factoring some duplicate code".
1933 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
1934 for deliveries of the second and subsequent messages over the same SMTP
1937 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
1938 ${reduce, with only minor "tidies".
1940 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
1942 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
1943 expansion side effects.
1945 PH/27 When a message times out after an over-quota error from an Exim-imposed
1946 quota, the bounce message says "mailbox is full". This message was not
1947 being given when it was a system quota that was exceeded. It now should
1950 MH/03 Made $recipients available in local_scan(). local_scan() already has
1951 better access to the recipient list through recipients_list[], but
1952 $recipients can be useful in postmaster-provided expansion strings.
1954 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
1955 in the case of a MAIL command with additional options following the
1956 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
1957 were accidentally chopped off.
1959 PH/29 SMTP synchronization checks are implemented when a command is read -
1960 there is a check that no more input is waiting when there shouldn't be
1961 any. However, for some commands, a delay in an ACL can mean that it is
1962 some time before the response is written. In this time, more input might
1963 arrive, invalidly. So now there are extra checks after an ACL has run for
1964 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
1965 pipelining has not been advertised.
1967 PH/30 MH's patch to allow iscntrl() characters to be list separators.
1969 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
1970 returned in the SMTP response when smtp_return_error_details was false.
1971 This has been fixed.
1973 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
1974 instead of the C I/O that was originally supplied, because problems were
1975 reported on Solaris.
1977 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
1978 Exim which did not show up earlier: it was assuming that a call to
1979 SSL_CTX_set_info_callback() might give an error value. In fact, there is
1980 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
1981 was a macro that became an assignment, so it seemed to work. This has
1982 changed to a proper function call with a void return, hence the compile
1983 error. Exim's code has been fixed.
1985 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
1988 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
1990 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
1992 PH/37 If a message is not accepted after it has had an id assigned (e.g.
1993 because it turns out to be too big or there is a timeout) there is no
1994 "Completed" line in the log. When some messages of this type were
1995 selected by exigrep, they were listed as "not completed". Others were
1996 picked up by some special patterns. I have improved the selection
1997 criteria to be more general.
1999 PH/38 The host_find_failed option in the manualroute router can now be set
2000 to "ignore", to completely ignore a host whose IP address cannot be
2001 found. If all hosts are ignored, the behaviour is controlled by the new
2002 host_all_ignored option.
2004 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
2005 homing or because of multiple MX records with /mx) generated more than
2006 one IP address, and the following item turned out to be the local host,
2007 all the secondary addresses of the first item were incorrectly removed
2008 from the list, along with the local host and any following hosts (which
2009 is what is supposed to happen).
2011 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
2012 whoever called Exim into the -H file. In the case of the daemon it was
2013 behaving confusingly. When first started, it used values for whoever
2014 started the daemon, but after a SIGHUP it used the Exim user (because it
2015 calls itself on a restart). I have changed the code so that it now always
2018 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
2019 message are rejected with the same error (e.g. no authentication or bad
2020 sender address), and a DATA command is nevertheless sent (as can happen
2021 with PIPELINING or a stupid MUA), the error message that was given to the
2022 RCPT commands is included in the rejection of the DATA command. This is
2023 intended to be helpful for MUAs that show only the final error to their
2026 PH/42 Another patch from the Sieve maintainer.
2028 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
2029 Eximstats - Fixed some broken HTML links and added missing column headers
2031 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
2032 columns for Rejects, Temp Rejects, Ham, and Spam rows.
2034 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
2036 PH/43 Yet another patch from the Sieve maintainer.
2038 PH/44 I found a way to check for a TCP/IP connection going away before sending
2039 the response to the final '.' that terminates a message, but only in the
2040 case where the client has not sent further data following the '.'
2041 (unfortunately, this is allowed). However, in many cases there won't be
2042 any further data because there won't be any more messages to send. A call
2043 to select() can be used: if it shows that the input is "ready", there is
2044 either input waiting, or the socket has been closed. An attempt to read
2045 the next input character can distinguish the two cases. Previously, Exim
2046 would have sent an OK response which the client would never have see.
2047 This could lead to message repetition. This fix should cure that, at
2048 least in a lot of common cases.
2050 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
2051 advertised in response to EHLO.
2057 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
2058 fixed by 4.65/MH/01 (is this a record?) are fixed:
2060 (i) An empty string was always treated as zero by the numeric comparison
2061 operators. This behaviour has been restored.
2063 (ii) It is documented that the numeric comparison operators always treat
2064 their arguments as decimal numbers. This was broken in that numbers
2065 starting with 0 were being interpreted as octal.
2067 While fixing these problems I realized that there was another issue that
2068 hadn't been noticed. Values of message_size_limit (both the global option
2069 and the transport option) were treated as octal if they started with 0.
2070 The documentation was vague. These values are now always treated as
2071 decimal, and I will make that clear in the documentation.
2077 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
2078 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
2081 MH/01 Don't check that the operands of numeric comparison operators are
2082 integers when their expansion is in "skipping" mode (fixes bug
2083 introduced by 4.64-PH/07).
2085 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
2086 child addresses, Exim now panics and dies. Previously, because the count
2087 is held in a short int, deliveries were likely to be lost. As such a
2088 large number of recipients for a single message is ridiculous
2089 (performance will be very, very poor), I have chosen to impose a limit
2090 rather than extend the field.
2096 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
2097 leftover -K file (the existence of which was triggered by #402).
2098 While we were at it, introduced process PID as part of the -K
2099 filename. This should rule out race conditions when creating
2102 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
2103 processing considerably. Previous code took too long for large mails,
2104 triggering a timeout which in turn triggers #401.
2106 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
2107 in the DK code in transports.c. sendfile() is not really portable,
2108 hence the _LINUX specificness.
2110 TF/01 In the add_headers option to the mail command in an Exim filter,
2111 there was a bug that Exim would claim a syntax error in any
2112 header after the first one which had an odd number of characters
2115 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
2116 callout verification, Exim cached a "reject" for the entire domain. This
2117 is correct for most verifications, but it is not correct for a recipient
2118 verification with use_sender or use_postmaster set, because in that case
2119 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
2120 case of MAIL FROM:<> rejection from other early rejections (e.g.
2121 rejection of HELO). When verifying a recipient using a non-null MAIL
2122 address, the cache is ignored if it shows MAIL FROM:<> rejection.
2123 Whatever the result of the callout, the value of the domain cache is
2124 left unchanged (for any other kind of callout, getting as far as trying
2125 RCPT means that the domain itself is ok).
2127 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
2130 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
2131 manifest itself as EPIPE rather than ECONNECT. When tidying away a
2132 session, the daemon ignores ECONNECT errors and logs others; it now
2133 ignores EPIPE as well.
2135 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
2136 (quoted-printable decoding).
2138 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
2139 later the small subsequent patch to fix an introduced bug.
2141 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
2143 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
2145 PH/08 An error is now given if message_size_limit is specified negative.
2147 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
2148 to be given (somewhat) arbitrary names.
2150 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
2153 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
2154 miscellaneous code fixes
2156 PH/10 Added the log_reject_target ACL modifier to specify where to log
2159 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
2160 hostname. This is wrong, because it relates to the incoming message (and
2161 probably the interface on which it is arriving) and not to the outgoing
2162 callout (which could be using a different interface). This has been
2163 changed to use the value of the helo_data option from the smtp transport
2164 instead - this is what is used when a message is actually being sent. If
2165 there is no remote transport (possible with a router that sets up host
2166 addresses), $smtp_active_hostname is used.
2168 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
2169 tweaks were necessary in order to get it to work (see also 21 below):
2170 (a) The code assumed that strncpy() returns a negative number on buffer
2171 overflow, which isn't the case. Replaced with Exim's string_format()
2173 (b) There were several signed/unsigned issues. I just did the minimum
2174 hacking in of casts. There is scope for a larger refactoring.
2175 (c) The code used strcasecmp() which is not a standard C function.
2176 Replaced with Exim's strcmpic() function.
2177 (d) The code set only $1; it now sets $auth1 as well.
2178 (e) A simple test gave the error "authentication client didn't specify
2179 service in request". It would seem that Dovecot has changed its
2180 interface. Fortunately there's a specification; I followed it and
2181 changed what the client sends and it appears to be working now.
2183 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
2186 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
2187 address A is aliased to B and C, where B exists and C does not. Without
2188 -v the output is "A verified" because verification stops after a
2189 successful redirection if more than one address is generated. However,
2190 with -v the child addresses are also verified. Exim was outputting "A
2191 failed to verify" and then showing the successful verification for C,
2192 with its parentage. It now outputs "B failed to verify", showing B's
2193 parentage before showing the successful verification of C.
2195 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
2196 look up a TXT record in a specific list after matching in a combined
2199 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
2200 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
2201 they consult the DNS. I had assumed they would set it the way they
2202 wanted; and indeed my experiments on Linux seem to show that in some
2203 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
2204 To be on the safe side, however, I have now made the interface to
2205 host_find_byname() similar to host_find_bydns(), with an argument
2206 containing the DNS resolver options. The host_find_byname() function now
2207 sets these options at its start, just as host_find_bydns() does. The smtp
2208 transport options dns_qualify_single and dns_search_parents are passed to
2209 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
2210 of host_find_byname() use the default settings of RES_DEFNAMES
2211 (qualify_single) but not RES_DNSRCH (search_parents).
2213 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
2214 spool_read_header() do less string testing, by means of a preliminary
2215 switch on the second character of optional "-foo" lines. (This is
2216 overdue, caused by the large number of possibilities that now exist.
2217 Originally there were few.) While I was there, I also converted the
2218 str(n)cmp tests so they don't re-test the leading "-" and the first
2219 character, in the hope this might squeeze out yet more improvement.
2221 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
2222 flag allowing group syntax was set by the header_syntax check but not
2223 turned off, possible causing trouble later; (2) The flag was not being
2224 set at all for the header_verify test, causing "group"-style headers to
2225 be rejected. I have now set it in this case, and also caused header_
2226 verify to ignore an empty address taken from a group. While doing this, I
2227 came across some other cases where the code for allowing group syntax
2228 while scanning a header line wasn't quite right (mostly, not resetting
2229 the flag correctly in the right place). These bugs could have caused
2230 trouble for malformed header lines. I hope it is now all correct.
2232 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
2233 with the "reply" argument non-NULL. The code, however (which originally
2234 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
2235 but it didn't always do it. This confused somebody who was copying the
2236 code for some other use. I have removed all the tests.
2238 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
2239 feature that was used to support insecure browsers during the U.S. crypto
2240 embargo. It requires special client support, and Exim is probably the
2241 only MTA that supported it -- and would never use it because real RSA is
2242 always available. This code has been removed, because it had the bad
2243 effect of slowing Exim down by computing (never used) parameters for the
2244 RSA_EXPORT functionality.
2246 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
2247 authenticator to fail if there's a tab character in the incoming data
2248 (there should never be unless someone is messing about, as it's supposed
2249 to be base64-encoded). Also added, on Timo's advice, the "secured" option
2250 if the connection is using TLS or if the remote IP is the same as the
2251 local IP, and the "valid-client-cert option" if a client certificate has
2254 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
2255 authenticators. This can be used for authorization after authentication
2256 succeeds. (In the case of plaintext, it servers for both authentication
2259 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
2260 if any retry times were supplied.
2262 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
2263 connection, where there is no client IP address to check. In this
2264 situation, the verify now always succeeds.
2266 PH/25 Applied John Jetmore's -Mset patch.
2268 PH/26 Added -bem to be like -Mset, but loading a message from a file.
2270 PH/27 In a string expansion for a processed (not raw) header when multiple
2271 headers of the same name were present, leading whitespace was being
2272 removed from all of them, but trailing whitespace was being removed only
2273 from the last one. Now trailing whitespace is removed from each header
2274 before concatenation. Completely empty headers in a concatenation (as
2275 before) are ignored.
2277 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
2278 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
2280 PH/29 [Removed. This was a change that I later backed out, and forgot to
2281 correct the ChangeLog entry (that I had efficiently created) before
2282 committing the later change.]
2284 PH/30 Exim was sometimes attempting to deliver messages that had suffered
2285 address errors (4xx response to RCPT) over the same connection as other
2286 messages routed to the same hosts. Such deliveries are always "forced",
2287 so retry times are not inspected. This resulted in far too many retries
2288 for the affected addresses. The effect occurred only when there were more
2289 hosts than the hosts_max_try setting in the smtp transport when it had
2290 the 4xx errors. Those hosts that it had tried were not added to the list
2291 of hosts for which the message was waiting, so if all were tried, there
2292 was no problem. Two fixes have been applied:
2294 (i) If there are any address or message errors in an SMTP delivery, none
2295 of the hosts (tried or untried) are now added to the list of hosts
2296 for which the message is waiting, so the message should not be a
2297 candidate for sending over the same connection that was used for a
2298 successful delivery of some other message. This seems entirely
2299 reasonable: after all the message is NOT "waiting for some host".
2300 This is so "obvious" that I'm not sure why it wasn't done
2301 previously. Hope I haven't missed anything, but it can't do any
2302 harm, as the worst effect is to miss an optimization.
2304 (ii) If, despite (i), such a delivery is accidentally attempted, the
2305 routing retry time is respected, so at least it doesn't keep
2306 hammering the server.
2308 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
2309 in ${readsocket because some servers need this prod.
2311 PH/32 Added some extra debug output when updating a wait-xxx database.
2313 PH/33 The hint "could be header name not terminated by colon", which has been
2314 given for certain expansion errors for a long time, was not being given
2315 for the ${if def:h_colon_omitted{... case.
2317 PH/34 The spec says: "With one important exception, whenever a domain list is
2318 being scanned, $domain contains the subject domain." There was at least
2319 one case where this was not true.
2321 PH/35 The error "getsockname() failed: connection reset by peer" was being
2322 written to the panic log as well as the main log, but it isn't really
2323 panic-worthy as it just means the connection died rather early on. I have
2324 removed the panic log writing for the ECONNRESET error when getsockname()
2327 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
2328 runs only) independently of the message's sender address. This meant
2329 that, if the 4xx error was in fact related to the sender, a different
2330 message to the same recipient with a different sender could confuse
2331 things. In particualar, this can happen when sending to a greylisting
2332 server, but other circumstances could also provoke similar problems.
2333 I have changed the default so that the retry time for these errors is now
2334 based a combination of the sender and recipient addresses. This change
2335 can be overridden by setting address_retry_include_sender=false in the
2338 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
2339 remote server are returned as part of bounce messages. This was not
2340 happening for LMTP over a pipe (the lmtp transport), but now it is the
2341 same for both kinds of LMTP.
2343 PH/38 Despite being documented as not happening, Exim was rewriting addresses
2344 in header lines that were in fact CNAMEs. This is no longer the case.
2346 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
2347 and queue runs started by the daemon processed all messages. This has
2348 been fixed so that -R and -S can now usefully be given with -q<time>.
2350 PH/40 Import PCRE release 6.7 (fixes some bugs).
2352 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
2354 PH/42 Give an error if -q is specified more than once.
2356 PH/43 Renamed the variables $interface_address and $interface_port as
2357 $received_ip_address and $received_port, to make it clear that these
2358 values apply to message reception, and not to the outgoing interface when
2359 a message is delivered. (The old names remain recognized, of course.)
2361 PH/44 There was no timeout on the connect() call when using a Unix domain
2362 socket in the ${readsocket expansion. There now is.
2364 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
2365 be meaningful with "accept".
2367 SC/01 Eximstats V1.43
2368 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
2370 SC/02 Eximstats V1.44
2371 Use a glob alias rather than an array ref in the generated
2372 parser. This improves both readability and performance.
2374 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
2375 Collect SpamAssassin and rejection statistics.
2376 Don't display local sender or destination tables unless
2377 there is data to show.
2378 Added average volumes into the top table text output.
2380 SC/04 Eximstats V1.46
2381 Collect data on the number of addresses (recipients)
2382 as well as the number of messages.
2384 SC/05 Eximstats V1.47
2385 Added 'Message too big' to the list of mail rejection
2386 reasons (thanks to Marco Gaiarin).
2388 SC/06 Eximstats V1.48
2389 Mainlog lines which have GMT offsets and are too short to
2390 have a flag are now skipped.
2392 SC/07 Eximstats V1.49 (Alain Williams)
2393 Added the -emptyok flag.
2395 SC/08 Eximstats V1.50
2396 Fixes for obtaining the IP address from reject messages.
2398 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
2399 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
2400 whitesspace changes from 4.64-PH/27
2402 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
2405 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
2406 are found, allow negative numbers in numeric criteria)
2408 JJ/06 exipick.20061117.2, added new $message_body_missing variable
2410 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
2411 to match changes made in 4.64-PH/43
2413 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
2415 PH/47 Put in an explicit test for a DNS lookup of an address record where the
2416 "domain" is actually an IP address, and force a failure. This locks out
2417 those revolvers/nameservers that support "A-for-A" lookups, in
2418 contravention of the specifications.
2420 PH/48 When a host name was looked up from an IP address, and the subsequent
2421 forward lookup of the name timed out, the host name was left in
2422 $sender_host_name, contrary to the specification.
2424 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
2425 restricted to single-key lookups, Exim was not diagnosing an error if
2426 * or *@ was used with a query-style lookup.
2428 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
2430 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
2431 long ago, but noone interested enough thought of it. Let's just say that
2432 the "1.1" means that there are some new functions that weren't there at
2433 some point in the past.
2435 PH/51 Error processing for expansion failure of helo_data from an smtp
2436 transport during callout processing was broken.
2438 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
2439 tested/used via the -bh/-bhc/-bs options.
2441 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
2442 bug, fixed in subsequent PCRE releases).
2444 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
2445 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
2447 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
2453 SC/01 Use a glob alias rather than an array ref in eximstats generated
2454 parser. This improves both readability and performance.
2456 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
2457 Don't display local sender or destination tables in eximstats unless
2458 there is data to show.
2459 Added average volumes into the eximstats top table text output.
2461 SC/03 Collect data on the number of addresses (recipients) as well
2462 as the number of messages in eximstats.
2464 TF/01 Correct an error in the documentation for the redirect router. Exim
2465 does (usually) call initgroups() when daemonizing.
2467 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
2468 with consistent privilege compared to when running as a daemon.
2470 TF/03 Note in the spec that $authenticated_id is not set for local
2471 submissions from trusted users.
2473 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
2474 Thanks to Dean Brooks <dean@iglou.com> for the patch.
2476 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
2477 by adding some example configuration directives to the default
2478 configuration file. A little bit of work is required to uncomment the
2479 directives and define how usernames and passwords are checked, but
2480 there is now a framework to start from.
2482 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
2483 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
2484 without this. I don't know how relevant this is to other LDAP libraries.
2486 PH/02 Add the verb name to the "unknown ACL verb" error.
2488 PH/03 Magnus Holmgren's patch for filter_prepend_home.
2490 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
2492 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
2493 directory not expanded when it should be if an expanded home directory
2494 was set for the address (which is overridden by the transport).
2496 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
2499 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
2500 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
2501 because it is too late at that time, and has no effect.
2503 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
2504 security issue with \' (bugzilla #107). I could not use the
2505 PQescapeStringConn() function, because it needs a PGconn value as one of
2508 PH/08 When testing addresses using -bt, indicate those final addresses that
2509 are duplicates that would not cause an additional delivery. At least one
2510 person was confused, thinking that -bt output corresponded to deliveries.
2511 (Suppressing duplicates isn't a good idea as you lose the information
2512 about possibly different redirections that led to the duplicates.)
2514 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
2515 systems where poll() doesn't work, in particular OS X.
2517 PH/10 Added more information to debugging output for retry time not reached.
2519 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
2520 operations in malware.c.
2522 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
2525 PH/13 If write_rejectlog was set false when logging was sent to syslog with
2526 syslog_duplication set false, log lines that would normally be written
2527 both the the main log and to the reject log were not written to syslog at
2530 PH/14 In the default configuration, change the use of "message" in ACL warn
2531 statements to "add_header".
2533 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
2534 not followed by a command (e.g. "seen endif").
2536 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
2537 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
2540 PH/17 Added extra conditions to the default value of delay_warning_condition
2544 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
2545 { match{$h_precedence:}{(?i)bulk|list|junk} } \
2546 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
2549 The Auto-Submitted: and various List- headers are standardised, whereas I
2550 don't think Precedence: ever was.
2552 PH/18 Refactored debugging code in route_finduser() to show more information,
2553 in particular, the error code if getpwnam() issues one.
2555 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
2556 This is apparently needed in addition to the PH/07 change above to avoid
2557 any possible encoding problems.
2559 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
2560 but not after initializing Perl.
2562 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
2563 output them only if debugging. By default they are written stderr,
2564 apparently, which is not desirable.
2566 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
2569 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
2572 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
2574 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
2575 authenticated or an ident call has been made. Suppress the default
2576 values for $authenticated_id and $authenticated_sender (but permit -oMai
2577 and -oMas) when testing with -bh.
2579 PH/24 Re-jigged the order of the tests in the default configuration so that the
2580 tests for valid domains and recipients precede the DNS black list and CSA
2581 tests, on the grounds that those ones are more expensive.
2583 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
2584 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
2585 This bug exists in every version of Exim that I still have, right back to
2588 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
2589 However, an attempt to turn on case-sensitivity in a regex key by
2590 including (?-i) didn't work because the subject string was already
2591 lowercased, and the effects were non-intuitive. It turns out that a
2592 one-line patch can be used to allow (?-i) to work as expected.
2598 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
2599 other effects) broke the use of negated acl sub-conditions.
2601 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
2604 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
2605 "Deny" causes Exim to reject the incoming connection with a 554 error.
2606 Unfortunately, if there is a major crisis, such as a disk failure,
2607 tcp-wrappers gives "deny", whereas what one would like would be some
2608 kind of temporary error. A kludge has been added to help with this.
2609 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
2610 554 error is used if errno is still zero or contains ENOENT (which occurs
2611 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
2614 PH/03 Add -lutil to the default FreeBSD LIBS setting.
2616 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
2617 errors. Otherwise a message that provokes a temporary error (when other
2618 messages do not) can cause a whole host to time out.
2620 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
2621 addresses were routed directly to files or pipes from a redirect router.
2622 File deliveries just didn't batch; pipe deliveries might have suffered
2625 PH/06 A failure to get a lock for a hints database would erroneously always say
2626 "Failed to get write lock", even when it was really a read lock.
2628 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
2629 of 0600. This has been changed to use the value of the lockfile_mode
2630 option (which defaults to 0600).
2632 PH/08 Applied small patch from the Sieve maintainer.
2634 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
2635 folder from quota calculations, a direct delivery into this folder messed
2636 up the contents of the maildirsize file. This was because the regex was
2637 used only to exclude .Trash (or whatever) when the size of the mailbox
2638 was calculated. There was no check that a delivery was happening into an
2639 excluded directory. This bug has been fixed by ignoring all quota
2640 processing for deliveries into excluded directories.
2642 PH/10 Added the maildirfolder_create_regex option to appendfile.
2648 PH/01 The code for finding all the local interface addresses on a FreeBSD
2649 system running IPv6 was broken. This may well have applied to all BSD
2650 systems, as well as to others that have similar system calls. The broken
2651 code found IPv4 interfaces correctly, but gave incorrect values for the
2652 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
2653 that it would not match correctly against @[] and not recognize the IPv6
2656 PH/02 The ipliteral router was not recognizing addresses of the form user@
2657 [ipv6:....] because it didn't know about the "ipv6:" prefix.
2659 PH/03 Added disable_ipv6.
2661 PH/04 Changed $reply_address to use the raw form of the headers instead of the
2662 decoded form, because it is most often used to construct To: headers
2663 lines in autoreplies, and the decoded form may well be syntactically
2664 invalid. However, $reply_address has leading white space removed, and all
2665 newlines turned into spaces so that the autoreply transport does not
2668 PH/05 If group was specified without a user on a router, and no group or user
2669 was specified on a transport, the group from the router was ignored.
2671 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
2672 for visible compile-time settings that can be used to change these
2673 numbers, for those that want even more. Backwards compatibility with old
2674 spool files has been maintained. However, going back to a previous Exim
2675 release will lost any variables that are in spool files.
2677 PH/07 Two small changes when running in the test harness: increase delay when
2678 passing a TCP/IP connection to a new process, in case the original
2679 process has to generate a bounce, and remove special handling of
2680 127.0.0.2 (sic), which is no longer necessary.
2682 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
2683 be the same on different OS.
2685 PH/09 Moved a debug statement in filter processing to avoid a race problem when
2688 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
2689 whether --show-vars was specified or not
2691 JJ/02 exipick: Added support for new ACL variable spool format introduced
2694 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
2695 syntactically invalid From: or Reply-to: line, and a filter used this to
2696 generate an autoreply, and therefore failed to obtain an address for the
2697 autoreply, Exim could try to deliver to a non-existent relative file
2698 name, causing unrelated and misleading errors. What now happens is that
2699 it logs this as a hard delivery error, but does not attempt to create a
2702 PH/11 The exinext utility has a -C option for testing purposes, but although
2703 the given file was scanned by exinext itself; it wasn't being passed on
2704 when Exim was called.
2706 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
2707 an end-of-file indication when reading a command response.
2709 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
2710 compiled. In many other places in Exim, IPv6 addresses are always
2711 recognized, so I have changed this. It also means that IPv4 domain
2712 literals of the form [IPV4:n.n.n.n] are now always recognized.
2714 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
2715 used if the router is not running as root, for example, when verifying at
2716 ACL time, or when using -bh. The debugging output from this situation was
2717 non-existent - all you got was a failure to exec. I have made two
2720 (a) Failures to set uid/gid, the current directory, or a process leader
2721 in a subprocess such as that created by queryprogram now generate
2722 suitable debugging ouput when -d is set.
2724 (b) The queryprogram router detects when it is not running as root,
2725 outputs suitable debugging information if -d is set, and then runs
2726 the subprocess without attempting to change uid/gid.
2728 PH/15 Minor change to Makefile for building test_host (undocumented testing
2731 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
2732 additional section of a DNS packet that returns MX or SRV records.
2733 Instead, it always explicitly searches for A/AAAA records. This avoids
2734 major problems that occur when a DNS server includes only records of one
2735 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
2736 fixed another bug: if SRV records were looked up and the corresponding
2737 address records were *not* found in the additional section, the port
2738 values from the SRV records were lost.
2740 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
2741 using the correct key (the original address) when searching the retry
2742 rules in order to find which one to use for generating the retry hint.
2744 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
2745 adding the default one. Similarly, if it contains a Reply-To: header, the
2746 errors_reply_to option, if set, is not used.
2748 PH/19 When calculating a retry time, Exim used to measure the "time since
2749 failure" by looking at the "first failed" field in the retry record. Now
2750 it does not use this if it is later than than the arrival time of the
2751 message. Instead it uses the arrival time. This makes for better
2752 behaviour in cases where some deliveries succeed, thus re-setting the
2753 "first failed" field. An example is a quota failure for a huge message
2754 when small messages continue to be delivered. Without this change, the
2755 "time since failure" will always be short, possible causing more frequent
2756 delivery attempts for the huge message than are intended.
2757 [Note: This change was subsequently modified - see PH/04 for 4.62.]
2759 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
2760 $1, $2, $3) because the numerical variables can be reset during some
2761 expansion items (e.g. "match"), thereby losing the authentication data.
2763 PH/21 Make -bV show the size of off_t variables so that the test suite can
2764 decide whether to run tests for quotas > 2G.
2766 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
2767 mailbox_size, and mailbox_filecount in the appendfile transport. If a
2768 filecount value is greater than 2G or if a quota value is greater than 2G
2769 on a system where the size of off_t is not greater than 4, a panic error
2772 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
2773 never match. The debug and -bh output now contains an explicit error
2774 message indicating a malformed IPv4 address or mask.
2776 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
2777 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
2778 PH/23 above applies.
2780 PH/25 Do not write to syslog when running in the test harness. The only
2781 occasion when this arises is a failure to open the main or panic logs
2782 (for which there is an explicit test).
2784 PH/26 Added the /no_tell option to "control=freeze".
2786 PH/27 If a host name lookup failed very early in a connection, for example, if
2787 the IP address matched host_lookup and the reverse lookup yielded a name
2788 that did not have a forward lookup, an error message of the form "no IP
2789 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
2790 could be logged. Now it outputs the IP address instead of "NULL".
2792 PH/28 An enabling patch from MH: add new function child_open_exim2() which
2793 allows the sender and the authenticated sender to be set when
2794 submitting a message from within Exim. Since child_open_exim() is
2795 documented for local_scan(), the new function should be too.
2797 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
2798 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
2799 results in an empty string is now treated as unset.
2801 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
2803 PH/31 Added sender_verify_fail logging option.
2805 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
2806 needed by RFC 822 but not by RFC 2822 was commented out. I have now
2807 tidied the source and removed it altogether.
2809 PH/33 When a queue run was abandoned because the load average was too high, a
2810 log line was always written; now it is written only if the queue_run log
2811 selector is set. In addition, the log line for abandonment now contains
2812 information about the queue run such as the pid. This is always present
2813 in "start" and "stop" lines but was omitted from the "abandon" line.
2815 PH/34 Omit spaces between a header name and the colon in the error message that
2816 is given when verify = headers_syntax fails (if there are lots of them,
2817 the message gets confusing).
2819 PH/35 Change the default for dns_check_names_pattern to allow slashes within
2820 names, as there are now some PTR records that contain slashes. This check
2821 is only to protect against broken name servers that fall over on strange
2822 characters, so the fact that it applies to all lookups doesn't matter.
2824 PH/36 Now that the new test suite is complete, we can remove some of the
2825 special code in Exim that was needed for the old test suite. For example,
2826 sorting DNS records because real resolvers return them in an arbitrary
2827 order. The new test suite's fake resolver always returns records in the
2830 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
2831 bounces) except when queue_only is set, to avoid logging races between
2832 the different processes.
2834 PH/38 Panic-die if .include specifies a non-absolute path.
2836 PH/39 A tweak to the "H" retry rule from its user.
2838 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
2839 a label. They prevented compilation on older perls.
2841 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
2842 a warning to be raised on newish perls.
2844 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
2845 on queue. Changes to match documented behaviour of showing count of
2846 messages matching specified criteria.
2848 PH/40 Changed the default ident timeout from 30s to 5s.
2850 PH/41 Added support for the use of login_cap features, on those BSD systems
2851 that have them, for controlling the resources used by pipe deliveries.
2853 PH/42 The content-scanning code uses fopen() to create files in which to put
2854 message data. Previously it was not paying any attention to the mode of
2855 the files. Exim runs with umask(0) because the rest of the code creates
2856 files with open(), and sets the required mode explicitly. Thus, these
2857 files were ending up world-writeable. This was not a big issue, because,
2858 being within the spool directory, they were not world-accessible. I have
2859 created a function called modefopen, which takes an additional mode
2860 argument. It sets umask(777), creates the file, chmods it to the required
2861 mode, then resets the umask. All the relevant calls to fopen() in the
2862 content scanning code have been changed to use this function.
2864 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
2865 to 24 hours. This avoids potential overflow problems when processing G
2866 and H retry rules. I suspect nobody ever tinkers with this value.
2868 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
2870 PH/45 When the plaintext authenticator is running as a client, the server's
2871 challenges are checked to ensure they are valid base64 strings. By
2872 default, the authentication attempt is cancelled if an invalid string is
2873 received. Setting client_ignore_invalid_base64 true ignores these errors.
2874 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
2875 they are received. Thus, the responses can be made to depend on the
2876 challenges. If an invalid string is ignored, an empty string is placed in
2879 PH/46 Messages that are created by the autoreply transport now contains a
2880 References: header, in accordance with RFCs 2822 and 3834.
2882 PH/47 Added authenticated_sender_force to the smtp transport.
2884 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
2886 PH/49 Installed latest patch from the Sieve maintainer.
2888 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
2889 was also set, the appendfile transport was unnecessarily scanning a
2890 directory of message files (e.g. for maildir delivery) to find the count
2891 of files (along with the size), even though it did not need this
2892 information. It now does the scan only if it needs to find either the
2893 size of the count of files.
2895 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
2897 PH/52 Two bugs concerned with error handling when the smtp transport is
2900 (i) Exim was not creating retry information for temporary errors given
2901 for individual recipients after the DATA command when the smtp transport
2902 was used in LMTP mode. This meant that they could be retried too
2903 frequently, and not timed out correctly.
2905 (ii) Exim was setting the flag that allows error details to be returned
2906 for LMTP errors on RCPT commands, but not for LMTP errors for individual
2907 recipients that were returned after the DATA command.
2909 PH/53 This is related to PH/52, but is more general: for any failing address,
2910 when detailed error information was permitted to be returned to the
2911 sender, but the error was temporary, then after the final timeout, only
2912 "retry timeout exceeded" was returned. Now it returns the full error as
2913 well as "retry timeout exceeded".
2915 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
2916 do this, and (what is worse) MTAs that accept it.
2918 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
2919 will now be deprecated.
2921 PH/56 New os.c-cygwin from the Cygwin maintainer.
2923 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
2924 formats (previously only available in exim formats via -bpr, -bpru,
2925 and -bpra. Now also available in native and exiqgrep formats)
2927 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
2928 with very large, slow to parse queues
2930 JJ/08 exipick: added ! as generic prefix to negate any criteria format
2932 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
2934 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
2935 responses to authentication challenges, though it was showing the
2936 challenges; (ii) I've removed the CR characters from the debug output for
2939 PH/58 Allow for the insertion of a newline as well as a space when a string
2940 is turned into more than one encoded-word during RFC 2047 encoding. The
2941 Sieve code now uses this.
2943 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
2944 data_4xx, lost_connection, tls_required.
2946 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
2947 message was being sent as an SMTP response.
2949 PH/61 Add -l and -k options to exicyclog.
2951 PH/62 When verifying, if an address was redirected to one new address, so that
2952 verification continued, and the new address failed or deferred after
2953 having set something in $address_data, the value of $address_data was not
2954 passed back to the ACL. This was different to the case when no
2955 redirection occurred. The value is now passed back in both cases.
2957 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
2958 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
2959 use login_cap.h, so on its own it isn't the distinguishing feature. The
2960 new name refers directly to the setclassresources() function.
2962 PH/65 Added configuration files for NetBSD3.
2964 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
2966 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
2967 is preferred over IPv4.
2969 PH/68 The bounce_return_message and bounce_return_body options were not being
2970 honoured for bounces generated during the reception of non-SMTP messages.
2971 In particular, this applied to messages rejected by the ACL. This bug has
2972 been fixed. However, if bounce_return_message is true and bounce_return_
2973 body is false, the headers that are returned for a non-SMTP message
2974 include only those that have been read before the error was detected.
2975 (In the case of an ACL rejection, they have all been read.)
2977 PH/69 The HTML version of the specification is now built in a directory called
2978 spec_html instead of spec.html, because the latter looks like a path with
2979 a MIME-type, and this confuses some software.
2981 PH/70 Catch two compiler warnings in sieve.c.
2983 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
2984 function verify_get_ident() calls ip_connect() to connect a socket, but
2985 if the "connect()" function timed out, ip_connect() used to close the
2986 socket. However, verify_get_ident() also closes the socket later, and in
2987 between Exim writes to the log, which may get opened at this point. When
2988 the socket was closed in ip_connect(), the log could get the same file
2989 descriptor number as the socket. This naturally causes chaos. The fix is
2990 not to close the socket in ip_connect(); the socket should be closed by
2991 the function that creates it. There was only one place in the code where
2992 this was missing, in the iplookup router, which I don't think anybody now
2993 uses, but I've fixed it anyway.
2995 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
2996 well as to direct DNS lookups. Otherwise the handling of names in host
2997 lists is inconsistent and therefore confusing.
3003 PH/01 Two changes to the default runtime configuration:
3005 (1) Move the checks for relay_from_hosts and authenticated clients from
3006 after to before the (commented out) DNS black list checks.
3008 (2) Add control=submission to the relay_from_hosts and authenticated
3009 clients checks, on the grounds that messages accepted by these
3010 statements are most likely to be submissions.
3012 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
3014 (1) Generate an error if the third argument for the ${prvs expansion is
3017 (2) Treat a missing third argument of ${prvscheck as if it were an empty
3020 (3) Reset the variables that are obtained from the first argument of
3021 ${prvscheck and used in the second argument before leaving the code,
3022 because their memory is reclaimed, so using them afterwards may do
3025 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
3026 one (it's much easier than Tom thought :-).
3028 (5) Because of (4), we can now allow for the use of $prvscheck_result
3029 inside the third argument.
3031 PH/03 For some reason, the default setting of PATH when running a command from
3032 a pipe transport was just "/usr/bin". I have changed it to
3035 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
3036 anything to be listed in the output from -bV.
3038 PH/05 When a filter generated an autoreply, the entire To: header line was
3039 quoted in the delivery log line, like this:
3041 => >A.N.Other <ano@some.domain> <original@ddress> ...
3043 This has been changed so that it extracts the operative address. There
3044 may be more than one such address. If so, they are comma-separated, like
3047 => >ano@some.domain,ona@other.domain <original@ddress> ...
3049 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
3050 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
3051 not being looked up in the rDNS to get a host name, Exim was showing the
3052 IP address twice in Received: lines, even though the IP addresses were
3053 identical. For example:
3055 Received: from [1.2.3.4] (helo=[1.2.3.4])
3057 However, if the real host name was known, it was omitting the HELO data
3058 if it matched the actual IP address. This has been tidied up so that it
3059 doesn't show the same IP address twice.
3061 PH/07 When both +timestamp and +memory debugging was on, the value given by
3062 $tod_xxx expansions could be wrong, because the tod_stamp() function was
3063 called by the debug printing, thereby overwriting the timestamp buffer.
3064 Debugging no longer uses the tod_stamp() function when +timestamp is set.
3066 PH/08 When the original message was included in an autoreply transport, it
3067 always said "this is a copy of the message, including all the headers",
3068 even if body_only or headers_only was set. It now gives an appropriate
3071 PH/09 Applied a patch from the Sieve maintainer which:
3073 o fixes some comments
3074 o adds the (disabled) notify extension core
3075 o adds some debug output for the result of if/elsif tests
3076 o points to the current vacation draft in the documentation
3077 and documents the missing references header update
3081 o fixes a bug in processing the envelope test (when testing
3082 multiple envelope elements, the last element determinted the
3085 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
3086 Electronic Mail") by including:
3088 Auto-submitted: auto-generated
3090 in the messages that it generates (bounce messages and others, such as
3091 warnings). In the case of bounce messages for non-SMTP mesages, there was
3092 also a typo: it was using "Auto_submitted" (underscore instead of
3093 hyphen). Since every message generated by Exim is necessarily in response
3094 to another message, thes have all been changed to:
3096 Auto-Submitted: auto-replied
3098 in accordance with these statements in the RFC:
3100 The auto-replied keyword:
3102 - SHOULD be used on messages sent in direct response to another
3103 message by an automatic process,
3105 - MUST NOT be used on manually-generated messages,
3107 - MAY be used on Delivery Status Notifications (DSNs) and Message
3108 Disposition Notifications (MDNs),
3110 - MUST NOT be used on messages generated by automatic or periodic
3111 processes, except for messages which are automatic responses to
3114 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
3115 to the default Received: header definition.
3117 PH/12 Added log selector acl_warn_skipped (default on).
3119 PH/13 After a successful wildlsearch lookup, discard the values of numeric
3120 variables because (a) they are in the wrong storage pool and (b) even if
3121 they were copied, it wouldn't work properly because of the caching.
3123 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
3124 checking when decoding. Apparently there are clients that generate
3125 overlong encoded strings. Why am I not surprised?
3127 PH/15 If the first argument of "${if match_address" was not empty, but did not
3128 contain an "@" character, Exim crashed. Now it writes a panic log message
3129 and treats the condition as false.
3131 PH/16 In autoreply, treat an empty string for "once" the same as unset.
3133 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
3134 extension "envelope-auth". The code is finished and in agreement with
3135 other implementations, but there is no documentation so far and in fact,
3136 nobody wrote the draft yet. This extension is currently #undef'ed, thus
3137 not changing the active code.
3139 Print executed "if" and "elsif" statements when debugging is used. This
3140 helps a great deal to understand what a filter does.
3142 Document more things not specified clearly in RFC3028. I had all this
3143 sorted out, when out of a sudden new issues came to my mind. Oops."
3145 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
3148 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
3149 canonical form (as documented). However, after a host name lookup from
3150 the IP address, check_host() was doing a simple string comparison with
3151 addresses acquired from the DNS when checking that the found name did
3152 have the original IP as one of its addresses. Since any found IPv6
3153 addresses are likely to be in abbreviated form, the comparison could
3154 fail. Luckily, there already exists a function for doing the comparison
3155 by converting both addresses to binary, so now that is used instead of
3156 the text comparison.
3158 PH/20 There was another similar case to PH/19, when a complete host name was
3159 given in a host list; looking up its IP address could give an abbreviated
3160 form, whereas the current host's name might or might not be abbreviated.
3161 The same fix has been applied.
3167 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
3168 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
3171 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
3172 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
3174 PH/03 Typo: missing ".o" in src/pcre/Makefile.
3176 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
3177 header line, restrict the check to what is listed in RFCs 2369 and 2929.
3178 Also, for "Auto-Submitted", treat anything other than "no" as
3179 non-personal, in accordance with RFC 3834. (Previously it treated
3180 anything starting "auto-" as non-personal.)
3182 TF/01 The control=submission/name=... option had a problem with syntax
3183 errors if the name included a slash character. The /name= option
3184 now slurps the rest of the string, so it can include any characters
3185 but it must come last in the list of options (after /sender_retain
3188 PH/05 Some modifications to the interface to the fake nameserver for the new
3196 TK/01 Added the "success_on_redirect" address verification option. See
3197 NewStuff for rationale and an example.
3199 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
3201 PH/02 Patch to exigrep to allow it to work on syslog lines.
3203 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
3204 fread() to skip over the body file's header line, because in Cygwin the
3205 header line is locked and is inaccessible.
3207 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
3208 co-exist for some time) to make it clear that it is the Exim ID that is
3209 referenced, not the Message-ID: header line.
3211 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
3212 string_format() function, because snprintf() does not exist on all
3215 PH/06 The use of forbid_filter_existstest now also locks out the use of the
3216 ${stat: expansion item.
3218 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
3219 protocol synchronization error", to keep the pedants happy.
3221 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
3222 well as for IRIX systems, when gcc is being used. See the host.c source
3225 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
3227 PH/10 Named domain lists were not working if used in a queue_smtp_domains
3230 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
3231 transport and to the smtp transport in LMTP mode.
3233 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
3235 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
3236 run a filter in a subprocess. This could lead to confusion in subsequent
3237 lookups in the parent process. There should also be a search_tidyup() at
3238 the end of the subprocess.
3240 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
3241 only if the host matched helo_try_verify_hosts, which caused the
3242 verification to occur when the EHLO/HELO command was issued. The ACL just
3243 tested the remembered result. Now, if a previous verification attempt has
3244 not happened, "verify = helo" does it there and then.
3246 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
3248 TK/03 Fix log output including CR from clamd.
3250 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
3251 exist provoked a memory error which could cause a segfault.
3253 PH/15 Installed PCRE 6.2
3255 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
3257 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
3258 of the problem. Specifically, suggested +O2 rather than +O1 for the
3261 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
3263 PH/20 If a delivery was routed to a non-standard port by means of an SRV
3264 record, the port was not correctly logged when the outgoing_port log
3265 selector was set (it logged the transort's default port).
3267 PH/21 Added support for host-specific ports to manualroute, queryprogram,
3268 fallback_hosts, and "hosts" in the smtp transport.
3270 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
3271 host errors such as "Connection refused".
3273 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
3274 authentication with radiusclient 0.4.9:
3276 - Error returned from rc_read_config was caught wrongly
3277 - Username/password not passed on to radius server due to wrong length.
3279 The presumption is that some radiusclient API changes for 4.51/PH/17
3280 were not taken care of correctly. The code is still untested by me (my
3281 Linux distribution still has 0.3.2 of radiusclient), but it was
3282 contributed by a Radius user.
3284 PH/24 When doing a callout, the value of $domain wasn't set correctly when
3285 expanding the "port" option of the smtp transport.
3287 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
3288 while reading a MIME header. Thanks to Tom Hughes for a patch.
3290 PH/24 Include config.h inside local_scan.h so that configuration settings are
3293 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
3294 that in an ACL for RCPT (for example), you can examine exactly what was
3297 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
3298 commands, but it was not correctly comparing the address with the actual
3299 client host address. Thus, it would show the EHLO address in Received:
3300 header lines when this was not necessary.
3302 PH/27 Added the % operator to ${eval:}.
3304 PH/28 Exim tries to create and chdir to its spool directory when it starts;
3305 it should be ignoring failures (because with -C, for example, it has lost
3306 privilege). It wasn't ignoring creation failures other than "already
3309 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
3312 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
3313 because an input line was too long, either on its own, or by virtue of
3314 too many continuations, the temporary file was not being removed, and the
3315 return code was incorrect.
3317 PH/31 Missing "BOOL" in function definition in filtertest.c.
3319 PH/32 Applied Sieve patches from the maintainer.
3321 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
3323 PH/33 Added "verify = not_blind".
3325 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
3326 Local/Makefile (with some defaults set). These are used in built scripts
3327 such as exicyclog, but they have never been used in the exim_install
3328 script (though there are many overriding facilities there). I have
3329 arranged that the exim_install script now takes note of these two
3332 PH/35 Installed configuration files for Dragonfly.
3334 PH/36 When a locally submitted message by a trusted user did not contain a
3335 From: header, and the sender address was obtained from -f or from an SMTP
3336 MAIL command, and the trusted user did not use -F to supply a sender
3337 name, $originator_name was incorrectly used when constructing a From:
3338 header. Furthermore, $originator_name was used for submission mode
3339 messages from external hosts without From: headers in a similar way,
3340 which is clearly wrong.
3342 PH/37 Added control=suppress_local_fixups.
3344 PH/38 When log_selector = +received_sender was set, and the addition of the
3345 sender made the log line's construction buffer exactly full, or one byte
3346 less than full, an overflow happened when the terminating "\n" was
3349 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
3350 when the result of a list match is failure because a DNS lookup failed.
3352 PH/40 RM_COMMAND is now used in the building process.
3354 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
3355 the "build-* directories that it finds.
3357 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
3358 address in a domain literal was a prefix of an interface address.
3360 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
3361 when verifying a sender address, unless rewrite_headers is false.
3363 PH/44 Wrote a long comment about why errors_to addresses are verified as
3364 recipients, not senders.
3366 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
3367 the ratelimit ACL was added.
3369 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
3371 PH/46 Added extra information about PostgreSQL errors to the error string.
3373 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
3374 avoiding the need to install special zones in a real server. This is
3375 backwards compatible; if it can't find the fake resolver, it drops back.
3376 Thus, both old and new test suites can be run.
3378 TF/02 Added util/ratelimit.pl
3380 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
3381 clock is set back in time.
3383 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
3384 Candler <B.Candler@pobox.com>.
3386 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
3387 OK for addresses that are the result of redirections.
3389 PH/48 A number of further additions for the benefit of the new test suite,
3390 including a fake gethostbyname() that interfaces to the fake DNS resolver
3393 TF/06 The fix for widen_domains has also been applied to qualify_single and
3394 search_parents which are the other dnslookup options that can cause
3397 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
3400 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
3402 TF/07 Exim produced the error message "an SRV record indicated no SMTP
3403 service" if it encountered an MX record with an empty target hostname.
3404 The message is now "an MX or SRV record indicated no SMTP service".
3406 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
3407 if the DNS of the sending site is misconfigured. This is quite a
3408 common situation. This change restores the behaviour of treating a
3409 helo verification defer as a failure.
3411 PH/51 If self=fail was set on a router, the bounce message did not include the
3412 actual error message.
3418 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
3420 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
3421 command itself ended in error, the underlying message about the transport
3422 filter timeout was being overwritten with the pipe command error. Now the
3423 underlying error message should be appended to the second error message.
3425 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
3427 PH/02 Reduce the amount of output that "make" produces by default. Full output
3428 can still be requested.
3430 PH/03 The warning log line about a condition test deferring for a "warn" verb
3431 was being output only once per connection, rather than after each
3432 occurrence (because it was using the same function as for successful
3433 "warn" verbs). This seems wrong, so I have changed it.
3435 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
3436 it should not have, which might have caused a crash in the right
3437 circumstances, but probably never did.
3439 PH/04 Installed a modified version of Tony Finch's patch to make submission
3440 mode fix the return path as well as the Sender: header line, and to
3441 add a /name= option so that you can make the user's friendly name appear
3444 TF/03 Added the control = fakedefer ACL modifier.
3446 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
3447 Mark Lowes for thorough testing.
3449 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
3451 TK/03 Merged latest SRS patch from Miles Wilton.
3453 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
3454 with the definition in sysexits.h (which is #included earlier).
3455 Fortunately, Exim does not actually use EX_OK. The code used to try to
3456 preserve the sysexits.h value, by assumimg that macro definitions were
3457 scanned for macro replacements. I have been disabused of this notion,
3458 so now the code just undefines EX_OK before #including unistd.h.
3460 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
3461 in the smtp transport. When a block could not be written in a single
3462 write() function, the timeout was being re-applied to each part-write.
3463 This seems wrong - if the receiver was accepting one byte at a time it
3464 would take for ever. The timeout is now adjusted when this happens. It
3465 doesn't have to be particularly precise.
3467 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
3468 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
3470 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
3471 without a domain if the check to <postmaster@domain> fails.
3473 SC/01 Eximstats: added -xls and the ability to specify output files
3474 (patch written by Frank Heydlauf).
3476 SC/02 Eximstats: use FileHandles for outputing results.
3478 SC/03 Eximstats: allow any combination of xls, txt, and html output.
3480 SC/04 Eximstats: fixed display of large numbers with -nvr option
3482 SC/05 Eximstats: fixed merging of reports with empty tables.
3484 SC/06 Eximstats: added the -include_original_destination flag
3486 SC/07 Eximstats: removed tabs and trailing whitespace.
3488 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
3490 TK/06 MBOX spool code: Add real "From " MBOX separator line
3491 so the .eml file is really in mbox format (even though
3492 most programs do not really care). Patch from Alex Miller.
3494 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
3495 The latter is generated from $received_to and is only set if the
3496 message has one envelope recipient. SA can use these headers,
3497 obviously out-of-the-box. Patch from Alex Miller.
3499 PH/08 The ${def test on a variable was returning false if the variable's
3500 value was "0", contrary to what the specification has always said!
3501 The result should be true unless the variable is empty.
3503 PH/09 The syntax error of a character other than { following "${if
3504 def:variable_name" (after optional whitespace) was not being diagnosed.
3505 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
3506 accidental colon was present, for example, could give incorrect results.
3508 PH/10 Tidied the code in a number of places where the st_size field of a stat()
3509 result is used (not including appendfile, where other changes are about
3512 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
3513 This involved changing a lot of size variables from int to off_t. It
3514 should work with maildirs and everything.
3516 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
3517 spamd dying while we are connected to it.
3519 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
3522 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
3523 function for BDB 4.3.
3525 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
3527 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
3528 the sources for PCRE have been re-arranged and more files are now
3531 PH/15 The code I had for printing potentially long long variables in PH/11
3532 above was not the best (it lost precision). The length of off_t variables
3533 is now inspected at build time, and an appropriate printing format (%ld
3534 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
3535 to be "long long int" or "long int". This is needed for the internal
3536 formatting function string_vformat().
3538 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
3539 the configuration file to be ":syslog", then the script "guesses" where
3540 the logs files are, rather than using the compiled in default. In our
3541 case the guess is not the same as the compiled default, so the script
3542 suddenly stopped working when I started to use syslog. The patch checks
3543 to see if log_file_path is "". If so, it attempts to read it from exim
3544 with no configuration file to get the compiled in version, before it
3545 falls back to the previous guessing code."
3547 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
3548 implementing BATV in an Exim configuration. See NewStuff for the gory
3551 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
3552 Makefile that are specific to HP-UX.
3554 PH/18 If the "use_postmaster" option was set for a recipient callout together
3555 with the "random" option, the postmaster address was used as the MAIL
3556 FROM address for the random test, but not for the subsequent recipient
3557 test. It is now used for both.
3559 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
3560 patch removes a few documentation additions to RFC 3028, because the
3561 latest draft now contains them. It adds the new en;ascii-case comparator
3562 and a new error check for 8bit text in MIME parts. Comparator and
3563 require names are now matched exactly. I enabled the subaddress
3564 extension, but it is not well tested yet (read: it works for me)."
3566 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
3567 rework some of the code of TK/09 above to avoid the hardwired use of
3568 "%lld" and "long long". Replaced the call to snprintf() with a call to
3571 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
3572 records point to non-existent hosts", "retry timeout exceeded", and
3573 "retry time not reached for any host after a long failure period".
3575 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
3576 experimental DomainKeys support:
3578 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
3579 (2) On an error such as an illegally used "control", the wrong name for
3580 the control was given.
3582 These problems did NOT occur unless DomainKeys support was compiled.
3584 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
3586 PH/24 Added ${if match_ip condition.
3588 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
3589 databases so that it will be absolutely obvious if a crash occurs in the
3590 DB library. This is a regular occurrence (often caused by mis-matched
3593 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
3594 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
3595 picked up on a user's system that detects such things. There doesn't seem
3596 to be a gcc warning option for this - only an attribute that has to be
3597 put on the function's prototype. It seems that in Fedora Core 4 they have
3598 set this on a number of new functions. No doubt there will be more in due
3601 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
3602 specify a transport. However, if an address that was verified by such a
3603 router was the subject of a callout, Exim crashed because it tried to
3604 read the rcpt_include_affixes from the non-existent transport. Now it
3605 just assumes that the setting of that option is false. This bug was
3606 introduced by 4.51/PH/31.
3608 PH/28 Changed -d+all to exclude +memory, because that information is very
3609 rarely of interest, but it makes the output a lot bigger. People tend to
3610 do -d+all out of habit.
3612 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
3613 code in os-type was giving problems when libc.so lives in lib64, like on
3616 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
3617 aren't the modern standard, and it seems that some systems' include files
3618 don't always have them. Exim was already checking for some of the newer
3619 ones like T_AAAA, and defining it itself. I've added checks for all the
3620 record types that Exim uses.
3622 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
3623 not automatically generating a new one, as it is supposed to. This
3624 prevented TLS from working. If the file did exist, but contained invalid
3625 data, a new version was generated, as expected. It was only the case of a
3626 non-existent file that was broken.
3628 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
3629 with a change in libdomainkeys > 0.64.
3631 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
3632 from DNS. If the selector record carries the flag, it now has
3633 precedence over the domain-wide flag.
3635 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
3637 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
3638 the use of a port name that isn't defined in /etc/services) occurred, the
3639 message was deferred as in a normal delivery, and thus remained on the
3640 spool, instead of being failed because of the mua_wrapper setting. This
3641 is now fixed, and I tidied up some of the mua_wrapper messages at the
3644 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
3645 the messages in a hash of arrays rather than using individual hashes.
3646 This is a bit cleaner and results in dramatic memory savings, albeit
3647 at a slight CPU cost.
3649 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
3650 as requested by Marc Sherman.
3652 SC/10 Eximstats: added histograms for user specified patterns as requested
3655 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
3657 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
3658 fopen() in the content-scanning modules that did not already have it.
3664 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
3665 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
3667 TK/02 Fix ACL "control" statement not being available in MIME ACL.
3669 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
3671 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
3672 to test Sieve filters that use "vacation".
3674 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
3675 that changes the way the GnuTLS parameters are stored in the cache file.
3676 The new format can be generated externally. For backward compatibility,
3677 if the data in the cache doesn't make sense, Exim assumes it has read an
3678 old-format file, and it generates new data and writes a new file. This
3679 means that you can't go back to an older release without removing the
3682 PH/03 A redirect router that has both "unseen" and "one_time" set does not
3683 work if there are any delivery delays because "one_time" forces the
3684 parent to be marked "delivered", so its unseen clone is never tried
3685 again. For this reason, Exim now forbids the simultaneous setting of
3688 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
3689 redirected to themselves ("homonym" addresses). Read the long ChangeLog
3690 entry if you want to know the details. The fix, however, neglected to
3691 consider the case when local delivery batching is involved. The test for
3692 "previously delivered" was not happening when checking to see if an
3693 address could be batched with a previous (undelivered) one; under
3694 certain circumstances this could lead to multiple deliveries to the same
3697 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
3698 in its include files, and this causes problems building Exim.
3700 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
3701 header_syntax) but Exim was just ignoring anything given after a slash.
3702 In particular, this caused confusion with an attempt to use "verify =
3703 reverse_host_lookup/defer_ok". An error is now given when options are
3704 supplied for verify items that do not have them. (Maybe reverse_host_
3705 lookup should have a defer_ok option, but that's a different point.)
3707 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
3708 defined by RFC 821) to 2048, because there were problems with some AUTH
3709 commands, and RFC 1869 says the size should be increased for extended
3710 SMTP commands that take arguments.
3712 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
3715 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
3716 "unknown" error; now it says that the functionality isn't in the binary.
3718 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
3719 an address' error message when a string expansion fails (syntax or
3720 whatever). Otherwise the password may appear in the log. Following change
3721 PH/42 below, there is no longer a chance of it appearing in a bounce
3724 PH/11 Installed exipick version 20050225.0 from John Jetmore.
3726 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
3727 of its addresses was ever tried. (Bugzilla bug #2.)
3729 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
3730 the result incorrectly in the debug output. (It correctly added a newline
3731 to what was transported.)
3733 TF/01 Added $received_time.
3735 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
3736 commented out examples of how to interface to a virus scanner and to
3737 SpamAssassin. Also added commented examples of av_scanner and
3738 spamd_address settings.
3740 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
3741 and controls are allowed in which ACLs. There were a couple of minor
3742 errors. Some of the entries in the conditions table (which is a table of
3743 where they are NOT allowed) were getting very unwieldy; rewrote them as a
3744 negation of where the condition IS allowed.
3746 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
3748 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
3749 header file does not have a version number, so I've had to invent a new
3750 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
3751 API. The code is untested by me (my Linux distribution still has 0.3.2 of
3752 radiusclient), but it was contributed by a Radius user.
3754 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
3755 files or directories, for OpenSSL.
3757 PH/19 When an Exim process that is running as root has to create an Exim log
3758 file, it does so in a subprocess that runs as exim:exim so as to get the
3759 ownership right at creation (otherwise, other Exim processes might see
3760 the file with the wrong ownership). There was no test for failure of this
3761 fork() call, which would lead to the process getting stuck as it waited
3762 for a non-existent subprocess. Forks do occasionally fail when resources
3763 run out. I reviewed all the other calls to fork(); they all seem to check
3766 PH/20 When checking for unexpected SMTP input at connect time (before writing
3767 the banner), Exim was not dealing correctly with a non-positive return
3768 from the read() function. If the client had disconnected by this time,
3769 the result was a log entry for a synchronization error with an empty
3770 string after "input=" when read() returned zero. If read() returned -1
3771 (an event I could not check), uninitialized data bytes were printed.
3772 There were reports of junk text (parts of files, etc) appearing after
3775 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
3777 PH/22 Added support for macro redefinition, and (re)definition in between
3778 driver and ACL definitions.
3780 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
3781 forgetting to use the resulting value; it was using the unexpanded value.
3783 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
3784 hadn't been configured. The fix is from Juergen Kreileder, who
3785 understands it better than I do:
3787 "Here's what I see happening with three configured cyrus_sasl
3788 authenticators configured (plain, login, cram-md5):
3790 On startup auth_cyrus_sasl_init() gets called for each of these.
3791 This means three calls to sasl_listmech() without a specified mech_list.
3792 => SASL tests which mechs of all available mechs actually work
3793 => three warnings about OTP not working
3794 => the returned list contains: plain, login, cram-md5, digest-md5, ...
3796 With the patch, sasl_listmech() also gets called three times. But now
3797 SASL's mech_list option is set to the server_mech specified in the the
3798 authenticator. Or in other words, the answer from sasl_listmech()
3799 gets limited to just the mech you're testing for (which is different
3801 => the return list contains just 'plain' or 'login', 'cram-md5' or
3802 nothing depending on the value of ob->server_mech.
3804 I've just tested the patch: Authentication still works fine,
3805 unavailable mechs specified in the exim configuration are still
3806 caught, and the auth.log warnings about OTP are gone."
3808 PH/25 When debugging is enabled, the contents of the command line are added
3809 to the debugging output, even when log_selector=+arguments is not
3812 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
3813 answer is "GNU", and only if the return is "GNU/something" is the answer
3816 PH/27 $acl_verify_message is now set immediately after the failure of a
3817 verification in an ACL, and so is available in subsequent modifiers. In
3818 particular, the message can be preserved by coding like this:
3820 warn !verify = sender
3821 set acl_m0 = $acl_verify_message
3823 Previously, $acl_verify_message was set only while expanding "message"
3824 and "log_message" when a very denied access.
3826 PH/28 Modified OS/os.c-Linux with
3828 -#ifndef OS_LOAD_AVERAGE
3829 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
3831 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
3832 nomenclature these days.)
3834 PH/29 Installed patch from the Sieve maintainer that adds the options
3835 sieve_useraddress and sieve_subaddress to the redirect router.
3837 PH/30 In these circumstances:
3838 . Two addresses routed to the same list of hosts;
3839 . First host does not offer TLS;
3840 . First host accepts first address;
3841 . First host gives temporary error to second address;
3842 . Second host offers TLS and a TLS session is established;
3843 . Second host accepts second address.
3844 Exim incorrectly logged both deliveries with the TLS parameters (cipher
3845 and peerdn, if requested) that were in fact used only for the second
3848 PH/31 When doing a callout as part of verifying an address, Exim was not paying
3849 attention to any local part prefix or suffix that was matched by the
3850 router that accepted the address. It now behaves in the same way as it
3851 does for delivery: the affixes are removed from the local part unless
3852 rcpt_include_affixes is set on the transport.
3854 PH/32 Add the sender address, as F=<...>, to the log line when logging a
3855 timeout during the DATA phase of an incoming message.
3857 PH/33 Sieve envelope tests were broken for match types other than :is. I have
3858 applied a patch sanctioned by the Sieve maintainer.
3860 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
3861 the uid or gid is negative. A case of a negative gid caused this to be
3862 noticed. The fix allows for either to be negative.
3864 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
3865 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
3868 PH/36 The change PH/12 above was broken. Fixed it.
3870 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
3871 the grounds that routing the same address twice would always produce the
3872 same answer. This might have been true once, but it is certainly no
3873 longer true now. Routing a child address may depend on the previous
3874 routing that produced that child. Some complicated redirection strategies
3875 went wrong when messages had multiple recipients, and made Exim's
3876 behaviour dependent on the order in which the addresses were given.
3878 I have moved the duplicate checking until after the routing is complete.
3879 Exim scans the addresses that are assigned to local and remote
3880 transports, and removes any duplicates. This means that more work will be
3881 done, as duplicates will always all be routed, but duplicates are
3882 presumably rare, so I don't expect this is of any significance.
3884 For deliveries to pipes, files, and autoreplies, the duplicate checking
3885 still happens during the routing process, since they are not going to be
3888 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
3889 It corrects a timeout issue with spamd. This is Ian's comment: "The
3890 background is that sometimes spamd either never reads data from a
3891 connection it has accepted, or it never writes response data. The exiscan
3892 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
3893 blindly assumes that writes won't block so it may never time out."
3895 PH/39 Allow G after quota size as well as K and M.
3897 PH/40 The value set for $authenticated_id in an authenticator may not contain
3898 binary zeroes or newlines because the value is written to log lines and
3899 to spool files. There was no check on this. Now the value is run through
3900 the string_printing() function so that such characters are converted to
3901 printable escape sequences.
3903 PH/41 $message_linecount is a new variable that contains the total number of
3904 lines in the message. Compare $body_linecount, which is the count for the
3907 PH/42 Exim no longer gives details of delivery errors for specific addresses in
3908 bounce and delay warning messages, except in certain special cases, which
3911 (a) An SMTP error message from a remote host;
3912 (b) A message specified in a :fail: redirection;
3913 (c) A message specified in a "fail" command in a system filter;
3914 (d) A message specified in a FAIL return from the queryprogram router;
3915 (e) A message specified by the cannot_route_message router option.
3917 In these cases only, Exim does include the error details in bounce and
3918 warning messages. There are also a few cases where bland messages such
3919 as "unrouteable address" or "local delivery error" are given.
3921 PH/43 $value is now also set for the "else" part of a ${run expansion.
3923 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
3924 being worked on, but at least Exim now implements the latest version to
3927 PH/45 In a pipe transport, although a timeout while waiting for the pipe
3928 process to complete was treated as a delivery failure, a timeout while
3929 writing the message to the pipe was logged, but erroneously treated as a
3930 successful delivery. Such timeouts include transport filter timeouts. For
3931 consistency with the overall process timeout, these timeouts are now
3932 treated as errors, giving rise to delivery failures by default. However,
3933 there is now a new Boolean option for the pipe transport called
3934 timeout_defer, which, if set TRUE, converts the failures into defers for
3935 both kinds of timeout. A transport filter timeout is now identified in
3938 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
3939 systems where "make" and "gmake" are different, calling "gmake" at top
3940 level broke things. I've arranged for the value of $(MAKE) to be passed
3941 from the Makefile to this script so that it can call the same version of
3945 A note about Exim versions 4.44 and 4.50
3946 ----------------------------------------
3948 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
3949 changes of various kinds. As a consequence, a big documentation update was
3950 needed. This delayed the release for rather longer than seemed good, especially
3951 in the light of a couple of (minor) security issues. Therefore, the changes
3952 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
3953 release. So 4.44 and 4.50 are in effect two different branches that both start
3956 I have left the 4.50 change log unchanged; it contains all the changes since
3957 4.43. The change log for 4.44 is below; many of its items are identical to
3958 those for 4.50. This seems to be the most sensible way to preserve the
3959 historical information.
3965 1. Minor wording change to the doc/README.SIEVE file.
3967 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
3968 computation of the current number of files was incorrect.
3970 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
3971 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
3974 4. Give more explanation in the error message when the command for a transport
3975 filter fails to execute.
3977 5. There are several places where Exim runs a non-Exim command in a
3978 subprocess. The SIGUSR1 signal should be disabled for these processes. This
3979 was being done only for the command run by the queryprogram router. It is
3980 now done for all such subprocesses. The other cases are: ${run, transport
3981 filters, and the commands run by the lmtp and pipe transports.
3983 6. Added CONFIGURE_GROUP build-time option.
3985 7. Some older OS have a limit of 256 on the maximum number of file
3986 descriptors. Exim was using setrlimit() to set 1000 as a large value
3987 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
3988 systems. I've change it so that if it can't get 1000, it tries for 256.
3990 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
3991 was an oversight, and furthermore, ever since the addition of extra
3992 controls (e.g. 4.43/32), the checks on when to allow different forms of
3993 "control" were broken. There should now be diagnostics for all cases when a
3994 control that does not make sense is encountered.
3996 9. Added the /retain_sender option to "control=submission".
3998 10. $recipients is now available in the predata ACL (oversight).
4000 11. Tidy the search cache before the fork to do a delivery from a message
4001 received from the command line. Otherwise the child will trigger a lookup
4002 failure and thereby defer the delivery if it tries to use (for example) a
4003 cached ldap connection that the parent has called unbind on.
4005 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4006 of $address_data from the recipient verification was clobbered by the
4007 sender verification.
4009 13. The value of address_data from a sender verification is now available in
4010 $sender_address_data in subsequent conditions in the ACL statement.
4012 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
4014 15. Added a new option "connect=<time>" to callout options, to set a different
4017 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4018 was its contents. (It was OK if the option was not defined at all.)
4020 17. A "Completed" log line is now written for messages that are removed from
4021 the spool by the -Mrm option.
4023 18. New variables $sender_verify_failure and $recipient_verify_failure contain
4024 information about exactly what failed.
4026 19. Added -dd to debug only the daemon process.
4028 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
4029 handles timeouts, both on the server side and network timeouts. Renamed the
4030 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
4032 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
4033 It is now set to "smtps".
4035 22. $host_address is now set to the target address during the checking of
4036 ignore_target_hosts.
4038 23. When checking ignore_target_hosts for an ipliteral router, no host name was
4039 being passed; this would have caused $sender_host_name to have been used if
4040 matching the list had actually called for a host name (not very likely,
4041 since this list is usually IP addresses). A host name is now passed as
4044 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4045 code that specifies a non-restarting handler (typically sigaction() in
4046 modern systems) in an attempt to fix a rare and obscure crash bug.
4048 25. Narrowed the window for a race in the daemon that could cause it to ignore
4049 SIGCHLD signals. This is not a major problem, because they are used only to
4050 wake it up if nothing else does.
4052 26. A malformed maildirsize file could cause Exim to calculate negative values
4053 for the mailbox size or file count. Odd effects could occur as a result.
4054 The maildirsize information is now recalculated if the size or filecount
4057 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4058 support for a long time. Removed HAVE_SYS_VFS_H.
4060 28. Installed the latest version of exipick from John Jetmore.
4062 29. In an address list, if the pattern was not a regular expression, an empty
4063 subject address (from a bounce message) matched only if the pattern was an
4064 empty string. Non-empty patterns were not even tested. This was the wrong
4065 because it is perfectly reasonable to use an empty address as part of a
4066 database query. An empty address is now tested by patterns that are
4067 lookups. However, all the other forms of pattern expect the subject to
4068 contain a local part and a domain, and therefore, for them, an empty
4069 address still always fails if the pattern is not itself empty.
4071 30. Exim went into a mad DNS loop when attempting to do a callout where the
4072 host was specified on an smtp transport, and looking it up yielded more
4073 than one IP address.
4075 31. Re-factored the code for checking spool and log partition space into a
4076 function that finds that data and another that does the check. The former
4077 is then used to implement four new variables: $spool_space, $log_space,
4078 $spool_inodes, and $log_inodes.
4080 32. The RFC2047 encoding function was originally intended for short strings
4081 such as real names; it was not keeping to the 75-character limit for
4082 encoded words that the RFC imposes. It now respects the limit, and
4083 generates multiple encoded words if necessary. To be on the safe side, I
4084 have increased the buffer size for the ${rfc2047: expansion operator from
4087 33. It is now permitted to omit both strings after an "if" condition; if the
4088 condition is true, the result is "true". As before, when the second string
4089 is omitted, a false condition yields an empty string. This makes it less
4090 cumbersome to write custom ACL and router conditions.
4092 34. Failure to deliver a bounce message always caused it to be frozen, even if
4093 there was an errors_to setting on the router. The errors_to setting is now
4096 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4097 canonical form (fully expanded) before being placed in
4098 $sender_host_address.
4100 36. The table in the code that translates DNS record types into text (T_A to
4101 "A" for instance) was missing entries for NS and CNAME. It is just possible
4102 that this could have caused confusion if both these types were looked up
4103 for the same domain, because the text type is used as part of Exim's
4104 per-process caching. But the chance of anyone hitting this buglet seems
4107 37. The dnsdb lookup has been extended in a number of ways.
4109 (1) There is a new type, "zns", which walks up the domain tree until it
4110 finds some nameserver records. It should be used with care.
4112 (2) There is a new type, "mxh", which is like "mx" except that it returns
4113 just the host names, not the priorities.
4115 (3) It is now possible to give a list of domains (or IP addresses) to be
4116 looked up. The behaviour when one of the lookups defers can be
4117 controlled by a keyword.
4119 (4) It is now possible to specify the separator character for use when
4120 multiple records are returned.
4122 38. The dnslists ACL condition has been extended: it is now possible to supply
4123 a list of IP addresses and/or domains to be looked up in a particular DNS
4126 39. Added log_selector=+queue_time_overall.
4128 40. When running the queue in the test harness, wait just a tad after forking a
4129 delivery process, to get repeatability of debugging output.
4131 41. Include certificate and key file names in error message when GnuTLS fails
4132 to set them up, because the GnuTLS error message doesn't include the name
4133 of the failing file when there is a problem reading it.
4135 42. Allow both -bf and -bF in the same test run.
4137 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
4139 44. The "Exiscan patch" is now merged into the mainline Exim source.
4141 45. Sometimes the final signoff response after QUIT could fail to get
4142 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4143 < 0 before doing a fflush(). This bug looks as though it goes back to the
4144 introduction of TLS in release 3.20, but "sometimes" must have been rare
4145 because the tests only now provoked it.
4147 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
4148 (this can affect the format of dates).
4150 47. exim_tidydb, when checking for the continued existence of a message for
4151 which it has found a message-specific retry record, was not finding
4152 messages that were in split spool directories. Consequently, it was
4153 deleting retry records that should have stayed in existence.
4155 48. Steve fixed some bugs in eximstats.
4157 49. The SPA authentication driver was not abandoning authentication and moving
4158 on to the next authenticator when an expansion was forced to fail,
4159 contradicting the general specification for all authenticators. Instead it
4160 was generating a temporary error. It now behaves as specified.
4162 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
4163 (the order specifies the preference for clients). The order is now AES256,
4164 AES128, 3DES, ARCFOUR128.
4166 51. Small patch to Sieve code - explicitly set From: when generating an
4169 52. Exim crashed if a remote delivery caused a very long error message to be
4170 recorded - for instance if somebody sent an entire SpamAssassin report back
4171 as a large number of 550 error lines. This bug was coincidentally fixed by
4172 increasing the size of one of Exim's internal buffers (big_buffer) that
4173 happened as part of the Exiscan merge. However, to be on the safe side, I
4174 have made the code more robust (and fixed the comments that describe what
4177 53. Now that there can be additional text after "Completed" in log lines (if
4178 the queue_time_overall log selector is set), a one-byte patch to exigrep
4179 was needed to allow it to recognize "Completed" as not the last thing in
4182 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
4183 patch that reportedly fixes this has been added. I am not expert enough to
4184 create a test for it. This is what the patch creator wrote:
4186 "I found a little strange behaviour of ldap code when working with
4187 Windows 2003 AD Domain, where users was placed in more than one
4188 Organization Units. When I tried to give exim partial DN, the exit code
4189 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
4190 But simultaneously result of request was absolutely normal ldap result,
4191 so I produce this patch..."
4193 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
4194 REFERENCE, so I have modified the code to exclude the patch when that macro
4197 55. Some experimental protocols are using DNS PTR records for new purposes. The
4198 keys for these records are domain names, not reversed IP addresses. The
4199 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4200 leaves it alone. Component reversal etc. now happens only for IP addresses.
4203 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
4205 57. Double the size of the debug message buffer (to 2048) so that more of very
4206 long debug lines gets shown.
4208 58. The exicyclog utility now does better if the number of log files to keep
4209 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
4211 59. Two changes related to the smtp_active_hostname option:
4213 (1) $smtp_active_hostname is now available as a variable.
4214 (2) The default for smtp_banner uses $smtp_active_hostname instead
4215 of $primary_hostname.
4217 60. The host_aton() function is supposed to be passed a string that is known
4218 to be a valid IP address. However, in the case of IPv6 addresses, it was
4219 not checking this. This is a hostage to fortune. Exim now panics and dies
4220 if the condition is not met. A case was found where this could be provoked
4221 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4222 components; fortuitously, this particular loophole had already been fixed
4223 by change 4.50/55 above.
4225 If there are any other similar loopholes, the new check in host_aton()
4226 itself should stop them being exploited. The report I received stated that
4227 data on the command line could provoke the exploit when Exim was running as
4228 exim, but did not say which command line option was involved. All I could
4229 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4230 running as the user.
4233 61. There was a buffer overflow vulnerability in the SPA authentication code
4234 (which came originally from the Samba project). I have added a test to the
4235 spa_base64_to_bits() function which I hope fixes it.
4238 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
4239 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
4241 63. The daemon start-up calls getloadavg() while still root for those OS that
4242 need the first call to be done as root, but it missed one case: when
4243 deliver_queue_load_max is set with deliver_drop_privilege. This is
4244 necessary for the benefit of the queue runner, because there is no re-exec
4245 when deliver_drop_privilege is set.
4247 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
4248 This has been fixed.
4250 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4251 was in use, was not putting the data itself into the right store pool;
4252 consequently, it could be overwritten for a subsequent message in the same
4253 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4256 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
4258 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
4259 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
4260 the function were treating the return as a boolean value, which happened to
4261 work because 0=false and not-0=true, but is not correct code.
4263 68. The host_aton() function was not handling scoped IPv6 addresses (those
4264 with, for example, "%eth0" on the end) correctly.
4266 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
4267 negated items (that is, ~something) in unsigned ints. Some compilers
4268 apparently mutter when there is no cast.
4270 70. If an address verification called from an ACL failed, and did not produce a
4271 user-specific message (i.e. there was only a "system" message), nothing was
4272 put in $acl_verify_message. In this situation, it now puts the system
4275 71. Change 4.23/11 added synchronization checking at the start of an SMTP
4276 session; change 4.31/43 added the unwanted input to the log line - except
4277 that it did not do this in the start of session case. It now does.
4279 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
4280 This is wrong and can cause the other end to generate a synchronization
4281 error if it is another Exim or anything else that does the synchronization
4282 check. A QUIT command is no longer sent after a timeout.
4284 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
4285 during host lookups.
4287 74. The defer_ok option of callout verification was not working if it was used
4288 when verifying addresses in header lines, that is, for this case:
4290 verify = header_sender/callout=defer_ok
4292 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
4293 those file descriptors could be used for SMTP connections. If anything
4294 wrote to stderr (the example that came up was "warn" in embedded Perl), it
4295 could be sent to the SMTP client, causing chaos. The daemon now opens
4296 stdin, stdout, and stderr to /dev/null when it puts itself into the
4299 76. Arrange for output from Perl's "warn" command to be written to Exim's main
4300 log by default. The user can override this with suitable Perl magic.
4302 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
4303 the log message when discard triggers, was not working for the DATA ACL or
4304 for the non-SMTP ACL.
4306 78. Error message wording change in sieve.c.
4308 79. If smtp_accept_max_per_host was set, the number of connections could be
4309 restricted to fewer than expected, because the daemon was trying to set up
4310 a new connection before checking whether the processes handling previous
4311 connections had finished. The check for completed processes is now done
4312 earlier. On busy systems, this bug wouldn't be noticed because something
4313 else would have woken the daemon, and it would have reaped the completed
4316 80. If a message was submitted locally by a user whose login name contained one
4317 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
4318 It caused a spool format error. I have fixed the spool reading code. A
4319 related problem was that the "from" clause in the Received: line became
4320 illegal because of the space(s). It is now covered by ${quote_local_part.
4322 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
4325 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
4327 83. Patch from Sieve maintainer to fix unterminated string problem in
4328 "vacation" handling.
4330 84. Some minor changes to the Linux configuration files to help with other
4331 OS variants using glibc.
4333 85. One more patch for Sieve to update vacation handling to latest spec.
4336 ----------------------------------------------------
4337 See the note above about the 4.44 and 4.50 releases.
4338 ----------------------------------------------------
4344 1. Change 4.43/35 introduced a bug that caused file counts to be
4345 incorrectly computed when quota_filecount was set in an appendfile
4348 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4349 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4352 3. Give more explanation in the error message when the command for a transport
4353 filter fails to execute.
4355 4. There are several places where Exim runs a non-Exim command in a
4356 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4357 was being done only for the command run by the queryprogram router. It is
4358 now done for all such subprocesses. The other cases are: ${run, transport
4359 filters, and the commands run by the lmtp and pipe transports.
4361 5. Some older OS have a limit of 256 on the maximum number of file
4362 descriptors. Exim was using setrlimit() to set 1000 as a large value
4363 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4364 systems. I've change it so that if it can't get 1000, it tries for 256.
4366 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4367 was an oversight, and furthermore, ever since the addition of extra
4368 controls (e.g. 4.43/32), the checks on when to allow different forms of
4369 "control" were broken. There should now be diagnostics for all cases when a
4370 control that does not make sense is encountered.
4372 7. $recipients is now available in the predata ACL (oversight).
4374 8. Tidy the search cache before the fork to do a delivery from a message
4375 received from the command line. Otherwise the child will trigger a lookup
4376 failure and thereby defer the delivery if it tries to use (for example) a
4377 cached ldap connection that the parent has called unbind on.
4379 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4380 of $address_data from the recipient verification was clobbered by the
4381 sender verification.
4383 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4384 was its contents. (It was OK if the option was not defined at all.)
4386 11. A "Completed" log line is now written for messages that are removed from
4387 the spool by the -Mrm option.
4389 12. $host_address is now set to the target address during the checking of
4390 ignore_target_hosts.
4392 13. When checking ignore_target_hosts for an ipliteral router, no host name was
4393 being passed; this would have caused $sender_host_name to have been used if
4394 matching the list had actually called for a host name (not very likely,
4395 since this list is usually IP addresses). A host name is now passed as
4398 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4399 code that specifies a non-restarting handler (typically sigaction() in
4400 modern systems) in an attempt to fix a rare and obscure crash bug.
4402 15. Narrowed the window for a race in the daemon that could cause it to ignore
4403 SIGCHLD signals. This is not a major problem, because they are used only to
4404 wake it up if nothing else does.
4406 16. A malformed maildirsize file could cause Exim to calculate negative values
4407 for the mailbox size or file count. Odd effects could occur as a result.
4408 The maildirsize information is now recalculated if the size or filecount
4411 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4412 support for a long time. Removed HAVE_SYS_VFS_H.
4414 18. Updated exipick to current release from John Jetmore.
4416 19. Allow an empty sender to be matched against a lookup in an address list.
4417 Previously the only cases considered were a regular expression, or an
4420 20. Exim went into a mad DNS lookup loop when doing a callout where the
4421 host was specified on the transport, if the DNS lookup yielded more than
4424 21. The RFC2047 encoding function was originally intended for short strings
4425 such as real names; it was not keeping to the 75-character limit for
4426 encoded words that the RFC imposes. It now respects the limit, and
4427 generates multiple encoded words if necessary. To be on the safe side, I
4428 have increased the buffer size for the ${rfc2047: expansion operator from
4431 22. Failure to deliver a bounce message always caused it to be frozen, even if
4432 there was an errors_to setting on the router. The errors_to setting is now
4435 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4436 canonical form (fully expanded) before being placed in
4437 $sender_host_address.
4439 24. Updated eximstats to version 1.33
4441 25. Include certificate and key file names in error message when GnuTLS fails
4442 to set them up, because the GnuTLS error message doesn't include the name
4443 of the failing file when there is a problem reading it.
4445 26. Expand error message when OpenSSL has problems setting up cert/key files.
4448 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
4449 (this can affect the format of dates).
4451 28. exim_tidydb, when checking for the continued existence of a message for
4452 which it has found a message-specific retry record, was not finding
4453 messages that were in split spool directories. Consequently, it was
4454 deleting retry records that should have stayed in existence.
4456 29. eximstats updated to version 1.35
4457 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
4458 1.35 - bugfix such that pie charts by volume are generated correctly
4460 30. The SPA authentication driver was not abandoning authentication and moving
4461 on to the next authenticator when an expansion was forced to fail,
4462 contradicting the general specification for all authenticators. Instead it
4463 was generating a temporary error. It now behaves as specified.
4465 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
4466 (the order specifies the preference for clients). The order is now AES256,
4467 AES128, 3DES, ARCFOUR128.
4469 31. Small patch to Sieve code - explicitly set From: when generating an
4472 32. Exim crashed if a remote delivery caused a very long error message to be
4473 recorded - for instance if somebody sent an entire SpamAssassin report back
4474 as a large number of 550 error lines. This bug was coincidentally fixed by
4475 increasing the size of one of Exim's internal buffers (big_buffer) that
4476 happened as part of the Exiscan merge. However, to be on the safe side, I
4477 have made the code more robust (and fixed the comments that describe what
4480 33. Some experimental protocols are using DNS PTR records for new purposes. The
4481 keys for these records are domain names, not reversed IP addresses. The
4482 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4483 leaves it alone. Component reversal etc. now happens only for IP addresses.
4486 34. The host_aton() function is supposed to be passed a string that is known
4487 to be a valid IP address. However, in the case of IPv6 addresses, it was
4488 not checking this. This is a hostage to fortune. Exim now panics and dies
4489 if the condition is not met. A case was found where this could be provoked
4490 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4491 components; fortuitously, this particular loophole had already been fixed
4492 by change 4.50/55 or 4.44/33 above.
4494 If there are any other similar loopholes, the new check in host_aton()
4495 itself should stop them being exploited. The report I received stated that
4496 data on the command line could provoke the exploit when Exim was running as
4497 exim, but did not say which command line option was involved. All I could
4498 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4499 running as the user.
4502 35. There was a buffer overflow vulnerability in the SPA authentication code
4503 (which came originally from the Samba project). I have added a test to the
4504 spa_base64_to_bits() function which I hope fixes it.
4507 36. The daemon start-up calls getloadavg() while still root for those OS that
4508 need the first call to be done as root, but it missed one case: when
4509 deliver_queue_load_max is set with deliver_drop_privilege. This is
4510 necessary for the benefit of the queue runner, because there is no re-exec
4511 when deliver_drop_privilege is set.
4513 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4514 was in use, was not putting the data itself into the right store pool;
4515 consequently, it could be overwritten for a subsequent message in the same
4516 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4519 38. Sometimes the final signoff response after QUIT could fail to get
4520 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4521 < 0 before doing a fflush(). This bug looks as though it goes back to the
4522 introduction of TLS in release 3.20, but "sometimes" must have been rare
4523 because the tests only now provoked it.
4529 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
4530 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
4531 Now it returns an integer. A number of calls were still expecting a T/F
4532 return. Fortuitously, in all cases, the tests worked in OK situations,
4533 which is the norm. However, things would have gone wrong on any write
4534 failures on the smtp file descriptor. This function is used when sending
4535 messages over SMTP and also when doing verify callouts.
4537 2. When Exim is called to do synchronous delivery of a locally submitted
4538 message (the -odf or -odi options), it no longer closes stderr before doing
4541 3. Implemented the mua_wrapper option.
4543 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
4545 5. Implemented the functions header_remove(), header_testname(),
4546 header_add_at_position(), and receive_remove_recipient(), and exported them
4549 6. If an ACL "warn" statement specified the addition of headers, Exim already
4550 inserted X-ACL-Warn: at the start if there was no header name. However, it
4551 was not making this test for the second and subsequent header lines if
4552 there were newlines in the string. This meant that an invalid header could
4553 be inserted if Exim was badly configured.
4555 7. Allow an ACL "warn" statement to add header lines at the start or after all
4556 the Received: headers, as well as at the end.
4558 8. Added the rcpt_4xx retry error code.
4560 9. Added postmaster_mailfrom=xxx to callout verification option.
4562 10. Added mailfrom=xxxx to the callout verification option, for verify=
4565 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
4566 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
4568 12. Inserted some casts to stop certain compilers warning when using pointer
4569 differences as field lengths or precisions in printf-type calls (mostly
4570 affecting debugging statements).
4572 13. Added optional readline() support for -be (dynamically loaded).
4574 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
4575 same clock tick as a message's arrival, so that its received time was the
4576 same as the "first fail" time on the retry record, and that message
4577 remained on the queue past the ultimate address timeout, every queue runner
4578 would try a delivery (because it was past the ultimate address timeout) but
4579 after another failure, the ultimate address timeout, which should have then
4580 bounced the address, did not kick in. This was a "< instead of <=" error;
4581 in most cases the first failure would have been in the next clock tick
4582 after the received time, and all would be well.
4584 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
4585 being recognized when the domain list was tested by the match_domain
4586 condition in an expansion string.
4588 16. Added the ${str2b64: operator.
4590 17. Exim was always calling setrlimit() to set a large limit for the number of
4591 processes, without checking whether the existing limit was already
4592 adequate. (It did check for the limit on file descriptors.) Furthermore,
4593 errors from getrlimit() and setrlimit() were being ignored. Now they are
4594 logged to the main and panic logs, but Exim does carry on, to try to do its
4595 job under whatever limits there are.
4597 18. Imported PCRE 5.0.
4599 19. Trivial typo in log message " temporarily refused connection" (the leading
4602 20. If the log selector return_path_on_delivery was set and an address was
4603 redirected to /dev/null, the delivery process crashed because it assumed
4604 that a return path would always be set for a "successful" delivery. In this
4605 case, the whole delivery is bypassed as an optimization, and therefore no
4608 21. Internal re-arrangement: the function for sending a challenge and reading
4609 a response while authentication was assuming a zero-terminated challenge
4610 string. It's now changed to take a pointer and a length, to allow for
4611 binary data in such strings.
4613 22. Added the cyrus_sasl authenticator (code supplied by MBM).
4615 23. Exim was not respecting finduser_retries when seeking the login of the
4616 uid under which it was called; it was always trying 10 times. (The default
4617 setting of finduser_retries is zero.) Also, it was sleeping after the final
4618 failure, which is pointless.
4620 24. Implemented tls_on_connect_ports.
4622 25. Implemented acl_smtp_predata.
4624 26. If the domain in control=submission is set empty, Exim assumes that the
4625 authenticated id is a complete email address when it generates From: or
4626 Sender: header lines.
4628 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
4629 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
4630 chown and chgrp in /bin and hostname in /usr/bin.
4632 28. Exim was keeping the "process log" file open after each use, just as it
4633 does for the main log. This opens the possibility of it remaining open for
4634 long periods when the USR1 signal hits a daemon. Occasional processlog
4635 errors were reported, that could have been caused by this. Anyway, it seems
4636 much more sensible not to leave this file open at all, so that is what now
4639 29. The long-running daemon process does not normally write to the log once it
4640 has entered its main loop, and it closes the log before doing so. This is
4641 so that log files can straightforwardly be renamed and moved. However,
4642 there are a couple of unusual error situations where the daemon does write
4643 log entries, and I had neglected to close the log afterwards.
4645 30. The text of an SMTP error response that was received during a remote
4646 delivery was being truncated at 512 bytes. This is too short for some of
4647 the long messages that one sometimes sees. I've increased the limit to
4650 31. It is now possible to make retry rules that apply only when a message has a
4651 specific sender, in particular, an empty sender.
4653 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
4654 it possible to be selective about when SMTP synchronization is enforced.
4656 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
4658 32. Implemented hosts_connection_nolog.
4660 33. Added an ACL for QUIT.
4662 34. Setting "delay_warning=" to disable warnings was not working; it gave a
4665 35. Added mailbox_size and mailbox_filecount to appendfile.
4667 36. Added control = no_multiline_responses to ACLs.
4669 37. There was a bug in the logic of the code that waits for the clock to tick
4670 in the case where the clock went backwards by a substantial amount such
4671 that the microsecond fraction of "now" was more than the microsecond
4672 fraction of "then" (but the whole seconds number was less).
4674 38. Added support for the libradius Radius client library this is found on
4675 FreeBSD (previously only the radiusclient library was supported).
4681 1. When certain lookups returned multiple values in the form name=value, the
4682 quoting of the values was not always being done properly. Specifically:
4683 (a) If the value started with a double quote, but contained no whitespace,
4685 (b) If the value contained whitespace other than a space character (i.e.
4686 tabs or newlines or carriage returns) it was not quoted.
4687 This fix has been applied to the mysql and pgsql lookups by writing a
4688 separate quoting function and calling it from the lookup code. The fix
4689 should probably also be applied to nisplus, ibase and oracle lookups, but
4690 since I cannot test any of those, I have not disturbed their existing code.
4692 2. A hit in the callout cache for a specific address caused a log line with no
4693 reason for rejecting RCPT. Now it says "Previous (cached) callout
4694 verification failure".
4696 3. There was an off-by-one bug in the queryprogram router. An over-long
4697 return line was truncated at 256 instead of 255 characters, thereby
4698 overflowing its buffer with the terminating zero. As well as fixing this, I
4699 have increased the buffer size to 1024 (and made a note to document this).
4701 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
4702 when Exim is waiting for an SMTP response from a remote server, Exim
4703 restarts its select() call on the socket, thereby resetting its timeout.
4704 This is not a problem when such interrupts are rare. Somebody set up a cron
4705 job to run exiwhat every 2 minutes, which is less than the normal select()
4706 timeout (5 or 10 minutes). This meant that the select() timeout never
4707 kicked in because it was always reset. I have fixed this by comparing the
4708 time when an interrupt arrives with the time at the start of the first call
4709 to select(). If more time than the timeout has elapsed, the interrupt is
4710 treated as a timeout.
4712 5. Some internal re-factoring in preparation for the addition of Sieve
4713 extensions (by MH). In particular, the "personal" test is moved to a
4714 separate function, and given an option for scanning Cc: and Bcc: (which is
4715 not set for Exim filters).
4717 6. When Exim created an email address using the login of the caller as the
4718 local part (e.g. when creating a From: or Sender: header line), it was not
4719 quoting the local part when it contained special characters such as @.
4721 7. Installed new OpenBSD configuration files.
4723 8. Reworded some messages for syntax errors in "and" and "or" conditions to
4724 try to make them clearer.
4726 9. Callout options, other than the timeout value, were being ignored when
4727 verifying sender addresses in header lines. For example, when using
4729 verify = header_sender/callout=no_cache
4731 the cache was (incorrectly) being used.
4733 10. Added a missing instance of ${EXE} to the exim_install script; this affects
4734 only the Cygwin environment.
4736 11. When return_path_on_delivery was set as a log selector, if different remote
4737 addresses in the same message used different return paths and parallel
4738 remote delivery occurred, the wrong values would sometimes be logged.
4739 (Whenever a remote delivery process finished, the return path value from
4740 the most recently started remote delivery process was logged.)
4742 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
4743 lines when AUTH and/or TLS are in use. This is the "received protocol"
4744 field. Exim used to use "asmtp" for authenticated SMTP, without any
4745 indication (in the protocol name) for TLS use. Now it follows the RFC and
4746 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
4747 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
4748 names appear in log lines as well as in Received: header lines.
4750 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
4751 extensions, and comparison tests, and to fix some bugs.
4753 14. Changes to the "personal" filter test:
4755 (1) The test was buggy in that it was just doing the equivalent of
4756 "contains" tests on header lines. For example, if a user's address was
4757 anne@some.where, the "personal" test would incorrectly be true for
4759 To: susanne@some.where
4761 This test is now done by extracting each address from the header in turn,
4762 and checking the entire address. Other tests that are part of "personal"
4763 are now done using regular expressions (for example, to check local parts
4764 of addresses in From: header lines).
4766 (2) The list of non-personal local parts in From: addresses has been
4767 extended to include "listserv", "majordomo", "*-request", and "owner-*",
4768 taken from the Sieve specification recommendations.
4770 (3) If the message contains any header line starting with "List-" it is
4771 treated as non-personal.
4773 (4) The test for "circular" in the Subject: header line has been removed
4774 because it now seems ill-conceived.
4776 15. Minor typos in src/EDITME comments corrected.
4778 16. Installed latest exipick from John Jetmore.
4780 17. If headers_add on a router specified a text string that was too long for
4781 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
4782 of string_sprintf() is now avoided.
4784 18. $message_body_size was not set (it was always zero) when running the DATA
4785 ACL and the local_scan() function.
4787 19. For the "mail" command in an Exim filter, no default was being set for
4788 the once_repeat time, causing a random time value to be used if "once" was
4789 specified. (If the value happened to be <= 0, no repeat happened.) The
4790 default is now 0s, meaning "never repeat". The "vacation" command was OK
4791 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
4792 (I found it when inspecting the code).
4794 20. There is now an overall timeout for performing a callout verification. It
4795 defaults to 4 times the callout timeout, which applies to individual SMTP
4796 commands during the callout. The overall timeout applies when there is more
4797 than one host that can be tried. The timeout is checked before trying the
4798 next host. This prevents very long delays if there are a large number of
4799 hosts and all are timing out (e.g. when the network connections are timing
4800 out). The value of the overall timeout can be changed by specifying an
4801 additional sub-option for "callout", called "maxwait". For example:
4803 verify = sender/callout=5s,maxwait=20s
4805 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
4806 to the end before writing, but this should make it even safer).
4808 22. Exim was forgetting that it had advertised PIPELINING for the second and
4809 subsequent messages on an SMTP connection. It was also not resetting its
4810 memory on STARTTLS and an internal HELO.
4812 23. When Exim logs an SMTP synchronization error within a session, it now
4813 records whether PIPELINING has been advertised or not.
4815 24. Added 3 instances of "(long int)" casts to time_t variables that were being
4816 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
4817 rather than long int.
4819 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
4821 26. Added the never_mail option to autoreply.
4827 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
4828 crash if the getsockname() call failed; this can happen if a connection is
4829 closed very soon after it is established. The problem was simply in the
4830 order in which certain operations were done, causing Exim to try to write
4831 to the SMTP stream before it had set up the file descriptor. The bug has
4832 been fixed by making things happen in the correct order.
4838 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
4839 before the connection was closed, thus losing the rejection response.
4841 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
4842 some early Solaris releases, but causes trouble in current releases where
4843 socklen_t is defined.
4845 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
4848 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
4851 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
4852 that releasing the top of it at the end releases what was used for sub-
4853 expansions (unless the block got too big). However, discard this block if
4854 the first thing is a variable or header, so that we can use its block when
4855 it is dynamic (useful for very large $message_headers, for example).
4857 6. Lookups now cache *every* query, not just the most recent. A new, separate
4858 store pool is used for this. It can be recovered when all lookup caches are
4859 flushed. Lookups now release memory at the end of their result strings.
4860 This has involved some general refactoring of the lookup sources.
4862 7. Some code has been added to the store_xxx() functions to reduce the amount
4863 of flapping under certain conditions.
4865 8. log_incoming_interface used to affect only the <= reception log lines. Now
4866 it causes the local interface and port to be added to several more SMTP log
4867 lines, for example "SMTP connection from", and rejection lines.
4869 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
4871 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
4873 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
4874 could be overwritten at the end of the current message (or the start of a
4875 new message if it was set in a HELO ACL). The value is now preserved for
4876 the duration of the SMTP connection.
4878 12. If a transport had a headers_rewrite setting, and a matching header line
4879 contained an unqualified address, that address was qualified, even if it
4880 did not match any rewriting rules. The underlying bug was that the values
4881 of the flags that permit the existence of unqualified sender and recipient
4882 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
4883 non-local messages, and by -bnq for local messages) were not being
4884 preserved with the message after it was received.
4886 13. When Exim was logging an SMTP synchronization error, it could sometimes log
4887 "next input=" as part of the text comprising the host identity instead of
4888 the correct text. The code was using the same buffer for two different
4889 strings. However, depending on which order the printing function evaluated
4890 its arguments, the bug did not always show up. Under Linux, for example, my
4891 test suite worked just fine.
4893 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
4894 doesn't work with some older versions of Perl. It has been changed to "my",
4895 which in any case is probably the better facility to use.
4897 15. A really picky compiler found some instances of statements for creating
4898 error messages that either had too many or two few arguments for the format
4901 16. The size of the buffer for calls to the DNS resolver has been increased
4902 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
4903 for addresses that have a lot of PTR records. This alleviates a problem; it
4904 does not fully solve it.
4906 17. A dnsdb lookup for PTR records that receives more data than will fit in the
4907 buffer now truncates the list and logs the incident, which is the same
4908 action as happens when Exim is looking up a host name and its aliases.
4909 Previously in this situation something unpredictable would happen;
4910 sometimes it was "internal error: store_reset failed".
4912 18. If a server dropped the connection unexpectedly when an Exim client was
4913 using GnuTLS and trying to read a response, the client delivery process
4914 crashed while trying to generate an error log message.
4916 19. If a "warn" verb in an ACL added multiple headers to a message in a single
4917 string, for example:
4919 warn message = H1: something\nH2: something
4921 the text was added as a single header line from Exim's point of view
4922 though it ended up OK in the delivered message. However, searching for the
4923 second and subsequent header lines using $h_h2: did not work. This has been
4924 fixed. Similarly, if a system filter added multiple headers in this way,
4925 the routers could not see them.
4927 20. Expanded the error message when iplsearch is called with an invalid key to
4928 suggest using net-iplsearch in a host list.
4930 21. When running tests using -bh, any delays imposed by "delay" modifiers in
4931 ACLs are no longer actually imposed (and a message to that effect is
4934 22. If a "gecos" field in a passwd entry contained escaped characters, in
4935 particular, if it contained a \" sequence, Exim got it wrong when building
4936 a From: or a Sender: header from that name. A second bug also caused
4937 incorrect handling when an unquoted " was present following a character
4938 that needed quoting.
4940 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
4941 was not being matched caselessly.
4943 24. Arranged for all hyphens in the exim.8 source to be escaped with
4946 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
4947 properly. Recipient callout cache records were still being keyed to include
4948 the sender, even when use_sender was set false. This led to far more
4949 callouts that were necessary. The sender is no longer included in the key
4950 when use_sender is false.
4952 26. Added "control = submission" modifier to ACLs.
4954 27. Added the ${base62d: operator to decode base 62 numbers.
4956 28. dnsdb lookups can now access SRV records.
4958 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
4959 the configuration file.
4961 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
4962 (-v) mode. This makes the output for a verbose queue run more intelligible.
4964 31. Added a use_postmaster feature to recipient callouts.
4966 32. Added the $body_zerocount variable, containing the number of binary zero
4967 bytes in the message body.
4969 33. The time of last modification of the "new" subdirectory is now used as the
4970 "mailbox time last read" when there is a quota error for a maildir
4973 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
4975 35. Added +ignore_unknown as a special item in host lists.
4977 36. Code for decoding IPv6 addresses in host lists is now included, even if
4978 IPv6 support is not being compiled. This fixes a bug in which an IPv6
4979 address was recognized as an IP address, but was then not correctly decoded
4980 into binary, causing unexpected and incorrect effects when compared with
4987 1. Very minor rewording of debugging text in manualroute to say "list of
4988 hosts" instead of "hostlist".
4990 2. If verify=header_syntax was set, and a header line with an unqualified
4991 address (no domain) and a large number of spaces between the end of the
4992 name and the colon was received, the reception process suffered a buffer
4993 overflow, and (when I tested it) crashed. This was caused by some obsolete
4994 code that should have been removed. The fix is to remove it!
4996 3. When running in the test harness, delay a bit after writing a bounce
4997 message to get a bit more predictability in the log output.
4999 4. Added a call to search_tidyup() just before forking a reception process. In
5000 theory, someone could use a lookup in the expansion of smtp_accept_max_
5001 per_host which, without the tidyup, could leave open a database connection.
5003 5. Added the variables $recipient_data and $sender_data which get set from a
5004 lookup success in an ACL "recipients" or "senders" condition, or a router
5005 "senders" option, similar to $domain_data and $local_part_data.
5007 6. Moved the writing of debug_print from before to after the "senders" test
5010 7. Change 4.31/66 (moving the time when the Received: is generated) caused
5011 problems for message scanning, either using a data ACL, or using
5012 local_scan() because the Received: header was not generated till after they
5013 were called (in order to set the time as the time of reception completion).
5014 I have revised the way this works. The header is now generated after the
5015 body is received, but before the ACL or local_scan() are called. After they
5016 are run, the timestamp in the header is updated.
5022 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
5023 before starting a queue runner without re-exec. This happened only when
5024 deliver_drop_privilege was set or when the Exim user was set to root. The
5025 effect of the bug was that timeouts during subsequent deliveries caused
5026 crashes instead of being properly handled. The handler is now left at its
5027 default (and expected) setting.
5029 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
5030 message, again when deliver_drop_privilege is set or Exim is run as root.
5031 The bug described in (1) was not present in this case, but the tidying up
5032 of the other signals was missing. I have made the two cases consistent.
5034 3. The ignore_target_hosts setting on a manualroute router was being ignored
5035 for hosts that were looked up using the /MX notation.
5037 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
5040 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
5041 operated on the sender address. After changing the $sender_address to <>
5042 for the sender address verify, Exim was re-instated it as the original
5043 (before rewriting) address, but remembering that it had rewritten it, so it
5044 wasn't rewriting it again. This bug also had the effect of breaking the
5045 sender address verification caching when the sender address was rewritten.
5047 6. The ignore_target_hosts option was being ignored by the ipliteral router.
5048 This has been changed so that if the ip literal address matches
5049 ignore_target_hosts, the router declines.
5051 7. Added expansion conditions match_domain, match_address, and match_local_
5052 part (NOT match_host).
5054 8. The placeholder for the Received: header didn't have a length field set.
5056 9. Added code to Exim itself and to exim_lock to test for a specific race
5057 condition that could lead to file corruption when using MBX delivery. The
5058 issue is with the lockfile that is created in /tmp. If this file is removed
5059 after a process has opened it but before that process has acquired a lock,
5060 there is the potential for a second process to recreate the file and also
5061 acquire a lock. This could lead to two Exim processes writing to the file
5062 at the same time. The added code performs the same test as UW imapd; it
5063 checks after acquiring the lock that its file descriptor still refers to
5064 the same named file.
5066 10. The buffer for building added header lines was of fixed size, 8192 bytes.
5067 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
5070 11. Added the smtp_active_hostname option. If used, this will typically be made
5071 to depend on the incoming interface address. Because $interface_address is
5072 not set up until the daemon has forked a reception process, error responses
5073 that can happen earlier (such as "too many connections") no longer contain
5076 12. If an expansion in a condition on a "warn" statement fails because a lookup
5077 defers, the "warn" statement is abandoned, and the next ACL statement is
5078 processed. Previously this caused the whole ACL to be aborted.
5080 13. Added the iplsearch lookup type.
5082 14. Added ident_timeout as a log selector.
5084 15. Added tls_certificate_verified as a log selector.
5086 16. Added a global option tls_require_ciphers (compare the smtp transport
5087 option of the same name). This controls incoming TLS connections.
5089 17. I finally figured out how to make tls_require_ciphers do a similar thing
5090 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
5091 before starting the TLS session.
5093 18. Tabs are now shown as \t in -bP output.
5095 19. If the log selector return_path_on_delivery was set, Exim crashed when
5096 bouncing a message because it had too many Received: header lines.
5098 20. If two routers both had headers_remove settings, and the first one included
5099 a superfluous trailing colon, the final name in the first list and the
5100 first name in the second list were incorrectly joined into one item (with a
5101 colon in the middle).
5107 1. Added -C and -D options to the exinext utility, mainly to make it easier
5108 to include in the automated testing, but these could be helpful when
5109 multiple configurations are in use.
5111 2. The exinext utility was not formatting the output nicely when there was
5112 an alternate port involved in the retry record key, nor when there was a
5113 message id as well (for retries that were specific to a specific message
5114 and a specific host). It was also confused by IPv6 addresses, because of
5115 the additional colons they contain. I have fixed the IPv4 problem, and
5116 patched it up to do a reasonable job for IPv6.
5118 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
5119 delivery, the log line now contains "pipelined" if PIPELINING was used.
5121 4. An SMTP transport process used to panic and die if the bind() call to set
5122 an explicit outgoing interface failed. This has been changed; it is now
5123 treated in the same way as a connect() failure.
5125 5. A reference to $sender_host_name in the part of a conditional expansion
5126 that was being skipped was still causing a DNS lookup. This no longer
5129 6. The def: expansion condition was not recognizing references to header lines
5130 that used bh_ and bheader_.
5132 7. Added the _cache feature to named lists.
5134 8. The code for checking quota_filecount in the appendfile transport was
5135 allowing one more file than it should have been.
5137 9. For compatibility with Sendmail, the command line option
5145 and sets the incoming protocol and host name (for trusted callers). The
5146 host name and its colon can be omitted when only the protocol is to be set.
5147 Note the Exim already has two private options, -pd and -ps, that refer to
5148 embedded Perl. It is therefore impossible to set a protocol value of "d" or
5149 "s", but I don't think that's a major issue.
5151 10. A number of refactoring changes to the code, none of which should affect
5154 (a) The number of logging options was getting close to filling up the
5155 32-bit word that was used as a bit map. I have split them into two classes:
5156 those that are passed in the argument to log_write(), and those that are
5157 only ever tested independently outside of that function. These are now in
5158 separate 32-bit words, so there is plenty of room for expansion again.
5159 There is no change in the user interface or the logging behaviour.
5161 (b) When building, for example, log lines, the code previously used a
5162 macro that called string_cat() twice, in order to add two strings. This is
5163 not really sufficiently general. Furthermore, there was one instance where
5164 it was actually wrong because one of the argument was used twice, and in
5165 one call a function was used. (As it happened, calling the function twice
5166 did not affect the overall behaviour.) The macro has been replaced by a
5167 function that can join an arbitrary number of extra strings onto a growing
5170 (c) The code for expansion conditions now uses a table and a binary chop
5171 instead of a serial search (which was left over from when there were very
5172 few conditions). Also, it now recognizes conditions like "pam" even when
5173 the relevant support is not compiled in: a suitably worded error message is
5174 given if an attempt is made to use such a condition.
5176 11. Added ${time_interval:xxxxx}.
5178 12. A bug was causing one of the ddress fields not to be passed back correctly
5179 from remote delivery subprocesses. The field in question was not being
5180 subsequently used, so this caused to problems in practice.
5182 13. Added new log selectors queue_time and deliver_time.
5184 14. Might have fixed a bug in maildirsizefile handling that threw up
5185 "unexpected character" debug warnings, and recalculated the data
5186 unnecessarily. In any case, I expanded the warning message to give more
5189 15. Added the message "Restricted characters in address" to the statements in
5190 the default ACL that block characters like @ and % in local parts.
5192 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
5193 Three changes have been made:
5195 (a) There was a serious bug; a negative response to MAIL caused the whole
5196 recipient domain to be cached as invalid, thereby blocking all messages
5197 to all local parts at the same domain, from all senders. This bug has
5198 been fixed. The domain is no longer cached after a negative response to
5199 MAIL if the sender used is not empty.
5201 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
5204 (c) A new callout option, "use_sender" has been added for people who want
5205 the modified behaviour.
5211 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
5214 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
5215 indeed breaks things for older releases.
5217 3. Added additional logging to the case where there is a problem reading data
5218 from a filter that is running in a subprocess using a pipe, in order to
5219 try to track down a specific problem.
5221 4. Testing facility fudge: when running in the test harness and attempting
5222 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
5223 getting "No route to host". Convert this to a timeout.
5225 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
5228 6. Some OS don't have socklen_t but use size_t instead. This affects the
5229 fifth argument of getsockopt() amongst other things. This is now
5230 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
5231 can be set for individual OS. I have set it for SunOS5, OSF1, and
5232 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
5233 some earlier ones do not.
5235 7. Change 4.30/15 was not doing the test caselessly.
5237 8. The standard form for an IPv6 address literal was being rejected by address
5238 parsing in, for example, MAIL and RCPT commands. An example of this kind of
5239 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5240 this, as well as the form without the "IPv6" on the front (but only when
5241 address literals are enabled, of course).
5243 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
5245 10. Exim crashed if a message with an empty sender address specified by -f
5246 encountered a router with an errors_to setting. This could be provoked only
5247 by a command such as
5251 where an empty string was supplied; "<>" did not hit this bug.
5253 11. Installed PCRE release 4.5.
5255 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
5256 remained set. It is now erased.
5258 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
5259 times from message ids (which are base 36 rather than the normal 62).
5261 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
5262 were being counted as actual protocol errors, and logged if the log
5263 selector +smtp_protocol_error was set. One cannot be perfect in this test,
5264 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
5265 and DATA following a set of rejected RCPTs do not count as protocol errors.
5266 In other words, Exim assumes they were pipelined, though this may not
5267 actually be the case. Of course, in all cases the client gets an
5268 appropriate error code.
5270 15. If a lookup fails in an ACL condition, a message about the failure may
5271 be available; it is used if testing the ACL cannot continue, because most
5272 such messages specify what the cause of the deferral is. However, some
5273 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
5274 that caused an old message to be retained and used if a later statement
5275 caused a defer, replacing the real cause of the deferral.
5277 16. If an IP address had so many PTR records that the DNS lookup buffer
5278 was not large enough to hold them, Exim could crash while trying to process
5279 the truncated data. It now detects and logs this case.
5281 17. Further to 4.21/58, another change has been made: if (and only if) the
5282 first line of a message (the first header line) ends with CRLF, a bare LF
5283 in a subsequent header line has a space inserted after it, so as not to
5284 terminate the header.
5286 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
5287 unnecessarily, used atoi() instead of strtol(), and didn't check the
5288 termination when getting file sizes from file names by regex.
5290 19. Completely re-implemented the support for maildirsize files, in the light
5291 of a number of problems with the previous contributed implementation
5292 (4.30/29). In particular:
5294 . If the quota is zero, the maildirsize file is maintained, but no quota is
5297 . If the maildir directory does not exist, it is created before any attempt
5298 to write a maildirsize file.
5300 . The quota value in the file is just a cache; if the quota is changed in
5301 the transport, the new value overrides.
5303 . A regular expression is available for excluding directories from the
5306 20. The autoreply transport checks the characters in options that define the
5307 message's headers; it allows continued headers, but it was checking with
5308 isspace() after an embedded newline instead of explicitly looking for a
5311 21. If all the "regular" hosts to which an address was routed had passed their
5312 expiry times, and had not reached their retry times, the address was
5313 bounced, even if fallback hosts were defined. Now Exim should go on to try
5316 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
5317 equivalent code in the SMTP transport. Some hosts send humungous responses
5318 to HELO/EHLO, more than 1024 it seems.
5320 23. Refactoring: code in filter.c used (void *) for "any old type" but this
5321 gives compiler warnings in some environments. I've now done it "properly",
5324 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
5325 (because of problems with the built-in one) was declared to return uschar *
5326 instead of char *, causing compiler failure.
5328 25. Fixed a file descriptor leak when processing alias/forward files.
5330 26. Fixed a minor format string issue in dbfn.c.
5332 27. Typo in exim.c: ("dmbnz" for "dbmnz").
5334 28. If a filter file refered to $h_xxx or $message_headers, and the headers
5335 contained RFC 2047 "words", Exim's memory could, under certain conditions,
5338 29. When a sender address is verified, it is cached, to save repeating the test
5339 when there is more than one recipient in a message. However, when the
5340 verification involves a callout, it is possible for different callout
5341 options to be set for different recipients. It is too complicated to keep
5342 track of this in the cache, so now Exim always runs a verification when a
5343 callout is required, relying on the callout cache for the optimization.
5344 The overhead is duplication of the address routing, but this should not be
5347 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
5348 to be verified with callout=postmaster, and the main callout worked but the
5349 postmaster check failed, the verification correctly failed. However, if a
5350 subsequent RCPT command asked for sender verification *without* the
5351 postmaster check, incorrect caching caused this verification also to fail,
5354 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
5355 it was not caching the DNS options (qualify_single, search_parents) that
5356 were used when the lookup failed. A subsequent lookup with different
5357 options therefore always gave the same answer, though there were cases
5358 where it should not have. (Example: a "domains = !$mx_any" option on a
5359 dnslookup router: the "domains" option is always processed without any
5360 widening, but the router might have qualify_single set.) Now Exim uses the
5361 cached value only when the same options are set.
5363 32. Added John Jetmore's "exipick" utility to the distribution.
5365 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
5366 than a timeout (e.g. a certificate is required, and is not provided), an
5367 Exim server now closes the connection immediately. Previously it waited for
5368 the client to close - but if the client is SSL, it seems that they each
5369 wait for each other, leading to a delay before one of them times out.
5371 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
5372 maintained 0.8.x compatibility because I don't think many are using it, and
5373 it is clearly obsolete.
5375 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
5378 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
5379 client certificate was expired. A simple patch fixes this, though I don't
5380 understand the full logic of why the verify callback is called multiple
5383 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
5384 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
5385 which causes problems with some clients (such as the Certicom SSL Plus
5386 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
5387 disables the coutermeasure allowing Eudora to connect."
5389 38. Exim was not checking that a write() to a log file succeeded. This could
5390 lead to Bad Things if a log got too big, in particular if it hit a file
5391 size limit. Exim now panics and dies if it cannot write to a log file, just
5392 as it does if it cannot open a log file.
5394 39. Modified OS/Makefile-Linux so that it now contains
5396 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
5398 The two -D definitions ensure that Exim is compiled with large file
5399 support, which makes it possible to handle log files that are bigger than
5402 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
5403 instance) a domain was checked against a named list that involved a lookup,
5404 causing $domain_data to be set, then another domain was checked against the
5405 same list, then the first domain was re-checked, the value of $domain_data
5406 after the final check could be wrong. In particular, if the second check
5407 failed, it could be set empty. This bug probably also applied to
5410 41. The strip_trailing_dot option was not being applied to the address given
5411 with the -f command-line option.
5413 42. The code for reading a message's header from the spool was incrementing
5414 $received_count, but never initializing it. This meant that the value was
5415 incorrect (doubled) while delivering a message in the same process in which
5416 it was received. In the most common configuration of Exim, this never
5417 happens - a fresh exec is done - but it can happen when
5418 deliver_drop_privilege is set.
5420 43. When Exim logs an SMTP synchronization error - client data sent too soon -
5421 it now includes up to 150 characters of the unexpected data in the log
5424 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
5425 and building data strings. The size of both of these buffers was 10 000
5426 bytes - far larger than anybody would *ever* want, thought I. Needless to
5427 say, somebody hit the limit. I have increased the maximum line length to
5428 20 000 and the maximum data length of concatenated lines to 100 000. I have
5429 also fixed two bugs, because there was no checking on these buffers. Tsk,
5430 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
5431 buffer is too small.
5433 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
5434 lsearch lookups. Now it does.
5436 46. When parsing a route_list item in a manualroute router, a fixed-length
5437 buffer was used for the list of hosts. I made this 1024 bytes long,
5438 thinking that nobody would ever have a list of hosts that long. Wrong.
5439 Somebody had a whole pile of complicated expansion conditions, and the
5440 string was silently truncated, leading to an expansion error. It turns out
5441 that it is easier to change to an unlimited length (owing to other changes
5442 that have happened since this code was originally written) than to build
5443 structure for giving a limitation error. The length of the item that
5444 expands into the list of hosts is now unlimited.
5446 47. The lsearch lookup could not handle data where the length of text line was
5447 more than 4095 characters. Such lines were truncated, leading to shortened
5448 data being returned. It should now handle lines of any length.
5450 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
5451 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
5454 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
5455 replacing strings such as BIN_DIRECTORY in a source file: the replacement
5456 no longer happens in comment lines. A list of replacements is now placed
5457 at the head of all of the source files, except those whose only change is
5458 to replace PERL_COMMAND in the very first #! line.
5460 50. Replaced the slow insertion sort in queue.c, for sorting the list of
5461 messages on the queue, with a bottom-up merge sort, using code contributed
5462 by Michael Haardt. This should make operations like -bp somewhat faster on
5463 large queues. It won't affect queue runners, except when queue_run_in_order
5466 51. Installed eximstats 1.31 in the distribution.
5468 52. Added support for SRV lookups to the dnslookup router.
5470 53. If an ACL referred to $message_body or $message_body_end, the value was not
5471 reset for any messages that followed in the same SMTP session.
5473 54. The store-handling optimization for building very long strings was not
5474 differentiating between the different store pools. I don't think this
5475 actually made any difference in practice, but I've tidied it.
5477 55. While running the routers to verify a sender address, $sender_address
5478 was still set to the sender address. This is wrong, because when routing to
5479 send a bounce to the sender, it would be empty. Therefore, I have changed
5480 it so that, while verifying a sender address, $sender_address is set to <>.
5481 (There is no change to what happens when verifying a recipient address.)
5483 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
5484 target A or AAAA records (if not already returned) without resetting the
5485 qualify_single or search_parents options of the DNS resolver. These are
5486 inappropriate in this case because the targets of MX and SRV records must
5487 be FQDNs. A broken DNS record could cause trouble if it happened to have a
5488 target that, when qualified, matched something in the local domain. These
5489 two options are now turned off when doing these lookups.
5491 57. It seems that at least some releases of Reiserfs (which does not have the
5492 concept of a fixed number of inodes) returns zero and not -1 for the
5493 number of available inodes. This interacted badly with check_spool_inodes,
5494 which assumed that -1 was the "no such thing" setting. What I have done is
5495 to check that the total number of inodes is greater than zero before doing
5496 the test of how many are available.
5498 58. When a "warn" ACL statement has a log_message modifier, the message is
5499 remembered, and not repeated. This is to avoid a lot of repetition when a
5500 message has many recipients that cause the same warning to be written.
5501 Howewer, Exim was preserving the list of already written lines for an
5502 entire SMTP session, which doesn't seem right. The memory is now reset if a
5503 new message is started.
5505 59. The "rewrite" debugging flag was not showing the result of rewriting in the
5506 debugging output unless log_rewrite was also set.
5508 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
5509 of (int)(handle) when we know that handle contains (void *)(-1).
5511 61. The Exim daemon panic-logs an error return when it closes the incoming
5512 connection. However "connection reset by peer" seems to be common, and
5513 isn't really an error worthy of noting specially, so that particular error
5516 62. When Exim is trying to find all the local interfaces, it used to panic and
5517 die if the ioctl to get the interface flags failed. However, it seems that
5518 on at least one OS (Solaris 9) it is possible to have an interface that is
5519 included in the list of interfaces, but for which you get a failure error
5520 for this call. This happens when the interface is not "plumbed" into a
5521 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
5522 failure of the "get flags" call assumes that the interface is down.
5524 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
5525 makes life easier for people who are doing arithmetic on fields extracted
5526 from dates, where you often get leading zeros that should not be
5527 interpreted as octal.
5529 64. Added qualify_domain to the redirect router, to override the global
5532 65. If a pathologically long header line contained very many addresses (the
5533 report of this problem mentioned 10 000) and each of them was rewritten,
5534 Exim could use up a very large amount of memory. (It kept on making new
5535 copies of the header line as it rewrote, and never released the old ones.)
5536 At the expense of a bit more processing, the header rewriting function has
5537 been changed so that it no longer eats memory in this way.
5539 66. The generation of the Received: header has been moved from the time that a
5540 message starts to be received, to the time that it finishes. The timestamp
5541 in the Received: header should now be very close to that of the <= log
5542 line. There are two side-effects of this change:
5544 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
5545 logged header lines no longer include the local Received: line, because
5546 it has not yet been created. The same applies to a copy of the message
5547 that is returned to a non-SMTP sender when a message is rejected.
5549 (b) When a filter file is tested using -bf, no additional Received: header
5550 is added to the test message. After some thought, I decided that this
5553 This change does not affect the value of $received_for. It is still set
5554 after address rewriting, but before local_scan() is called.
5556 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
5558 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
5559 gave an unhelpful panic error message, and a defer error. I have managed to
5560 change this behaviour so that it now rejects any supplied certificate,
5561 which seems right, as the list of acceptable certificates is empty.
5563 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
5564 gave an unhelpful defer error. I have not managed to make this reject any
5565 supplied certificates, but the error message it gives is "no certificate
5566 supplied", which is not helpful.
5568 70. exigrep's output now also includes lines that are not associated with any
5569 message, but which match the given pattern. Implemented by a patch from
5570 Martin Sluka, which also tidied up the Perl a bit.
5572 71. Recipient callout verification, like sender verification, was using <> in
5573 the MAIL FROM command. This isn't really the right thing, since the actual
5574 sender may affect whether the remote host accepts the recipient or not. I
5575 have changed it to use the actual sender in the callout; this means that
5576 the cache record is now keyed on a recipient/sender pair, not just the
5577 recipient address. There doesn't seem to be a real danger of callout loops,
5578 since a callout by the remote host to check the sender would use <>.
5579 [SEE ABOVE: changed after hitting problems.]
5581 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
5582 temporary errors. However, in the case of such a code being given after
5583 the end of a data transmission (i.e. after ".") Exim was failing to write
5584 a retry record for the message. (Yes, there was some broken host that was
5585 actually sending 8xx at this point.)
5587 73. An unknown lookup type in a host list could cause Exim to panic-die when
5588 the list was checked. (An example that provoked this was putting <; in the
5589 middle of a list instead of at the start.) If this happened during a DATA
5590 ACL check, a -D file could be left lying around. This kind of configuration
5591 error no longer causes Exim to die; instead it causes a defer errror. The
5592 incident is still logged to the main and panic logs.
5594 74. Buglet left over from Exim 3 conversion. The message "too many messages
5595 in one connection" was written to the rejectlog but not the mainlog, except
5596 when address rewriting (yes!) was being logged.
5598 75. Added write_rejectlog option.
5600 76. When a system filter was run not as root (that is, when system_filter_user
5601 was set), the values of the $n variables were not being returned to the
5602 main process; thus, they were not subsequently available in the $sn
5605 77. Added +return_path_on_delivery log selector.
5607 78. A connection timeout was being treated differently from recipients deferred
5608 when testing hosts_max_try with a message that was older than the host's
5609 retry timeout. (The host should not be counted, thus allowing all hosts to
5610 be tried at least once before bouncing.) This may have been the cause of an
5611 occasionally reported bug whereby a message would remain on the queue
5612 longer than the retry timeout, but would be bounced if a delivery was
5613 forced. I say "may" because I never totally pinned down the problem;
5614 setting up timeout/retry tests is difficult. See also the next item.
5616 79. The ultimate address timeout was not being applied to errors that involved
5617 a combination of host plus message (for example, a timeout on a MAIL
5618 command). When an address resolved to a number of possible hosts, and they
5619 were not all tried for each delivery (e.g. because of hosts_max_try), a
5620 message could remain on the queue longer than the retry timeout.
5622 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
5625 81. Fixed an obscure SMTP outgoing bug which required at least the following
5626 conditions: (a) there was another message waiting for the same server;
5627 (b) the server returned 5xx to all RCPT commands in the first message so
5628 that the message was not completed; (c) the server dropped the connection
5629 or gave a negative response to the RSET that Exim sends to abort the
5630 transaction. The observed case was a dropped connection after DATA that had
5631 been sent in pipelining mode. That is, the server had advertised PIPELINING
5632 but was not implementing it correctly. The effect of the bug was incorrect
5633 behaviour, such as trying another host, and this could lead to a crash.
5639 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
5640 and daemon.c were passed as pointers to ints; they should have been
5641 pointers to socklen_t variables (which are typically unsigned ints).
5643 2. Some signed/unsigned type warnings in the os.c file for Linux have been
5646 3. Fixed a really odd bug that affected only the testing scheme; patching a
5647 certain fixed string in the binary changed the value of another string that
5648 happened to be identical to the end of the original first string.
5650 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
5651 name", it returns that address as the IP address. On some operating
5652 systems (e.g. Solaris), it also passes back the IP address string as the
5653 "host name". However, on others (e.g. Linux), it passes back an empty
5654 string. Exim wasn't checking for this, and was changing the host name to an
5655 empty string, assuming it had been canonicized.
5657 5. Although rare, it is permitted to have more than one PTR record for a given
5658 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
5659 all the names associated with an address, because they do in Solaris.
5660 However, it seems that they do not in Linux for data that comes from the
5661 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
5662 I found this out when I moved to a new Linux workstation and tried to run
5663 the Exim test suite.
5665 To get round this problem I have changed the code so that it now does its
5666 own call to the DNS to look up PTR records when searching for a host name.
5667 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
5668 addresses that are only in /etc/hosts are still found.
5670 This behaviour is, however, controlled by an option called host_lookup_
5671 order, which defaults to "bydns:byaddr". If people want to use the other
5672 order, or indeed, just use one or the other means of lookup, they can
5673 specify it in this variable.
5675 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
5676 some operating systems, this comes back from gethostbyaddr() as an empty
5677 string, and this is what Exim used to test for. However, it seems that in
5678 other systems, "." is yielded. Exim now tests for this case too.
5680 7. The values of check_spool_space and check_log_space are now held internally
5681 as a number of kilobytes instead of an absolute number of bytes. If a
5682 numbers is specified without 'K' or 'M', it is rounded up to the nearest
5683 kilobyte. This means that much larger values can be stored.
5685 8. Exim monitor: an attempt to get the action menu when not actually pointing
5686 at a message produces an empty menu entitled "No message selected". This
5687 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
5688 no entries in it ("Shell widget menu has zero width and/or height"). So I
5689 have added a single, blank menu entry in this case.
5691 9. Added ${quote_local_part.
5693 10. MIME decoding is now applied to the contents of Subject: header lines when
5696 11. Now that a reference to $sender_host_address automatically causes a reverse
5697 lookup to occur if necessary (4.13/18), there is no need to arrange for a
5698 host lookup before query-style lookups in lists that might use this
5699 variable. This has therefore been abolished, and the "net-" prefix is no
5700 longer necessary for query-style lookups.
5702 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
5703 have been a typo for LFLAGS, so it has been changed.
5705 13. The install script calls Exim with "-C /dev/null" in order to find the
5706 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
5707 to be output. Howeve, since Exim outputs its version number before the
5708 error, it didn't break the script. It just looked ugly. I fixed this by
5709 always allowing "-C /dev/null" if the caller is root.
5711 14. Ignore overlarge ACL variable number when reading spool file - insurance
5712 against a later release with more variables having written the file.
5714 15. The standard form for an IPv6 address literal was being rejected by EHLO.
5715 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5716 this, as well as the form without the "IPv6" on the front.
5718 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
5719 OS/Makefile-Darwin file.
5721 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
5722 only to LDAP libraries that do not have LDAP_OPT_DEREF.
5724 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
5725 variable. However, some OS use ints for inodes. I've added cast to long int
5726 to get rid of the compiler warning.
5728 19. I had forgotten to lock out "/../" in configuration file names when
5729 ALT_CONFIG_PREFIX was set.
5731 20. Routers used for verification do not need to specify transports. However,
5732 if such a router generated a host list, and callout was configured, Exim
5733 crashed, because it could not find a port number from the (non-existent)
5734 transport. It now assumes port 25 in this circumstance.
5736 21. Added the -t option to exigrep.
5738 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
5739 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
5740 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
5741 them is set, LOOKUP_LSEARCH is forced.
5743 23. "exim -bV" now outputs a list of lookups that are included in the binary.
5745 24. Added sender and host information to the "rejected by local_scan()" log
5746 line; previously there was no indication of these.
5748 25. Added .include_if_exists.
5750 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
5751 Linux. It turns out that not all file systems support this. Apparently some
5752 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
5753 do it.) To cope with this, the error EINVAL, which means that sync-ing is
5754 not supported on the file descriptor, is now ignored when Exim is trying to
5755 sync a directory. This applies only to Linux.
5757 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
5759 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
5760 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
5763 29. Added support for maildirsize files from supplied patch (modified a bit).
5765 30. The use of :fail: followed by an empty string could lead Exim to respond to
5766 sender verification failures with (e.g.):
5768 550 Verification failed for <xxx>
5769 550 Sender verify failed
5771 where the first response line was missing the '-' that indicates it is not
5772 the final line of the response.
5774 31. The loop for finding the name of the user that called Exim had a hardwired
5775 limit of 10; it now uses the value of finduser_retries, which is used for
5776 all other user lookups.
5778 32. Added $received_count variable, available in data and not_smtp ACLs, and at
5781 33. Exim was neglecting to zero errno before one call of strtol() when
5782 expanding a string and expecting an integer value. On some systems this
5783 resulted in spurious "integer overflow" errors. Also, it was casting the
5784 result into an int without checking.
5786 34. Testing for a connection timeout using "timeout_connect" in the retry rules
5787 did not work. The code looks as if it has *never* worked, though it appears
5788 to have been documented since at least releast 1.62. I have made it work.
5790 35. The "timeout_DNS" error in retry rules, also documented since at least
5791 1.62, also never worked. As it isn't clear exactly what this means, and
5792 clearly it isn't a major issue, I have abolished the feature by treating it
5793 as "timeout", and writing a warning to the main and panic logs.
5795 36. The display of retry rules for -brt wasn't always showing the error code
5798 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
5799 timeout_connect_A, timeout_connect_MX.
5801 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
5802 to the empty sender.
5804 39. The daemon was not analysing the content of -oX till after it had closed
5805 stderr and disconnected from the controlling terminal. This meant that any
5806 syntax errors were only noted on the panic log, and the return code from
5807 the command was 0. By re-arranging the code a little, I've made the
5808 decoding happen first, so such errors now appear on stderr, and the return
5809 code is 1. However, the actual setting up of the sockets still happens in
5810 the disconnected process, so errors there are still only recorded on the
5813 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
5814 connections (as happens on some IP stacks) was logged at start up time as
5815 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
5816 it from "IPv6 and IPv4", which means that two separate sockets are being
5819 41. The debug output for gethostbyname2() or getipnodebyname() failures now
5820 says whether AF_INET or AF_INET6 was passed as an argument.
5822 42. Exiwhat output was messed up when time zones were included in log
5825 43. Exiwhat now gives more information about the daemon's listening ports,
5826 and whether -tls-on-connect was used.
5828 44. The "port" option of the smtp transport is now expanded.
5830 45. A "message" modifier in a "warn" statement in a non-message ACL was being
5831 silently ignored. Now an error message is written to the main and panic
5834 46. There's a new ACL modifier called "logwrite" which writes to a log file
5835 as soon as it is encountered.
5837 47. Added $local_user_uid and $local_user_gid at routing time.
5839 48. Exim crashed when trying to verify a sender address that was being
5842 49. Exim was recognizing only a space character after ".include". It now also
5843 recognizes a tab character.
5845 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
5846 extracting the relevant information from the specification. The man page no
5847 longer contains scrambled data for the -d option, and I've added a section
5848 at the front about calling Exim under different names.
5850 51. Added "extra_headers" argument to the "mail" command in filter files.
5852 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
5855 53. Installed eximstats 1.29.
5857 54. Added transport_filter_timeout as a generic transport option.
5859 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
5860 Cc: header lines. This was required by RFC 822, but it not required by RFC
5863 56. Exim used to add From:, Date:, and Message-Id: header lines to any
5864 incoming messages that did not have them. Now it does so only if the
5865 message originates locally, that is, if there is no associated remote host
5866 address. When Resent- header lines are present, this applies to the Resent-
5867 lines rather than the non-Resent- lines.
5869 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
5870 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
5872 58. Messages for configuration errors now include the name of the main
5873 configuration file - useful now that there may be more than one file in a
5874 list (.included file names were always shown).
5876 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
5877 for those rare installations that do not start the daemon as root or run it
5878 setuid root. I've cut out the call to initgroups() if the daemon is not
5881 60. The Exim user and group can now be bound into the binary as text strings
5882 that are looked up at the start of Exim's processing.
5884 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
5886 62. Added $mailstore_basename variable.
5888 63. Installed patch to sieve.c from Michael Haardt.
5890 64. When Exim failed to open the panic log after failing to open the main log,
5891 the original message it was trying to log was written to stderr and debug
5892 output, but if they were not available (the usual case in production), it
5893 was lost. Now it is written to syslog before the two lines that record the
5894 failures to open the logs.
5896 65. Users' Exim filters run in subprocesses under the user's uid. It is
5897 possible for a "deliver" command or an alias in a "personal" command to
5898 provoke an address rewrite. If logging of address rewriting is configured,
5899 this fails because the process is not running as root or exim. There may be
5900 a better way of dealing with this, but for the moment (because 4.30 needs
5901 to be released), I have disabled address rewrite logging when running a
5902 filter in a non-root, non-exim process.
5908 1. The buildconfig auxiliary program wasn't quoting the value set for
5909 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
5910 not defined. This bug was masked in 4.22 by the effect that was fixed in
5913 2. Some messages that were rejected after a message id was allocated were
5914 shown as "incomplete" by exigrep. It no longer does this for messages that
5915 are rejected by local_scan() or the DATA or non-SMTP ACLs.
5917 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
5918 have allow_domain_literals set, the ID did not get logged in the <= line.
5919 Domain literals are now always recognized in Message-ID: header lines.
5921 4. The first argument for a ${extract expansion item is the key name or field
5922 number. Leading and trailing spaces in this item were not being ignored,
5923 causing some misleading effects.
5925 5. When deliver_drop_privilege was set, single queue runner processes started
5926 manually (i.e. by the command "exim -q") or by the daemon (which uses the
5927 same command in the process it spins off) were not dropping privilege.
5929 6. When the daemon running as "exim" started a queue runner, it always
5930 re-executed Exim in the spun-off process. This is a waste of effort when
5931 deliver_drop_privilege is set. The new process now just calls the
5932 queue-runner function directly.
5938 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
5941 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
5942 ignored. Though the use of -oP was forcing the writing of a pid file, it
5943 was always written to the default place.
5945 3. If the message "no IP address found for host xxxx" is generated during
5946 incoming verification, it is now followed by identification of the incoming
5947 connection (so you can more easily find what provoked it).
5949 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
5951 5. Added some features to "harden" Exim a bit more against certain attacks:
5953 (a) There is now a build-time option called FIXED_NEVER_USERS that can
5954 be put in Local/Makefile. This is like the never_users runtime option,
5955 but it cannot be overridden. The default setting is "root".
5957 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
5958 prefix string with which any file named in a -C command line option
5961 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
5962 is retained for -C and -D only if the caller of Exim is root. Without
5963 it, the exim user may also use -C and -D and retain privilege.
5965 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
5966 command line option is disabled.
5968 6. Macro names set by the -D option must start with an upper case letter, just
5969 like macro names defined in the configuration file.
5971 7. Added "dereference=" facility to LDAP.
5973 8. Two instances of the typo "uknown" in the source files are fixed.
5975 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
5976 the Configure-Makefile script screwed up while processing it.
5978 10. Incorporated PCRE 4.4.
5980 11. The SMTP synchronization check was not operating right at the start of an
5981 SMTP session. For example, it could not catch a HELO sent before the client
5982 waited for the greeting. There is now a check for outstanding input at the
5983 point when the greeting is written. Because of the duplex, asynchronous
5984 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
5985 way, but not yet received, when the check is performed.
5987 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
5988 on TCP/IP sockets, because this apparently causes some broken clients to
5991 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
5992 unchanged) from the Cygwin maintainer.
5994 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
5995 was supported instead of testing for mbx. Effectively a typo.
5997 15. The spa authenticator server code was not checking that the input it
5998 received was valid base64.
6000 16. The debug output line for the "set" modifier in ACLs was not showing the
6001 name of the variable that was being set.
6003 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
6005 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
6006 lookup on its own. Something else was needed to trigger the lookup. For
6007 example, a match in host_lookup or the need for a host name in a host list.
6008 Now, if $sender_host_name is referenced and the host name has not yet been
6009 looked up, a lookup is performed. If the lookup fails, the variable remains
6010 empty, and $host_lookup_failed is set to "1".
6012 19. Added "eqi" as a case-independent comparison operator.
6014 20. The saslauthd authentication condition could segfault if neither service
6015 nor realm was specified.
6017 21. If an overflowing value such as "2048M" was set for message_size_limit, the
6018 error message that was logged was misleading, and incoming SMTP
6019 connections were dropped. The message is now more accurate, and temporary
6020 errors are given to SMTP connections.
6022 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
6023 (except RSET) with a 421 error, until QUIT is received. However, it was
6024 failing to send a response to QUIT.
6026 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
6027 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
6028 ACL is now run after the helo_try_verify_hosts code.
6030 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
6031 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
6032 case-independent, so other case variants are also recognized). Apparently
6033 some systems use these upper case variants.
6035 25. If more than two messages were waiting for the same host, and a transport
6036 filter was specified for the transport, Exim sent two messages over the
6037 same TCP/IP connection, and then failed with "socket operation on non-
6038 socket" when it tried to send the third.
6040 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
6042 27. The extern definition of crypt16() in expand.c was not being excluded when
6043 the OS had its own crypt16() function.
6045 28. Added bounce_return_body as a new option, and bounce_return_size_limit
6046 as a preferred synonym for return_size_limit, both as an option and as an
6049 29. Added LIBS=-liconv to OS/Makefile-OSF1.
6051 30. Changed the default configuration ACL to relax the local part checking rule
6052 for addresses that are not in any local domains. For these addresses,
6053 slashes and pipe symbols are allowed within local parts, but the sequence
6054 /../ is explicitly forbidden.
6056 31. SPA server authentication was not clearing the challenge buffer before
6059 32. log_message in a "warn" ACL statement was writing to the reject log as
6060 well as to the main log, which contradicts the documentation and doesn't
6061 seem right (because no rejection is happening). So I have stopped it.
6063 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
6064 However, I am unable to do any testing of this.
6066 34. Fixed an infelicity in the appendfile transport. When checking directories
6067 for a mailbox, to see if any needed to be created, it was accidentally
6068 using path names with one or more superfluous leading slashes; tracing
6069 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
6071 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
6072 contents are added to the log line that is written for every discarded
6073 recipient. (Previously a log_message setting was ignored.)
6075 36. The ${quote: operator now quotes the string if it is empty.
6077 37. The install script runs exim in order to find its version number. If for
6078 some reason other than non-existence or emptiness, which it checks, it
6079 could not run './exim', it was installing it with an empty version number,
6080 i.e. as "exim-". This error state is now caught, and the installation is
6083 38. An argument was missing from the function that creates an error message
6084 when Exim fails to connect to the socket for saslauthd authentication.
6085 This could cause Exim to crash, or give a corrupted message.
6087 39. Added isip, isip4, and isip6 to ${if conditions.
6089 40. The ACL variables $acl_xx are now saved with the message, and can be
6090 accessed later in routers, transports, and filters.
6092 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
6093 strings in the file are not string-expanded.
6095 42. If a MAIL command specified a SIZE value that was too large to fit into an
6096 int variable, the check against message_size_limit failed. Such values are
6097 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
6098 day this will have to be increased, but I don't think I want to be around
6099 when emails are that large.
6106 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
6107 iconv() is not standard in FreeBSD.
6109 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
6110 IPv6 enabled. The observed symptom was a segmentation fault on return from
6111 the function os_common_find_running_interfaces() in src/os.c.
6113 3. In the check_special_case() function in daemon.c I had used "errno" as an
6114 argument name, which causes warnings on some systems. This was basically a
6115 typo, since it was named "eno" in the comments!
6117 4. The code that waits for the clock to tick (at a resolution of some fraction
6118 of a second) so as to ensure message-id uniqueness was always waiting for
6119 at least one whole tick, when it could have waited for less. [This is
6120 almost certainly not relevant at current processor speeds, where it is
6121 unlikely to ever wait at all. But we try to future-proof.]
6123 5. The function that sleeps for a time interval that includes fractions of a
6124 second contained a race. It did not block SIGALRM between setting the
6125 timer, and suspending (a couple of lines later). If the interval was short
6126 and the sigsuspend() was delayed until after it had expired, the suspension
6127 never ended. On busy systems this could lead to processes getting stuck for
6130 6. Some uncommon configurations may cause a lookup to happen in a queue runner
6131 process, before it forks any delivery processes. The open lookup caching
6132 mechanism meant that the open file or database connection was passed into
6133 the delivery process. The problem was that delivery processes always tidy
6134 up cached lookup data. This could cause a problem for the next delivery
6135 process started by the queue runner, because the external queue runner
6136 process does not know about the closure. So the next delivery process
6137 still has data in the lookup cache. In the case of a file lookup, there was
6138 no problem because closing a file descriptor in a subprocess doesn't affect
6139 the parent. However, if the lookup was caching a connection to a database,
6140 the connection was closed, and the second delivery process was likely to
6141 see errors such as "PGSQL: query failed: server closed the connection
6142 unexpectedly". The problem has been fixed by closing all cached lookups
6143 in a queue runner before running a delivery process.
6145 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
6146 seem to have the "const" qualifier which it has on other OS. I've
6149 8. Change 4.21/2 was too strict. It is only if there are two authenticators
6150 *of the same type* (client or server) with the same public name that an
6151 error should be diagnosed.
6153 9. When Exim looked up a host name for an IP address, but failed to find the
6154 original IP address when looking up the host name (a safety check), it
6155 output the message "<ip address> does not match any IP for NULL", which was
6156 confusing, to say the least. The bug was that the host name should have
6157 appeared instead of "NULL".
6159 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
6160 user that is built into the binary, and the -C or -D options is used, root
6161 privilege is dropped before the configuration file is read. In addition,
6162 logging is switched to stderr instead of the normal log files. If the
6163 configuration then re-defines the Exim user, the unprivileged environment
6164 is probably not what is expected, so Exim logs a panic warning message (but
6167 However, if deliver_drop_privilege is set, the unprivileged state may well
6168 be exactly what is intended, so the warning has been cut out in that case,
6169 and Exim is allowed to try to write to its normal log files.
6175 1. smtp_return_error_details was not giving details for temporary sender
6176 or receiver verification errors.
6178 2. Diagnose a configuration error if two authenticators have the same public
6181 3. Exim used not to create the message log file for a message until the first
6182 delivery attempt. This could be confusing when incoming messages were held
6183 for policy or load reasons. The message log file is now created at the time
6184 the message is received, and an initial "Received" line is written to it.
6186 4. The automatically generated man page for command line options had a minor
6187 bug that caused no ill effects; however, a more serious problem was that
6188 the procedure for building the man page automatically didn't always
6189 operate. Consequently, release 4.20 contains an out-of-date version. This
6190 shouldn't happen again.
6192 5. When building Exim with embedded Perl support, the script that builds the
6193 Makefile was calling 'perl' to find its compile-time parameters, ignoring
6194 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
6196 6. The freeze_tell option was not being used for messages that were frozen on
6197 arrival, either by an ACL or by local_scan().
6199 7. Added the smtp_incomplete_transaction log selector.
6201 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
6202 was accepting AUTH without a new EHLO.
6204 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
6205 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
6208 10. Logging of TCP/IP connections (when configured) now happens in the main
6209 daemon process instead of the child process, so that the TCP/IP connection
6210 count is more accurate (but it can never be perfect).
6212 11. The use of "drop" in a nested ACL was not being handled correctly in the
6213 outer ACL. Now, if condition failure induced by the nested "drop" causes
6214 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
6215 or "require"), the connection is dropped.
6217 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
6218 that yield "discard" can now be used with an "accept" or a "discard" verb,
6219 but an error is generated for any others (because I can't see a useful way
6220 to define what should happen).
6222 13. When an ACL is read dynamically from a file (or anywhere else), the lines
6223 are now processed in the same way as lines in the Exim configuration file.
6224 In particular, continuation lines are supported.
6226 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
6228 15. Added -ti meaning -t -i.
6230 16. Check for letters, digits, hyphens, and dots in the names of dnslist
6231 domains, and warn by logging if others are found.
6233 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
6234 returned from GIFCONF when Exim is trying to find the list of interfaces on
6235 a host. The code in os.c has been modified to copy each ifreq to an aligned
6236 structure in all cases.
6238 Also, in some cases, the returned ifreq's were being copied to a 'struct
6239 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
6240 means the last couple of bytes of an IPv6 address could be chopped if the
6241 ifreq contained only a normal sockaddr (14 bytes storage).
6243 18. Named domain lists were not supported in the hosts_treat_as_local option.
6244 An entry such as +xxxx was not recognized, and was treated as a literal
6247 19. Ensure that header lines added by a DATA ACL are included in the reject log
6248 if the ACL subsequently rejects the message.
6250 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
6251 MD5 (which is deprecated).
6253 21. When testing a filter file using -bf, Exim was writing a message when it
6254 took the sender from a "From " line in the message, but it was not doing so
6255 when it took $return_path from a Return-Path: header line. It now does.
6257 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
6258 with a valid header line field name (a series of printing characters
6259 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
6261 23. Changed "disc" in the source to "disk" to conform to the documentation and
6262 the book and for uniformity.
6264 24. Ignore Sendmail's -Ooption=value command line item.
6266 25. When execve() failed while trying to run a command in a pipe transport,
6267 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
6268 could be confused with a return value of 69 from the command itself. This
6269 has been changed to 127, the value the shell returns if it is asked to run
6270 a non-existent command. The wording for the related log line suggests a
6271 non-existent command as the problem.
6273 26. If received_header_text expands to an empty string, do not add a Received:
6274 header line to the message. (Well, it adds a token one on the spool, but
6275 marks it "old" so that it doesn't get used or transmitted.)
6277 27. Installed eximstats 1.28 (addition of -nt option).
6279 28. There was no check for failure on the call to getsockname() in the daemon
6280 code. This can fail if there is a shortage of resources on the system, with
6281 ENOMEM, for example. A temporary error is now given on failure.
6283 29. Contrary to the C standard, it seems that in some environments, the
6284 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
6285 program. Exim now does this explicitly; it affects the formatting of
6286 timestamps using strftime().
6288 30. If exiqsumm was given junk data, it threw up some uninitialized variable
6289 complaints. I've now initialized all the variables, to avoid this.
6291 32. Header lines added by a system filter were not being "seen" during
6292 transport-time rewrites.
6294 33. The info_callback() function passed to OpenSSL is set up with type void
6295 (*)(SSL *, int, int), as described somewhere. However, when calling the
6296 function (actually a macro) that sets it up, the type void(*)() is
6297 expected. I've put in a cast to prevent warnings from picky compilers.
6299 34. If a DNS black list lookup found a CNAME record, but there were no A
6300 records associated with the domain it pointed at, Exim crashed.
6302 35. If a DNS black list lookup returned more than one A record, Exim ignored
6303 all but the first. It now scans all returned addresses if a particular IP
6304 value is being sought. In this situation, the contents of the
6305 $dnslist_value variable are a list of all the addresses, separated by a
6308 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
6309 to accept a host name and all its aliases if the forward lookup for any of
6310 them yielded the IP address of the incoming connection. Now it accepts only
6311 those names whose forward lookup yields the correct IP address. Any other
6312 names are discarded. This closes a loophole whereby a rogue DNS
6313 administrator could create reverse DNS records to break through a
6314 wildcarded host restriction in an ACL.
6316 37. If a user filter or a system filter that ran in a subprocess used any of
6317 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
6318 the wrong values were passed to the pipe command ($thisaddress had the
6319 value of $0, $0 had the value of $1, etc). This bug was introduced by
6320 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
6322 38. Improved the line breaking for long SMTP error messages from ACLs.
6323 Previously, if there was no break point between 40 and 75 characters, Exim
6324 left the rest of the message alone. Two changes have been made: (a) I've
6325 reduced the minimum length to 35 characters; (b) if it can't find a break
6326 point between 35 and 75 characters, it looks ahead and uses the first one
6327 that it finds. This may give the occasional overlong line, but at least the
6328 remaining text gets split now.
6330 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
6331 file descriptors might be low, and that setting 1000 would always raise it.
6332 It turns out that in some environments, the limit is already over 1000 and
6333 that lowering it causes trouble. So now Exim takes care not to decrease it.
6335 40. When delivering a message, the value of $return_path is set to $sender_
6336 address at the start of routing (routers may change the value). By an
6337 oversight, this default was not being set up when an address was tested by
6338 -bt or -bv, which affected the outcome if any router or filter referred to
6341 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
6342 log only when "message" or "log_message" are set. However, if one of the
6343 conditions was an address verification, or a call to a nested ACL, the
6344 messages generated by the underlying test were being passed through. This
6345 no longer happens. The underlying message is available in $acl_verify_
6346 message for both "message" and "log_message" expansions, so it can be
6347 passed through if needed.
6349 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
6350 new expansion $bh_ to give the encoded byte string without charset
6351 translation. Translation happens only if iconv() is available; HAVE_ICONV
6352 indicates this at build time. HEADERS_CHARSET gives the charset to
6353 translate to; headers_charset can change it in the configuration, and
6354 "headers charset" can change it in an individual filter file.
6356 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
6357 that creates RFC 2047 encoded "words" labels them as that charset instead
6358 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
6359 expansion operator; (ii) when Exim creates a From: line for a local
6360 message; (iii) when a header line is rewritten to include a "phrase" part.
6362 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
6363 buggy, causing it to skip the first lines of messages whose message ID
6364 ended in 'D'. This would not have bitten before Exim release 4.14, because
6365 message IDs were unlikely to end in 'D' before then. The effect was to have
6366 incorrect size information for certain domains.
6368 45. #include "config.h" was missing at the start of the crypt16.c module. This
6369 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
6372 46. If there was a timeout during a "random" callout check, Exim treated it as
6373 a failure of the random address, and carried on sending RSET and the real
6374 address. If the delay was just some slowness somewhere, the response to the
6375 original RCPT would be taken as a response to RSET and so on, causing
6376 mayhem of various kinds.
6378 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
6379 when I implemented it. It didn't allow for the fact that some option values
6380 may legitimatetly be negative (e.g. size_addition), and it didn't even do
6381 the right test for positive values.
6383 48. Domain names in DNS records are case-independent. Exim always looks them up
6384 in lower case. Some resolvers return domain names in exactly the case they
6385 appear in the zone file, that is, they may contain uppercase letters. Not
6386 all resolvers do this - some return always lower case. Exim was treating a
6387 change of case by a resolver as a change of domain, similar to a widening
6388 of a domain abbreviation. This triggered its re-routing code and so it was
6389 trying to route what was effectively the same domain again. This normally
6390 caused routing to fail (because the router wouldn't handle the domain
6391 twice). Now Exim checks for this case specially, and just changes the
6392 casing of the domain that it ultimately uses when it transmits the message
6395 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
6398 50. If a filter generated a file delivery with a non-absolute name (possible if
6399 no home directory exists for the router), the forbid_file option was not
6402 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
6403 the existing equality matching.
6405 52. Exim was using ints instead of ino_t variables in some places where it was
6406 dealing with inode numbers.
6408 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
6409 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
6410 called TMPDIR, and if it finds it is different, it changes its value.
6412 54. The smtp_printf() function is now made available to local_scan() so
6413 additional output lines can be written before returning. There is also an
6414 smtp_fflush() function to enable the detection of a dropped connection.
6415 The variables smtp_input and smtp_batched_input are exported to
6418 55. Changed the default runtime configuration: the message "Unknown user"
6419 has been removed from the ACL, and instead placed on the localuser router,
6420 using the cannot_route_message feature. This means that any verification
6421 failures that generate their own messages won't get overridden. Similarly,
6422 the "Unrouteable address" message that was in the ACL for unverifiable
6423 relay addresses has also been removed.
6425 56. Added hosts_avoid_esmtp to the smtp transport.
6427 57. The exicyclog script was not checking for the esoteric option
6428 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
6429 will work only if exicyclog is run under the appropriate euid.
6431 58. Following a discussion on the list, the rules by which Exim recognises line
6432 endings on incoming messages have been changed. The -dropcr and drop_cr
6433 options are now no-ops, retained only for backwards compatibility. The
6434 following line terminators are recognized: LF CRLF CR. However, special
6435 processing applies to CR:
6437 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
6438 nor a local message in the state where . is a terminator.
6440 (ii) If a bare CR is encountered in a header line, an extra space is added
6441 after the line terminator so as not to end the header. The reasoning
6442 behind this is that bare CRs in header lines are most likely either
6443 to be mistakes, or people trying to play silly games.
6445 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
6446 was being incorrectly given as 18 bytes larger than it should have been.
6447 This is a VOB (very old bug).
6449 60. This may never have affected anything current, but just in case it has:
6450 When the local host is found other than at the start of a list of hosts,
6451 the local host, those with the same MX, and any that follow, are discarded.
6452 When the list in question was part of a longer list of hosts, the following
6453 hosts (not currently being processed) were also being discarded. This no
6454 longer happens. I'm not sure if this situation could ever has previously
6457 61. Added the "/MX" feature to lists of hosts in the manualroute and query
6460 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
6461 header. This is something that is recommended in a new Internet Draft, and
6462 is something that is documented as being done by Sendmail. There are two
6463 possible values. For messages generated by the autoreply transport, Exim
6466 Auto-Submitted: auto-replied
6468 whereas for all other generated messages (e.g. bounces) it adds
6470 Auto-Submitted: auto-generated
6472 63. The "personal" condition in filters now includes a test for the
6473 Auto-Submitted: header. If it contains the string "auto-" the message it
6474 not considered personal.
6476 64. Added rcpt_include_affixes as a generic transport option.
6478 65. Added queue_only_override (default true).
6480 66. Added the syslog_duplication option.
6482 67. If what should have been the first header line of a message consisted of
6483 a space followed by a colon, Exim was mis-interpreting it as a header line.
6484 It isn't of course - it is syntactically invalid and should therefore be
6485 treated as the start of the message body. The misbehaviour could have
6486 caused a number of strange effects, including loss of data in subsequent
6487 header lines, and spool format errors.
6489 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
6490 client host had authenticated. This control can now be exercised by an ACL
6491 for more flexibility.
6493 69. By default, callouts do not happen when testing with -bh. There is now a
6494 variant, -bhc, which does actually run the callout code, including
6495 consulting and updating the callout cache.
6497 70. Added support for saslauthd authentication, courtesy of Alexander
6500 71. If statvfs() failed on the spool or log directories while checking their
6501 size for availability, Exim confusingly gave the error "space shortage".
6502 Furthermore, in debugging mode it crashed with a floating point exception.
6503 These checks are done if check_{spool,log}_{space,inodes} are set, and when
6504 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
6505 serious problem, Exim now writes to the main and panic logs when this
6506 happens, with details of the failure. It then refuses to accept the
6507 incoming message, giving the message "spool directory problem" or "log
6508 directory problem" with a 421 code for SMTP messages.
6510 72. When Exim is about to re-exec itself, it ensures that the file descriptors
6511 0, 1, and 2 exist, because some OS complain for execs without them (see
6512 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
6513 descriptors. However, the code omitted to check that the open succeeded,
6514 causing mysterious errors if for some reason the permissions on /dev/null
6515 got screwed. Now Exim writes a message to the main and panic logs, and
6516 bombs out if it can't open /dev/null.
6518 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
6519 interact so that it is all more flexible. It is supposed to remain
6520 backwards compatible. Also added extra_local_interfaces.
6522 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
6523 to bomb out with an assertion failure - to the client this appears as a
6524 connection drop. This problem occurs in the part of the code that was taken
6525 from the Samba project. Fortunately, the assertion is in a very simple
6526 function, so I have fixed this by reproducing the function inline in the
6527 one place where it is called, and arranging for authentication to fail
6528 instead of killing the process with assert().
6530 75. The SPA client code was not working when the server requested OEM rather
6531 than Unicode encoding.
6533 76. Added code to make require_files with a specific uid setting more usable in
6534 the case where statting the file as root fails - usually a non-root-mounted
6535 NFS file system. When this happens and the failure is EACCES, Exim now
6536 forks a subprocess and does the per-uid checking as the relevant uid.
6538 77. Added process_log_path.
6540 78. If log_file_path was not explicitly set, a setting of check_log_space or
6541 check_log_inodes was ignored.
6543 79. If a space check for the spool or log partitions fails, the incident is now
6544 logged. Of course, in the latter case the data may get lost...
6546 80. Added the %p formatting code to string_format() so that it can be used to
6547 print addresses in debug_print(). Adjusted all the address printing in the
6548 debugging in store.c to use %p rather than %d.
6550 81. There was a concern that a line of code in smtp_in.c could overflow a
6551 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
6552 initially expressed, the concern was not well-founded, because trailing
6553 spaces are removed early. However, if the trailing spaces were followed by
6554 a NULL, they did not get removed, so the overflow was possible. Two fixes
6557 (a) I re-wrote the offending code in a cleaner fashion.
6558 (b) If an incoming SMTP command contains a NULL character, it is rejected
6561 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
6562 runs initgroups(), so that if the Exim user is in any additional groups,
6563 they will be used during message reception.
6569 The change log for 4.20 and earlier releases has been archived.