1 Change log file for Exim from version 4.21
2 -------------------------------------------
7 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
8 and 3.4.4 - once the server is enabled to respond to an OCSP request
9 it does even when not requested, resulting in a stapling non-aware
10 client dropping the TLS connection.
12 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
13 support variable-length bit vectors. No functional change.
15 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
16 The I= interface field on outgoing lines is now after the H= remote
17 host field, same as incoming lines. There is a separate
18 outgoing_interface log selector which allows you to disable the
21 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
22 If not running log_selector +smtp_connection the mainlog would be held
23 open indefinitely after a "too many connections" event, including to a
24 deleted file after a log rotate. Leave the per net connection logging
25 leaving it open for efficiency as that will be quickly detected by the
26 check on the next write.
28 HS/01 Bug 1671: Fix post transport crash.
29 Processing the wait-<transport> messages could crash the delivery
30 process if the message IDs didn't exist for some reason. When
31 using 'split_spool_directory=yes' the construction of the spool
32 file name failed already, exposing the same netto behaviour.
34 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
35 mime_regex ACL conditions.
37 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
38 to DSN fail messages (bounces): remote IP, remote greeting, remote response
39 to HELO, local diagnostic string.
41 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
42 log line to debug. Even when configured with a tls authenticator many
43 client connections are expected to not authenticate in this way, so
44 an authenticate fail is not an error.
46 HS/02 Add the Exim version string to the process info. This way exiwhat
47 gives some more detail about the running daemon.
49 JH/06 Bug 1395: time-limit cacheing of DNS lookups, to the TTL value. This may
50 matter for fast-change records such as DNSBLs.
52 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
53 retry record, even if constant. There may be multiple transports with
54 different interface settings and the retry behaviour needs to be kept
57 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
59 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
61 JH/10 Bug 840: fix log_defer_output option of pipe transport
63 JH/11 Bug 830: use same host for all RCPTS of a message, even under
64 hosts_randomize. This matters a lot when combined with mua_wrapper.
66 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
67 ${quote_pgsql:<string>} operator.
69 JH/13 Bug 1708: avoid misaligned access in cached lookup.
71 JH/14 Change header file name for freeradius-client. Relevant if compiling
72 with Radius support; from the Gentoo tree and checked under Fedora.
74 JH/15 Bug 1712: Introduce $prdr_requested flag variable
76 JH/16 Bug 1714: Permit an empty string as expansion result for transport
77 option transport_filter, meaning no filtering.
79 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
81 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
82 defaults to "*" (all hosts). The variable is now available when not built
83 with TLS, default unset, mainly to enable keeping the testuite sane.
84 If a server certificate is not supplied (via tls_certificate) an error is
85 logged, and clients will find TLS connections fail on startup. Presumably
86 they will retry in-clear.
87 Packagers of Exim are strongly encouraged to create a server certificate
90 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
91 with the $config_file variable.
93 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
94 in transport context, after the attempt, and per-recipient. The latter type
95 is per host attempted. The event data is the error message, and the errno
96 information encodes the lookup type (A vs. MX) used for the (first) host,
97 and the trailing two digits of the smtp 4xx reponse.
99 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
100 to write to mainlog (or rejectlog, paniclog) in the window between file
101 creation and permissions/ownership being changed. Particularly affects
102 installations where exicyclog is run as root, rather than exim user;
103 result is that the running daemon panics and dies.
105 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
107 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
108 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
109 "pri" and "weight". Note that the previous implicit priority given by the
110 list order is no longer honoured.
112 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalisation
115 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
116 by defining SUPPORT_SOCKS.
118 JH/26 Move PROXY support from Experimental to mainline, enabled for a build
119 by defining SUPPORT_PROXY. Note that the proxy_required_hosts option
120 is renamed to hosts_proxy, and the proxy_{host,target}_{address,port}.
121 variables are renamed to proxy_{local,external}_{address,port}.
126 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
129 JH/02 The smtp transport option "multi_domain" is now expanded.
131 JH/03 The smtp transport now requests PRDR by default, if the server offers
134 JH/04 Certificate name checking on server certificates, when exim is a client,
135 is now done by default. The transport option tls_verify_cert_hostnames
136 can be used to disable this per-host. The build option
137 EXPERIMENTAL_CERTNAMES is withdrawn.
139 JH/05 The value of the tls_verify_certificates smtp transport and main options
140 default to the word "system" to access the system default CA bundle.
141 For GnuTLS, only version 3.0.20 or later.
143 JH/06 Verification of the server certificate for a TLS connection is now tried
144 (but not required) by default. The verification status is now logged by
145 default, for both outbound TLS and client-certificate supplying inbound
148 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
151 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
152 Status Notification (bounce) messages are now MIME format per RFC 3464.
153 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
154 under the control of the dsn_advertise_hosts option, and routers may
155 have a dsn_lasthop option.
157 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
158 default, modifiable by a malware= option. The list separator for
159 the options can now be changed in the usual way. Bug 68.
161 JH/10 The smtp_receive_timeout main option is now expanded before use.
163 JH/11 The incoming_interface log option now also enables logging of the
164 local interface on delivery outgoing connections.
166 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
167 if the interface and destination host and port all match.
169 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
172 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
173 Patch from Andrew Lewis.
175 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
176 now supports optional time-restrictions, weighting, and priority
177 modifiers per server. Patch originally by <rommer@active.by>.
179 JH/16 The spamd_address main option now supports a mixed list of local
180 and remote servers. Remote servers can be IPv6 addresses, and
181 specify a port-range.
183 JH/17 Bug 68: The spamd_address main option now supports an optional
184 timeout value per server.
186 JH/18 Bug 1581: Router and transport options headers_add/remove can
187 now have the list separator specified.
189 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
192 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
195 JH/21 Support for the A6 type of dns record is withdrawn.
197 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
198 rather than the verbs used.
200 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
201 from 255 to 1024 chars.
203 JH/24 Verification callouts now attempt to use TLS by default.
205 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
206 are generic router options now. The defaults didn't change.
208 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
209 Original patch from Alexander Shikoff, worked over by JH.
211 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
212 files if ZCAT_COMMAND is not executable.
214 JH/26 Bug 1539: Add timout/retry options on dnsdb lookups.
216 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
218 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
219 Normally benign, it bites when the pair was led to by a CNAME;
220 modern usage is to not canoicalize the domain to a CNAME target
221 (and we were inconsistent anyway for A-only vs AAAA+A).
223 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
225 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
226 when evaluating $sender_host_dnssec.
228 JH/31 Check the HELO verification lookup for DNSSEC, adding new
229 $sender_helo_dnssec variable.
231 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
233 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
235 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
237 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
238 documented as working, but never had. Support all but $spam_report.
240 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
241 added for tls authenticator.
246 TL/01 When running the test suite, the README says that variables such as
247 no_msglog_check are global and can be placed anywhere in a specific
248 test's script, however it was observed that placement needed to be near
249 the beginning for it to behave that way. Changed the runtest perl
250 script to read through the entire script once to detect and set these
251 variables, reset to the beginning of the script, and then run through
252 the script parsing/test process like normal.
254 TL/02 The BSD's have an arc4random API. One of the functions to induce
255 adding randomness was arc4random_stir(), but it has been removed in
256 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
257 function when detected.
259 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
260 cause callback expansion.
262 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
263 syntax errors in an expansion can be treated as a string instead of
264 logging or causing an error, due to the internal use of bool_lax
265 instead of bool when processing it.
267 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
268 server certificates when making smtp deliveries.
270 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
272 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
274 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
276 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
277 Merged patch from Sebastian Wiedenroth.
279 JH/05 Fix results-pipe from transport process. Several recipients, combined
280 with certificate use, exposed issues where response data items split
281 over buffer boundaries were not parsed properly. This eventually
282 resulted in duplicates being sent. This issue only became common enough
283 to notice due to the introduction of conection certificate information,
284 the item size being so much larger. Found and fixed by Wolfgang Breyha.
286 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
287 size buffer was used, resulting in syntax errors when an expansion
290 JH/07 Add support for directories of certificates when compiled with a GnuTLS
291 version 3.3.6 or later.
293 JH/08 Rename the TPDA expermimental facility to Event Actions. The #ifdef
294 is EXPERIMENTAL_EVENT, the main-configuration and transport options
295 both become "event_action", the variables become $event_name, $event_data
296 and $event_defer_errno. There is a new variable $verify_mode, usable in
297 routers, transports and related events. The tls:cert event is now also
298 raised for inbound connections, if the main configuration event_action
301 TL/06 In test suite, disable OCSP for old versions of openssl which contained
302 early OCSP support, but no stapling (appears to be less than 1.0.0).
304 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
305 server certificate names available under the smtp transport option
306 "tls_verify_cert_hostname" now do not permit multi-component wildcard
309 JH/10 Time-related extraction expansions from certificates now use the main
310 option "timezone" setting for output formatting, and are consistent
311 between OpenSSL and GnuTLS compilations. Bug 1541.
313 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
314 encoded parameter in the incoming message. Bug 1558.
316 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
317 include certificate info, eximon was claiming there were spoolfile
320 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
322 JH/14 Log delivery-related information more consistently, using the sequence
323 "H=<name> [<ip>]" wherever possible.
325 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
326 are problematic for Debian distribution, omit them from the release
329 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
331 JH/16 Fix string representation of time values on 64bit time_t anchitectures.
334 JH/17 Fix a null-indirection in certextract expansions when a nondefault
335 output list separator was used.
340 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
341 checkers that were complaining about end of non-void function with no
344 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
345 This was a regression intruduced in 4.83 by another bugfix.
347 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
349 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
350 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
356 TF/01 Correctly close the server side of TLS when forking for delivery.
358 When a message was received over SMTP with TLS, Exim failed to clear up
359 the incoming connection properly after forking off the child process to
360 deliver the message. In some situations the subsequent outgoing
361 delivery connection happened to have the same fd number as the incoming
362 connection previously had. Exim would try to use TLS and fail, logging
363 a "Bad file descriptor" error.
365 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
366 utilities have not been installed.
368 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
369 temporary space as the ACL may create new global variables.
371 TL/01 LDAP support uses per connection or global context settings, depending
372 upon the detected version of the libraries at build time.
374 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
375 to extract and use the src ip:port in logging and expansions as if it
376 were a direct connection from the outside internet. PPv2 support was
377 updated based on HAProxy spec change in May 2014.
379 JH/02 Add ${listextract {number}{list}{success}{fail}}.
381 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
382 Properly escape header and check for NULL return.
384 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
387 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
389 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
390 characters in header names, implemented as a verify condition.
391 Contributed by Michael Fischer v. Mollard.
393 TL/05 Rename SPF condition results err_perm and err_temp to standardized
394 results permerror and temperror. Previous values are deprecated but
395 still accepted. In a future release, err_perm and err_temp will be
396 completely removed, which will be a backward incompatibility if the
397 ACL tests for either of these two old results. Patch contributed by
398 user bes-internal on the mailing list.
400 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
402 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
403 selectors, in both main and reject logs.
405 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
408 JH/07 Add malware type "sock" for talking to simple daemon.
410 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
412 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
413 routers/transports under cutthrough routing.
415 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
416 numbers. Touch up "bool" conditional to keep the same definition.
418 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
420 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
422 JH/12 Expand items in router/transport headers_add or headers_remove lists
423 individually rather than the list as a whole. Bug 1452.
425 Required for reasonable handling of multiple headers_ options when
426 they may be empty; requires that headers_remove items with embedded
427 colons must have them doubled (or the list-separator changed).
429 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
430 view the policy declared in the DMARC record. Currently, $dmarc_status
431 is a combined value of both the record presence and the result of the
434 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
436 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
437 dnslookup router and the smtp transport (applying to the forward
440 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
441 of ldap servers used for a specific lookup. Patch provided by Heiko
444 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
445 New variable $lookup_dnssec_authenticated for observability.
447 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
448 Patch submitted by Lars Timman.
450 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
452 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
453 Requires trusted mode and valid format message id, aborts otherwise.
454 Patch contributed by Heiko Schlichting.
456 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
457 certextract with support for various fields. Bug 1358.
459 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
460 is requested by default, modifiable by smtp transport option
463 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
464 operate on certificate variables to give certificate fingerprints
465 Also new ${sha256:cert_variable}.
467 JH/23 The PRDR feature is moved from being Experimental into the mainline.
469 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
470 Christian Aistleitner.
472 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
474 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
475 file. Patch from Wolfgang Breyha.
477 JH/25 Expand the coverage of the delivery $host and $host_address to
478 client authenticators run in verify callout. Bug 1476.
480 JH/26 Port service names are now accepted for tls_on_connect_ports, to
481 align with daemon_smtp_ports. Bug 72.
483 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
484 support and error reporting did not work properly.
486 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
487 and is readable. Patch from Andrew Colin Kissa.
489 TL/14 Enhance documentation of ${run expansion and how it parses the
490 commandline after expansion, particularly in the case when an
491 unquoted variable expansion results in an empty value.
493 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
495 PP/02 Fix internal collision of T_APL on systems which support RFC3123
496 by renaming away from it. Addresses GH issue 15, reported by
499 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
501 TL/15 SECURITY: prevent double expansion in math comparison functions
502 (can expand unsanitized data). Not remotely exploitable.
509 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
511 PP/02 Make -n do something, by making it not do something.
512 When combined with -bP, the name of an option is not output.
514 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
517 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
518 $sender_host_name and config options to manage this, and basic check
521 PP/05 DSCP support for outbound connections and control modifier for inbound.
523 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
524 (Only plugin which currently uses this is kerberos4, which nobody should
525 be using, but we should make it available and other future plugins might
526 conceivably use it, even though it would break NAT; stuff *should* be
527 using channel bindings instead).
529 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
530 name; added for Sendmail compatibility; requires admin caller.
531 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
532 just ignore it); requires trusted caller.
533 Also parse but ignore: -Ac -Am -X<logfile>
536 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
538 TL/02 Add +smtp_confirmation as a default logging option.
540 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
541 Patch by Magnus Holmgren from 2007-02-20.
543 TL/04 Bugzilla 1281 - Spec typo.
544 Bugzilla 1283 - Spec typo.
545 Bugzilla 1290 - Spec grammar fixes.
547 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
549 TL/06 Add Experimental DMARC support using libopendmarc libraries.
551 TL/07 Fix an out of order global option causing a segfault. Reported to dev
552 mailing list by by Dmitry Isaikin.
554 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
556 JH/02 Support "G" suffix to numbers in ${if comparisons.
558 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
560 NM/01 Bugzilla 1197 - Spec typo
561 Bugzilla 1196 - Spec examples corrections
563 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
565 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
566 gnutls_enable_pkcs11, but renamed to more accurately indicate its
569 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
570 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
572 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
573 "acl {{name}{arg}...}", and optional args on acl condition
576 JH/05 Permit multiple router/transport headers_add/remove lines.
578 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
580 JH/07 Avoid using a waiting database for a single-message-only transport.
581 Performance patch from Paul Fisher. Bugzilla 1262.
583 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
586 JH/09 Add $headers_added variable, with content from use of ACL modifier
587 add_header (but not yet added to the message). Bugzilla 199.
589 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
590 Pulled from Bugzilla 817 by Wolfgang Breyha.
592 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
594 (nb: this is the same fix as in Exim 4.80.1)
596 JH/11 Add A= logging on delivery lines, and a client_set_id option on
599 JH/12 Add optional authenticated_sender logging to A= and a log_selector
602 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
604 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
605 advertise SMTP AUTH mechanism to us, instead of a generic
606 protocol violation error. Also, make Exim more robust to bad
607 data from the Dovecot auth socket.
609 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
611 When a queue runner is handling a message, Exim first routes the
612 recipient addresses, during which it prunes them based on the retry
613 hints database. After that it attempts to deliver the message to
614 any remaining recipients. It then updates the hints database using
617 So if a recipient address works intermittently, it can get repeatedly
618 deferred at routing time. The retry hints record remains fresh so the
619 address never reaches the final cutoff time.
621 This is a fairly common occurrence when a user is bumping up against
622 their storage quota. Exim had some logic in its local delivery code
623 to deal with this. However it did not apply to per-recipient defers
624 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
626 This change adds a proper retry rule check during routing so that the
627 final cutoff time is checked against the message's age. We only do
628 this check if there is an address retry record and there is not a
629 domain retry record; this implies that previous attempts to handle
630 the address had the retry_use_local_parts option turned on. We use
631 this as an approximation for the destination being like a local
632 delivery, as in LMTP.
634 I suspect this new check makes the old local delivery cutoff check
635 redundant, but I have not verified this so I left the code in place.
637 TF/02 Correct gecos expansion when From: is a prefix of the username.
639 Test 0254 submits a message to Exim with the header
643 When I ran the test suite under the user fanf2, Exim expanded
644 the header to contain my full name, whereas it should have added
645 a Resent-Sender: header. It erroneously treats any prefix of the
646 username as equal to the username.
648 This change corrects that bug.
650 GF/01 DCC debug and logging tidyup
651 Error conditions log to paniclog rather than rejectlog.
652 Debug lines prefixed by "DCC: " to remove any ambiguity.
654 TF/03 Avoid unnecessary rebuilds of lookup-related code.
656 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
657 Bug spotted by Jeremy Harris; was flawed since initial commit.
658 Would have resulted in OCSP responses post-SNI triggering an Exim
659 NULL dereference and crash.
661 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
663 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
664 Bug detection, analysis and fix by Samuel Thibault.
665 Bugzilla 1331, Debian bug #698092.
667 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
669 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
670 Server implementation by Todd Lyons, client by JH.
671 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
672 config variable "prdr_enable" controls whether the server
673 advertises the facility. If the client requests PRDR a new
674 acl_data_smtp_prdr ACL is called once for each recipient, after
675 the body content is received and before the acl_smtp_data ACL.
676 The client is controlled by bolth of: a hosts_try_prdr option
677 on the smtp transport, and the server advertisement.
678 Default client logging of deliveries and rejections involving
679 PRDR are flagged with the string "PRDR".
681 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
682 fclose(). Diagnosis by Todd Lyons.
684 PP/17 Update configure.default to handle IPv6 localhost better.
685 Patch by Alain Williams (plus minor tweaks).
688 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
689 This is now consistent with GnuTLS, and is now documented: the
690 previous undocumented portable approach to treating the option as
691 unset was to force an expansion failure. That still works, and
692 an empty string is now equivalent.
694 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
695 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
696 not performing validation itself.
698 PP/20 Added force_command boolean option to pipe transport.
699 Patch from Nick Koston, of cPanel Inc.
701 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
704 TF/04 Added udpsend ACL modifer and hexquote expansion operator
706 PP/21 Fix eximon continuous updating with timestamped log-files.
707 Broken in a format-string cleanup in 4.80, missed when I repaired the
708 other false fix of the same issue.
709 Report and fix from Heiko Schlichting.
712 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
713 Report from Prashanth Katuri.
715 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
716 It's SecureTransport, so affects any MacOS clients which use the
717 system-integrated TLS libraries, including email clients.
719 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
720 using a MIME ACL for non-SMTP local injection.
721 Report and assistance in diagnosis by Warren Baker.
723 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
725 JH/16 Fix comparisons for 64b. Bugzilla 1385.
727 TL/09 Add expansion variable $authenticated_fail_id to keep track of
728 last id that failed so it may be referenced in subsequent ACL's.
730 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
733 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
734 ldap library initialization, allowing self-signed CA's to be
735 used. Also properly sets require_cert option later in code by
736 using NULL (global ldap config) instead of ldap handle (per
737 session). Bug diagnosis and testing by alxgomz.
739 TL/12 Enhanced documentation in the ratelimit.pl script provided in
740 the src/util/ subdirectory.
742 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
743 renamed to Transport Post Delivery Action by Jeremy Harris, as
746 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
747 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
748 redis_servers = needs to be configured which will be used by the redis
749 lookup. Patch from Warren Baker, of The Packet Hub.
751 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
753 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
754 hostname or reverse DNS when processing a host list. Used suggestions
755 from multiple comments on this bug.
757 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
759 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
760 Missed a few lines, added it to make the runtest require no keyboard
763 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
764 contains upper case chars. Make router use caseful_local_part.
766 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
767 support when GnuTLS has been built with p11-kit.
773 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
775 This, or similar/improved, will also be change PP/11 of 4.82.
781 PP/01 Handle short writes when writing local log-files.
782 In practice, only affects FreeBSD (8 onwards).
783 Bugzilla 1053, with thanks to Dmitry Isaikin.
785 NM/01 Bugzilla 949 - Documentation tweak
787 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
790 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
792 PP/02 Implemented gsasl authenticator.
794 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
796 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
797 `pkg-config foo` for cflags/libs.
799 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
800 with rest of GSASL and with heimdal_gssapi.
802 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
803 `pkg-config foo` for cflags/libs for the TLS implementation.
805 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
806 properties get this fed in as external SSF. A number of robustness
807 and debugging improvements to the cyrus_sasl authenticator.
809 PP/08 cyrus_sasl server now expands the server_realm option.
811 PP/09 Bugzilla 1214 - Log authentication information in reject log.
812 Patch by Jeremy Harris.
814 PP/10 Added dbmjz lookup type.
816 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
818 PP/12 MAIL args handles TAB as well as SP, for better interop with
819 non-compliant senders.
820 Analysis and variant patch by Todd Lyons.
822 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
823 Bug report from Lars Müller <lars@samba.org> (via SUSE),
824 Patch from Dirk Mueller <dmueller@suse.com>
826 PP/13 tls_peerdn now print-escaped for spool files.
827 Observed some $tls_peerdn in wild which contained \n, which resulted
828 in spool file corruption.
830 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
831 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
832 or write after TLS renegotiation, which otherwise led to messages
835 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
836 as a tracking header (ie: a signed header comes before the signature).
837 Patch from Wolfgang Breyha.
839 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
840 comma-sep list; embedded commas doubled.
842 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
844 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
846 Report and patch from Dmitry Banschikov.
848 PP/16 Removed "dont_insert_empty_fragments" fron "openssl_options".
849 Removed SSL_clear() after SSL_new() which led to protocol negotiation
850 failures. We appear to now support TLS1.1+ with Exim.
852 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
853 lets Exim select keys and certificates based upon TLS SNI from client.
854 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
855 before an outbound SMTP session. New log_selector, +tls_sni.
857 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
858 NULL dereference. Report and patch from Alun Jones.
860 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
861 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
862 Not seeing resolver debug output on NetBSD, but suspect this is a
863 resolver implementation change.
865 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
866 Left warnings. Added "eximon gdb" invocation mode.
868 PP/21 Defaulting "accept_8bitmime" to true, not false.
870 PP/22 Added -bw for inetd wait mode support.
872 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
873 locate the relevant includes and libraries. Made this the default.
875 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
876 Bugzilla 1246, report and most of solution from Tomasz Kusy.
878 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
879 This may cause build issues on older platforms.
881 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
882 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
883 gnutls_require_mac & gnutls_require_protocols (no longer supported).
884 Added SNI support via GnuTLS too.
885 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
887 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
889 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
890 Applied second patch from Janne, implementing suggestion to default
891 multiple-strings-in-record handling to match SPF spec.
893 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
895 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
896 read-only, out of scope).
897 Patch from Wolfgang Breyha, report from Stuart Northfield.
899 PP/29 Fix three issues highlighted by clang analyser static analysis.
900 Only crash-plausible issue would require the Cambridge-specific
901 iplookup router and a misconfiguration.
902 Report from Marcin Mirosław.
904 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
906 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
907 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
908 As part of this, removing so much warning spew let me fix some minor
909 real issues in debug logging.
911 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
912 assignment on my part. Fixed.
914 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
915 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
916 Janne Snabb (who went above and beyond: thank you).
918 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
919 string otherwise requires a connection and a bunch more work and it's
920 relatively easy to get wrong. Should also expose TLS library linkage
923 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
924 64-bit ${eval} (JH/03).
926 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
927 GNU libc to support some of the 64-bit stuff, should not lead to
928 conflicts. Defined before os.h is pulled in, so if a given platform
929 needs to override this, it can.
931 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
932 protection layer was required, which is not implemented.
933 Bugzilla 1254, patch from Wolfgang Breyha.
935 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
936 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
937 tls_dhparam take prime identifiers. Also unbreak combination of
938 OpenSSL+DH_params+TLSSNI.
940 PP/39 Disable SSLv2 by default in OpenSSL support.
946 PP/01 Solaris build fix for Oracle's LDAP libraries.
947 Bugzilla 1109, patch from Stephen Usher.
949 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
951 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
954 TF/02 Fix a couple more cases where we did not log the error message
955 when unlink() failed. See also change 4.74-TF/03.
957 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
958 lock up or crash if it happened to be inside a call to libc when it
959 got a SIGUSR1 from exiwhat.
961 The SIGUSR1 handler appends the current process status to the process
962 log which is later printed by exiwhat. It used to use the general
963 purpose logging code to do this, but several functions it calls are
964 not safe for signals.
966 The new output code in the SIGUSR1 handler is specific to the process
967 log, and simple enough that it's easy to inspect for signal safety.
968 Removing some special cases also simplifies the general logging code.
969 Removing the spurious timestamps from the process log simplifies
972 TF/04 Improved ratelimit ACL condition.
974 The /noupdate option has been deprecated in favour of /readonly which
975 has clearer semantics. The /leaky, /strict, and /readonly update modes
976 are mutually exclusive. The update mode is no longer included in the
977 database key; it just determines when the database is updated. (This
978 means that when you upgrde Exim will forget old rate measurements.)
980 Exim now checks that the per_* options are used with an update mode that
981 makes sense for the current ACL. For example, when Exim is processing a
982 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
983 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
984 must specify per_mail/readonly. If you omit the update mode it defaults to
985 /leaky where that makes sense (as before) or /readonly where required.
987 The /noupdate option is now undocumented but still supported for
988 backwards compatibility. It is equivalent to /readonly except that in
989 ACLs where /readonly is required you may specify /leaky/noupdate or
990 /strict/noupdate which are treated the same as /readonly.
992 A useful new feature is the /count= option. This is a generalization
993 of the per_byte option, so that you can measure the throughput of other
994 aggregate values. For example, the per_byte option is now equivalent
995 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
997 The per_rcpt option has been generalized using the /count= mechanism
998 (though it's more complicated than the per_byte equivalence). When it is
999 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
1000 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
1001 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
1002 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
1003 Note that using per_rcpt with a non-readonly update mode in more than
1004 one ACL will cause the recipients to be double-counted. (The per_mail
1005 and per_byte options don't have this problem.)
1007 The handling of very low rates has changed slightly. If the computed rate
1008 is less than the event's count (usually one) then this event is the first
1009 after a long gap. In this case the rate is set to the same as this event's
1010 count, so that the first message of a spam run is counted properly.
1012 The major new feature is a mechanism for counting the rate of unique
1013 events. The new per_addr option counts the number of different
1014 recipients that someone has sent messages to in the last time period. It
1015 behaves like per_rcpt if all the recipient addresses are different, but
1016 duplicate recipient addresses do not increase the measured rate. Like
1017 the /count= option this is a general mechanism, so the per_addr option
1018 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
1019 example, measure the rate that a client uses different sender addresses
1020 with the options per_mail/unique=$sender_address. There are further
1021 details in the main documentation.
1023 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
1025 TF/06 Removed a few PCRE remnants.
1027 TF/07 Automatically extract Exim's version number from tags in the git
1028 repository when doing development or release builds.
1030 PP/02 Raise smtp_cmd_buffer_size to 16kB.
1031 Bugzilla 879. Patch from Paul Fisher.
1033 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
1034 Heavily based on revision 40f9a89a from Simon Arlott's tree.
1037 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
1039 PP/05 Variable $av_failed, true if the AV scanner deferred.
1040 Bugzilla 1078. Patch from John Horne.
1042 PP/06 Stop make process more reliably on build failure.
1043 Bugzilla 1087. Patch from Heiko Schlittermann.
1045 PP/07 Make maildir_use_size_file an _expandable_ boolean.
1046 Bugzilla 1089. Patch from Heiko Schlittermann.
1048 PP/08 Handle ${run} returning more data than OS pipe buffer size.
1049 Bugzilla 1131. Patch from Holger Weiß.
1051 PP/09 Handle IPv6 addresses with SPF.
1052 Bugzilla 860. Patch from Wolfgang Breyha.
1054 PP/10 GnuTLS: support TLS 1.2 & 1.1.
1056 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
1059 PP/11 match_* no longer expand right-hand-side by default.
1060 New compile-time build option, EXPAND_LISTMATCH_RHS.
1061 New expansion conditions, "inlist", "inlisti".
1063 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
1065 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
1067 PP/14 fix log_write() format string regression from TF/03.
1068 Bugzilla 1152. Patch from Dmitry Isaikin.
1074 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
1076 PP/02 Harmonised TLS library version reporting; only show if debugging.
1077 Layout now matches that introduced for other libraries in 4.74 PP/03.
1079 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
1081 PP/04 New "dns_use_edns0" global option.
1083 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
1086 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
1087 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
1089 TK/01 Updated PolarSSL code to 0.14.2.
1090 Bugzilla 1097. Patch from Andreas Metzler.
1092 PP/07 Catch divide-by-zero in ${eval:...}.
1093 Fixes bugzilla 1102.
1095 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
1098 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
1099 format-string attack -- SECURITY: remote arbitrary code execution.
1101 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
1102 time unintentionally subject to list matching rules, letting the header
1103 cause arbitrary Exim lookups (of items which can occur in lists, *not*
1104 arbitrary string expansion). This allowed for information disclosure.
1106 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
1107 INT_MIN/-1 -- value coerced to INT_MAX.
1113 NM/01 Workround for PCRE version dependency in version reporting
1116 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
1117 This fixes portability to compilers other than gcc, notably
1118 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
1120 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
1121 makefiles for portability to HP-UX and POSIX correctness.
1123 PP/01 Permit LOOKUP_foo enabling on the make command-line.
1124 Also via indented variable definition in the Makefile.
1125 (Debugging by Oliver Heesakkers).
1127 PP/02 Restore caching of spamd results with expanded spamd_address.
1128 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
1130 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
1131 Improves build reliability. Fix from: Frank Elsner
1133 NM/02 Fix wide character breakage in the rfc2047 coding
1134 Fixes bug 1064. Patch from Andrey N. Oktyabrski
1136 NM/03 Allow underscore in dnslist lookups
1137 Fixes bug 1026. Patch from Graeme Fowler
1139 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
1140 Code patches from Adam Ciarcinski of NetBSD.
1142 NM/04 Fixed exiqgrep to cope with mailq missing size issue
1145 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
1146 is logged, to avoid truncation. Patch from John Horne.
1148 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
1149 Patch from Jakob Hirsch.
1151 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
1152 SQL string expansion failure details.
1153 Patch from Andrey Oktyabrski.
1155 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
1156 Patch from Simon Arlott.
1158 PP/09 New lookups functionality failed to compile on old gcc which rejects
1159 extern declarations in function scope.
1160 Patch from Oliver Fleischmann
1162 PP/10 Use sig_atomic_t for flags set from signal handlers.
1163 Check getgroups() return and improve debugging.
1164 Fixed developed for diagnosis in bug 927 (which turned out to be
1167 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
1168 Patch from Mark Zealey.
1170 PP/12 Bugzilla 1056: Improved spamd server selection.
1171 Patch from Mark Zealey.
1173 PP/13 Bugzilla 1086: Deal with maildir quota file races.
1174 Based on patch from Heiko Schlittermann.
1176 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
1177 Patch from Uwe Doering, sign-off by Michael Haardt.
1179 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
1180 variable declaration deep within a block. Bug and patch from
1183 PP/15 lookups-Makefile IRIX compatibilty coercion.
1185 PP/16 Make DISABLE_DKIM build knob functional.
1187 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
1188 Patch by Simon Arlott
1190 TF/03 Fix valgrind.h portability to C89 compilers that do not support
1191 variable argument macros. Our copy now differs from upstream.
1197 TF/01 Failure to get a lock on a hints database can have serious
1198 consequences so log it to the panic log.
1200 TF/02 Log LMTP confirmation messages in the same way as SMTP,
1201 controlled using the smtp_confirmation log selector.
1203 TF/03 Include the error message when we fail to unlink a spool file.
1205 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
1206 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
1207 for maintaining out-of-tree patches for some time.
1209 PP/01 Bugzilla 139: Documentation and portability issues.
1210 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
1211 Handle per-OS dynamic-module compilation flags.
1213 PP/02 Let /dev/null have normal permissions.
1214 The 4.73 fixes were a little too stringent and complained about the
1215 permissions on /dev/null. Exempt it from some checks.
1216 Reported by Andreas M. Kirchwitz.
1218 PP/03 Report version information for many libraries, including
1219 Exim version information for dynamically loaded libraries. Created
1220 version.h, now support a version extension string for distributors
1221 who patch heavily. Dynamic module ABI change.
1223 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
1224 privilege escalation vulnerability whereby the Exim run-time user
1225 can cause root to append content of the attacker's choosing to
1228 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
1231 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
1232 If dropping privileges for untrusted macros, we disabled normal logging
1233 on the basis that it would fail; for the Exim run-time user, this is not
1234 the case, and it resulted in successful deliveries going unlogged.
1235 Fixed. Reported by Andreas Metzler.
1241 PP/01 Date: & Message-Id: revert to normally being appended to a message,
1242 only prepend for the Resent-* case. Fixes regression introduced in
1243 Exim 4.70 by NM/22 for Bugzilla 607.
1245 PP/02 Include check_rfc2047_length in configure.default because we're seeing
1246 increasing numbers of administrators be bitten by this.
1248 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
1250 PP/03 Bugzilla 994: added openssl_options main configuration option.
1252 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
1254 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
1256 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
1258 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
1259 without a peer certificate, leading to a segfault because of an
1260 assumption that peers always have certificates. Be a little more
1261 paranoid. Problem reported by Martin Tscholak.
1263 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
1264 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
1265 NB: ClamAV planning to remove STREAM in "middle of 2010".
1266 CL also introduces -bmalware, various -d+acl logging additions and
1267 more caution in buffer sizes.
1269 PP/09 Implemented reverse_ip expansion operator.
1271 PP/10 Bugzilla 937: provide a "debug" ACL control.
1273 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
1275 PP/12 Bugzilla 973: Implement --version.
1277 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
1279 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
1281 PP/15 Bugzilla 816: support multiple condition rules on Routers.
1283 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
1284 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
1285 ignore trailing whitespace.
1287 JJ/02 prevent non-panic DKIM error from being sent to paniclog
1289 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
1292 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
1293 Notification from Dr Andrew Aitchison.
1295 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
1296 ExtendedDetectionInfo response format.
1297 Notification from John Horne.
1299 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
1302 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
1303 XSL and documented dependency on system catalogs, with examples of how
1306 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
1309 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
1310 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
1311 configuration file which is writeable by the Exim user or group.
1313 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
1314 of configuration files to cover files specified with the -C option if
1315 they are going to be used with root privileges, not just the default
1318 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
1319 option (effectively making it always true).
1321 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
1322 files to be used while preserving root privileges.
1324 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
1325 that rogue child processes cannot use them.
1327 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
1328 run-time user, instead of root.
1330 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
1331 Exim run-time user without dropping privileges.
1333 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
1334 result string, instead of calling string_vformat() twice with the same
1337 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
1338 for other users. Others should always drop root privileges if they use
1339 -C on the command line, even for a whitelisted configure file.
1341 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
1343 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
1349 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
1350 $data_path, and $header_path variables; fixed documentation bugs and
1353 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
1354 exipick to access non-standard spools, including the "frozen" queue
1357 NM/01 Bugzilla 965: Support mysql stored procedures.
1358 Patch from Alain Williams
1360 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
1362 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
1363 Patch from Andreas Metzler
1365 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
1366 Patch from Kirill Miazine
1368 NM/05 Bugzilla 671: Added umask to procmail example.
1370 JJ/03 installed exipick 20100323.0, fixing doc bug
1372 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
1373 directory. Notification and patch from Dan Rosenberg.
1375 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
1377 TK/02 Improve log output when DKIM signing operation fails.
1379 MH/01 Treat the transport option dkim_domain as a colon separated
1380 list, not as a single string, and sign the message with each element,
1381 omitting multiple occurences of the same signer.
1383 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
1384 Bugzilla 985, 986. Patch by Simon Arlott
1386 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
1387 Patch by Simon Arlott
1389 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
1390 MBX locking. Notification from Dan Rosenberg.
1396 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
1398 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
1400 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
1402 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
1404 NM/04 Bugzilla 915: Flag broken perl installation during build.
1410 TK/01 Added patch by Johannes Berg that expands the main option
1411 "spamd_address" if it starts with a dollar sign.
1413 TK/02 Write list of recipients to X-Envelope-Sender header when building
1414 the mbox-format spool file for content scanning (suggested by Jakob
1417 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
1418 (http://www.dcc-servers.net/) support via dccifd. Activated by
1419 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
1421 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
1422 by Mark Daniel Reidel <mr@df.eu>.
1424 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
1425 When building exim an external PCRE library is now needed -
1426 PCRE is a system library on the majority of modern systems.
1427 See entry on PCRE_LIBS in EDITME file.
1429 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
1430 conversation. Added nologin parameter to request.
1431 Patch contributed by Kirill Miazine.
1433 TF/01 Do not log submission mode rewrites if they do not change the address.
1435 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
1437 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
1438 log files in place. Contributed by Roberto Lima.
1440 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
1442 TF/03 Bugzilla 615: When checking the local_parts router precondition
1443 after a local_part_suffix or local_part_prefix option, Exim now
1444 does not use the address's named list lookup cache, since this
1445 contains cached lookups for the whole local part.
1447 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
1448 Robert Millan. Documentation is in experimental-spec.txt.
1450 TF/04 Bugzilla 668: Fix parallel build (make -j).
1452 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
1454 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
1455 Patch provided by Jan Srzednicki.
1457 TF/05 Leading white space used to be stripped from $spam_report which
1458 wrecked the formatting. Now it is preserved.
1460 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
1461 that they are available at delivery time.
1463 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
1465 TF/08 TLS error reporting now respects the incoming_interface and
1466 incoming_port log selectors.
1468 TF/09 Produce a more useful error message if an SMTP transport's hosts
1469 setting expands to an empty string.
1471 NM/06 Bugzilla 744: EXPN did not work under TLS.
1472 Patch provided by Phil Pennock.
1474 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
1475 Patch provided by Richard Godbee.
1477 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
1478 acl_smtp_notquit, added index entry.
1480 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
1481 Patch provided by Eugene Bujak.
1483 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
1484 accept(). Patch provided by Maxim Dounin.
1486 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
1487 Patch provided by Phil Pennock.
1489 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
1491 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
1492 Patch provided by Brad "anomie" Jorsch.
1494 NM/14 Bugzilla 622: Added timeout setting to transport filter.
1495 Patch provided by Dean Brooks.
1497 TK/05 Add native DKIM support (does not depend on external libraries).
1499 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
1500 Patch provided by Graeme Fowler.
1502 NM/16 Bugzilla 851: Documentation example syntax fix.
1504 NM/17 Changed NOTICE file to remove references to embedded PCRE.
1506 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
1509 NM/19 Bugzilla 745: TLS version reporting.
1510 Patch provided by Phil Pennock.
1512 NM/20 Bugzilla 167: bool: condition support.
1513 Patch provided by Phil Pennock.
1515 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
1516 clients. Patch provided by Phil Pennock.
1518 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
1519 Patch provided by Brad "anomie" Jorsch.
1521 NM/23 Bugzilla 687: Fix misparses in eximstats.
1522 Patch provided by Heiko Schlittermann.
1524 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
1525 Patch provided by Heiko Schlittermann.
1527 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
1528 plus update to original patch.
1530 NM/26 Bugzilla 799: Documentation correction for ratelimit.
1532 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
1533 Patch provided by David Brownlee.
1535 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
1537 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
1539 NM/30 Bugzilla 888: TLS documentation bugfixes.
1541 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
1543 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
1544 Unlike the original bugzilla I have changed all shell scripts in src tree.
1546 NM/33 Bugzilla 898: Transport filter timeout fix.
1547 Patch by Todd Rinaldo.
1549 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
1550 Patch by Serge Demonchaux.
1552 NM/35 Bugzilla 39: Base64 decode bug fixes.
1553 Patch by Jakob Hirsch.
1555 NM/36 Bugzilla 909: Correct connect() call in dcc code.
1557 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
1559 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
1561 NM/39 Bugzilla 911: Fixed MakeLinks build script.
1567 TK/01 Add preliminary DKIM support. Currently requires a forked version of
1568 ALT-N's libdkim that I have put here:
1569 http://duncanthrax.net/exim-experimental/
1571 Note to Michael Haardt: I had to rename some vars in sieve.c. They
1572 were called 'true' and it seems that C99 defines that as a reserved
1573 keyword to be used with 'bool' variable types. That means you could
1574 not include C99-style headers which use bools without triggering
1575 build errors in sieve.c.
1577 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
1578 as mailq or other aliases. Changed the --help handling significantly
1579 to do whats expected. exim_usage() emits usage/help information.
1581 SC/01 Added the -bylocaldomain option to eximstats.
1583 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
1585 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
1587 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
1593 PH/01 Another patch from the Sieve maintainer.
1595 PH/02 When an IPv6 address is converted to a string for single-key lookup
1596 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
1597 dots are used instead of colons so that keys in lsearch files need not
1598 contain colons. This was done some time before quoting was made available
1599 in lsearch files. However, iplsearch files do require colons in IPv6 keys
1600 (notated using the quote facility) so as to distinguish them from IPv4
1601 keys. This meant that lookups for IP addresses in host lists did not work
1602 for iplsearch lookups.
1604 This has been fixed by arranging for IPv6 addresses to be expressed with
1605 colons if the lookup type is iplsearch. This is not incompatible, because
1606 previously such lookups could never work.
1608 The situation is now rather anomolous, since one *can* have colons in
1609 ordinary lsearch keys. However, making the change in all cases is
1610 incompatible and would probably break a number of configurations.
1612 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
1615 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
1616 conversion specification without a maximum field width, thereby enabling
1617 a rogue spamd server to cause a buffer overflow. While nobody in their
1618 right mind would setup Exim to query an untrusted spamd server, an
1619 attacker that gains access to a server running spamd could potentially
1620 exploit this vulnerability to run arbitrary code as the Exim user.
1622 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
1623 $primary_hostname instead of what libspf2 thinks the hosts name is.
1625 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
1626 a directory entry by the name of the lookup key. Previously, if a
1627 symlink pointed to a non-existing file or a file in a directory that
1628 Exim lacked permissions to read, a lookup for a key matching that
1629 symlink would fail. Now it is enough that a matching directory entry
1630 exists, symlink or not. (Bugzilla 503.)
1632 PH/03 The body_linecount and body_zerocount variables are now exported in the
1635 PH/04 Added the $dnslist_matched variable.
1637 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
1638 This means they are set thereafter only if the connection becomes
1641 PH/06 Added the client_condition to authenticators so that some can be skipped
1642 by clients under certain conditions.
1644 PH/07 The error message for a badly-placed control=no_multiline_responses left
1645 "_responses" off the end of the name.
1647 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
1649 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
1650 (without spaces) instead of just copying the configuration text.
1652 PH/10 Added the /noupdate option to the ratelimit ACL condition.
1654 PH/11 Added $max_received_linelength.
1656 PH/12 Added +ignore_defer and +include_defer to host lists.
1658 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
1659 way in which PCRE > 7.0 is built.
1661 PH/14 Implemented queue_only_load_latch.
1663 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
1664 MAIL command. The effect was to mangle the value on 64-bit systems.
1666 PH/16 Another patch from the Sieve maintainer.
1668 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
1670 PH/18 If a system quota error occurred while trying to create the file for
1671 a maildir delivery, the message "Mailbox is full" was not appended to the
1672 bounce if the delivery eventually timed out. Change 4.67/27 below applied
1673 only to a quota excession during the actual writing of the file.
1675 PH/19 It seems that peer DN values may contain newlines (and other non-printing
1676 characters?) which causes problems in log lines. The DN values are now
1677 passed through string_printing() before being added to log lines.
1679 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
1680 and InterBase are left for another time.)
1682 PH/21 Added message_body_newlines option.
1684 PH/22 Guard against possible overflow in moan_check_errorcopy().
1686 PH/23 POSIX allows open() to be a macro; guard against that.
1688 PH/24 If the recipient of an error message contained an @ in the local part
1689 (suitably quoted, of course), incorrect values were put in $domain and
1690 $local_part during the evaluation of errors_copy.
1696 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
1697 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
1700 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
1701 issue a MAIL command.
1703 PH/02 In an ACL statement such as
1705 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
1707 if a client was not listed at all, or was listed with a value other than
1708 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
1709 the condition was not true (as it should be), so access was not denied.
1710 The bug was that the ! inversion was incorrectly passed on to the second
1711 item. This has been fixed.
1713 PH/03 Added additional dnslists conditions == and =& which are different from
1714 = and & when the dns lookup returns more than one IP address.
1716 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
1717 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
1719 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
1720 FSYNC, which compiles an option called disable_fsync that allows for
1721 bypassing fsync(). The documentation is heavily laced with warnings.
1723 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
1725 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
1726 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
1727 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
1728 including adding "make clean"; (3) Added -fPIC when compiling the test
1729 dynamically loaded module, to get rid of a warning.
1731 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
1732 message fails, move_frozen_messages = true and ignore_bounce_errors_after
1733 = 0s. The bug is otherwise harmless.
1735 PH/07 There was a bug in the dovecot authenticator such that the value of
1736 $auth1 could be overwritten, and so not correctly preserved, after a
1737 successful authentication. This usually meant that the value preserved by
1738 the server_setid option was incorrect.
1740 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
1742 PH/09 Installed PCRE release 7.0.
1744 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
1745 run for batched SMTP input. It is now run at the start of every message
1746 in the batch. While fixing this I discovered that the process information
1747 (output by running exiwhat) was not always getting set for -bs and -bS
1748 input. This is fixed, and it now also says "batched" for BSMTP.
1750 PH/11 Added control=no_pipelining.
1752 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
1753 patch, slightly modified), and move the expansion of helo_data till after
1754 the connection is made in the smtp transport (so it can use these
1757 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
1759 PH/14 Added log_selector = +pid.
1761 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
1763 PH/16 Add ${if forany and ${if forall.
1765 PH/17 Added dsn_from option to vary the From: line in DSNs.
1767 PH/18 Flush SMTP output before performing a callout, unless control =
1768 no_callout_flush is set.
1770 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
1771 was true (the default) a successful delivery failed to delete the retry
1772 item, thus causing premature timeout of the address. The bug is now
1775 PH/20 Added hosts_avoid_pipelining to the smtp transport.
1777 PH/21 Long custom messages for fakedefer and fakereject are now split up
1778 into multiline reponses in the same way that messages for "deny" and
1779 other ACL rejections are.
1781 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
1782 with slight modification.
1784 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
1785 draft, changing the syntax and factoring some duplicate code".
1787 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
1788 for deliveries of the second and subsequent messages over the same SMTP
1791 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
1792 ${reduce, with only minor "tidies".
1794 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
1796 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
1797 expansion side effects.
1799 PH/27 When a message times out after an over-quota error from an Exim-imposed
1800 quota, the bounce message says "mailbox is full". This message was not
1801 being given when it was a system quota that was exceeded. It now should
1804 MH/03 Made $recipients available in local_scan(). local_scan() already has
1805 better access to the recipient list through recipients_list[], but
1806 $recipients can be useful in postmaster-provided expansion strings.
1808 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
1809 in the case of a MAIL command with additional options following the
1810 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
1811 were accidentally chopped off.
1813 PH/29 SMTP synchronization checks are implemented when a command is read -
1814 there is a check that no more input is waiting when there shouldn't be
1815 any. However, for some commands, a delay in an ACL can mean that it is
1816 some time before the response is written. In this time, more input might
1817 arrive, invalidly. So now there are extra checks after an ACL has run for
1818 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
1819 pipelining has not been advertised.
1821 PH/30 MH's patch to allow iscntrl() characters to be list separators.
1823 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
1824 returned in the SMTP response when smtp_return_error_details was false.
1825 This has been fixed.
1827 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
1828 instead of the C I/O that was originally supplied, because problems were
1829 reported on Solaris.
1831 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
1832 Exim which did not show up earlier: it was assuming that a call to
1833 SSL_CTX_set_info_callback() might give an error value. In fact, there is
1834 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
1835 was a macro that became an assignment, so it seemed to work. This has
1836 changed to a proper function call with a void return, hence the compile
1837 error. Exim's code has been fixed.
1839 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
1842 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
1844 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
1846 PH/37 If a message is not accepted after it has had an id assigned (e.g.
1847 because it turns out to be too big or there is a timeout) there is no
1848 "Completed" line in the log. When some messages of this type were
1849 selected by exigrep, they were listed as "not completed". Others were
1850 picked up by some special patterns. I have improved the selection
1851 criteria to be more general.
1853 PH/38 The host_find_failed option in the manualroute router can now be set
1854 to "ignore", to completely ignore a host whose IP address cannot be
1855 found. If all hosts are ignored, the behaviour is controlled by the new
1856 host_all_ignored option.
1858 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
1859 homing or because of multiple MX records with /mx) generated more than
1860 one IP address, and the following item turned out to be the local host,
1861 all the secondary addresses of the first item were incorrectly removed
1862 from the list, along with the local host and any following hosts (which
1863 is what is supposed to happen).
1865 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
1866 whoever called Exim into the -H file. In the case of the daemon it was
1867 behaving confusingly. When first started, it used values for whoever
1868 started the daemon, but after a SIGHUP it used the Exim user (because it
1869 calls itself on a restart). I have changed the code so that it now always
1872 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
1873 message are rejected with the same error (e.g. no authentication or bad
1874 sender address), and a DATA command is nevertheless sent (as can happen
1875 with PIPELINING or a stupid MUA), the error message that was given to the
1876 RCPT commands is included in the rejection of the DATA command. This is
1877 intended to be helpful for MUAs that show only the final error to their
1880 PH/42 Another patch from the Sieve maintainer.
1882 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
1883 Eximstats - Fixed some broken HTML links and added missing column headers
1885 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
1886 columns for Rejects, Temp Rejects, Ham, and Spam rows.
1888 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
1890 PH/43 Yet another patch from the Sieve maintainer.
1892 PH/44 I found a way to check for a TCP/IP connection going away before sending
1893 the response to the final '.' that terminates a message, but only in the
1894 case where the client has not sent further data following the '.'
1895 (unfortunately, this is allowed). However, in many cases there won't be
1896 any further data because there won't be any more messages to send. A call
1897 to select() can be used: if it shows that the input is "ready", there is
1898 either input waiting, or the socket has been closed. An attempt to read
1899 the next input character can distinguish the two cases. Previously, Exim
1900 would have sent an OK response which the client would never have see.
1901 This could lead to message repetition. This fix should cure that, at
1902 least in a lot of common cases.
1904 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
1905 advertised in response to EHLO.
1911 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
1912 fixed by 4.65/MH/01 (is this a record?) are fixed:
1914 (i) An empty string was always treated as zero by the numeric comparison
1915 operators. This behaviour has been restored.
1917 (ii) It is documented that the numeric comparison operators always treat
1918 their arguments as decimal numbers. This was broken in that numbers
1919 starting with 0 were being interpreted as octal.
1921 While fixing these problems I realized that there was another issue that
1922 hadn't been noticed. Values of message_size_limit (both the global option
1923 and the transport option) were treated as octal if they started with 0.
1924 The documentation was vague. These values are now always treated as
1925 decimal, and I will make that clear in the documentation.
1931 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
1932 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
1935 MH/01 Don't check that the operands of numeric comparison operators are
1936 integers when their expansion is in "skipping" mode (fixes bug
1937 introduced by 4.64-PH/07).
1939 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
1940 child addresses, Exim now panics and dies. Previously, because the count
1941 is held in a short int, deliveries were likely to be lost. As such a
1942 large number of recipients for a single message is ridiculous
1943 (performance will be very, very poor), I have chosen to impose a limit
1944 rather than extend the field.
1950 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
1951 leftover -K file (the existence of which was triggered by #402).
1952 While we were at it, introduced process PID as part of the -K
1953 filename. This should rule out race conditions when creating
1956 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
1957 processing considerably. Previous code took too long for large mails,
1958 triggering a timeout which in turn triggers #401.
1960 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
1961 in the DK code in transports.c. sendfile() is not really portable,
1962 hence the _LINUX specificness.
1964 TF/01 In the add_headers option to the mail command in an Exim filter,
1965 there was a bug that Exim would claim a syntax error in any
1966 header after the first one which had an odd number of characters
1969 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
1970 callout verification, Exim cached a "reject" for the entire domain. This
1971 is correct for most verifications, but it is not correct for a recipient
1972 verification with use_sender or use_postmaster set, because in that case
1973 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
1974 case of MAIL FROM:<> rejection from other early rejections (e.g.
1975 rejection of HELO). When verifying a recipient using a non-null MAIL
1976 address, the cache is ignored if it shows MAIL FROM:<> rejection.
1977 Whatever the result of the callout, the value of the domain cache is
1978 left unchanged (for any other kind of callout, getting as far as trying
1979 RCPT means that the domain itself is ok).
1981 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
1984 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
1985 manifest itself as EPIPE rather than ECONNECT. When tidying away a
1986 session, the daemon ignores ECONNECT errors and logs others; it now
1987 ignores EPIPE as well.
1989 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
1990 (quoted-printable decoding).
1992 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
1993 later the small subsequent patch to fix an introduced bug.
1995 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
1997 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
1999 PH/08 An error is now given if message_size_limit is specified negative.
2001 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
2002 to be given (somewhat) arbitrary names.
2004 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
2007 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
2008 miscellaneous code fixes
2010 PH/10 Added the log_reject_target ACL modifier to specify where to log
2013 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
2014 hostname. This is wrong, because it relates to the incoming message (and
2015 probably the interface on which it is arriving) and not to the outgoing
2016 callout (which could be using a different interface). This has been
2017 changed to use the value of the helo_data option from the smtp transport
2018 instead - this is what is used when a message is actually being sent. If
2019 there is no remote transport (possible with a router that sets up host
2020 addresses), $smtp_active_hostname is used.
2022 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
2023 tweaks were necessary in order to get it to work (see also 21 below):
2024 (a) The code assumed that strncpy() returns a negative number on buffer
2025 overflow, which isn't the case. Replaced with Exim's string_format()
2027 (b) There were several signed/unsigned issues. I just did the minimum
2028 hacking in of casts. There is scope for a larger refactoring.
2029 (c) The code used strcasecmp() which is not a standard C function.
2030 Replaced with Exim's strcmpic() function.
2031 (d) The code set only $1; it now sets $auth1 as well.
2032 (e) A simple test gave the error "authentication client didn't specify
2033 service in request". It would seem that Dovecot has changed its
2034 interface. Fortunately there's a specification; I followed it and
2035 changed what the client sends and it appears to be working now.
2037 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
2040 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
2041 address A is aliased to B and C, where B exists and C does not. Without
2042 -v the output is "A verified" because verification stops after a
2043 successful redirection if more than one address is generated. However,
2044 with -v the child addresses are also verified. Exim was outputting "A
2045 failed to verify" and then showing the successful verification for C,
2046 with its parentage. It now outputs "B failed to verify", showing B's
2047 parentage before showing the successful verification of C.
2049 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
2050 look up a TXT record in a specific list after matching in a combined
2053 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
2054 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
2055 they consult the DNS. I had assumed they would set it the way they
2056 wanted; and indeed my experiments on Linux seem to show that in some
2057 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
2058 To be on the safe side, however, I have now made the interface to
2059 host_find_byname() similar to host_find_bydns(), with an argument
2060 containing the DNS resolver options. The host_find_byname() function now
2061 sets these options at its start, just as host_find_bydns() does. The smtp
2062 transport options dns_qualify_single and dns_search_parents are passed to
2063 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
2064 of host_find_byname() use the default settings of RES_DEFNAMES
2065 (qualify_single) but not RES_DNSRCH (search_parents).
2067 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
2068 spool_read_header() do less string testing, by means of a preliminary
2069 switch on the second character of optional "-foo" lines. (This is
2070 overdue, caused by the large number of possibilities that now exist.
2071 Originally there were few.) While I was there, I also converted the
2072 str(n)cmp tests so they don't re-test the leading "-" and the first
2073 character, in the hope this might squeeze out yet more improvement.
2075 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
2076 flag allowing group syntax was set by the header_syntax check but not
2077 turned off, possible causing trouble later; (2) The flag was not being
2078 set at all for the header_verify test, causing "group"-style headers to
2079 be rejected. I have now set it in this case, and also caused header_
2080 verify to ignore an empty address taken from a group. While doing this, I
2081 came across some other cases where the code for allowing group syntax
2082 while scanning a header line wasn't quite right (mostly, not resetting
2083 the flag correctly in the right place). These bugs could have caused
2084 trouble for malformed header lines. I hope it is now all correct.
2086 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
2087 with the "reply" argument non-NULL. The code, however (which originally
2088 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
2089 but it didn't always do it. This confused somebody who was copying the
2090 code for some other use. I have removed all the tests.
2092 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
2093 feature that was used to support insecure browsers during the U.S. crypto
2094 embargo. It requires special client support, and Exim is probably the
2095 only MTA that supported it -- and would never use it because real RSA is
2096 always available. This code has been removed, because it had the bad
2097 effect of slowing Exim down by computing (never used) parameters for the
2098 RSA_EXPORT functionality.
2100 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
2101 authenticator to fail if there's a tab character in the incoming data
2102 (there should never be unless someone is messing about, as it's supposed
2103 to be base64-encoded). Also added, on Timo's advice, the "secured" option
2104 if the connection is using TLS or if the remote IP is the same as the
2105 local IP, and the "valid-client-cert option" if a client certificate has
2108 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
2109 authenticators. This can be used for authorization after authentication
2110 succeeds. (In the case of plaintext, it servers for both authentication
2113 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
2114 if any retry times were supplied.
2116 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
2117 connection, where there is no client IP address to check. In this
2118 situation, the verify now always succeeds.
2120 PH/25 Applied John Jetmore's -Mset patch.
2122 PH/26 Added -bem to be like -Mset, but loading a message from a file.
2124 PH/27 In a string expansion for a processed (not raw) header when multiple
2125 headers of the same name were present, leading whitespace was being
2126 removed from all of them, but trailing whitespace was being removed only
2127 from the last one. Now trailing whitespace is removed from each header
2128 before concatenation. Completely empty headers in a concatenation (as
2129 before) are ignored.
2131 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
2132 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
2134 PH/29 [Removed. This was a change that I later backed out, and forgot to
2135 correct the ChangeLog entry (that I had efficiently created) before
2136 committing the later change.]
2138 PH/30 Exim was sometimes attempting to deliver messages that had suffered
2139 address errors (4xx response to RCPT) over the same connection as other
2140 messages routed to the same hosts. Such deliveries are always "forced",
2141 so retry times are not inspected. This resulted in far too many retries
2142 for the affected addresses. The effect occurred only when there were more
2143 hosts than the hosts_max_try setting in the smtp transport when it had
2144 the 4xx errors. Those hosts that it had tried were not added to the list
2145 of hosts for which the message was waiting, so if all were tried, there
2146 was no problem. Two fixes have been applied:
2148 (i) If there are any address or message errors in an SMTP delivery, none
2149 of the hosts (tried or untried) are now added to the list of hosts
2150 for which the message is waiting, so the message should not be a
2151 candidate for sending over the same connection that was used for a
2152 successful delivery of some other message. This seems entirely
2153 reasonable: after all the message is NOT "waiting for some host".
2154 This is so "obvious" that I'm not sure why it wasn't done
2155 previously. Hope I haven't missed anything, but it can't do any
2156 harm, as the worst effect is to miss an optimization.
2158 (ii) If, despite (i), such a delivery is accidentally attempted, the
2159 routing retry time is respected, so at least it doesn't keep
2160 hammering the server.
2162 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
2163 in ${readsocket because some servers need this prod.
2165 PH/32 Added some extra debug output when updating a wait-xxx database.
2167 PH/33 The hint "could be header name not terminated by colon", which has been
2168 given for certain expansion errors for a long time, was not being given
2169 for the ${if def:h_colon_omitted{... case.
2171 PH/34 The spec says: "With one important exception, whenever a domain list is
2172 being scanned, $domain contains the subject domain." There was at least
2173 one case where this was not true.
2175 PH/35 The error "getsockname() failed: connection reset by peer" was being
2176 written to the panic log as well as the main log, but it isn't really
2177 panic-worthy as it just means the connection died rather early on. I have
2178 removed the panic log writing for the ECONNRESET error when getsockname()
2181 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
2182 runs only) independently of the message's sender address. This meant
2183 that, if the 4xx error was in fact related to the sender, a different
2184 message to the same recipient with a different sender could confuse
2185 things. In particualar, this can happen when sending to a greylisting
2186 server, but other circumstances could also provoke similar problems.
2187 I have changed the default so that the retry time for these errors is now
2188 based a combination of the sender and recipient addresses. This change
2189 can be overridden by setting address_retry_include_sender=false in the
2192 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
2193 remote server are returned as part of bounce messages. This was not
2194 happening for LMTP over a pipe (the lmtp transport), but now it is the
2195 same for both kinds of LMTP.
2197 PH/38 Despite being documented as not happening, Exim was rewriting addresses
2198 in header lines that were in fact CNAMEs. This is no longer the case.
2200 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
2201 and queue runs started by the daemon processed all messages. This has
2202 been fixed so that -R and -S can now usefully be given with -q<time>.
2204 PH/40 Import PCRE release 6.7 (fixes some bugs).
2206 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
2208 PH/42 Give an error if -q is specified more than once.
2210 PH/43 Renamed the variables $interface_address and $interface_port as
2211 $received_ip_address and $received_port, to make it clear that these
2212 values apply to message reception, and not to the outgoing interface when
2213 a message is delivered. (The old names remain recognized, of course.)
2215 PH/44 There was no timeout on the connect() call when using a Unix domain
2216 socket in the ${readsocket expansion. There now is.
2218 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
2219 be meaningful with "accept".
2221 SC/01 Eximstats V1.43
2222 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
2224 SC/02 Eximstats V1.44
2225 Use a glob alias rather than an array ref in the generated
2226 parser. This improves both readability and performance.
2228 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
2229 Collect SpamAssassin and rejection statistics.
2230 Don't display local sender or destination tables unless
2231 there is data to show.
2232 Added average volumes into the top table text output.
2234 SC/04 Eximstats V1.46
2235 Collect data on the number of addresses (recipients)
2236 as well as the number of messages.
2238 SC/05 Eximstats V1.47
2239 Added 'Message too big' to the list of mail rejection
2240 reasons (thanks to Marco Gaiarin).
2242 SC/06 Eximstats V1.48
2243 Mainlog lines which have GMT offsets and are too short to
2244 have a flag are now skipped.
2246 SC/07 Eximstats V1.49 (Alain Williams)
2247 Added the -emptyok flag.
2249 SC/08 Eximstats V1.50
2250 Fixes for obtaining the IP address from reject messages.
2252 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
2253 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
2254 whitesspace changes from 4.64-PH/27
2256 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
2259 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
2260 are found, allow negative numbers in numeric criteria)
2262 JJ/06 exipick.20061117.2, added new $message_body_missing variable
2264 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
2265 to match changes made in 4.64-PH/43
2267 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
2269 PH/47 Put in an explicit test for a DNS lookup of an address record where the
2270 "domain" is actually an IP address, and force a failure. This locks out
2271 those revolvers/nameservers that support "A-for-A" lookups, in
2272 contravention of the specifications.
2274 PH/48 When a host name was looked up from an IP address, and the subsequent
2275 forward lookup of the name timed out, the host name was left in
2276 $sender_host_name, contrary to the specification.
2278 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
2279 restricted to single-key lookups, Exim was not diagnosing an error if
2280 * or *@ was used with a query-style lookup.
2282 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
2284 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
2285 long ago, but noone interested enough thought of it. Let's just say that
2286 the "1.1" means that there are some new functions that weren't there at
2287 some point in the past.
2289 PH/51 Error processing for expansion failure of helo_data from an smtp
2290 transport during callout processing was broken.
2292 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
2293 tested/used via the -bh/-bhc/-bs options.
2295 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
2296 bug, fixed in subsequent PCRE releases).
2298 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
2299 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
2301 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
2307 SC/01 Use a glob alias rather than an array ref in eximstats generated
2308 parser. This improves both readability and performance.
2310 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
2311 Don't display local sender or destination tables in eximstats unless
2312 there is data to show.
2313 Added average volumes into the eximstats top table text output.
2315 SC/03 Collect data on the number of addresses (recipients) as well
2316 as the number of messages in eximstats.
2318 TF/01 Correct an error in the documentation for the redirect router. Exim
2319 does (usually) call initgroups() when daemonizing.
2321 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
2322 with consistent privilege compared to when running as a daemon.
2324 TF/03 Note in the spec that $authenticated_id is not set for local
2325 submissions from trusted users.
2327 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
2328 Thanks to Dean Brooks <dean@iglou.com> for the patch.
2330 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
2331 by adding some example configuration directives to the default
2332 configuration file. A little bit of work is required to uncomment the
2333 directives and define how usernames and passwords are checked, but
2334 there is now a framework to start from.
2336 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
2337 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
2338 without this. I don't know how relevant this is to other LDAP libraries.
2340 PH/02 Add the verb name to the "unknown ACL verb" error.
2342 PH/03 Magnus Holmgren's patch for filter_prepend_home.
2344 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
2346 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
2347 directory not expanded when it should be if an expanded home directory
2348 was set for the address (which is overridden by the transport).
2350 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
2353 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
2354 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
2355 because it is too late at that time, and has no effect.
2357 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
2358 security issue with \' (bugzilla #107). I could not use the
2359 PQescapeStringConn() function, because it needs a PGconn value as one of
2362 PH/08 When testing addresses using -bt, indicate those final addresses that
2363 are duplicates that would not cause an additional delivery. At least one
2364 person was confused, thinking that -bt output corresponded to deliveries.
2365 (Suppressing duplicates isn't a good idea as you lose the information
2366 about possibly different redirections that led to the duplicates.)
2368 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
2369 systems where poll() doesn't work, in particular OS X.
2371 PH/10 Added more information to debugging output for retry time not reached.
2373 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
2374 operations in malware.c.
2376 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
2379 PH/13 If write_rejectlog was set false when logging was sent to syslog with
2380 syslog_duplication set false, log lines that would normally be written
2381 both the the main log and to the reject log were not written to syslog at
2384 PH/14 In the default configuration, change the use of "message" in ACL warn
2385 statements to "add_header".
2387 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
2388 not followed by a command (e.g. "seen endif").
2390 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
2391 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
2394 PH/17 Added extra conditions to the default value of delay_warning_condition
2398 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
2399 { match{$h_precedence:}{(?i)bulk|list|junk} } \
2400 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
2403 The Auto-Submitted: and various List- headers are standardised, whereas I
2404 don't think Precedence: ever was.
2406 PH/18 Refactored debugging code in route_finduser() to show more information,
2407 in particular, the error code if getpwnam() issues one.
2409 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
2410 This is apparently needed in addition to the PH/07 change above to avoid
2411 any possible encoding problems.
2413 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
2414 but not after initializing Perl.
2416 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
2417 output them only if debugging. By default they are written stderr,
2418 apparently, which is not desirable.
2420 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
2423 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
2426 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
2428 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
2429 authenticated or an ident call has been made. Suppress the default
2430 values for $authenticated_id and $authenticated_sender (but permit -oMai
2431 and -oMas) when testing with -bh.
2433 PH/24 Re-jigged the order of the tests in the default configuration so that the
2434 tests for valid domains and recipients precede the DNS black list and CSA
2435 tests, on the grounds that those ones are more expensive.
2437 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
2438 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
2439 This bug exists in every version of Exim that I still have, right back to
2442 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
2443 However, an attempt to turn on case-sensitivity in a regex key by
2444 including (?-i) didn't work because the subject string was already
2445 lowercased, and the effects were non-intuitive. It turns out that a
2446 one-line patch can be used to allow (?-i) to work as expected.
2452 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
2453 other effects) broke the use of negated acl sub-conditions.
2455 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
2458 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
2459 "Deny" causes Exim to reject the incoming connection with a 554 error.
2460 Unfortunately, if there is a major crisis, such as a disk failure,
2461 tcp-wrappers gives "deny", whereas what one would like would be some
2462 kind of temporary error. A kludge has been added to help with this.
2463 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
2464 554 error is used if errno is still zero or contains ENOENT (which occurs
2465 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
2468 PH/03 Add -lutil to the default FreeBSD LIBS setting.
2470 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
2471 errors. Otherwise a message that provokes a temporary error (when other
2472 messages do not) can cause a whole host to time out.
2474 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
2475 addresses were routed directly to files or pipes from a redirect router.
2476 File deliveries just didn't batch; pipe deliveries might have suffered
2479 PH/06 A failure to get a lock for a hints database would erroneously always say
2480 "Failed to get write lock", even when it was really a read lock.
2482 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
2483 of 0600. This has been changed to use the value of the lockfile_mode
2484 option (which defaults to 0600).
2486 PH/08 Applied small patch from the Sieve maintainer.
2488 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
2489 folder from quota calculations, a direct delivery into this folder messed
2490 up the contents of the maildirsize file. This was because the regex was
2491 used only to exclude .Trash (or whatever) when the size of the mailbox
2492 was calculated. There was no check that a delivery was happening into an
2493 excluded directory. This bug has been fixed by ignoring all quota
2494 processing for deliveries into excluded directories.
2496 PH/10 Added the maildirfolder_create_regex option to appendfile.
2502 PH/01 The code for finding all the local interface addresses on a FreeBSD
2503 system running IPv6 was broken. This may well have applied to all BSD
2504 systems, as well as to others that have similar system calls. The broken
2505 code found IPv4 interfaces correctly, but gave incorrect values for the
2506 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
2507 that it would not match correctly against @[] and not recognize the IPv6
2510 PH/02 The ipliteral router was not recognizing addresses of the form user@
2511 [ipv6:....] because it didn't know about the "ipv6:" prefix.
2513 PH/03 Added disable_ipv6.
2515 PH/04 Changed $reply_address to use the raw form of the headers instead of the
2516 decoded form, because it is most often used to construct To: headers
2517 lines in autoreplies, and the decoded form may well be syntactically
2518 invalid. However, $reply_address has leading white space removed, and all
2519 newlines turned into spaces so that the autoreply transport does not
2522 PH/05 If group was specified without a user on a router, and no group or user
2523 was specified on a transport, the group from the router was ignored.
2525 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
2526 for visible compile-time settings that can be used to change these
2527 numbers, for those that want even more. Backwards compatibility with old
2528 spool files has been maintained. However, going back to a previous Exim
2529 release will lost any variables that are in spool files.
2531 PH/07 Two small changes when running in the test harness: increase delay when
2532 passing a TCP/IP connection to a new process, in case the original
2533 process has to generate a bounce, and remove special handling of
2534 127.0.0.2 (sic), which is no longer necessary.
2536 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
2537 be the same on different OS.
2539 PH/09 Moved a debug statement in filter processing to avoid a race problem when
2542 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
2543 whether --show-vars was specified or not
2545 JJ/02 exipick: Added support for new ACL variable spool format introduced
2548 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
2549 syntactically invalid From: or Reply-to: line, and a filter used this to
2550 generate an autoreply, and therefore failed to obtain an address for the
2551 autoreply, Exim could try to deliver to a non-existent relative file
2552 name, causing unrelated and misleading errors. What now happens is that
2553 it logs this as a hard delivery error, but does not attempt to create a
2556 PH/11 The exinext utility has a -C option for testing purposes, but although
2557 the given file was scanned by exinext itself; it wasn't being passed on
2558 when Exim was called.
2560 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
2561 an end-of-file indication when reading a command response.
2563 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
2564 compiled. In many other places in Exim, IPv6 addresses are always
2565 recognized, so I have changed this. It also means that IPv4 domain
2566 literals of the form [IPV4:n.n.n.n] are now always recognized.
2568 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
2569 used if the router is not running as root, for example, when verifying at
2570 ACL time, or when using -bh. The debugging output from this situation was
2571 non-existent - all you got was a failure to exec. I have made two
2574 (a) Failures to set uid/gid, the current directory, or a process leader
2575 in a subprocess such as that created by queryprogram now generate
2576 suitable debugging ouput when -d is set.
2578 (b) The queryprogram router detects when it is not running as root,
2579 outputs suitable debugging information if -d is set, and then runs
2580 the subprocess without attempting to change uid/gid.
2582 PH/15 Minor change to Makefile for building test_host (undocumented testing
2585 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
2586 additional section of a DNS packet that returns MX or SRV records.
2587 Instead, it always explicitly searches for A/AAAA records. This avoids
2588 major problems that occur when a DNS server includes only records of one
2589 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
2590 fixed another bug: if SRV records were looked up and the corresponding
2591 address records were *not* found in the additional section, the port
2592 values from the SRV records were lost.
2594 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
2595 using the correct key (the original address) when searching the retry
2596 rules in order to find which one to use for generating the retry hint.
2598 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
2599 adding the default one. Similarly, if it contains a Reply-To: header, the
2600 errors_reply_to option, if set, is not used.
2602 PH/19 When calculating a retry time, Exim used to measure the "time since
2603 failure" by looking at the "first failed" field in the retry record. Now
2604 it does not use this if it is later than than the arrival time of the
2605 message. Instead it uses the arrival time. This makes for better
2606 behaviour in cases where some deliveries succeed, thus re-setting the
2607 "first failed" field. An example is a quota failure for a huge message
2608 when small messages continue to be delivered. Without this change, the
2609 "time since failure" will always be short, possible causing more frequent
2610 delivery attempts for the huge message than are intended.
2611 [Note: This change was subsequently modified - see PH/04 for 4.62.]
2613 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
2614 $1, $2, $3) because the numerical variables can be reset during some
2615 expansion items (e.g. "match"), thereby losing the authentication data.
2617 PH/21 Make -bV show the size of off_t variables so that the test suite can
2618 decide whether to run tests for quotas > 2G.
2620 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
2621 mailbox_size, and mailbox_filecount in the appendfile transport. If a
2622 filecount value is greater than 2G or if a quota value is greater than 2G
2623 on a system where the size of off_t is not greater than 4, a panic error
2626 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
2627 never match. The debug and -bh output now contains an explicit error
2628 message indicating a malformed IPv4 address or mask.
2630 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
2631 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
2632 PH/23 above applies.
2634 PH/25 Do not write to syslog when running in the test harness. The only
2635 occasion when this arises is a failure to open the main or panic logs
2636 (for which there is an explicit test).
2638 PH/26 Added the /no_tell option to "control=freeze".
2640 PH/27 If a host name lookup failed very early in a connection, for example, if
2641 the IP address matched host_lookup and the reverse lookup yielded a name
2642 that did not have a forward lookup, an error message of the form "no IP
2643 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
2644 could be logged. Now it outputs the IP address instead of "NULL".
2646 PH/28 An enabling patch from MH: add new function child_open_exim2() which
2647 allows the sender and the authenticated sender to be set when
2648 submitting a message from within Exim. Since child_open_exim() is
2649 documented for local_scan(), the new function should be too.
2651 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
2652 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
2653 results in an empty string is now treated as unset.
2655 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
2657 PH/31 Added sender_verify_fail logging option.
2659 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
2660 needed by RFC 822 but not by RFC 2822 was commented out. I have now
2661 tidied the source and removed it altogether.
2663 PH/33 When a queue run was abandoned because the load average was too high, a
2664 log line was always written; now it is written only if the queue_run log
2665 selector is set. In addition, the log line for abandonment now contains
2666 information about the queue run such as the pid. This is always present
2667 in "start" and "stop" lines but was omitted from the "abandon" line.
2669 PH/34 Omit spaces between a header name and the colon in the error message that
2670 is given when verify = headers_syntax fails (if there are lots of them,
2671 the message gets confusing).
2673 PH/35 Change the default for dns_check_names_pattern to allow slashes within
2674 names, as there are now some PTR records that contain slashes. This check
2675 is only to protect against broken name servers that fall over on strange
2676 characters, so the fact that it applies to all lookups doesn't matter.
2678 PH/36 Now that the new test suite is complete, we can remove some of the
2679 special code in Exim that was needed for the old test suite. For example,
2680 sorting DNS records because real resolvers return them in an arbitrary
2681 order. The new test suite's fake resolver always returns records in the
2684 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
2685 bounces) except when queue_only is set, to avoid logging races between
2686 the different processes.
2688 PH/38 Panic-die if .include specifies a non-absolute path.
2690 PH/39 A tweak to the "H" retry rule from its user.
2692 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
2693 a label. They prevented compilation on older perls.
2695 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
2696 a warning to be raised on newish perls.
2698 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
2699 on queue. Changes to match documented behaviour of showing count of
2700 messages matching specified criteria.
2702 PH/40 Changed the default ident timeout from 30s to 5s.
2704 PH/41 Added support for the use of login_cap features, on those BSD systems
2705 that have them, for controlling the resources used by pipe deliveries.
2707 PH/42 The content-scanning code uses fopen() to create files in which to put
2708 message data. Previously it was not paying any attention to the mode of
2709 the files. Exim runs with umask(0) because the rest of the code creates
2710 files with open(), and sets the required mode explicitly. Thus, these
2711 files were ending up world-writeable. This was not a big issue, because,
2712 being within the spool directory, they were not world-accessible. I have
2713 created a function called modefopen, which takes an additional mode
2714 argument. It sets umask(777), creates the file, chmods it to the required
2715 mode, then resets the umask. All the relevant calls to fopen() in the
2716 content scanning code have been changed to use this function.
2718 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
2719 to 24 hours. This avoids potential overflow problems when processing G
2720 and H retry rules. I suspect nobody ever tinkers with this value.
2722 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
2724 PH/45 When the plaintext authenticator is running as a client, the server's
2725 challenges are checked to ensure they are valid base64 strings. By
2726 default, the authentication attempt is cancelled if an invalid string is
2727 received. Setting client_ignore_invalid_base64 true ignores these errors.
2728 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
2729 they are received. Thus, the responses can be made to depend on the
2730 challenges. If an invalid string is ignored, an empty string is placed in
2733 PH/46 Messages that are created by the autoreply transport now contains a
2734 References: header, in accordance with RFCs 2822 and 3834.
2736 PH/47 Added authenticated_sender_force to the smtp transport.
2738 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
2740 PH/49 Installed latest patch from the Sieve maintainer.
2742 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
2743 was also set, the appendfile transport was unnecessarily scanning a
2744 directory of message files (e.g. for maildir delivery) to find the count
2745 of files (along with the size), even though it did not need this
2746 information. It now does the scan only if it needs to find either the
2747 size of the count of files.
2749 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
2751 PH/52 Two bugs concerned with error handling when the smtp transport is
2754 (i) Exim was not creating retry information for temporary errors given
2755 for individual recipients after the DATA command when the smtp transport
2756 was used in LMTP mode. This meant that they could be retried too
2757 frequently, and not timed out correctly.
2759 (ii) Exim was setting the flag that allows error details to be returned
2760 for LMTP errors on RCPT commands, but not for LMTP errors for individual
2761 recipients that were returned after the DATA command.
2763 PH/53 This is related to PH/52, but is more general: for any failing address,
2764 when detailed error information was permitted to be returned to the
2765 sender, but the error was temporary, then after the final timeout, only
2766 "retry timeout exceeded" was returned. Now it returns the full error as
2767 well as "retry timeout exceeded".
2769 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
2770 do this, and (what is worse) MTAs that accept it.
2772 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
2773 will now be deprecated.
2775 PH/56 New os.c-cygwin from the Cygwin maintainer.
2777 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
2778 formats (previously only available in exim formats via -bpr, -bpru,
2779 and -bpra. Now also available in native and exiqgrep formats)
2781 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
2782 with very large, slow to parse queues
2784 JJ/08 exipick: added ! as generic prefix to negate any criteria format
2786 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
2788 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
2789 responses to authentication challenges, though it was showing the
2790 challenges; (ii) I've removed the CR characters from the debug output for
2793 PH/58 Allow for the insertion of a newline as well as a space when a string
2794 is turned into more than one encoded-word during RFC 2047 encoding. The
2795 Sieve code now uses this.
2797 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
2798 data_4xx, lost_connection, tls_required.
2800 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
2801 message was being sent as an SMTP response.
2803 PH/61 Add -l and -k options to exicyclog.
2805 PH/62 When verifying, if an address was redirected to one new address, so that
2806 verification continued, and the new address failed or deferred after
2807 having set something in $address_data, the value of $address_data was not
2808 passed back to the ACL. This was different to the case when no
2809 redirection occurred. The value is now passed back in both cases.
2811 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
2812 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
2813 use login_cap.h, so on its own it isn't the distinguishing feature. The
2814 new name refers directly to the setclassresources() function.
2816 PH/65 Added configuration files for NetBSD3.
2818 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
2820 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
2821 is preferred over IPv4.
2823 PH/68 The bounce_return_message and bounce_return_body options were not being
2824 honoured for bounces generated during the reception of non-SMTP messages.
2825 In particular, this applied to messages rejected by the ACL. This bug has
2826 been fixed. However, if bounce_return_message is true and bounce_return_
2827 body is false, the headers that are returned for a non-SMTP message
2828 include only those that have been read before the error was detected.
2829 (In the case of an ACL rejection, they have all been read.)
2831 PH/69 The HTML version of the specification is now built in a directory called
2832 spec_html instead of spec.html, because the latter looks like a path with
2833 a MIME-type, and this confuses some software.
2835 PH/70 Catch two compiler warnings in sieve.c.
2837 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
2838 function verify_get_ident() calls ip_connect() to connect a socket, but
2839 if the "connect()" function timed out, ip_connect() used to close the
2840 socket. However, verify_get_ident() also closes the socket later, and in
2841 between Exim writes to the log, which may get opened at this point. When
2842 the socket was closed in ip_connect(), the log could get the same file
2843 descriptor number as the socket. This naturally causes chaos. The fix is
2844 not to close the socket in ip_connect(); the socket should be closed by
2845 the function that creates it. There was only one place in the code where
2846 this was missing, in the iplookup router, which I don't think anybody now
2847 uses, but I've fixed it anyway.
2849 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
2850 well as to direct DNS lookups. Otherwise the handling of names in host
2851 lists is inconsistent and therefore confusing.
2857 PH/01 Two changes to the default runtime configuration:
2859 (1) Move the checks for relay_from_hosts and authenticated clients from
2860 after to before the (commented out) DNS black list checks.
2862 (2) Add control=submission to the relay_from_hosts and authenticated
2863 clients checks, on the grounds that messages accepted by these
2864 statements are most likely to be submissions.
2866 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
2868 (1) Generate an error if the third argument for the ${prvs expansion is
2871 (2) Treat a missing third argument of ${prvscheck as if it were an empty
2874 (3) Reset the variables that are obtained from the first argument of
2875 ${prvscheck and used in the second argument before leaving the code,
2876 because their memory is reclaimed, so using them afterwards may do
2879 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
2880 one (it's much easier than Tom thought :-).
2882 (5) Because of (4), we can now allow for the use of $prvscheck_result
2883 inside the third argument.
2885 PH/03 For some reason, the default setting of PATH when running a command from
2886 a pipe transport was just "/usr/bin". I have changed it to
2889 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
2890 anything to be listed in the output from -bV.
2892 PH/05 When a filter generated an autoreply, the entire To: header line was
2893 quoted in the delivery log line, like this:
2895 => >A.N.Other <ano@some.domain> <original@ddress> ...
2897 This has been changed so that it extracts the operative address. There
2898 may be more than one such address. If so, they are comma-separated, like
2901 => >ano@some.domain,ona@other.domain <original@ddress> ...
2903 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
2904 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
2905 not being looked up in the rDNS to get a host name, Exim was showing the
2906 IP address twice in Received: lines, even though the IP addresses were
2907 identical. For example:
2909 Received: from [1.2.3.4] (helo=[1.2.3.4])
2911 However, if the real host name was known, it was omitting the HELO data
2912 if it matched the actual IP address. This has been tidied up so that it
2913 doesn't show the same IP address twice.
2915 PH/07 When both +timestamp and +memory debugging was on, the value given by
2916 $tod_xxx expansions could be wrong, because the tod_stamp() function was
2917 called by the debug printing, thereby overwriting the timestamp buffer.
2918 Debugging no longer uses the tod_stamp() function when +timestamp is set.
2920 PH/08 When the original message was included in an autoreply transport, it
2921 always said "this is a copy of the message, including all the headers",
2922 even if body_only or headers_only was set. It now gives an appropriate
2925 PH/09 Applied a patch from the Sieve maintainer which:
2927 o fixes some comments
2928 o adds the (disabled) notify extension core
2929 o adds some debug output for the result of if/elsif tests
2930 o points to the current vacation draft in the documentation
2931 and documents the missing references header update
2935 o fixes a bug in processing the envelope test (when testing
2936 multiple envelope elements, the last element determinted the
2939 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
2940 Electronic Mail") by including:
2942 Auto-submitted: auto-generated
2944 in the messages that it generates (bounce messages and others, such as
2945 warnings). In the case of bounce messages for non-SMTP mesages, there was
2946 also a typo: it was using "Auto_submitted" (underscore instead of
2947 hyphen). Since every message generated by Exim is necessarily in response
2948 to another message, thes have all been changed to:
2950 Auto-Submitted: auto-replied
2952 in accordance with these statements in the RFC:
2954 The auto-replied keyword:
2956 - SHOULD be used on messages sent in direct response to another
2957 message by an automatic process,
2959 - MUST NOT be used on manually-generated messages,
2961 - MAY be used on Delivery Status Notifications (DSNs) and Message
2962 Disposition Notifications (MDNs),
2964 - MUST NOT be used on messages generated by automatic or periodic
2965 processes, except for messages which are automatic responses to
2968 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
2969 to the default Received: header definition.
2971 PH/12 Added log selector acl_warn_skipped (default on).
2973 PH/13 After a successful wildlsearch lookup, discard the values of numeric
2974 variables because (a) they are in the wrong storage pool and (b) even if
2975 they were copied, it wouldn't work properly because of the caching.
2977 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
2978 checking when decoding. Apparently there are clients that generate
2979 overlong encoded strings. Why am I not surprised?
2981 PH/15 If the first argument of "${if match_address" was not empty, but did not
2982 contain an "@" character, Exim crashed. Now it writes a panic log message
2983 and treats the condition as false.
2985 PH/16 In autoreply, treat an empty string for "once" the same as unset.
2987 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
2988 extension "envelope-auth". The code is finished and in agreement with
2989 other implementations, but there is no documentation so far and in fact,
2990 nobody wrote the draft yet. This extension is currently #undef'ed, thus
2991 not changing the active code.
2993 Print executed "if" and "elsif" statements when debugging is used. This
2994 helps a great deal to understand what a filter does.
2996 Document more things not specified clearly in RFC3028. I had all this
2997 sorted out, when out of a sudden new issues came to my mind. Oops."
2999 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
3002 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
3003 canonical form (as documented). However, after a host name lookup from
3004 the IP address, check_host() was doing a simple string comparison with
3005 addresses acquired from the DNS when checking that the found name did
3006 have the original IP as one of its addresses. Since any found IPv6
3007 addresses are likely to be in abbreviated form, the comparison could
3008 fail. Luckily, there already exists a function for doing the comparison
3009 by converting both addresses to binary, so now that is used instead of
3010 the text comparison.
3012 PH/20 There was another similar case to PH/19, when a complete host name was
3013 given in a host list; looking up its IP address could give an abbreviated
3014 form, whereas the current host's name might or might not be abbreviated.
3015 The same fix has been applied.
3021 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
3022 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
3025 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
3026 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
3028 PH/03 Typo: missing ".o" in src/pcre/Makefile.
3030 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
3031 header line, restrict the check to what is listed in RFCs 2369 and 2929.
3032 Also, for "Auto-Submitted", treat anything other than "no" as
3033 non-personal, in accordance with RFC 3834. (Previously it treated
3034 anything starting "auto-" as non-personal.)
3036 TF/01 The control=submission/name=... option had a problem with syntax
3037 errors if the name included a slash character. The /name= option
3038 now slurps the rest of the string, so it can include any characters
3039 but it must come last in the list of options (after /sender_retain
3042 PH/05 Some modifications to the interface to the fake nameserver for the new
3050 TK/01 Added the "success_on_redirect" address verification option. See
3051 NewStuff for rationale and an example.
3053 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
3055 PH/02 Patch to exigrep to allow it to work on syslog lines.
3057 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
3058 fread() to skip over the body file's header line, because in Cygwin the
3059 header line is locked and is inaccessible.
3061 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
3062 co-exist for some time) to make it clear that it is the Exim ID that is
3063 referenced, not the Message-ID: header line.
3065 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
3066 string_format() function, because snprintf() does not exist on all
3069 PH/06 The use of forbid_filter_existstest now also locks out the use of the
3070 ${stat: expansion item.
3072 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
3073 protocol synchronization error", to keep the pedants happy.
3075 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
3076 well as for IRIX systems, when gcc is being used. See the host.c source
3079 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
3081 PH/10 Named domain lists were not working if used in a queue_smtp_domains
3084 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
3085 transport and to the smtp transport in LMTP mode.
3087 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
3089 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
3090 run a filter in a subprocess. This could lead to confusion in subsequent
3091 lookups in the parent process. There should also be a search_tidyup() at
3092 the end of the subprocess.
3094 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
3095 only if the host matched helo_try_verify_hosts, which caused the
3096 verification to occur when the EHLO/HELO command was issued. The ACL just
3097 tested the remembered result. Now, if a previous verification attempt has
3098 not happened, "verify = helo" does it there and then.
3100 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
3102 TK/03 Fix log output including CR from clamd.
3104 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
3105 exist provoked a memory error which could cause a segfault.
3107 PH/15 Installed PCRE 6.2
3109 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
3111 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
3112 of the problem. Specifically, suggested +O2 rather than +O1 for the
3115 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
3117 PH/20 If a delivery was routed to a non-standard port by means of an SRV
3118 record, the port was not correctly logged when the outgoing_port log
3119 selector was set (it logged the transort's default port).
3121 PH/21 Added support for host-specific ports to manualroute, queryprogram,
3122 fallback_hosts, and "hosts" in the smtp transport.
3124 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
3125 host errors such as "Connection refused".
3127 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
3128 authentication with radiusclient 0.4.9:
3130 - Error returned from rc_read_config was caught wrongly
3131 - Username/password not passed on to radius server due to wrong length.
3133 The presumption is that some radiusclient API changes for 4.51/PH/17
3134 were not taken care of correctly. The code is still untested by me (my
3135 Linux distribution still has 0.3.2 of radiusclient), but it was
3136 contributed by a Radius user.
3138 PH/24 When doing a callout, the value of $domain wasn't set correctly when
3139 expanding the "port" option of the smtp transport.
3141 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
3142 while reading a MIME header. Thanks to Tom Hughes for a patch.
3144 PH/24 Include config.h inside local_scan.h so that configuration settings are
3147 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
3148 that in an ACL for RCPT (for example), you can examine exactly what was
3151 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
3152 commands, but it was not correctly comparing the address with the actual
3153 client host address. Thus, it would show the EHLO address in Received:
3154 header lines when this was not necessary.
3156 PH/27 Added the % operator to ${eval:}.
3158 PH/28 Exim tries to create and chdir to its spool directory when it starts;
3159 it should be ignoring failures (because with -C, for example, it has lost
3160 privilege). It wasn't ignoring creation failures other than "already
3163 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
3166 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
3167 because an input line was too long, either on its own, or by virtue of
3168 too many continuations, the temporary file was not being removed, and the
3169 return code was incorrect.
3171 PH/31 Missing "BOOL" in function definition in filtertest.c.
3173 PH/32 Applied Sieve patches from the maintainer.
3175 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
3177 PH/33 Added "verify = not_blind".
3179 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
3180 Local/Makefile (with some defaults set). These are used in built scripts
3181 such as exicyclog, but they have never been used in the exim_install
3182 script (though there are many overriding facilities there). I have
3183 arranged that the exim_install script now takes note of these two
3186 PH/35 Installed configuration files for Dragonfly.
3188 PH/36 When a locally submitted message by a trusted user did not contain a
3189 From: header, and the sender address was obtained from -f or from an SMTP
3190 MAIL command, and the trusted user did not use -F to supply a sender
3191 name, $originator_name was incorrectly used when constructing a From:
3192 header. Furthermore, $originator_name was used for submission mode
3193 messages from external hosts without From: headers in a similar way,
3194 which is clearly wrong.
3196 PH/37 Added control=suppress_local_fixups.
3198 PH/38 When log_selector = +received_sender was set, and the addition of the
3199 sender made the log line's construction buffer exactly full, or one byte
3200 less than full, an overflow happened when the terminating "\n" was
3203 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
3204 when the result of a list match is failure because a DNS lookup failed.
3206 PH/40 RM_COMMAND is now used in the building process.
3208 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
3209 the "build-* directories that it finds.
3211 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
3212 address in a domain literal was a prefix of an interface address.
3214 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
3215 when verifying a sender address, unless rewrite_headers is false.
3217 PH/44 Wrote a long comment about why errors_to addresses are verified as
3218 recipients, not senders.
3220 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
3221 the ratelimit ACL was added.
3223 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
3225 PH/46 Added extra information about PostgreSQL errors to the error string.
3227 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
3228 avoiding the need to install special zones in a real server. This is
3229 backwards compatible; if it can't find the fake resolver, it drops back.
3230 Thus, both old and new test suites can be run.
3232 TF/02 Added util/ratelimit.pl
3234 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
3235 clock is set back in time.
3237 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
3238 Candler <B.Candler@pobox.com>.
3240 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
3241 OK for addresses that are the result of redirections.
3243 PH/48 A number of further additions for the benefit of the new test suite,
3244 including a fake gethostbyname() that interfaces to the fake DNS resolver
3247 TF/06 The fix for widen_domains has also been applied to qualify_single and
3248 search_parents which are the other dnslookup options that can cause
3251 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
3254 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
3256 TF/07 Exim produced the error message "an SRV record indicated no SMTP
3257 service" if it encountered an MX record with an empty target hostname.
3258 The message is now "an MX or SRV record indicated no SMTP service".
3260 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
3261 if the DNS of the sending site is misconfigured. This is quite a
3262 common situation. This change restores the behaviour of treating a
3263 helo verification defer as a failure.
3265 PH/51 If self=fail was set on a router, the bounce message did not include the
3266 actual error message.
3272 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
3274 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
3275 command itself ended in error, the underlying message about the transport
3276 filter timeout was being overwritten with the pipe command error. Now the
3277 underlying error message should be appended to the second error message.
3279 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
3281 PH/02 Reduce the amount of output that "make" produces by default. Full output
3282 can still be requested.
3284 PH/03 The warning log line about a condition test deferring for a "warn" verb
3285 was being output only once per connection, rather than after each
3286 occurrence (because it was using the same function as for successful
3287 "warn" verbs). This seems wrong, so I have changed it.
3289 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
3290 it should not have, which might have caused a crash in the right
3291 circumstances, but probably never did.
3293 PH/04 Installed a modified version of Tony Finch's patch to make submission
3294 mode fix the return path as well as the Sender: header line, and to
3295 add a /name= option so that you can make the user's friendly name appear
3298 TF/03 Added the control = fakedefer ACL modifier.
3300 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
3301 Mark Lowes for thorough testing.
3303 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
3305 TK/03 Merged latest SRS patch from Miles Wilton.
3307 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
3308 with the definition in sysexits.h (which is #included earlier).
3309 Fortunately, Exim does not actually use EX_OK. The code used to try to
3310 preserve the sysexits.h value, by assumimg that macro definitions were
3311 scanned for macro replacements. I have been disabused of this notion,
3312 so now the code just undefines EX_OK before #including unistd.h.
3314 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
3315 in the smtp transport. When a block could not be written in a single
3316 write() function, the timeout was being re-applied to each part-write.
3317 This seems wrong - if the receiver was accepting one byte at a time it
3318 would take for ever. The timeout is now adjusted when this happens. It
3319 doesn't have to be particularly precise.
3321 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
3322 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
3324 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
3325 without a domain if the check to <postmaster@domain> fails.
3327 SC/01 Eximstats: added -xls and the ability to specify output files
3328 (patch written by Frank Heydlauf).
3330 SC/02 Eximstats: use FileHandles for outputing results.
3332 SC/03 Eximstats: allow any combination of xls, txt, and html output.
3334 SC/04 Eximstats: fixed display of large numbers with -nvr option
3336 SC/05 Eximstats: fixed merging of reports with empty tables.
3338 SC/06 Eximstats: added the -include_original_destination flag
3340 SC/07 Eximstats: removed tabs and trailing whitespace.
3342 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
3344 TK/06 MBOX spool code: Add real "From " MBOX separator line
3345 so the .eml file is really in mbox format (even though
3346 most programs do not really care). Patch from Alex Miller.
3348 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
3349 The latter is generated from $received_to and is only set if the
3350 message has one envelope recipient. SA can use these headers,
3351 obviously out-of-the-box. Patch from Alex Miller.
3353 PH/08 The ${def test on a variable was returning false if the variable's
3354 value was "0", contrary to what the specification has always said!
3355 The result should be true unless the variable is empty.
3357 PH/09 The syntax error of a character other than { following "${if
3358 def:variable_name" (after optional whitespace) was not being diagnosed.
3359 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
3360 accidental colon was present, for example, could give incorrect results.
3362 PH/10 Tidied the code in a number of places where the st_size field of a stat()
3363 result is used (not including appendfile, where other changes are about
3366 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
3367 This involved changing a lot of size variables from int to off_t. It
3368 should work with maildirs and everything.
3370 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
3371 spamd dying while we are connected to it.
3373 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
3376 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
3377 function for BDB 4.3.
3379 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
3381 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
3382 the sources for PCRE have been re-arranged and more files are now
3385 PH/15 The code I had for printing potentially long long variables in PH/11
3386 above was not the best (it lost precision). The length of off_t variables
3387 is now inspected at build time, and an appropriate printing format (%ld
3388 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
3389 to be "long long int" or "long int". This is needed for the internal
3390 formatting function string_vformat().
3392 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
3393 the configuration file to be ":syslog", then the script "guesses" where
3394 the logs files are, rather than using the compiled in default. In our
3395 case the guess is not the same as the compiled default, so the script
3396 suddenly stopped working when I started to use syslog. The patch checks
3397 to see if log_file_path is "". If so, it attempts to read it from exim
3398 with no configuration file to get the compiled in version, before it
3399 falls back to the previous guessing code."
3401 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
3402 implementing BATV in an Exim configuration. See NewStuff for the gory
3405 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
3406 Makefile that are specific to HP-UX.
3408 PH/18 If the "use_postmaster" option was set for a recipient callout together
3409 with the "random" option, the postmaster address was used as the MAIL
3410 FROM address for the random test, but not for the subsequent recipient
3411 test. It is now used for both.
3413 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
3414 patch removes a few documentation additions to RFC 3028, because the
3415 latest draft now contains them. It adds the new en;ascii-case comparator
3416 and a new error check for 8bit text in MIME parts. Comparator and
3417 require names are now matched exactly. I enabled the subaddress
3418 extension, but it is not well tested yet (read: it works for me)."
3420 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
3421 rework some of the code of TK/09 above to avoid the hardwired use of
3422 "%lld" and "long long". Replaced the call to snprintf() with a call to
3425 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
3426 records point to non-existent hosts", "retry timeout exceeded", and
3427 "retry time not reached for any host after a long failure period".
3429 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
3430 experimental DomainKeys support:
3432 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
3433 (2) On an error such as an illegally used "control", the wrong name for
3434 the control was given.
3436 These problems did NOT occur unless DomainKeys support was compiled.
3438 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
3440 PH/24 Added ${if match_ip condition.
3442 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
3443 databases so that it will be absolutely obvious if a crash occurs in the
3444 DB library. This is a regular occurrence (often caused by mis-matched
3447 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
3448 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
3449 picked up on a user's system that detects such things. There doesn't seem
3450 to be a gcc warning option for this - only an attribute that has to be
3451 put on the function's prototype. It seems that in Fedora Core 4 they have
3452 set this on a number of new functions. No doubt there will be more in due
3455 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
3456 specify a transport. However, if an address that was verified by such a
3457 router was the subject of a callout, Exim crashed because it tried to
3458 read the rcpt_include_affixes from the non-existent transport. Now it
3459 just assumes that the setting of that option is false. This bug was
3460 introduced by 4.51/PH/31.
3462 PH/28 Changed -d+all to exclude +memory, because that information is very
3463 rarely of interest, but it makes the output a lot bigger. People tend to
3464 do -d+all out of habit.
3466 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
3467 code in os-type was giving problems when libc.so lives in lib64, like on
3470 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
3471 aren't the modern standard, and it seems that some systems' include files
3472 don't always have them. Exim was already checking for some of the newer
3473 ones like T_AAAA, and defining it itself. I've added checks for all the
3474 record types that Exim uses.
3476 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
3477 not automatically generating a new one, as it is supposed to. This
3478 prevented TLS from working. If the file did exist, but contained invalid
3479 data, a new version was generated, as expected. It was only the case of a
3480 non-existent file that was broken.
3482 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
3483 with a change in libdomainkeys > 0.64.
3485 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
3486 from DNS. If the selector record carries the flag, it now has
3487 precedence over the domain-wide flag.
3489 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
3491 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
3492 the use of a port name that isn't defined in /etc/services) occurred, the
3493 message was deferred as in a normal delivery, and thus remained on the
3494 spool, instead of being failed because of the mua_wrapper setting. This
3495 is now fixed, and I tidied up some of the mua_wrapper messages at the
3498 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
3499 the messages in a hash of arrays rather than using individual hashes.
3500 This is a bit cleaner and results in dramatic memory savings, albeit
3501 at a slight CPU cost.
3503 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
3504 as requested by Marc Sherman.
3506 SC/10 Eximstats: added histograms for user specified patterns as requested
3509 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
3511 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
3512 fopen() in the content-scanning modules that did not already have it.
3518 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
3519 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
3521 TK/02 Fix ACL "control" statement not being available in MIME ACL.
3523 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
3525 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
3526 to test Sieve filters that use "vacation".
3528 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
3529 that changes the way the GnuTLS parameters are stored in the cache file.
3530 The new format can be generated externally. For backward compatibility,
3531 if the data in the cache doesn't make sense, Exim assumes it has read an
3532 old-format file, and it generates new data and writes a new file. This
3533 means that you can't go back to an older release without removing the
3536 PH/03 A redirect router that has both "unseen" and "one_time" set does not
3537 work if there are any delivery delays because "one_time" forces the
3538 parent to be marked "delivered", so its unseen clone is never tried
3539 again. For this reason, Exim now forbids the simultaneous setting of
3542 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
3543 redirected to themselves ("homonym" addresses). Read the long ChangeLog
3544 entry if you want to know the details. The fix, however, neglected to
3545 consider the case when local delivery batching is involved. The test for
3546 "previously delivered" was not happening when checking to see if an
3547 address could be batched with a previous (undelivered) one; under
3548 certain circumstances this could lead to multiple deliveries to the same
3551 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
3552 in its include files, and this causes problems building Exim.
3554 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
3555 header_syntax) but Exim was just ignoring anything given after a slash.
3556 In particular, this caused confusion with an attempt to use "verify =
3557 reverse_host_lookup/defer_ok". An error is now given when options are
3558 supplied for verify items that do not have them. (Maybe reverse_host_
3559 lookup should have a defer_ok option, but that's a different point.)
3561 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
3562 defined by RFC 821) to 2048, because there were problems with some AUTH
3563 commands, and RFC 1869 says the size should be increased for extended
3564 SMTP commands that take arguments.
3566 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
3569 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
3570 "unknown" error; now it says that the functionality isn't in the binary.
3572 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
3573 an address' error message when a string expansion fails (syntax or
3574 whatever). Otherwise the password may appear in the log. Following change
3575 PH/42 below, there is no longer a chance of it appearing in a bounce
3578 PH/11 Installed exipick version 20050225.0 from John Jetmore.
3580 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
3581 of its addresses was ever tried. (Bugzilla bug #2.)
3583 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
3584 the result incorrectly in the debug output. (It correctly added a newline
3585 to what was transported.)
3587 TF/01 Added $received_time.
3589 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
3590 commented out examples of how to interface to a virus scanner and to
3591 SpamAssassin. Also added commented examples of av_scanner and
3592 spamd_address settings.
3594 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
3595 and controls are allowed in which ACLs. There were a couple of minor
3596 errors. Some of the entries in the conditions table (which is a table of
3597 where they are NOT allowed) were getting very unwieldy; rewrote them as a
3598 negation of where the condition IS allowed.
3600 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
3602 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
3603 header file does not have a version number, so I've had to invent a new
3604 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
3605 API. The code is untested by me (my Linux distribution still has 0.3.2 of
3606 radiusclient), but it was contributed by a Radius user.
3608 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
3609 files or directories, for OpenSSL.
3611 PH/19 When an Exim process that is running as root has to create an Exim log
3612 file, it does so in a subprocess that runs as exim:exim so as to get the
3613 ownership right at creation (otherwise, other Exim processes might see
3614 the file with the wrong ownership). There was no test for failure of this
3615 fork() call, which would lead to the process getting stuck as it waited
3616 for a non-existent subprocess. Forks do occasionally fail when resources
3617 run out. I reviewed all the other calls to fork(); they all seem to check
3620 PH/20 When checking for unexpected SMTP input at connect time (before writing
3621 the banner), Exim was not dealing correctly with a non-positive return
3622 from the read() function. If the client had disconnected by this time,
3623 the result was a log entry for a synchronization error with an empty
3624 string after "input=" when read() returned zero. If read() returned -1
3625 (an event I could not check), uninitialized data bytes were printed.
3626 There were reports of junk text (parts of files, etc) appearing after
3629 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
3631 PH/22 Added support for macro redefinition, and (re)definition in between
3632 driver and ACL definitions.
3634 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
3635 forgetting to use the resulting value; it was using the unexpanded value.
3637 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
3638 hadn't been configured. The fix is from Juergen Kreileder, who
3639 understands it better than I do:
3641 "Here's what I see happening with three configured cyrus_sasl
3642 authenticators configured (plain, login, cram-md5):
3644 On startup auth_cyrus_sasl_init() gets called for each of these.
3645 This means three calls to sasl_listmech() without a specified mech_list.
3646 => SASL tests which mechs of all available mechs actually work
3647 => three warnings about OTP not working
3648 => the returned list contains: plain, login, cram-md5, digest-md5, ...
3650 With the patch, sasl_listmech() also gets called three times. But now
3651 SASL's mech_list option is set to the server_mech specified in the the
3652 authenticator. Or in other words, the answer from sasl_listmech()
3653 gets limited to just the mech you're testing for (which is different
3655 => the return list contains just 'plain' or 'login', 'cram-md5' or
3656 nothing depending on the value of ob->server_mech.
3658 I've just tested the patch: Authentication still works fine,
3659 unavailable mechs specified in the exim configuration are still
3660 caught, and the auth.log warnings about OTP are gone."
3662 PH/25 When debugging is enabled, the contents of the command line are added
3663 to the debugging output, even when log_selector=+arguments is not
3666 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
3667 answer is "GNU", and only if the return is "GNU/something" is the answer
3670 PH/27 $acl_verify_message is now set immediately after the failure of a
3671 verification in an ACL, and so is available in subsequent modifiers. In
3672 particular, the message can be preserved by coding like this:
3674 warn !verify = sender
3675 set acl_m0 = $acl_verify_message
3677 Previously, $acl_verify_message was set only while expanding "message"
3678 and "log_message" when a very denied access.
3680 PH/28 Modified OS/os.c-Linux with
3682 -#ifndef OS_LOAD_AVERAGE
3683 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
3685 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
3686 nomenclature these days.)
3688 PH/29 Installed patch from the Sieve maintainer that adds the options
3689 sieve_useraddress and sieve_subaddress to the redirect router.
3691 PH/30 In these circumstances:
3692 . Two addresses routed to the same list of hosts;
3693 . First host does not offer TLS;
3694 . First host accepts first address;
3695 . First host gives temporary error to second address;
3696 . Second host offers TLS and a TLS session is established;
3697 . Second host accepts second address.
3698 Exim incorrectly logged both deliveries with the TLS parameters (cipher
3699 and peerdn, if requested) that were in fact used only for the second
3702 PH/31 When doing a callout as part of verifying an address, Exim was not paying
3703 attention to any local part prefix or suffix that was matched by the
3704 router that accepted the address. It now behaves in the same way as it
3705 does for delivery: the affixes are removed from the local part unless
3706 rcpt_include_affixes is set on the transport.
3708 PH/32 Add the sender address, as F=<...>, to the log line when logging a
3709 timeout during the DATA phase of an incoming message.
3711 PH/33 Sieve envelope tests were broken for match types other than :is. I have
3712 applied a patch sanctioned by the Sieve maintainer.
3714 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
3715 the uid or gid is negative. A case of a negative gid caused this to be
3716 noticed. The fix allows for either to be negative.
3718 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
3719 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
3722 PH/36 The change PH/12 above was broken. Fixed it.
3724 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
3725 the grounds that routing the same address twice would always produce the
3726 same answer. This might have been true once, but it is certainly no
3727 longer true now. Routing a child address may depend on the previous
3728 routing that produced that child. Some complicated redirection strategies
3729 went wrong when messages had multiple recipients, and made Exim's
3730 behaviour dependent on the order in which the addresses were given.
3732 I have moved the duplicate checking until after the routing is complete.
3733 Exim scans the addresses that are assigned to local and remote
3734 transports, and removes any duplicates. This means that more work will be
3735 done, as duplicates will always all be routed, but duplicates are
3736 presumably rare, so I don't expect this is of any significance.
3738 For deliveries to pipes, files, and autoreplies, the duplicate checking
3739 still happens during the routing process, since they are not going to be
3742 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
3743 It corrects a timeout issue with spamd. This is Ian's comment: "The
3744 background is that sometimes spamd either never reads data from a
3745 connection it has accepted, or it never writes response data. The exiscan
3746 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
3747 blindly assumes that writes won't block so it may never time out."
3749 PH/39 Allow G after quota size as well as K and M.
3751 PH/40 The value set for $authenticated_id in an authenticator may not contain
3752 binary zeroes or newlines because the value is written to log lines and
3753 to spool files. There was no check on this. Now the value is run through
3754 the string_printing() function so that such characters are converted to
3755 printable escape sequences.
3757 PH/41 $message_linecount is a new variable that contains the total number of
3758 lines in the message. Compare $body_linecount, which is the count for the
3761 PH/42 Exim no longer gives details of delivery errors for specific addresses in
3762 bounce and delay warning messages, except in certain special cases, which
3765 (a) An SMTP error message from a remote host;
3766 (b) A message specified in a :fail: redirection;
3767 (c) A message specified in a "fail" command in a system filter;
3768 (d) A message specified in a FAIL return from the queryprogram router;
3769 (e) A message specified by the cannot_route_message router option.
3771 In these cases only, Exim does include the error details in bounce and
3772 warning messages. There are also a few cases where bland messages such
3773 as "unrouteable address" or "local delivery error" are given.
3775 PH/43 $value is now also set for the "else" part of a ${run expansion.
3777 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
3778 being worked on, but at least Exim now implements the latest version to
3781 PH/45 In a pipe transport, although a timeout while waiting for the pipe
3782 process to complete was treated as a delivery failure, a timeout while
3783 writing the message to the pipe was logged, but erroneously treated as a
3784 successful delivery. Such timeouts include transport filter timeouts. For
3785 consistency with the overall process timeout, these timeouts are now
3786 treated as errors, giving rise to delivery failures by default. However,
3787 there is now a new Boolean option for the pipe transport called
3788 timeout_defer, which, if set TRUE, converts the failures into defers for
3789 both kinds of timeout. A transport filter timeout is now identified in
3792 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
3793 systems where "make" and "gmake" are different, calling "gmake" at top
3794 level broke things. I've arranged for the value of $(MAKE) to be passed
3795 from the Makefile to this script so that it can call the same version of
3799 A note about Exim versions 4.44 and 4.50
3800 ----------------------------------------
3802 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
3803 changes of various kinds. As a consequence, a big documentation update was
3804 needed. This delayed the release for rather longer than seemed good, especially
3805 in the light of a couple of (minor) security issues. Therefore, the changes
3806 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
3807 release. So 4.44 and 4.50 are in effect two different branches that both start
3810 I have left the 4.50 change log unchanged; it contains all the changes since
3811 4.43. The change log for 4.44 is below; many of its items are identical to
3812 those for 4.50. This seems to be the most sensible way to preserve the
3813 historical information.
3819 1. Minor wording change to the doc/README.SIEVE file.
3821 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
3822 computation of the current number of files was incorrect.
3824 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
3825 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
3828 4. Give more explanation in the error message when the command for a transport
3829 filter fails to execute.
3831 5. There are several places where Exim runs a non-Exim command in a
3832 subprocess. The SIGUSR1 signal should be disabled for these processes. This
3833 was being done only for the command run by the queryprogram router. It is
3834 now done for all such subprocesses. The other cases are: ${run, transport
3835 filters, and the commands run by the lmtp and pipe transports.
3837 6. Added CONFIGURE_GROUP build-time option.
3839 7. Some older OS have a limit of 256 on the maximum number of file
3840 descriptors. Exim was using setrlimit() to set 1000 as a large value
3841 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
3842 systems. I've change it so that if it can't get 1000, it tries for 256.
3844 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
3845 was an oversight, and furthermore, ever since the addition of extra
3846 controls (e.g. 4.43/32), the checks on when to allow different forms of
3847 "control" were broken. There should now be diagnostics for all cases when a
3848 control that does not make sense is encountered.
3850 9. Added the /retain_sender option to "control=submission".
3852 10. $recipients is now available in the predata ACL (oversight).
3854 11. Tidy the search cache before the fork to do a delivery from a message
3855 received from the command line. Otherwise the child will trigger a lookup
3856 failure and thereby defer the delivery if it tries to use (for example) a
3857 cached ldap connection that the parent has called unbind on.
3859 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
3860 of $address_data from the recipient verification was clobbered by the
3861 sender verification.
3863 13. The value of address_data from a sender verification is now available in
3864 $sender_address_data in subsequent conditions in the ACL statement.
3866 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
3868 15. Added a new option "connect=<time>" to callout options, to set a different
3871 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
3872 was its contents. (It was OK if the option was not defined at all.)
3874 17. A "Completed" log line is now written for messages that are removed from
3875 the spool by the -Mrm option.
3877 18. New variables $sender_verify_failure and $recipient_verify_failure contain
3878 information about exactly what failed.
3880 19. Added -dd to debug only the daemon process.
3882 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
3883 handles timeouts, both on the server side and network timeouts. Renamed the
3884 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
3886 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
3887 It is now set to "smtps".
3889 22. $host_address is now set to the target address during the checking of
3890 ignore_target_hosts.
3892 23. When checking ignore_target_hosts for an ipliteral router, no host name was
3893 being passed; this would have caused $sender_host_name to have been used if
3894 matching the list had actually called for a host name (not very likely,
3895 since this list is usually IP addresses). A host name is now passed as
3898 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
3899 code that specifies a non-restarting handler (typically sigaction() in
3900 modern systems) in an attempt to fix a rare and obscure crash bug.
3902 25. Narrowed the window for a race in the daemon that could cause it to ignore
3903 SIGCHLD signals. This is not a major problem, because they are used only to
3904 wake it up if nothing else does.
3906 26. A malformed maildirsize file could cause Exim to calculate negative values
3907 for the mailbox size or file count. Odd effects could occur as a result.
3908 The maildirsize information is now recalculated if the size or filecount
3911 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
3912 support for a long time. Removed HAVE_SYS_VFS_H.
3914 28. Installed the latest version of exipick from John Jetmore.
3916 29. In an address list, if the pattern was not a regular expression, an empty
3917 subject address (from a bounce message) matched only if the pattern was an
3918 empty string. Non-empty patterns were not even tested. This was the wrong
3919 because it is perfectly reasonable to use an empty address as part of a
3920 database query. An empty address is now tested by patterns that are
3921 lookups. However, all the other forms of pattern expect the subject to
3922 contain a local part and a domain, and therefore, for them, an empty
3923 address still always fails if the pattern is not itself empty.
3925 30. Exim went into a mad DNS loop when attempting to do a callout where the
3926 host was specified on an smtp transport, and looking it up yielded more
3927 than one IP address.
3929 31. Re-factored the code for checking spool and log partition space into a
3930 function that finds that data and another that does the check. The former
3931 is then used to implement four new variables: $spool_space, $log_space,
3932 $spool_inodes, and $log_inodes.
3934 32. The RFC2047 encoding function was originally intended for short strings
3935 such as real names; it was not keeping to the 75-character limit for
3936 encoded words that the RFC imposes. It now respects the limit, and
3937 generates multiple encoded words if necessary. To be on the safe side, I
3938 have increased the buffer size for the ${rfc2047: expansion operator from
3941 33. It is now permitted to omit both strings after an "if" condition; if the
3942 condition is true, the result is "true". As before, when the second string
3943 is omitted, a false condition yields an empty string. This makes it less
3944 cumbersome to write custom ACL and router conditions.
3946 34. Failure to deliver a bounce message always caused it to be frozen, even if
3947 there was an errors_to setting on the router. The errors_to setting is now
3950 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
3951 canonical form (fully expanded) before being placed in
3952 $sender_host_address.
3954 36. The table in the code that translates DNS record types into text (T_A to
3955 "A" for instance) was missing entries for NS and CNAME. It is just possible
3956 that this could have caused confusion if both these types were looked up
3957 for the same domain, because the text type is used as part of Exim's
3958 per-process caching. But the chance of anyone hitting this buglet seems
3961 37. The dnsdb lookup has been extended in a number of ways.
3963 (1) There is a new type, "zns", which walks up the domain tree until it
3964 finds some nameserver records. It should be used with care.
3966 (2) There is a new type, "mxh", which is like "mx" except that it returns
3967 just the host names, not the priorities.
3969 (3) It is now possible to give a list of domains (or IP addresses) to be
3970 looked up. The behaviour when one of the lookups defers can be
3971 controlled by a keyword.
3973 (4) It is now possible to specify the separator character for use when
3974 multiple records are returned.
3976 38. The dnslists ACL condition has been extended: it is now possible to supply
3977 a list of IP addresses and/or domains to be looked up in a particular DNS
3980 39. Added log_selector=+queue_time_overall.
3982 40. When running the queue in the test harness, wait just a tad after forking a
3983 delivery process, to get repeatability of debugging output.
3985 41. Include certificate and key file names in error message when GnuTLS fails
3986 to set them up, because the GnuTLS error message doesn't include the name
3987 of the failing file when there is a problem reading it.
3989 42. Allow both -bf and -bF in the same test run.
3991 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
3993 44. The "Exiscan patch" is now merged into the mainline Exim source.
3995 45. Sometimes the final signoff response after QUIT could fail to get
3996 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
3997 < 0 before doing a fflush(). This bug looks as though it goes back to the
3998 introduction of TLS in release 3.20, but "sometimes" must have been rare
3999 because the tests only now provoked it.
4001 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
4002 (this can affect the format of dates).
4004 47. exim_tidydb, when checking for the continued existence of a message for
4005 which it has found a message-specific retry record, was not finding
4006 messages that were in split spool directories. Consequently, it was
4007 deleting retry records that should have stayed in existence.
4009 48. Steve fixed some bugs in eximstats.
4011 49. The SPA authentication driver was not abandoning authentication and moving
4012 on to the next authenticator when an expansion was forced to fail,
4013 contradicting the general specification for all authenticators. Instead it
4014 was generating a temporary error. It now behaves as specified.
4016 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
4017 (the order specifies the preference for clients). The order is now AES256,
4018 AES128, 3DES, ARCFOUR128.
4020 51. Small patch to Sieve code - explicitly set From: when generating an
4023 52. Exim crashed if a remote delivery caused a very long error message to be
4024 recorded - for instance if somebody sent an entire SpamAssassin report back
4025 as a large number of 550 error lines. This bug was coincidentally fixed by
4026 increasing the size of one of Exim's internal buffers (big_buffer) that
4027 happened as part of the Exiscan merge. However, to be on the safe side, I
4028 have made the code more robust (and fixed the comments that describe what
4031 53. Now that there can be additional text after "Completed" in log lines (if
4032 the queue_time_overall log selector is set), a one-byte patch to exigrep
4033 was needed to allow it to recognize "Completed" as not the last thing in
4036 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
4037 patch that reportedly fixes this has been added. I am not expert enough to
4038 create a test for it. This is what the patch creator wrote:
4040 "I found a little strange behaviour of ldap code when working with
4041 Windows 2003 AD Domain, where users was placed in more than one
4042 Organization Units. When I tried to give exim partial DN, the exit code
4043 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
4044 But simultaneously result of request was absolutely normal ldap result,
4045 so I produce this patch..."
4047 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
4048 REFERENCE, so I have modified the code to exclude the patch when that macro
4051 55. Some experimental protocols are using DNS PTR records for new purposes. The
4052 keys for these records are domain names, not reversed IP addresses. The
4053 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4054 leaves it alone. Component reversal etc. now happens only for IP addresses.
4057 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
4059 57. Double the size of the debug message buffer (to 2048) so that more of very
4060 long debug lines gets shown.
4062 58. The exicyclog utility now does better if the number of log files to keep
4063 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
4065 59. Two changes related to the smtp_active_hostname option:
4067 (1) $smtp_active_hostname is now available as a variable.
4068 (2) The default for smtp_banner uses $smtp_active_hostname instead
4069 of $primary_hostname.
4071 60. The host_aton() function is supposed to be passed a string that is known
4072 to be a valid IP address. However, in the case of IPv6 addresses, it was
4073 not checking this. This is a hostage to fortune. Exim now panics and dies
4074 if the condition is not met. A case was found where this could be provoked
4075 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4076 components; fortuitously, this particular loophole had already been fixed
4077 by change 4.50/55 above.
4079 If there are any other similar loopholes, the new check in host_aton()
4080 itself should stop them being exploited. The report I received stated that
4081 data on the command line could provoke the exploit when Exim was running as
4082 exim, but did not say which command line option was involved. All I could
4083 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4084 running as the user.
4087 61. There was a buffer overflow vulnerability in the SPA authentication code
4088 (which came originally from the Samba project). I have added a test to the
4089 spa_base64_to_bits() function which I hope fixes it.
4092 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
4093 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
4095 63. The daemon start-up calls getloadavg() while still root for those OS that
4096 need the first call to be done as root, but it missed one case: when
4097 deliver_queue_load_max is set with deliver_drop_privilege. This is
4098 necessary for the benefit of the queue runner, because there is no re-exec
4099 when deliver_drop_privilege is set.
4101 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
4102 This has been fixed.
4104 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4105 was in use, was not putting the data itself into the right store pool;
4106 consequently, it could be overwritten for a subsequent message in the same
4107 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4110 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
4112 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
4113 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
4114 the function were treating the return as a boolean value, which happened to
4115 work because 0=false and not-0=true, but is not correct code.
4117 68. The host_aton() function was not handling scoped IPv6 addresses (those
4118 with, for example, "%eth0" on the end) correctly.
4120 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
4121 negated items (that is, ~something) in unsigned ints. Some compilers
4122 apparently mutter when there is no cast.
4124 70. If an address verification called from an ACL failed, and did not produce a
4125 user-specific message (i.e. there was only a "system" message), nothing was
4126 put in $acl_verify_message. In this situation, it now puts the system
4129 71. Change 4.23/11 added synchronization checking at the start of an SMTP
4130 session; change 4.31/43 added the unwanted input to the log line - except
4131 that it did not do this in the start of session case. It now does.
4133 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
4134 This is wrong and can cause the other end to generate a synchronization
4135 error if it is another Exim or anything else that does the synchronization
4136 check. A QUIT command is no longer sent after a timeout.
4138 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
4139 during host lookups.
4141 74. The defer_ok option of callout verification was not working if it was used
4142 when verifying addresses in header lines, that is, for this case:
4144 verify = header_sender/callout=defer_ok
4146 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
4147 those file descriptors could be used for SMTP connections. If anything
4148 wrote to stderr (the example that came up was "warn" in embedded Perl), it
4149 could be sent to the SMTP client, causing chaos. The daemon now opens
4150 stdin, stdout, and stderr to /dev/null when it puts itself into the
4153 76. Arrange for output from Perl's "warn" command to be written to Exim's main
4154 log by default. The user can override this with suitable Perl magic.
4156 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
4157 the log message when discard triggers, was not working for the DATA ACL or
4158 for the non-SMTP ACL.
4160 78. Error message wording change in sieve.c.
4162 79. If smtp_accept_max_per_host was set, the number of connections could be
4163 restricted to fewer than expected, because the daemon was trying to set up
4164 a new connection before checking whether the processes handling previous
4165 connections had finished. The check for completed processes is now done
4166 earlier. On busy systems, this bug wouldn't be noticed because something
4167 else would have woken the daemon, and it would have reaped the completed
4170 80. If a message was submitted locally by a user whose login name contained one
4171 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
4172 It caused a spool format error. I have fixed the spool reading code. A
4173 related problem was that the "from" clause in the Received: line became
4174 illegal because of the space(s). It is now covered by ${quote_local_part.
4176 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
4179 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
4181 83. Patch from Sieve maintainer to fix unterminated string problem in
4182 "vacation" handling.
4184 84. Some minor changes to the Linux configuration files to help with other
4185 OS variants using glibc.
4187 85. One more patch for Sieve to update vacation handling to latest spec.
4190 ----------------------------------------------------
4191 See the note above about the 4.44 and 4.50 releases.
4192 ----------------------------------------------------
4198 1. Change 4.43/35 introduced a bug that caused file counts to be
4199 incorrectly computed when quota_filecount was set in an appendfile
4202 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4203 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4206 3. Give more explanation in the error message when the command for a transport
4207 filter fails to execute.
4209 4. There are several places where Exim runs a non-Exim command in a
4210 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4211 was being done only for the command run by the queryprogram router. It is
4212 now done for all such subprocesses. The other cases are: ${run, transport
4213 filters, and the commands run by the lmtp and pipe transports.
4215 5. Some older OS have a limit of 256 on the maximum number of file
4216 descriptors. Exim was using setrlimit() to set 1000 as a large value
4217 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4218 systems. I've change it so that if it can't get 1000, it tries for 256.
4220 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4221 was an oversight, and furthermore, ever since the addition of extra
4222 controls (e.g. 4.43/32), the checks on when to allow different forms of
4223 "control" were broken. There should now be diagnostics for all cases when a
4224 control that does not make sense is encountered.
4226 7. $recipients is now available in the predata ACL (oversight).
4228 8. Tidy the search cache before the fork to do a delivery from a message
4229 received from the command line. Otherwise the child will trigger a lookup
4230 failure and thereby defer the delivery if it tries to use (for example) a
4231 cached ldap connection that the parent has called unbind on.
4233 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4234 of $address_data from the recipient verification was clobbered by the
4235 sender verification.
4237 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4238 was its contents. (It was OK if the option was not defined at all.)
4240 11. A "Completed" log line is now written for messages that are removed from
4241 the spool by the -Mrm option.
4243 12. $host_address is now set to the target address during the checking of
4244 ignore_target_hosts.
4246 13. When checking ignore_target_hosts for an ipliteral router, no host name was
4247 being passed; this would have caused $sender_host_name to have been used if
4248 matching the list had actually called for a host name (not very likely,
4249 since this list is usually IP addresses). A host name is now passed as
4252 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4253 code that specifies a non-restarting handler (typically sigaction() in
4254 modern systems) in an attempt to fix a rare and obscure crash bug.
4256 15. Narrowed the window for a race in the daemon that could cause it to ignore
4257 SIGCHLD signals. This is not a major problem, because they are used only to
4258 wake it up if nothing else does.
4260 16. A malformed maildirsize file could cause Exim to calculate negative values
4261 for the mailbox size or file count. Odd effects could occur as a result.
4262 The maildirsize information is now recalculated if the size or filecount
4265 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4266 support for a long time. Removed HAVE_SYS_VFS_H.
4268 18. Updated exipick to current release from John Jetmore.
4270 19. Allow an empty sender to be matched against a lookup in an address list.
4271 Previously the only cases considered were a regular expression, or an
4274 20. Exim went into a mad DNS lookup loop when doing a callout where the
4275 host was specified on the transport, if the DNS lookup yielded more than
4278 21. The RFC2047 encoding function was originally intended for short strings
4279 such as real names; it was not keeping to the 75-character limit for
4280 encoded words that the RFC imposes. It now respects the limit, and
4281 generates multiple encoded words if necessary. To be on the safe side, I
4282 have increased the buffer size for the ${rfc2047: expansion operator from
4285 22. Failure to deliver a bounce message always caused it to be frozen, even if
4286 there was an errors_to setting on the router. The errors_to setting is now
4289 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4290 canonical form (fully expanded) before being placed in
4291 $sender_host_address.
4293 24. Updated eximstats to version 1.33
4295 25. Include certificate and key file names in error message when GnuTLS fails
4296 to set them up, because the GnuTLS error message doesn't include the name
4297 of the failing file when there is a problem reading it.
4299 26. Expand error message when OpenSSL has problems setting up cert/key files.
4302 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
4303 (this can affect the format of dates).
4305 28. exim_tidydb, when checking for the continued existence of a message for
4306 which it has found a message-specific retry record, was not finding
4307 messages that were in split spool directories. Consequently, it was
4308 deleting retry records that should have stayed in existence.
4310 29. eximstats updated to version 1.35
4311 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
4312 1.35 - bugfix such that pie charts by volume are generated correctly
4314 30. The SPA authentication driver was not abandoning authentication and moving
4315 on to the next authenticator when an expansion was forced to fail,
4316 contradicting the general specification for all authenticators. Instead it
4317 was generating a temporary error. It now behaves as specified.
4319 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
4320 (the order specifies the preference for clients). The order is now AES256,
4321 AES128, 3DES, ARCFOUR128.
4323 31. Small patch to Sieve code - explicitly set From: when generating an
4326 32. Exim crashed if a remote delivery caused a very long error message to be
4327 recorded - for instance if somebody sent an entire SpamAssassin report back
4328 as a large number of 550 error lines. This bug was coincidentally fixed by
4329 increasing the size of one of Exim's internal buffers (big_buffer) that
4330 happened as part of the Exiscan merge. However, to be on the safe side, I
4331 have made the code more robust (and fixed the comments that describe what
4334 33. Some experimental protocols are using DNS PTR records for new purposes. The
4335 keys for these records are domain names, not reversed IP addresses. The
4336 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4337 leaves it alone. Component reversal etc. now happens only for IP addresses.
4340 34. The host_aton() function is supposed to be passed a string that is known
4341 to be a valid IP address. However, in the case of IPv6 addresses, it was
4342 not checking this. This is a hostage to fortune. Exim now panics and dies
4343 if the condition is not met. A case was found where this could be provoked
4344 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4345 components; fortuitously, this particular loophole had already been fixed
4346 by change 4.50/55 or 4.44/33 above.
4348 If there are any other similar loopholes, the new check in host_aton()
4349 itself should stop them being exploited. The report I received stated that
4350 data on the command line could provoke the exploit when Exim was running as
4351 exim, but did not say which command line option was involved. All I could
4352 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4353 running as the user.
4356 35. There was a buffer overflow vulnerability in the SPA authentication code
4357 (which came originally from the Samba project). I have added a test to the
4358 spa_base64_to_bits() function which I hope fixes it.
4361 36. The daemon start-up calls getloadavg() while still root for those OS that
4362 need the first call to be done as root, but it missed one case: when
4363 deliver_queue_load_max is set with deliver_drop_privilege. This is
4364 necessary for the benefit of the queue runner, because there is no re-exec
4365 when deliver_drop_privilege is set.
4367 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4368 was in use, was not putting the data itself into the right store pool;
4369 consequently, it could be overwritten for a subsequent message in the same
4370 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4373 38. Sometimes the final signoff response after QUIT could fail to get
4374 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4375 < 0 before doing a fflush(). This bug looks as though it goes back to the
4376 introduction of TLS in release 3.20, but "sometimes" must have been rare
4377 because the tests only now provoked it.
4383 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
4384 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
4385 Now it returns an integer. A number of calls were still expecting a T/F
4386 return. Fortuitously, in all cases, the tests worked in OK situations,
4387 which is the norm. However, things would have gone wrong on any write
4388 failures on the smtp file descriptor. This function is used when sending
4389 messages over SMTP and also when doing verify callouts.
4391 2. When Exim is called to do synchronous delivery of a locally submitted
4392 message (the -odf or -odi options), it no longer closes stderr before doing
4395 3. Implemented the mua_wrapper option.
4397 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
4399 5. Implemented the functions header_remove(), header_testname(),
4400 header_add_at_position(), and receive_remove_recipient(), and exported them
4403 6. If an ACL "warn" statement specified the addition of headers, Exim already
4404 inserted X-ACL-Warn: at the start if there was no header name. However, it
4405 was not making this test for the second and subsequent header lines if
4406 there were newlines in the string. This meant that an invalid header could
4407 be inserted if Exim was badly configured.
4409 7. Allow an ACL "warn" statement to add header lines at the start or after all
4410 the Received: headers, as well as at the end.
4412 8. Added the rcpt_4xx retry error code.
4414 9. Added postmaster_mailfrom=xxx to callout verification option.
4416 10. Added mailfrom=xxxx to the callout verification option, for verify=
4419 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
4420 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
4422 12. Inserted some casts to stop certain compilers warning when using pointer
4423 differences as field lengths or precisions in printf-type calls (mostly
4424 affecting debugging statements).
4426 13. Added optional readline() support for -be (dynamically loaded).
4428 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
4429 same clock tick as a message's arrival, so that its received time was the
4430 same as the "first fail" time on the retry record, and that message
4431 remained on the queue past the ultimate address timeout, every queue runner
4432 would try a delivery (because it was past the ultimate address timeout) but
4433 after another failure, the ultimate address timeout, which should have then
4434 bounced the address, did not kick in. This was a "< instead of <=" error;
4435 in most cases the first failure would have been in the next clock tick
4436 after the received time, and all would be well.
4438 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
4439 being recognized when the domain list was tested by the match_domain
4440 condition in an expansion string.
4442 16. Added the ${str2b64: operator.
4444 17. Exim was always calling setrlimit() to set a large limit for the number of
4445 processes, without checking whether the existing limit was already
4446 adequate. (It did check for the limit on file descriptors.) Furthermore,
4447 errors from getrlimit() and setrlimit() were being ignored. Now they are
4448 logged to the main and panic logs, but Exim does carry on, to try to do its
4449 job under whatever limits there are.
4451 18. Imported PCRE 5.0.
4453 19. Trivial typo in log message " temporarily refused connection" (the leading
4456 20. If the log selector return_path_on_delivery was set and an address was
4457 redirected to /dev/null, the delivery process crashed because it assumed
4458 that a return path would always be set for a "successful" delivery. In this
4459 case, the whole delivery is bypassed as an optimization, and therefore no
4462 21. Internal re-arrangement: the function for sending a challenge and reading
4463 a response while authentication was assuming a zero-terminated challenge
4464 string. It's now changed to take a pointer and a length, to allow for
4465 binary data in such strings.
4467 22. Added the cyrus_sasl authenticator (code supplied by MBM).
4469 23. Exim was not respecting finduser_retries when seeking the login of the
4470 uid under which it was called; it was always trying 10 times. (The default
4471 setting of finduser_retries is zero.) Also, it was sleeping after the final
4472 failure, which is pointless.
4474 24. Implemented tls_on_connect_ports.
4476 25. Implemented acl_smtp_predata.
4478 26. If the domain in control=submission is set empty, Exim assumes that the
4479 authenticated id is a complete email address when it generates From: or
4480 Sender: header lines.
4482 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
4483 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
4484 chown and chgrp in /bin and hostname in /usr/bin.
4486 28. Exim was keeping the "process log" file open after each use, just as it
4487 does for the main log. This opens the possibility of it remaining open for
4488 long periods when the USR1 signal hits a daemon. Occasional processlog
4489 errors were reported, that could have been caused by this. Anyway, it seems
4490 much more sensible not to leave this file open at all, so that is what now
4493 29. The long-running daemon process does not normally write to the log once it
4494 has entered its main loop, and it closes the log before doing so. This is
4495 so that log files can straightforwardly be renamed and moved. However,
4496 there are a couple of unusual error situations where the daemon does write
4497 log entries, and I had neglected to close the log afterwards.
4499 30. The text of an SMTP error response that was received during a remote
4500 delivery was being truncated at 512 bytes. This is too short for some of
4501 the long messages that one sometimes sees. I've increased the limit to
4504 31. It is now possible to make retry rules that apply only when a message has a
4505 specific sender, in particular, an empty sender.
4507 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
4508 it possible to be selective about when SMTP synchronization is enforced.
4510 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
4512 32. Implemented hosts_connection_nolog.
4514 33. Added an ACL for QUIT.
4516 34. Setting "delay_warning=" to disable warnings was not working; it gave a
4519 35. Added mailbox_size and mailbox_filecount to appendfile.
4521 36. Added control = no_multiline_responses to ACLs.
4523 37. There was a bug in the logic of the code that waits for the clock to tick
4524 in the case where the clock went backwards by a substantial amount such
4525 that the microsecond fraction of "now" was more than the microsecond
4526 fraction of "then" (but the whole seconds number was less).
4528 38. Added support for the libradius Radius client library this is found on
4529 FreeBSD (previously only the radiusclient library was supported).
4535 1. When certain lookups returned multiple values in the form name=value, the
4536 quoting of the values was not always being done properly. Specifically:
4537 (a) If the value started with a double quote, but contained no whitespace,
4539 (b) If the value contained whitespace other than a space character (i.e.
4540 tabs or newlines or carriage returns) it was not quoted.
4541 This fix has been applied to the mysql and pgsql lookups by writing a
4542 separate quoting function and calling it from the lookup code. The fix
4543 should probably also be applied to nisplus, ibase and oracle lookups, but
4544 since I cannot test any of those, I have not disturbed their existing code.
4546 2. A hit in the callout cache for a specific address caused a log line with no
4547 reason for rejecting RCPT. Now it says "Previous (cached) callout
4548 verification failure".
4550 3. There was an off-by-one bug in the queryprogram router. An over-long
4551 return line was truncated at 256 instead of 255 characters, thereby
4552 overflowing its buffer with the terminating zero. As well as fixing this, I
4553 have increased the buffer size to 1024 (and made a note to document this).
4555 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
4556 when Exim is waiting for an SMTP response from a remote server, Exim
4557 restarts its select() call on the socket, thereby resetting its timeout.
4558 This is not a problem when such interrupts are rare. Somebody set up a cron
4559 job to run exiwhat every 2 minutes, which is less than the normal select()
4560 timeout (5 or 10 minutes). This meant that the select() timeout never
4561 kicked in because it was always reset. I have fixed this by comparing the
4562 time when an interrupt arrives with the time at the start of the first call
4563 to select(). If more time than the timeout has elapsed, the interrupt is
4564 treated as a timeout.
4566 5. Some internal re-factoring in preparation for the addition of Sieve
4567 extensions (by MH). In particular, the "personal" test is moved to a
4568 separate function, and given an option for scanning Cc: and Bcc: (which is
4569 not set for Exim filters).
4571 6. When Exim created an email address using the login of the caller as the
4572 local part (e.g. when creating a From: or Sender: header line), it was not
4573 quoting the local part when it contained special characters such as @.
4575 7. Installed new OpenBSD configuration files.
4577 8. Reworded some messages for syntax errors in "and" and "or" conditions to
4578 try to make them clearer.
4580 9. Callout options, other than the timeout value, were being ignored when
4581 verifying sender addresses in header lines. For example, when using
4583 verify = header_sender/callout=no_cache
4585 the cache was (incorrectly) being used.
4587 10. Added a missing instance of ${EXE} to the exim_install script; this affects
4588 only the Cygwin environment.
4590 11. When return_path_on_delivery was set as a log selector, if different remote
4591 addresses in the same message used different return paths and parallel
4592 remote delivery occurred, the wrong values would sometimes be logged.
4593 (Whenever a remote delivery process finished, the return path value from
4594 the most recently started remote delivery process was logged.)
4596 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
4597 lines when AUTH and/or TLS are in use. This is the "received protocol"
4598 field. Exim used to use "asmtp" for authenticated SMTP, without any
4599 indication (in the protocol name) for TLS use. Now it follows the RFC and
4600 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
4601 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
4602 names appear in log lines as well as in Received: header lines.
4604 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
4605 extensions, and comparison tests, and to fix some bugs.
4607 14. Changes to the "personal" filter test:
4609 (1) The test was buggy in that it was just doing the equivalent of
4610 "contains" tests on header lines. For example, if a user's address was
4611 anne@some.where, the "personal" test would incorrectly be true for
4613 To: susanne@some.where
4615 This test is now done by extracting each address from the header in turn,
4616 and checking the entire address. Other tests that are part of "personal"
4617 are now done using regular expressions (for example, to check local parts
4618 of addresses in From: header lines).
4620 (2) The list of non-personal local parts in From: addresses has been
4621 extended to include "listserv", "majordomo", "*-request", and "owner-*",
4622 taken from the Sieve specification recommendations.
4624 (3) If the message contains any header line starting with "List-" it is
4625 treated as non-personal.
4627 (4) The test for "circular" in the Subject: header line has been removed
4628 because it now seems ill-conceived.
4630 15. Minor typos in src/EDITME comments corrected.
4632 16. Installed latest exipick from John Jetmore.
4634 17. If headers_add on a router specified a text string that was too long for
4635 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
4636 of string_sprintf() is now avoided.
4638 18. $message_body_size was not set (it was always zero) when running the DATA
4639 ACL and the local_scan() function.
4641 19. For the "mail" command in an Exim filter, no default was being set for
4642 the once_repeat time, causing a random time value to be used if "once" was
4643 specified. (If the value happened to be <= 0, no repeat happened.) The
4644 default is now 0s, meaning "never repeat". The "vacation" command was OK
4645 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
4646 (I found it when inspecting the code).
4648 20. There is now an overall timeout for performing a callout verification. It
4649 defaults to 4 times the callout timeout, which applies to individual SMTP
4650 commands during the callout. The overall timeout applies when there is more
4651 than one host that can be tried. The timeout is checked before trying the
4652 next host. This prevents very long delays if there are a large number of
4653 hosts and all are timing out (e.g. when the network connections are timing
4654 out). The value of the overall timeout can be changed by specifying an
4655 additional sub-option for "callout", called "maxwait". For example:
4657 verify = sender/callout=5s,maxwait=20s
4659 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
4660 to the end before writing, but this should make it even safer).
4662 22. Exim was forgetting that it had advertised PIPELINING for the second and
4663 subsequent messages on an SMTP connection. It was also not resetting its
4664 memory on STARTTLS and an internal HELO.
4666 23. When Exim logs an SMTP synchronization error within a session, it now
4667 records whether PIPELINING has been advertised or not.
4669 24. Added 3 instances of "(long int)" casts to time_t variables that were being
4670 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
4671 rather than long int.
4673 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
4675 26. Added the never_mail option to autoreply.
4681 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
4682 crash if the getsockname() call failed; this can happen if a connection is
4683 closed very soon after it is established. The problem was simply in the
4684 order in which certain operations were done, causing Exim to try to write
4685 to the SMTP stream before it had set up the file descriptor. The bug has
4686 been fixed by making things happen in the correct order.
4692 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
4693 before the connection was closed, thus losing the rejection response.
4695 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
4696 some early Solaris releases, but causes trouble in current releases where
4697 socklen_t is defined.
4699 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
4702 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
4705 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
4706 that releasing the top of it at the end releases what was used for sub-
4707 expansions (unless the block got too big). However, discard this block if
4708 the first thing is a variable or header, so that we can use its block when
4709 it is dynamic (useful for very large $message_headers, for example).
4711 6. Lookups now cache *every* query, not just the most recent. A new, separate
4712 store pool is used for this. It can be recovered when all lookup caches are
4713 flushed. Lookups now release memory at the end of their result strings.
4714 This has involved some general refactoring of the lookup sources.
4716 7. Some code has been added to the store_xxx() functions to reduce the amount
4717 of flapping under certain conditions.
4719 8. log_incoming_interface used to affect only the <= reception log lines. Now
4720 it causes the local interface and port to be added to several more SMTP log
4721 lines, for example "SMTP connection from", and rejection lines.
4723 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
4725 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
4727 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
4728 could be overwritten at the end of the current message (or the start of a
4729 new message if it was set in a HELO ACL). The value is now preserved for
4730 the duration of the SMTP connection.
4732 12. If a transport had a headers_rewrite setting, and a matching header line
4733 contained an unqualified address, that address was qualified, even if it
4734 did not match any rewriting rules. The underlying bug was that the values
4735 of the flags that permit the existence of unqualified sender and recipient
4736 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
4737 non-local messages, and by -bnq for local messages) were not being
4738 preserved with the message after it was received.
4740 13. When Exim was logging an SMTP synchronization error, it could sometimes log
4741 "next input=" as part of the text comprising the host identity instead of
4742 the correct text. The code was using the same buffer for two different
4743 strings. However, depending on which order the printing function evaluated
4744 its arguments, the bug did not always show up. Under Linux, for example, my
4745 test suite worked just fine.
4747 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
4748 doesn't work with some older versions of Perl. It has been changed to "my",
4749 which in any case is probably the better facility to use.
4751 15. A really picky compiler found some instances of statements for creating
4752 error messages that either had too many or two few arguments for the format
4755 16. The size of the buffer for calls to the DNS resolver has been increased
4756 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
4757 for addresses that have a lot of PTR records. This alleviates a problem; it
4758 does not fully solve it.
4760 17. A dnsdb lookup for PTR records that receives more data than will fit in the
4761 buffer now truncates the list and logs the incident, which is the same
4762 action as happens when Exim is looking up a host name and its aliases.
4763 Previously in this situation something unpredictable would happen;
4764 sometimes it was "internal error: store_reset failed".
4766 18. If a server dropped the connection unexpectedly when an Exim client was
4767 using GnuTLS and trying to read a response, the client delivery process
4768 crashed while trying to generate an error log message.
4770 19. If a "warn" verb in an ACL added multiple headers to a message in a single
4771 string, for example:
4773 warn message = H1: something\nH2: something
4775 the text was added as a single header line from Exim's point of view
4776 though it ended up OK in the delivered message. However, searching for the
4777 second and subsequent header lines using $h_h2: did not work. This has been
4778 fixed. Similarly, if a system filter added multiple headers in this way,
4779 the routers could not see them.
4781 20. Expanded the error message when iplsearch is called with an invalid key to
4782 suggest using net-iplsearch in a host list.
4784 21. When running tests using -bh, any delays imposed by "delay" modifiers in
4785 ACLs are no longer actually imposed (and a message to that effect is
4788 22. If a "gecos" field in a passwd entry contained escaped characters, in
4789 particular, if it contained a \" sequence, Exim got it wrong when building
4790 a From: or a Sender: header from that name. A second bug also caused
4791 incorrect handling when an unquoted " was present following a character
4792 that needed quoting.
4794 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
4795 was not being matched caselessly.
4797 24. Arranged for all hyphens in the exim.8 source to be escaped with
4800 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
4801 properly. Recipient callout cache records were still being keyed to include
4802 the sender, even when use_sender was set false. This led to far more
4803 callouts that were necessary. The sender is no longer included in the key
4804 when use_sender is false.
4806 26. Added "control = submission" modifier to ACLs.
4808 27. Added the ${base62d: operator to decode base 62 numbers.
4810 28. dnsdb lookups can now access SRV records.
4812 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
4813 the configuration file.
4815 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
4816 (-v) mode. This makes the output for a verbose queue run more intelligible.
4818 31. Added a use_postmaster feature to recipient callouts.
4820 32. Added the $body_zerocount variable, containing the number of binary zero
4821 bytes in the message body.
4823 33. The time of last modification of the "new" subdirectory is now used as the
4824 "mailbox time last read" when there is a quota error for a maildir
4827 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
4829 35. Added +ignore_unknown as a special item in host lists.
4831 36. Code for decoding IPv6 addresses in host lists is now included, even if
4832 IPv6 support is not being compiled. This fixes a bug in which an IPv6
4833 address was recognized as an IP address, but was then not correctly decoded
4834 into binary, causing unexpected and incorrect effects when compared with
4841 1. Very minor rewording of debugging text in manualroute to say "list of
4842 hosts" instead of "hostlist".
4844 2. If verify=header_syntax was set, and a header line with an unqualified
4845 address (no domain) and a large number of spaces between the end of the
4846 name and the colon was received, the reception process suffered a buffer
4847 overflow, and (when I tested it) crashed. This was caused by some obsolete
4848 code that should have been removed. The fix is to remove it!
4850 3. When running in the test harness, delay a bit after writing a bounce
4851 message to get a bit more predictability in the log output.
4853 4. Added a call to search_tidyup() just before forking a reception process. In
4854 theory, someone could use a lookup in the expansion of smtp_accept_max_
4855 per_host which, without the tidyup, could leave open a database connection.
4857 5. Added the variables $recipient_data and $sender_data which get set from a
4858 lookup success in an ACL "recipients" or "senders" condition, or a router
4859 "senders" option, similar to $domain_data and $local_part_data.
4861 6. Moved the writing of debug_print from before to after the "senders" test
4864 7. Change 4.31/66 (moving the time when the Received: is generated) caused
4865 problems for message scanning, either using a data ACL, or using
4866 local_scan() because the Received: header was not generated till after they
4867 were called (in order to set the time as the time of reception completion).
4868 I have revised the way this works. The header is now generated after the
4869 body is received, but before the ACL or local_scan() are called. After they
4870 are run, the timestamp in the header is updated.
4876 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
4877 before starting a queue runner without re-exec. This happened only when
4878 deliver_drop_privilege was set or when the Exim user was set to root. The
4879 effect of the bug was that timeouts during subsequent deliveries caused
4880 crashes instead of being properly handled. The handler is now left at its
4881 default (and expected) setting.
4883 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
4884 message, again when deliver_drop_privilege is set or Exim is run as root.
4885 The bug described in (1) was not present in this case, but the tidying up
4886 of the other signals was missing. I have made the two cases consistent.
4888 3. The ignore_target_hosts setting on a manualroute router was being ignored
4889 for hosts that were looked up using the /MX notation.
4891 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
4894 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
4895 operated on the sender address. After changing the $sender_address to <>
4896 for the sender address verify, Exim was re-instated it as the original
4897 (before rewriting) address, but remembering that it had rewritten it, so it
4898 wasn't rewriting it again. This bug also had the effect of breaking the
4899 sender address verification caching when the sender address was rewritten.
4901 6. The ignore_target_hosts option was being ignored by the ipliteral router.
4902 This has been changed so that if the ip literal address matches
4903 ignore_target_hosts, the router declines.
4905 7. Added expansion conditions match_domain, match_address, and match_local_
4906 part (NOT match_host).
4908 8. The placeholder for the Received: header didn't have a length field set.
4910 9. Added code to Exim itself and to exim_lock to test for a specific race
4911 condition that could lead to file corruption when using MBX delivery. The
4912 issue is with the lockfile that is created in /tmp. If this file is removed
4913 after a process has opened it but before that process has acquired a lock,
4914 there is the potential for a second process to recreate the file and also
4915 acquire a lock. This could lead to two Exim processes writing to the file
4916 at the same time. The added code performs the same test as UW imapd; it
4917 checks after acquiring the lock that its file descriptor still refers to
4918 the same named file.
4920 10. The buffer for building added header lines was of fixed size, 8192 bytes.
4921 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
4924 11. Added the smtp_active_hostname option. If used, this will typically be made
4925 to depend on the incoming interface address. Because $interface_address is
4926 not set up until the daemon has forked a reception process, error responses
4927 that can happen earlier (such as "too many connections") no longer contain
4930 12. If an expansion in a condition on a "warn" statement fails because a lookup
4931 defers, the "warn" statement is abandoned, and the next ACL statement is
4932 processed. Previously this caused the whole ACL to be aborted.
4934 13. Added the iplsearch lookup type.
4936 14. Added ident_timeout as a log selector.
4938 15. Added tls_certificate_verified as a log selector.
4940 16. Added a global option tls_require_ciphers (compare the smtp transport
4941 option of the same name). This controls incoming TLS connections.
4943 17. I finally figured out how to make tls_require_ciphers do a similar thing
4944 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
4945 before starting the TLS session.
4947 18. Tabs are now shown as \t in -bP output.
4949 19. If the log selector return_path_on_delivery was set, Exim crashed when
4950 bouncing a message because it had too many Received: header lines.
4952 20. If two routers both had headers_remove settings, and the first one included
4953 a superfluous trailing colon, the final name in the first list and the
4954 first name in the second list were incorrectly joined into one item (with a
4955 colon in the middle).
4961 1. Added -C and -D options to the exinext utility, mainly to make it easier
4962 to include in the automated testing, but these could be helpful when
4963 multiple configurations are in use.
4965 2. The exinext utility was not formatting the output nicely when there was
4966 an alternate port involved in the retry record key, nor when there was a
4967 message id as well (for retries that were specific to a specific message
4968 and a specific host). It was also confused by IPv6 addresses, because of
4969 the additional colons they contain. I have fixed the IPv4 problem, and
4970 patched it up to do a reasonable job for IPv6.
4972 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
4973 delivery, the log line now contains "pipelined" if PIPELINING was used.
4975 4. An SMTP transport process used to panic and die if the bind() call to set
4976 an explicit outgoing interface failed. This has been changed; it is now
4977 treated in the same way as a connect() failure.
4979 5. A reference to $sender_host_name in the part of a conditional expansion
4980 that was being skipped was still causing a DNS lookup. This no longer
4983 6. The def: expansion condition was not recognizing references to header lines
4984 that used bh_ and bheader_.
4986 7. Added the _cache feature to named lists.
4988 8. The code for checking quota_filecount in the appendfile transport was
4989 allowing one more file than it should have been.
4991 9. For compatibility with Sendmail, the command line option
4999 and sets the incoming protocol and host name (for trusted callers). The
5000 host name and its colon can be omitted when only the protocol is to be set.
5001 Note the Exim already has two private options, -pd and -ps, that refer to
5002 embedded Perl. It is therefore impossible to set a protocol value of "d" or
5003 "s", but I don't think that's a major issue.
5005 10. A number of refactoring changes to the code, none of which should affect
5008 (a) The number of logging options was getting close to filling up the
5009 32-bit word that was used as a bit map. I have split them into two classes:
5010 those that are passed in the argument to log_write(), and those that are
5011 only ever tested independently outside of that function. These are now in
5012 separate 32-bit words, so there is plenty of room for expansion again.
5013 There is no change in the user interface or the logging behaviour.
5015 (b) When building, for example, log lines, the code previously used a
5016 macro that called string_cat() twice, in order to add two strings. This is
5017 not really sufficiently general. Furthermore, there was one instance where
5018 it was actually wrong because one of the argument was used twice, and in
5019 one call a function was used. (As it happened, calling the function twice
5020 did not affect the overall behaviour.) The macro has been replaced by a
5021 function that can join an arbitrary number of extra strings onto a growing
5024 (c) The code for expansion conditions now uses a table and a binary chop
5025 instead of a serial search (which was left over from when there were very
5026 few conditions). Also, it now recognizes conditions like "pam" even when
5027 the relevant support is not compiled in: a suitably worded error message is
5028 given if an attempt is made to use such a condition.
5030 11. Added ${time_interval:xxxxx}.
5032 12. A bug was causing one of the ddress fields not to be passed back correctly
5033 from remote delivery subprocesses. The field in question was not being
5034 subsequently used, so this caused to problems in practice.
5036 13. Added new log selectors queue_time and deliver_time.
5038 14. Might have fixed a bug in maildirsizefile handling that threw up
5039 "unexpected character" debug warnings, and recalculated the data
5040 unnecessarily. In any case, I expanded the warning message to give more
5043 15. Added the message "Restricted characters in address" to the statements in
5044 the default ACL that block characters like @ and % in local parts.
5046 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
5047 Three changes have been made:
5049 (a) There was a serious bug; a negative response to MAIL caused the whole
5050 recipient domain to be cached as invalid, thereby blocking all messages
5051 to all local parts at the same domain, from all senders. This bug has
5052 been fixed. The domain is no longer cached after a negative response to
5053 MAIL if the sender used is not empty.
5055 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
5058 (c) A new callout option, "use_sender" has been added for people who want
5059 the modified behaviour.
5065 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
5068 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
5069 indeed breaks things for older releases.
5071 3. Added additional logging to the case where there is a problem reading data
5072 from a filter that is running in a subprocess using a pipe, in order to
5073 try to track down a specific problem.
5075 4. Testing facility fudge: when running in the test harness and attempting
5076 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
5077 getting "No route to host". Convert this to a timeout.
5079 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
5082 6. Some OS don't have socklen_t but use size_t instead. This affects the
5083 fifth argument of getsockopt() amongst other things. This is now
5084 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
5085 can be set for individual OS. I have set it for SunOS5, OSF1, and
5086 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
5087 some earlier ones do not.
5089 7. Change 4.30/15 was not doing the test caselessly.
5091 8. The standard form for an IPv6 address literal was being rejected by address
5092 parsing in, for example, MAIL and RCPT commands. An example of this kind of
5093 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5094 this, as well as the form without the "IPv6" on the front (but only when
5095 address literals are enabled, of course).
5097 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
5099 10. Exim crashed if a message with an empty sender address specified by -f
5100 encountered a router with an errors_to setting. This could be provoked only
5101 by a command such as
5105 where an empty string was supplied; "<>" did not hit this bug.
5107 11. Installed PCRE release 4.5.
5109 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
5110 remained set. It is now erased.
5112 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
5113 times from message ids (which are base 36 rather than the normal 62).
5115 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
5116 were being counted as actual protocol errors, and logged if the log
5117 selector +smtp_protocol_error was set. One cannot be perfect in this test,
5118 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
5119 and DATA following a set of rejected RCPTs do not count as protocol errors.
5120 In other words, Exim assumes they were pipelined, though this may not
5121 actually be the case. Of course, in all cases the client gets an
5122 appropriate error code.
5124 15. If a lookup fails in an ACL condition, a message about the failure may
5125 be available; it is used if testing the ACL cannot continue, because most
5126 such messages specify what the cause of the deferral is. However, some
5127 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
5128 that caused an old message to be retained and used if a later statement
5129 caused a defer, replacing the real cause of the deferral.
5131 16. If an IP address had so many PTR records that the DNS lookup buffer
5132 was not large enough to hold them, Exim could crash while trying to process
5133 the truncated data. It now detects and logs this case.
5135 17. Further to 4.21/58, another change has been made: if (and only if) the
5136 first line of a message (the first header line) ends with CRLF, a bare LF
5137 in a subsequent header line has a space inserted after it, so as not to
5138 terminate the header.
5140 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
5141 unnecessarily, used atoi() instead of strtol(), and didn't check the
5142 termination when getting file sizes from file names by regex.
5144 19. Completely re-implemented the support for maildirsize files, in the light
5145 of a number of problems with the previous contributed implementation
5146 (4.30/29). In particular:
5148 . If the quota is zero, the maildirsize file is maintained, but no quota is
5151 . If the maildir directory does not exist, it is created before any attempt
5152 to write a maildirsize file.
5154 . The quota value in the file is just a cache; if the quota is changed in
5155 the transport, the new value overrides.
5157 . A regular expression is available for excluding directories from the
5160 20. The autoreply transport checks the characters in options that define the
5161 message's headers; it allows continued headers, but it was checking with
5162 isspace() after an embedded newline instead of explicitly looking for a
5165 21. If all the "regular" hosts to which an address was routed had passed their
5166 expiry times, and had not reached their retry times, the address was
5167 bounced, even if fallback hosts were defined. Now Exim should go on to try
5170 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
5171 equivalent code in the SMTP transport. Some hosts send humungous responses
5172 to HELO/EHLO, more than 1024 it seems.
5174 23. Refactoring: code in filter.c used (void *) for "any old type" but this
5175 gives compiler warnings in some environments. I've now done it "properly",
5178 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
5179 (because of problems with the built-in one) was declared to return uschar *
5180 instead of char *, causing compiler failure.
5182 25. Fixed a file descriptor leak when processing alias/forward files.
5184 26. Fixed a minor format string issue in dbfn.c.
5186 27. Typo in exim.c: ("dmbnz" for "dbmnz").
5188 28. If a filter file refered to $h_xxx or $message_headers, and the headers
5189 contained RFC 2047 "words", Exim's memory could, under certain conditions,
5192 29. When a sender address is verified, it is cached, to save repeating the test
5193 when there is more than one recipient in a message. However, when the
5194 verification involves a callout, it is possible for different callout
5195 options to be set for different recipients. It is too complicated to keep
5196 track of this in the cache, so now Exim always runs a verification when a
5197 callout is required, relying on the callout cache for the optimization.
5198 The overhead is duplication of the address routing, but this should not be
5201 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
5202 to be verified with callout=postmaster, and the main callout worked but the
5203 postmaster check failed, the verification correctly failed. However, if a
5204 subsequent RCPT command asked for sender verification *without* the
5205 postmaster check, incorrect caching caused this verification also to fail,
5208 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
5209 it was not caching the DNS options (qualify_single, search_parents) that
5210 were used when the lookup failed. A subsequent lookup with different
5211 options therefore always gave the same answer, though there were cases
5212 where it should not have. (Example: a "domains = !$mx_any" option on a
5213 dnslookup router: the "domains" option is always processed without any
5214 widening, but the router might have qualify_single set.) Now Exim uses the
5215 cached value only when the same options are set.
5217 32. Added John Jetmore's "exipick" utility to the distribution.
5219 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
5220 than a timeout (e.g. a certificate is required, and is not provided), an
5221 Exim server now closes the connection immediately. Previously it waited for
5222 the client to close - but if the client is SSL, it seems that they each
5223 wait for each other, leading to a delay before one of them times out.
5225 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
5226 maintained 0.8.x compatibility because I don't think many are using it, and
5227 it is clearly obsolete.
5229 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
5232 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
5233 client certificate was expired. A simple patch fixes this, though I don't
5234 understand the full logic of why the verify callback is called multiple
5237 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
5238 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
5239 which causes problems with some clients (such as the Certicom SSL Plus
5240 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
5241 disables the coutermeasure allowing Eudora to connect."
5243 38. Exim was not checking that a write() to a log file succeeded. This could
5244 lead to Bad Things if a log got too big, in particular if it hit a file
5245 size limit. Exim now panics and dies if it cannot write to a log file, just
5246 as it does if it cannot open a log file.
5248 39. Modified OS/Makefile-Linux so that it now contains
5250 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
5252 The two -D definitions ensure that Exim is compiled with large file
5253 support, which makes it possible to handle log files that are bigger than
5256 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
5257 instance) a domain was checked against a named list that involved a lookup,
5258 causing $domain_data to be set, then another domain was checked against the
5259 same list, then the first domain was re-checked, the value of $domain_data
5260 after the final check could be wrong. In particular, if the second check
5261 failed, it could be set empty. This bug probably also applied to
5264 41. The strip_trailing_dot option was not being applied to the address given
5265 with the -f command-line option.
5267 42. The code for reading a message's header from the spool was incrementing
5268 $received_count, but never initializing it. This meant that the value was
5269 incorrect (doubled) while delivering a message in the same process in which
5270 it was received. In the most common configuration of Exim, this never
5271 happens - a fresh exec is done - but it can happen when
5272 deliver_drop_privilege is set.
5274 43. When Exim logs an SMTP synchronization error - client data sent too soon -
5275 it now includes up to 150 characters of the unexpected data in the log
5278 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
5279 and building data strings. The size of both of these buffers was 10 000
5280 bytes - far larger than anybody would *ever* want, thought I. Needless to
5281 say, somebody hit the limit. I have increased the maximum line length to
5282 20 000 and the maximum data length of concatenated lines to 100 000. I have
5283 also fixed two bugs, because there was no checking on these buffers. Tsk,
5284 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
5285 buffer is too small.
5287 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
5288 lsearch lookups. Now it does.
5290 46. When parsing a route_list item in a manualroute router, a fixed-length
5291 buffer was used for the list of hosts. I made this 1024 bytes long,
5292 thinking that nobody would ever have a list of hosts that long. Wrong.
5293 Somebody had a whole pile of complicated expansion conditions, and the
5294 string was silently truncated, leading to an expansion error. It turns out
5295 that it is easier to change to an unlimited length (owing to other changes
5296 that have happened since this code was originally written) than to build
5297 structure for giving a limitation error. The length of the item that
5298 expands into the list of hosts is now unlimited.
5300 47. The lsearch lookup could not handle data where the length of text line was
5301 more than 4095 characters. Such lines were truncated, leading to shortened
5302 data being returned. It should now handle lines of any length.
5304 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
5305 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
5308 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
5309 replacing strings such as BIN_DIRECTORY in a source file: the replacement
5310 no longer happens in comment lines. A list of replacements is now placed
5311 at the head of all of the source files, except those whose only change is
5312 to replace PERL_COMMAND in the very first #! line.
5314 50. Replaced the slow insertion sort in queue.c, for sorting the list of
5315 messages on the queue, with a bottom-up merge sort, using code contributed
5316 by Michael Haardt. This should make operations like -bp somewhat faster on
5317 large queues. It won't affect queue runners, except when queue_run_in_order
5320 51. Installed eximstats 1.31 in the distribution.
5322 52. Added support for SRV lookups to the dnslookup router.
5324 53. If an ACL referred to $message_body or $message_body_end, the value was not
5325 reset for any messages that followed in the same SMTP session.
5327 54. The store-handling optimization for building very long strings was not
5328 differentiating between the different store pools. I don't think this
5329 actually made any difference in practice, but I've tidied it.
5331 55. While running the routers to verify a sender address, $sender_address
5332 was still set to the sender address. This is wrong, because when routing to
5333 send a bounce to the sender, it would be empty. Therefore, I have changed
5334 it so that, while verifying a sender address, $sender_address is set to <>.
5335 (There is no change to what happens when verifying a recipient address.)
5337 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
5338 target A or AAAA records (if not already returned) without resetting the
5339 qualify_single or search_parents options of the DNS resolver. These are
5340 inappropriate in this case because the targets of MX and SRV records must
5341 be FQDNs. A broken DNS record could cause trouble if it happened to have a
5342 target that, when qualified, matched something in the local domain. These
5343 two options are now turned off when doing these lookups.
5345 57. It seems that at least some releases of Reiserfs (which does not have the
5346 concept of a fixed number of inodes) returns zero and not -1 for the
5347 number of available inodes. This interacted badly with check_spool_inodes,
5348 which assumed that -1 was the "no such thing" setting. What I have done is
5349 to check that the total number of inodes is greater than zero before doing
5350 the test of how many are available.
5352 58. When a "warn" ACL statement has a log_message modifier, the message is
5353 remembered, and not repeated. This is to avoid a lot of repetition when a
5354 message has many recipients that cause the same warning to be written.
5355 Howewer, Exim was preserving the list of already written lines for an
5356 entire SMTP session, which doesn't seem right. The memory is now reset if a
5357 new message is started.
5359 59. The "rewrite" debugging flag was not showing the result of rewriting in the
5360 debugging output unless log_rewrite was also set.
5362 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
5363 of (int)(handle) when we know that handle contains (void *)(-1).
5365 61. The Exim daemon panic-logs an error return when it closes the incoming
5366 connection. However "connection reset by peer" seems to be common, and
5367 isn't really an error worthy of noting specially, so that particular error
5370 62. When Exim is trying to find all the local interfaces, it used to panic and
5371 die if the ioctl to get the interface flags failed. However, it seems that
5372 on at least one OS (Solaris 9) it is possible to have an interface that is
5373 included in the list of interfaces, but for which you get a failure error
5374 for this call. This happens when the interface is not "plumbed" into a
5375 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
5376 failure of the "get flags" call assumes that the interface is down.
5378 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
5379 makes life easier for people who are doing arithmetic on fields extracted
5380 from dates, where you often get leading zeros that should not be
5381 interpreted as octal.
5383 64. Added qualify_domain to the redirect router, to override the global
5386 65. If a pathologically long header line contained very many addresses (the
5387 report of this problem mentioned 10 000) and each of them was rewritten,
5388 Exim could use up a very large amount of memory. (It kept on making new
5389 copies of the header line as it rewrote, and never released the old ones.)
5390 At the expense of a bit more processing, the header rewriting function has
5391 been changed so that it no longer eats memory in this way.
5393 66. The generation of the Received: header has been moved from the time that a
5394 message starts to be received, to the time that it finishes. The timestamp
5395 in the Received: header should now be very close to that of the <= log
5396 line. There are two side-effects of this change:
5398 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
5399 logged header lines no longer include the local Received: line, because
5400 it has not yet been created. The same applies to a copy of the message
5401 that is returned to a non-SMTP sender when a message is rejected.
5403 (b) When a filter file is tested using -bf, no additional Received: header
5404 is added to the test message. After some thought, I decided that this
5407 This change does not affect the value of $received_for. It is still set
5408 after address rewriting, but before local_scan() is called.
5410 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
5412 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
5413 gave an unhelpful panic error message, and a defer error. I have managed to
5414 change this behaviour so that it now rejects any supplied certificate,
5415 which seems right, as the list of acceptable certificates is empty.
5417 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
5418 gave an unhelpful defer error. I have not managed to make this reject any
5419 supplied certificates, but the error message it gives is "no certificate
5420 supplied", which is not helpful.
5422 70. exigrep's output now also includes lines that are not associated with any
5423 message, but which match the given pattern. Implemented by a patch from
5424 Martin Sluka, which also tidied up the Perl a bit.
5426 71. Recipient callout verification, like sender verification, was using <> in
5427 the MAIL FROM command. This isn't really the right thing, since the actual
5428 sender may affect whether the remote host accepts the recipient or not. I
5429 have changed it to use the actual sender in the callout; this means that
5430 the cache record is now keyed on a recipient/sender pair, not just the
5431 recipient address. There doesn't seem to be a real danger of callout loops,
5432 since a callout by the remote host to check the sender would use <>.
5433 [SEE ABOVE: changed after hitting problems.]
5435 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
5436 temporary errors. However, in the case of such a code being given after
5437 the end of a data transmission (i.e. after ".") Exim was failing to write
5438 a retry record for the message. (Yes, there was some broken host that was
5439 actually sending 8xx at this point.)
5441 73. An unknown lookup type in a host list could cause Exim to panic-die when
5442 the list was checked. (An example that provoked this was putting <; in the
5443 middle of a list instead of at the start.) If this happened during a DATA
5444 ACL check, a -D file could be left lying around. This kind of configuration
5445 error no longer causes Exim to die; instead it causes a defer errror. The
5446 incident is still logged to the main and panic logs.
5448 74. Buglet left over from Exim 3 conversion. The message "too many messages
5449 in one connection" was written to the rejectlog but not the mainlog, except
5450 when address rewriting (yes!) was being logged.
5452 75. Added write_rejectlog option.
5454 76. When a system filter was run not as root (that is, when system_filter_user
5455 was set), the values of the $n variables were not being returned to the
5456 main process; thus, they were not subsequently available in the $sn
5459 77. Added +return_path_on_delivery log selector.
5461 78. A connection timeout was being treated differently from recipients deferred
5462 when testing hosts_max_try with a message that was older than the host's
5463 retry timeout. (The host should not be counted, thus allowing all hosts to
5464 be tried at least once before bouncing.) This may have been the cause of an
5465 occasionally reported bug whereby a message would remain on the queue
5466 longer than the retry timeout, but would be bounced if a delivery was
5467 forced. I say "may" because I never totally pinned down the problem;
5468 setting up timeout/retry tests is difficult. See also the next item.
5470 79. The ultimate address timeout was not being applied to errors that involved
5471 a combination of host plus message (for example, a timeout on a MAIL
5472 command). When an address resolved to a number of possible hosts, and they
5473 were not all tried for each delivery (e.g. because of hosts_max_try), a
5474 message could remain on the queue longer than the retry timeout.
5476 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
5479 81. Fixed an obscure SMTP outgoing bug which required at least the following
5480 conditions: (a) there was another message waiting for the same server;
5481 (b) the server returned 5xx to all RCPT commands in the first message so
5482 that the message was not completed; (c) the server dropped the connection
5483 or gave a negative response to the RSET that Exim sends to abort the
5484 transaction. The observed case was a dropped connection after DATA that had
5485 been sent in pipelining mode. That is, the server had advertised PIPELINING
5486 but was not implementing it correctly. The effect of the bug was incorrect
5487 behaviour, such as trying another host, and this could lead to a crash.
5493 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
5494 and daemon.c were passed as pointers to ints; they should have been
5495 pointers to socklen_t variables (which are typically unsigned ints).
5497 2. Some signed/unsigned type warnings in the os.c file for Linux have been
5500 3. Fixed a really odd bug that affected only the testing scheme; patching a
5501 certain fixed string in the binary changed the value of another string that
5502 happened to be identical to the end of the original first string.
5504 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
5505 name", it returns that address as the IP address. On some operating
5506 systems (e.g. Solaris), it also passes back the IP address string as the
5507 "host name". However, on others (e.g. Linux), it passes back an empty
5508 string. Exim wasn't checking for this, and was changing the host name to an
5509 empty string, assuming it had been canonicized.
5511 5. Although rare, it is permitted to have more than one PTR record for a given
5512 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
5513 all the names associated with an address, because they do in Solaris.
5514 However, it seems that they do not in Linux for data that comes from the
5515 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
5516 I found this out when I moved to a new Linux workstation and tried to run
5517 the Exim test suite.
5519 To get round this problem I have changed the code so that it now does its
5520 own call to the DNS to look up PTR records when searching for a host name.
5521 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
5522 addresses that are only in /etc/hosts are still found.
5524 This behaviour is, however, controlled by an option called host_lookup_
5525 order, which defaults to "bydns:byaddr". If people want to use the other
5526 order, or indeed, just use one or the other means of lookup, they can
5527 specify it in this variable.
5529 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
5530 some operating systems, this comes back from gethostbyaddr() as an empty
5531 string, and this is what Exim used to test for. However, it seems that in
5532 other systems, "." is yielded. Exim now tests for this case too.
5534 7. The values of check_spool_space and check_log_space are now held internally
5535 as a number of kilobytes instead of an absolute number of bytes. If a
5536 numbers is specified without 'K' or 'M', it is rounded up to the nearest
5537 kilobyte. This means that much larger values can be stored.
5539 8. Exim monitor: an attempt to get the action menu when not actually pointing
5540 at a message produces an empty menu entitled "No message selected". This
5541 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
5542 no entries in it ("Shell widget menu has zero width and/or height"). So I
5543 have added a single, blank menu entry in this case.
5545 9. Added ${quote_local_part.
5547 10. MIME decoding is now applied to the contents of Subject: header lines when
5550 11. Now that a reference to $sender_host_address automatically causes a reverse
5551 lookup to occur if necessary (4.13/18), there is no need to arrange for a
5552 host lookup before query-style lookups in lists that might use this
5553 variable. This has therefore been abolished, and the "net-" prefix is no
5554 longer necessary for query-style lookups.
5556 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
5557 have been a typo for LFLAGS, so it has been changed.
5559 13. The install script calls Exim with "-C /dev/null" in order to find the
5560 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
5561 to be output. Howeve, since Exim outputs its version number before the
5562 error, it didn't break the script. It just looked ugly. I fixed this by
5563 always allowing "-C /dev/null" if the caller is root.
5565 14. Ignore overlarge ACL variable number when reading spool file - insurance
5566 against a later release with more variables having written the file.
5568 15. The standard form for an IPv6 address literal was being rejected by EHLO.
5569 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5570 this, as well as the form without the "IPv6" on the front.
5572 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
5573 OS/Makefile-Darwin file.
5575 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
5576 only to LDAP libraries that do not have LDAP_OPT_DEREF.
5578 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
5579 variable. However, some OS use ints for inodes. I've added cast to long int
5580 to get rid of the compiler warning.
5582 19. I had forgotten to lock out "/../" in configuration file names when
5583 ALT_CONFIG_PREFIX was set.
5585 20. Routers used for verification do not need to specify transports. However,
5586 if such a router generated a host list, and callout was configured, Exim
5587 crashed, because it could not find a port number from the (non-existent)
5588 transport. It now assumes port 25 in this circumstance.
5590 21. Added the -t option to exigrep.
5592 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
5593 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
5594 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
5595 them is set, LOOKUP_LSEARCH is forced.
5597 23. "exim -bV" now outputs a list of lookups that are included in the binary.
5599 24. Added sender and host information to the "rejected by local_scan()" log
5600 line; previously there was no indication of these.
5602 25. Added .include_if_exists.
5604 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
5605 Linux. It turns out that not all file systems support this. Apparently some
5606 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
5607 do it.) To cope with this, the error EINVAL, which means that sync-ing is
5608 not supported on the file descriptor, is now ignored when Exim is trying to
5609 sync a directory. This applies only to Linux.
5611 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
5613 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
5614 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
5617 29. Added support for maildirsize files from supplied patch (modified a bit).
5619 30. The use of :fail: followed by an empty string could lead Exim to respond to
5620 sender verification failures with (e.g.):
5622 550 Verification failed for <xxx>
5623 550 Sender verify failed
5625 where the first response line was missing the '-' that indicates it is not
5626 the final line of the response.
5628 31. The loop for finding the name of the user that called Exim had a hardwired
5629 limit of 10; it now uses the value of finduser_retries, which is used for
5630 all other user lookups.
5632 32. Added $received_count variable, available in data and not_smtp ACLs, and at
5635 33. Exim was neglecting to zero errno before one call of strtol() when
5636 expanding a string and expecting an integer value. On some systems this
5637 resulted in spurious "integer overflow" errors. Also, it was casting the
5638 result into an int without checking.
5640 34. Testing for a connection timeout using "timeout_connect" in the retry rules
5641 did not work. The code looks as if it has *never* worked, though it appears
5642 to have been documented since at least releast 1.62. I have made it work.
5644 35. The "timeout_DNS" error in retry rules, also documented since at least
5645 1.62, also never worked. As it isn't clear exactly what this means, and
5646 clearly it isn't a major issue, I have abolished the feature by treating it
5647 as "timeout", and writing a warning to the main and panic logs.
5649 36. The display of retry rules for -brt wasn't always showing the error code
5652 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
5653 timeout_connect_A, timeout_connect_MX.
5655 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
5656 to the empty sender.
5658 39. The daemon was not analysing the content of -oX till after it had closed
5659 stderr and disconnected from the controlling terminal. This meant that any
5660 syntax errors were only noted on the panic log, and the return code from
5661 the command was 0. By re-arranging the code a little, I've made the
5662 decoding happen first, so such errors now appear on stderr, and the return
5663 code is 1. However, the actual setting up of the sockets still happens in
5664 the disconnected process, so errors there are still only recorded on the
5667 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
5668 connections (as happens on some IP stacks) was logged at start up time as
5669 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
5670 it from "IPv6 and IPv4", which means that two separate sockets are being
5673 41. The debug output for gethostbyname2() or getipnodebyname() failures now
5674 says whether AF_INET or AF_INET6 was passed as an argument.
5676 42. Exiwhat output was messed up when time zones were included in log
5679 43. Exiwhat now gives more information about the daemon's listening ports,
5680 and whether -tls-on-connect was used.
5682 44. The "port" option of the smtp transport is now expanded.
5684 45. A "message" modifier in a "warn" statement in a non-message ACL was being
5685 silently ignored. Now an error message is written to the main and panic
5688 46. There's a new ACL modifier called "logwrite" which writes to a log file
5689 as soon as it is encountered.
5691 47. Added $local_user_uid and $local_user_gid at routing time.
5693 48. Exim crashed when trying to verify a sender address that was being
5696 49. Exim was recognizing only a space character after ".include". It now also
5697 recognizes a tab character.
5699 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
5700 extracting the relevant information from the specification. The man page no
5701 longer contains scrambled data for the -d option, and I've added a section
5702 at the front about calling Exim under different names.
5704 51. Added "extra_headers" argument to the "mail" command in filter files.
5706 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
5709 53. Installed eximstats 1.29.
5711 54. Added transport_filter_timeout as a generic transport option.
5713 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
5714 Cc: header lines. This was required by RFC 822, but it not required by RFC
5717 56. Exim used to add From:, Date:, and Message-Id: header lines to any
5718 incoming messages that did not have them. Now it does so only if the
5719 message originates locally, that is, if there is no associated remote host
5720 address. When Resent- header lines are present, this applies to the Resent-
5721 lines rather than the non-Resent- lines.
5723 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
5724 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
5726 58. Messages for configuration errors now include the name of the main
5727 configuration file - useful now that there may be more than one file in a
5728 list (.included file names were always shown).
5730 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
5731 for those rare installations that do not start the daemon as root or run it
5732 setuid root. I've cut out the call to initgroups() if the daemon is not
5735 60. The Exim user and group can now be bound into the binary as text strings
5736 that are looked up at the start of Exim's processing.
5738 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
5740 62. Added $mailstore_basename variable.
5742 63. Installed patch to sieve.c from Michael Haardt.
5744 64. When Exim failed to open the panic log after failing to open the main log,
5745 the original message it was trying to log was written to stderr and debug
5746 output, but if they were not available (the usual case in production), it
5747 was lost. Now it is written to syslog before the two lines that record the
5748 failures to open the logs.
5750 65. Users' Exim filters run in subprocesses under the user's uid. It is
5751 possible for a "deliver" command or an alias in a "personal" command to
5752 provoke an address rewrite. If logging of address rewriting is configured,
5753 this fails because the process is not running as root or exim. There may be
5754 a better way of dealing with this, but for the moment (because 4.30 needs
5755 to be released), I have disabled address rewrite logging when running a
5756 filter in a non-root, non-exim process.
5762 1. The buildconfig auxiliary program wasn't quoting the value set for
5763 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
5764 not defined. This bug was masked in 4.22 by the effect that was fixed in
5767 2. Some messages that were rejected after a message id was allocated were
5768 shown as "incomplete" by exigrep. It no longer does this for messages that
5769 are rejected by local_scan() or the DATA or non-SMTP ACLs.
5771 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
5772 have allow_domain_literals set, the ID did not get logged in the <= line.
5773 Domain literals are now always recognized in Message-ID: header lines.
5775 4. The first argument for a ${extract expansion item is the key name or field
5776 number. Leading and trailing spaces in this item were not being ignored,
5777 causing some misleading effects.
5779 5. When deliver_drop_privilege was set, single queue runner processes started
5780 manually (i.e. by the command "exim -q") or by the daemon (which uses the
5781 same command in the process it spins off) were not dropping privilege.
5783 6. When the daemon running as "exim" started a queue runner, it always
5784 re-executed Exim in the spun-off process. This is a waste of effort when
5785 deliver_drop_privilege is set. The new process now just calls the
5786 queue-runner function directly.
5792 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
5795 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
5796 ignored. Though the use of -oP was forcing the writing of a pid file, it
5797 was always written to the default place.
5799 3. If the message "no IP address found for host xxxx" is generated during
5800 incoming verification, it is now followed by identification of the incoming
5801 connection (so you can more easily find what provoked it).
5803 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
5805 5. Added some features to "harden" Exim a bit more against certain attacks:
5807 (a) There is now a build-time option called FIXED_NEVER_USERS that can
5808 be put in Local/Makefile. This is like the never_users runtime option,
5809 but it cannot be overridden. The default setting is "root".
5811 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
5812 prefix string with which any file named in a -C command line option
5815 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
5816 is retained for -C and -D only if the caller of Exim is root. Without
5817 it, the exim user may also use -C and -D and retain privilege.
5819 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
5820 command line option is disabled.
5822 6. Macro names set by the -D option must start with an upper case letter, just
5823 like macro names defined in the configuration file.
5825 7. Added "dereference=" facility to LDAP.
5827 8. Two instances of the typo "uknown" in the source files are fixed.
5829 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
5830 the Configure-Makefile script screwed up while processing it.
5832 10. Incorporated PCRE 4.4.
5834 11. The SMTP synchronization check was not operating right at the start of an
5835 SMTP session. For example, it could not catch a HELO sent before the client
5836 waited for the greeting. There is now a check for outstanding input at the
5837 point when the greeting is written. Because of the duplex, asynchronous
5838 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
5839 way, but not yet received, when the check is performed.
5841 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
5842 on TCP/IP sockets, because this apparently causes some broken clients to
5845 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
5846 unchanged) from the Cygwin maintainer.
5848 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
5849 was supported instead of testing for mbx. Effectively a typo.
5851 15. The spa authenticator server code was not checking that the input it
5852 received was valid base64.
5854 16. The debug output line for the "set" modifier in ACLs was not showing the
5855 name of the variable that was being set.
5857 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
5859 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
5860 lookup on its own. Something else was needed to trigger the lookup. For
5861 example, a match in host_lookup or the need for a host name in a host list.
5862 Now, if $sender_host_name is referenced and the host name has not yet been
5863 looked up, a lookup is performed. If the lookup fails, the variable remains
5864 empty, and $host_lookup_failed is set to "1".
5866 19. Added "eqi" as a case-independent comparison operator.
5868 20. The saslauthd authentication condition could segfault if neither service
5869 nor realm was specified.
5871 21. If an overflowing value such as "2048M" was set for message_size_limit, the
5872 error message that was logged was misleading, and incoming SMTP
5873 connections were dropped. The message is now more accurate, and temporary
5874 errors are given to SMTP connections.
5876 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
5877 (except RSET) with a 421 error, until QUIT is received. However, it was
5878 failing to send a response to QUIT.
5880 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
5881 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
5882 ACL is now run after the helo_try_verify_hosts code.
5884 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
5885 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
5886 case-independent, so other case variants are also recognized). Apparently
5887 some systems use these upper case variants.
5889 25. If more than two messages were waiting for the same host, and a transport
5890 filter was specified for the transport, Exim sent two messages over the
5891 same TCP/IP connection, and then failed with "socket operation on non-
5892 socket" when it tried to send the third.
5894 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
5896 27. The extern definition of crypt16() in expand.c was not being excluded when
5897 the OS had its own crypt16() function.
5899 28. Added bounce_return_body as a new option, and bounce_return_size_limit
5900 as a preferred synonym for return_size_limit, both as an option and as an
5903 29. Added LIBS=-liconv to OS/Makefile-OSF1.
5905 30. Changed the default configuration ACL to relax the local part checking rule
5906 for addresses that are not in any local domains. For these addresses,
5907 slashes and pipe symbols are allowed within local parts, but the sequence
5908 /../ is explicitly forbidden.
5910 31. SPA server authentication was not clearing the challenge buffer before
5913 32. log_message in a "warn" ACL statement was writing to the reject log as
5914 well as to the main log, which contradicts the documentation and doesn't
5915 seem right (because no rejection is happening). So I have stopped it.
5917 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
5918 However, I am unable to do any testing of this.
5920 34. Fixed an infelicity in the appendfile transport. When checking directories
5921 for a mailbox, to see if any needed to be created, it was accidentally
5922 using path names with one or more superfluous leading slashes; tracing
5923 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
5925 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
5926 contents are added to the log line that is written for every discarded
5927 recipient. (Previously a log_message setting was ignored.)
5929 36. The ${quote: operator now quotes the string if it is empty.
5931 37. The install script runs exim in order to find its version number. If for
5932 some reason other than non-existence or emptiness, which it checks, it
5933 could not run './exim', it was installing it with an empty version number,
5934 i.e. as "exim-". This error state is now caught, and the installation is
5937 38. An argument was missing from the function that creates an error message
5938 when Exim fails to connect to the socket for saslauthd authentication.
5939 This could cause Exim to crash, or give a corrupted message.
5941 39. Added isip, isip4, and isip6 to ${if conditions.
5943 40. The ACL variables $acl_xx are now saved with the message, and can be
5944 accessed later in routers, transports, and filters.
5946 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
5947 strings in the file are not string-expanded.
5949 42. If a MAIL command specified a SIZE value that was too large to fit into an
5950 int variable, the check against message_size_limit failed. Such values are
5951 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
5952 day this will have to be increased, but I don't think I want to be around
5953 when emails are that large.
5960 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
5961 iconv() is not standard in FreeBSD.
5963 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
5964 IPv6 enabled. The observed symptom was a segmentation fault on return from
5965 the function os_common_find_running_interfaces() in src/os.c.
5967 3. In the check_special_case() function in daemon.c I had used "errno" as an
5968 argument name, which causes warnings on some systems. This was basically a
5969 typo, since it was named "eno" in the comments!
5971 4. The code that waits for the clock to tick (at a resolution of some fraction
5972 of a second) so as to ensure message-id uniqueness was always waiting for
5973 at least one whole tick, when it could have waited for less. [This is
5974 almost certainly not relevant at current processor speeds, where it is
5975 unlikely to ever wait at all. But we try to future-proof.]
5977 5. The function that sleeps for a time interval that includes fractions of a
5978 second contained a race. It did not block SIGALRM between setting the
5979 timer, and suspending (a couple of lines later). If the interval was short
5980 and the sigsuspend() was delayed until after it had expired, the suspension
5981 never ended. On busy systems this could lead to processes getting stuck for
5984 6. Some uncommon configurations may cause a lookup to happen in a queue runner
5985 process, before it forks any delivery processes. The open lookup caching
5986 mechanism meant that the open file or database connection was passed into
5987 the delivery process. The problem was that delivery processes always tidy
5988 up cached lookup data. This could cause a problem for the next delivery
5989 process started by the queue runner, because the external queue runner
5990 process does not know about the closure. So the next delivery process
5991 still has data in the lookup cache. In the case of a file lookup, there was
5992 no problem because closing a file descriptor in a subprocess doesn't affect
5993 the parent. However, if the lookup was caching a connection to a database,
5994 the connection was closed, and the second delivery process was likely to
5995 see errors such as "PGSQL: query failed: server closed the connection
5996 unexpectedly". The problem has been fixed by closing all cached lookups
5997 in a queue runner before running a delivery process.
5999 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
6000 seem to have the "const" qualifier which it has on other OS. I've
6003 8. Change 4.21/2 was too strict. It is only if there are two authenticators
6004 *of the same type* (client or server) with the same public name that an
6005 error should be diagnosed.
6007 9. When Exim looked up a host name for an IP address, but failed to find the
6008 original IP address when looking up the host name (a safety check), it
6009 output the message "<ip address> does not match any IP for NULL", which was
6010 confusing, to say the least. The bug was that the host name should have
6011 appeared instead of "NULL".
6013 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
6014 user that is built into the binary, and the -C or -D options is used, root
6015 privilege is dropped before the configuration file is read. In addition,
6016 logging is switched to stderr instead of the normal log files. If the
6017 configuration then re-defines the Exim user, the unprivileged environment
6018 is probably not what is expected, so Exim logs a panic warning message (but
6021 However, if deliver_drop_privilege is set, the unprivileged state may well
6022 be exactly what is intended, so the warning has been cut out in that case,
6023 and Exim is allowed to try to write to its normal log files.
6029 1. smtp_return_error_details was not giving details for temporary sender
6030 or receiver verification errors.
6032 2. Diagnose a configuration error if two authenticators have the same public
6035 3. Exim used not to create the message log file for a message until the first
6036 delivery attempt. This could be confusing when incoming messages were held
6037 for policy or load reasons. The message log file is now created at the time
6038 the message is received, and an initial "Received" line is written to it.
6040 4. The automatically generated man page for command line options had a minor
6041 bug that caused no ill effects; however, a more serious problem was that
6042 the procedure for building the man page automatically didn't always
6043 operate. Consequently, release 4.20 contains an out-of-date version. This
6044 shouldn't happen again.
6046 5. When building Exim with embedded Perl support, the script that builds the
6047 Makefile was calling 'perl' to find its compile-time parameters, ignoring
6048 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
6050 6. The freeze_tell option was not being used for messages that were frozen on
6051 arrival, either by an ACL or by local_scan().
6053 7. Added the smtp_incomplete_transaction log selector.
6055 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
6056 was accepting AUTH without a new EHLO.
6058 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
6059 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
6062 10. Logging of TCP/IP connections (when configured) now happens in the main
6063 daemon process instead of the child process, so that the TCP/IP connection
6064 count is more accurate (but it can never be perfect).
6066 11. The use of "drop" in a nested ACL was not being handled correctly in the
6067 outer ACL. Now, if condition failure induced by the nested "drop" causes
6068 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
6069 or "require"), the connection is dropped.
6071 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
6072 that yield "discard" can now be used with an "accept" or a "discard" verb,
6073 but an error is generated for any others (because I can't see a useful way
6074 to define what should happen).
6076 13. When an ACL is read dynamically from a file (or anywhere else), the lines
6077 are now processed in the same way as lines in the Exim configuration file.
6078 In particular, continuation lines are supported.
6080 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
6082 15. Added -ti meaning -t -i.
6084 16. Check for letters, digits, hyphens, and dots in the names of dnslist
6085 domains, and warn by logging if others are found.
6087 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
6088 returned from GIFCONF when Exim is trying to find the list of interfaces on
6089 a host. The code in os.c has been modified to copy each ifreq to an aligned
6090 structure in all cases.
6092 Also, in some cases, the returned ifreq's were being copied to a 'struct
6093 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
6094 means the last couple of bytes of an IPv6 address could be chopped if the
6095 ifreq contained only a normal sockaddr (14 bytes storage).
6097 18. Named domain lists were not supported in the hosts_treat_as_local option.
6098 An entry such as +xxxx was not recognized, and was treated as a literal
6101 19. Ensure that header lines added by a DATA ACL are included in the reject log
6102 if the ACL subsequently rejects the message.
6104 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
6105 MD5 (which is deprecated).
6107 21. When testing a filter file using -bf, Exim was writing a message when it
6108 took the sender from a "From " line in the message, but it was not doing so
6109 when it took $return_path from a Return-Path: header line. It now does.
6111 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
6112 with a valid header line field name (a series of printing characters
6113 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
6115 23. Changed "disc" in the source to "disk" to conform to the documentation and
6116 the book and for uniformity.
6118 24. Ignore Sendmail's -Ooption=value command line item.
6120 25. When execve() failed while trying to run a command in a pipe transport,
6121 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
6122 could be confused with a return value of 69 from the command itself. This
6123 has been changed to 127, the value the shell returns if it is asked to run
6124 a non-existent command. The wording for the related log line suggests a
6125 non-existent command as the problem.
6127 26. If received_header_text expands to an empty string, do not add a Received:
6128 header line to the message. (Well, it adds a token one on the spool, but
6129 marks it "old" so that it doesn't get used or transmitted.)
6131 27. Installed eximstats 1.28 (addition of -nt option).
6133 28. There was no check for failure on the call to getsockname() in the daemon
6134 code. This can fail if there is a shortage of resources on the system, with
6135 ENOMEM, for example. A temporary error is now given on failure.
6137 29. Contrary to the C standard, it seems that in some environments, the
6138 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
6139 program. Exim now does this explicitly; it affects the formatting of
6140 timestamps using strftime().
6142 30. If exiqsumm was given junk data, it threw up some uninitialized variable
6143 complaints. I've now initialized all the variables, to avoid this.
6145 32. Header lines added by a system filter were not being "seen" during
6146 transport-time rewrites.
6148 33. The info_callback() function passed to OpenSSL is set up with type void
6149 (*)(SSL *, int, int), as described somewhere. However, when calling the
6150 function (actually a macro) that sets it up, the type void(*)() is
6151 expected. I've put in a cast to prevent warnings from picky compilers.
6153 34. If a DNS black list lookup found a CNAME record, but there were no A
6154 records associated with the domain it pointed at, Exim crashed.
6156 35. If a DNS black list lookup returned more than one A record, Exim ignored
6157 all but the first. It now scans all returned addresses if a particular IP
6158 value is being sought. In this situation, the contents of the
6159 $dnslist_value variable are a list of all the addresses, separated by a
6162 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
6163 to accept a host name and all its aliases if the forward lookup for any of
6164 them yielded the IP address of the incoming connection. Now it accepts only
6165 those names whose forward lookup yields the correct IP address. Any other
6166 names are discarded. This closes a loophole whereby a rogue DNS
6167 administrator could create reverse DNS records to break through a
6168 wildcarded host restriction in an ACL.
6170 37. If a user filter or a system filter that ran in a subprocess used any of
6171 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
6172 the wrong values were passed to the pipe command ($thisaddress had the
6173 value of $0, $0 had the value of $1, etc). This bug was introduced by
6174 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
6176 38. Improved the line breaking for long SMTP error messages from ACLs.
6177 Previously, if there was no break point between 40 and 75 characters, Exim
6178 left the rest of the message alone. Two changes have been made: (a) I've
6179 reduced the minimum length to 35 characters; (b) if it can't find a break
6180 point between 35 and 75 characters, it looks ahead and uses the first one
6181 that it finds. This may give the occasional overlong line, but at least the
6182 remaining text gets split now.
6184 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
6185 file descriptors might be low, and that setting 1000 would always raise it.
6186 It turns out that in some environments, the limit is already over 1000 and
6187 that lowering it causes trouble. So now Exim takes care not to decrease it.
6189 40. When delivering a message, the value of $return_path is set to $sender_
6190 address at the start of routing (routers may change the value). By an
6191 oversight, this default was not being set up when an address was tested by
6192 -bt or -bv, which affected the outcome if any router or filter referred to
6195 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
6196 log only when "message" or "log_message" are set. However, if one of the
6197 conditions was an address verification, or a call to a nested ACL, the
6198 messages generated by the underlying test were being passed through. This
6199 no longer happens. The underlying message is available in $acl_verify_
6200 message for both "message" and "log_message" expansions, so it can be
6201 passed through if needed.
6203 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
6204 new expansion $bh_ to give the encoded byte string without charset
6205 translation. Translation happens only if iconv() is available; HAVE_ICONV
6206 indicates this at build time. HEADERS_CHARSET gives the charset to
6207 translate to; headers_charset can change it in the configuration, and
6208 "headers charset" can change it in an individual filter file.
6210 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
6211 that creates RFC 2047 encoded "words" labels them as that charset instead
6212 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
6213 expansion operator; (ii) when Exim creates a From: line for a local
6214 message; (iii) when a header line is rewritten to include a "phrase" part.
6216 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
6217 buggy, causing it to skip the first lines of messages whose message ID
6218 ended in 'D'. This would not have bitten before Exim release 4.14, because
6219 message IDs were unlikely to end in 'D' before then. The effect was to have
6220 incorrect size information for certain domains.
6222 45. #include "config.h" was missing at the start of the crypt16.c module. This
6223 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
6226 46. If there was a timeout during a "random" callout check, Exim treated it as
6227 a failure of the random address, and carried on sending RSET and the real
6228 address. If the delay was just some slowness somewhere, the response to the
6229 original RCPT would be taken as a response to RSET and so on, causing
6230 mayhem of various kinds.
6232 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
6233 when I implemented it. It didn't allow for the fact that some option values
6234 may legitimatetly be negative (e.g. size_addition), and it didn't even do
6235 the right test for positive values.
6237 48. Domain names in DNS records are case-independent. Exim always looks them up
6238 in lower case. Some resolvers return domain names in exactly the case they
6239 appear in the zone file, that is, they may contain uppercase letters. Not
6240 all resolvers do this - some return always lower case. Exim was treating a
6241 change of case by a resolver as a change of domain, similar to a widening
6242 of a domain abbreviation. This triggered its re-routing code and so it was
6243 trying to route what was effectively the same domain again. This normally
6244 caused routing to fail (because the router wouldn't handle the domain
6245 twice). Now Exim checks for this case specially, and just changes the
6246 casing of the domain that it ultimately uses when it transmits the message
6249 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
6252 50. If a filter generated a file delivery with a non-absolute name (possible if
6253 no home directory exists for the router), the forbid_file option was not
6256 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
6257 the existing equality matching.
6259 52. Exim was using ints instead of ino_t variables in some places where it was
6260 dealing with inode numbers.
6262 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
6263 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
6264 called TMPDIR, and if it finds it is different, it changes its value.
6266 54. The smtp_printf() function is now made available to local_scan() so
6267 additional output lines can be written before returning. There is also an
6268 smtp_fflush() function to enable the detection of a dropped connection.
6269 The variables smtp_input and smtp_batched_input are exported to
6272 55. Changed the default runtime configuration: the message "Unknown user"
6273 has been removed from the ACL, and instead placed on the localuser router,
6274 using the cannot_route_message feature. This means that any verification
6275 failures that generate their own messages won't get overridden. Similarly,
6276 the "Unrouteable address" message that was in the ACL for unverifiable
6277 relay addresses has also been removed.
6279 56. Added hosts_avoid_esmtp to the smtp transport.
6281 57. The exicyclog script was not checking for the esoteric option
6282 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
6283 will work only if exicyclog is run under the appropriate euid.
6285 58. Following a discussion on the list, the rules by which Exim recognises line
6286 endings on incoming messages have been changed. The -dropcr and drop_cr
6287 options are now no-ops, retained only for backwards compatibility. The
6288 following line terminators are recognized: LF CRLF CR. However, special
6289 processing applies to CR:
6291 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
6292 nor a local message in the state where . is a terminator.
6294 (ii) If a bare CR is encountered in a header line, an extra space is added
6295 after the line terminator so as not to end the header. The reasoning
6296 behind this is that bare CRs in header lines are most likely either
6297 to be mistakes, or people trying to play silly games.
6299 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
6300 was being incorrectly given as 18 bytes larger than it should have been.
6301 This is a VOB (very old bug).
6303 60. This may never have affected anything current, but just in case it has:
6304 When the local host is found other than at the start of a list of hosts,
6305 the local host, those with the same MX, and any that follow, are discarded.
6306 When the list in question was part of a longer list of hosts, the following
6307 hosts (not currently being processed) were also being discarded. This no
6308 longer happens. I'm not sure if this situation could ever has previously
6311 61. Added the "/MX" feature to lists of hosts in the manualroute and query
6314 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
6315 header. This is something that is recommended in a new Internet Draft, and
6316 is something that is documented as being done by Sendmail. There are two
6317 possible values. For messages generated by the autoreply transport, Exim
6320 Auto-Submitted: auto-replied
6322 whereas for all other generated messages (e.g. bounces) it adds
6324 Auto-Submitted: auto-generated
6326 63. The "personal" condition in filters now includes a test for the
6327 Auto-Submitted: header. If it contains the string "auto-" the message it
6328 not considered personal.
6330 64. Added rcpt_include_affixes as a generic transport option.
6332 65. Added queue_only_override (default true).
6334 66. Added the syslog_duplication option.
6336 67. If what should have been the first header line of a message consisted of
6337 a space followed by a colon, Exim was mis-interpreting it as a header line.
6338 It isn't of course - it is syntactically invalid and should therefore be
6339 treated as the start of the message body. The misbehaviour could have
6340 caused a number of strange effects, including loss of data in subsequent
6341 header lines, and spool format errors.
6343 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
6344 client host had authenticated. This control can now be exercised by an ACL
6345 for more flexibility.
6347 69. By default, callouts do not happen when testing with -bh. There is now a
6348 variant, -bhc, which does actually run the callout code, including
6349 consulting and updating the callout cache.
6351 70. Added support for saslauthd authentication, courtesy of Alexander
6354 71. If statvfs() failed on the spool or log directories while checking their
6355 size for availability, Exim confusingly gave the error "space shortage".
6356 Furthermore, in debugging mode it crashed with a floating point exception.
6357 These checks are done if check_{spool,log}_{space,inodes} are set, and when
6358 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
6359 serious problem, Exim now writes to the main and panic logs when this
6360 happens, with details of the failure. It then refuses to accept the
6361 incoming message, giving the message "spool directory problem" or "log
6362 directory problem" with a 421 code for SMTP messages.
6364 72. When Exim is about to re-exec itself, it ensures that the file descriptors
6365 0, 1, and 2 exist, because some OS complain for execs without them (see
6366 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
6367 descriptors. However, the code omitted to check that the open succeeded,
6368 causing mysterious errors if for some reason the permissions on /dev/null
6369 got screwed. Now Exim writes a message to the main and panic logs, and
6370 bombs out if it can't open /dev/null.
6372 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
6373 interact so that it is all more flexible. It is supposed to remain
6374 backwards compatible. Also added extra_local_interfaces.
6376 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
6377 to bomb out with an assertion failure - to the client this appears as a
6378 connection drop. This problem occurs in the part of the code that was taken
6379 from the Samba project. Fortunately, the assertion is in a very simple
6380 function, so I have fixed this by reproducing the function inline in the
6381 one place where it is called, and arranging for authentication to fail
6382 instead of killing the process with assert().
6384 75. The SPA client code was not working when the server requested OEM rather
6385 than Unicode encoding.
6387 76. Added code to make require_files with a specific uid setting more usable in
6388 the case where statting the file as root fails - usually a non-root-mounted
6389 NFS file system. When this happens and the failure is EACCES, Exim now
6390 forks a subprocess and does the per-uid checking as the relevant uid.
6392 77. Added process_log_path.
6394 78. If log_file_path was not explicitly set, a setting of check_log_space or
6395 check_log_inodes was ignored.
6397 79. If a space check for the spool or log partitions fails, the incident is now
6398 logged. Of course, in the latter case the data may get lost...
6400 80. Added the %p formatting code to string_format() so that it can be used to
6401 print addresses in debug_print(). Adjusted all the address printing in the
6402 debugging in store.c to use %p rather than %d.
6404 81. There was a concern that a line of code in smtp_in.c could overflow a
6405 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
6406 initially expressed, the concern was not well-founded, because trailing
6407 spaces are removed early. However, if the trailing spaces were followed by
6408 a NULL, they did not get removed, so the overflow was possible. Two fixes
6411 (a) I re-wrote the offending code in a cleaner fashion.
6412 (b) If an incoming SMTP command contains a NULL character, it is rejected
6415 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
6416 runs initgroups(), so that if the Exim user is in any additional groups,
6417 they will be used during message reception.
6423 The change log for 4.20 and earlier releases has been archived.
projects / exim.git / blob