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.
115 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
118 JH/02 The smtp transport option "multi_domain" is now expanded.
120 JH/03 The smtp transport now requests PRDR by default, if the server offers
123 JH/04 Certificate name checking on server certificates, when exim is a client,
124 is now done by default. The transport option tls_verify_cert_hostnames
125 can be used to disable this per-host. The build option
126 EXPERIMENTAL_CERTNAMES is withdrawn.
128 JH/05 The value of the tls_verify_certificates smtp transport and main options
129 default to the word "system" to access the system default CA bundle.
130 For GnuTLS, only version 3.0.20 or later.
132 JH/06 Verification of the server certificate for a TLS connection is now tried
133 (but not required) by default. The verification status is now logged by
134 default, for both outbound TLS and client-certificate supplying inbound
137 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
140 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
141 Status Notification (bounce) messages are now MIME format per RFC 3464.
142 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
143 under the control of the dsn_advertise_hosts option, and routers may
144 have a dsn_lasthop option.
146 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
147 default, modifiable by a malware= option. The list separator for
148 the options can now be changed in the usual way. Bug 68.
150 JH/10 The smtp_receive_timeout main option is now expanded before use.
152 JH/11 The incoming_interface log option now also enables logging of the
153 local interface on delivery outgoing connections.
155 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
156 if the interface and destination host and port all match.
158 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
161 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
162 Patch from Andrew Lewis.
164 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
165 now supports optional time-restrictions, weighting, and priority
166 modifiers per server. Patch originally by <rommer@active.by>.
168 JH/16 The spamd_address main option now supports a mixed list of local
169 and remote servers. Remote servers can be IPv6 addresses, and
170 specify a port-range.
172 JH/17 Bug 68: The spamd_address main option now supports an optional
173 timeout value per server.
175 JH/18 Bug 1581: Router and transport options headers_add/remove can
176 now have the list separator specified.
178 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
181 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
184 JH/21 Support for the A6 type of dns record is withdrawn.
186 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
187 rather than the verbs used.
189 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
190 from 255 to 1024 chars.
192 JH/24 Verification callouts now attempt to use TLS by default.
194 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
195 are generic router options now. The defaults didn't change.
197 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
198 Original patch from Alexander Shikoff, worked over by JH.
200 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
201 files if ZCAT_COMMAND is not executable.
203 JH/26 Bug 1539: Add timout/retry options on dnsdb lookups.
205 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
207 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
208 Normally benign, it bites when the pair was led to by a CNAME;
209 modern usage is to not canoicalize the domain to a CNAME target
210 (and we were inconsistent anyway for A-only vs AAAA+A).
212 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
214 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
215 when evaluating $sender_host_dnssec.
217 JH/31 Check the HELO verification lookup for DNSSEC, adding new
218 $sender_helo_dnssec variable.
220 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
222 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
224 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
226 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
227 documented as working, but never had. Support all but $spam_report.
229 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
230 added for tls authenticator.
235 TL/01 When running the test suite, the README says that variables such as
236 no_msglog_check are global and can be placed anywhere in a specific
237 test's script, however it was observed that placement needed to be near
238 the beginning for it to behave that way. Changed the runtest perl
239 script to read through the entire script once to detect and set these
240 variables, reset to the beginning of the script, and then run through
241 the script parsing/test process like normal.
243 TL/02 The BSD's have an arc4random API. One of the functions to induce
244 adding randomness was arc4random_stir(), but it has been removed in
245 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
246 function when detected.
248 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
249 cause callback expansion.
251 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
252 syntax errors in an expansion can be treated as a string instead of
253 logging or causing an error, due to the internal use of bool_lax
254 instead of bool when processing it.
256 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
257 server certificates when making smtp deliveries.
259 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
261 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
263 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
265 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
266 Merged patch from Sebastian Wiedenroth.
268 JH/05 Fix results-pipe from transport process. Several recipients, combined
269 with certificate use, exposed issues where response data items split
270 over buffer boundaries were not parsed properly. This eventually
271 resulted in duplicates being sent. This issue only became common enough
272 to notice due to the introduction of conection certificate information,
273 the item size being so much larger. Found and fixed by Wolfgang Breyha.
275 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
276 size buffer was used, resulting in syntax errors when an expansion
279 JH/07 Add support for directories of certificates when compiled with a GnuTLS
280 version 3.3.6 or later.
282 JH/08 Rename the TPDA expermimental facility to Event Actions. The #ifdef
283 is EXPERIMENTAL_EVENT, the main-configuration and transport options
284 both become "event_action", the variables become $event_name, $event_data
285 and $event_defer_errno. There is a new variable $verify_mode, usable in
286 routers, transports and related events. The tls:cert event is now also
287 raised for inbound connections, if the main configuration event_action
290 TL/06 In test suite, disable OCSP for old versions of openssl which contained
291 early OCSP support, but no stapling (appears to be less than 1.0.0).
293 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
294 server certificate names available under the smtp transport option
295 "tls_verify_cert_hostname" now do not permit multi-component wildcard
298 JH/10 Time-related extraction expansions from certificates now use the main
299 option "timezone" setting for output formatting, and are consistent
300 between OpenSSL and GnuTLS compilations. Bug 1541.
302 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
303 encoded parameter in the incoming message. Bug 1558.
305 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
306 include certificate info, eximon was claiming there were spoolfile
309 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
311 JH/14 Log delivery-related information more consistently, using the sequence
312 "H=<name> [<ip>]" wherever possible.
314 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
315 are problematic for Debian distribution, omit them from the release
318 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
320 JH/16 Fix string representation of time values on 64bit time_t anchitectures.
323 JH/17 Fix a null-indirection in certextract expansions when a nondefault
324 output list separator was used.
329 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
330 checkers that were complaining about end of non-void function with no
333 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
334 This was a regression intruduced in 4.83 by another bugfix.
336 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
338 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
339 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
345 TF/01 Correctly close the server side of TLS when forking for delivery.
347 When a message was received over SMTP with TLS, Exim failed to clear up
348 the incoming connection properly after forking off the child process to
349 deliver the message. In some situations the subsequent outgoing
350 delivery connection happened to have the same fd number as the incoming
351 connection previously had. Exim would try to use TLS and fail, logging
352 a "Bad file descriptor" error.
354 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
355 utilities have not been installed.
357 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
358 temporary space as the ACL may create new global variables.
360 TL/01 LDAP support uses per connection or global context settings, depending
361 upon the detected version of the libraries at build time.
363 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
364 to extract and use the src ip:port in logging and expansions as if it
365 were a direct connection from the outside internet. PPv2 support was
366 updated based on HAProxy spec change in May 2014.
368 JH/02 Add ${listextract {number}{list}{success}{fail}}.
370 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
371 Properly escape header and check for NULL return.
373 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
376 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
378 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
379 characters in header names, implemented as a verify condition.
380 Contributed by Michael Fischer v. Mollard.
382 TL/05 Rename SPF condition results err_perm and err_temp to standardized
383 results permerror and temperror. Previous values are deprecated but
384 still accepted. In a future release, err_perm and err_temp will be
385 completely removed, which will be a backward incompatibility if the
386 ACL tests for either of these two old results. Patch contributed by
387 user bes-internal on the mailing list.
389 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
391 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
392 selectors, in both main and reject logs.
394 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
397 JH/07 Add malware type "sock" for talking to simple daemon.
399 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
401 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
402 routers/transports under cutthrough routing.
404 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
405 numbers. Touch up "bool" conditional to keep the same definition.
407 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
409 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
411 JH/12 Expand items in router/transport headers_add or headers_remove lists
412 individually rather than the list as a whole. Bug 1452.
414 Required for reasonable handling of multiple headers_ options when
415 they may be empty; requires that headers_remove items with embedded
416 colons must have them doubled (or the list-separator changed).
418 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
419 view the policy declared in the DMARC record. Currently, $dmarc_status
420 is a combined value of both the record presence and the result of the
423 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
425 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
426 dnslookup router and the smtp transport (applying to the forward
429 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
430 of ldap servers used for a specific lookup. Patch provided by Heiko
433 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
434 New variable $lookup_dnssec_authenticated for observability.
436 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
437 Patch submitted by Lars Timman.
439 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
441 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
442 Requires trusted mode and valid format message id, aborts otherwise.
443 Patch contributed by Heiko Schlichting.
445 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
446 certextract with support for various fields. Bug 1358.
448 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
449 is requested by default, modifiable by smtp transport option
452 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
453 operate on certificate variables to give certificate fingerprints
454 Also new ${sha256:cert_variable}.
456 JH/23 The PRDR feature is moved from being Experimental into the mainline.
458 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
459 Christian Aistleitner.
461 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
463 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
464 file. Patch from Wolfgang Breyha.
466 JH/25 Expand the coverage of the delivery $host and $host_address to
467 client authenticators run in verify callout. Bug 1476.
469 JH/26 Port service names are now accepted for tls_on_connect_ports, to
470 align with daemon_smtp_ports. Bug 72.
472 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
473 support and error reporting did not work properly.
475 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
476 and is readable. Patch from Andrew Colin Kissa.
478 TL/14 Enhance documentation of ${run expansion and how it parses the
479 commandline after expansion, particularly in the case when an
480 unquoted variable expansion results in an empty value.
482 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
484 PP/02 Fix internal collision of T_APL on systems which support RFC3123
485 by renaming away from it. Addresses GH issue 15, reported by
488 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
490 TL/15 SECURITY: prevent double expansion in math comparison functions
491 (can expand unsanitized data). Not remotely exploitable.
498 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
500 PP/02 Make -n do something, by making it not do something.
501 When combined with -bP, the name of an option is not output.
503 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
506 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
507 $sender_host_name and config options to manage this, and basic check
510 PP/05 DSCP support for outbound connections and control modifier for inbound.
512 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
513 (Only plugin which currently uses this is kerberos4, which nobody should
514 be using, but we should make it available and other future plugins might
515 conceivably use it, even though it would break NAT; stuff *should* be
516 using channel bindings instead).
518 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
519 name; added for Sendmail compatibility; requires admin caller.
520 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
521 just ignore it); requires trusted caller.
522 Also parse but ignore: -Ac -Am -X<logfile>
525 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
527 TL/02 Add +smtp_confirmation as a default logging option.
529 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
530 Patch by Magnus Holmgren from 2007-02-20.
532 TL/04 Bugzilla 1281 - Spec typo.
533 Bugzilla 1283 - Spec typo.
534 Bugzilla 1290 - Spec grammar fixes.
536 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
538 TL/06 Add Experimental DMARC support using libopendmarc libraries.
540 TL/07 Fix an out of order global option causing a segfault. Reported to dev
541 mailing list by by Dmitry Isaikin.
543 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
545 JH/02 Support "G" suffix to numbers in ${if comparisons.
547 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
549 NM/01 Bugzilla 1197 - Spec typo
550 Bugzilla 1196 - Spec examples corrections
552 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
554 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
555 gnutls_enable_pkcs11, but renamed to more accurately indicate its
558 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
559 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
561 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
562 "acl {{name}{arg}...}", and optional args on acl condition
565 JH/05 Permit multiple router/transport headers_add/remove lines.
567 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
569 JH/07 Avoid using a waiting database for a single-message-only transport.
570 Performance patch from Paul Fisher. Bugzilla 1262.
572 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
575 JH/09 Add $headers_added variable, with content from use of ACL modifier
576 add_header (but not yet added to the message). Bugzilla 199.
578 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
579 Pulled from Bugzilla 817 by Wolfgang Breyha.
581 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
583 (nb: this is the same fix as in Exim 4.80.1)
585 JH/11 Add A= logging on delivery lines, and a client_set_id option on
588 JH/12 Add optional authenticated_sender logging to A= and a log_selector
591 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
593 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
594 advertise SMTP AUTH mechanism to us, instead of a generic
595 protocol violation error. Also, make Exim more robust to bad
596 data from the Dovecot auth socket.
598 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
600 When a queue runner is handling a message, Exim first routes the
601 recipient addresses, during which it prunes them based on the retry
602 hints database. After that it attempts to deliver the message to
603 any remaining recipients. It then updates the hints database using
606 So if a recipient address works intermittently, it can get repeatedly
607 deferred at routing time. The retry hints record remains fresh so the
608 address never reaches the final cutoff time.
610 This is a fairly common occurrence when a user is bumping up against
611 their storage quota. Exim had some logic in its local delivery code
612 to deal with this. However it did not apply to per-recipient defers
613 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
615 This change adds a proper retry rule check during routing so that the
616 final cutoff time is checked against the message's age. We only do
617 this check if there is an address retry record and there is not a
618 domain retry record; this implies that previous attempts to handle
619 the address had the retry_use_local_parts option turned on. We use
620 this as an approximation for the destination being like a local
621 delivery, as in LMTP.
623 I suspect this new check makes the old local delivery cutoff check
624 redundant, but I have not verified this so I left the code in place.
626 TF/02 Correct gecos expansion when From: is a prefix of the username.
628 Test 0254 submits a message to Exim with the header
632 When I ran the test suite under the user fanf2, Exim expanded
633 the header to contain my full name, whereas it should have added
634 a Resent-Sender: header. It erroneously treats any prefix of the
635 username as equal to the username.
637 This change corrects that bug.
639 GF/01 DCC debug and logging tidyup
640 Error conditions log to paniclog rather than rejectlog.
641 Debug lines prefixed by "DCC: " to remove any ambiguity.
643 TF/03 Avoid unnecessary rebuilds of lookup-related code.
645 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
646 Bug spotted by Jeremy Harris; was flawed since initial commit.
647 Would have resulted in OCSP responses post-SNI triggering an Exim
648 NULL dereference and crash.
650 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
652 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
653 Bug detection, analysis and fix by Samuel Thibault.
654 Bugzilla 1331, Debian bug #698092.
656 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
658 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
659 Server implementation by Todd Lyons, client by JH.
660 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
661 config variable "prdr_enable" controls whether the server
662 advertises the facility. If the client requests PRDR a new
663 acl_data_smtp_prdr ACL is called once for each recipient, after
664 the body content is received and before the acl_smtp_data ACL.
665 The client is controlled by bolth of: a hosts_try_prdr option
666 on the smtp transport, and the server advertisement.
667 Default client logging of deliveries and rejections involving
668 PRDR are flagged with the string "PRDR".
670 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
671 fclose(). Diagnosis by Todd Lyons.
673 PP/17 Update configure.default to handle IPv6 localhost better.
674 Patch by Alain Williams (plus minor tweaks).
677 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
678 This is now consistent with GnuTLS, and is now documented: the
679 previous undocumented portable approach to treating the option as
680 unset was to force an expansion failure. That still works, and
681 an empty string is now equivalent.
683 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
684 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
685 not performing validation itself.
687 PP/20 Added force_command boolean option to pipe transport.
688 Patch from Nick Koston, of cPanel Inc.
690 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
693 TF/04 Added udpsend ACL modifer and hexquote expansion operator
695 PP/21 Fix eximon continuous updating with timestamped log-files.
696 Broken in a format-string cleanup in 4.80, missed when I repaired the
697 other false fix of the same issue.
698 Report and fix from Heiko Schlichting.
701 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
702 Report from Prashanth Katuri.
704 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
705 It's SecureTransport, so affects any MacOS clients which use the
706 system-integrated TLS libraries, including email clients.
708 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
709 using a MIME ACL for non-SMTP local injection.
710 Report and assistance in diagnosis by Warren Baker.
712 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
714 JH/16 Fix comparisons for 64b. Bugzilla 1385.
716 TL/09 Add expansion variable $authenticated_fail_id to keep track of
717 last id that failed so it may be referenced in subsequent ACL's.
719 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
722 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
723 ldap library initialization, allowing self-signed CA's to be
724 used. Also properly sets require_cert option later in code by
725 using NULL (global ldap config) instead of ldap handle (per
726 session). Bug diagnosis and testing by alxgomz.
728 TL/12 Enhanced documentation in the ratelimit.pl script provided in
729 the src/util/ subdirectory.
731 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
732 renamed to Transport Post Delivery Action by Jeremy Harris, as
735 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
736 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
737 redis_servers = needs to be configured which will be used by the redis
738 lookup. Patch from Warren Baker, of The Packet Hub.
740 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
742 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
743 hostname or reverse DNS when processing a host list. Used suggestions
744 from multiple comments on this bug.
746 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
748 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
749 Missed a few lines, added it to make the runtest require no keyboard
752 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
753 contains upper case chars. Make router use caseful_local_part.
755 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
756 support when GnuTLS has been built with p11-kit.
762 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
764 This, or similar/improved, will also be change PP/11 of 4.82.
770 PP/01 Handle short writes when writing local log-files.
771 In practice, only affects FreeBSD (8 onwards).
772 Bugzilla 1053, with thanks to Dmitry Isaikin.
774 NM/01 Bugzilla 949 - Documentation tweak
776 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
779 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
781 PP/02 Implemented gsasl authenticator.
783 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
785 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
786 `pkg-config foo` for cflags/libs.
788 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
789 with rest of GSASL and with heimdal_gssapi.
791 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
792 `pkg-config foo` for cflags/libs for the TLS implementation.
794 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
795 properties get this fed in as external SSF. A number of robustness
796 and debugging improvements to the cyrus_sasl authenticator.
798 PP/08 cyrus_sasl server now expands the server_realm option.
800 PP/09 Bugzilla 1214 - Log authentication information in reject log.
801 Patch by Jeremy Harris.
803 PP/10 Added dbmjz lookup type.
805 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
807 PP/12 MAIL args handles TAB as well as SP, for better interop with
808 non-compliant senders.
809 Analysis and variant patch by Todd Lyons.
811 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
812 Bug report from Lars Müller <lars@samba.org> (via SUSE),
813 Patch from Dirk Mueller <dmueller@suse.com>
815 PP/13 tls_peerdn now print-escaped for spool files.
816 Observed some $tls_peerdn in wild which contained \n, which resulted
817 in spool file corruption.
819 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
820 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
821 or write after TLS renegotiation, which otherwise led to messages
824 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
825 as a tracking header (ie: a signed header comes before the signature).
826 Patch from Wolfgang Breyha.
828 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
829 comma-sep list; embedded commas doubled.
831 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
833 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
835 Report and patch from Dmitry Banschikov.
837 PP/16 Removed "dont_insert_empty_fragments" fron "openssl_options".
838 Removed SSL_clear() after SSL_new() which led to protocol negotiation
839 failures. We appear to now support TLS1.1+ with Exim.
841 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
842 lets Exim select keys and certificates based upon TLS SNI from client.
843 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
844 before an outbound SMTP session. New log_selector, +tls_sni.
846 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
847 NULL dereference. Report and patch from Alun Jones.
849 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
850 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
851 Not seeing resolver debug output on NetBSD, but suspect this is a
852 resolver implementation change.
854 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
855 Left warnings. Added "eximon gdb" invocation mode.
857 PP/21 Defaulting "accept_8bitmime" to true, not false.
859 PP/22 Added -bw for inetd wait mode support.
861 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
862 locate the relevant includes and libraries. Made this the default.
864 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
865 Bugzilla 1246, report and most of solution from Tomasz Kusy.
867 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
868 This may cause build issues on older platforms.
870 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
871 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
872 gnutls_require_mac & gnutls_require_protocols (no longer supported).
873 Added SNI support via GnuTLS too.
874 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
876 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
878 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
879 Applied second patch from Janne, implementing suggestion to default
880 multiple-strings-in-record handling to match SPF spec.
882 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
884 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
885 read-only, out of scope).
886 Patch from Wolfgang Breyha, report from Stuart Northfield.
888 PP/29 Fix three issues highlighted by clang analyser static analysis.
889 Only crash-plausible issue would require the Cambridge-specific
890 iplookup router and a misconfiguration.
891 Report from Marcin Mirosław.
893 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
895 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
896 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
897 As part of this, removing so much warning spew let me fix some minor
898 real issues in debug logging.
900 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
901 assignment on my part. Fixed.
903 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
904 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
905 Janne Snabb (who went above and beyond: thank you).
907 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
908 string otherwise requires a connection and a bunch more work and it's
909 relatively easy to get wrong. Should also expose TLS library linkage
912 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
913 64-bit ${eval} (JH/03).
915 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
916 GNU libc to support some of the 64-bit stuff, should not lead to
917 conflicts. Defined before os.h is pulled in, so if a given platform
918 needs to override this, it can.
920 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
921 protection layer was required, which is not implemented.
922 Bugzilla 1254, patch from Wolfgang Breyha.
924 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
925 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
926 tls_dhparam take prime identifiers. Also unbreak combination of
927 OpenSSL+DH_params+TLSSNI.
929 PP/39 Disable SSLv2 by default in OpenSSL support.
935 PP/01 Solaris build fix for Oracle's LDAP libraries.
936 Bugzilla 1109, patch from Stephen Usher.
938 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
940 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
943 TF/02 Fix a couple more cases where we did not log the error message
944 when unlink() failed. See also change 4.74-TF/03.
946 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
947 lock up or crash if it happened to be inside a call to libc when it
948 got a SIGUSR1 from exiwhat.
950 The SIGUSR1 handler appends the current process status to the process
951 log which is later printed by exiwhat. It used to use the general
952 purpose logging code to do this, but several functions it calls are
953 not safe for signals.
955 The new output code in the SIGUSR1 handler is specific to the process
956 log, and simple enough that it's easy to inspect for signal safety.
957 Removing some special cases also simplifies the general logging code.
958 Removing the spurious timestamps from the process log simplifies
961 TF/04 Improved ratelimit ACL condition.
963 The /noupdate option has been deprecated in favour of /readonly which
964 has clearer semantics. The /leaky, /strict, and /readonly update modes
965 are mutually exclusive. The update mode is no longer included in the
966 database key; it just determines when the database is updated. (This
967 means that when you upgrde Exim will forget old rate measurements.)
969 Exim now checks that the per_* options are used with an update mode that
970 makes sense for the current ACL. For example, when Exim is processing a
971 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
972 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
973 must specify per_mail/readonly. If you omit the update mode it defaults to
974 /leaky where that makes sense (as before) or /readonly where required.
976 The /noupdate option is now undocumented but still supported for
977 backwards compatibility. It is equivalent to /readonly except that in
978 ACLs where /readonly is required you may specify /leaky/noupdate or
979 /strict/noupdate which are treated the same as /readonly.
981 A useful new feature is the /count= option. This is a generalization
982 of the per_byte option, so that you can measure the throughput of other
983 aggregate values. For example, the per_byte option is now equivalent
984 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
986 The per_rcpt option has been generalized using the /count= mechanism
987 (though it's more complicated than the per_byte equivalence). When it is
988 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
989 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
990 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
991 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
992 Note that using per_rcpt with a non-readonly update mode in more than
993 one ACL will cause the recipients to be double-counted. (The per_mail
994 and per_byte options don't have this problem.)
996 The handling of very low rates has changed slightly. If the computed rate
997 is less than the event's count (usually one) then this event is the first
998 after a long gap. In this case the rate is set to the same as this event's
999 count, so that the first message of a spam run is counted properly.
1001 The major new feature is a mechanism for counting the rate of unique
1002 events. The new per_addr option counts the number of different
1003 recipients that someone has sent messages to in the last time period. It
1004 behaves like per_rcpt if all the recipient addresses are different, but
1005 duplicate recipient addresses do not increase the measured rate. Like
1006 the /count= option this is a general mechanism, so the per_addr option
1007 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
1008 example, measure the rate that a client uses different sender addresses
1009 with the options per_mail/unique=$sender_address. There are further
1010 details in the main documentation.
1012 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
1014 TF/06 Removed a few PCRE remnants.
1016 TF/07 Automatically extract Exim's version number from tags in the git
1017 repository when doing development or release builds.
1019 PP/02 Raise smtp_cmd_buffer_size to 16kB.
1020 Bugzilla 879. Patch from Paul Fisher.
1022 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
1023 Heavily based on revision 40f9a89a from Simon Arlott's tree.
1026 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
1028 PP/05 Variable $av_failed, true if the AV scanner deferred.
1029 Bugzilla 1078. Patch from John Horne.
1031 PP/06 Stop make process more reliably on build failure.
1032 Bugzilla 1087. Patch from Heiko Schlittermann.
1034 PP/07 Make maildir_use_size_file an _expandable_ boolean.
1035 Bugzilla 1089. Patch from Heiko Schlittermann.
1037 PP/08 Handle ${run} returning more data than OS pipe buffer size.
1038 Bugzilla 1131. Patch from Holger Weiß.
1040 PP/09 Handle IPv6 addresses with SPF.
1041 Bugzilla 860. Patch from Wolfgang Breyha.
1043 PP/10 GnuTLS: support TLS 1.2 & 1.1.
1045 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
1048 PP/11 match_* no longer expand right-hand-side by default.
1049 New compile-time build option, EXPAND_LISTMATCH_RHS.
1050 New expansion conditions, "inlist", "inlisti".
1052 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
1054 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
1056 PP/14 fix log_write() format string regression from TF/03.
1057 Bugzilla 1152. Patch from Dmitry Isaikin.
1063 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
1065 PP/02 Harmonised TLS library version reporting; only show if debugging.
1066 Layout now matches that introduced for other libraries in 4.74 PP/03.
1068 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
1070 PP/04 New "dns_use_edns0" global option.
1072 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
1075 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
1076 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
1078 TK/01 Updated PolarSSL code to 0.14.2.
1079 Bugzilla 1097. Patch from Andreas Metzler.
1081 PP/07 Catch divide-by-zero in ${eval:...}.
1082 Fixes bugzilla 1102.
1084 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
1087 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
1088 format-string attack -- SECURITY: remote arbitrary code execution.
1090 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
1091 time unintentionally subject to list matching rules, letting the header
1092 cause arbitrary Exim lookups (of items which can occur in lists, *not*
1093 arbitrary string expansion). This allowed for information disclosure.
1095 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
1096 INT_MIN/-1 -- value coerced to INT_MAX.
1102 NM/01 Workround for PCRE version dependency in version reporting
1105 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
1106 This fixes portability to compilers other than gcc, notably
1107 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
1109 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
1110 makefiles for portability to HP-UX and POSIX correctness.
1112 PP/01 Permit LOOKUP_foo enabling on the make command-line.
1113 Also via indented variable definition in the Makefile.
1114 (Debugging by Oliver Heesakkers).
1116 PP/02 Restore caching of spamd results with expanded spamd_address.
1117 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
1119 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
1120 Improves build reliability. Fix from: Frank Elsner
1122 NM/02 Fix wide character breakage in the rfc2047 coding
1123 Fixes bug 1064. Patch from Andrey N. Oktyabrski
1125 NM/03 Allow underscore in dnslist lookups
1126 Fixes bug 1026. Patch from Graeme Fowler
1128 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
1129 Code patches from Adam Ciarcinski of NetBSD.
1131 NM/04 Fixed exiqgrep to cope with mailq missing size issue
1134 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
1135 is logged, to avoid truncation. Patch from John Horne.
1137 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
1138 Patch from Jakob Hirsch.
1140 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
1141 SQL string expansion failure details.
1142 Patch from Andrey Oktyabrski.
1144 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
1145 Patch from Simon Arlott.
1147 PP/09 New lookups functionality failed to compile on old gcc which rejects
1148 extern declarations in function scope.
1149 Patch from Oliver Fleischmann
1151 PP/10 Use sig_atomic_t for flags set from signal handlers.
1152 Check getgroups() return and improve debugging.
1153 Fixed developed for diagnosis in bug 927 (which turned out to be
1156 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
1157 Patch from Mark Zealey.
1159 PP/12 Bugzilla 1056: Improved spamd server selection.
1160 Patch from Mark Zealey.
1162 PP/13 Bugzilla 1086: Deal with maildir quota file races.
1163 Based on patch from Heiko Schlittermann.
1165 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
1166 Patch from Uwe Doering, sign-off by Michael Haardt.
1168 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
1169 variable declaration deep within a block. Bug and patch from
1172 PP/15 lookups-Makefile IRIX compatibilty coercion.
1174 PP/16 Make DISABLE_DKIM build knob functional.
1176 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
1177 Patch by Simon Arlott
1179 TF/03 Fix valgrind.h portability to C89 compilers that do not support
1180 variable argument macros. Our copy now differs from upstream.
1186 TF/01 Failure to get a lock on a hints database can have serious
1187 consequences so log it to the panic log.
1189 TF/02 Log LMTP confirmation messages in the same way as SMTP,
1190 controlled using the smtp_confirmation log selector.
1192 TF/03 Include the error message when we fail to unlink a spool file.
1194 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
1195 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
1196 for maintaining out-of-tree patches for some time.
1198 PP/01 Bugzilla 139: Documentation and portability issues.
1199 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
1200 Handle per-OS dynamic-module compilation flags.
1202 PP/02 Let /dev/null have normal permissions.
1203 The 4.73 fixes were a little too stringent and complained about the
1204 permissions on /dev/null. Exempt it from some checks.
1205 Reported by Andreas M. Kirchwitz.
1207 PP/03 Report version information for many libraries, including
1208 Exim version information for dynamically loaded libraries. Created
1209 version.h, now support a version extension string for distributors
1210 who patch heavily. Dynamic module ABI change.
1212 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
1213 privilege escalation vulnerability whereby the Exim run-time user
1214 can cause root to append content of the attacker's choosing to
1217 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
1220 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
1221 If dropping privileges for untrusted macros, we disabled normal logging
1222 on the basis that it would fail; for the Exim run-time user, this is not
1223 the case, and it resulted in successful deliveries going unlogged.
1224 Fixed. Reported by Andreas Metzler.
1230 PP/01 Date: & Message-Id: revert to normally being appended to a message,
1231 only prepend for the Resent-* case. Fixes regression introduced in
1232 Exim 4.70 by NM/22 for Bugzilla 607.
1234 PP/02 Include check_rfc2047_length in configure.default because we're seeing
1235 increasing numbers of administrators be bitten by this.
1237 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
1239 PP/03 Bugzilla 994: added openssl_options main configuration option.
1241 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
1243 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
1245 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
1247 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
1248 without a peer certificate, leading to a segfault because of an
1249 assumption that peers always have certificates. Be a little more
1250 paranoid. Problem reported by Martin Tscholak.
1252 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
1253 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
1254 NB: ClamAV planning to remove STREAM in "middle of 2010".
1255 CL also introduces -bmalware, various -d+acl logging additions and
1256 more caution in buffer sizes.
1258 PP/09 Implemented reverse_ip expansion operator.
1260 PP/10 Bugzilla 937: provide a "debug" ACL control.
1262 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
1264 PP/12 Bugzilla 973: Implement --version.
1266 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
1268 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
1270 PP/15 Bugzilla 816: support multiple condition rules on Routers.
1272 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
1273 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
1274 ignore trailing whitespace.
1276 JJ/02 prevent non-panic DKIM error from being sent to paniclog
1278 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
1281 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
1282 Notification from Dr Andrew Aitchison.
1284 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
1285 ExtendedDetectionInfo response format.
1286 Notification from John Horne.
1288 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
1291 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
1292 XSL and documented dependency on system catalogs, with examples of how
1295 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
1298 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
1299 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
1300 configuration file which is writeable by the Exim user or group.
1302 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
1303 of configuration files to cover files specified with the -C option if
1304 they are going to be used with root privileges, not just the default
1307 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
1308 option (effectively making it always true).
1310 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
1311 files to be used while preserving root privileges.
1313 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
1314 that rogue child processes cannot use them.
1316 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
1317 run-time user, instead of root.
1319 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
1320 Exim run-time user without dropping privileges.
1322 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
1323 result string, instead of calling string_vformat() twice with the same
1326 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
1327 for other users. Others should always drop root privileges if they use
1328 -C on the command line, even for a whitelisted configure file.
1330 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
1332 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
1338 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
1339 $data_path, and $header_path variables; fixed documentation bugs and
1342 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
1343 exipick to access non-standard spools, including the "frozen" queue
1346 NM/01 Bugzilla 965: Support mysql stored procedures.
1347 Patch from Alain Williams
1349 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
1351 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
1352 Patch from Andreas Metzler
1354 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
1355 Patch from Kirill Miazine
1357 NM/05 Bugzilla 671: Added umask to procmail example.
1359 JJ/03 installed exipick 20100323.0, fixing doc bug
1361 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
1362 directory. Notification and patch from Dan Rosenberg.
1364 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
1366 TK/02 Improve log output when DKIM signing operation fails.
1368 MH/01 Treat the transport option dkim_domain as a colon separated
1369 list, not as a single string, and sign the message with each element,
1370 omitting multiple occurences of the same signer.
1372 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
1373 Bugzilla 985, 986. Patch by Simon Arlott
1375 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
1376 Patch by Simon Arlott
1378 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
1379 MBX locking. Notification from Dan Rosenberg.
1385 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
1387 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
1389 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
1391 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
1393 NM/04 Bugzilla 915: Flag broken perl installation during build.
1399 TK/01 Added patch by Johannes Berg that expands the main option
1400 "spamd_address" if it starts with a dollar sign.
1402 TK/02 Write list of recipients to X-Envelope-Sender header when building
1403 the mbox-format spool file for content scanning (suggested by Jakob
1406 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
1407 (http://www.dcc-servers.net/) support via dccifd. Activated by
1408 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
1410 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
1411 by Mark Daniel Reidel <mr@df.eu>.
1413 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
1414 When building exim an external PCRE library is now needed -
1415 PCRE is a system library on the majority of modern systems.
1416 See entry on PCRE_LIBS in EDITME file.
1418 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
1419 conversation. Added nologin parameter to request.
1420 Patch contributed by Kirill Miazine.
1422 TF/01 Do not log submission mode rewrites if they do not change the address.
1424 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
1426 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
1427 log files in place. Contributed by Roberto Lima.
1429 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
1431 TF/03 Bugzilla 615: When checking the local_parts router precondition
1432 after a local_part_suffix or local_part_prefix option, Exim now
1433 does not use the address's named list lookup cache, since this
1434 contains cached lookups for the whole local part.
1436 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
1437 Robert Millan. Documentation is in experimental-spec.txt.
1439 TF/04 Bugzilla 668: Fix parallel build (make -j).
1441 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
1443 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
1444 Patch provided by Jan Srzednicki.
1446 TF/05 Leading white space used to be stripped from $spam_report which
1447 wrecked the formatting. Now it is preserved.
1449 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
1450 that they are available at delivery time.
1452 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
1454 TF/08 TLS error reporting now respects the incoming_interface and
1455 incoming_port log selectors.
1457 TF/09 Produce a more useful error message if an SMTP transport's hosts
1458 setting expands to an empty string.
1460 NM/06 Bugzilla 744: EXPN did not work under TLS.
1461 Patch provided by Phil Pennock.
1463 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
1464 Patch provided by Richard Godbee.
1466 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
1467 acl_smtp_notquit, added index entry.
1469 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
1470 Patch provided by Eugene Bujak.
1472 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
1473 accept(). Patch provided by Maxim Dounin.
1475 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
1476 Patch provided by Phil Pennock.
1478 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
1480 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
1481 Patch provided by Brad "anomie" Jorsch.
1483 NM/14 Bugzilla 622: Added timeout setting to transport filter.
1484 Patch provided by Dean Brooks.
1486 TK/05 Add native DKIM support (does not depend on external libraries).
1488 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
1489 Patch provided by Graeme Fowler.
1491 NM/16 Bugzilla 851: Documentation example syntax fix.
1493 NM/17 Changed NOTICE file to remove references to embedded PCRE.
1495 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
1498 NM/19 Bugzilla 745: TLS version reporting.
1499 Patch provided by Phil Pennock.
1501 NM/20 Bugzilla 167: bool: condition support.
1502 Patch provided by Phil Pennock.
1504 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
1505 clients. Patch provided by Phil Pennock.
1507 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
1508 Patch provided by Brad "anomie" Jorsch.
1510 NM/23 Bugzilla 687: Fix misparses in eximstats.
1511 Patch provided by Heiko Schlittermann.
1513 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
1514 Patch provided by Heiko Schlittermann.
1516 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
1517 plus update to original patch.
1519 NM/26 Bugzilla 799: Documentation correction for ratelimit.
1521 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
1522 Patch provided by David Brownlee.
1524 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
1526 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
1528 NM/30 Bugzilla 888: TLS documentation bugfixes.
1530 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
1532 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
1533 Unlike the original bugzilla I have changed all shell scripts in src tree.
1535 NM/33 Bugzilla 898: Transport filter timeout fix.
1536 Patch by Todd Rinaldo.
1538 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
1539 Patch by Serge Demonchaux.
1541 NM/35 Bugzilla 39: Base64 decode bug fixes.
1542 Patch by Jakob Hirsch.
1544 NM/36 Bugzilla 909: Correct connect() call in dcc code.
1546 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
1548 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
1550 NM/39 Bugzilla 911: Fixed MakeLinks build script.
1556 TK/01 Add preliminary DKIM support. Currently requires a forked version of
1557 ALT-N's libdkim that I have put here:
1558 http://duncanthrax.net/exim-experimental/
1560 Note to Michael Haardt: I had to rename some vars in sieve.c. They
1561 were called 'true' and it seems that C99 defines that as a reserved
1562 keyword to be used with 'bool' variable types. That means you could
1563 not include C99-style headers which use bools without triggering
1564 build errors in sieve.c.
1566 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
1567 as mailq or other aliases. Changed the --help handling significantly
1568 to do whats expected. exim_usage() emits usage/help information.
1570 SC/01 Added the -bylocaldomain option to eximstats.
1572 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
1574 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
1576 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
1582 PH/01 Another patch from the Sieve maintainer.
1584 PH/02 When an IPv6 address is converted to a string for single-key lookup
1585 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
1586 dots are used instead of colons so that keys in lsearch files need not
1587 contain colons. This was done some time before quoting was made available
1588 in lsearch files. However, iplsearch files do require colons in IPv6 keys
1589 (notated using the quote facility) so as to distinguish them from IPv4
1590 keys. This meant that lookups for IP addresses in host lists did not work
1591 for iplsearch lookups.
1593 This has been fixed by arranging for IPv6 addresses to be expressed with
1594 colons if the lookup type is iplsearch. This is not incompatible, because
1595 previously such lookups could never work.
1597 The situation is now rather anomolous, since one *can* have colons in
1598 ordinary lsearch keys. However, making the change in all cases is
1599 incompatible and would probably break a number of configurations.
1601 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
1604 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
1605 conversion specification without a maximum field width, thereby enabling
1606 a rogue spamd server to cause a buffer overflow. While nobody in their
1607 right mind would setup Exim to query an untrusted spamd server, an
1608 attacker that gains access to a server running spamd could potentially
1609 exploit this vulnerability to run arbitrary code as the Exim user.
1611 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
1612 $primary_hostname instead of what libspf2 thinks the hosts name is.
1614 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
1615 a directory entry by the name of the lookup key. Previously, if a
1616 symlink pointed to a non-existing file or a file in a directory that
1617 Exim lacked permissions to read, a lookup for a key matching that
1618 symlink would fail. Now it is enough that a matching directory entry
1619 exists, symlink or not. (Bugzilla 503.)
1621 PH/03 The body_linecount and body_zerocount variables are now exported in the
1624 PH/04 Added the $dnslist_matched variable.
1626 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
1627 This means they are set thereafter only if the connection becomes
1630 PH/06 Added the client_condition to authenticators so that some can be skipped
1631 by clients under certain conditions.
1633 PH/07 The error message for a badly-placed control=no_multiline_responses left
1634 "_responses" off the end of the name.
1636 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
1638 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
1639 (without spaces) instead of just copying the configuration text.
1641 PH/10 Added the /noupdate option to the ratelimit ACL condition.
1643 PH/11 Added $max_received_linelength.
1645 PH/12 Added +ignore_defer and +include_defer to host lists.
1647 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
1648 way in which PCRE > 7.0 is built.
1650 PH/14 Implemented queue_only_load_latch.
1652 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
1653 MAIL command. The effect was to mangle the value on 64-bit systems.
1655 PH/16 Another patch from the Sieve maintainer.
1657 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
1659 PH/18 If a system quota error occurred while trying to create the file for
1660 a maildir delivery, the message "Mailbox is full" was not appended to the
1661 bounce if the delivery eventually timed out. Change 4.67/27 below applied
1662 only to a quota excession during the actual writing of the file.
1664 PH/19 It seems that peer DN values may contain newlines (and other non-printing
1665 characters?) which causes problems in log lines. The DN values are now
1666 passed through string_printing() before being added to log lines.
1668 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
1669 and InterBase are left for another time.)
1671 PH/21 Added message_body_newlines option.
1673 PH/22 Guard against possible overflow in moan_check_errorcopy().
1675 PH/23 POSIX allows open() to be a macro; guard against that.
1677 PH/24 If the recipient of an error message contained an @ in the local part
1678 (suitably quoted, of course), incorrect values were put in $domain and
1679 $local_part during the evaluation of errors_copy.
1685 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
1686 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
1689 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
1690 issue a MAIL command.
1692 PH/02 In an ACL statement such as
1694 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
1696 if a client was not listed at all, or was listed with a value other than
1697 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
1698 the condition was not true (as it should be), so access was not denied.
1699 The bug was that the ! inversion was incorrectly passed on to the second
1700 item. This has been fixed.
1702 PH/03 Added additional dnslists conditions == and =& which are different from
1703 = and & when the dns lookup returns more than one IP address.
1705 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
1706 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
1708 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
1709 FSYNC, which compiles an option called disable_fsync that allows for
1710 bypassing fsync(). The documentation is heavily laced with warnings.
1712 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
1714 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
1715 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
1716 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
1717 including adding "make clean"; (3) Added -fPIC when compiling the test
1718 dynamically loaded module, to get rid of a warning.
1720 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
1721 message fails, move_frozen_messages = true and ignore_bounce_errors_after
1722 = 0s. The bug is otherwise harmless.
1724 PH/07 There was a bug in the dovecot authenticator such that the value of
1725 $auth1 could be overwritten, and so not correctly preserved, after a
1726 successful authentication. This usually meant that the value preserved by
1727 the server_setid option was incorrect.
1729 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
1731 PH/09 Installed PCRE release 7.0.
1733 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
1734 run for batched SMTP input. It is now run at the start of every message
1735 in the batch. While fixing this I discovered that the process information
1736 (output by running exiwhat) was not always getting set for -bs and -bS
1737 input. This is fixed, and it now also says "batched" for BSMTP.
1739 PH/11 Added control=no_pipelining.
1741 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
1742 patch, slightly modified), and move the expansion of helo_data till after
1743 the connection is made in the smtp transport (so it can use these
1746 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
1748 PH/14 Added log_selector = +pid.
1750 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
1752 PH/16 Add ${if forany and ${if forall.
1754 PH/17 Added dsn_from option to vary the From: line in DSNs.
1756 PH/18 Flush SMTP output before performing a callout, unless control =
1757 no_callout_flush is set.
1759 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
1760 was true (the default) a successful delivery failed to delete the retry
1761 item, thus causing premature timeout of the address. The bug is now
1764 PH/20 Added hosts_avoid_pipelining to the smtp transport.
1766 PH/21 Long custom messages for fakedefer and fakereject are now split up
1767 into multiline reponses in the same way that messages for "deny" and
1768 other ACL rejections are.
1770 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
1771 with slight modification.
1773 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
1774 draft, changing the syntax and factoring some duplicate code".
1776 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
1777 for deliveries of the second and subsequent messages over the same SMTP
1780 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
1781 ${reduce, with only minor "tidies".
1783 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
1785 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
1786 expansion side effects.
1788 PH/27 When a message times out after an over-quota error from an Exim-imposed
1789 quota, the bounce message says "mailbox is full". This message was not
1790 being given when it was a system quota that was exceeded. It now should
1793 MH/03 Made $recipients available in local_scan(). local_scan() already has
1794 better access to the recipient list through recipients_list[], but
1795 $recipients can be useful in postmaster-provided expansion strings.
1797 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
1798 in the case of a MAIL command with additional options following the
1799 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
1800 were accidentally chopped off.
1802 PH/29 SMTP synchronization checks are implemented when a command is read -
1803 there is a check that no more input is waiting when there shouldn't be
1804 any. However, for some commands, a delay in an ACL can mean that it is
1805 some time before the response is written. In this time, more input might
1806 arrive, invalidly. So now there are extra checks after an ACL has run for
1807 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
1808 pipelining has not been advertised.
1810 PH/30 MH's patch to allow iscntrl() characters to be list separators.
1812 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
1813 returned in the SMTP response when smtp_return_error_details was false.
1814 This has been fixed.
1816 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
1817 instead of the C I/O that was originally supplied, because problems were
1818 reported on Solaris.
1820 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
1821 Exim which did not show up earlier: it was assuming that a call to
1822 SSL_CTX_set_info_callback() might give an error value. In fact, there is
1823 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
1824 was a macro that became an assignment, so it seemed to work. This has
1825 changed to a proper function call with a void return, hence the compile
1826 error. Exim's code has been fixed.
1828 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
1831 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
1833 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
1835 PH/37 If a message is not accepted after it has had an id assigned (e.g.
1836 because it turns out to be too big or there is a timeout) there is no
1837 "Completed" line in the log. When some messages of this type were
1838 selected by exigrep, they were listed as "not completed". Others were
1839 picked up by some special patterns. I have improved the selection
1840 criteria to be more general.
1842 PH/38 The host_find_failed option in the manualroute router can now be set
1843 to "ignore", to completely ignore a host whose IP address cannot be
1844 found. If all hosts are ignored, the behaviour is controlled by the new
1845 host_all_ignored option.
1847 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
1848 homing or because of multiple MX records with /mx) generated more than
1849 one IP address, and the following item turned out to be the local host,
1850 all the secondary addresses of the first item were incorrectly removed
1851 from the list, along with the local host and any following hosts (which
1852 is what is supposed to happen).
1854 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
1855 whoever called Exim into the -H file. In the case of the daemon it was
1856 behaving confusingly. When first started, it used values for whoever
1857 started the daemon, but after a SIGHUP it used the Exim user (because it
1858 calls itself on a restart). I have changed the code so that it now always
1861 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
1862 message are rejected with the same error (e.g. no authentication or bad
1863 sender address), and a DATA command is nevertheless sent (as can happen
1864 with PIPELINING or a stupid MUA), the error message that was given to the
1865 RCPT commands is included in the rejection of the DATA command. This is
1866 intended to be helpful for MUAs that show only the final error to their
1869 PH/42 Another patch from the Sieve maintainer.
1871 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
1872 Eximstats - Fixed some broken HTML links and added missing column headers
1874 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
1875 columns for Rejects, Temp Rejects, Ham, and Spam rows.
1877 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
1879 PH/43 Yet another patch from the Sieve maintainer.
1881 PH/44 I found a way to check for a TCP/IP connection going away before sending
1882 the response to the final '.' that terminates a message, but only in the
1883 case where the client has not sent further data following the '.'
1884 (unfortunately, this is allowed). However, in many cases there won't be
1885 any further data because there won't be any more messages to send. A call
1886 to select() can be used: if it shows that the input is "ready", there is
1887 either input waiting, or the socket has been closed. An attempt to read
1888 the next input character can distinguish the two cases. Previously, Exim
1889 would have sent an OK response which the client would never have see.
1890 This could lead to message repetition. This fix should cure that, at
1891 least in a lot of common cases.
1893 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
1894 advertised in response to EHLO.
1900 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
1901 fixed by 4.65/MH/01 (is this a record?) are fixed:
1903 (i) An empty string was always treated as zero by the numeric comparison
1904 operators. This behaviour has been restored.
1906 (ii) It is documented that the numeric comparison operators always treat
1907 their arguments as decimal numbers. This was broken in that numbers
1908 starting with 0 were being interpreted as octal.
1910 While fixing these problems I realized that there was another issue that
1911 hadn't been noticed. Values of message_size_limit (both the global option
1912 and the transport option) were treated as octal if they started with 0.
1913 The documentation was vague. These values are now always treated as
1914 decimal, and I will make that clear in the documentation.
1920 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
1921 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
1924 MH/01 Don't check that the operands of numeric comparison operators are
1925 integers when their expansion is in "skipping" mode (fixes bug
1926 introduced by 4.64-PH/07).
1928 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
1929 child addresses, Exim now panics and dies. Previously, because the count
1930 is held in a short int, deliveries were likely to be lost. As such a
1931 large number of recipients for a single message is ridiculous
1932 (performance will be very, very poor), I have chosen to impose a limit
1933 rather than extend the field.
1939 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
1940 leftover -K file (the existence of which was triggered by #402).
1941 While we were at it, introduced process PID as part of the -K
1942 filename. This should rule out race conditions when creating
1945 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
1946 processing considerably. Previous code took too long for large mails,
1947 triggering a timeout which in turn triggers #401.
1949 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
1950 in the DK code in transports.c. sendfile() is not really portable,
1951 hence the _LINUX specificness.
1953 TF/01 In the add_headers option to the mail command in an Exim filter,
1954 there was a bug that Exim would claim a syntax error in any
1955 header after the first one which had an odd number of characters
1958 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
1959 callout verification, Exim cached a "reject" for the entire domain. This
1960 is correct for most verifications, but it is not correct for a recipient
1961 verification with use_sender or use_postmaster set, because in that case
1962 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
1963 case of MAIL FROM:<> rejection from other early rejections (e.g.
1964 rejection of HELO). When verifying a recipient using a non-null MAIL
1965 address, the cache is ignored if it shows MAIL FROM:<> rejection.
1966 Whatever the result of the callout, the value of the domain cache is
1967 left unchanged (for any other kind of callout, getting as far as trying
1968 RCPT means that the domain itself is ok).
1970 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
1973 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
1974 manifest itself as EPIPE rather than ECONNECT. When tidying away a
1975 session, the daemon ignores ECONNECT errors and logs others; it now
1976 ignores EPIPE as well.
1978 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
1979 (quoted-printable decoding).
1981 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
1982 later the small subsequent patch to fix an introduced bug.
1984 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
1986 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
1988 PH/08 An error is now given if message_size_limit is specified negative.
1990 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
1991 to be given (somewhat) arbitrary names.
1993 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
1996 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
1997 miscellaneous code fixes
1999 PH/10 Added the log_reject_target ACL modifier to specify where to log
2002 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
2003 hostname. This is wrong, because it relates to the incoming message (and
2004 probably the interface on which it is arriving) and not to the outgoing
2005 callout (which could be using a different interface). This has been
2006 changed to use the value of the helo_data option from the smtp transport
2007 instead - this is what is used when a message is actually being sent. If
2008 there is no remote transport (possible with a router that sets up host
2009 addresses), $smtp_active_hostname is used.
2011 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
2012 tweaks were necessary in order to get it to work (see also 21 below):
2013 (a) The code assumed that strncpy() returns a negative number on buffer
2014 overflow, which isn't the case. Replaced with Exim's string_format()
2016 (b) There were several signed/unsigned issues. I just did the minimum
2017 hacking in of casts. There is scope for a larger refactoring.
2018 (c) The code used strcasecmp() which is not a standard C function.
2019 Replaced with Exim's strcmpic() function.
2020 (d) The code set only $1; it now sets $auth1 as well.
2021 (e) A simple test gave the error "authentication client didn't specify
2022 service in request". It would seem that Dovecot has changed its
2023 interface. Fortunately there's a specification; I followed it and
2024 changed what the client sends and it appears to be working now.
2026 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
2029 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
2030 address A is aliased to B and C, where B exists and C does not. Without
2031 -v the output is "A verified" because verification stops after a
2032 successful redirection if more than one address is generated. However,
2033 with -v the child addresses are also verified. Exim was outputting "A
2034 failed to verify" and then showing the successful verification for C,
2035 with its parentage. It now outputs "B failed to verify", showing B's
2036 parentage before showing the successful verification of C.
2038 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
2039 look up a TXT record in a specific list after matching in a combined
2042 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
2043 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
2044 they consult the DNS. I had assumed they would set it the way they
2045 wanted; and indeed my experiments on Linux seem to show that in some
2046 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
2047 To be on the safe side, however, I have now made the interface to
2048 host_find_byname() similar to host_find_bydns(), with an argument
2049 containing the DNS resolver options. The host_find_byname() function now
2050 sets these options at its start, just as host_find_bydns() does. The smtp
2051 transport options dns_qualify_single and dns_search_parents are passed to
2052 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
2053 of host_find_byname() use the default settings of RES_DEFNAMES
2054 (qualify_single) but not RES_DNSRCH (search_parents).
2056 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
2057 spool_read_header() do less string testing, by means of a preliminary
2058 switch on the second character of optional "-foo" lines. (This is
2059 overdue, caused by the large number of possibilities that now exist.
2060 Originally there were few.) While I was there, I also converted the
2061 str(n)cmp tests so they don't re-test the leading "-" and the first
2062 character, in the hope this might squeeze out yet more improvement.
2064 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
2065 flag allowing group syntax was set by the header_syntax check but not
2066 turned off, possible causing trouble later; (2) The flag was not being
2067 set at all for the header_verify test, causing "group"-style headers to
2068 be rejected. I have now set it in this case, and also caused header_
2069 verify to ignore an empty address taken from a group. While doing this, I
2070 came across some other cases where the code for allowing group syntax
2071 while scanning a header line wasn't quite right (mostly, not resetting
2072 the flag correctly in the right place). These bugs could have caused
2073 trouble for malformed header lines. I hope it is now all correct.
2075 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
2076 with the "reply" argument non-NULL. The code, however (which originally
2077 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
2078 but it didn't always do it. This confused somebody who was copying the
2079 code for some other use. I have removed all the tests.
2081 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
2082 feature that was used to support insecure browsers during the U.S. crypto
2083 embargo. It requires special client support, and Exim is probably the
2084 only MTA that supported it -- and would never use it because real RSA is
2085 always available. This code has been removed, because it had the bad
2086 effect of slowing Exim down by computing (never used) parameters for the
2087 RSA_EXPORT functionality.
2089 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
2090 authenticator to fail if there's a tab character in the incoming data
2091 (there should never be unless someone is messing about, as it's supposed
2092 to be base64-encoded). Also added, on Timo's advice, the "secured" option
2093 if the connection is using TLS or if the remote IP is the same as the
2094 local IP, and the "valid-client-cert option" if a client certificate has
2097 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
2098 authenticators. This can be used for authorization after authentication
2099 succeeds. (In the case of plaintext, it servers for both authentication
2102 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
2103 if any retry times were supplied.
2105 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
2106 connection, where there is no client IP address to check. In this
2107 situation, the verify now always succeeds.
2109 PH/25 Applied John Jetmore's -Mset patch.
2111 PH/26 Added -bem to be like -Mset, but loading a message from a file.
2113 PH/27 In a string expansion for a processed (not raw) header when multiple
2114 headers of the same name were present, leading whitespace was being
2115 removed from all of them, but trailing whitespace was being removed only
2116 from the last one. Now trailing whitespace is removed from each header
2117 before concatenation. Completely empty headers in a concatenation (as
2118 before) are ignored.
2120 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
2121 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
2123 PH/29 [Removed. This was a change that I later backed out, and forgot to
2124 correct the ChangeLog entry (that I had efficiently created) before
2125 committing the later change.]
2127 PH/30 Exim was sometimes attempting to deliver messages that had suffered
2128 address errors (4xx response to RCPT) over the same connection as other
2129 messages routed to the same hosts. Such deliveries are always "forced",
2130 so retry times are not inspected. This resulted in far too many retries
2131 for the affected addresses. The effect occurred only when there were more
2132 hosts than the hosts_max_try setting in the smtp transport when it had
2133 the 4xx errors. Those hosts that it had tried were not added to the list
2134 of hosts for which the message was waiting, so if all were tried, there
2135 was no problem. Two fixes have been applied:
2137 (i) If there are any address or message errors in an SMTP delivery, none
2138 of the hosts (tried or untried) are now added to the list of hosts
2139 for which the message is waiting, so the message should not be a
2140 candidate for sending over the same connection that was used for a
2141 successful delivery of some other message. This seems entirely
2142 reasonable: after all the message is NOT "waiting for some host".
2143 This is so "obvious" that I'm not sure why it wasn't done
2144 previously. Hope I haven't missed anything, but it can't do any
2145 harm, as the worst effect is to miss an optimization.
2147 (ii) If, despite (i), such a delivery is accidentally attempted, the
2148 routing retry time is respected, so at least it doesn't keep
2149 hammering the server.
2151 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
2152 in ${readsocket because some servers need this prod.
2154 PH/32 Added some extra debug output when updating a wait-xxx database.
2156 PH/33 The hint "could be header name not terminated by colon", which has been
2157 given for certain expansion errors for a long time, was not being given
2158 for the ${if def:h_colon_omitted{... case.
2160 PH/34 The spec says: "With one important exception, whenever a domain list is
2161 being scanned, $domain contains the subject domain." There was at least
2162 one case where this was not true.
2164 PH/35 The error "getsockname() failed: connection reset by peer" was being
2165 written to the panic log as well as the main log, but it isn't really
2166 panic-worthy as it just means the connection died rather early on. I have
2167 removed the panic log writing for the ECONNRESET error when getsockname()
2170 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
2171 runs only) independently of the message's sender address. This meant
2172 that, if the 4xx error was in fact related to the sender, a different
2173 message to the same recipient with a different sender could confuse
2174 things. In particualar, this can happen when sending to a greylisting
2175 server, but other circumstances could also provoke similar problems.
2176 I have changed the default so that the retry time for these errors is now
2177 based a combination of the sender and recipient addresses. This change
2178 can be overridden by setting address_retry_include_sender=false in the
2181 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
2182 remote server are returned as part of bounce messages. This was not
2183 happening for LMTP over a pipe (the lmtp transport), but now it is the
2184 same for both kinds of LMTP.
2186 PH/38 Despite being documented as not happening, Exim was rewriting addresses
2187 in header lines that were in fact CNAMEs. This is no longer the case.
2189 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
2190 and queue runs started by the daemon processed all messages. This has
2191 been fixed so that -R and -S can now usefully be given with -q<time>.
2193 PH/40 Import PCRE release 6.7 (fixes some bugs).
2195 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
2197 PH/42 Give an error if -q is specified more than once.
2199 PH/43 Renamed the variables $interface_address and $interface_port as
2200 $received_ip_address and $received_port, to make it clear that these
2201 values apply to message reception, and not to the outgoing interface when
2202 a message is delivered. (The old names remain recognized, of course.)
2204 PH/44 There was no timeout on the connect() call when using a Unix domain
2205 socket in the ${readsocket expansion. There now is.
2207 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
2208 be meaningful with "accept".
2210 SC/01 Eximstats V1.43
2211 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
2213 SC/02 Eximstats V1.44
2214 Use a glob alias rather than an array ref in the generated
2215 parser. This improves both readability and performance.
2217 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
2218 Collect SpamAssassin and rejection statistics.
2219 Don't display local sender or destination tables unless
2220 there is data to show.
2221 Added average volumes into the top table text output.
2223 SC/04 Eximstats V1.46
2224 Collect data on the number of addresses (recipients)
2225 as well as the number of messages.
2227 SC/05 Eximstats V1.47
2228 Added 'Message too big' to the list of mail rejection
2229 reasons (thanks to Marco Gaiarin).
2231 SC/06 Eximstats V1.48
2232 Mainlog lines which have GMT offsets and are too short to
2233 have a flag are now skipped.
2235 SC/07 Eximstats V1.49 (Alain Williams)
2236 Added the -emptyok flag.
2238 SC/08 Eximstats V1.50
2239 Fixes for obtaining the IP address from reject messages.
2241 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
2242 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
2243 whitesspace changes from 4.64-PH/27
2245 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
2248 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
2249 are found, allow negative numbers in numeric criteria)
2251 JJ/06 exipick.20061117.2, added new $message_body_missing variable
2253 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
2254 to match changes made in 4.64-PH/43
2256 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
2258 PH/47 Put in an explicit test for a DNS lookup of an address record where the
2259 "domain" is actually an IP address, and force a failure. This locks out
2260 those revolvers/nameservers that support "A-for-A" lookups, in
2261 contravention of the specifications.
2263 PH/48 When a host name was looked up from an IP address, and the subsequent
2264 forward lookup of the name timed out, the host name was left in
2265 $sender_host_name, contrary to the specification.
2267 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
2268 restricted to single-key lookups, Exim was not diagnosing an error if
2269 * or *@ was used with a query-style lookup.
2271 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
2273 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
2274 long ago, but noone interested enough thought of it. Let's just say that
2275 the "1.1" means that there are some new functions that weren't there at
2276 some point in the past.
2278 PH/51 Error processing for expansion failure of helo_data from an smtp
2279 transport during callout processing was broken.
2281 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
2282 tested/used via the -bh/-bhc/-bs options.
2284 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
2285 bug, fixed in subsequent PCRE releases).
2287 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
2288 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
2290 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
2296 SC/01 Use a glob alias rather than an array ref in eximstats generated
2297 parser. This improves both readability and performance.
2299 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
2300 Don't display local sender or destination tables in eximstats unless
2301 there is data to show.
2302 Added average volumes into the eximstats top table text output.
2304 SC/03 Collect data on the number of addresses (recipients) as well
2305 as the number of messages in eximstats.
2307 TF/01 Correct an error in the documentation for the redirect router. Exim
2308 does (usually) call initgroups() when daemonizing.
2310 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
2311 with consistent privilege compared to when running as a daemon.
2313 TF/03 Note in the spec that $authenticated_id is not set for local
2314 submissions from trusted users.
2316 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
2317 Thanks to Dean Brooks <dean@iglou.com> for the patch.
2319 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
2320 by adding some example configuration directives to the default
2321 configuration file. A little bit of work is required to uncomment the
2322 directives and define how usernames and passwords are checked, but
2323 there is now a framework to start from.
2325 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
2326 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
2327 without this. I don't know how relevant this is to other LDAP libraries.
2329 PH/02 Add the verb name to the "unknown ACL verb" error.
2331 PH/03 Magnus Holmgren's patch for filter_prepend_home.
2333 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
2335 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
2336 directory not expanded when it should be if an expanded home directory
2337 was set for the address (which is overridden by the transport).
2339 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
2342 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
2343 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
2344 because it is too late at that time, and has no effect.
2346 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
2347 security issue with \' (bugzilla #107). I could not use the
2348 PQescapeStringConn() function, because it needs a PGconn value as one of
2351 PH/08 When testing addresses using -bt, indicate those final addresses that
2352 are duplicates that would not cause an additional delivery. At least one
2353 person was confused, thinking that -bt output corresponded to deliveries.
2354 (Suppressing duplicates isn't a good idea as you lose the information
2355 about possibly different redirections that led to the duplicates.)
2357 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
2358 systems where poll() doesn't work, in particular OS X.
2360 PH/10 Added more information to debugging output for retry time not reached.
2362 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
2363 operations in malware.c.
2365 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
2368 PH/13 If write_rejectlog was set false when logging was sent to syslog with
2369 syslog_duplication set false, log lines that would normally be written
2370 both the the main log and to the reject log were not written to syslog at
2373 PH/14 In the default configuration, change the use of "message" in ACL warn
2374 statements to "add_header".
2376 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
2377 not followed by a command (e.g. "seen endif").
2379 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
2380 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
2383 PH/17 Added extra conditions to the default value of delay_warning_condition
2387 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
2388 { match{$h_precedence:}{(?i)bulk|list|junk} } \
2389 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
2392 The Auto-Submitted: and various List- headers are standardised, whereas I
2393 don't think Precedence: ever was.
2395 PH/18 Refactored debugging code in route_finduser() to show more information,
2396 in particular, the error code if getpwnam() issues one.
2398 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
2399 This is apparently needed in addition to the PH/07 change above to avoid
2400 any possible encoding problems.
2402 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
2403 but not after initializing Perl.
2405 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
2406 output them only if debugging. By default they are written stderr,
2407 apparently, which is not desirable.
2409 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
2412 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
2415 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
2417 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
2418 authenticated or an ident call has been made. Suppress the default
2419 values for $authenticated_id and $authenticated_sender (but permit -oMai
2420 and -oMas) when testing with -bh.
2422 PH/24 Re-jigged the order of the tests in the default configuration so that the
2423 tests for valid domains and recipients precede the DNS black list and CSA
2424 tests, on the grounds that those ones are more expensive.
2426 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
2427 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
2428 This bug exists in every version of Exim that I still have, right back to
2431 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
2432 However, an attempt to turn on case-sensitivity in a regex key by
2433 including (?-i) didn't work because the subject string was already
2434 lowercased, and the effects were non-intuitive. It turns out that a
2435 one-line patch can be used to allow (?-i) to work as expected.
2441 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
2442 other effects) broke the use of negated acl sub-conditions.
2444 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
2447 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
2448 "Deny" causes Exim to reject the incoming connection with a 554 error.
2449 Unfortunately, if there is a major crisis, such as a disk failure,
2450 tcp-wrappers gives "deny", whereas what one would like would be some
2451 kind of temporary error. A kludge has been added to help with this.
2452 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
2453 554 error is used if errno is still zero or contains ENOENT (which occurs
2454 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
2457 PH/03 Add -lutil to the default FreeBSD LIBS setting.
2459 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
2460 errors. Otherwise a message that provokes a temporary error (when other
2461 messages do not) can cause a whole host to time out.
2463 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
2464 addresses were routed directly to files or pipes from a redirect router.
2465 File deliveries just didn't batch; pipe deliveries might have suffered
2468 PH/06 A failure to get a lock for a hints database would erroneously always say
2469 "Failed to get write lock", even when it was really a read lock.
2471 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
2472 of 0600. This has been changed to use the value of the lockfile_mode
2473 option (which defaults to 0600).
2475 PH/08 Applied small patch from the Sieve maintainer.
2477 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
2478 folder from quota calculations, a direct delivery into this folder messed
2479 up the contents of the maildirsize file. This was because the regex was
2480 used only to exclude .Trash (or whatever) when the size of the mailbox
2481 was calculated. There was no check that a delivery was happening into an
2482 excluded directory. This bug has been fixed by ignoring all quota
2483 processing for deliveries into excluded directories.
2485 PH/10 Added the maildirfolder_create_regex option to appendfile.
2491 PH/01 The code for finding all the local interface addresses on a FreeBSD
2492 system running IPv6 was broken. This may well have applied to all BSD
2493 systems, as well as to others that have similar system calls. The broken
2494 code found IPv4 interfaces correctly, but gave incorrect values for the
2495 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
2496 that it would not match correctly against @[] and not recognize the IPv6
2499 PH/02 The ipliteral router was not recognizing addresses of the form user@
2500 [ipv6:....] because it didn't know about the "ipv6:" prefix.
2502 PH/03 Added disable_ipv6.
2504 PH/04 Changed $reply_address to use the raw form of the headers instead of the
2505 decoded form, because it is most often used to construct To: headers
2506 lines in autoreplies, and the decoded form may well be syntactically
2507 invalid. However, $reply_address has leading white space removed, and all
2508 newlines turned into spaces so that the autoreply transport does not
2511 PH/05 If group was specified without a user on a router, and no group or user
2512 was specified on a transport, the group from the router was ignored.
2514 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
2515 for visible compile-time settings that can be used to change these
2516 numbers, for those that want even more. Backwards compatibility with old
2517 spool files has been maintained. However, going back to a previous Exim
2518 release will lost any variables that are in spool files.
2520 PH/07 Two small changes when running in the test harness: increase delay when
2521 passing a TCP/IP connection to a new process, in case the original
2522 process has to generate a bounce, and remove special handling of
2523 127.0.0.2 (sic), which is no longer necessary.
2525 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
2526 be the same on different OS.
2528 PH/09 Moved a debug statement in filter processing to avoid a race problem when
2531 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
2532 whether --show-vars was specified or not
2534 JJ/02 exipick: Added support for new ACL variable spool format introduced
2537 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
2538 syntactically invalid From: or Reply-to: line, and a filter used this to
2539 generate an autoreply, and therefore failed to obtain an address for the
2540 autoreply, Exim could try to deliver to a non-existent relative file
2541 name, causing unrelated and misleading errors. What now happens is that
2542 it logs this as a hard delivery error, but does not attempt to create a
2545 PH/11 The exinext utility has a -C option for testing purposes, but although
2546 the given file was scanned by exinext itself; it wasn't being passed on
2547 when Exim was called.
2549 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
2550 an end-of-file indication when reading a command response.
2552 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
2553 compiled. In many other places in Exim, IPv6 addresses are always
2554 recognized, so I have changed this. It also means that IPv4 domain
2555 literals of the form [IPV4:n.n.n.n] are now always recognized.
2557 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
2558 used if the router is not running as root, for example, when verifying at
2559 ACL time, or when using -bh. The debugging output from this situation was
2560 non-existent - all you got was a failure to exec. I have made two
2563 (a) Failures to set uid/gid, the current directory, or a process leader
2564 in a subprocess such as that created by queryprogram now generate
2565 suitable debugging ouput when -d is set.
2567 (b) The queryprogram router detects when it is not running as root,
2568 outputs suitable debugging information if -d is set, and then runs
2569 the subprocess without attempting to change uid/gid.
2571 PH/15 Minor change to Makefile for building test_host (undocumented testing
2574 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
2575 additional section of a DNS packet that returns MX or SRV records.
2576 Instead, it always explicitly searches for A/AAAA records. This avoids
2577 major problems that occur when a DNS server includes only records of one
2578 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
2579 fixed another bug: if SRV records were looked up and the corresponding
2580 address records were *not* found in the additional section, the port
2581 values from the SRV records were lost.
2583 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
2584 using the correct key (the original address) when searching the retry
2585 rules in order to find which one to use for generating the retry hint.
2587 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
2588 adding the default one. Similarly, if it contains a Reply-To: header, the
2589 errors_reply_to option, if set, is not used.
2591 PH/19 When calculating a retry time, Exim used to measure the "time since
2592 failure" by looking at the "first failed" field in the retry record. Now
2593 it does not use this if it is later than than the arrival time of the
2594 message. Instead it uses the arrival time. This makes for better
2595 behaviour in cases where some deliveries succeed, thus re-setting the
2596 "first failed" field. An example is a quota failure for a huge message
2597 when small messages continue to be delivered. Without this change, the
2598 "time since failure" will always be short, possible causing more frequent
2599 delivery attempts for the huge message than are intended.
2600 [Note: This change was subsequently modified - see PH/04 for 4.62.]
2602 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
2603 $1, $2, $3) because the numerical variables can be reset during some
2604 expansion items (e.g. "match"), thereby losing the authentication data.
2606 PH/21 Make -bV show the size of off_t variables so that the test suite can
2607 decide whether to run tests for quotas > 2G.
2609 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
2610 mailbox_size, and mailbox_filecount in the appendfile transport. If a
2611 filecount value is greater than 2G or if a quota value is greater than 2G
2612 on a system where the size of off_t is not greater than 4, a panic error
2615 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
2616 never match. The debug and -bh output now contains an explicit error
2617 message indicating a malformed IPv4 address or mask.
2619 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
2620 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
2621 PH/23 above applies.
2623 PH/25 Do not write to syslog when running in the test harness. The only
2624 occasion when this arises is a failure to open the main or panic logs
2625 (for which there is an explicit test).
2627 PH/26 Added the /no_tell option to "control=freeze".
2629 PH/27 If a host name lookup failed very early in a connection, for example, if
2630 the IP address matched host_lookup and the reverse lookup yielded a name
2631 that did not have a forward lookup, an error message of the form "no IP
2632 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
2633 could be logged. Now it outputs the IP address instead of "NULL".
2635 PH/28 An enabling patch from MH: add new function child_open_exim2() which
2636 allows the sender and the authenticated sender to be set when
2637 submitting a message from within Exim. Since child_open_exim() is
2638 documented for local_scan(), the new function should be too.
2640 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
2641 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
2642 results in an empty string is now treated as unset.
2644 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
2646 PH/31 Added sender_verify_fail logging option.
2648 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
2649 needed by RFC 822 but not by RFC 2822 was commented out. I have now
2650 tidied the source and removed it altogether.
2652 PH/33 When a queue run was abandoned because the load average was too high, a
2653 log line was always written; now it is written only if the queue_run log
2654 selector is set. In addition, the log line for abandonment now contains
2655 information about the queue run such as the pid. This is always present
2656 in "start" and "stop" lines but was omitted from the "abandon" line.
2658 PH/34 Omit spaces between a header name and the colon in the error message that
2659 is given when verify = headers_syntax fails (if there are lots of them,
2660 the message gets confusing).
2662 PH/35 Change the default for dns_check_names_pattern to allow slashes within
2663 names, as there are now some PTR records that contain slashes. This check
2664 is only to protect against broken name servers that fall over on strange
2665 characters, so the fact that it applies to all lookups doesn't matter.
2667 PH/36 Now that the new test suite is complete, we can remove some of the
2668 special code in Exim that was needed for the old test suite. For example,
2669 sorting DNS records because real resolvers return them in an arbitrary
2670 order. The new test suite's fake resolver always returns records in the
2673 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
2674 bounces) except when queue_only is set, to avoid logging races between
2675 the different processes.
2677 PH/38 Panic-die if .include specifies a non-absolute path.
2679 PH/39 A tweak to the "H" retry rule from its user.
2681 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
2682 a label. They prevented compilation on older perls.
2684 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
2685 a warning to be raised on newish perls.
2687 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
2688 on queue. Changes to match documented behaviour of showing count of
2689 messages matching specified criteria.
2691 PH/40 Changed the default ident timeout from 30s to 5s.
2693 PH/41 Added support for the use of login_cap features, on those BSD systems
2694 that have them, for controlling the resources used by pipe deliveries.
2696 PH/42 The content-scanning code uses fopen() to create files in which to put
2697 message data. Previously it was not paying any attention to the mode of
2698 the files. Exim runs with umask(0) because the rest of the code creates
2699 files with open(), and sets the required mode explicitly. Thus, these
2700 files were ending up world-writeable. This was not a big issue, because,
2701 being within the spool directory, they were not world-accessible. I have
2702 created a function called modefopen, which takes an additional mode
2703 argument. It sets umask(777), creates the file, chmods it to the required
2704 mode, then resets the umask. All the relevant calls to fopen() in the
2705 content scanning code have been changed to use this function.
2707 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
2708 to 24 hours. This avoids potential overflow problems when processing G
2709 and H retry rules. I suspect nobody ever tinkers with this value.
2711 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
2713 PH/45 When the plaintext authenticator is running as a client, the server's
2714 challenges are checked to ensure they are valid base64 strings. By
2715 default, the authentication attempt is cancelled if an invalid string is
2716 received. Setting client_ignore_invalid_base64 true ignores these errors.
2717 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
2718 they are received. Thus, the responses can be made to depend on the
2719 challenges. If an invalid string is ignored, an empty string is placed in
2722 PH/46 Messages that are created by the autoreply transport now contains a
2723 References: header, in accordance with RFCs 2822 and 3834.
2725 PH/47 Added authenticated_sender_force to the smtp transport.
2727 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
2729 PH/49 Installed latest patch from the Sieve maintainer.
2731 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
2732 was also set, the appendfile transport was unnecessarily scanning a
2733 directory of message files (e.g. for maildir delivery) to find the count
2734 of files (along with the size), even though it did not need this
2735 information. It now does the scan only if it needs to find either the
2736 size of the count of files.
2738 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
2740 PH/52 Two bugs concerned with error handling when the smtp transport is
2743 (i) Exim was not creating retry information for temporary errors given
2744 for individual recipients after the DATA command when the smtp transport
2745 was used in LMTP mode. This meant that they could be retried too
2746 frequently, and not timed out correctly.
2748 (ii) Exim was setting the flag that allows error details to be returned
2749 for LMTP errors on RCPT commands, but not for LMTP errors for individual
2750 recipients that were returned after the DATA command.
2752 PH/53 This is related to PH/52, but is more general: for any failing address,
2753 when detailed error information was permitted to be returned to the
2754 sender, but the error was temporary, then after the final timeout, only
2755 "retry timeout exceeded" was returned. Now it returns the full error as
2756 well as "retry timeout exceeded".
2758 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
2759 do this, and (what is worse) MTAs that accept it.
2761 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
2762 will now be deprecated.
2764 PH/56 New os.c-cygwin from the Cygwin maintainer.
2766 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
2767 formats (previously only available in exim formats via -bpr, -bpru,
2768 and -bpra. Now also available in native and exiqgrep formats)
2770 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
2771 with very large, slow to parse queues
2773 JJ/08 exipick: added ! as generic prefix to negate any criteria format
2775 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
2777 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
2778 responses to authentication challenges, though it was showing the
2779 challenges; (ii) I've removed the CR characters from the debug output for
2782 PH/58 Allow for the insertion of a newline as well as a space when a string
2783 is turned into more than one encoded-word during RFC 2047 encoding. The
2784 Sieve code now uses this.
2786 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
2787 data_4xx, lost_connection, tls_required.
2789 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
2790 message was being sent as an SMTP response.
2792 PH/61 Add -l and -k options to exicyclog.
2794 PH/62 When verifying, if an address was redirected to one new address, so that
2795 verification continued, and the new address failed or deferred after
2796 having set something in $address_data, the value of $address_data was not
2797 passed back to the ACL. This was different to the case when no
2798 redirection occurred. The value is now passed back in both cases.
2800 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
2801 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
2802 use login_cap.h, so on its own it isn't the distinguishing feature. The
2803 new name refers directly to the setclassresources() function.
2805 PH/65 Added configuration files for NetBSD3.
2807 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
2809 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
2810 is preferred over IPv4.
2812 PH/68 The bounce_return_message and bounce_return_body options were not being
2813 honoured for bounces generated during the reception of non-SMTP messages.
2814 In particular, this applied to messages rejected by the ACL. This bug has
2815 been fixed. However, if bounce_return_message is true and bounce_return_
2816 body is false, the headers that are returned for a non-SMTP message
2817 include only those that have been read before the error was detected.
2818 (In the case of an ACL rejection, they have all been read.)
2820 PH/69 The HTML version of the specification is now built in a directory called
2821 spec_html instead of spec.html, because the latter looks like a path with
2822 a MIME-type, and this confuses some software.
2824 PH/70 Catch two compiler warnings in sieve.c.
2826 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
2827 function verify_get_ident() calls ip_connect() to connect a socket, but
2828 if the "connect()" function timed out, ip_connect() used to close the
2829 socket. However, verify_get_ident() also closes the socket later, and in
2830 between Exim writes to the log, which may get opened at this point. When
2831 the socket was closed in ip_connect(), the log could get the same file
2832 descriptor number as the socket. This naturally causes chaos. The fix is
2833 not to close the socket in ip_connect(); the socket should be closed by
2834 the function that creates it. There was only one place in the code where
2835 this was missing, in the iplookup router, which I don't think anybody now
2836 uses, but I've fixed it anyway.
2838 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
2839 well as to direct DNS lookups. Otherwise the handling of names in host
2840 lists is inconsistent and therefore confusing.
2846 PH/01 Two changes to the default runtime configuration:
2848 (1) Move the checks for relay_from_hosts and authenticated clients from
2849 after to before the (commented out) DNS black list checks.
2851 (2) Add control=submission to the relay_from_hosts and authenticated
2852 clients checks, on the grounds that messages accepted by these
2853 statements are most likely to be submissions.
2855 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
2857 (1) Generate an error if the third argument for the ${prvs expansion is
2860 (2) Treat a missing third argument of ${prvscheck as if it were an empty
2863 (3) Reset the variables that are obtained from the first argument of
2864 ${prvscheck and used in the second argument before leaving the code,
2865 because their memory is reclaimed, so using them afterwards may do
2868 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
2869 one (it's much easier than Tom thought :-).
2871 (5) Because of (4), we can now allow for the use of $prvscheck_result
2872 inside the third argument.
2874 PH/03 For some reason, the default setting of PATH when running a command from
2875 a pipe transport was just "/usr/bin". I have changed it to
2878 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
2879 anything to be listed in the output from -bV.
2881 PH/05 When a filter generated an autoreply, the entire To: header line was
2882 quoted in the delivery log line, like this:
2884 => >A.N.Other <ano@some.domain> <original@ddress> ...
2886 This has been changed so that it extracts the operative address. There
2887 may be more than one such address. If so, they are comma-separated, like
2890 => >ano@some.domain,ona@other.domain <original@ddress> ...
2892 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
2893 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
2894 not being looked up in the rDNS to get a host name, Exim was showing the
2895 IP address twice in Received: lines, even though the IP addresses were
2896 identical. For example:
2898 Received: from [1.2.3.4] (helo=[1.2.3.4])
2900 However, if the real host name was known, it was omitting the HELO data
2901 if it matched the actual IP address. This has been tidied up so that it
2902 doesn't show the same IP address twice.
2904 PH/07 When both +timestamp and +memory debugging was on, the value given by
2905 $tod_xxx expansions could be wrong, because the tod_stamp() function was
2906 called by the debug printing, thereby overwriting the timestamp buffer.
2907 Debugging no longer uses the tod_stamp() function when +timestamp is set.
2909 PH/08 When the original message was included in an autoreply transport, it
2910 always said "this is a copy of the message, including all the headers",
2911 even if body_only or headers_only was set. It now gives an appropriate
2914 PH/09 Applied a patch from the Sieve maintainer which:
2916 o fixes some comments
2917 o adds the (disabled) notify extension core
2918 o adds some debug output for the result of if/elsif tests
2919 o points to the current vacation draft in the documentation
2920 and documents the missing references header update
2924 o fixes a bug in processing the envelope test (when testing
2925 multiple envelope elements, the last element determinted the
2928 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
2929 Electronic Mail") by including:
2931 Auto-submitted: auto-generated
2933 in the messages that it generates (bounce messages and others, such as
2934 warnings). In the case of bounce messages for non-SMTP mesages, there was
2935 also a typo: it was using "Auto_submitted" (underscore instead of
2936 hyphen). Since every message generated by Exim is necessarily in response
2937 to another message, thes have all been changed to:
2939 Auto-Submitted: auto-replied
2941 in accordance with these statements in the RFC:
2943 The auto-replied keyword:
2945 - SHOULD be used on messages sent in direct response to another
2946 message by an automatic process,
2948 - MUST NOT be used on manually-generated messages,
2950 - MAY be used on Delivery Status Notifications (DSNs) and Message
2951 Disposition Notifications (MDNs),
2953 - MUST NOT be used on messages generated by automatic or periodic
2954 processes, except for messages which are automatic responses to
2957 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
2958 to the default Received: header definition.
2960 PH/12 Added log selector acl_warn_skipped (default on).
2962 PH/13 After a successful wildlsearch lookup, discard the values of numeric
2963 variables because (a) they are in the wrong storage pool and (b) even if
2964 they were copied, it wouldn't work properly because of the caching.
2966 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
2967 checking when decoding. Apparently there are clients that generate
2968 overlong encoded strings. Why am I not surprised?
2970 PH/15 If the first argument of "${if match_address" was not empty, but did not
2971 contain an "@" character, Exim crashed. Now it writes a panic log message
2972 and treats the condition as false.
2974 PH/16 In autoreply, treat an empty string for "once" the same as unset.
2976 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
2977 extension "envelope-auth". The code is finished and in agreement with
2978 other implementations, but there is no documentation so far and in fact,
2979 nobody wrote the draft yet. This extension is currently #undef'ed, thus
2980 not changing the active code.
2982 Print executed "if" and "elsif" statements when debugging is used. This
2983 helps a great deal to understand what a filter does.
2985 Document more things not specified clearly in RFC3028. I had all this
2986 sorted out, when out of a sudden new issues came to my mind. Oops."
2988 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
2991 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
2992 canonical form (as documented). However, after a host name lookup from
2993 the IP address, check_host() was doing a simple string comparison with
2994 addresses acquired from the DNS when checking that the found name did
2995 have the original IP as one of its addresses. Since any found IPv6
2996 addresses are likely to be in abbreviated form, the comparison could
2997 fail. Luckily, there already exists a function for doing the comparison
2998 by converting both addresses to binary, so now that is used instead of
2999 the text comparison.
3001 PH/20 There was another similar case to PH/19, when a complete host name was
3002 given in a host list; looking up its IP address could give an abbreviated
3003 form, whereas the current host's name might or might not be abbreviated.
3004 The same fix has been applied.
3010 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
3011 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
3014 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
3015 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
3017 PH/03 Typo: missing ".o" in src/pcre/Makefile.
3019 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
3020 header line, restrict the check to what is listed in RFCs 2369 and 2929.
3021 Also, for "Auto-Submitted", treat anything other than "no" as
3022 non-personal, in accordance with RFC 3834. (Previously it treated
3023 anything starting "auto-" as non-personal.)
3025 TF/01 The control=submission/name=... option had a problem with syntax
3026 errors if the name included a slash character. The /name= option
3027 now slurps the rest of the string, so it can include any characters
3028 but it must come last in the list of options (after /sender_retain
3031 PH/05 Some modifications to the interface to the fake nameserver for the new
3039 TK/01 Added the "success_on_redirect" address verification option. See
3040 NewStuff for rationale and an example.
3042 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
3044 PH/02 Patch to exigrep to allow it to work on syslog lines.
3046 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
3047 fread() to skip over the body file's header line, because in Cygwin the
3048 header line is locked and is inaccessible.
3050 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
3051 co-exist for some time) to make it clear that it is the Exim ID that is
3052 referenced, not the Message-ID: header line.
3054 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
3055 string_format() function, because snprintf() does not exist on all
3058 PH/06 The use of forbid_filter_existstest now also locks out the use of the
3059 ${stat: expansion item.
3061 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
3062 protocol synchronization error", to keep the pedants happy.
3064 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
3065 well as for IRIX systems, when gcc is being used. See the host.c source
3068 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
3070 PH/10 Named domain lists were not working if used in a queue_smtp_domains
3073 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
3074 transport and to the smtp transport in LMTP mode.
3076 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
3078 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
3079 run a filter in a subprocess. This could lead to confusion in subsequent
3080 lookups in the parent process. There should also be a search_tidyup() at
3081 the end of the subprocess.
3083 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
3084 only if the host matched helo_try_verify_hosts, which caused the
3085 verification to occur when the EHLO/HELO command was issued. The ACL just
3086 tested the remembered result. Now, if a previous verification attempt has
3087 not happened, "verify = helo" does it there and then.
3089 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
3091 TK/03 Fix log output including CR from clamd.
3093 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
3094 exist provoked a memory error which could cause a segfault.
3096 PH/15 Installed PCRE 6.2
3098 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
3100 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
3101 of the problem. Specifically, suggested +O2 rather than +O1 for the
3104 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
3106 PH/20 If a delivery was routed to a non-standard port by means of an SRV
3107 record, the port was not correctly logged when the outgoing_port log
3108 selector was set (it logged the transort's default port).
3110 PH/21 Added support for host-specific ports to manualroute, queryprogram,
3111 fallback_hosts, and "hosts" in the smtp transport.
3113 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
3114 host errors such as "Connection refused".
3116 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
3117 authentication with radiusclient 0.4.9:
3119 - Error returned from rc_read_config was caught wrongly
3120 - Username/password not passed on to radius server due to wrong length.
3122 The presumption is that some radiusclient API changes for 4.51/PH/17
3123 were not taken care of correctly. The code is still untested by me (my
3124 Linux distribution still has 0.3.2 of radiusclient), but it was
3125 contributed by a Radius user.
3127 PH/24 When doing a callout, the value of $domain wasn't set correctly when
3128 expanding the "port" option of the smtp transport.
3130 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
3131 while reading a MIME header. Thanks to Tom Hughes for a patch.
3133 PH/24 Include config.h inside local_scan.h so that configuration settings are
3136 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
3137 that in an ACL for RCPT (for example), you can examine exactly what was
3140 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
3141 commands, but it was not correctly comparing the address with the actual
3142 client host address. Thus, it would show the EHLO address in Received:
3143 header lines when this was not necessary.
3145 PH/27 Added the % operator to ${eval:}.
3147 PH/28 Exim tries to create and chdir to its spool directory when it starts;
3148 it should be ignoring failures (because with -C, for example, it has lost
3149 privilege). It wasn't ignoring creation failures other than "already
3152 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
3155 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
3156 because an input line was too long, either on its own, or by virtue of
3157 too many continuations, the temporary file was not being removed, and the
3158 return code was incorrect.
3160 PH/31 Missing "BOOL" in function definition in filtertest.c.
3162 PH/32 Applied Sieve patches from the maintainer.
3164 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
3166 PH/33 Added "verify = not_blind".
3168 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
3169 Local/Makefile (with some defaults set). These are used in built scripts
3170 such as exicyclog, but they have never been used in the exim_install
3171 script (though there are many overriding facilities there). I have
3172 arranged that the exim_install script now takes note of these two
3175 PH/35 Installed configuration files for Dragonfly.
3177 PH/36 When a locally submitted message by a trusted user did not contain a
3178 From: header, and the sender address was obtained from -f or from an SMTP
3179 MAIL command, and the trusted user did not use -F to supply a sender
3180 name, $originator_name was incorrectly used when constructing a From:
3181 header. Furthermore, $originator_name was used for submission mode
3182 messages from external hosts without From: headers in a similar way,
3183 which is clearly wrong.
3185 PH/37 Added control=suppress_local_fixups.
3187 PH/38 When log_selector = +received_sender was set, and the addition of the
3188 sender made the log line's construction buffer exactly full, or one byte
3189 less than full, an overflow happened when the terminating "\n" was
3192 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
3193 when the result of a list match is failure because a DNS lookup failed.
3195 PH/40 RM_COMMAND is now used in the building process.
3197 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
3198 the "build-* directories that it finds.
3200 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
3201 address in a domain literal was a prefix of an interface address.
3203 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
3204 when verifying a sender address, unless rewrite_headers is false.
3206 PH/44 Wrote a long comment about why errors_to addresses are verified as
3207 recipients, not senders.
3209 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
3210 the ratelimit ACL was added.
3212 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
3214 PH/46 Added extra information about PostgreSQL errors to the error string.
3216 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
3217 avoiding the need to install special zones in a real server. This is
3218 backwards compatible; if it can't find the fake resolver, it drops back.
3219 Thus, both old and new test suites can be run.
3221 TF/02 Added util/ratelimit.pl
3223 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
3224 clock is set back in time.
3226 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
3227 Candler <B.Candler@pobox.com>.
3229 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
3230 OK for addresses that are the result of redirections.
3232 PH/48 A number of further additions for the benefit of the new test suite,
3233 including a fake gethostbyname() that interfaces to the fake DNS resolver
3236 TF/06 The fix for widen_domains has also been applied to qualify_single and
3237 search_parents which are the other dnslookup options that can cause
3240 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
3243 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
3245 TF/07 Exim produced the error message "an SRV record indicated no SMTP
3246 service" if it encountered an MX record with an empty target hostname.
3247 The message is now "an MX or SRV record indicated no SMTP service".
3249 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
3250 if the DNS of the sending site is misconfigured. This is quite a
3251 common situation. This change restores the behaviour of treating a
3252 helo verification defer as a failure.
3254 PH/51 If self=fail was set on a router, the bounce message did not include the
3255 actual error message.
3261 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
3263 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
3264 command itself ended in error, the underlying message about the transport
3265 filter timeout was being overwritten with the pipe command error. Now the
3266 underlying error message should be appended to the second error message.
3268 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
3270 PH/02 Reduce the amount of output that "make" produces by default. Full output
3271 can still be requested.
3273 PH/03 The warning log line about a condition test deferring for a "warn" verb
3274 was being output only once per connection, rather than after each
3275 occurrence (because it was using the same function as for successful
3276 "warn" verbs). This seems wrong, so I have changed it.
3278 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
3279 it should not have, which might have caused a crash in the right
3280 circumstances, but probably never did.
3282 PH/04 Installed a modified version of Tony Finch's patch to make submission
3283 mode fix the return path as well as the Sender: header line, and to
3284 add a /name= option so that you can make the user's friendly name appear
3287 TF/03 Added the control = fakedefer ACL modifier.
3289 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
3290 Mark Lowes for thorough testing.
3292 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
3294 TK/03 Merged latest SRS patch from Miles Wilton.
3296 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
3297 with the definition in sysexits.h (which is #included earlier).
3298 Fortunately, Exim does not actually use EX_OK. The code used to try to
3299 preserve the sysexits.h value, by assumimg that macro definitions were
3300 scanned for macro replacements. I have been disabused of this notion,
3301 so now the code just undefines EX_OK before #including unistd.h.
3303 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
3304 in the smtp transport. When a block could not be written in a single
3305 write() function, the timeout was being re-applied to each part-write.
3306 This seems wrong - if the receiver was accepting one byte at a time it
3307 would take for ever. The timeout is now adjusted when this happens. It
3308 doesn't have to be particularly precise.
3310 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
3311 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
3313 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
3314 without a domain if the check to <postmaster@domain> fails.
3316 SC/01 Eximstats: added -xls and the ability to specify output files
3317 (patch written by Frank Heydlauf).
3319 SC/02 Eximstats: use FileHandles for outputing results.
3321 SC/03 Eximstats: allow any combination of xls, txt, and html output.
3323 SC/04 Eximstats: fixed display of large numbers with -nvr option
3325 SC/05 Eximstats: fixed merging of reports with empty tables.
3327 SC/06 Eximstats: added the -include_original_destination flag
3329 SC/07 Eximstats: removed tabs and trailing whitespace.
3331 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
3333 TK/06 MBOX spool code: Add real "From " MBOX separator line
3334 so the .eml file is really in mbox format (even though
3335 most programs do not really care). Patch from Alex Miller.
3337 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
3338 The latter is generated from $received_to and is only set if the
3339 message has one envelope recipient. SA can use these headers,
3340 obviously out-of-the-box. Patch from Alex Miller.
3342 PH/08 The ${def test on a variable was returning false if the variable's
3343 value was "0", contrary to what the specification has always said!
3344 The result should be true unless the variable is empty.
3346 PH/09 The syntax error of a character other than { following "${if
3347 def:variable_name" (after optional whitespace) was not being diagnosed.
3348 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
3349 accidental colon was present, for example, could give incorrect results.
3351 PH/10 Tidied the code in a number of places where the st_size field of a stat()
3352 result is used (not including appendfile, where other changes are about
3355 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
3356 This involved changing a lot of size variables from int to off_t. It
3357 should work with maildirs and everything.
3359 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
3360 spamd dying while we are connected to it.
3362 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
3365 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
3366 function for BDB 4.3.
3368 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
3370 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
3371 the sources for PCRE have been re-arranged and more files are now
3374 PH/15 The code I had for printing potentially long long variables in PH/11
3375 above was not the best (it lost precision). The length of off_t variables
3376 is now inspected at build time, and an appropriate printing format (%ld
3377 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
3378 to be "long long int" or "long int". This is needed for the internal
3379 formatting function string_vformat().
3381 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
3382 the configuration file to be ":syslog", then the script "guesses" where
3383 the logs files are, rather than using the compiled in default. In our
3384 case the guess is not the same as the compiled default, so the script
3385 suddenly stopped working when I started to use syslog. The patch checks
3386 to see if log_file_path is "". If so, it attempts to read it from exim
3387 with no configuration file to get the compiled in version, before it
3388 falls back to the previous guessing code."
3390 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
3391 implementing BATV in an Exim configuration. See NewStuff for the gory
3394 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
3395 Makefile that are specific to HP-UX.
3397 PH/18 If the "use_postmaster" option was set for a recipient callout together
3398 with the "random" option, the postmaster address was used as the MAIL
3399 FROM address for the random test, but not for the subsequent recipient
3400 test. It is now used for both.
3402 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
3403 patch removes a few documentation additions to RFC 3028, because the
3404 latest draft now contains them. It adds the new en;ascii-case comparator
3405 and a new error check for 8bit text in MIME parts. Comparator and
3406 require names are now matched exactly. I enabled the subaddress
3407 extension, but it is not well tested yet (read: it works for me)."
3409 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
3410 rework some of the code of TK/09 above to avoid the hardwired use of
3411 "%lld" and "long long". Replaced the call to snprintf() with a call to
3414 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
3415 records point to non-existent hosts", "retry timeout exceeded", and
3416 "retry time not reached for any host after a long failure period".
3418 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
3419 experimental DomainKeys support:
3421 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
3422 (2) On an error such as an illegally used "control", the wrong name for
3423 the control was given.
3425 These problems did NOT occur unless DomainKeys support was compiled.
3427 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
3429 PH/24 Added ${if match_ip condition.
3431 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
3432 databases so that it will be absolutely obvious if a crash occurs in the
3433 DB library. This is a regular occurrence (often caused by mis-matched
3436 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
3437 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
3438 picked up on a user's system that detects such things. There doesn't seem
3439 to be a gcc warning option for this - only an attribute that has to be
3440 put on the function's prototype. It seems that in Fedora Core 4 they have
3441 set this on a number of new functions. No doubt there will be more in due
3444 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
3445 specify a transport. However, if an address that was verified by such a
3446 router was the subject of a callout, Exim crashed because it tried to
3447 read the rcpt_include_affixes from the non-existent transport. Now it
3448 just assumes that the setting of that option is false. This bug was
3449 introduced by 4.51/PH/31.
3451 PH/28 Changed -d+all to exclude +memory, because that information is very
3452 rarely of interest, but it makes the output a lot bigger. People tend to
3453 do -d+all out of habit.
3455 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
3456 code in os-type was giving problems when libc.so lives in lib64, like on
3459 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
3460 aren't the modern standard, and it seems that some systems' include files
3461 don't always have them. Exim was already checking for some of the newer
3462 ones like T_AAAA, and defining it itself. I've added checks for all the
3463 record types that Exim uses.
3465 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
3466 not automatically generating a new one, as it is supposed to. This
3467 prevented TLS from working. If the file did exist, but contained invalid
3468 data, a new version was generated, as expected. It was only the case of a
3469 non-existent file that was broken.
3471 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
3472 with a change in libdomainkeys > 0.64.
3474 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
3475 from DNS. If the selector record carries the flag, it now has
3476 precedence over the domain-wide flag.
3478 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
3480 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
3481 the use of a port name that isn't defined in /etc/services) occurred, the
3482 message was deferred as in a normal delivery, and thus remained on the
3483 spool, instead of being failed because of the mua_wrapper setting. This
3484 is now fixed, and I tidied up some of the mua_wrapper messages at the
3487 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
3488 the messages in a hash of arrays rather than using individual hashes.
3489 This is a bit cleaner and results in dramatic memory savings, albeit
3490 at a slight CPU cost.
3492 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
3493 as requested by Marc Sherman.
3495 SC/10 Eximstats: added histograms for user specified patterns as requested
3498 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
3500 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
3501 fopen() in the content-scanning modules that did not already have it.
3507 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
3508 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
3510 TK/02 Fix ACL "control" statement not being available in MIME ACL.
3512 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
3514 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
3515 to test Sieve filters that use "vacation".
3517 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
3518 that changes the way the GnuTLS parameters are stored in the cache file.
3519 The new format can be generated externally. For backward compatibility,
3520 if the data in the cache doesn't make sense, Exim assumes it has read an
3521 old-format file, and it generates new data and writes a new file. This
3522 means that you can't go back to an older release without removing the
3525 PH/03 A redirect router that has both "unseen" and "one_time" set does not
3526 work if there are any delivery delays because "one_time" forces the
3527 parent to be marked "delivered", so its unseen clone is never tried
3528 again. For this reason, Exim now forbids the simultaneous setting of
3531 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
3532 redirected to themselves ("homonym" addresses). Read the long ChangeLog
3533 entry if you want to know the details. The fix, however, neglected to
3534 consider the case when local delivery batching is involved. The test for
3535 "previously delivered" was not happening when checking to see if an
3536 address could be batched with a previous (undelivered) one; under
3537 certain circumstances this could lead to multiple deliveries to the same
3540 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
3541 in its include files, and this causes problems building Exim.
3543 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
3544 header_syntax) but Exim was just ignoring anything given after a slash.
3545 In particular, this caused confusion with an attempt to use "verify =
3546 reverse_host_lookup/defer_ok". An error is now given when options are
3547 supplied for verify items that do not have them. (Maybe reverse_host_
3548 lookup should have a defer_ok option, but that's a different point.)
3550 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
3551 defined by RFC 821) to 2048, because there were problems with some AUTH
3552 commands, and RFC 1869 says the size should be increased for extended
3553 SMTP commands that take arguments.
3555 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
3558 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
3559 "unknown" error; now it says that the functionality isn't in the binary.
3561 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
3562 an address' error message when a string expansion fails (syntax or
3563 whatever). Otherwise the password may appear in the log. Following change
3564 PH/42 below, there is no longer a chance of it appearing in a bounce
3567 PH/11 Installed exipick version 20050225.0 from John Jetmore.
3569 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
3570 of its addresses was ever tried. (Bugzilla bug #2.)
3572 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
3573 the result incorrectly in the debug output. (It correctly added a newline
3574 to what was transported.)
3576 TF/01 Added $received_time.
3578 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
3579 commented out examples of how to interface to a virus scanner and to
3580 SpamAssassin. Also added commented examples of av_scanner and
3581 spamd_address settings.
3583 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
3584 and controls are allowed in which ACLs. There were a couple of minor
3585 errors. Some of the entries in the conditions table (which is a table of
3586 where they are NOT allowed) were getting very unwieldy; rewrote them as a
3587 negation of where the condition IS allowed.
3589 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
3591 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
3592 header file does not have a version number, so I've had to invent a new
3593 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
3594 API. The code is untested by me (my Linux distribution still has 0.3.2 of
3595 radiusclient), but it was contributed by a Radius user.
3597 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
3598 files or directories, for OpenSSL.
3600 PH/19 When an Exim process that is running as root has to create an Exim log
3601 file, it does so in a subprocess that runs as exim:exim so as to get the
3602 ownership right at creation (otherwise, other Exim processes might see
3603 the file with the wrong ownership). There was no test for failure of this
3604 fork() call, which would lead to the process getting stuck as it waited
3605 for a non-existent subprocess. Forks do occasionally fail when resources
3606 run out. I reviewed all the other calls to fork(); they all seem to check
3609 PH/20 When checking for unexpected SMTP input at connect time (before writing
3610 the banner), Exim was not dealing correctly with a non-positive return
3611 from the read() function. If the client had disconnected by this time,
3612 the result was a log entry for a synchronization error with an empty
3613 string after "input=" when read() returned zero. If read() returned -1
3614 (an event I could not check), uninitialized data bytes were printed.
3615 There were reports of junk text (parts of files, etc) appearing after
3618 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
3620 PH/22 Added support for macro redefinition, and (re)definition in between
3621 driver and ACL definitions.
3623 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
3624 forgetting to use the resulting value; it was using the unexpanded value.
3626 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
3627 hadn't been configured. The fix is from Juergen Kreileder, who
3628 understands it better than I do:
3630 "Here's what I see happening with three configured cyrus_sasl
3631 authenticators configured (plain, login, cram-md5):
3633 On startup auth_cyrus_sasl_init() gets called for each of these.
3634 This means three calls to sasl_listmech() without a specified mech_list.
3635 => SASL tests which mechs of all available mechs actually work
3636 => three warnings about OTP not working
3637 => the returned list contains: plain, login, cram-md5, digest-md5, ...
3639 With the patch, sasl_listmech() also gets called three times. But now
3640 SASL's mech_list option is set to the server_mech specified in the the
3641 authenticator. Or in other words, the answer from sasl_listmech()
3642 gets limited to just the mech you're testing for (which is different
3644 => the return list contains just 'plain' or 'login', 'cram-md5' or
3645 nothing depending on the value of ob->server_mech.
3647 I've just tested the patch: Authentication still works fine,
3648 unavailable mechs specified in the exim configuration are still
3649 caught, and the auth.log warnings about OTP are gone."
3651 PH/25 When debugging is enabled, the contents of the command line are added
3652 to the debugging output, even when log_selector=+arguments is not
3655 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
3656 answer is "GNU", and only if the return is "GNU/something" is the answer
3659 PH/27 $acl_verify_message is now set immediately after the failure of a
3660 verification in an ACL, and so is available in subsequent modifiers. In
3661 particular, the message can be preserved by coding like this:
3663 warn !verify = sender
3664 set acl_m0 = $acl_verify_message
3666 Previously, $acl_verify_message was set only while expanding "message"
3667 and "log_message" when a very denied access.
3669 PH/28 Modified OS/os.c-Linux with
3671 -#ifndef OS_LOAD_AVERAGE
3672 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
3674 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
3675 nomenclature these days.)
3677 PH/29 Installed patch from the Sieve maintainer that adds the options
3678 sieve_useraddress and sieve_subaddress to the redirect router.
3680 PH/30 In these circumstances:
3681 . Two addresses routed to the same list of hosts;
3682 . First host does not offer TLS;
3683 . First host accepts first address;
3684 . First host gives temporary error to second address;
3685 . Second host offers TLS and a TLS session is established;
3686 . Second host accepts second address.
3687 Exim incorrectly logged both deliveries with the TLS parameters (cipher
3688 and peerdn, if requested) that were in fact used only for the second
3691 PH/31 When doing a callout as part of verifying an address, Exim was not paying
3692 attention to any local part prefix or suffix that was matched by the
3693 router that accepted the address. It now behaves in the same way as it
3694 does for delivery: the affixes are removed from the local part unless
3695 rcpt_include_affixes is set on the transport.
3697 PH/32 Add the sender address, as F=<...>, to the log line when logging a
3698 timeout during the DATA phase of an incoming message.
3700 PH/33 Sieve envelope tests were broken for match types other than :is. I have
3701 applied a patch sanctioned by the Sieve maintainer.
3703 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
3704 the uid or gid is negative. A case of a negative gid caused this to be
3705 noticed. The fix allows for either to be negative.
3707 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
3708 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
3711 PH/36 The change PH/12 above was broken. Fixed it.
3713 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
3714 the grounds that routing the same address twice would always produce the
3715 same answer. This might have been true once, but it is certainly no
3716 longer true now. Routing a child address may depend on the previous
3717 routing that produced that child. Some complicated redirection strategies
3718 went wrong when messages had multiple recipients, and made Exim's
3719 behaviour dependent on the order in which the addresses were given.
3721 I have moved the duplicate checking until after the routing is complete.
3722 Exim scans the addresses that are assigned to local and remote
3723 transports, and removes any duplicates. This means that more work will be
3724 done, as duplicates will always all be routed, but duplicates are
3725 presumably rare, so I don't expect this is of any significance.
3727 For deliveries to pipes, files, and autoreplies, the duplicate checking
3728 still happens during the routing process, since they are not going to be
3731 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
3732 It corrects a timeout issue with spamd. This is Ian's comment: "The
3733 background is that sometimes spamd either never reads data from a
3734 connection it has accepted, or it never writes response data. The exiscan
3735 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
3736 blindly assumes that writes won't block so it may never time out."
3738 PH/39 Allow G after quota size as well as K and M.
3740 PH/40 The value set for $authenticated_id in an authenticator may not contain
3741 binary zeroes or newlines because the value is written to log lines and
3742 to spool files. There was no check on this. Now the value is run through
3743 the string_printing() function so that such characters are converted to
3744 printable escape sequences.
3746 PH/41 $message_linecount is a new variable that contains the total number of
3747 lines in the message. Compare $body_linecount, which is the count for the
3750 PH/42 Exim no longer gives details of delivery errors for specific addresses in
3751 bounce and delay warning messages, except in certain special cases, which
3754 (a) An SMTP error message from a remote host;
3755 (b) A message specified in a :fail: redirection;
3756 (c) A message specified in a "fail" command in a system filter;
3757 (d) A message specified in a FAIL return from the queryprogram router;
3758 (e) A message specified by the cannot_route_message router option.
3760 In these cases only, Exim does include the error details in bounce and
3761 warning messages. There are also a few cases where bland messages such
3762 as "unrouteable address" or "local delivery error" are given.
3764 PH/43 $value is now also set for the "else" part of a ${run expansion.
3766 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
3767 being worked on, but at least Exim now implements the latest version to
3770 PH/45 In a pipe transport, although a timeout while waiting for the pipe
3771 process to complete was treated as a delivery failure, a timeout while
3772 writing the message to the pipe was logged, but erroneously treated as a
3773 successful delivery. Such timeouts include transport filter timeouts. For
3774 consistency with the overall process timeout, these timeouts are now
3775 treated as errors, giving rise to delivery failures by default. However,
3776 there is now a new Boolean option for the pipe transport called
3777 timeout_defer, which, if set TRUE, converts the failures into defers for
3778 both kinds of timeout. A transport filter timeout is now identified in
3781 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
3782 systems where "make" and "gmake" are different, calling "gmake" at top
3783 level broke things. I've arranged for the value of $(MAKE) to be passed
3784 from the Makefile to this script so that it can call the same version of
3788 A note about Exim versions 4.44 and 4.50
3789 ----------------------------------------
3791 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
3792 changes of various kinds. As a consequence, a big documentation update was
3793 needed. This delayed the release for rather longer than seemed good, especially
3794 in the light of a couple of (minor) security issues. Therefore, the changes
3795 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
3796 release. So 4.44 and 4.50 are in effect two different branches that both start
3799 I have left the 4.50 change log unchanged; it contains all the changes since
3800 4.43. The change log for 4.44 is below; many of its items are identical to
3801 those for 4.50. This seems to be the most sensible way to preserve the
3802 historical information.
3808 1. Minor wording change to the doc/README.SIEVE file.
3810 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
3811 computation of the current number of files was incorrect.
3813 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
3814 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
3817 4. Give more explanation in the error message when the command for a transport
3818 filter fails to execute.
3820 5. There are several places where Exim runs a non-Exim command in a
3821 subprocess. The SIGUSR1 signal should be disabled for these processes. This
3822 was being done only for the command run by the queryprogram router. It is
3823 now done for all such subprocesses. The other cases are: ${run, transport
3824 filters, and the commands run by the lmtp and pipe transports.
3826 6. Added CONFIGURE_GROUP build-time option.
3828 7. Some older OS have a limit of 256 on the maximum number of file
3829 descriptors. Exim was using setrlimit() to set 1000 as a large value
3830 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
3831 systems. I've change it so that if it can't get 1000, it tries for 256.
3833 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
3834 was an oversight, and furthermore, ever since the addition of extra
3835 controls (e.g. 4.43/32), the checks on when to allow different forms of
3836 "control" were broken. There should now be diagnostics for all cases when a
3837 control that does not make sense is encountered.
3839 9. Added the /retain_sender option to "control=submission".
3841 10. $recipients is now available in the predata ACL (oversight).
3843 11. Tidy the search cache before the fork to do a delivery from a message
3844 received from the command line. Otherwise the child will trigger a lookup
3845 failure and thereby defer the delivery if it tries to use (for example) a
3846 cached ldap connection that the parent has called unbind on.
3848 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
3849 of $address_data from the recipient verification was clobbered by the
3850 sender verification.
3852 13. The value of address_data from a sender verification is now available in
3853 $sender_address_data in subsequent conditions in the ACL statement.
3855 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
3857 15. Added a new option "connect=<time>" to callout options, to set a different
3860 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
3861 was its contents. (It was OK if the option was not defined at all.)
3863 17. A "Completed" log line is now written for messages that are removed from
3864 the spool by the -Mrm option.
3866 18. New variables $sender_verify_failure and $recipient_verify_failure contain
3867 information about exactly what failed.
3869 19. Added -dd to debug only the daemon process.
3871 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
3872 handles timeouts, both on the server side and network timeouts. Renamed the
3873 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
3875 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
3876 It is now set to "smtps".
3878 22. $host_address is now set to the target address during the checking of
3879 ignore_target_hosts.
3881 23. When checking ignore_target_hosts for an ipliteral router, no host name was
3882 being passed; this would have caused $sender_host_name to have been used if
3883 matching the list had actually called for a host name (not very likely,
3884 since this list is usually IP addresses). A host name is now passed as
3887 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
3888 code that specifies a non-restarting handler (typically sigaction() in
3889 modern systems) in an attempt to fix a rare and obscure crash bug.
3891 25. Narrowed the window for a race in the daemon that could cause it to ignore
3892 SIGCHLD signals. This is not a major problem, because they are used only to
3893 wake it up if nothing else does.
3895 26. A malformed maildirsize file could cause Exim to calculate negative values
3896 for the mailbox size or file count. Odd effects could occur as a result.
3897 The maildirsize information is now recalculated if the size or filecount
3900 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
3901 support for a long time. Removed HAVE_SYS_VFS_H.
3903 28. Installed the latest version of exipick from John Jetmore.
3905 29. In an address list, if the pattern was not a regular expression, an empty
3906 subject address (from a bounce message) matched only if the pattern was an
3907 empty string. Non-empty patterns were not even tested. This was the wrong
3908 because it is perfectly reasonable to use an empty address as part of a
3909 database query. An empty address is now tested by patterns that are
3910 lookups. However, all the other forms of pattern expect the subject to
3911 contain a local part and a domain, and therefore, for them, an empty
3912 address still always fails if the pattern is not itself empty.
3914 30. Exim went into a mad DNS loop when attempting to do a callout where the
3915 host was specified on an smtp transport, and looking it up yielded more
3916 than one IP address.
3918 31. Re-factored the code for checking spool and log partition space into a
3919 function that finds that data and another that does the check. The former
3920 is then used to implement four new variables: $spool_space, $log_space,
3921 $spool_inodes, and $log_inodes.
3923 32. The RFC2047 encoding function was originally intended for short strings
3924 such as real names; it was not keeping to the 75-character limit for
3925 encoded words that the RFC imposes. It now respects the limit, and
3926 generates multiple encoded words if necessary. To be on the safe side, I
3927 have increased the buffer size for the ${rfc2047: expansion operator from
3930 33. It is now permitted to omit both strings after an "if" condition; if the
3931 condition is true, the result is "true". As before, when the second string
3932 is omitted, a false condition yields an empty string. This makes it less
3933 cumbersome to write custom ACL and router conditions.
3935 34. Failure to deliver a bounce message always caused it to be frozen, even if
3936 there was an errors_to setting on the router. The errors_to setting is now
3939 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
3940 canonical form (fully expanded) before being placed in
3941 $sender_host_address.
3943 36. The table in the code that translates DNS record types into text (T_A to
3944 "A" for instance) was missing entries for NS and CNAME. It is just possible
3945 that this could have caused confusion if both these types were looked up
3946 for the same domain, because the text type is used as part of Exim's
3947 per-process caching. But the chance of anyone hitting this buglet seems
3950 37. The dnsdb lookup has been extended in a number of ways.
3952 (1) There is a new type, "zns", which walks up the domain tree until it
3953 finds some nameserver records. It should be used with care.
3955 (2) There is a new type, "mxh", which is like "mx" except that it returns
3956 just the host names, not the priorities.
3958 (3) It is now possible to give a list of domains (or IP addresses) to be
3959 looked up. The behaviour when one of the lookups defers can be
3960 controlled by a keyword.
3962 (4) It is now possible to specify the separator character for use when
3963 multiple records are returned.
3965 38. The dnslists ACL condition has been extended: it is now possible to supply
3966 a list of IP addresses and/or domains to be looked up in a particular DNS
3969 39. Added log_selector=+queue_time_overall.
3971 40. When running the queue in the test harness, wait just a tad after forking a
3972 delivery process, to get repeatability of debugging output.
3974 41. Include certificate and key file names in error message when GnuTLS fails
3975 to set them up, because the GnuTLS error message doesn't include the name
3976 of the failing file when there is a problem reading it.
3978 42. Allow both -bf and -bF in the same test run.
3980 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
3982 44. The "Exiscan patch" is now merged into the mainline Exim source.
3984 45. Sometimes the final signoff response after QUIT could fail to get
3985 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
3986 < 0 before doing a fflush(). This bug looks as though it goes back to the
3987 introduction of TLS in release 3.20, but "sometimes" must have been rare
3988 because the tests only now provoked it.
3990 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
3991 (this can affect the format of dates).
3993 47. exim_tidydb, when checking for the continued existence of a message for
3994 which it has found a message-specific retry record, was not finding
3995 messages that were in split spool directories. Consequently, it was
3996 deleting retry records that should have stayed in existence.
3998 48. Steve fixed some bugs in eximstats.
4000 49. The SPA authentication driver was not abandoning authentication and moving
4001 on to the next authenticator when an expansion was forced to fail,
4002 contradicting the general specification for all authenticators. Instead it
4003 was generating a temporary error. It now behaves as specified.
4005 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
4006 (the order specifies the preference for clients). The order is now AES256,
4007 AES128, 3DES, ARCFOUR128.
4009 51. Small patch to Sieve code - explicitly set From: when generating an
4012 52. Exim crashed if a remote delivery caused a very long error message to be
4013 recorded - for instance if somebody sent an entire SpamAssassin report back
4014 as a large number of 550 error lines. This bug was coincidentally fixed by
4015 increasing the size of one of Exim's internal buffers (big_buffer) that
4016 happened as part of the Exiscan merge. However, to be on the safe side, I
4017 have made the code more robust (and fixed the comments that describe what
4020 53. Now that there can be additional text after "Completed" in log lines (if
4021 the queue_time_overall log selector is set), a one-byte patch to exigrep
4022 was needed to allow it to recognize "Completed" as not the last thing in
4025 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
4026 patch that reportedly fixes this has been added. I am not expert enough to
4027 create a test for it. This is what the patch creator wrote:
4029 "I found a little strange behaviour of ldap code when working with
4030 Windows 2003 AD Domain, where users was placed in more than one
4031 Organization Units. When I tried to give exim partial DN, the exit code
4032 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
4033 But simultaneously result of request was absolutely normal ldap result,
4034 so I produce this patch..."
4036 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
4037 REFERENCE, so I have modified the code to exclude the patch when that macro
4040 55. Some experimental protocols are using DNS PTR records for new purposes. The
4041 keys for these records are domain names, not reversed IP addresses. The
4042 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4043 leaves it alone. Component reversal etc. now happens only for IP addresses.
4046 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
4048 57. Double the size of the debug message buffer (to 2048) so that more of very
4049 long debug lines gets shown.
4051 58. The exicyclog utility now does better if the number of log files to keep
4052 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
4054 59. Two changes related to the smtp_active_hostname option:
4056 (1) $smtp_active_hostname is now available as a variable.
4057 (2) The default for smtp_banner uses $smtp_active_hostname instead
4058 of $primary_hostname.
4060 60. The host_aton() function is supposed to be passed a string that is known
4061 to be a valid IP address. However, in the case of IPv6 addresses, it was
4062 not checking this. This is a hostage to fortune. Exim now panics and dies
4063 if the condition is not met. A case was found where this could be provoked
4064 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4065 components; fortuitously, this particular loophole had already been fixed
4066 by change 4.50/55 above.
4068 If there are any other similar loopholes, the new check in host_aton()
4069 itself should stop them being exploited. The report I received stated that
4070 data on the command line could provoke the exploit when Exim was running as
4071 exim, but did not say which command line option was involved. All I could
4072 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4073 running as the user.
4076 61. There was a buffer overflow vulnerability in the SPA authentication code
4077 (which came originally from the Samba project). I have added a test to the
4078 spa_base64_to_bits() function which I hope fixes it.
4081 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
4082 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
4084 63. The daemon start-up calls getloadavg() while still root for those OS that
4085 need the first call to be done as root, but it missed one case: when
4086 deliver_queue_load_max is set with deliver_drop_privilege. This is
4087 necessary for the benefit of the queue runner, because there is no re-exec
4088 when deliver_drop_privilege is set.
4090 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
4091 This has been fixed.
4093 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4094 was in use, was not putting the data itself into the right store pool;
4095 consequently, it could be overwritten for a subsequent message in the same
4096 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4099 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
4101 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
4102 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
4103 the function were treating the return as a boolean value, which happened to
4104 work because 0=false and not-0=true, but is not correct code.
4106 68. The host_aton() function was not handling scoped IPv6 addresses (those
4107 with, for example, "%eth0" on the end) correctly.
4109 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
4110 negated items (that is, ~something) in unsigned ints. Some compilers
4111 apparently mutter when there is no cast.
4113 70. If an address verification called from an ACL failed, and did not produce a
4114 user-specific message (i.e. there was only a "system" message), nothing was
4115 put in $acl_verify_message. In this situation, it now puts the system
4118 71. Change 4.23/11 added synchronization checking at the start of an SMTP
4119 session; change 4.31/43 added the unwanted input to the log line - except
4120 that it did not do this in the start of session case. It now does.
4122 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
4123 This is wrong and can cause the other end to generate a synchronization
4124 error if it is another Exim or anything else that does the synchronization
4125 check. A QUIT command is no longer sent after a timeout.
4127 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
4128 during host lookups.
4130 74. The defer_ok option of callout verification was not working if it was used
4131 when verifying addresses in header lines, that is, for this case:
4133 verify = header_sender/callout=defer_ok
4135 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
4136 those file descriptors could be used for SMTP connections. If anything
4137 wrote to stderr (the example that came up was "warn" in embedded Perl), it
4138 could be sent to the SMTP client, causing chaos. The daemon now opens
4139 stdin, stdout, and stderr to /dev/null when it puts itself into the
4142 76. Arrange for output from Perl's "warn" command to be written to Exim's main
4143 log by default. The user can override this with suitable Perl magic.
4145 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
4146 the log message when discard triggers, was not working for the DATA ACL or
4147 for the non-SMTP ACL.
4149 78. Error message wording change in sieve.c.
4151 79. If smtp_accept_max_per_host was set, the number of connections could be
4152 restricted to fewer than expected, because the daemon was trying to set up
4153 a new connection before checking whether the processes handling previous
4154 connections had finished. The check for completed processes is now done
4155 earlier. On busy systems, this bug wouldn't be noticed because something
4156 else would have woken the daemon, and it would have reaped the completed
4159 80. If a message was submitted locally by a user whose login name contained one
4160 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
4161 It caused a spool format error. I have fixed the spool reading code. A
4162 related problem was that the "from" clause in the Received: line became
4163 illegal because of the space(s). It is now covered by ${quote_local_part.
4165 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
4168 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
4170 83. Patch from Sieve maintainer to fix unterminated string problem in
4171 "vacation" handling.
4173 84. Some minor changes to the Linux configuration files to help with other
4174 OS variants using glibc.
4176 85. One more patch for Sieve to update vacation handling to latest spec.
4179 ----------------------------------------------------
4180 See the note above about the 4.44 and 4.50 releases.
4181 ----------------------------------------------------
4187 1. Change 4.43/35 introduced a bug that caused file counts to be
4188 incorrectly computed when quota_filecount was set in an appendfile
4191 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4192 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4195 3. Give more explanation in the error message when the command for a transport
4196 filter fails to execute.
4198 4. There are several places where Exim runs a non-Exim command in a
4199 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4200 was being done only for the command run by the queryprogram router. It is
4201 now done for all such subprocesses. The other cases are: ${run, transport
4202 filters, and the commands run by the lmtp and pipe transports.
4204 5. Some older OS have a limit of 256 on the maximum number of file
4205 descriptors. Exim was using setrlimit() to set 1000 as a large value
4206 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4207 systems. I've change it so that if it can't get 1000, it tries for 256.
4209 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4210 was an oversight, and furthermore, ever since the addition of extra
4211 controls (e.g. 4.43/32), the checks on when to allow different forms of
4212 "control" were broken. There should now be diagnostics for all cases when a
4213 control that does not make sense is encountered.
4215 7. $recipients is now available in the predata ACL (oversight).
4217 8. Tidy the search cache before the fork to do a delivery from a message
4218 received from the command line. Otherwise the child will trigger a lookup
4219 failure and thereby defer the delivery if it tries to use (for example) a
4220 cached ldap connection that the parent has called unbind on.
4222 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4223 of $address_data from the recipient verification was clobbered by the
4224 sender verification.
4226 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4227 was its contents. (It was OK if the option was not defined at all.)
4229 11. A "Completed" log line is now written for messages that are removed from
4230 the spool by the -Mrm option.
4232 12. $host_address is now set to the target address during the checking of
4233 ignore_target_hosts.
4235 13. When checking ignore_target_hosts for an ipliteral router, no host name was
4236 being passed; this would have caused $sender_host_name to have been used if
4237 matching the list had actually called for a host name (not very likely,
4238 since this list is usually IP addresses). A host name is now passed as
4241 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4242 code that specifies a non-restarting handler (typically sigaction() in
4243 modern systems) in an attempt to fix a rare and obscure crash bug.
4245 15. Narrowed the window for a race in the daemon that could cause it to ignore
4246 SIGCHLD signals. This is not a major problem, because they are used only to
4247 wake it up if nothing else does.
4249 16. A malformed maildirsize file could cause Exim to calculate negative values
4250 for the mailbox size or file count. Odd effects could occur as a result.
4251 The maildirsize information is now recalculated if the size or filecount
4254 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4255 support for a long time. Removed HAVE_SYS_VFS_H.
4257 18. Updated exipick to current release from John Jetmore.
4259 19. Allow an empty sender to be matched against a lookup in an address list.
4260 Previously the only cases considered were a regular expression, or an
4263 20. Exim went into a mad DNS lookup loop when doing a callout where the
4264 host was specified on the transport, if the DNS lookup yielded more than
4267 21. The RFC2047 encoding function was originally intended for short strings
4268 such as real names; it was not keeping to the 75-character limit for
4269 encoded words that the RFC imposes. It now respects the limit, and
4270 generates multiple encoded words if necessary. To be on the safe side, I
4271 have increased the buffer size for the ${rfc2047: expansion operator from
4274 22. Failure to deliver a bounce message always caused it to be frozen, even if
4275 there was an errors_to setting on the router. The errors_to setting is now
4278 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4279 canonical form (fully expanded) before being placed in
4280 $sender_host_address.
4282 24. Updated eximstats to version 1.33
4284 25. Include certificate and key file names in error message when GnuTLS fails
4285 to set them up, because the GnuTLS error message doesn't include the name
4286 of the failing file when there is a problem reading it.
4288 26. Expand error message when OpenSSL has problems setting up cert/key files.
4291 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
4292 (this can affect the format of dates).
4294 28. exim_tidydb, when checking for the continued existence of a message for
4295 which it has found a message-specific retry record, was not finding
4296 messages that were in split spool directories. Consequently, it was
4297 deleting retry records that should have stayed in existence.
4299 29. eximstats updated to version 1.35
4300 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
4301 1.35 - bugfix such that pie charts by volume are generated correctly
4303 30. The SPA authentication driver was not abandoning authentication and moving
4304 on to the next authenticator when an expansion was forced to fail,
4305 contradicting the general specification for all authenticators. Instead it
4306 was generating a temporary error. It now behaves as specified.
4308 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
4309 (the order specifies the preference for clients). The order is now AES256,
4310 AES128, 3DES, ARCFOUR128.
4312 31. Small patch to Sieve code - explicitly set From: when generating an
4315 32. Exim crashed if a remote delivery caused a very long error message to be
4316 recorded - for instance if somebody sent an entire SpamAssassin report back
4317 as a large number of 550 error lines. This bug was coincidentally fixed by
4318 increasing the size of one of Exim's internal buffers (big_buffer) that
4319 happened as part of the Exiscan merge. However, to be on the safe side, I
4320 have made the code more robust (and fixed the comments that describe what
4323 33. Some experimental protocols are using DNS PTR records for new purposes. The
4324 keys for these records are domain names, not reversed IP addresses. The
4325 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4326 leaves it alone. Component reversal etc. now happens only for IP addresses.
4329 34. The host_aton() function is supposed to be passed a string that is known
4330 to be a valid IP address. However, in the case of IPv6 addresses, it was
4331 not checking this. This is a hostage to fortune. Exim now panics and dies
4332 if the condition is not met. A case was found where this could be provoked
4333 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4334 components; fortuitously, this particular loophole had already been fixed
4335 by change 4.50/55 or 4.44/33 above.
4337 If there are any other similar loopholes, the new check in host_aton()
4338 itself should stop them being exploited. The report I received stated that
4339 data on the command line could provoke the exploit when Exim was running as
4340 exim, but did not say which command line option was involved. All I could
4341 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4342 running as the user.
4345 35. There was a buffer overflow vulnerability in the SPA authentication code
4346 (which came originally from the Samba project). I have added a test to the
4347 spa_base64_to_bits() function which I hope fixes it.
4350 36. The daemon start-up calls getloadavg() while still root for those OS that
4351 need the first call to be done as root, but it missed one case: when
4352 deliver_queue_load_max is set with deliver_drop_privilege. This is
4353 necessary for the benefit of the queue runner, because there is no re-exec
4354 when deliver_drop_privilege is set.
4356 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4357 was in use, was not putting the data itself into the right store pool;
4358 consequently, it could be overwritten for a subsequent message in the same
4359 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4362 38. Sometimes the final signoff response after QUIT could fail to get
4363 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4364 < 0 before doing a fflush(). This bug looks as though it goes back to the
4365 introduction of TLS in release 3.20, but "sometimes" must have been rare
4366 because the tests only now provoked it.
4372 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
4373 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
4374 Now it returns an integer. A number of calls were still expecting a T/F
4375 return. Fortuitously, in all cases, the tests worked in OK situations,
4376 which is the norm. However, things would have gone wrong on any write
4377 failures on the smtp file descriptor. This function is used when sending
4378 messages over SMTP and also when doing verify callouts.
4380 2. When Exim is called to do synchronous delivery of a locally submitted
4381 message (the -odf or -odi options), it no longer closes stderr before doing
4384 3. Implemented the mua_wrapper option.
4386 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
4388 5. Implemented the functions header_remove(), header_testname(),
4389 header_add_at_position(), and receive_remove_recipient(), and exported them
4392 6. If an ACL "warn" statement specified the addition of headers, Exim already
4393 inserted X-ACL-Warn: at the start if there was no header name. However, it
4394 was not making this test for the second and subsequent header lines if
4395 there were newlines in the string. This meant that an invalid header could
4396 be inserted if Exim was badly configured.
4398 7. Allow an ACL "warn" statement to add header lines at the start or after all
4399 the Received: headers, as well as at the end.
4401 8. Added the rcpt_4xx retry error code.
4403 9. Added postmaster_mailfrom=xxx to callout verification option.
4405 10. Added mailfrom=xxxx to the callout verification option, for verify=
4408 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
4409 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
4411 12. Inserted some casts to stop certain compilers warning when using pointer
4412 differences as field lengths or precisions in printf-type calls (mostly
4413 affecting debugging statements).
4415 13. Added optional readline() support for -be (dynamically loaded).
4417 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
4418 same clock tick as a message's arrival, so that its received time was the
4419 same as the "first fail" time on the retry record, and that message
4420 remained on the queue past the ultimate address timeout, every queue runner
4421 would try a delivery (because it was past the ultimate address timeout) but
4422 after another failure, the ultimate address timeout, which should have then
4423 bounced the address, did not kick in. This was a "< instead of <=" error;
4424 in most cases the first failure would have been in the next clock tick
4425 after the received time, and all would be well.
4427 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
4428 being recognized when the domain list was tested by the match_domain
4429 condition in an expansion string.
4431 16. Added the ${str2b64: operator.
4433 17. Exim was always calling setrlimit() to set a large limit for the number of
4434 processes, without checking whether the existing limit was already
4435 adequate. (It did check for the limit on file descriptors.) Furthermore,
4436 errors from getrlimit() and setrlimit() were being ignored. Now they are
4437 logged to the main and panic logs, but Exim does carry on, to try to do its
4438 job under whatever limits there are.
4440 18. Imported PCRE 5.0.
4442 19. Trivial typo in log message " temporarily refused connection" (the leading
4445 20. If the log selector return_path_on_delivery was set and an address was
4446 redirected to /dev/null, the delivery process crashed because it assumed
4447 that a return path would always be set for a "successful" delivery. In this
4448 case, the whole delivery is bypassed as an optimization, and therefore no
4451 21. Internal re-arrangement: the function for sending a challenge and reading
4452 a response while authentication was assuming a zero-terminated challenge
4453 string. It's now changed to take a pointer and a length, to allow for
4454 binary data in such strings.
4456 22. Added the cyrus_sasl authenticator (code supplied by MBM).
4458 23. Exim was not respecting finduser_retries when seeking the login of the
4459 uid under which it was called; it was always trying 10 times. (The default
4460 setting of finduser_retries is zero.) Also, it was sleeping after the final
4461 failure, which is pointless.
4463 24. Implemented tls_on_connect_ports.
4465 25. Implemented acl_smtp_predata.
4467 26. If the domain in control=submission is set empty, Exim assumes that the
4468 authenticated id is a complete email address when it generates From: or
4469 Sender: header lines.
4471 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
4472 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
4473 chown and chgrp in /bin and hostname in /usr/bin.
4475 28. Exim was keeping the "process log" file open after each use, just as it
4476 does for the main log. This opens the possibility of it remaining open for
4477 long periods when the USR1 signal hits a daemon. Occasional processlog
4478 errors were reported, that could have been caused by this. Anyway, it seems
4479 much more sensible not to leave this file open at all, so that is what now
4482 29. The long-running daemon process does not normally write to the log once it
4483 has entered its main loop, and it closes the log before doing so. This is
4484 so that log files can straightforwardly be renamed and moved. However,
4485 there are a couple of unusual error situations where the daemon does write
4486 log entries, and I had neglected to close the log afterwards.
4488 30. The text of an SMTP error response that was received during a remote
4489 delivery was being truncated at 512 bytes. This is too short for some of
4490 the long messages that one sometimes sees. I've increased the limit to
4493 31. It is now possible to make retry rules that apply only when a message has a
4494 specific sender, in particular, an empty sender.
4496 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
4497 it possible to be selective about when SMTP synchronization is enforced.
4499 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
4501 32. Implemented hosts_connection_nolog.
4503 33. Added an ACL for QUIT.
4505 34. Setting "delay_warning=" to disable warnings was not working; it gave a
4508 35. Added mailbox_size and mailbox_filecount to appendfile.
4510 36. Added control = no_multiline_responses to ACLs.
4512 37. There was a bug in the logic of the code that waits for the clock to tick
4513 in the case where the clock went backwards by a substantial amount such
4514 that the microsecond fraction of "now" was more than the microsecond
4515 fraction of "then" (but the whole seconds number was less).
4517 38. Added support for the libradius Radius client library this is found on
4518 FreeBSD (previously only the radiusclient library was supported).
4524 1. When certain lookups returned multiple values in the form name=value, the
4525 quoting of the values was not always being done properly. Specifically:
4526 (a) If the value started with a double quote, but contained no whitespace,
4528 (b) If the value contained whitespace other than a space character (i.e.
4529 tabs or newlines or carriage returns) it was not quoted.
4530 This fix has been applied to the mysql and pgsql lookups by writing a
4531 separate quoting function and calling it from the lookup code. The fix
4532 should probably also be applied to nisplus, ibase and oracle lookups, but
4533 since I cannot test any of those, I have not disturbed their existing code.
4535 2. A hit in the callout cache for a specific address caused a log line with no
4536 reason for rejecting RCPT. Now it says "Previous (cached) callout
4537 verification failure".
4539 3. There was an off-by-one bug in the queryprogram router. An over-long
4540 return line was truncated at 256 instead of 255 characters, thereby
4541 overflowing its buffer with the terminating zero. As well as fixing this, I
4542 have increased the buffer size to 1024 (and made a note to document this).
4544 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
4545 when Exim is waiting for an SMTP response from a remote server, Exim
4546 restarts its select() call on the socket, thereby resetting its timeout.
4547 This is not a problem when such interrupts are rare. Somebody set up a cron
4548 job to run exiwhat every 2 minutes, which is less than the normal select()
4549 timeout (5 or 10 minutes). This meant that the select() timeout never
4550 kicked in because it was always reset. I have fixed this by comparing the
4551 time when an interrupt arrives with the time at the start of the first call
4552 to select(). If more time than the timeout has elapsed, the interrupt is
4553 treated as a timeout.
4555 5. Some internal re-factoring in preparation for the addition of Sieve
4556 extensions (by MH). In particular, the "personal" test is moved to a
4557 separate function, and given an option for scanning Cc: and Bcc: (which is
4558 not set for Exim filters).
4560 6. When Exim created an email address using the login of the caller as the
4561 local part (e.g. when creating a From: or Sender: header line), it was not
4562 quoting the local part when it contained special characters such as @.
4564 7. Installed new OpenBSD configuration files.
4566 8. Reworded some messages for syntax errors in "and" and "or" conditions to
4567 try to make them clearer.
4569 9. Callout options, other than the timeout value, were being ignored when
4570 verifying sender addresses in header lines. For example, when using
4572 verify = header_sender/callout=no_cache
4574 the cache was (incorrectly) being used.
4576 10. Added a missing instance of ${EXE} to the exim_install script; this affects
4577 only the Cygwin environment.
4579 11. When return_path_on_delivery was set as a log selector, if different remote
4580 addresses in the same message used different return paths and parallel
4581 remote delivery occurred, the wrong values would sometimes be logged.
4582 (Whenever a remote delivery process finished, the return path value from
4583 the most recently started remote delivery process was logged.)
4585 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
4586 lines when AUTH and/or TLS are in use. This is the "received protocol"
4587 field. Exim used to use "asmtp" for authenticated SMTP, without any
4588 indication (in the protocol name) for TLS use. Now it follows the RFC and
4589 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
4590 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
4591 names appear in log lines as well as in Received: header lines.
4593 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
4594 extensions, and comparison tests, and to fix some bugs.
4596 14. Changes to the "personal" filter test:
4598 (1) The test was buggy in that it was just doing the equivalent of
4599 "contains" tests on header lines. For example, if a user's address was
4600 anne@some.where, the "personal" test would incorrectly be true for
4602 To: susanne@some.where
4604 This test is now done by extracting each address from the header in turn,
4605 and checking the entire address. Other tests that are part of "personal"
4606 are now done using regular expressions (for example, to check local parts
4607 of addresses in From: header lines).
4609 (2) The list of non-personal local parts in From: addresses has been
4610 extended to include "listserv", "majordomo", "*-request", and "owner-*",
4611 taken from the Sieve specification recommendations.
4613 (3) If the message contains any header line starting with "List-" it is
4614 treated as non-personal.
4616 (4) The test for "circular" in the Subject: header line has been removed
4617 because it now seems ill-conceived.
4619 15. Minor typos in src/EDITME comments corrected.
4621 16. Installed latest exipick from John Jetmore.
4623 17. If headers_add on a router specified a text string that was too long for
4624 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
4625 of string_sprintf() is now avoided.
4627 18. $message_body_size was not set (it was always zero) when running the DATA
4628 ACL and the local_scan() function.
4630 19. For the "mail" command in an Exim filter, no default was being set for
4631 the once_repeat time, causing a random time value to be used if "once" was
4632 specified. (If the value happened to be <= 0, no repeat happened.) The
4633 default is now 0s, meaning "never repeat". The "vacation" command was OK
4634 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
4635 (I found it when inspecting the code).
4637 20. There is now an overall timeout for performing a callout verification. It
4638 defaults to 4 times the callout timeout, which applies to individual SMTP
4639 commands during the callout. The overall timeout applies when there is more
4640 than one host that can be tried. The timeout is checked before trying the
4641 next host. This prevents very long delays if there are a large number of
4642 hosts and all are timing out (e.g. when the network connections are timing
4643 out). The value of the overall timeout can be changed by specifying an
4644 additional sub-option for "callout", called "maxwait". For example:
4646 verify = sender/callout=5s,maxwait=20s
4648 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
4649 to the end before writing, but this should make it even safer).
4651 22. Exim was forgetting that it had advertised PIPELINING for the second and
4652 subsequent messages on an SMTP connection. It was also not resetting its
4653 memory on STARTTLS and an internal HELO.
4655 23. When Exim logs an SMTP synchronization error within a session, it now
4656 records whether PIPELINING has been advertised or not.
4658 24. Added 3 instances of "(long int)" casts to time_t variables that were being
4659 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
4660 rather than long int.
4662 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
4664 26. Added the never_mail option to autoreply.
4670 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
4671 crash if the getsockname() call failed; this can happen if a connection is
4672 closed very soon after it is established. The problem was simply in the
4673 order in which certain operations were done, causing Exim to try to write
4674 to the SMTP stream before it had set up the file descriptor. The bug has
4675 been fixed by making things happen in the correct order.
4681 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
4682 before the connection was closed, thus losing the rejection response.
4684 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
4685 some early Solaris releases, but causes trouble in current releases where
4686 socklen_t is defined.
4688 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
4691 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
4694 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
4695 that releasing the top of it at the end releases what was used for sub-
4696 expansions (unless the block got too big). However, discard this block if
4697 the first thing is a variable or header, so that we can use its block when
4698 it is dynamic (useful for very large $message_headers, for example).
4700 6. Lookups now cache *every* query, not just the most recent. A new, separate
4701 store pool is used for this. It can be recovered when all lookup caches are
4702 flushed. Lookups now release memory at the end of their result strings.
4703 This has involved some general refactoring of the lookup sources.
4705 7. Some code has been added to the store_xxx() functions to reduce the amount
4706 of flapping under certain conditions.
4708 8. log_incoming_interface used to affect only the <= reception log lines. Now
4709 it causes the local interface and port to be added to several more SMTP log
4710 lines, for example "SMTP connection from", and rejection lines.
4712 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
4714 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
4716 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
4717 could be overwritten at the end of the current message (or the start of a
4718 new message if it was set in a HELO ACL). The value is now preserved for
4719 the duration of the SMTP connection.
4721 12. If a transport had a headers_rewrite setting, and a matching header line
4722 contained an unqualified address, that address was qualified, even if it
4723 did not match any rewriting rules. The underlying bug was that the values
4724 of the flags that permit the existence of unqualified sender and recipient
4725 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
4726 non-local messages, and by -bnq for local messages) were not being
4727 preserved with the message after it was received.
4729 13. When Exim was logging an SMTP synchronization error, it could sometimes log
4730 "next input=" as part of the text comprising the host identity instead of
4731 the correct text. The code was using the same buffer for two different
4732 strings. However, depending on which order the printing function evaluated
4733 its arguments, the bug did not always show up. Under Linux, for example, my
4734 test suite worked just fine.
4736 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
4737 doesn't work with some older versions of Perl. It has been changed to "my",
4738 which in any case is probably the better facility to use.
4740 15. A really picky compiler found some instances of statements for creating
4741 error messages that either had too many or two few arguments for the format
4744 16. The size of the buffer for calls to the DNS resolver has been increased
4745 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
4746 for addresses that have a lot of PTR records. This alleviates a problem; it
4747 does not fully solve it.
4749 17. A dnsdb lookup for PTR records that receives more data than will fit in the
4750 buffer now truncates the list and logs the incident, which is the same
4751 action as happens when Exim is looking up a host name and its aliases.
4752 Previously in this situation something unpredictable would happen;
4753 sometimes it was "internal error: store_reset failed".
4755 18. If a server dropped the connection unexpectedly when an Exim client was
4756 using GnuTLS and trying to read a response, the client delivery process
4757 crashed while trying to generate an error log message.
4759 19. If a "warn" verb in an ACL added multiple headers to a message in a single
4760 string, for example:
4762 warn message = H1: something\nH2: something
4764 the text was added as a single header line from Exim's point of view
4765 though it ended up OK in the delivered message. However, searching for the
4766 second and subsequent header lines using $h_h2: did not work. This has been
4767 fixed. Similarly, if a system filter added multiple headers in this way,
4768 the routers could not see them.
4770 20. Expanded the error message when iplsearch is called with an invalid key to
4771 suggest using net-iplsearch in a host list.
4773 21. When running tests using -bh, any delays imposed by "delay" modifiers in
4774 ACLs are no longer actually imposed (and a message to that effect is
4777 22. If a "gecos" field in a passwd entry contained escaped characters, in
4778 particular, if it contained a \" sequence, Exim got it wrong when building
4779 a From: or a Sender: header from that name. A second bug also caused
4780 incorrect handling when an unquoted " was present following a character
4781 that needed quoting.
4783 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
4784 was not being matched caselessly.
4786 24. Arranged for all hyphens in the exim.8 source to be escaped with
4789 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
4790 properly. Recipient callout cache records were still being keyed to include
4791 the sender, even when use_sender was set false. This led to far more
4792 callouts that were necessary. The sender is no longer included in the key
4793 when use_sender is false.
4795 26. Added "control = submission" modifier to ACLs.
4797 27. Added the ${base62d: operator to decode base 62 numbers.
4799 28. dnsdb lookups can now access SRV records.
4801 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
4802 the configuration file.
4804 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
4805 (-v) mode. This makes the output for a verbose queue run more intelligible.
4807 31. Added a use_postmaster feature to recipient callouts.
4809 32. Added the $body_zerocount variable, containing the number of binary zero
4810 bytes in the message body.
4812 33. The time of last modification of the "new" subdirectory is now used as the
4813 "mailbox time last read" when there is a quota error for a maildir
4816 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
4818 35. Added +ignore_unknown as a special item in host lists.
4820 36. Code for decoding IPv6 addresses in host lists is now included, even if
4821 IPv6 support is not being compiled. This fixes a bug in which an IPv6
4822 address was recognized as an IP address, but was then not correctly decoded
4823 into binary, causing unexpected and incorrect effects when compared with
4830 1. Very minor rewording of debugging text in manualroute to say "list of
4831 hosts" instead of "hostlist".
4833 2. If verify=header_syntax was set, and a header line with an unqualified
4834 address (no domain) and a large number of spaces between the end of the
4835 name and the colon was received, the reception process suffered a buffer
4836 overflow, and (when I tested it) crashed. This was caused by some obsolete
4837 code that should have been removed. The fix is to remove it!
4839 3. When running in the test harness, delay a bit after writing a bounce
4840 message to get a bit more predictability in the log output.
4842 4. Added a call to search_tidyup() just before forking a reception process. In
4843 theory, someone could use a lookup in the expansion of smtp_accept_max_
4844 per_host which, without the tidyup, could leave open a database connection.
4846 5. Added the variables $recipient_data and $sender_data which get set from a
4847 lookup success in an ACL "recipients" or "senders" condition, or a router
4848 "senders" option, similar to $domain_data and $local_part_data.
4850 6. Moved the writing of debug_print from before to after the "senders" test
4853 7. Change 4.31/66 (moving the time when the Received: is generated) caused
4854 problems for message scanning, either using a data ACL, or using
4855 local_scan() because the Received: header was not generated till after they
4856 were called (in order to set the time as the time of reception completion).
4857 I have revised the way this works. The header is now generated after the
4858 body is received, but before the ACL or local_scan() are called. After they
4859 are run, the timestamp in the header is updated.
4865 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
4866 before starting a queue runner without re-exec. This happened only when
4867 deliver_drop_privilege was set or when the Exim user was set to root. The
4868 effect of the bug was that timeouts during subsequent deliveries caused
4869 crashes instead of being properly handled. The handler is now left at its
4870 default (and expected) setting.
4872 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
4873 message, again when deliver_drop_privilege is set or Exim is run as root.
4874 The bug described in (1) was not present in this case, but the tidying up
4875 of the other signals was missing. I have made the two cases consistent.
4877 3. The ignore_target_hosts setting on a manualroute router was being ignored
4878 for hosts that were looked up using the /MX notation.
4880 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
4883 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
4884 operated on the sender address. After changing the $sender_address to <>
4885 for the sender address verify, Exim was re-instated it as the original
4886 (before rewriting) address, but remembering that it had rewritten it, so it
4887 wasn't rewriting it again. This bug also had the effect of breaking the
4888 sender address verification caching when the sender address was rewritten.
4890 6. The ignore_target_hosts option was being ignored by the ipliteral router.
4891 This has been changed so that if the ip literal address matches
4892 ignore_target_hosts, the router declines.
4894 7. Added expansion conditions match_domain, match_address, and match_local_
4895 part (NOT match_host).
4897 8. The placeholder for the Received: header didn't have a length field set.
4899 9. Added code to Exim itself and to exim_lock to test for a specific race
4900 condition that could lead to file corruption when using MBX delivery. The
4901 issue is with the lockfile that is created in /tmp. If this file is removed
4902 after a process has opened it but before that process has acquired a lock,
4903 there is the potential for a second process to recreate the file and also
4904 acquire a lock. This could lead to two Exim processes writing to the file
4905 at the same time. The added code performs the same test as UW imapd; it
4906 checks after acquiring the lock that its file descriptor still refers to
4907 the same named file.
4909 10. The buffer for building added header lines was of fixed size, 8192 bytes.
4910 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
4913 11. Added the smtp_active_hostname option. If used, this will typically be made
4914 to depend on the incoming interface address. Because $interface_address is
4915 not set up until the daemon has forked a reception process, error responses
4916 that can happen earlier (such as "too many connections") no longer contain
4919 12. If an expansion in a condition on a "warn" statement fails because a lookup
4920 defers, the "warn" statement is abandoned, and the next ACL statement is
4921 processed. Previously this caused the whole ACL to be aborted.
4923 13. Added the iplsearch lookup type.
4925 14. Added ident_timeout as a log selector.
4927 15. Added tls_certificate_verified as a log selector.
4929 16. Added a global option tls_require_ciphers (compare the smtp transport
4930 option of the same name). This controls incoming TLS connections.
4932 17. I finally figured out how to make tls_require_ciphers do a similar thing
4933 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
4934 before starting the TLS session.
4936 18. Tabs are now shown as \t in -bP output.
4938 19. If the log selector return_path_on_delivery was set, Exim crashed when
4939 bouncing a message because it had too many Received: header lines.
4941 20. If two routers both had headers_remove settings, and the first one included
4942 a superfluous trailing colon, the final name in the first list and the
4943 first name in the second list were incorrectly joined into one item (with a
4944 colon in the middle).
4950 1. Added -C and -D options to the exinext utility, mainly to make it easier
4951 to include in the automated testing, but these could be helpful when
4952 multiple configurations are in use.
4954 2. The exinext utility was not formatting the output nicely when there was
4955 an alternate port involved in the retry record key, nor when there was a
4956 message id as well (for retries that were specific to a specific message
4957 and a specific host). It was also confused by IPv6 addresses, because of
4958 the additional colons they contain. I have fixed the IPv4 problem, and
4959 patched it up to do a reasonable job for IPv6.
4961 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
4962 delivery, the log line now contains "pipelined" if PIPELINING was used.
4964 4. An SMTP transport process used to panic and die if the bind() call to set
4965 an explicit outgoing interface failed. This has been changed; it is now
4966 treated in the same way as a connect() failure.
4968 5. A reference to $sender_host_name in the part of a conditional expansion
4969 that was being skipped was still causing a DNS lookup. This no longer
4972 6. The def: expansion condition was not recognizing references to header lines
4973 that used bh_ and bheader_.
4975 7. Added the _cache feature to named lists.
4977 8. The code for checking quota_filecount in the appendfile transport was
4978 allowing one more file than it should have been.
4980 9. For compatibility with Sendmail, the command line option
4988 and sets the incoming protocol and host name (for trusted callers). The
4989 host name and its colon can be omitted when only the protocol is to be set.
4990 Note the Exim already has two private options, -pd and -ps, that refer to
4991 embedded Perl. It is therefore impossible to set a protocol value of "d" or
4992 "s", but I don't think that's a major issue.
4994 10. A number of refactoring changes to the code, none of which should affect
4997 (a) The number of logging options was getting close to filling up the
4998 32-bit word that was used as a bit map. I have split them into two classes:
4999 those that are passed in the argument to log_write(), and those that are
5000 only ever tested independently outside of that function. These are now in
5001 separate 32-bit words, so there is plenty of room for expansion again.
5002 There is no change in the user interface or the logging behaviour.
5004 (b) When building, for example, log lines, the code previously used a
5005 macro that called string_cat() twice, in order to add two strings. This is
5006 not really sufficiently general. Furthermore, there was one instance where
5007 it was actually wrong because one of the argument was used twice, and in
5008 one call a function was used. (As it happened, calling the function twice
5009 did not affect the overall behaviour.) The macro has been replaced by a
5010 function that can join an arbitrary number of extra strings onto a growing
5013 (c) The code for expansion conditions now uses a table and a binary chop
5014 instead of a serial search (which was left over from when there were very
5015 few conditions). Also, it now recognizes conditions like "pam" even when
5016 the relevant support is not compiled in: a suitably worded error message is
5017 given if an attempt is made to use such a condition.
5019 11. Added ${time_interval:xxxxx}.
5021 12. A bug was causing one of the ddress fields not to be passed back correctly
5022 from remote delivery subprocesses. The field in question was not being
5023 subsequently used, so this caused to problems in practice.
5025 13. Added new log selectors queue_time and deliver_time.
5027 14. Might have fixed a bug in maildirsizefile handling that threw up
5028 "unexpected character" debug warnings, and recalculated the data
5029 unnecessarily. In any case, I expanded the warning message to give more
5032 15. Added the message "Restricted characters in address" to the statements in
5033 the default ACL that block characters like @ and % in local parts.
5035 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
5036 Three changes have been made:
5038 (a) There was a serious bug; a negative response to MAIL caused the whole
5039 recipient domain to be cached as invalid, thereby blocking all messages
5040 to all local parts at the same domain, from all senders. This bug has
5041 been fixed. The domain is no longer cached after a negative response to
5042 MAIL if the sender used is not empty.
5044 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
5047 (c) A new callout option, "use_sender" has been added for people who want
5048 the modified behaviour.
5054 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
5057 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
5058 indeed breaks things for older releases.
5060 3. Added additional logging to the case where there is a problem reading data
5061 from a filter that is running in a subprocess using a pipe, in order to
5062 try to track down a specific problem.
5064 4. Testing facility fudge: when running in the test harness and attempting
5065 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
5066 getting "No route to host". Convert this to a timeout.
5068 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
5071 6. Some OS don't have socklen_t but use size_t instead. This affects the
5072 fifth argument of getsockopt() amongst other things. This is now
5073 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
5074 can be set for individual OS. I have set it for SunOS5, OSF1, and
5075 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
5076 some earlier ones do not.
5078 7. Change 4.30/15 was not doing the test caselessly.
5080 8. The standard form for an IPv6 address literal was being rejected by address
5081 parsing in, for example, MAIL and RCPT commands. An example of this kind of
5082 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5083 this, as well as the form without the "IPv6" on the front (but only when
5084 address literals are enabled, of course).
5086 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
5088 10. Exim crashed if a message with an empty sender address specified by -f
5089 encountered a router with an errors_to setting. This could be provoked only
5090 by a command such as
5094 where an empty string was supplied; "<>" did not hit this bug.
5096 11. Installed PCRE release 4.5.
5098 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
5099 remained set. It is now erased.
5101 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
5102 times from message ids (which are base 36 rather than the normal 62).
5104 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
5105 were being counted as actual protocol errors, and logged if the log
5106 selector +smtp_protocol_error was set. One cannot be perfect in this test,
5107 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
5108 and DATA following a set of rejected RCPTs do not count as protocol errors.
5109 In other words, Exim assumes they were pipelined, though this may not
5110 actually be the case. Of course, in all cases the client gets an
5111 appropriate error code.
5113 15. If a lookup fails in an ACL condition, a message about the failure may
5114 be available; it is used if testing the ACL cannot continue, because most
5115 such messages specify what the cause of the deferral is. However, some
5116 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
5117 that caused an old message to be retained and used if a later statement
5118 caused a defer, replacing the real cause of the deferral.
5120 16. If an IP address had so many PTR records that the DNS lookup buffer
5121 was not large enough to hold them, Exim could crash while trying to process
5122 the truncated data. It now detects and logs this case.
5124 17. Further to 4.21/58, another change has been made: if (and only if) the
5125 first line of a message (the first header line) ends with CRLF, a bare LF
5126 in a subsequent header line has a space inserted after it, so as not to
5127 terminate the header.
5129 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
5130 unnecessarily, used atoi() instead of strtol(), and didn't check the
5131 termination when getting file sizes from file names by regex.
5133 19. Completely re-implemented the support for maildirsize files, in the light
5134 of a number of problems with the previous contributed implementation
5135 (4.30/29). In particular:
5137 . If the quota is zero, the maildirsize file is maintained, but no quota is
5140 . If the maildir directory does not exist, it is created before any attempt
5141 to write a maildirsize file.
5143 . The quota value in the file is just a cache; if the quota is changed in
5144 the transport, the new value overrides.
5146 . A regular expression is available for excluding directories from the
5149 20. The autoreply transport checks the characters in options that define the
5150 message's headers; it allows continued headers, but it was checking with
5151 isspace() after an embedded newline instead of explicitly looking for a
5154 21. If all the "regular" hosts to which an address was routed had passed their
5155 expiry times, and had not reached their retry times, the address was
5156 bounced, even if fallback hosts were defined. Now Exim should go on to try
5159 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
5160 equivalent code in the SMTP transport. Some hosts send humungous responses
5161 to HELO/EHLO, more than 1024 it seems.
5163 23. Refactoring: code in filter.c used (void *) for "any old type" but this
5164 gives compiler warnings in some environments. I've now done it "properly",
5167 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
5168 (because of problems with the built-in one) was declared to return uschar *
5169 instead of char *, causing compiler failure.
5171 25. Fixed a file descriptor leak when processing alias/forward files.
5173 26. Fixed a minor format string issue in dbfn.c.
5175 27. Typo in exim.c: ("dmbnz" for "dbmnz").
5177 28. If a filter file refered to $h_xxx or $message_headers, and the headers
5178 contained RFC 2047 "words", Exim's memory could, under certain conditions,
5181 29. When a sender address is verified, it is cached, to save repeating the test
5182 when there is more than one recipient in a message. However, when the
5183 verification involves a callout, it is possible for different callout
5184 options to be set for different recipients. It is too complicated to keep
5185 track of this in the cache, so now Exim always runs a verification when a
5186 callout is required, relying on the callout cache for the optimization.
5187 The overhead is duplication of the address routing, but this should not be
5190 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
5191 to be verified with callout=postmaster, and the main callout worked but the
5192 postmaster check failed, the verification correctly failed. However, if a
5193 subsequent RCPT command asked for sender verification *without* the
5194 postmaster check, incorrect caching caused this verification also to fail,
5197 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
5198 it was not caching the DNS options (qualify_single, search_parents) that
5199 were used when the lookup failed. A subsequent lookup with different
5200 options therefore always gave the same answer, though there were cases
5201 where it should not have. (Example: a "domains = !$mx_any" option on a
5202 dnslookup router: the "domains" option is always processed without any
5203 widening, but the router might have qualify_single set.) Now Exim uses the
5204 cached value only when the same options are set.
5206 32. Added John Jetmore's "exipick" utility to the distribution.
5208 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
5209 than a timeout (e.g. a certificate is required, and is not provided), an
5210 Exim server now closes the connection immediately. Previously it waited for
5211 the client to close - but if the client is SSL, it seems that they each
5212 wait for each other, leading to a delay before one of them times out.
5214 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
5215 maintained 0.8.x compatibility because I don't think many are using it, and
5216 it is clearly obsolete.
5218 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
5221 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
5222 client certificate was expired. A simple patch fixes this, though I don't
5223 understand the full logic of why the verify callback is called multiple
5226 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
5227 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
5228 which causes problems with some clients (such as the Certicom SSL Plus
5229 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
5230 disables the coutermeasure allowing Eudora to connect."
5232 38. Exim was not checking that a write() to a log file succeeded. This could
5233 lead to Bad Things if a log got too big, in particular if it hit a file
5234 size limit. Exim now panics and dies if it cannot write to a log file, just
5235 as it does if it cannot open a log file.
5237 39. Modified OS/Makefile-Linux so that it now contains
5239 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
5241 The two -D definitions ensure that Exim is compiled with large file
5242 support, which makes it possible to handle log files that are bigger than
5245 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
5246 instance) a domain was checked against a named list that involved a lookup,
5247 causing $domain_data to be set, then another domain was checked against the
5248 same list, then the first domain was re-checked, the value of $domain_data
5249 after the final check could be wrong. In particular, if the second check
5250 failed, it could be set empty. This bug probably also applied to
5253 41. The strip_trailing_dot option was not being applied to the address given
5254 with the -f command-line option.
5256 42. The code for reading a message's header from the spool was incrementing
5257 $received_count, but never initializing it. This meant that the value was
5258 incorrect (doubled) while delivering a message in the same process in which
5259 it was received. In the most common configuration of Exim, this never
5260 happens - a fresh exec is done - but it can happen when
5261 deliver_drop_privilege is set.
5263 43. When Exim logs an SMTP synchronization error - client data sent too soon -
5264 it now includes up to 150 characters of the unexpected data in the log
5267 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
5268 and building data strings. The size of both of these buffers was 10 000
5269 bytes - far larger than anybody would *ever* want, thought I. Needless to
5270 say, somebody hit the limit. I have increased the maximum line length to
5271 20 000 and the maximum data length of concatenated lines to 100 000. I have
5272 also fixed two bugs, because there was no checking on these buffers. Tsk,
5273 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
5274 buffer is too small.
5276 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
5277 lsearch lookups. Now it does.
5279 46. When parsing a route_list item in a manualroute router, a fixed-length
5280 buffer was used for the list of hosts. I made this 1024 bytes long,
5281 thinking that nobody would ever have a list of hosts that long. Wrong.
5282 Somebody had a whole pile of complicated expansion conditions, and the
5283 string was silently truncated, leading to an expansion error. It turns out
5284 that it is easier to change to an unlimited length (owing to other changes
5285 that have happened since this code was originally written) than to build
5286 structure for giving a limitation error. The length of the item that
5287 expands into the list of hosts is now unlimited.
5289 47. The lsearch lookup could not handle data where the length of text line was
5290 more than 4095 characters. Such lines were truncated, leading to shortened
5291 data being returned. It should now handle lines of any length.
5293 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
5294 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
5297 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
5298 replacing strings such as BIN_DIRECTORY in a source file: the replacement
5299 no longer happens in comment lines. A list of replacements is now placed
5300 at the head of all of the source files, except those whose only change is
5301 to replace PERL_COMMAND in the very first #! line.
5303 50. Replaced the slow insertion sort in queue.c, for sorting the list of
5304 messages on the queue, with a bottom-up merge sort, using code contributed
5305 by Michael Haardt. This should make operations like -bp somewhat faster on
5306 large queues. It won't affect queue runners, except when queue_run_in_order
5309 51. Installed eximstats 1.31 in the distribution.
5311 52. Added support for SRV lookups to the dnslookup router.
5313 53. If an ACL referred to $message_body or $message_body_end, the value was not
5314 reset for any messages that followed in the same SMTP session.
5316 54. The store-handling optimization for building very long strings was not
5317 differentiating between the different store pools. I don't think this
5318 actually made any difference in practice, but I've tidied it.
5320 55. While running the routers to verify a sender address, $sender_address
5321 was still set to the sender address. This is wrong, because when routing to
5322 send a bounce to the sender, it would be empty. Therefore, I have changed
5323 it so that, while verifying a sender address, $sender_address is set to <>.
5324 (There is no change to what happens when verifying a recipient address.)
5326 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
5327 target A or AAAA records (if not already returned) without resetting the
5328 qualify_single or search_parents options of the DNS resolver. These are
5329 inappropriate in this case because the targets of MX and SRV records must
5330 be FQDNs. A broken DNS record could cause trouble if it happened to have a
5331 target that, when qualified, matched something in the local domain. These
5332 two options are now turned off when doing these lookups.
5334 57. It seems that at least some releases of Reiserfs (which does not have the
5335 concept of a fixed number of inodes) returns zero and not -1 for the
5336 number of available inodes. This interacted badly with check_spool_inodes,
5337 which assumed that -1 was the "no such thing" setting. What I have done is
5338 to check that the total number of inodes is greater than zero before doing
5339 the test of how many are available.
5341 58. When a "warn" ACL statement has a log_message modifier, the message is
5342 remembered, and not repeated. This is to avoid a lot of repetition when a
5343 message has many recipients that cause the same warning to be written.
5344 Howewer, Exim was preserving the list of already written lines for an
5345 entire SMTP session, which doesn't seem right. The memory is now reset if a
5346 new message is started.
5348 59. The "rewrite" debugging flag was not showing the result of rewriting in the
5349 debugging output unless log_rewrite was also set.
5351 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
5352 of (int)(handle) when we know that handle contains (void *)(-1).
5354 61. The Exim daemon panic-logs an error return when it closes the incoming
5355 connection. However "connection reset by peer" seems to be common, and
5356 isn't really an error worthy of noting specially, so that particular error
5359 62. When Exim is trying to find all the local interfaces, it used to panic and
5360 die if the ioctl to get the interface flags failed. However, it seems that
5361 on at least one OS (Solaris 9) it is possible to have an interface that is
5362 included in the list of interfaces, but for which you get a failure error
5363 for this call. This happens when the interface is not "plumbed" into a
5364 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
5365 failure of the "get flags" call assumes that the interface is down.
5367 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
5368 makes life easier for people who are doing arithmetic on fields extracted
5369 from dates, where you often get leading zeros that should not be
5370 interpreted as octal.
5372 64. Added qualify_domain to the redirect router, to override the global
5375 65. If a pathologically long header line contained very many addresses (the
5376 report of this problem mentioned 10 000) and each of them was rewritten,
5377 Exim could use up a very large amount of memory. (It kept on making new
5378 copies of the header line as it rewrote, and never released the old ones.)
5379 At the expense of a bit more processing, the header rewriting function has
5380 been changed so that it no longer eats memory in this way.
5382 66. The generation of the Received: header has been moved from the time that a
5383 message starts to be received, to the time that it finishes. The timestamp
5384 in the Received: header should now be very close to that of the <= log
5385 line. There are two side-effects of this change:
5387 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
5388 logged header lines no longer include the local Received: line, because
5389 it has not yet been created. The same applies to a copy of the message
5390 that is returned to a non-SMTP sender when a message is rejected.
5392 (b) When a filter file is tested using -bf, no additional Received: header
5393 is added to the test message. After some thought, I decided that this
5396 This change does not affect the value of $received_for. It is still set
5397 after address rewriting, but before local_scan() is called.
5399 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
5401 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
5402 gave an unhelpful panic error message, and a defer error. I have managed to
5403 change this behaviour so that it now rejects any supplied certificate,
5404 which seems right, as the list of acceptable certificates is empty.
5406 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
5407 gave an unhelpful defer error. I have not managed to make this reject any
5408 supplied certificates, but the error message it gives is "no certificate
5409 supplied", which is not helpful.
5411 70. exigrep's output now also includes lines that are not associated with any
5412 message, but which match the given pattern. Implemented by a patch from
5413 Martin Sluka, which also tidied up the Perl a bit.
5415 71. Recipient callout verification, like sender verification, was using <> in
5416 the MAIL FROM command. This isn't really the right thing, since the actual
5417 sender may affect whether the remote host accepts the recipient or not. I
5418 have changed it to use the actual sender in the callout; this means that
5419 the cache record is now keyed on a recipient/sender pair, not just the
5420 recipient address. There doesn't seem to be a real danger of callout loops,
5421 since a callout by the remote host to check the sender would use <>.
5422 [SEE ABOVE: changed after hitting problems.]
5424 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
5425 temporary errors. However, in the case of such a code being given after
5426 the end of a data transmission (i.e. after ".") Exim was failing to write
5427 a retry record for the message. (Yes, there was some broken host that was
5428 actually sending 8xx at this point.)
5430 73. An unknown lookup type in a host list could cause Exim to panic-die when
5431 the list was checked. (An example that provoked this was putting <; in the
5432 middle of a list instead of at the start.) If this happened during a DATA
5433 ACL check, a -D file could be left lying around. This kind of configuration
5434 error no longer causes Exim to die; instead it causes a defer errror. The
5435 incident is still logged to the main and panic logs.
5437 74. Buglet left over from Exim 3 conversion. The message "too many messages
5438 in one connection" was written to the rejectlog but not the mainlog, except
5439 when address rewriting (yes!) was being logged.
5441 75. Added write_rejectlog option.
5443 76. When a system filter was run not as root (that is, when system_filter_user
5444 was set), the values of the $n variables were not being returned to the
5445 main process; thus, they were not subsequently available in the $sn
5448 77. Added +return_path_on_delivery log selector.
5450 78. A connection timeout was being treated differently from recipients deferred
5451 when testing hosts_max_try with a message that was older than the host's
5452 retry timeout. (The host should not be counted, thus allowing all hosts to
5453 be tried at least once before bouncing.) This may have been the cause of an
5454 occasionally reported bug whereby a message would remain on the queue
5455 longer than the retry timeout, but would be bounced if a delivery was
5456 forced. I say "may" because I never totally pinned down the problem;
5457 setting up timeout/retry tests is difficult. See also the next item.
5459 79. The ultimate address timeout was not being applied to errors that involved
5460 a combination of host plus message (for example, a timeout on a MAIL
5461 command). When an address resolved to a number of possible hosts, and they
5462 were not all tried for each delivery (e.g. because of hosts_max_try), a
5463 message could remain on the queue longer than the retry timeout.
5465 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
5468 81. Fixed an obscure SMTP outgoing bug which required at least the following
5469 conditions: (a) there was another message waiting for the same server;
5470 (b) the server returned 5xx to all RCPT commands in the first message so
5471 that the message was not completed; (c) the server dropped the connection
5472 or gave a negative response to the RSET that Exim sends to abort the
5473 transaction. The observed case was a dropped connection after DATA that had
5474 been sent in pipelining mode. That is, the server had advertised PIPELINING
5475 but was not implementing it correctly. The effect of the bug was incorrect
5476 behaviour, such as trying another host, and this could lead to a crash.
5482 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
5483 and daemon.c were passed as pointers to ints; they should have been
5484 pointers to socklen_t variables (which are typically unsigned ints).
5486 2. Some signed/unsigned type warnings in the os.c file for Linux have been
5489 3. Fixed a really odd bug that affected only the testing scheme; patching a
5490 certain fixed string in the binary changed the value of another string that
5491 happened to be identical to the end of the original first string.
5493 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
5494 name", it returns that address as the IP address. On some operating
5495 systems (e.g. Solaris), it also passes back the IP address string as the
5496 "host name". However, on others (e.g. Linux), it passes back an empty
5497 string. Exim wasn't checking for this, and was changing the host name to an
5498 empty string, assuming it had been canonicized.
5500 5. Although rare, it is permitted to have more than one PTR record for a given
5501 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
5502 all the names associated with an address, because they do in Solaris.
5503 However, it seems that they do not in Linux for data that comes from the
5504 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
5505 I found this out when I moved to a new Linux workstation and tried to run
5506 the Exim test suite.
5508 To get round this problem I have changed the code so that it now does its
5509 own call to the DNS to look up PTR records when searching for a host name.
5510 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
5511 addresses that are only in /etc/hosts are still found.
5513 This behaviour is, however, controlled by an option called host_lookup_
5514 order, which defaults to "bydns:byaddr". If people want to use the other
5515 order, or indeed, just use one or the other means of lookup, they can
5516 specify it in this variable.
5518 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
5519 some operating systems, this comes back from gethostbyaddr() as an empty
5520 string, and this is what Exim used to test for. However, it seems that in
5521 other systems, "." is yielded. Exim now tests for this case too.
5523 7. The values of check_spool_space and check_log_space are now held internally
5524 as a number of kilobytes instead of an absolute number of bytes. If a
5525 numbers is specified without 'K' or 'M', it is rounded up to the nearest
5526 kilobyte. This means that much larger values can be stored.
5528 8. Exim monitor: an attempt to get the action menu when not actually pointing
5529 at a message produces an empty menu entitled "No message selected". This
5530 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
5531 no entries in it ("Shell widget menu has zero width and/or height"). So I
5532 have added a single, blank menu entry in this case.
5534 9. Added ${quote_local_part.
5536 10. MIME decoding is now applied to the contents of Subject: header lines when
5539 11. Now that a reference to $sender_host_address automatically causes a reverse
5540 lookup to occur if necessary (4.13/18), there is no need to arrange for a
5541 host lookup before query-style lookups in lists that might use this
5542 variable. This has therefore been abolished, and the "net-" prefix is no
5543 longer necessary for query-style lookups.
5545 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
5546 have been a typo for LFLAGS, so it has been changed.
5548 13. The install script calls Exim with "-C /dev/null" in order to find the
5549 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
5550 to be output. Howeve, since Exim outputs its version number before the
5551 error, it didn't break the script. It just looked ugly. I fixed this by
5552 always allowing "-C /dev/null" if the caller is root.
5554 14. Ignore overlarge ACL variable number when reading spool file - insurance
5555 against a later release with more variables having written the file.
5557 15. The standard form for an IPv6 address literal was being rejected by EHLO.
5558 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5559 this, as well as the form without the "IPv6" on the front.
5561 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
5562 OS/Makefile-Darwin file.
5564 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
5565 only to LDAP libraries that do not have LDAP_OPT_DEREF.
5567 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
5568 variable. However, some OS use ints for inodes. I've added cast to long int
5569 to get rid of the compiler warning.
5571 19. I had forgotten to lock out "/../" in configuration file names when
5572 ALT_CONFIG_PREFIX was set.
5574 20. Routers used for verification do not need to specify transports. However,
5575 if such a router generated a host list, and callout was configured, Exim
5576 crashed, because it could not find a port number from the (non-existent)
5577 transport. It now assumes port 25 in this circumstance.
5579 21. Added the -t option to exigrep.
5581 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
5582 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
5583 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
5584 them is set, LOOKUP_LSEARCH is forced.
5586 23. "exim -bV" now outputs a list of lookups that are included in the binary.
5588 24. Added sender and host information to the "rejected by local_scan()" log
5589 line; previously there was no indication of these.
5591 25. Added .include_if_exists.
5593 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
5594 Linux. It turns out that not all file systems support this. Apparently some
5595 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
5596 do it.) To cope with this, the error EINVAL, which means that sync-ing is
5597 not supported on the file descriptor, is now ignored when Exim is trying to
5598 sync a directory. This applies only to Linux.
5600 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
5602 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
5603 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
5606 29. Added support for maildirsize files from supplied patch (modified a bit).
5608 30. The use of :fail: followed by an empty string could lead Exim to respond to
5609 sender verification failures with (e.g.):
5611 550 Verification failed for <xxx>
5612 550 Sender verify failed
5614 where the first response line was missing the '-' that indicates it is not
5615 the final line of the response.
5617 31. The loop for finding the name of the user that called Exim had a hardwired
5618 limit of 10; it now uses the value of finduser_retries, which is used for
5619 all other user lookups.
5621 32. Added $received_count variable, available in data and not_smtp ACLs, and at
5624 33. Exim was neglecting to zero errno before one call of strtol() when
5625 expanding a string and expecting an integer value. On some systems this
5626 resulted in spurious "integer overflow" errors. Also, it was casting the
5627 result into an int without checking.
5629 34. Testing for a connection timeout using "timeout_connect" in the retry rules
5630 did not work. The code looks as if it has *never* worked, though it appears
5631 to have been documented since at least releast 1.62. I have made it work.
5633 35. The "timeout_DNS" error in retry rules, also documented since at least
5634 1.62, also never worked. As it isn't clear exactly what this means, and
5635 clearly it isn't a major issue, I have abolished the feature by treating it
5636 as "timeout", and writing a warning to the main and panic logs.
5638 36. The display of retry rules for -brt wasn't always showing the error code
5641 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
5642 timeout_connect_A, timeout_connect_MX.
5644 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
5645 to the empty sender.
5647 39. The daemon was not analysing the content of -oX till after it had closed
5648 stderr and disconnected from the controlling terminal. This meant that any
5649 syntax errors were only noted on the panic log, and the return code from
5650 the command was 0. By re-arranging the code a little, I've made the
5651 decoding happen first, so such errors now appear on stderr, and the return
5652 code is 1. However, the actual setting up of the sockets still happens in
5653 the disconnected process, so errors there are still only recorded on the
5656 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
5657 connections (as happens on some IP stacks) was logged at start up time as
5658 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
5659 it from "IPv6 and IPv4", which means that two separate sockets are being
5662 41. The debug output for gethostbyname2() or getipnodebyname() failures now
5663 says whether AF_INET or AF_INET6 was passed as an argument.
5665 42. Exiwhat output was messed up when time zones were included in log
5668 43. Exiwhat now gives more information about the daemon's listening ports,
5669 and whether -tls-on-connect was used.
5671 44. The "port" option of the smtp transport is now expanded.
5673 45. A "message" modifier in a "warn" statement in a non-message ACL was being
5674 silently ignored. Now an error message is written to the main and panic
5677 46. There's a new ACL modifier called "logwrite" which writes to a log file
5678 as soon as it is encountered.
5680 47. Added $local_user_uid and $local_user_gid at routing time.
5682 48. Exim crashed when trying to verify a sender address that was being
5685 49. Exim was recognizing only a space character after ".include". It now also
5686 recognizes a tab character.
5688 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
5689 extracting the relevant information from the specification. The man page no
5690 longer contains scrambled data for the -d option, and I've added a section
5691 at the front about calling Exim under different names.
5693 51. Added "extra_headers" argument to the "mail" command in filter files.
5695 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
5698 53. Installed eximstats 1.29.
5700 54. Added transport_filter_timeout as a generic transport option.
5702 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
5703 Cc: header lines. This was required by RFC 822, but it not required by RFC
5706 56. Exim used to add From:, Date:, and Message-Id: header lines to any
5707 incoming messages that did not have them. Now it does so only if the
5708 message originates locally, that is, if there is no associated remote host
5709 address. When Resent- header lines are present, this applies to the Resent-
5710 lines rather than the non-Resent- lines.
5712 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
5713 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
5715 58. Messages for configuration errors now include the name of the main
5716 configuration file - useful now that there may be more than one file in a
5717 list (.included file names were always shown).
5719 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
5720 for those rare installations that do not start the daemon as root or run it
5721 setuid root. I've cut out the call to initgroups() if the daemon is not
5724 60. The Exim user and group can now be bound into the binary as text strings
5725 that are looked up at the start of Exim's processing.
5727 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
5729 62. Added $mailstore_basename variable.
5731 63. Installed patch to sieve.c from Michael Haardt.
5733 64. When Exim failed to open the panic log after failing to open the main log,
5734 the original message it was trying to log was written to stderr and debug
5735 output, but if they were not available (the usual case in production), it
5736 was lost. Now it is written to syslog before the two lines that record the
5737 failures to open the logs.
5739 65. Users' Exim filters run in subprocesses under the user's uid. It is
5740 possible for a "deliver" command or an alias in a "personal" command to
5741 provoke an address rewrite. If logging of address rewriting is configured,
5742 this fails because the process is not running as root or exim. There may be
5743 a better way of dealing with this, but for the moment (because 4.30 needs
5744 to be released), I have disabled address rewrite logging when running a
5745 filter in a non-root, non-exim process.
5751 1. The buildconfig auxiliary program wasn't quoting the value set for
5752 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
5753 not defined. This bug was masked in 4.22 by the effect that was fixed in
5756 2. Some messages that were rejected after a message id was allocated were
5757 shown as "incomplete" by exigrep. It no longer does this for messages that
5758 are rejected by local_scan() or the DATA or non-SMTP ACLs.
5760 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
5761 have allow_domain_literals set, the ID did not get logged in the <= line.
5762 Domain literals are now always recognized in Message-ID: header lines.
5764 4. The first argument for a ${extract expansion item is the key name or field
5765 number. Leading and trailing spaces in this item were not being ignored,
5766 causing some misleading effects.
5768 5. When deliver_drop_privilege was set, single queue runner processes started
5769 manually (i.e. by the command "exim -q") or by the daemon (which uses the
5770 same command in the process it spins off) were not dropping privilege.
5772 6. When the daemon running as "exim" started a queue runner, it always
5773 re-executed Exim in the spun-off process. This is a waste of effort when
5774 deliver_drop_privilege is set. The new process now just calls the
5775 queue-runner function directly.
5781 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
5784 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
5785 ignored. Though the use of -oP was forcing the writing of a pid file, it
5786 was always written to the default place.
5788 3. If the message "no IP address found for host xxxx" is generated during
5789 incoming verification, it is now followed by identification of the incoming
5790 connection (so you can more easily find what provoked it).
5792 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
5794 5. Added some features to "harden" Exim a bit more against certain attacks:
5796 (a) There is now a build-time option called FIXED_NEVER_USERS that can
5797 be put in Local/Makefile. This is like the never_users runtime option,
5798 but it cannot be overridden. The default setting is "root".
5800 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
5801 prefix string with which any file named in a -C command line option
5804 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
5805 is retained for -C and -D only if the caller of Exim is root. Without
5806 it, the exim user may also use -C and -D and retain privilege.
5808 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
5809 command line option is disabled.
5811 6. Macro names set by the -D option must start with an upper case letter, just
5812 like macro names defined in the configuration file.
5814 7. Added "dereference=" facility to LDAP.
5816 8. Two instances of the typo "uknown" in the source files are fixed.
5818 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
5819 the Configure-Makefile script screwed up while processing it.
5821 10. Incorporated PCRE 4.4.
5823 11. The SMTP synchronization check was not operating right at the start of an
5824 SMTP session. For example, it could not catch a HELO sent before the client
5825 waited for the greeting. There is now a check for outstanding input at the
5826 point when the greeting is written. Because of the duplex, asynchronous
5827 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
5828 way, but not yet received, when the check is performed.
5830 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
5831 on TCP/IP sockets, because this apparently causes some broken clients to
5834 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
5835 unchanged) from the Cygwin maintainer.
5837 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
5838 was supported instead of testing for mbx. Effectively a typo.
5840 15. The spa authenticator server code was not checking that the input it
5841 received was valid base64.
5843 16. The debug output line for the "set" modifier in ACLs was not showing the
5844 name of the variable that was being set.
5846 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
5848 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
5849 lookup on its own. Something else was needed to trigger the lookup. For
5850 example, a match in host_lookup or the need for a host name in a host list.
5851 Now, if $sender_host_name is referenced and the host name has not yet been
5852 looked up, a lookup is performed. If the lookup fails, the variable remains
5853 empty, and $host_lookup_failed is set to "1".
5855 19. Added "eqi" as a case-independent comparison operator.
5857 20. The saslauthd authentication condition could segfault if neither service
5858 nor realm was specified.
5860 21. If an overflowing value such as "2048M" was set for message_size_limit, the
5861 error message that was logged was misleading, and incoming SMTP
5862 connections were dropped. The message is now more accurate, and temporary
5863 errors are given to SMTP connections.
5865 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
5866 (except RSET) with a 421 error, until QUIT is received. However, it was
5867 failing to send a response to QUIT.
5869 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
5870 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
5871 ACL is now run after the helo_try_verify_hosts code.
5873 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
5874 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
5875 case-independent, so other case variants are also recognized). Apparently
5876 some systems use these upper case variants.
5878 25. If more than two messages were waiting for the same host, and a transport
5879 filter was specified for the transport, Exim sent two messages over the
5880 same TCP/IP connection, and then failed with "socket operation on non-
5881 socket" when it tried to send the third.
5883 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
5885 27. The extern definition of crypt16() in expand.c was not being excluded when
5886 the OS had its own crypt16() function.
5888 28. Added bounce_return_body as a new option, and bounce_return_size_limit
5889 as a preferred synonym for return_size_limit, both as an option and as an
5892 29. Added LIBS=-liconv to OS/Makefile-OSF1.
5894 30. Changed the default configuration ACL to relax the local part checking rule
5895 for addresses that are not in any local domains. For these addresses,
5896 slashes and pipe symbols are allowed within local parts, but the sequence
5897 /../ is explicitly forbidden.
5899 31. SPA server authentication was not clearing the challenge buffer before
5902 32. log_message in a "warn" ACL statement was writing to the reject log as
5903 well as to the main log, which contradicts the documentation and doesn't
5904 seem right (because no rejection is happening). So I have stopped it.
5906 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
5907 However, I am unable to do any testing of this.
5909 34. Fixed an infelicity in the appendfile transport. When checking directories
5910 for a mailbox, to see if any needed to be created, it was accidentally
5911 using path names with one or more superfluous leading slashes; tracing
5912 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
5914 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
5915 contents are added to the log line that is written for every discarded
5916 recipient. (Previously a log_message setting was ignored.)
5918 36. The ${quote: operator now quotes the string if it is empty.
5920 37. The install script runs exim in order to find its version number. If for
5921 some reason other than non-existence or emptiness, which it checks, it
5922 could not run './exim', it was installing it with an empty version number,
5923 i.e. as "exim-". This error state is now caught, and the installation is
5926 38. An argument was missing from the function that creates an error message
5927 when Exim fails to connect to the socket for saslauthd authentication.
5928 This could cause Exim to crash, or give a corrupted message.
5930 39. Added isip, isip4, and isip6 to ${if conditions.
5932 40. The ACL variables $acl_xx are now saved with the message, and can be
5933 accessed later in routers, transports, and filters.
5935 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
5936 strings in the file are not string-expanded.
5938 42. If a MAIL command specified a SIZE value that was too large to fit into an
5939 int variable, the check against message_size_limit failed. Such values are
5940 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
5941 day this will have to be increased, but I don't think I want to be around
5942 when emails are that large.
5949 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
5950 iconv() is not standard in FreeBSD.
5952 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
5953 IPv6 enabled. The observed symptom was a segmentation fault on return from
5954 the function os_common_find_running_interfaces() in src/os.c.
5956 3. In the check_special_case() function in daemon.c I had used "errno" as an
5957 argument name, which causes warnings on some systems. This was basically a
5958 typo, since it was named "eno" in the comments!
5960 4. The code that waits for the clock to tick (at a resolution of some fraction
5961 of a second) so as to ensure message-id uniqueness was always waiting for
5962 at least one whole tick, when it could have waited for less. [This is
5963 almost certainly not relevant at current processor speeds, where it is
5964 unlikely to ever wait at all. But we try to future-proof.]
5966 5. The function that sleeps for a time interval that includes fractions of a
5967 second contained a race. It did not block SIGALRM between setting the
5968 timer, and suspending (a couple of lines later). If the interval was short
5969 and the sigsuspend() was delayed until after it had expired, the suspension
5970 never ended. On busy systems this could lead to processes getting stuck for
5973 6. Some uncommon configurations may cause a lookup to happen in a queue runner
5974 process, before it forks any delivery processes. The open lookup caching
5975 mechanism meant that the open file or database connection was passed into
5976 the delivery process. The problem was that delivery processes always tidy
5977 up cached lookup data. This could cause a problem for the next delivery
5978 process started by the queue runner, because the external queue runner
5979 process does not know about the closure. So the next delivery process
5980 still has data in the lookup cache. In the case of a file lookup, there was
5981 no problem because closing a file descriptor in a subprocess doesn't affect
5982 the parent. However, if the lookup was caching a connection to a database,
5983 the connection was closed, and the second delivery process was likely to
5984 see errors such as "PGSQL: query failed: server closed the connection
5985 unexpectedly". The problem has been fixed by closing all cached lookups
5986 in a queue runner before running a delivery process.
5988 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
5989 seem to have the "const" qualifier which it has on other OS. I've
5992 8. Change 4.21/2 was too strict. It is only if there are two authenticators
5993 *of the same type* (client or server) with the same public name that an
5994 error should be diagnosed.
5996 9. When Exim looked up a host name for an IP address, but failed to find the
5997 original IP address when looking up the host name (a safety check), it
5998 output the message "<ip address> does not match any IP for NULL", which was
5999 confusing, to say the least. The bug was that the host name should have
6000 appeared instead of "NULL".
6002 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
6003 user that is built into the binary, and the -C or -D options is used, root
6004 privilege is dropped before the configuration file is read. In addition,
6005 logging is switched to stderr instead of the normal log files. If the
6006 configuration then re-defines the Exim user, the unprivileged environment
6007 is probably not what is expected, so Exim logs a panic warning message (but
6010 However, if deliver_drop_privilege is set, the unprivileged state may well
6011 be exactly what is intended, so the warning has been cut out in that case,
6012 and Exim is allowed to try to write to its normal log files.
6018 1. smtp_return_error_details was not giving details for temporary sender
6019 or receiver verification errors.
6021 2. Diagnose a configuration error if two authenticators have the same public
6024 3. Exim used not to create the message log file for a message until the first
6025 delivery attempt. This could be confusing when incoming messages were held
6026 for policy or load reasons. The message log file is now created at the time
6027 the message is received, and an initial "Received" line is written to it.
6029 4. The automatically generated man page for command line options had a minor
6030 bug that caused no ill effects; however, a more serious problem was that
6031 the procedure for building the man page automatically didn't always
6032 operate. Consequently, release 4.20 contains an out-of-date version. This
6033 shouldn't happen again.
6035 5. When building Exim with embedded Perl support, the script that builds the
6036 Makefile was calling 'perl' to find its compile-time parameters, ignoring
6037 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
6039 6. The freeze_tell option was not being used for messages that were frozen on
6040 arrival, either by an ACL or by local_scan().
6042 7. Added the smtp_incomplete_transaction log selector.
6044 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
6045 was accepting AUTH without a new EHLO.
6047 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
6048 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
6051 10. Logging of TCP/IP connections (when configured) now happens in the main
6052 daemon process instead of the child process, so that the TCP/IP connection
6053 count is more accurate (but it can never be perfect).
6055 11. The use of "drop" in a nested ACL was not being handled correctly in the
6056 outer ACL. Now, if condition failure induced by the nested "drop" causes
6057 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
6058 or "require"), the connection is dropped.
6060 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
6061 that yield "discard" can now be used with an "accept" or a "discard" verb,
6062 but an error is generated for any others (because I can't see a useful way
6063 to define what should happen).
6065 13. When an ACL is read dynamically from a file (or anywhere else), the lines
6066 are now processed in the same way as lines in the Exim configuration file.
6067 In particular, continuation lines are supported.
6069 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
6071 15. Added -ti meaning -t -i.
6073 16. Check for letters, digits, hyphens, and dots in the names of dnslist
6074 domains, and warn by logging if others are found.
6076 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
6077 returned from GIFCONF when Exim is trying to find the list of interfaces on
6078 a host. The code in os.c has been modified to copy each ifreq to an aligned
6079 structure in all cases.
6081 Also, in some cases, the returned ifreq's were being copied to a 'struct
6082 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
6083 means the last couple of bytes of an IPv6 address could be chopped if the
6084 ifreq contained only a normal sockaddr (14 bytes storage).
6086 18. Named domain lists were not supported in the hosts_treat_as_local option.
6087 An entry such as +xxxx was not recognized, and was treated as a literal
6090 19. Ensure that header lines added by a DATA ACL are included in the reject log
6091 if the ACL subsequently rejects the message.
6093 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
6094 MD5 (which is deprecated).
6096 21. When testing a filter file using -bf, Exim was writing a message when it
6097 took the sender from a "From " line in the message, but it was not doing so
6098 when it took $return_path from a Return-Path: header line. It now does.
6100 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
6101 with a valid header line field name (a series of printing characters
6102 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
6104 23. Changed "disc" in the source to "disk" to conform to the documentation and
6105 the book and for uniformity.
6107 24. Ignore Sendmail's -Ooption=value command line item.
6109 25. When execve() failed while trying to run a command in a pipe transport,
6110 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
6111 could be confused with a return value of 69 from the command itself. This
6112 has been changed to 127, the value the shell returns if it is asked to run
6113 a non-existent command. The wording for the related log line suggests a
6114 non-existent command as the problem.
6116 26. If received_header_text expands to an empty string, do not add a Received:
6117 header line to the message. (Well, it adds a token one on the spool, but
6118 marks it "old" so that it doesn't get used or transmitted.)
6120 27. Installed eximstats 1.28 (addition of -nt option).
6122 28. There was no check for failure on the call to getsockname() in the daemon
6123 code. This can fail if there is a shortage of resources on the system, with
6124 ENOMEM, for example. A temporary error is now given on failure.
6126 29. Contrary to the C standard, it seems that in some environments, the
6127 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
6128 program. Exim now does this explicitly; it affects the formatting of
6129 timestamps using strftime().
6131 30. If exiqsumm was given junk data, it threw up some uninitialized variable
6132 complaints. I've now initialized all the variables, to avoid this.
6134 32. Header lines added by a system filter were not being "seen" during
6135 transport-time rewrites.
6137 33. The info_callback() function passed to OpenSSL is set up with type void
6138 (*)(SSL *, int, int), as described somewhere. However, when calling the
6139 function (actually a macro) that sets it up, the type void(*)() is
6140 expected. I've put in a cast to prevent warnings from picky compilers.
6142 34. If a DNS black list lookup found a CNAME record, but there were no A
6143 records associated with the domain it pointed at, Exim crashed.
6145 35. If a DNS black list lookup returned more than one A record, Exim ignored
6146 all but the first. It now scans all returned addresses if a particular IP
6147 value is being sought. In this situation, the contents of the
6148 $dnslist_value variable are a list of all the addresses, separated by a
6151 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
6152 to accept a host name and all its aliases if the forward lookup for any of
6153 them yielded the IP address of the incoming connection. Now it accepts only
6154 those names whose forward lookup yields the correct IP address. Any other
6155 names are discarded. This closes a loophole whereby a rogue DNS
6156 administrator could create reverse DNS records to break through a
6157 wildcarded host restriction in an ACL.
6159 37. If a user filter or a system filter that ran in a subprocess used any of
6160 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
6161 the wrong values were passed to the pipe command ($thisaddress had the
6162 value of $0, $0 had the value of $1, etc). This bug was introduced by
6163 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
6165 38. Improved the line breaking for long SMTP error messages from ACLs.
6166 Previously, if there was no break point between 40 and 75 characters, Exim
6167 left the rest of the message alone. Two changes have been made: (a) I've
6168 reduced the minimum length to 35 characters; (b) if it can't find a break
6169 point between 35 and 75 characters, it looks ahead and uses the first one
6170 that it finds. This may give the occasional overlong line, but at least the
6171 remaining text gets split now.
6173 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
6174 file descriptors might be low, and that setting 1000 would always raise it.
6175 It turns out that in some environments, the limit is already over 1000 and
6176 that lowering it causes trouble. So now Exim takes care not to decrease it.
6178 40. When delivering a message, the value of $return_path is set to $sender_
6179 address at the start of routing (routers may change the value). By an
6180 oversight, this default was not being set up when an address was tested by
6181 -bt or -bv, which affected the outcome if any router or filter referred to
6184 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
6185 log only when "message" or "log_message" are set. However, if one of the
6186 conditions was an address verification, or a call to a nested ACL, the
6187 messages generated by the underlying test were being passed through. This
6188 no longer happens. The underlying message is available in $acl_verify_
6189 message for both "message" and "log_message" expansions, so it can be
6190 passed through if needed.
6192 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
6193 new expansion $bh_ to give the encoded byte string without charset
6194 translation. Translation happens only if iconv() is available; HAVE_ICONV
6195 indicates this at build time. HEADERS_CHARSET gives the charset to
6196 translate to; headers_charset can change it in the configuration, and
6197 "headers charset" can change it in an individual filter file.
6199 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
6200 that creates RFC 2047 encoded "words" labels them as that charset instead
6201 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
6202 expansion operator; (ii) when Exim creates a From: line for a local
6203 message; (iii) when a header line is rewritten to include a "phrase" part.
6205 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
6206 buggy, causing it to skip the first lines of messages whose message ID
6207 ended in 'D'. This would not have bitten before Exim release 4.14, because
6208 message IDs were unlikely to end in 'D' before then. The effect was to have
6209 incorrect size information for certain domains.
6211 45. #include "config.h" was missing at the start of the crypt16.c module. This
6212 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
6215 46. If there was a timeout during a "random" callout check, Exim treated it as
6216 a failure of the random address, and carried on sending RSET and the real
6217 address. If the delay was just some slowness somewhere, the response to the
6218 original RCPT would be taken as a response to RSET and so on, causing
6219 mayhem of various kinds.
6221 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
6222 when I implemented it. It didn't allow for the fact that some option values
6223 may legitimatetly be negative (e.g. size_addition), and it didn't even do
6224 the right test for positive values.
6226 48. Domain names in DNS records are case-independent. Exim always looks them up
6227 in lower case. Some resolvers return domain names in exactly the case they
6228 appear in the zone file, that is, they may contain uppercase letters. Not
6229 all resolvers do this - some return always lower case. Exim was treating a
6230 change of case by a resolver as a change of domain, similar to a widening
6231 of a domain abbreviation. This triggered its re-routing code and so it was
6232 trying to route what was effectively the same domain again. This normally
6233 caused routing to fail (because the router wouldn't handle the domain
6234 twice). Now Exim checks for this case specially, and just changes the
6235 casing of the domain that it ultimately uses when it transmits the message
6238 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
6241 50. If a filter generated a file delivery with a non-absolute name (possible if
6242 no home directory exists for the router), the forbid_file option was not
6245 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
6246 the existing equality matching.
6248 52. Exim was using ints instead of ino_t variables in some places where it was
6249 dealing with inode numbers.
6251 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
6252 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
6253 called TMPDIR, and if it finds it is different, it changes its value.
6255 54. The smtp_printf() function is now made available to local_scan() so
6256 additional output lines can be written before returning. There is also an
6257 smtp_fflush() function to enable the detection of a dropped connection.
6258 The variables smtp_input and smtp_batched_input are exported to
6261 55. Changed the default runtime configuration: the message "Unknown user"
6262 has been removed from the ACL, and instead placed on the localuser router,
6263 using the cannot_route_message feature. This means that any verification
6264 failures that generate their own messages won't get overridden. Similarly,
6265 the "Unrouteable address" message that was in the ACL for unverifiable
6266 relay addresses has also been removed.
6268 56. Added hosts_avoid_esmtp to the smtp transport.
6270 57. The exicyclog script was not checking for the esoteric option
6271 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
6272 will work only if exicyclog is run under the appropriate euid.
6274 58. Following a discussion on the list, the rules by which Exim recognises line
6275 endings on incoming messages have been changed. The -dropcr and drop_cr
6276 options are now no-ops, retained only for backwards compatibility. The
6277 following line terminators are recognized: LF CRLF CR. However, special
6278 processing applies to CR:
6280 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
6281 nor a local message in the state where . is a terminator.
6283 (ii) If a bare CR is encountered in a header line, an extra space is added
6284 after the line terminator so as not to end the header. The reasoning
6285 behind this is that bare CRs in header lines are most likely either
6286 to be mistakes, or people trying to play silly games.
6288 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
6289 was being incorrectly given as 18 bytes larger than it should have been.
6290 This is a VOB (very old bug).
6292 60. This may never have affected anything current, but just in case it has:
6293 When the local host is found other than at the start of a list of hosts,
6294 the local host, those with the same MX, and any that follow, are discarded.
6295 When the list in question was part of a longer list of hosts, the following
6296 hosts (not currently being processed) were also being discarded. This no
6297 longer happens. I'm not sure if this situation could ever has previously
6300 61. Added the "/MX" feature to lists of hosts in the manualroute and query
6303 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
6304 header. This is something that is recommended in a new Internet Draft, and
6305 is something that is documented as being done by Sendmail. There are two
6306 possible values. For messages generated by the autoreply transport, Exim
6309 Auto-Submitted: auto-replied
6311 whereas for all other generated messages (e.g. bounces) it adds
6313 Auto-Submitted: auto-generated
6315 63. The "personal" condition in filters now includes a test for the
6316 Auto-Submitted: header. If it contains the string "auto-" the message it
6317 not considered personal.
6319 64. Added rcpt_include_affixes as a generic transport option.
6321 65. Added queue_only_override (default true).
6323 66. Added the syslog_duplication option.
6325 67. If what should have been the first header line of a message consisted of
6326 a space followed by a colon, Exim was mis-interpreting it as a header line.
6327 It isn't of course - it is syntactically invalid and should therefore be
6328 treated as the start of the message body. The misbehaviour could have
6329 caused a number of strange effects, including loss of data in subsequent
6330 header lines, and spool format errors.
6332 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
6333 client host had authenticated. This control can now be exercised by an ACL
6334 for more flexibility.
6336 69. By default, callouts do not happen when testing with -bh. There is now a
6337 variant, -bhc, which does actually run the callout code, including
6338 consulting and updating the callout cache.
6340 70. Added support for saslauthd authentication, courtesy of Alexander
6343 71. If statvfs() failed on the spool or log directories while checking their
6344 size for availability, Exim confusingly gave the error "space shortage".
6345 Furthermore, in debugging mode it crashed with a floating point exception.
6346 These checks are done if check_{spool,log}_{space,inodes} are set, and when
6347 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
6348 serious problem, Exim now writes to the main and panic logs when this
6349 happens, with details of the failure. It then refuses to accept the
6350 incoming message, giving the message "spool directory problem" or "log
6351 directory problem" with a 421 code for SMTP messages.
6353 72. When Exim is about to re-exec itself, it ensures that the file descriptors
6354 0, 1, and 2 exist, because some OS complain for execs without them (see
6355 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
6356 descriptors. However, the code omitted to check that the open succeeded,
6357 causing mysterious errors if for some reason the permissions on /dev/null
6358 got screwed. Now Exim writes a message to the main and panic logs, and
6359 bombs out if it can't open /dev/null.
6361 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
6362 interact so that it is all more flexible. It is supposed to remain
6363 backwards compatible. Also added extra_local_interfaces.
6365 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
6366 to bomb out with an assertion failure - to the client this appears as a
6367 connection drop. This problem occurs in the part of the code that was taken
6368 from the Samba project. Fortunately, the assertion is in a very simple
6369 function, so I have fixed this by reproducing the function inline in the
6370 one place where it is called, and arranging for authentication to fail
6371 instead of killing the process with assert().
6373 75. The SPA client code was not working when the server requested OEM rather
6374 than Unicode encoding.
6376 76. Added code to make require_files with a specific uid setting more usable in
6377 the case where statting the file as root fails - usually a non-root-mounted
6378 NFS file system. When this happens and the failure is EACCES, Exim now
6379 forks a subprocess and does the per-uid checking as the relevant uid.
6381 77. Added process_log_path.
6383 78. If log_file_path was not explicitly set, a setting of check_log_space or
6384 check_log_inodes was ignored.
6386 79. If a space check for the spool or log partitions fails, the incident is now
6387 logged. Of course, in the latter case the data may get lost...
6389 80. Added the %p formatting code to string_format() so that it can be used to
6390 print addresses in debug_print(). Adjusted all the address printing in the
6391 debugging in store.c to use %p rather than %d.
6393 81. There was a concern that a line of code in smtp_in.c could overflow a
6394 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
6395 initially expressed, the concern was not well-founded, because trailing
6396 spaces are removed early. However, if the trailing spaces were followed by
6397 a NULL, they did not get removed, so the overflow was possible. Two fixes
6400 (a) I re-wrote the offending code in a cleaner fashion.
6401 (b) If an incoming SMTP command contains a NULL character, it is rejected
6404 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
6405 runs initgroups(), so that if the Exim user is in any additional groups,
6406 they will be used during message reception.
6412 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob