1 Change log file for Exim from version 4.21
2 -------------------------------------------
7 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
8 and 3.4.4 - once the server is enabled to respond to an OCSP request
9 it does even when not requested, resulting in a stapling non-aware
10 client dropping the TLS connection.
12 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
13 support variable-length bit vectors. No functional change.
15 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
16 The I= interface field on outgoing lines is now after the H= remote
17 host field, same as incoming lines. There is a separate
18 outgoing_interface log selector which allows you to disable the
21 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
22 If not running log_selector +smtp_connection the mainlog would be held
23 open indefinitely after a "too many connections" event, including to a
24 deleted file after a log rotate. Leave the per net connection logging
25 leaving it open for efficiency as that will be quickly detected by the
26 check on the next write.
28 HS/01 Bug 1671: Fix post transport crash.
29 Processing the wait-<transport> messages could crash the delivery
30 process if the message IDs didn't exist for some reason. When
31 using 'split_spool_directory=yes' the construction of the spool
32 file name failed already, exposing the same netto behaviour.
34 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
35 mime_regex ACL conditions.
37 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
38 to DSN fail messages (bounces): remote IP, remote greeting, remote response
39 to HELO, local diagnostic string.
41 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
42 log line to debug. Even when configured with a tls authenticator many
43 client connections are expected to not authenticate in this way, so
44 an authenticate fail is not an error.
46 HS/02 Add the Exim version string to the process info. This way exiwhat
47 gives some more detail about the running daemon.
49 JH/06 Bug 1395: time-limit cacheing of DNS lookups, to the TTL value. This may
50 matter for fast-change records such as DNSBLs.
52 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
53 retry record, even if constant. There may be multiple transports with
54 different interface settings and the retry behaviour needs to be kept
57 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
59 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
61 JH/10 Bug 840: fix log_defer_output option of pipe transport
63 JH/11 Bug 830: use same host for all RCPTS of a message, even under
64 hosts_randomize. This matters a lot when combined with mua_wrapper.
66 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
67 ${quote_pgsql:<string>} operator.
69 JH/13 Bug 1708: avoid misaligned access in cached lookup.
71 JH/14 Change header file name for freeradius-client. Relevant if compiling
72 with Radius support; from the Gentoo tree and checked under Fedora.
74 JH/15 Bug 1712: Introduce $prdr_requested flag variable
76 JH/16 Bug 1714: Permit an empty string as expansion result for transport
77 option transport_filter, meaning no filtering.
79 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
81 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
82 defaults to "*" (all hosts). The variable is now available when not built
83 with TLS, default unset, mainly to enable keeping the testuite sane.
84 If a server certificate is not supplied (via tls_certificate) an error is
85 logged, and clients will find TLS connections fail on startup. Presumably
86 they will retry in-clear.
87 Packagers of Exim are strongly encouraged to create a server certificate
90 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
91 with the $config_file variable.
93 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
94 in transport context, after the attempt, and per-recipient. The latter type
95 is per host attempted. The event data is the error message, and the errno
96 information encodes the lookup type (A vs. MX) used for the (first) host,
97 and the trailing two digits of the smtp 4xx reponse.
99 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
100 to write to mainlog (or rejectlog, paniclog) in the window between file
101 creation and permissions/ownership being changed. Particularly affects
102 installations where exicyclog is run as root, rather than exim user;
103 result is that the running daemon panics and dies.
105 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
107 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
108 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
109 "pri" and "weight". Note that the previous implicit priority given by the
110 list order is no longer honoured.
112 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalisation
115 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
116 by defining SUPPORT_SOCKS.
121 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
124 JH/02 The smtp transport option "multi_domain" is now expanded.
126 JH/03 The smtp transport now requests PRDR by default, if the server offers
129 JH/04 Certificate name checking on server certificates, when exim is a client,
130 is now done by default. The transport option tls_verify_cert_hostnames
131 can be used to disable this per-host. The build option
132 EXPERIMENTAL_CERTNAMES is withdrawn.
134 JH/05 The value of the tls_verify_certificates smtp transport and main options
135 default to the word "system" to access the system default CA bundle.
136 For GnuTLS, only version 3.0.20 or later.
138 JH/06 Verification of the server certificate for a TLS connection is now tried
139 (but not required) by default. The verification status is now logged by
140 default, for both outbound TLS and client-certificate supplying inbound
143 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
146 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
147 Status Notification (bounce) messages are now MIME format per RFC 3464.
148 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
149 under the control of the dsn_advertise_hosts option, and routers may
150 have a dsn_lasthop option.
152 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
153 default, modifiable by a malware= option. The list separator for
154 the options can now be changed in the usual way. Bug 68.
156 JH/10 The smtp_receive_timeout main option is now expanded before use.
158 JH/11 The incoming_interface log option now also enables logging of the
159 local interface on delivery outgoing connections.
161 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
162 if the interface and destination host and port all match.
164 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
167 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
168 Patch from Andrew Lewis.
170 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
171 now supports optional time-restrictions, weighting, and priority
172 modifiers per server. Patch originally by <rommer@active.by>.
174 JH/16 The spamd_address main option now supports a mixed list of local
175 and remote servers. Remote servers can be IPv6 addresses, and
176 specify a port-range.
178 JH/17 Bug 68: The spamd_address main option now supports an optional
179 timeout value per server.
181 JH/18 Bug 1581: Router and transport options headers_add/remove can
182 now have the list separator specified.
184 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
187 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
190 JH/21 Support for the A6 type of dns record is withdrawn.
192 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
193 rather than the verbs used.
195 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
196 from 255 to 1024 chars.
198 JH/24 Verification callouts now attempt to use TLS by default.
200 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
201 are generic router options now. The defaults didn't change.
203 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
204 Original patch from Alexander Shikoff, worked over by JH.
206 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
207 files if ZCAT_COMMAND is not executable.
209 JH/26 Bug 1539: Add timout/retry options on dnsdb lookups.
211 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
213 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
214 Normally benign, it bites when the pair was led to by a CNAME;
215 modern usage is to not canoicalize the domain to a CNAME target
216 (and we were inconsistent anyway for A-only vs AAAA+A).
218 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
220 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
221 when evaluating $sender_host_dnssec.
223 JH/31 Check the HELO verification lookup for DNSSEC, adding new
224 $sender_helo_dnssec variable.
226 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
228 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
230 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
232 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
233 documented as working, but never had. Support all but $spam_report.
235 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
236 added for tls authenticator.
241 TL/01 When running the test suite, the README says that variables such as
242 no_msglog_check are global and can be placed anywhere in a specific
243 test's script, however it was observed that placement needed to be near
244 the beginning for it to behave that way. Changed the runtest perl
245 script to read through the entire script once to detect and set these
246 variables, reset to the beginning of the script, and then run through
247 the script parsing/test process like normal.
249 TL/02 The BSD's have an arc4random API. One of the functions to induce
250 adding randomness was arc4random_stir(), but it has been removed in
251 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
252 function when detected.
254 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
255 cause callback expansion.
257 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
258 syntax errors in an expansion can be treated as a string instead of
259 logging or causing an error, due to the internal use of bool_lax
260 instead of bool when processing it.
262 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
263 server certificates when making smtp deliveries.
265 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
267 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
269 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
271 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
272 Merged patch from Sebastian Wiedenroth.
274 JH/05 Fix results-pipe from transport process. Several recipients, combined
275 with certificate use, exposed issues where response data items split
276 over buffer boundaries were not parsed properly. This eventually
277 resulted in duplicates being sent. This issue only became common enough
278 to notice due to the introduction of conection certificate information,
279 the item size being so much larger. Found and fixed by Wolfgang Breyha.
281 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
282 size buffer was used, resulting in syntax errors when an expansion
285 JH/07 Add support for directories of certificates when compiled with a GnuTLS
286 version 3.3.6 or later.
288 JH/08 Rename the TPDA expermimental facility to Event Actions. The #ifdef
289 is EXPERIMENTAL_EVENT, the main-configuration and transport options
290 both become "event_action", the variables become $event_name, $event_data
291 and $event_defer_errno. There is a new variable $verify_mode, usable in
292 routers, transports and related events. The tls:cert event is now also
293 raised for inbound connections, if the main configuration event_action
296 TL/06 In test suite, disable OCSP for old versions of openssl which contained
297 early OCSP support, but no stapling (appears to be less than 1.0.0).
299 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
300 server certificate names available under the smtp transport option
301 "tls_verify_cert_hostname" now do not permit multi-component wildcard
304 JH/10 Time-related extraction expansions from certificates now use the main
305 option "timezone" setting for output formatting, and are consistent
306 between OpenSSL and GnuTLS compilations. Bug 1541.
308 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
309 encoded parameter in the incoming message. Bug 1558.
311 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
312 include certificate info, eximon was claiming there were spoolfile
315 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
317 JH/14 Log delivery-related information more consistently, using the sequence
318 "H=<name> [<ip>]" wherever possible.
320 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
321 are problematic for Debian distribution, omit them from the release
324 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
326 JH/16 Fix string representation of time values on 64bit time_t anchitectures.
329 JH/17 Fix a null-indirection in certextract expansions when a nondefault
330 output list separator was used.
335 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
336 checkers that were complaining about end of non-void function with no
339 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
340 This was a regression intruduced in 4.83 by another bugfix.
342 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
344 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
345 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
351 TF/01 Correctly close the server side of TLS when forking for delivery.
353 When a message was received over SMTP with TLS, Exim failed to clear up
354 the incoming connection properly after forking off the child process to
355 deliver the message. In some situations the subsequent outgoing
356 delivery connection happened to have the same fd number as the incoming
357 connection previously had. Exim would try to use TLS and fail, logging
358 a "Bad file descriptor" error.
360 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
361 utilities have not been installed.
363 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
364 temporary space as the ACL may create new global variables.
366 TL/01 LDAP support uses per connection or global context settings, depending
367 upon the detected version of the libraries at build time.
369 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
370 to extract and use the src ip:port in logging and expansions as if it
371 were a direct connection from the outside internet. PPv2 support was
372 updated based on HAProxy spec change in May 2014.
374 JH/02 Add ${listextract {number}{list}{success}{fail}}.
376 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
377 Properly escape header and check for NULL return.
379 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
382 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
384 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
385 characters in header names, implemented as a verify condition.
386 Contributed by Michael Fischer v. Mollard.
388 TL/05 Rename SPF condition results err_perm and err_temp to standardized
389 results permerror and temperror. Previous values are deprecated but
390 still accepted. In a future release, err_perm and err_temp will be
391 completely removed, which will be a backward incompatibility if the
392 ACL tests for either of these two old results. Patch contributed by
393 user bes-internal on the mailing list.
395 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
397 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
398 selectors, in both main and reject logs.
400 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
403 JH/07 Add malware type "sock" for talking to simple daemon.
405 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
407 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
408 routers/transports under cutthrough routing.
410 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
411 numbers. Touch up "bool" conditional to keep the same definition.
413 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
415 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
417 JH/12 Expand items in router/transport headers_add or headers_remove lists
418 individually rather than the list as a whole. Bug 1452.
420 Required for reasonable handling of multiple headers_ options when
421 they may be empty; requires that headers_remove items with embedded
422 colons must have them doubled (or the list-separator changed).
424 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
425 view the policy declared in the DMARC record. Currently, $dmarc_status
426 is a combined value of both the record presence and the result of the
429 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
431 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
432 dnslookup router and the smtp transport (applying to the forward
435 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
436 of ldap servers used for a specific lookup. Patch provided by Heiko
439 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
440 New variable $lookup_dnssec_authenticated for observability.
442 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
443 Patch submitted by Lars Timman.
445 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
447 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
448 Requires trusted mode and valid format message id, aborts otherwise.
449 Patch contributed by Heiko Schlichting.
451 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
452 certextract with support for various fields. Bug 1358.
454 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
455 is requested by default, modifiable by smtp transport option
458 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
459 operate on certificate variables to give certificate fingerprints
460 Also new ${sha256:cert_variable}.
462 JH/23 The PRDR feature is moved from being Experimental into the mainline.
464 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
465 Christian Aistleitner.
467 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
469 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
470 file. Patch from Wolfgang Breyha.
472 JH/25 Expand the coverage of the delivery $host and $host_address to
473 client authenticators run in verify callout. Bug 1476.
475 JH/26 Port service names are now accepted for tls_on_connect_ports, to
476 align with daemon_smtp_ports. Bug 72.
478 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
479 support and error reporting did not work properly.
481 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
482 and is readable. Patch from Andrew Colin Kissa.
484 TL/14 Enhance documentation of ${run expansion and how it parses the
485 commandline after expansion, particularly in the case when an
486 unquoted variable expansion results in an empty value.
488 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
490 PP/02 Fix internal collision of T_APL on systems which support RFC3123
491 by renaming away from it. Addresses GH issue 15, reported by
494 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
496 TL/15 SECURITY: prevent double expansion in math comparison functions
497 (can expand unsanitized data). Not remotely exploitable.
504 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
506 PP/02 Make -n do something, by making it not do something.
507 When combined with -bP, the name of an option is not output.
509 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
512 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
513 $sender_host_name and config options to manage this, and basic check
516 PP/05 DSCP support for outbound connections and control modifier for inbound.
518 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
519 (Only plugin which currently uses this is kerberos4, which nobody should
520 be using, but we should make it available and other future plugins might
521 conceivably use it, even though it would break NAT; stuff *should* be
522 using channel bindings instead).
524 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
525 name; added for Sendmail compatibility; requires admin caller.
526 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
527 just ignore it); requires trusted caller.
528 Also parse but ignore: -Ac -Am -X<logfile>
531 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
533 TL/02 Add +smtp_confirmation as a default logging option.
535 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
536 Patch by Magnus Holmgren from 2007-02-20.
538 TL/04 Bugzilla 1281 - Spec typo.
539 Bugzilla 1283 - Spec typo.
540 Bugzilla 1290 - Spec grammar fixes.
542 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
544 TL/06 Add Experimental DMARC support using libopendmarc libraries.
546 TL/07 Fix an out of order global option causing a segfault. Reported to dev
547 mailing list by by Dmitry Isaikin.
549 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
551 JH/02 Support "G" suffix to numbers in ${if comparisons.
553 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
555 NM/01 Bugzilla 1197 - Spec typo
556 Bugzilla 1196 - Spec examples corrections
558 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
560 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
561 gnutls_enable_pkcs11, but renamed to more accurately indicate its
564 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
565 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
567 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
568 "acl {{name}{arg}...}", and optional args on acl condition
571 JH/05 Permit multiple router/transport headers_add/remove lines.
573 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
575 JH/07 Avoid using a waiting database for a single-message-only transport.
576 Performance patch from Paul Fisher. Bugzilla 1262.
578 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
581 JH/09 Add $headers_added variable, with content from use of ACL modifier
582 add_header (but not yet added to the message). Bugzilla 199.
584 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
585 Pulled from Bugzilla 817 by Wolfgang Breyha.
587 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
589 (nb: this is the same fix as in Exim 4.80.1)
591 JH/11 Add A= logging on delivery lines, and a client_set_id option on
594 JH/12 Add optional authenticated_sender logging to A= and a log_selector
597 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
599 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
600 advertise SMTP AUTH mechanism to us, instead of a generic
601 protocol violation error. Also, make Exim more robust to bad
602 data from the Dovecot auth socket.
604 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
606 When a queue runner is handling a message, Exim first routes the
607 recipient addresses, during which it prunes them based on the retry
608 hints database. After that it attempts to deliver the message to
609 any remaining recipients. It then updates the hints database using
612 So if a recipient address works intermittently, it can get repeatedly
613 deferred at routing time. The retry hints record remains fresh so the
614 address never reaches the final cutoff time.
616 This is a fairly common occurrence when a user is bumping up against
617 their storage quota. Exim had some logic in its local delivery code
618 to deal with this. However it did not apply to per-recipient defers
619 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
621 This change adds a proper retry rule check during routing so that the
622 final cutoff time is checked against the message's age. We only do
623 this check if there is an address retry record and there is not a
624 domain retry record; this implies that previous attempts to handle
625 the address had the retry_use_local_parts option turned on. We use
626 this as an approximation for the destination being like a local
627 delivery, as in LMTP.
629 I suspect this new check makes the old local delivery cutoff check
630 redundant, but I have not verified this so I left the code in place.
632 TF/02 Correct gecos expansion when From: is a prefix of the username.
634 Test 0254 submits a message to Exim with the header
638 When I ran the test suite under the user fanf2, Exim expanded
639 the header to contain my full name, whereas it should have added
640 a Resent-Sender: header. It erroneously treats any prefix of the
641 username as equal to the username.
643 This change corrects that bug.
645 GF/01 DCC debug and logging tidyup
646 Error conditions log to paniclog rather than rejectlog.
647 Debug lines prefixed by "DCC: " to remove any ambiguity.
649 TF/03 Avoid unnecessary rebuilds of lookup-related code.
651 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
652 Bug spotted by Jeremy Harris; was flawed since initial commit.
653 Would have resulted in OCSP responses post-SNI triggering an Exim
654 NULL dereference and crash.
656 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
658 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
659 Bug detection, analysis and fix by Samuel Thibault.
660 Bugzilla 1331, Debian bug #698092.
662 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
664 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
665 Server implementation by Todd Lyons, client by JH.
666 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
667 config variable "prdr_enable" controls whether the server
668 advertises the facility. If the client requests PRDR a new
669 acl_data_smtp_prdr ACL is called once for each recipient, after
670 the body content is received and before the acl_smtp_data ACL.
671 The client is controlled by bolth of: a hosts_try_prdr option
672 on the smtp transport, and the server advertisement.
673 Default client logging of deliveries and rejections involving
674 PRDR are flagged with the string "PRDR".
676 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
677 fclose(). Diagnosis by Todd Lyons.
679 PP/17 Update configure.default to handle IPv6 localhost better.
680 Patch by Alain Williams (plus minor tweaks).
683 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
684 This is now consistent with GnuTLS, and is now documented: the
685 previous undocumented portable approach to treating the option as
686 unset was to force an expansion failure. That still works, and
687 an empty string is now equivalent.
689 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
690 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
691 not performing validation itself.
693 PP/20 Added force_command boolean option to pipe transport.
694 Patch from Nick Koston, of cPanel Inc.
696 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
699 TF/04 Added udpsend ACL modifer and hexquote expansion operator
701 PP/21 Fix eximon continuous updating with timestamped log-files.
702 Broken in a format-string cleanup in 4.80, missed when I repaired the
703 other false fix of the same issue.
704 Report and fix from Heiko Schlichting.
707 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
708 Report from Prashanth Katuri.
710 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
711 It's SecureTransport, so affects any MacOS clients which use the
712 system-integrated TLS libraries, including email clients.
714 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
715 using a MIME ACL for non-SMTP local injection.
716 Report and assistance in diagnosis by Warren Baker.
718 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
720 JH/16 Fix comparisons for 64b. Bugzilla 1385.
722 TL/09 Add expansion variable $authenticated_fail_id to keep track of
723 last id that failed so it may be referenced in subsequent ACL's.
725 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
728 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
729 ldap library initialization, allowing self-signed CA's to be
730 used. Also properly sets require_cert option later in code by
731 using NULL (global ldap config) instead of ldap handle (per
732 session). Bug diagnosis and testing by alxgomz.
734 TL/12 Enhanced documentation in the ratelimit.pl script provided in
735 the src/util/ subdirectory.
737 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
738 renamed to Transport Post Delivery Action by Jeremy Harris, as
741 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
742 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
743 redis_servers = needs to be configured which will be used by the redis
744 lookup. Patch from Warren Baker, of The Packet Hub.
746 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
748 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
749 hostname or reverse DNS when processing a host list. Used suggestions
750 from multiple comments on this bug.
752 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
754 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
755 Missed a few lines, added it to make the runtest require no keyboard
758 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
759 contains upper case chars. Make router use caseful_local_part.
761 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
762 support when GnuTLS has been built with p11-kit.
768 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
770 This, or similar/improved, will also be change PP/11 of 4.82.
776 PP/01 Handle short writes when writing local log-files.
777 In practice, only affects FreeBSD (8 onwards).
778 Bugzilla 1053, with thanks to Dmitry Isaikin.
780 NM/01 Bugzilla 949 - Documentation tweak
782 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
785 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
787 PP/02 Implemented gsasl authenticator.
789 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
791 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
792 `pkg-config foo` for cflags/libs.
794 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
795 with rest of GSASL and with heimdal_gssapi.
797 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
798 `pkg-config foo` for cflags/libs for the TLS implementation.
800 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
801 properties get this fed in as external SSF. A number of robustness
802 and debugging improvements to the cyrus_sasl authenticator.
804 PP/08 cyrus_sasl server now expands the server_realm option.
806 PP/09 Bugzilla 1214 - Log authentication information in reject log.
807 Patch by Jeremy Harris.
809 PP/10 Added dbmjz lookup type.
811 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
813 PP/12 MAIL args handles TAB as well as SP, for better interop with
814 non-compliant senders.
815 Analysis and variant patch by Todd Lyons.
817 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
818 Bug report from Lars Müller <lars@samba.org> (via SUSE),
819 Patch from Dirk Mueller <dmueller@suse.com>
821 PP/13 tls_peerdn now print-escaped for spool files.
822 Observed some $tls_peerdn in wild which contained \n, which resulted
823 in spool file corruption.
825 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
826 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
827 or write after TLS renegotiation, which otherwise led to messages
830 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
831 as a tracking header (ie: a signed header comes before the signature).
832 Patch from Wolfgang Breyha.
834 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
835 comma-sep list; embedded commas doubled.
837 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
839 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
841 Report and patch from Dmitry Banschikov.
843 PP/16 Removed "dont_insert_empty_fragments" fron "openssl_options".
844 Removed SSL_clear() after SSL_new() which led to protocol negotiation
845 failures. We appear to now support TLS1.1+ with Exim.
847 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
848 lets Exim select keys and certificates based upon TLS SNI from client.
849 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
850 before an outbound SMTP session. New log_selector, +tls_sni.
852 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
853 NULL dereference. Report and patch from Alun Jones.
855 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
856 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
857 Not seeing resolver debug output on NetBSD, but suspect this is a
858 resolver implementation change.
860 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
861 Left warnings. Added "eximon gdb" invocation mode.
863 PP/21 Defaulting "accept_8bitmime" to true, not false.
865 PP/22 Added -bw for inetd wait mode support.
867 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
868 locate the relevant includes and libraries. Made this the default.
870 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
871 Bugzilla 1246, report and most of solution from Tomasz Kusy.
873 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
874 This may cause build issues on older platforms.
876 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
877 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
878 gnutls_require_mac & gnutls_require_protocols (no longer supported).
879 Added SNI support via GnuTLS too.
880 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
882 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
884 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
885 Applied second patch from Janne, implementing suggestion to default
886 multiple-strings-in-record handling to match SPF spec.
888 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
890 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
891 read-only, out of scope).
892 Patch from Wolfgang Breyha, report from Stuart Northfield.
894 PP/29 Fix three issues highlighted by clang analyser static analysis.
895 Only crash-plausible issue would require the Cambridge-specific
896 iplookup router and a misconfiguration.
897 Report from Marcin Mirosław.
899 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
901 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
902 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
903 As part of this, removing so much warning spew let me fix some minor
904 real issues in debug logging.
906 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
907 assignment on my part. Fixed.
909 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
910 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
911 Janne Snabb (who went above and beyond: thank you).
913 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
914 string otherwise requires a connection and a bunch more work and it's
915 relatively easy to get wrong. Should also expose TLS library linkage
918 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
919 64-bit ${eval} (JH/03).
921 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
922 GNU libc to support some of the 64-bit stuff, should not lead to
923 conflicts. Defined before os.h is pulled in, so if a given platform
924 needs to override this, it can.
926 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
927 protection layer was required, which is not implemented.
928 Bugzilla 1254, patch from Wolfgang Breyha.
930 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
931 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
932 tls_dhparam take prime identifiers. Also unbreak combination of
933 OpenSSL+DH_params+TLSSNI.
935 PP/39 Disable SSLv2 by default in OpenSSL support.
941 PP/01 Solaris build fix for Oracle's LDAP libraries.
942 Bugzilla 1109, patch from Stephen Usher.
944 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
946 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
949 TF/02 Fix a couple more cases where we did not log the error message
950 when unlink() failed. See also change 4.74-TF/03.
952 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
953 lock up or crash if it happened to be inside a call to libc when it
954 got a SIGUSR1 from exiwhat.
956 The SIGUSR1 handler appends the current process status to the process
957 log which is later printed by exiwhat. It used to use the general
958 purpose logging code to do this, but several functions it calls are
959 not safe for signals.
961 The new output code in the SIGUSR1 handler is specific to the process
962 log, and simple enough that it's easy to inspect for signal safety.
963 Removing some special cases also simplifies the general logging code.
964 Removing the spurious timestamps from the process log simplifies
967 TF/04 Improved ratelimit ACL condition.
969 The /noupdate option has been deprecated in favour of /readonly which
970 has clearer semantics. The /leaky, /strict, and /readonly update modes
971 are mutually exclusive. The update mode is no longer included in the
972 database key; it just determines when the database is updated. (This
973 means that when you upgrde Exim will forget old rate measurements.)
975 Exim now checks that the per_* options are used with an update mode that
976 makes sense for the current ACL. For example, when Exim is processing a
977 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
978 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
979 must specify per_mail/readonly. If you omit the update mode it defaults to
980 /leaky where that makes sense (as before) or /readonly where required.
982 The /noupdate option is now undocumented but still supported for
983 backwards compatibility. It is equivalent to /readonly except that in
984 ACLs where /readonly is required you may specify /leaky/noupdate or
985 /strict/noupdate which are treated the same as /readonly.
987 A useful new feature is the /count= option. This is a generalization
988 of the per_byte option, so that you can measure the throughput of other
989 aggregate values. For example, the per_byte option is now equivalent
990 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
992 The per_rcpt option has been generalized using the /count= mechanism
993 (though it's more complicated than the per_byte equivalence). When it is
994 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
995 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
996 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
997 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
998 Note that using per_rcpt with a non-readonly update mode in more than
999 one ACL will cause the recipients to be double-counted. (The per_mail
1000 and per_byte options don't have this problem.)
1002 The handling of very low rates has changed slightly. If the computed rate
1003 is less than the event's count (usually one) then this event is the first
1004 after a long gap. In this case the rate is set to the same as this event's
1005 count, so that the first message of a spam run is counted properly.
1007 The major new feature is a mechanism for counting the rate of unique
1008 events. The new per_addr option counts the number of different
1009 recipients that someone has sent messages to in the last time period. It
1010 behaves like per_rcpt if all the recipient addresses are different, but
1011 duplicate recipient addresses do not increase the measured rate. Like
1012 the /count= option this is a general mechanism, so the per_addr option
1013 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
1014 example, measure the rate that a client uses different sender addresses
1015 with the options per_mail/unique=$sender_address. There are further
1016 details in the main documentation.
1018 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
1020 TF/06 Removed a few PCRE remnants.
1022 TF/07 Automatically extract Exim's version number from tags in the git
1023 repository when doing development or release builds.
1025 PP/02 Raise smtp_cmd_buffer_size to 16kB.
1026 Bugzilla 879. Patch from Paul Fisher.
1028 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
1029 Heavily based on revision 40f9a89a from Simon Arlott's tree.
1032 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
1034 PP/05 Variable $av_failed, true if the AV scanner deferred.
1035 Bugzilla 1078. Patch from John Horne.
1037 PP/06 Stop make process more reliably on build failure.
1038 Bugzilla 1087. Patch from Heiko Schlittermann.
1040 PP/07 Make maildir_use_size_file an _expandable_ boolean.
1041 Bugzilla 1089. Patch from Heiko Schlittermann.
1043 PP/08 Handle ${run} returning more data than OS pipe buffer size.
1044 Bugzilla 1131. Patch from Holger Weiß.
1046 PP/09 Handle IPv6 addresses with SPF.
1047 Bugzilla 860. Patch from Wolfgang Breyha.
1049 PP/10 GnuTLS: support TLS 1.2 & 1.1.
1051 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
1054 PP/11 match_* no longer expand right-hand-side by default.
1055 New compile-time build option, EXPAND_LISTMATCH_RHS.
1056 New expansion conditions, "inlist", "inlisti".
1058 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
1060 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
1062 PP/14 fix log_write() format string regression from TF/03.
1063 Bugzilla 1152. Patch from Dmitry Isaikin.
1069 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
1071 PP/02 Harmonised TLS library version reporting; only show if debugging.
1072 Layout now matches that introduced for other libraries in 4.74 PP/03.
1074 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
1076 PP/04 New "dns_use_edns0" global option.
1078 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
1081 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
1082 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
1084 TK/01 Updated PolarSSL code to 0.14.2.
1085 Bugzilla 1097. Patch from Andreas Metzler.
1087 PP/07 Catch divide-by-zero in ${eval:...}.
1088 Fixes bugzilla 1102.
1090 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
1093 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
1094 format-string attack -- SECURITY: remote arbitrary code execution.
1096 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
1097 time unintentionally subject to list matching rules, letting the header
1098 cause arbitrary Exim lookups (of items which can occur in lists, *not*
1099 arbitrary string expansion). This allowed for information disclosure.
1101 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
1102 INT_MIN/-1 -- value coerced to INT_MAX.
1108 NM/01 Workround for PCRE version dependency in version reporting
1111 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
1112 This fixes portability to compilers other than gcc, notably
1113 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
1115 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
1116 makefiles for portability to HP-UX and POSIX correctness.
1118 PP/01 Permit LOOKUP_foo enabling on the make command-line.
1119 Also via indented variable definition in the Makefile.
1120 (Debugging by Oliver Heesakkers).
1122 PP/02 Restore caching of spamd results with expanded spamd_address.
1123 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
1125 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
1126 Improves build reliability. Fix from: Frank Elsner
1128 NM/02 Fix wide character breakage in the rfc2047 coding
1129 Fixes bug 1064. Patch from Andrey N. Oktyabrski
1131 NM/03 Allow underscore in dnslist lookups
1132 Fixes bug 1026. Patch from Graeme Fowler
1134 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
1135 Code patches from Adam Ciarcinski of NetBSD.
1137 NM/04 Fixed exiqgrep to cope with mailq missing size issue
1140 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
1141 is logged, to avoid truncation. Patch from John Horne.
1143 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
1144 Patch from Jakob Hirsch.
1146 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
1147 SQL string expansion failure details.
1148 Patch from Andrey Oktyabrski.
1150 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
1151 Patch from Simon Arlott.
1153 PP/09 New lookups functionality failed to compile on old gcc which rejects
1154 extern declarations in function scope.
1155 Patch from Oliver Fleischmann
1157 PP/10 Use sig_atomic_t for flags set from signal handlers.
1158 Check getgroups() return and improve debugging.
1159 Fixed developed for diagnosis in bug 927 (which turned out to be
1162 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
1163 Patch from Mark Zealey.
1165 PP/12 Bugzilla 1056: Improved spamd server selection.
1166 Patch from Mark Zealey.
1168 PP/13 Bugzilla 1086: Deal with maildir quota file races.
1169 Based on patch from Heiko Schlittermann.
1171 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
1172 Patch from Uwe Doering, sign-off by Michael Haardt.
1174 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
1175 variable declaration deep within a block. Bug and patch from
1178 PP/15 lookups-Makefile IRIX compatibilty coercion.
1180 PP/16 Make DISABLE_DKIM build knob functional.
1182 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
1183 Patch by Simon Arlott
1185 TF/03 Fix valgrind.h portability to C89 compilers that do not support
1186 variable argument macros. Our copy now differs from upstream.
1192 TF/01 Failure to get a lock on a hints database can have serious
1193 consequences so log it to the panic log.
1195 TF/02 Log LMTP confirmation messages in the same way as SMTP,
1196 controlled using the smtp_confirmation log selector.
1198 TF/03 Include the error message when we fail to unlink a spool file.
1200 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
1201 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
1202 for maintaining out-of-tree patches for some time.
1204 PP/01 Bugzilla 139: Documentation and portability issues.
1205 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
1206 Handle per-OS dynamic-module compilation flags.
1208 PP/02 Let /dev/null have normal permissions.
1209 The 4.73 fixes were a little too stringent and complained about the
1210 permissions on /dev/null. Exempt it from some checks.
1211 Reported by Andreas M. Kirchwitz.
1213 PP/03 Report version information for many libraries, including
1214 Exim version information for dynamically loaded libraries. Created
1215 version.h, now support a version extension string for distributors
1216 who patch heavily. Dynamic module ABI change.
1218 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
1219 privilege escalation vulnerability whereby the Exim run-time user
1220 can cause root to append content of the attacker's choosing to
1223 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
1226 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
1227 If dropping privileges for untrusted macros, we disabled normal logging
1228 on the basis that it would fail; for the Exim run-time user, this is not
1229 the case, and it resulted in successful deliveries going unlogged.
1230 Fixed. Reported by Andreas Metzler.
1236 PP/01 Date: & Message-Id: revert to normally being appended to a message,
1237 only prepend for the Resent-* case. Fixes regression introduced in
1238 Exim 4.70 by NM/22 for Bugzilla 607.
1240 PP/02 Include check_rfc2047_length in configure.default because we're seeing
1241 increasing numbers of administrators be bitten by this.
1243 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
1245 PP/03 Bugzilla 994: added openssl_options main configuration option.
1247 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
1249 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
1251 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
1253 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
1254 without a peer certificate, leading to a segfault because of an
1255 assumption that peers always have certificates. Be a little more
1256 paranoid. Problem reported by Martin Tscholak.
1258 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
1259 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
1260 NB: ClamAV planning to remove STREAM in "middle of 2010".
1261 CL also introduces -bmalware, various -d+acl logging additions and
1262 more caution in buffer sizes.
1264 PP/09 Implemented reverse_ip expansion operator.
1266 PP/10 Bugzilla 937: provide a "debug" ACL control.
1268 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
1270 PP/12 Bugzilla 973: Implement --version.
1272 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
1274 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
1276 PP/15 Bugzilla 816: support multiple condition rules on Routers.
1278 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
1279 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
1280 ignore trailing whitespace.
1282 JJ/02 prevent non-panic DKIM error from being sent to paniclog
1284 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
1287 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
1288 Notification from Dr Andrew Aitchison.
1290 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
1291 ExtendedDetectionInfo response format.
1292 Notification from John Horne.
1294 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
1297 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
1298 XSL and documented dependency on system catalogs, with examples of how
1301 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
1304 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
1305 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
1306 configuration file which is writeable by the Exim user or group.
1308 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
1309 of configuration files to cover files specified with the -C option if
1310 they are going to be used with root privileges, not just the default
1313 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
1314 option (effectively making it always true).
1316 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
1317 files to be used while preserving root privileges.
1319 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
1320 that rogue child processes cannot use them.
1322 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
1323 run-time user, instead of root.
1325 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
1326 Exim run-time user without dropping privileges.
1328 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
1329 result string, instead of calling string_vformat() twice with the same
1332 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
1333 for other users. Others should always drop root privileges if they use
1334 -C on the command line, even for a whitelisted configure file.
1336 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
1338 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
1344 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
1345 $data_path, and $header_path variables; fixed documentation bugs and
1348 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
1349 exipick to access non-standard spools, including the "frozen" queue
1352 NM/01 Bugzilla 965: Support mysql stored procedures.
1353 Patch from Alain Williams
1355 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
1357 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
1358 Patch from Andreas Metzler
1360 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
1361 Patch from Kirill Miazine
1363 NM/05 Bugzilla 671: Added umask to procmail example.
1365 JJ/03 installed exipick 20100323.0, fixing doc bug
1367 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
1368 directory. Notification and patch from Dan Rosenberg.
1370 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
1372 TK/02 Improve log output when DKIM signing operation fails.
1374 MH/01 Treat the transport option dkim_domain as a colon separated
1375 list, not as a single string, and sign the message with each element,
1376 omitting multiple occurences of the same signer.
1378 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
1379 Bugzilla 985, 986. Patch by Simon Arlott
1381 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
1382 Patch by Simon Arlott
1384 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
1385 MBX locking. Notification from Dan Rosenberg.
1391 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
1393 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
1395 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
1397 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
1399 NM/04 Bugzilla 915: Flag broken perl installation during build.
1405 TK/01 Added patch by Johannes Berg that expands the main option
1406 "spamd_address" if it starts with a dollar sign.
1408 TK/02 Write list of recipients to X-Envelope-Sender header when building
1409 the mbox-format spool file for content scanning (suggested by Jakob
1412 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
1413 (http://www.dcc-servers.net/) support via dccifd. Activated by
1414 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
1416 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
1417 by Mark Daniel Reidel <mr@df.eu>.
1419 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
1420 When building exim an external PCRE library is now needed -
1421 PCRE is a system library on the majority of modern systems.
1422 See entry on PCRE_LIBS in EDITME file.
1424 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
1425 conversation. Added nologin parameter to request.
1426 Patch contributed by Kirill Miazine.
1428 TF/01 Do not log submission mode rewrites if they do not change the address.
1430 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
1432 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
1433 log files in place. Contributed by Roberto Lima.
1435 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
1437 TF/03 Bugzilla 615: When checking the local_parts router precondition
1438 after a local_part_suffix or local_part_prefix option, Exim now
1439 does not use the address's named list lookup cache, since this
1440 contains cached lookups for the whole local part.
1442 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
1443 Robert Millan. Documentation is in experimental-spec.txt.
1445 TF/04 Bugzilla 668: Fix parallel build (make -j).
1447 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
1449 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
1450 Patch provided by Jan Srzednicki.
1452 TF/05 Leading white space used to be stripped from $spam_report which
1453 wrecked the formatting. Now it is preserved.
1455 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
1456 that they are available at delivery time.
1458 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
1460 TF/08 TLS error reporting now respects the incoming_interface and
1461 incoming_port log selectors.
1463 TF/09 Produce a more useful error message if an SMTP transport's hosts
1464 setting expands to an empty string.
1466 NM/06 Bugzilla 744: EXPN did not work under TLS.
1467 Patch provided by Phil Pennock.
1469 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
1470 Patch provided by Richard Godbee.
1472 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
1473 acl_smtp_notquit, added index entry.
1475 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
1476 Patch provided by Eugene Bujak.
1478 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
1479 accept(). Patch provided by Maxim Dounin.
1481 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
1482 Patch provided by Phil Pennock.
1484 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
1486 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
1487 Patch provided by Brad "anomie" Jorsch.
1489 NM/14 Bugzilla 622: Added timeout setting to transport filter.
1490 Patch provided by Dean Brooks.
1492 TK/05 Add native DKIM support (does not depend on external libraries).
1494 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
1495 Patch provided by Graeme Fowler.
1497 NM/16 Bugzilla 851: Documentation example syntax fix.
1499 NM/17 Changed NOTICE file to remove references to embedded PCRE.
1501 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
1504 NM/19 Bugzilla 745: TLS version reporting.
1505 Patch provided by Phil Pennock.
1507 NM/20 Bugzilla 167: bool: condition support.
1508 Patch provided by Phil Pennock.
1510 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
1511 clients. Patch provided by Phil Pennock.
1513 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
1514 Patch provided by Brad "anomie" Jorsch.
1516 NM/23 Bugzilla 687: Fix misparses in eximstats.
1517 Patch provided by Heiko Schlittermann.
1519 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
1520 Patch provided by Heiko Schlittermann.
1522 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
1523 plus update to original patch.
1525 NM/26 Bugzilla 799: Documentation correction for ratelimit.
1527 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
1528 Patch provided by David Brownlee.
1530 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
1532 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
1534 NM/30 Bugzilla 888: TLS documentation bugfixes.
1536 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
1538 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
1539 Unlike the original bugzilla I have changed all shell scripts in src tree.
1541 NM/33 Bugzilla 898: Transport filter timeout fix.
1542 Patch by Todd Rinaldo.
1544 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
1545 Patch by Serge Demonchaux.
1547 NM/35 Bugzilla 39: Base64 decode bug fixes.
1548 Patch by Jakob Hirsch.
1550 NM/36 Bugzilla 909: Correct connect() call in dcc code.
1552 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
1554 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
1556 NM/39 Bugzilla 911: Fixed MakeLinks build script.
1562 TK/01 Add preliminary DKIM support. Currently requires a forked version of
1563 ALT-N's libdkim that I have put here:
1564 http://duncanthrax.net/exim-experimental/
1566 Note to Michael Haardt: I had to rename some vars in sieve.c. They
1567 were called 'true' and it seems that C99 defines that as a reserved
1568 keyword to be used with 'bool' variable types. That means you could
1569 not include C99-style headers which use bools without triggering
1570 build errors in sieve.c.
1572 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
1573 as mailq or other aliases. Changed the --help handling significantly
1574 to do whats expected. exim_usage() emits usage/help information.
1576 SC/01 Added the -bylocaldomain option to eximstats.
1578 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
1580 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
1582 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
1588 PH/01 Another patch from the Sieve maintainer.
1590 PH/02 When an IPv6 address is converted to a string for single-key lookup
1591 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
1592 dots are used instead of colons so that keys in lsearch files need not
1593 contain colons. This was done some time before quoting was made available
1594 in lsearch files. However, iplsearch files do require colons in IPv6 keys
1595 (notated using the quote facility) so as to distinguish them from IPv4
1596 keys. This meant that lookups for IP addresses in host lists did not work
1597 for iplsearch lookups.
1599 This has been fixed by arranging for IPv6 addresses to be expressed with
1600 colons if the lookup type is iplsearch. This is not incompatible, because
1601 previously such lookups could never work.
1603 The situation is now rather anomolous, since one *can* have colons in
1604 ordinary lsearch keys. However, making the change in all cases is
1605 incompatible and would probably break a number of configurations.
1607 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
1610 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
1611 conversion specification without a maximum field width, thereby enabling
1612 a rogue spamd server to cause a buffer overflow. While nobody in their
1613 right mind would setup Exim to query an untrusted spamd server, an
1614 attacker that gains access to a server running spamd could potentially
1615 exploit this vulnerability to run arbitrary code as the Exim user.
1617 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
1618 $primary_hostname instead of what libspf2 thinks the hosts name is.
1620 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
1621 a directory entry by the name of the lookup key. Previously, if a
1622 symlink pointed to a non-existing file or a file in a directory that
1623 Exim lacked permissions to read, a lookup for a key matching that
1624 symlink would fail. Now it is enough that a matching directory entry
1625 exists, symlink or not. (Bugzilla 503.)
1627 PH/03 The body_linecount and body_zerocount variables are now exported in the
1630 PH/04 Added the $dnslist_matched variable.
1632 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
1633 This means they are set thereafter only if the connection becomes
1636 PH/06 Added the client_condition to authenticators so that some can be skipped
1637 by clients under certain conditions.
1639 PH/07 The error message for a badly-placed control=no_multiline_responses left
1640 "_responses" off the end of the name.
1642 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
1644 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
1645 (without spaces) instead of just copying the configuration text.
1647 PH/10 Added the /noupdate option to the ratelimit ACL condition.
1649 PH/11 Added $max_received_linelength.
1651 PH/12 Added +ignore_defer and +include_defer to host lists.
1653 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
1654 way in which PCRE > 7.0 is built.
1656 PH/14 Implemented queue_only_load_latch.
1658 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
1659 MAIL command. The effect was to mangle the value on 64-bit systems.
1661 PH/16 Another patch from the Sieve maintainer.
1663 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
1665 PH/18 If a system quota error occurred while trying to create the file for
1666 a maildir delivery, the message "Mailbox is full" was not appended to the
1667 bounce if the delivery eventually timed out. Change 4.67/27 below applied
1668 only to a quota excession during the actual writing of the file.
1670 PH/19 It seems that peer DN values may contain newlines (and other non-printing
1671 characters?) which causes problems in log lines. The DN values are now
1672 passed through string_printing() before being added to log lines.
1674 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
1675 and InterBase are left for another time.)
1677 PH/21 Added message_body_newlines option.
1679 PH/22 Guard against possible overflow in moan_check_errorcopy().
1681 PH/23 POSIX allows open() to be a macro; guard against that.
1683 PH/24 If the recipient of an error message contained an @ in the local part
1684 (suitably quoted, of course), incorrect values were put in $domain and
1685 $local_part during the evaluation of errors_copy.
1691 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
1692 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
1695 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
1696 issue a MAIL command.
1698 PH/02 In an ACL statement such as
1700 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
1702 if a client was not listed at all, or was listed with a value other than
1703 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
1704 the condition was not true (as it should be), so access was not denied.
1705 The bug was that the ! inversion was incorrectly passed on to the second
1706 item. This has been fixed.
1708 PH/03 Added additional dnslists conditions == and =& which are different from
1709 = and & when the dns lookup returns more than one IP address.
1711 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
1712 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
1714 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
1715 FSYNC, which compiles an option called disable_fsync that allows for
1716 bypassing fsync(). The documentation is heavily laced with warnings.
1718 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
1720 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
1721 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
1722 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
1723 including adding "make clean"; (3) Added -fPIC when compiling the test
1724 dynamically loaded module, to get rid of a warning.
1726 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
1727 message fails, move_frozen_messages = true and ignore_bounce_errors_after
1728 = 0s. The bug is otherwise harmless.
1730 PH/07 There was a bug in the dovecot authenticator such that the value of
1731 $auth1 could be overwritten, and so not correctly preserved, after a
1732 successful authentication. This usually meant that the value preserved by
1733 the server_setid option was incorrect.
1735 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
1737 PH/09 Installed PCRE release 7.0.
1739 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
1740 run for batched SMTP input. It is now run at the start of every message
1741 in the batch. While fixing this I discovered that the process information
1742 (output by running exiwhat) was not always getting set for -bs and -bS
1743 input. This is fixed, and it now also says "batched" for BSMTP.
1745 PH/11 Added control=no_pipelining.
1747 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
1748 patch, slightly modified), and move the expansion of helo_data till after
1749 the connection is made in the smtp transport (so it can use these
1752 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
1754 PH/14 Added log_selector = +pid.
1756 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
1758 PH/16 Add ${if forany and ${if forall.
1760 PH/17 Added dsn_from option to vary the From: line in DSNs.
1762 PH/18 Flush SMTP output before performing a callout, unless control =
1763 no_callout_flush is set.
1765 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
1766 was true (the default) a successful delivery failed to delete the retry
1767 item, thus causing premature timeout of the address. The bug is now
1770 PH/20 Added hosts_avoid_pipelining to the smtp transport.
1772 PH/21 Long custom messages for fakedefer and fakereject are now split up
1773 into multiline reponses in the same way that messages for "deny" and
1774 other ACL rejections are.
1776 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
1777 with slight modification.
1779 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
1780 draft, changing the syntax and factoring some duplicate code".
1782 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
1783 for deliveries of the second and subsequent messages over the same SMTP
1786 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
1787 ${reduce, with only minor "tidies".
1789 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
1791 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
1792 expansion side effects.
1794 PH/27 When a message times out after an over-quota error from an Exim-imposed
1795 quota, the bounce message says "mailbox is full". This message was not
1796 being given when it was a system quota that was exceeded. It now should
1799 MH/03 Made $recipients available in local_scan(). local_scan() already has
1800 better access to the recipient list through recipients_list[], but
1801 $recipients can be useful in postmaster-provided expansion strings.
1803 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
1804 in the case of a MAIL command with additional options following the
1805 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
1806 were accidentally chopped off.
1808 PH/29 SMTP synchronization checks are implemented when a command is read -
1809 there is a check that no more input is waiting when there shouldn't be
1810 any. However, for some commands, a delay in an ACL can mean that it is
1811 some time before the response is written. In this time, more input might
1812 arrive, invalidly. So now there are extra checks after an ACL has run for
1813 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
1814 pipelining has not been advertised.
1816 PH/30 MH's patch to allow iscntrl() characters to be list separators.
1818 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
1819 returned in the SMTP response when smtp_return_error_details was false.
1820 This has been fixed.
1822 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
1823 instead of the C I/O that was originally supplied, because problems were
1824 reported on Solaris.
1826 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
1827 Exim which did not show up earlier: it was assuming that a call to
1828 SSL_CTX_set_info_callback() might give an error value. In fact, there is
1829 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
1830 was a macro that became an assignment, so it seemed to work. This has
1831 changed to a proper function call with a void return, hence the compile
1832 error. Exim's code has been fixed.
1834 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
1837 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
1839 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
1841 PH/37 If a message is not accepted after it has had an id assigned (e.g.
1842 because it turns out to be too big or there is a timeout) there is no
1843 "Completed" line in the log. When some messages of this type were
1844 selected by exigrep, they were listed as "not completed". Others were
1845 picked up by some special patterns. I have improved the selection
1846 criteria to be more general.
1848 PH/38 The host_find_failed option in the manualroute router can now be set
1849 to "ignore", to completely ignore a host whose IP address cannot be
1850 found. If all hosts are ignored, the behaviour is controlled by the new
1851 host_all_ignored option.
1853 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
1854 homing or because of multiple MX records with /mx) generated more than
1855 one IP address, and the following item turned out to be the local host,
1856 all the secondary addresses of the first item were incorrectly removed
1857 from the list, along with the local host and any following hosts (which
1858 is what is supposed to happen).
1860 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
1861 whoever called Exim into the -H file. In the case of the daemon it was
1862 behaving confusingly. When first started, it used values for whoever
1863 started the daemon, but after a SIGHUP it used the Exim user (because it
1864 calls itself on a restart). I have changed the code so that it now always
1867 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
1868 message are rejected with the same error (e.g. no authentication or bad
1869 sender address), and a DATA command is nevertheless sent (as can happen
1870 with PIPELINING or a stupid MUA), the error message that was given to the
1871 RCPT commands is included in the rejection of the DATA command. This is
1872 intended to be helpful for MUAs that show only the final error to their
1875 PH/42 Another patch from the Sieve maintainer.
1877 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
1878 Eximstats - Fixed some broken HTML links and added missing column headers
1880 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
1881 columns for Rejects, Temp Rejects, Ham, and Spam rows.
1883 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
1885 PH/43 Yet another patch from the Sieve maintainer.
1887 PH/44 I found a way to check for a TCP/IP connection going away before sending
1888 the response to the final '.' that terminates a message, but only in the
1889 case where the client has not sent further data following the '.'
1890 (unfortunately, this is allowed). However, in many cases there won't be
1891 any further data because there won't be any more messages to send. A call
1892 to select() can be used: if it shows that the input is "ready", there is
1893 either input waiting, or the socket has been closed. An attempt to read
1894 the next input character can distinguish the two cases. Previously, Exim
1895 would have sent an OK response which the client would never have see.
1896 This could lead to message repetition. This fix should cure that, at
1897 least in a lot of common cases.
1899 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
1900 advertised in response to EHLO.
1906 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
1907 fixed by 4.65/MH/01 (is this a record?) are fixed:
1909 (i) An empty string was always treated as zero by the numeric comparison
1910 operators. This behaviour has been restored.
1912 (ii) It is documented that the numeric comparison operators always treat
1913 their arguments as decimal numbers. This was broken in that numbers
1914 starting with 0 were being interpreted as octal.
1916 While fixing these problems I realized that there was another issue that
1917 hadn't been noticed. Values of message_size_limit (both the global option
1918 and the transport option) were treated as octal if they started with 0.
1919 The documentation was vague. These values are now always treated as
1920 decimal, and I will make that clear in the documentation.
1926 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
1927 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
1930 MH/01 Don't check that the operands of numeric comparison operators are
1931 integers when their expansion is in "skipping" mode (fixes bug
1932 introduced by 4.64-PH/07).
1934 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
1935 child addresses, Exim now panics and dies. Previously, because the count
1936 is held in a short int, deliveries were likely to be lost. As such a
1937 large number of recipients for a single message is ridiculous
1938 (performance will be very, very poor), I have chosen to impose a limit
1939 rather than extend the field.
1945 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
1946 leftover -K file (the existence of which was triggered by #402).
1947 While we were at it, introduced process PID as part of the -K
1948 filename. This should rule out race conditions when creating
1951 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
1952 processing considerably. Previous code took too long for large mails,
1953 triggering a timeout which in turn triggers #401.
1955 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
1956 in the DK code in transports.c. sendfile() is not really portable,
1957 hence the _LINUX specificness.
1959 TF/01 In the add_headers option to the mail command in an Exim filter,
1960 there was a bug that Exim would claim a syntax error in any
1961 header after the first one which had an odd number of characters
1964 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
1965 callout verification, Exim cached a "reject" for the entire domain. This
1966 is correct for most verifications, but it is not correct for a recipient
1967 verification with use_sender or use_postmaster set, because in that case
1968 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
1969 case of MAIL FROM:<> rejection from other early rejections (e.g.
1970 rejection of HELO). When verifying a recipient using a non-null MAIL
1971 address, the cache is ignored if it shows MAIL FROM:<> rejection.
1972 Whatever the result of the callout, the value of the domain cache is
1973 left unchanged (for any other kind of callout, getting as far as trying
1974 RCPT means that the domain itself is ok).
1976 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
1979 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
1980 manifest itself as EPIPE rather than ECONNECT. When tidying away a
1981 session, the daemon ignores ECONNECT errors and logs others; it now
1982 ignores EPIPE as well.
1984 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
1985 (quoted-printable decoding).
1987 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
1988 later the small subsequent patch to fix an introduced bug.
1990 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
1992 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
1994 PH/08 An error is now given if message_size_limit is specified negative.
1996 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
1997 to be given (somewhat) arbitrary names.
1999 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
2002 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
2003 miscellaneous code fixes
2005 PH/10 Added the log_reject_target ACL modifier to specify where to log
2008 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
2009 hostname. This is wrong, because it relates to the incoming message (and
2010 probably the interface on which it is arriving) and not to the outgoing
2011 callout (which could be using a different interface). This has been
2012 changed to use the value of the helo_data option from the smtp transport
2013 instead - this is what is used when a message is actually being sent. If
2014 there is no remote transport (possible with a router that sets up host
2015 addresses), $smtp_active_hostname is used.
2017 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
2018 tweaks were necessary in order to get it to work (see also 21 below):
2019 (a) The code assumed that strncpy() returns a negative number on buffer
2020 overflow, which isn't the case. Replaced with Exim's string_format()
2022 (b) There were several signed/unsigned issues. I just did the minimum
2023 hacking in of casts. There is scope for a larger refactoring.
2024 (c) The code used strcasecmp() which is not a standard C function.
2025 Replaced with Exim's strcmpic() function.
2026 (d) The code set only $1; it now sets $auth1 as well.
2027 (e) A simple test gave the error "authentication client didn't specify
2028 service in request". It would seem that Dovecot has changed its
2029 interface. Fortunately there's a specification; I followed it and
2030 changed what the client sends and it appears to be working now.
2032 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
2035 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
2036 address A is aliased to B and C, where B exists and C does not. Without
2037 -v the output is "A verified" because verification stops after a
2038 successful redirection if more than one address is generated. However,
2039 with -v the child addresses are also verified. Exim was outputting "A
2040 failed to verify" and then showing the successful verification for C,
2041 with its parentage. It now outputs "B failed to verify", showing B's
2042 parentage before showing the successful verification of C.
2044 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
2045 look up a TXT record in a specific list after matching in a combined
2048 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
2049 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
2050 they consult the DNS. I had assumed they would set it the way they
2051 wanted; and indeed my experiments on Linux seem to show that in some
2052 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
2053 To be on the safe side, however, I have now made the interface to
2054 host_find_byname() similar to host_find_bydns(), with an argument
2055 containing the DNS resolver options. The host_find_byname() function now
2056 sets these options at its start, just as host_find_bydns() does. The smtp
2057 transport options dns_qualify_single and dns_search_parents are passed to
2058 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
2059 of host_find_byname() use the default settings of RES_DEFNAMES
2060 (qualify_single) but not RES_DNSRCH (search_parents).
2062 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
2063 spool_read_header() do less string testing, by means of a preliminary
2064 switch on the second character of optional "-foo" lines. (This is
2065 overdue, caused by the large number of possibilities that now exist.
2066 Originally there were few.) While I was there, I also converted the
2067 str(n)cmp tests so they don't re-test the leading "-" and the first
2068 character, in the hope this might squeeze out yet more improvement.
2070 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
2071 flag allowing group syntax was set by the header_syntax check but not
2072 turned off, possible causing trouble later; (2) The flag was not being
2073 set at all for the header_verify test, causing "group"-style headers to
2074 be rejected. I have now set it in this case, and also caused header_
2075 verify to ignore an empty address taken from a group. While doing this, I
2076 came across some other cases where the code for allowing group syntax
2077 while scanning a header line wasn't quite right (mostly, not resetting
2078 the flag correctly in the right place). These bugs could have caused
2079 trouble for malformed header lines. I hope it is now all correct.
2081 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
2082 with the "reply" argument non-NULL. The code, however (which originally
2083 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
2084 but it didn't always do it. This confused somebody who was copying the
2085 code for some other use. I have removed all the tests.
2087 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
2088 feature that was used to support insecure browsers during the U.S. crypto
2089 embargo. It requires special client support, and Exim is probably the
2090 only MTA that supported it -- and would never use it because real RSA is
2091 always available. This code has been removed, because it had the bad
2092 effect of slowing Exim down by computing (never used) parameters for the
2093 RSA_EXPORT functionality.
2095 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
2096 authenticator to fail if there's a tab character in the incoming data
2097 (there should never be unless someone is messing about, as it's supposed
2098 to be base64-encoded). Also added, on Timo's advice, the "secured" option
2099 if the connection is using TLS or if the remote IP is the same as the
2100 local IP, and the "valid-client-cert option" if a client certificate has
2103 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
2104 authenticators. This can be used for authorization after authentication
2105 succeeds. (In the case of plaintext, it servers for both authentication
2108 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
2109 if any retry times were supplied.
2111 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
2112 connection, where there is no client IP address to check. In this
2113 situation, the verify now always succeeds.
2115 PH/25 Applied John Jetmore's -Mset patch.
2117 PH/26 Added -bem to be like -Mset, but loading a message from a file.
2119 PH/27 In a string expansion for a processed (not raw) header when multiple
2120 headers of the same name were present, leading whitespace was being
2121 removed from all of them, but trailing whitespace was being removed only
2122 from the last one. Now trailing whitespace is removed from each header
2123 before concatenation. Completely empty headers in a concatenation (as
2124 before) are ignored.
2126 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
2127 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
2129 PH/29 [Removed. This was a change that I later backed out, and forgot to
2130 correct the ChangeLog entry (that I had efficiently created) before
2131 committing the later change.]
2133 PH/30 Exim was sometimes attempting to deliver messages that had suffered
2134 address errors (4xx response to RCPT) over the same connection as other
2135 messages routed to the same hosts. Such deliveries are always "forced",
2136 so retry times are not inspected. This resulted in far too many retries
2137 for the affected addresses. The effect occurred only when there were more
2138 hosts than the hosts_max_try setting in the smtp transport when it had
2139 the 4xx errors. Those hosts that it had tried were not added to the list
2140 of hosts for which the message was waiting, so if all were tried, there
2141 was no problem. Two fixes have been applied:
2143 (i) If there are any address or message errors in an SMTP delivery, none
2144 of the hosts (tried or untried) are now added to the list of hosts
2145 for which the message is waiting, so the message should not be a
2146 candidate for sending over the same connection that was used for a
2147 successful delivery of some other message. This seems entirely
2148 reasonable: after all the message is NOT "waiting for some host".
2149 This is so "obvious" that I'm not sure why it wasn't done
2150 previously. Hope I haven't missed anything, but it can't do any
2151 harm, as the worst effect is to miss an optimization.
2153 (ii) If, despite (i), such a delivery is accidentally attempted, the
2154 routing retry time is respected, so at least it doesn't keep
2155 hammering the server.
2157 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
2158 in ${readsocket because some servers need this prod.
2160 PH/32 Added some extra debug output when updating a wait-xxx database.
2162 PH/33 The hint "could be header name not terminated by colon", which has been
2163 given for certain expansion errors for a long time, was not being given
2164 for the ${if def:h_colon_omitted{... case.
2166 PH/34 The spec says: "With one important exception, whenever a domain list is
2167 being scanned, $domain contains the subject domain." There was at least
2168 one case where this was not true.
2170 PH/35 The error "getsockname() failed: connection reset by peer" was being
2171 written to the panic log as well as the main log, but it isn't really
2172 panic-worthy as it just means the connection died rather early on. I have
2173 removed the panic log writing for the ECONNRESET error when getsockname()
2176 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
2177 runs only) independently of the message's sender address. This meant
2178 that, if the 4xx error was in fact related to the sender, a different
2179 message to the same recipient with a different sender could confuse
2180 things. In particualar, this can happen when sending to a greylisting
2181 server, but other circumstances could also provoke similar problems.
2182 I have changed the default so that the retry time for these errors is now
2183 based a combination of the sender and recipient addresses. This change
2184 can be overridden by setting address_retry_include_sender=false in the
2187 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
2188 remote server are returned as part of bounce messages. This was not
2189 happening for LMTP over a pipe (the lmtp transport), but now it is the
2190 same for both kinds of LMTP.
2192 PH/38 Despite being documented as not happening, Exim was rewriting addresses
2193 in header lines that were in fact CNAMEs. This is no longer the case.
2195 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
2196 and queue runs started by the daemon processed all messages. This has
2197 been fixed so that -R and -S can now usefully be given with -q<time>.
2199 PH/40 Import PCRE release 6.7 (fixes some bugs).
2201 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
2203 PH/42 Give an error if -q is specified more than once.
2205 PH/43 Renamed the variables $interface_address and $interface_port as
2206 $received_ip_address and $received_port, to make it clear that these
2207 values apply to message reception, and not to the outgoing interface when
2208 a message is delivered. (The old names remain recognized, of course.)
2210 PH/44 There was no timeout on the connect() call when using a Unix domain
2211 socket in the ${readsocket expansion. There now is.
2213 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
2214 be meaningful with "accept".
2216 SC/01 Eximstats V1.43
2217 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
2219 SC/02 Eximstats V1.44
2220 Use a glob alias rather than an array ref in the generated
2221 parser. This improves both readability and performance.
2223 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
2224 Collect SpamAssassin and rejection statistics.
2225 Don't display local sender or destination tables unless
2226 there is data to show.
2227 Added average volumes into the top table text output.
2229 SC/04 Eximstats V1.46
2230 Collect data on the number of addresses (recipients)
2231 as well as the number of messages.
2233 SC/05 Eximstats V1.47
2234 Added 'Message too big' to the list of mail rejection
2235 reasons (thanks to Marco Gaiarin).
2237 SC/06 Eximstats V1.48
2238 Mainlog lines which have GMT offsets and are too short to
2239 have a flag are now skipped.
2241 SC/07 Eximstats V1.49 (Alain Williams)
2242 Added the -emptyok flag.
2244 SC/08 Eximstats V1.50
2245 Fixes for obtaining the IP address from reject messages.
2247 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
2248 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
2249 whitesspace changes from 4.64-PH/27
2251 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
2254 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
2255 are found, allow negative numbers in numeric criteria)
2257 JJ/06 exipick.20061117.2, added new $message_body_missing variable
2259 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
2260 to match changes made in 4.64-PH/43
2262 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
2264 PH/47 Put in an explicit test for a DNS lookup of an address record where the
2265 "domain" is actually an IP address, and force a failure. This locks out
2266 those revolvers/nameservers that support "A-for-A" lookups, in
2267 contravention of the specifications.
2269 PH/48 When a host name was looked up from an IP address, and the subsequent
2270 forward lookup of the name timed out, the host name was left in
2271 $sender_host_name, contrary to the specification.
2273 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
2274 restricted to single-key lookups, Exim was not diagnosing an error if
2275 * or *@ was used with a query-style lookup.
2277 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
2279 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
2280 long ago, but noone interested enough thought of it. Let's just say that
2281 the "1.1" means that there are some new functions that weren't there at
2282 some point in the past.
2284 PH/51 Error processing for expansion failure of helo_data from an smtp
2285 transport during callout processing was broken.
2287 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
2288 tested/used via the -bh/-bhc/-bs options.
2290 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
2291 bug, fixed in subsequent PCRE releases).
2293 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
2294 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
2296 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
2302 SC/01 Use a glob alias rather than an array ref in eximstats generated
2303 parser. This improves both readability and performance.
2305 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
2306 Don't display local sender or destination tables in eximstats unless
2307 there is data to show.
2308 Added average volumes into the eximstats top table text output.
2310 SC/03 Collect data on the number of addresses (recipients) as well
2311 as the number of messages in eximstats.
2313 TF/01 Correct an error in the documentation for the redirect router. Exim
2314 does (usually) call initgroups() when daemonizing.
2316 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
2317 with consistent privilege compared to when running as a daemon.
2319 TF/03 Note in the spec that $authenticated_id is not set for local
2320 submissions from trusted users.
2322 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
2323 Thanks to Dean Brooks <dean@iglou.com> for the patch.
2325 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
2326 by adding some example configuration directives to the default
2327 configuration file. A little bit of work is required to uncomment the
2328 directives and define how usernames and passwords are checked, but
2329 there is now a framework to start from.
2331 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
2332 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
2333 without this. I don't know how relevant this is to other LDAP libraries.
2335 PH/02 Add the verb name to the "unknown ACL verb" error.
2337 PH/03 Magnus Holmgren's patch for filter_prepend_home.
2339 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
2341 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
2342 directory not expanded when it should be if an expanded home directory
2343 was set for the address (which is overridden by the transport).
2345 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
2348 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
2349 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
2350 because it is too late at that time, and has no effect.
2352 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
2353 security issue with \' (bugzilla #107). I could not use the
2354 PQescapeStringConn() function, because it needs a PGconn value as one of
2357 PH/08 When testing addresses using -bt, indicate those final addresses that
2358 are duplicates that would not cause an additional delivery. At least one
2359 person was confused, thinking that -bt output corresponded to deliveries.
2360 (Suppressing duplicates isn't a good idea as you lose the information
2361 about possibly different redirections that led to the duplicates.)
2363 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
2364 systems where poll() doesn't work, in particular OS X.
2366 PH/10 Added more information to debugging output for retry time not reached.
2368 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
2369 operations in malware.c.
2371 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
2374 PH/13 If write_rejectlog was set false when logging was sent to syslog with
2375 syslog_duplication set false, log lines that would normally be written
2376 both the the main log and to the reject log were not written to syslog at
2379 PH/14 In the default configuration, change the use of "message" in ACL warn
2380 statements to "add_header".
2382 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
2383 not followed by a command (e.g. "seen endif").
2385 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
2386 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
2389 PH/17 Added extra conditions to the default value of delay_warning_condition
2393 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
2394 { match{$h_precedence:}{(?i)bulk|list|junk} } \
2395 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
2398 The Auto-Submitted: and various List- headers are standardised, whereas I
2399 don't think Precedence: ever was.
2401 PH/18 Refactored debugging code in route_finduser() to show more information,
2402 in particular, the error code if getpwnam() issues one.
2404 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
2405 This is apparently needed in addition to the PH/07 change above to avoid
2406 any possible encoding problems.
2408 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
2409 but not after initializing Perl.
2411 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
2412 output them only if debugging. By default they are written stderr,
2413 apparently, which is not desirable.
2415 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
2418 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
2421 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
2423 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
2424 authenticated or an ident call has been made. Suppress the default
2425 values for $authenticated_id and $authenticated_sender (but permit -oMai
2426 and -oMas) when testing with -bh.
2428 PH/24 Re-jigged the order of the tests in the default configuration so that the
2429 tests for valid domains and recipients precede the DNS black list and CSA
2430 tests, on the grounds that those ones are more expensive.
2432 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
2433 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
2434 This bug exists in every version of Exim that I still have, right back to
2437 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
2438 However, an attempt to turn on case-sensitivity in a regex key by
2439 including (?-i) didn't work because the subject string was already
2440 lowercased, and the effects were non-intuitive. It turns out that a
2441 one-line patch can be used to allow (?-i) to work as expected.
2447 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
2448 other effects) broke the use of negated acl sub-conditions.
2450 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
2453 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
2454 "Deny" causes Exim to reject the incoming connection with a 554 error.
2455 Unfortunately, if there is a major crisis, such as a disk failure,
2456 tcp-wrappers gives "deny", whereas what one would like would be some
2457 kind of temporary error. A kludge has been added to help with this.
2458 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
2459 554 error is used if errno is still zero or contains ENOENT (which occurs
2460 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
2463 PH/03 Add -lutil to the default FreeBSD LIBS setting.
2465 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
2466 errors. Otherwise a message that provokes a temporary error (when other
2467 messages do not) can cause a whole host to time out.
2469 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
2470 addresses were routed directly to files or pipes from a redirect router.
2471 File deliveries just didn't batch; pipe deliveries might have suffered
2474 PH/06 A failure to get a lock for a hints database would erroneously always say
2475 "Failed to get write lock", even when it was really a read lock.
2477 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
2478 of 0600. This has been changed to use the value of the lockfile_mode
2479 option (which defaults to 0600).
2481 PH/08 Applied small patch from the Sieve maintainer.
2483 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
2484 folder from quota calculations, a direct delivery into this folder messed
2485 up the contents of the maildirsize file. This was because the regex was
2486 used only to exclude .Trash (or whatever) when the size of the mailbox
2487 was calculated. There was no check that a delivery was happening into an
2488 excluded directory. This bug has been fixed by ignoring all quota
2489 processing for deliveries into excluded directories.
2491 PH/10 Added the maildirfolder_create_regex option to appendfile.
2497 PH/01 The code for finding all the local interface addresses on a FreeBSD
2498 system running IPv6 was broken. This may well have applied to all BSD
2499 systems, as well as to others that have similar system calls. The broken
2500 code found IPv4 interfaces correctly, but gave incorrect values for the
2501 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
2502 that it would not match correctly against @[] and not recognize the IPv6
2505 PH/02 The ipliteral router was not recognizing addresses of the form user@
2506 [ipv6:....] because it didn't know about the "ipv6:" prefix.
2508 PH/03 Added disable_ipv6.
2510 PH/04 Changed $reply_address to use the raw form of the headers instead of the
2511 decoded form, because it is most often used to construct To: headers
2512 lines in autoreplies, and the decoded form may well be syntactically
2513 invalid. However, $reply_address has leading white space removed, and all
2514 newlines turned into spaces so that the autoreply transport does not
2517 PH/05 If group was specified without a user on a router, and no group or user
2518 was specified on a transport, the group from the router was ignored.
2520 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
2521 for visible compile-time settings that can be used to change these
2522 numbers, for those that want even more. Backwards compatibility with old
2523 spool files has been maintained. However, going back to a previous Exim
2524 release will lost any variables that are in spool files.
2526 PH/07 Two small changes when running in the test harness: increase delay when
2527 passing a TCP/IP connection to a new process, in case the original
2528 process has to generate a bounce, and remove special handling of
2529 127.0.0.2 (sic), which is no longer necessary.
2531 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
2532 be the same on different OS.
2534 PH/09 Moved a debug statement in filter processing to avoid a race problem when
2537 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
2538 whether --show-vars was specified or not
2540 JJ/02 exipick: Added support for new ACL variable spool format introduced
2543 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
2544 syntactically invalid From: or Reply-to: line, and a filter used this to
2545 generate an autoreply, and therefore failed to obtain an address for the
2546 autoreply, Exim could try to deliver to a non-existent relative file
2547 name, causing unrelated and misleading errors. What now happens is that
2548 it logs this as a hard delivery error, but does not attempt to create a
2551 PH/11 The exinext utility has a -C option for testing purposes, but although
2552 the given file was scanned by exinext itself; it wasn't being passed on
2553 when Exim was called.
2555 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
2556 an end-of-file indication when reading a command response.
2558 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
2559 compiled. In many other places in Exim, IPv6 addresses are always
2560 recognized, so I have changed this. It also means that IPv4 domain
2561 literals of the form [IPV4:n.n.n.n] are now always recognized.
2563 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
2564 used if the router is not running as root, for example, when verifying at
2565 ACL time, or when using -bh. The debugging output from this situation was
2566 non-existent - all you got was a failure to exec. I have made two
2569 (a) Failures to set uid/gid, the current directory, or a process leader
2570 in a subprocess such as that created by queryprogram now generate
2571 suitable debugging ouput when -d is set.
2573 (b) The queryprogram router detects when it is not running as root,
2574 outputs suitable debugging information if -d is set, and then runs
2575 the subprocess without attempting to change uid/gid.
2577 PH/15 Minor change to Makefile for building test_host (undocumented testing
2580 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
2581 additional section of a DNS packet that returns MX or SRV records.
2582 Instead, it always explicitly searches for A/AAAA records. This avoids
2583 major problems that occur when a DNS server includes only records of one
2584 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
2585 fixed another bug: if SRV records were looked up and the corresponding
2586 address records were *not* found in the additional section, the port
2587 values from the SRV records were lost.
2589 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
2590 using the correct key (the original address) when searching the retry
2591 rules in order to find which one to use for generating the retry hint.
2593 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
2594 adding the default one. Similarly, if it contains a Reply-To: header, the
2595 errors_reply_to option, if set, is not used.
2597 PH/19 When calculating a retry time, Exim used to measure the "time since
2598 failure" by looking at the "first failed" field in the retry record. Now
2599 it does not use this if it is later than than the arrival time of the
2600 message. Instead it uses the arrival time. This makes for better
2601 behaviour in cases where some deliveries succeed, thus re-setting the
2602 "first failed" field. An example is a quota failure for a huge message
2603 when small messages continue to be delivered. Without this change, the
2604 "time since failure" will always be short, possible causing more frequent
2605 delivery attempts for the huge message than are intended.
2606 [Note: This change was subsequently modified - see PH/04 for 4.62.]
2608 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
2609 $1, $2, $3) because the numerical variables can be reset during some
2610 expansion items (e.g. "match"), thereby losing the authentication data.
2612 PH/21 Make -bV show the size of off_t variables so that the test suite can
2613 decide whether to run tests for quotas > 2G.
2615 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
2616 mailbox_size, and mailbox_filecount in the appendfile transport. If a
2617 filecount value is greater than 2G or if a quota value is greater than 2G
2618 on a system where the size of off_t is not greater than 4, a panic error
2621 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
2622 never match. The debug and -bh output now contains an explicit error
2623 message indicating a malformed IPv4 address or mask.
2625 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
2626 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
2627 PH/23 above applies.
2629 PH/25 Do not write to syslog when running in the test harness. The only
2630 occasion when this arises is a failure to open the main or panic logs
2631 (for which there is an explicit test).
2633 PH/26 Added the /no_tell option to "control=freeze".
2635 PH/27 If a host name lookup failed very early in a connection, for example, if
2636 the IP address matched host_lookup and the reverse lookup yielded a name
2637 that did not have a forward lookup, an error message of the form "no IP
2638 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
2639 could be logged. Now it outputs the IP address instead of "NULL".
2641 PH/28 An enabling patch from MH: add new function child_open_exim2() which
2642 allows the sender and the authenticated sender to be set when
2643 submitting a message from within Exim. Since child_open_exim() is
2644 documented for local_scan(), the new function should be too.
2646 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
2647 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
2648 results in an empty string is now treated as unset.
2650 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
2652 PH/31 Added sender_verify_fail logging option.
2654 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
2655 needed by RFC 822 but not by RFC 2822 was commented out. I have now
2656 tidied the source and removed it altogether.
2658 PH/33 When a queue run was abandoned because the load average was too high, a
2659 log line was always written; now it is written only if the queue_run log
2660 selector is set. In addition, the log line for abandonment now contains
2661 information about the queue run such as the pid. This is always present
2662 in "start" and "stop" lines but was omitted from the "abandon" line.
2664 PH/34 Omit spaces between a header name and the colon in the error message that
2665 is given when verify = headers_syntax fails (if there are lots of them,
2666 the message gets confusing).
2668 PH/35 Change the default for dns_check_names_pattern to allow slashes within
2669 names, as there are now some PTR records that contain slashes. This check
2670 is only to protect against broken name servers that fall over on strange
2671 characters, so the fact that it applies to all lookups doesn't matter.
2673 PH/36 Now that the new test suite is complete, we can remove some of the
2674 special code in Exim that was needed for the old test suite. For example,
2675 sorting DNS records because real resolvers return them in an arbitrary
2676 order. The new test suite's fake resolver always returns records in the
2679 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
2680 bounces) except when queue_only is set, to avoid logging races between
2681 the different processes.
2683 PH/38 Panic-die if .include specifies a non-absolute path.
2685 PH/39 A tweak to the "H" retry rule from its user.
2687 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
2688 a label. They prevented compilation on older perls.
2690 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
2691 a warning to be raised on newish perls.
2693 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
2694 on queue. Changes to match documented behaviour of showing count of
2695 messages matching specified criteria.
2697 PH/40 Changed the default ident timeout from 30s to 5s.
2699 PH/41 Added support for the use of login_cap features, on those BSD systems
2700 that have them, for controlling the resources used by pipe deliveries.
2702 PH/42 The content-scanning code uses fopen() to create files in which to put
2703 message data. Previously it was not paying any attention to the mode of
2704 the files. Exim runs with umask(0) because the rest of the code creates
2705 files with open(), and sets the required mode explicitly. Thus, these
2706 files were ending up world-writeable. This was not a big issue, because,
2707 being within the spool directory, they were not world-accessible. I have
2708 created a function called modefopen, which takes an additional mode
2709 argument. It sets umask(777), creates the file, chmods it to the required
2710 mode, then resets the umask. All the relevant calls to fopen() in the
2711 content scanning code have been changed to use this function.
2713 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
2714 to 24 hours. This avoids potential overflow problems when processing G
2715 and H retry rules. I suspect nobody ever tinkers with this value.
2717 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
2719 PH/45 When the plaintext authenticator is running as a client, the server's
2720 challenges are checked to ensure they are valid base64 strings. By
2721 default, the authentication attempt is cancelled if an invalid string is
2722 received. Setting client_ignore_invalid_base64 true ignores these errors.
2723 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
2724 they are received. Thus, the responses can be made to depend on the
2725 challenges. If an invalid string is ignored, an empty string is placed in
2728 PH/46 Messages that are created by the autoreply transport now contains a
2729 References: header, in accordance with RFCs 2822 and 3834.
2731 PH/47 Added authenticated_sender_force to the smtp transport.
2733 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
2735 PH/49 Installed latest patch from the Sieve maintainer.
2737 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
2738 was also set, the appendfile transport was unnecessarily scanning a
2739 directory of message files (e.g. for maildir delivery) to find the count
2740 of files (along with the size), even though it did not need this
2741 information. It now does the scan only if it needs to find either the
2742 size of the count of files.
2744 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
2746 PH/52 Two bugs concerned with error handling when the smtp transport is
2749 (i) Exim was not creating retry information for temporary errors given
2750 for individual recipients after the DATA command when the smtp transport
2751 was used in LMTP mode. This meant that they could be retried too
2752 frequently, and not timed out correctly.
2754 (ii) Exim was setting the flag that allows error details to be returned
2755 for LMTP errors on RCPT commands, but not for LMTP errors for individual
2756 recipients that were returned after the DATA command.
2758 PH/53 This is related to PH/52, but is more general: for any failing address,
2759 when detailed error information was permitted to be returned to the
2760 sender, but the error was temporary, then after the final timeout, only
2761 "retry timeout exceeded" was returned. Now it returns the full error as
2762 well as "retry timeout exceeded".
2764 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
2765 do this, and (what is worse) MTAs that accept it.
2767 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
2768 will now be deprecated.
2770 PH/56 New os.c-cygwin from the Cygwin maintainer.
2772 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
2773 formats (previously only available in exim formats via -bpr, -bpru,
2774 and -bpra. Now also available in native and exiqgrep formats)
2776 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
2777 with very large, slow to parse queues
2779 JJ/08 exipick: added ! as generic prefix to negate any criteria format
2781 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
2783 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
2784 responses to authentication challenges, though it was showing the
2785 challenges; (ii) I've removed the CR characters from the debug output for
2788 PH/58 Allow for the insertion of a newline as well as a space when a string
2789 is turned into more than one encoded-word during RFC 2047 encoding. The
2790 Sieve code now uses this.
2792 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
2793 data_4xx, lost_connection, tls_required.
2795 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
2796 message was being sent as an SMTP response.
2798 PH/61 Add -l and -k options to exicyclog.
2800 PH/62 When verifying, if an address was redirected to one new address, so that
2801 verification continued, and the new address failed or deferred after
2802 having set something in $address_data, the value of $address_data was not
2803 passed back to the ACL. This was different to the case when no
2804 redirection occurred. The value is now passed back in both cases.
2806 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
2807 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
2808 use login_cap.h, so on its own it isn't the distinguishing feature. The
2809 new name refers directly to the setclassresources() function.
2811 PH/65 Added configuration files for NetBSD3.
2813 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
2815 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
2816 is preferred over IPv4.
2818 PH/68 The bounce_return_message and bounce_return_body options were not being
2819 honoured for bounces generated during the reception of non-SMTP messages.
2820 In particular, this applied to messages rejected by the ACL. This bug has
2821 been fixed. However, if bounce_return_message is true and bounce_return_
2822 body is false, the headers that are returned for a non-SMTP message
2823 include only those that have been read before the error was detected.
2824 (In the case of an ACL rejection, they have all been read.)
2826 PH/69 The HTML version of the specification is now built in a directory called
2827 spec_html instead of spec.html, because the latter looks like a path with
2828 a MIME-type, and this confuses some software.
2830 PH/70 Catch two compiler warnings in sieve.c.
2832 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
2833 function verify_get_ident() calls ip_connect() to connect a socket, but
2834 if the "connect()" function timed out, ip_connect() used to close the
2835 socket. However, verify_get_ident() also closes the socket later, and in
2836 between Exim writes to the log, which may get opened at this point. When
2837 the socket was closed in ip_connect(), the log could get the same file
2838 descriptor number as the socket. This naturally causes chaos. The fix is
2839 not to close the socket in ip_connect(); the socket should be closed by
2840 the function that creates it. There was only one place in the code where
2841 this was missing, in the iplookup router, which I don't think anybody now
2842 uses, but I've fixed it anyway.
2844 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
2845 well as to direct DNS lookups. Otherwise the handling of names in host
2846 lists is inconsistent and therefore confusing.
2852 PH/01 Two changes to the default runtime configuration:
2854 (1) Move the checks for relay_from_hosts and authenticated clients from
2855 after to before the (commented out) DNS black list checks.
2857 (2) Add control=submission to the relay_from_hosts and authenticated
2858 clients checks, on the grounds that messages accepted by these
2859 statements are most likely to be submissions.
2861 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
2863 (1) Generate an error if the third argument for the ${prvs expansion is
2866 (2) Treat a missing third argument of ${prvscheck as if it were an empty
2869 (3) Reset the variables that are obtained from the first argument of
2870 ${prvscheck and used in the second argument before leaving the code,
2871 because their memory is reclaimed, so using them afterwards may do
2874 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
2875 one (it's much easier than Tom thought :-).
2877 (5) Because of (4), we can now allow for the use of $prvscheck_result
2878 inside the third argument.
2880 PH/03 For some reason, the default setting of PATH when running a command from
2881 a pipe transport was just "/usr/bin". I have changed it to
2884 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
2885 anything to be listed in the output from -bV.
2887 PH/05 When a filter generated an autoreply, the entire To: header line was
2888 quoted in the delivery log line, like this:
2890 => >A.N.Other <ano@some.domain> <original@ddress> ...
2892 This has been changed so that it extracts the operative address. There
2893 may be more than one such address. If so, they are comma-separated, like
2896 => >ano@some.domain,ona@other.domain <original@ddress> ...
2898 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
2899 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
2900 not being looked up in the rDNS to get a host name, Exim was showing the
2901 IP address twice in Received: lines, even though the IP addresses were
2902 identical. For example:
2904 Received: from [1.2.3.4] (helo=[1.2.3.4])
2906 However, if the real host name was known, it was omitting the HELO data
2907 if it matched the actual IP address. This has been tidied up so that it
2908 doesn't show the same IP address twice.
2910 PH/07 When both +timestamp and +memory debugging was on, the value given by
2911 $tod_xxx expansions could be wrong, because the tod_stamp() function was
2912 called by the debug printing, thereby overwriting the timestamp buffer.
2913 Debugging no longer uses the tod_stamp() function when +timestamp is set.
2915 PH/08 When the original message was included in an autoreply transport, it
2916 always said "this is a copy of the message, including all the headers",
2917 even if body_only or headers_only was set. It now gives an appropriate
2920 PH/09 Applied a patch from the Sieve maintainer which:
2922 o fixes some comments
2923 o adds the (disabled) notify extension core
2924 o adds some debug output for the result of if/elsif tests
2925 o points to the current vacation draft in the documentation
2926 and documents the missing references header update
2930 o fixes a bug in processing the envelope test (when testing
2931 multiple envelope elements, the last element determinted the
2934 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
2935 Electronic Mail") by including:
2937 Auto-submitted: auto-generated
2939 in the messages that it generates (bounce messages and others, such as
2940 warnings). In the case of bounce messages for non-SMTP mesages, there was
2941 also a typo: it was using "Auto_submitted" (underscore instead of
2942 hyphen). Since every message generated by Exim is necessarily in response
2943 to another message, thes have all been changed to:
2945 Auto-Submitted: auto-replied
2947 in accordance with these statements in the RFC:
2949 The auto-replied keyword:
2951 - SHOULD be used on messages sent in direct response to another
2952 message by an automatic process,
2954 - MUST NOT be used on manually-generated messages,
2956 - MAY be used on Delivery Status Notifications (DSNs) and Message
2957 Disposition Notifications (MDNs),
2959 - MUST NOT be used on messages generated by automatic or periodic
2960 processes, except for messages which are automatic responses to
2963 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
2964 to the default Received: header definition.
2966 PH/12 Added log selector acl_warn_skipped (default on).
2968 PH/13 After a successful wildlsearch lookup, discard the values of numeric
2969 variables because (a) they are in the wrong storage pool and (b) even if
2970 they were copied, it wouldn't work properly because of the caching.
2972 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
2973 checking when decoding. Apparently there are clients that generate
2974 overlong encoded strings. Why am I not surprised?
2976 PH/15 If the first argument of "${if match_address" was not empty, but did not
2977 contain an "@" character, Exim crashed. Now it writes a panic log message
2978 and treats the condition as false.
2980 PH/16 In autoreply, treat an empty string for "once" the same as unset.
2982 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
2983 extension "envelope-auth". The code is finished and in agreement with
2984 other implementations, but there is no documentation so far and in fact,
2985 nobody wrote the draft yet. This extension is currently #undef'ed, thus
2986 not changing the active code.
2988 Print executed "if" and "elsif" statements when debugging is used. This
2989 helps a great deal to understand what a filter does.
2991 Document more things not specified clearly in RFC3028. I had all this
2992 sorted out, when out of a sudden new issues came to my mind. Oops."
2994 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
2997 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
2998 canonical form (as documented). However, after a host name lookup from
2999 the IP address, check_host() was doing a simple string comparison with
3000 addresses acquired from the DNS when checking that the found name did
3001 have the original IP as one of its addresses. Since any found IPv6
3002 addresses are likely to be in abbreviated form, the comparison could
3003 fail. Luckily, there already exists a function for doing the comparison
3004 by converting both addresses to binary, so now that is used instead of
3005 the text comparison.
3007 PH/20 There was another similar case to PH/19, when a complete host name was
3008 given in a host list; looking up its IP address could give an abbreviated
3009 form, whereas the current host's name might or might not be abbreviated.
3010 The same fix has been applied.
3016 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
3017 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
3020 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
3021 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
3023 PH/03 Typo: missing ".o" in src/pcre/Makefile.
3025 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
3026 header line, restrict the check to what is listed in RFCs 2369 and 2929.
3027 Also, for "Auto-Submitted", treat anything other than "no" as
3028 non-personal, in accordance with RFC 3834. (Previously it treated
3029 anything starting "auto-" as non-personal.)
3031 TF/01 The control=submission/name=... option had a problem with syntax
3032 errors if the name included a slash character. The /name= option
3033 now slurps the rest of the string, so it can include any characters
3034 but it must come last in the list of options (after /sender_retain
3037 PH/05 Some modifications to the interface to the fake nameserver for the new
3045 TK/01 Added the "success_on_redirect" address verification option. See
3046 NewStuff for rationale and an example.
3048 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
3050 PH/02 Patch to exigrep to allow it to work on syslog lines.
3052 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
3053 fread() to skip over the body file's header line, because in Cygwin the
3054 header line is locked and is inaccessible.
3056 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
3057 co-exist for some time) to make it clear that it is the Exim ID that is
3058 referenced, not the Message-ID: header line.
3060 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
3061 string_format() function, because snprintf() does not exist on all
3064 PH/06 The use of forbid_filter_existstest now also locks out the use of the
3065 ${stat: expansion item.
3067 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
3068 protocol synchronization error", to keep the pedants happy.
3070 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
3071 well as for IRIX systems, when gcc is being used. See the host.c source
3074 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
3076 PH/10 Named domain lists were not working if used in a queue_smtp_domains
3079 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
3080 transport and to the smtp transport in LMTP mode.
3082 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
3084 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
3085 run a filter in a subprocess. This could lead to confusion in subsequent
3086 lookups in the parent process. There should also be a search_tidyup() at
3087 the end of the subprocess.
3089 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
3090 only if the host matched helo_try_verify_hosts, which caused the
3091 verification to occur when the EHLO/HELO command was issued. The ACL just
3092 tested the remembered result. Now, if a previous verification attempt has
3093 not happened, "verify = helo" does it there and then.
3095 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
3097 TK/03 Fix log output including CR from clamd.
3099 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
3100 exist provoked a memory error which could cause a segfault.
3102 PH/15 Installed PCRE 6.2
3104 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
3106 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
3107 of the problem. Specifically, suggested +O2 rather than +O1 for the
3110 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
3112 PH/20 If a delivery was routed to a non-standard port by means of an SRV
3113 record, the port was not correctly logged when the outgoing_port log
3114 selector was set (it logged the transort's default port).
3116 PH/21 Added support for host-specific ports to manualroute, queryprogram,
3117 fallback_hosts, and "hosts" in the smtp transport.
3119 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
3120 host errors such as "Connection refused".
3122 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
3123 authentication with radiusclient 0.4.9:
3125 - Error returned from rc_read_config was caught wrongly
3126 - Username/password not passed on to radius server due to wrong length.
3128 The presumption is that some radiusclient API changes for 4.51/PH/17
3129 were not taken care of correctly. The code is still untested by me (my
3130 Linux distribution still has 0.3.2 of radiusclient), but it was
3131 contributed by a Radius user.
3133 PH/24 When doing a callout, the value of $domain wasn't set correctly when
3134 expanding the "port" option of the smtp transport.
3136 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
3137 while reading a MIME header. Thanks to Tom Hughes for a patch.
3139 PH/24 Include config.h inside local_scan.h so that configuration settings are
3142 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
3143 that in an ACL for RCPT (for example), you can examine exactly what was
3146 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
3147 commands, but it was not correctly comparing the address with the actual
3148 client host address. Thus, it would show the EHLO address in Received:
3149 header lines when this was not necessary.
3151 PH/27 Added the % operator to ${eval:}.
3153 PH/28 Exim tries to create and chdir to its spool directory when it starts;
3154 it should be ignoring failures (because with -C, for example, it has lost
3155 privilege). It wasn't ignoring creation failures other than "already
3158 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
3161 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
3162 because an input line was too long, either on its own, or by virtue of
3163 too many continuations, the temporary file was not being removed, and the
3164 return code was incorrect.
3166 PH/31 Missing "BOOL" in function definition in filtertest.c.
3168 PH/32 Applied Sieve patches from the maintainer.
3170 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
3172 PH/33 Added "verify = not_blind".
3174 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
3175 Local/Makefile (with some defaults set). These are used in built scripts
3176 such as exicyclog, but they have never been used in the exim_install
3177 script (though there are many overriding facilities there). I have
3178 arranged that the exim_install script now takes note of these two
3181 PH/35 Installed configuration files for Dragonfly.
3183 PH/36 When a locally submitted message by a trusted user did not contain a
3184 From: header, and the sender address was obtained from -f or from an SMTP
3185 MAIL command, and the trusted user did not use -F to supply a sender
3186 name, $originator_name was incorrectly used when constructing a From:
3187 header. Furthermore, $originator_name was used for submission mode
3188 messages from external hosts without From: headers in a similar way,
3189 which is clearly wrong.
3191 PH/37 Added control=suppress_local_fixups.
3193 PH/38 When log_selector = +received_sender was set, and the addition of the
3194 sender made the log line's construction buffer exactly full, or one byte
3195 less than full, an overflow happened when the terminating "\n" was
3198 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
3199 when the result of a list match is failure because a DNS lookup failed.
3201 PH/40 RM_COMMAND is now used in the building process.
3203 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
3204 the "build-* directories that it finds.
3206 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
3207 address in a domain literal was a prefix of an interface address.
3209 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
3210 when verifying a sender address, unless rewrite_headers is false.
3212 PH/44 Wrote a long comment about why errors_to addresses are verified as
3213 recipients, not senders.
3215 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
3216 the ratelimit ACL was added.
3218 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
3220 PH/46 Added extra information about PostgreSQL errors to the error string.
3222 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
3223 avoiding the need to install special zones in a real server. This is
3224 backwards compatible; if it can't find the fake resolver, it drops back.
3225 Thus, both old and new test suites can be run.
3227 TF/02 Added util/ratelimit.pl
3229 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
3230 clock is set back in time.
3232 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
3233 Candler <B.Candler@pobox.com>.
3235 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
3236 OK for addresses that are the result of redirections.
3238 PH/48 A number of further additions for the benefit of the new test suite,
3239 including a fake gethostbyname() that interfaces to the fake DNS resolver
3242 TF/06 The fix for widen_domains has also been applied to qualify_single and
3243 search_parents which are the other dnslookup options that can cause
3246 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
3249 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
3251 TF/07 Exim produced the error message "an SRV record indicated no SMTP
3252 service" if it encountered an MX record with an empty target hostname.
3253 The message is now "an MX or SRV record indicated no SMTP service".
3255 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
3256 if the DNS of the sending site is misconfigured. This is quite a
3257 common situation. This change restores the behaviour of treating a
3258 helo verification defer as a failure.
3260 PH/51 If self=fail was set on a router, the bounce message did not include the
3261 actual error message.
3267 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
3269 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
3270 command itself ended in error, the underlying message about the transport
3271 filter timeout was being overwritten with the pipe command error. Now the
3272 underlying error message should be appended to the second error message.
3274 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
3276 PH/02 Reduce the amount of output that "make" produces by default. Full output
3277 can still be requested.
3279 PH/03 The warning log line about a condition test deferring for a "warn" verb
3280 was being output only once per connection, rather than after each
3281 occurrence (because it was using the same function as for successful
3282 "warn" verbs). This seems wrong, so I have changed it.
3284 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
3285 it should not have, which might have caused a crash in the right
3286 circumstances, but probably never did.
3288 PH/04 Installed a modified version of Tony Finch's patch to make submission
3289 mode fix the return path as well as the Sender: header line, and to
3290 add a /name= option so that you can make the user's friendly name appear
3293 TF/03 Added the control = fakedefer ACL modifier.
3295 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
3296 Mark Lowes for thorough testing.
3298 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
3300 TK/03 Merged latest SRS patch from Miles Wilton.
3302 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
3303 with the definition in sysexits.h (which is #included earlier).
3304 Fortunately, Exim does not actually use EX_OK. The code used to try to
3305 preserve the sysexits.h value, by assumimg that macro definitions were
3306 scanned for macro replacements. I have been disabused of this notion,
3307 so now the code just undefines EX_OK before #including unistd.h.
3309 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
3310 in the smtp transport. When a block could not be written in a single
3311 write() function, the timeout was being re-applied to each part-write.
3312 This seems wrong - if the receiver was accepting one byte at a time it
3313 would take for ever. The timeout is now adjusted when this happens. It
3314 doesn't have to be particularly precise.
3316 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
3317 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
3319 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
3320 without a domain if the check to <postmaster@domain> fails.
3322 SC/01 Eximstats: added -xls and the ability to specify output files
3323 (patch written by Frank Heydlauf).
3325 SC/02 Eximstats: use FileHandles for outputing results.
3327 SC/03 Eximstats: allow any combination of xls, txt, and html output.
3329 SC/04 Eximstats: fixed display of large numbers with -nvr option
3331 SC/05 Eximstats: fixed merging of reports with empty tables.
3333 SC/06 Eximstats: added the -include_original_destination flag
3335 SC/07 Eximstats: removed tabs and trailing whitespace.
3337 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
3339 TK/06 MBOX spool code: Add real "From " MBOX separator line
3340 so the .eml file is really in mbox format (even though
3341 most programs do not really care). Patch from Alex Miller.
3343 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
3344 The latter is generated from $received_to and is only set if the
3345 message has one envelope recipient. SA can use these headers,
3346 obviously out-of-the-box. Patch from Alex Miller.
3348 PH/08 The ${def test on a variable was returning false if the variable's
3349 value was "0", contrary to what the specification has always said!
3350 The result should be true unless the variable is empty.
3352 PH/09 The syntax error of a character other than { following "${if
3353 def:variable_name" (after optional whitespace) was not being diagnosed.
3354 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
3355 accidental colon was present, for example, could give incorrect results.
3357 PH/10 Tidied the code in a number of places where the st_size field of a stat()
3358 result is used (not including appendfile, where other changes are about
3361 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
3362 This involved changing a lot of size variables from int to off_t. It
3363 should work with maildirs and everything.
3365 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
3366 spamd dying while we are connected to it.
3368 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
3371 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
3372 function for BDB 4.3.
3374 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
3376 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
3377 the sources for PCRE have been re-arranged and more files are now
3380 PH/15 The code I had for printing potentially long long variables in PH/11
3381 above was not the best (it lost precision). The length of off_t variables
3382 is now inspected at build time, and an appropriate printing format (%ld
3383 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
3384 to be "long long int" or "long int". This is needed for the internal
3385 formatting function string_vformat().
3387 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
3388 the configuration file to be ":syslog", then the script "guesses" where
3389 the logs files are, rather than using the compiled in default. In our
3390 case the guess is not the same as the compiled default, so the script
3391 suddenly stopped working when I started to use syslog. The patch checks
3392 to see if log_file_path is "". If so, it attempts to read it from exim
3393 with no configuration file to get the compiled in version, before it
3394 falls back to the previous guessing code."
3396 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
3397 implementing BATV in an Exim configuration. See NewStuff for the gory
3400 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
3401 Makefile that are specific to HP-UX.
3403 PH/18 If the "use_postmaster" option was set for a recipient callout together
3404 with the "random" option, the postmaster address was used as the MAIL
3405 FROM address for the random test, but not for the subsequent recipient
3406 test. It is now used for both.
3408 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
3409 patch removes a few documentation additions to RFC 3028, because the
3410 latest draft now contains them. It adds the new en;ascii-case comparator
3411 and a new error check for 8bit text in MIME parts. Comparator and
3412 require names are now matched exactly. I enabled the subaddress
3413 extension, but it is not well tested yet (read: it works for me)."
3415 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
3416 rework some of the code of TK/09 above to avoid the hardwired use of
3417 "%lld" and "long long". Replaced the call to snprintf() with a call to
3420 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
3421 records point to non-existent hosts", "retry timeout exceeded", and
3422 "retry time not reached for any host after a long failure period".
3424 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
3425 experimental DomainKeys support:
3427 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
3428 (2) On an error such as an illegally used "control", the wrong name for
3429 the control was given.
3431 These problems did NOT occur unless DomainKeys support was compiled.
3433 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
3435 PH/24 Added ${if match_ip condition.
3437 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
3438 databases so that it will be absolutely obvious if a crash occurs in the
3439 DB library. This is a regular occurrence (often caused by mis-matched
3442 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
3443 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
3444 picked up on a user's system that detects such things. There doesn't seem
3445 to be a gcc warning option for this - only an attribute that has to be
3446 put on the function's prototype. It seems that in Fedora Core 4 they have
3447 set this on a number of new functions. No doubt there will be more in due
3450 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
3451 specify a transport. However, if an address that was verified by such a
3452 router was the subject of a callout, Exim crashed because it tried to
3453 read the rcpt_include_affixes from the non-existent transport. Now it
3454 just assumes that the setting of that option is false. This bug was
3455 introduced by 4.51/PH/31.
3457 PH/28 Changed -d+all to exclude +memory, because that information is very
3458 rarely of interest, but it makes the output a lot bigger. People tend to
3459 do -d+all out of habit.
3461 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
3462 code in os-type was giving problems when libc.so lives in lib64, like on
3465 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
3466 aren't the modern standard, and it seems that some systems' include files
3467 don't always have them. Exim was already checking for some of the newer
3468 ones like T_AAAA, and defining it itself. I've added checks for all the
3469 record types that Exim uses.
3471 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
3472 not automatically generating a new one, as it is supposed to. This
3473 prevented TLS from working. If the file did exist, but contained invalid
3474 data, a new version was generated, as expected. It was only the case of a
3475 non-existent file that was broken.
3477 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
3478 with a change in libdomainkeys > 0.64.
3480 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
3481 from DNS. If the selector record carries the flag, it now has
3482 precedence over the domain-wide flag.
3484 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
3486 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
3487 the use of a port name that isn't defined in /etc/services) occurred, the
3488 message was deferred as in a normal delivery, and thus remained on the
3489 spool, instead of being failed because of the mua_wrapper setting. This
3490 is now fixed, and I tidied up some of the mua_wrapper messages at the
3493 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
3494 the messages in a hash of arrays rather than using individual hashes.
3495 This is a bit cleaner and results in dramatic memory savings, albeit
3496 at a slight CPU cost.
3498 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
3499 as requested by Marc Sherman.
3501 SC/10 Eximstats: added histograms for user specified patterns as requested
3504 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
3506 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
3507 fopen() in the content-scanning modules that did not already have it.
3513 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
3514 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
3516 TK/02 Fix ACL "control" statement not being available in MIME ACL.
3518 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
3520 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
3521 to test Sieve filters that use "vacation".
3523 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
3524 that changes the way the GnuTLS parameters are stored in the cache file.
3525 The new format can be generated externally. For backward compatibility,
3526 if the data in the cache doesn't make sense, Exim assumes it has read an
3527 old-format file, and it generates new data and writes a new file. This
3528 means that you can't go back to an older release without removing the
3531 PH/03 A redirect router that has both "unseen" and "one_time" set does not
3532 work if there are any delivery delays because "one_time" forces the
3533 parent to be marked "delivered", so its unseen clone is never tried
3534 again. For this reason, Exim now forbids the simultaneous setting of
3537 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
3538 redirected to themselves ("homonym" addresses). Read the long ChangeLog
3539 entry if you want to know the details. The fix, however, neglected to
3540 consider the case when local delivery batching is involved. The test for
3541 "previously delivered" was not happening when checking to see if an
3542 address could be batched with a previous (undelivered) one; under
3543 certain circumstances this could lead to multiple deliveries to the same
3546 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
3547 in its include files, and this causes problems building Exim.
3549 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
3550 header_syntax) but Exim was just ignoring anything given after a slash.
3551 In particular, this caused confusion with an attempt to use "verify =
3552 reverse_host_lookup/defer_ok". An error is now given when options are
3553 supplied for verify items that do not have them. (Maybe reverse_host_
3554 lookup should have a defer_ok option, but that's a different point.)
3556 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
3557 defined by RFC 821) to 2048, because there were problems with some AUTH
3558 commands, and RFC 1869 says the size should be increased for extended
3559 SMTP commands that take arguments.
3561 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
3564 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
3565 "unknown" error; now it says that the functionality isn't in the binary.
3567 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
3568 an address' error message when a string expansion fails (syntax or
3569 whatever). Otherwise the password may appear in the log. Following change
3570 PH/42 below, there is no longer a chance of it appearing in a bounce
3573 PH/11 Installed exipick version 20050225.0 from John Jetmore.
3575 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
3576 of its addresses was ever tried. (Bugzilla bug #2.)
3578 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
3579 the result incorrectly in the debug output. (It correctly added a newline
3580 to what was transported.)
3582 TF/01 Added $received_time.
3584 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
3585 commented out examples of how to interface to a virus scanner and to
3586 SpamAssassin. Also added commented examples of av_scanner and
3587 spamd_address settings.
3589 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
3590 and controls are allowed in which ACLs. There were a couple of minor
3591 errors. Some of the entries in the conditions table (which is a table of
3592 where they are NOT allowed) were getting very unwieldy; rewrote them as a
3593 negation of where the condition IS allowed.
3595 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
3597 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
3598 header file does not have a version number, so I've had to invent a new
3599 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
3600 API. The code is untested by me (my Linux distribution still has 0.3.2 of
3601 radiusclient), but it was contributed by a Radius user.
3603 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
3604 files or directories, for OpenSSL.
3606 PH/19 When an Exim process that is running as root has to create an Exim log
3607 file, it does so in a subprocess that runs as exim:exim so as to get the
3608 ownership right at creation (otherwise, other Exim processes might see
3609 the file with the wrong ownership). There was no test for failure of this
3610 fork() call, which would lead to the process getting stuck as it waited
3611 for a non-existent subprocess. Forks do occasionally fail when resources
3612 run out. I reviewed all the other calls to fork(); they all seem to check
3615 PH/20 When checking for unexpected SMTP input at connect time (before writing
3616 the banner), Exim was not dealing correctly with a non-positive return
3617 from the read() function. If the client had disconnected by this time,
3618 the result was a log entry for a synchronization error with an empty
3619 string after "input=" when read() returned zero. If read() returned -1
3620 (an event I could not check), uninitialized data bytes were printed.
3621 There were reports of junk text (parts of files, etc) appearing after
3624 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
3626 PH/22 Added support for macro redefinition, and (re)definition in between
3627 driver and ACL definitions.
3629 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
3630 forgetting to use the resulting value; it was using the unexpanded value.
3632 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
3633 hadn't been configured. The fix is from Juergen Kreileder, who
3634 understands it better than I do:
3636 "Here's what I see happening with three configured cyrus_sasl
3637 authenticators configured (plain, login, cram-md5):
3639 On startup auth_cyrus_sasl_init() gets called for each of these.
3640 This means three calls to sasl_listmech() without a specified mech_list.
3641 => SASL tests which mechs of all available mechs actually work
3642 => three warnings about OTP not working
3643 => the returned list contains: plain, login, cram-md5, digest-md5, ...
3645 With the patch, sasl_listmech() also gets called three times. But now
3646 SASL's mech_list option is set to the server_mech specified in the the
3647 authenticator. Or in other words, the answer from sasl_listmech()
3648 gets limited to just the mech you're testing for (which is different
3650 => the return list contains just 'plain' or 'login', 'cram-md5' or
3651 nothing depending on the value of ob->server_mech.
3653 I've just tested the patch: Authentication still works fine,
3654 unavailable mechs specified in the exim configuration are still
3655 caught, and the auth.log warnings about OTP are gone."
3657 PH/25 When debugging is enabled, the contents of the command line are added
3658 to the debugging output, even when log_selector=+arguments is not
3661 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
3662 answer is "GNU", and only if the return is "GNU/something" is the answer
3665 PH/27 $acl_verify_message is now set immediately after the failure of a
3666 verification in an ACL, and so is available in subsequent modifiers. In
3667 particular, the message can be preserved by coding like this:
3669 warn !verify = sender
3670 set acl_m0 = $acl_verify_message
3672 Previously, $acl_verify_message was set only while expanding "message"
3673 and "log_message" when a very denied access.
3675 PH/28 Modified OS/os.c-Linux with
3677 -#ifndef OS_LOAD_AVERAGE
3678 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
3680 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
3681 nomenclature these days.)
3683 PH/29 Installed patch from the Sieve maintainer that adds the options
3684 sieve_useraddress and sieve_subaddress to the redirect router.
3686 PH/30 In these circumstances:
3687 . Two addresses routed to the same list of hosts;
3688 . First host does not offer TLS;
3689 . First host accepts first address;
3690 . First host gives temporary error to second address;
3691 . Second host offers TLS and a TLS session is established;
3692 . Second host accepts second address.
3693 Exim incorrectly logged both deliveries with the TLS parameters (cipher
3694 and peerdn, if requested) that were in fact used only for the second
3697 PH/31 When doing a callout as part of verifying an address, Exim was not paying
3698 attention to any local part prefix or suffix that was matched by the
3699 router that accepted the address. It now behaves in the same way as it
3700 does for delivery: the affixes are removed from the local part unless
3701 rcpt_include_affixes is set on the transport.
3703 PH/32 Add the sender address, as F=<...>, to the log line when logging a
3704 timeout during the DATA phase of an incoming message.
3706 PH/33 Sieve envelope tests were broken for match types other than :is. I have
3707 applied a patch sanctioned by the Sieve maintainer.
3709 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
3710 the uid or gid is negative. A case of a negative gid caused this to be
3711 noticed. The fix allows for either to be negative.
3713 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
3714 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
3717 PH/36 The change PH/12 above was broken. Fixed it.
3719 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
3720 the grounds that routing the same address twice would always produce the
3721 same answer. This might have been true once, but it is certainly no
3722 longer true now. Routing a child address may depend on the previous
3723 routing that produced that child. Some complicated redirection strategies
3724 went wrong when messages had multiple recipients, and made Exim's
3725 behaviour dependent on the order in which the addresses were given.
3727 I have moved the duplicate checking until after the routing is complete.
3728 Exim scans the addresses that are assigned to local and remote
3729 transports, and removes any duplicates. This means that more work will be
3730 done, as duplicates will always all be routed, but duplicates are
3731 presumably rare, so I don't expect this is of any significance.
3733 For deliveries to pipes, files, and autoreplies, the duplicate checking
3734 still happens during the routing process, since they are not going to be
3737 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
3738 It corrects a timeout issue with spamd. This is Ian's comment: "The
3739 background is that sometimes spamd either never reads data from a
3740 connection it has accepted, or it never writes response data. The exiscan
3741 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
3742 blindly assumes that writes won't block so it may never time out."
3744 PH/39 Allow G after quota size as well as K and M.
3746 PH/40 The value set for $authenticated_id in an authenticator may not contain
3747 binary zeroes or newlines because the value is written to log lines and
3748 to spool files. There was no check on this. Now the value is run through
3749 the string_printing() function so that such characters are converted to
3750 printable escape sequences.
3752 PH/41 $message_linecount is a new variable that contains the total number of
3753 lines in the message. Compare $body_linecount, which is the count for the
3756 PH/42 Exim no longer gives details of delivery errors for specific addresses in
3757 bounce and delay warning messages, except in certain special cases, which
3760 (a) An SMTP error message from a remote host;
3761 (b) A message specified in a :fail: redirection;
3762 (c) A message specified in a "fail" command in a system filter;
3763 (d) A message specified in a FAIL return from the queryprogram router;
3764 (e) A message specified by the cannot_route_message router option.
3766 In these cases only, Exim does include the error details in bounce and
3767 warning messages. There are also a few cases where bland messages such
3768 as "unrouteable address" or "local delivery error" are given.
3770 PH/43 $value is now also set for the "else" part of a ${run expansion.
3772 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
3773 being worked on, but at least Exim now implements the latest version to
3776 PH/45 In a pipe transport, although a timeout while waiting for the pipe
3777 process to complete was treated as a delivery failure, a timeout while
3778 writing the message to the pipe was logged, but erroneously treated as a
3779 successful delivery. Such timeouts include transport filter timeouts. For
3780 consistency with the overall process timeout, these timeouts are now
3781 treated as errors, giving rise to delivery failures by default. However,
3782 there is now a new Boolean option for the pipe transport called
3783 timeout_defer, which, if set TRUE, converts the failures into defers for
3784 both kinds of timeout. A transport filter timeout is now identified in
3787 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
3788 systems where "make" and "gmake" are different, calling "gmake" at top
3789 level broke things. I've arranged for the value of $(MAKE) to be passed
3790 from the Makefile to this script so that it can call the same version of
3794 A note about Exim versions 4.44 and 4.50
3795 ----------------------------------------
3797 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
3798 changes of various kinds. As a consequence, a big documentation update was
3799 needed. This delayed the release for rather longer than seemed good, especially
3800 in the light of a couple of (minor) security issues. Therefore, the changes
3801 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
3802 release. So 4.44 and 4.50 are in effect two different branches that both start
3805 I have left the 4.50 change log unchanged; it contains all the changes since
3806 4.43. The change log for 4.44 is below; many of its items are identical to
3807 those for 4.50. This seems to be the most sensible way to preserve the
3808 historical information.
3814 1. Minor wording change to the doc/README.SIEVE file.
3816 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
3817 computation of the current number of files was incorrect.
3819 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
3820 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
3823 4. Give more explanation in the error message when the command for a transport
3824 filter fails to execute.
3826 5. There are several places where Exim runs a non-Exim command in a
3827 subprocess. The SIGUSR1 signal should be disabled for these processes. This
3828 was being done only for the command run by the queryprogram router. It is
3829 now done for all such subprocesses. The other cases are: ${run, transport
3830 filters, and the commands run by the lmtp and pipe transports.
3832 6. Added CONFIGURE_GROUP build-time option.
3834 7. Some older OS have a limit of 256 on the maximum number of file
3835 descriptors. Exim was using setrlimit() to set 1000 as a large value
3836 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
3837 systems. I've change it so that if it can't get 1000, it tries for 256.
3839 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
3840 was an oversight, and furthermore, ever since the addition of extra
3841 controls (e.g. 4.43/32), the checks on when to allow different forms of
3842 "control" were broken. There should now be diagnostics for all cases when a
3843 control that does not make sense is encountered.
3845 9. Added the /retain_sender option to "control=submission".
3847 10. $recipients is now available in the predata ACL (oversight).
3849 11. Tidy the search cache before the fork to do a delivery from a message
3850 received from the command line. Otherwise the child will trigger a lookup
3851 failure and thereby defer the delivery if it tries to use (for example) a
3852 cached ldap connection that the parent has called unbind on.
3854 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
3855 of $address_data from the recipient verification was clobbered by the
3856 sender verification.
3858 13. The value of address_data from a sender verification is now available in
3859 $sender_address_data in subsequent conditions in the ACL statement.
3861 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
3863 15. Added a new option "connect=<time>" to callout options, to set a different
3866 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
3867 was its contents. (It was OK if the option was not defined at all.)
3869 17. A "Completed" log line is now written for messages that are removed from
3870 the spool by the -Mrm option.
3872 18. New variables $sender_verify_failure and $recipient_verify_failure contain
3873 information about exactly what failed.
3875 19. Added -dd to debug only the daemon process.
3877 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
3878 handles timeouts, both on the server side and network timeouts. Renamed the
3879 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
3881 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
3882 It is now set to "smtps".
3884 22. $host_address is now set to the target address during the checking of
3885 ignore_target_hosts.
3887 23. When checking ignore_target_hosts for an ipliteral router, no host name was
3888 being passed; this would have caused $sender_host_name to have been used if
3889 matching the list had actually called for a host name (not very likely,
3890 since this list is usually IP addresses). A host name is now passed as
3893 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
3894 code that specifies a non-restarting handler (typically sigaction() in
3895 modern systems) in an attempt to fix a rare and obscure crash bug.
3897 25. Narrowed the window for a race in the daemon that could cause it to ignore
3898 SIGCHLD signals. This is not a major problem, because they are used only to
3899 wake it up if nothing else does.
3901 26. A malformed maildirsize file could cause Exim to calculate negative values
3902 for the mailbox size or file count. Odd effects could occur as a result.
3903 The maildirsize information is now recalculated if the size or filecount
3906 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
3907 support for a long time. Removed HAVE_SYS_VFS_H.
3909 28. Installed the latest version of exipick from John Jetmore.
3911 29. In an address list, if the pattern was not a regular expression, an empty
3912 subject address (from a bounce message) matched only if the pattern was an
3913 empty string. Non-empty patterns were not even tested. This was the wrong
3914 because it is perfectly reasonable to use an empty address as part of a
3915 database query. An empty address is now tested by patterns that are
3916 lookups. However, all the other forms of pattern expect the subject to
3917 contain a local part and a domain, and therefore, for them, an empty
3918 address still always fails if the pattern is not itself empty.
3920 30. Exim went into a mad DNS loop when attempting to do a callout where the
3921 host was specified on an smtp transport, and looking it up yielded more
3922 than one IP address.
3924 31. Re-factored the code for checking spool and log partition space into a
3925 function that finds that data and another that does the check. The former
3926 is then used to implement four new variables: $spool_space, $log_space,
3927 $spool_inodes, and $log_inodes.
3929 32. The RFC2047 encoding function was originally intended for short strings
3930 such as real names; it was not keeping to the 75-character limit for
3931 encoded words that the RFC imposes. It now respects the limit, and
3932 generates multiple encoded words if necessary. To be on the safe side, I
3933 have increased the buffer size for the ${rfc2047: expansion operator from
3936 33. It is now permitted to omit both strings after an "if" condition; if the
3937 condition is true, the result is "true". As before, when the second string
3938 is omitted, a false condition yields an empty string. This makes it less
3939 cumbersome to write custom ACL and router conditions.
3941 34. Failure to deliver a bounce message always caused it to be frozen, even if
3942 there was an errors_to setting on the router. The errors_to setting is now
3945 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
3946 canonical form (fully expanded) before being placed in
3947 $sender_host_address.
3949 36. The table in the code that translates DNS record types into text (T_A to
3950 "A" for instance) was missing entries for NS and CNAME. It is just possible
3951 that this could have caused confusion if both these types were looked up
3952 for the same domain, because the text type is used as part of Exim's
3953 per-process caching. But the chance of anyone hitting this buglet seems
3956 37. The dnsdb lookup has been extended in a number of ways.
3958 (1) There is a new type, "zns", which walks up the domain tree until it
3959 finds some nameserver records. It should be used with care.
3961 (2) There is a new type, "mxh", which is like "mx" except that it returns
3962 just the host names, not the priorities.
3964 (3) It is now possible to give a list of domains (or IP addresses) to be
3965 looked up. The behaviour when one of the lookups defers can be
3966 controlled by a keyword.
3968 (4) It is now possible to specify the separator character for use when
3969 multiple records are returned.
3971 38. The dnslists ACL condition has been extended: it is now possible to supply
3972 a list of IP addresses and/or domains to be looked up in a particular DNS
3975 39. Added log_selector=+queue_time_overall.
3977 40. When running the queue in the test harness, wait just a tad after forking a
3978 delivery process, to get repeatability of debugging output.
3980 41. Include certificate and key file names in error message when GnuTLS fails
3981 to set them up, because the GnuTLS error message doesn't include the name
3982 of the failing file when there is a problem reading it.
3984 42. Allow both -bf and -bF in the same test run.
3986 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
3988 44. The "Exiscan patch" is now merged into the mainline Exim source.
3990 45. Sometimes the final signoff response after QUIT could fail to get
3991 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
3992 < 0 before doing a fflush(). This bug looks as though it goes back to the
3993 introduction of TLS in release 3.20, but "sometimes" must have been rare
3994 because the tests only now provoked it.
3996 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
3997 (this can affect the format of dates).
3999 47. exim_tidydb, when checking for the continued existence of a message for
4000 which it has found a message-specific retry record, was not finding
4001 messages that were in split spool directories. Consequently, it was
4002 deleting retry records that should have stayed in existence.
4004 48. Steve fixed some bugs in eximstats.
4006 49. The SPA authentication driver was not abandoning authentication and moving
4007 on to the next authenticator when an expansion was forced to fail,
4008 contradicting the general specification for all authenticators. Instead it
4009 was generating a temporary error. It now behaves as specified.
4011 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
4012 (the order specifies the preference for clients). The order is now AES256,
4013 AES128, 3DES, ARCFOUR128.
4015 51. Small patch to Sieve code - explicitly set From: when generating an
4018 52. Exim crashed if a remote delivery caused a very long error message to be
4019 recorded - for instance if somebody sent an entire SpamAssassin report back
4020 as a large number of 550 error lines. This bug was coincidentally fixed by
4021 increasing the size of one of Exim's internal buffers (big_buffer) that
4022 happened as part of the Exiscan merge. However, to be on the safe side, I
4023 have made the code more robust (and fixed the comments that describe what
4026 53. Now that there can be additional text after "Completed" in log lines (if
4027 the queue_time_overall log selector is set), a one-byte patch to exigrep
4028 was needed to allow it to recognize "Completed" as not the last thing in
4031 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
4032 patch that reportedly fixes this has been added. I am not expert enough to
4033 create a test for it. This is what the patch creator wrote:
4035 "I found a little strange behaviour of ldap code when working with
4036 Windows 2003 AD Domain, where users was placed in more than one
4037 Organization Units. When I tried to give exim partial DN, the exit code
4038 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
4039 But simultaneously result of request was absolutely normal ldap result,
4040 so I produce this patch..."
4042 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
4043 REFERENCE, so I have modified the code to exclude the patch when that macro
4046 55. Some experimental protocols are using DNS PTR records for new purposes. The
4047 keys for these records are domain names, not reversed IP addresses. The
4048 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4049 leaves it alone. Component reversal etc. now happens only for IP addresses.
4052 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
4054 57. Double the size of the debug message buffer (to 2048) so that more of very
4055 long debug lines gets shown.
4057 58. The exicyclog utility now does better if the number of log files to keep
4058 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
4060 59. Two changes related to the smtp_active_hostname option:
4062 (1) $smtp_active_hostname is now available as a variable.
4063 (2) The default for smtp_banner uses $smtp_active_hostname instead
4064 of $primary_hostname.
4066 60. The host_aton() function is supposed to be passed a string that is known
4067 to be a valid IP address. However, in the case of IPv6 addresses, it was
4068 not checking this. This is a hostage to fortune. Exim now panics and dies
4069 if the condition is not met. A case was found where this could be provoked
4070 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4071 components; fortuitously, this particular loophole had already been fixed
4072 by change 4.50/55 above.
4074 If there are any other similar loopholes, the new check in host_aton()
4075 itself should stop them being exploited. The report I received stated that
4076 data on the command line could provoke the exploit when Exim was running as
4077 exim, but did not say which command line option was involved. All I could
4078 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4079 running as the user.
4082 61. There was a buffer overflow vulnerability in the SPA authentication code
4083 (which came originally from the Samba project). I have added a test to the
4084 spa_base64_to_bits() function which I hope fixes it.
4087 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
4088 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
4090 63. The daemon start-up calls getloadavg() while still root for those OS that
4091 need the first call to be done as root, but it missed one case: when
4092 deliver_queue_load_max is set with deliver_drop_privilege. This is
4093 necessary for the benefit of the queue runner, because there is no re-exec
4094 when deliver_drop_privilege is set.
4096 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
4097 This has been fixed.
4099 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4100 was in use, was not putting the data itself into the right store pool;
4101 consequently, it could be overwritten for a subsequent message in the same
4102 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4105 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
4107 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
4108 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
4109 the function were treating the return as a boolean value, which happened to
4110 work because 0=false and not-0=true, but is not correct code.
4112 68. The host_aton() function was not handling scoped IPv6 addresses (those
4113 with, for example, "%eth0" on the end) correctly.
4115 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
4116 negated items (that is, ~something) in unsigned ints. Some compilers
4117 apparently mutter when there is no cast.
4119 70. If an address verification called from an ACL failed, and did not produce a
4120 user-specific message (i.e. there was only a "system" message), nothing was
4121 put in $acl_verify_message. In this situation, it now puts the system
4124 71. Change 4.23/11 added synchronization checking at the start of an SMTP
4125 session; change 4.31/43 added the unwanted input to the log line - except
4126 that it did not do this in the start of session case. It now does.
4128 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
4129 This is wrong and can cause the other end to generate a synchronization
4130 error if it is another Exim or anything else that does the synchronization
4131 check. A QUIT command is no longer sent after a timeout.
4133 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
4134 during host lookups.
4136 74. The defer_ok option of callout verification was not working if it was used
4137 when verifying addresses in header lines, that is, for this case:
4139 verify = header_sender/callout=defer_ok
4141 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
4142 those file descriptors could be used for SMTP connections. If anything
4143 wrote to stderr (the example that came up was "warn" in embedded Perl), it
4144 could be sent to the SMTP client, causing chaos. The daemon now opens
4145 stdin, stdout, and stderr to /dev/null when it puts itself into the
4148 76. Arrange for output from Perl's "warn" command to be written to Exim's main
4149 log by default. The user can override this with suitable Perl magic.
4151 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
4152 the log message when discard triggers, was not working for the DATA ACL or
4153 for the non-SMTP ACL.
4155 78. Error message wording change in sieve.c.
4157 79. If smtp_accept_max_per_host was set, the number of connections could be
4158 restricted to fewer than expected, because the daemon was trying to set up
4159 a new connection before checking whether the processes handling previous
4160 connections had finished. The check for completed processes is now done
4161 earlier. On busy systems, this bug wouldn't be noticed because something
4162 else would have woken the daemon, and it would have reaped the completed
4165 80. If a message was submitted locally by a user whose login name contained one
4166 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
4167 It caused a spool format error. I have fixed the spool reading code. A
4168 related problem was that the "from" clause in the Received: line became
4169 illegal because of the space(s). It is now covered by ${quote_local_part.
4171 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
4174 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
4176 83. Patch from Sieve maintainer to fix unterminated string problem in
4177 "vacation" handling.
4179 84. Some minor changes to the Linux configuration files to help with other
4180 OS variants using glibc.
4182 85. One more patch for Sieve to update vacation handling to latest spec.
4185 ----------------------------------------------------
4186 See the note above about the 4.44 and 4.50 releases.
4187 ----------------------------------------------------
4193 1. Change 4.43/35 introduced a bug that caused file counts to be
4194 incorrectly computed when quota_filecount was set in an appendfile
4197 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4198 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4201 3. Give more explanation in the error message when the command for a transport
4202 filter fails to execute.
4204 4. There are several places where Exim runs a non-Exim command in a
4205 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4206 was being done only for the command run by the queryprogram router. It is
4207 now done for all such subprocesses. The other cases are: ${run, transport
4208 filters, and the commands run by the lmtp and pipe transports.
4210 5. Some older OS have a limit of 256 on the maximum number of file
4211 descriptors. Exim was using setrlimit() to set 1000 as a large value
4212 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4213 systems. I've change it so that if it can't get 1000, it tries for 256.
4215 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4216 was an oversight, and furthermore, ever since the addition of extra
4217 controls (e.g. 4.43/32), the checks on when to allow different forms of
4218 "control" were broken. There should now be diagnostics for all cases when a
4219 control that does not make sense is encountered.
4221 7. $recipients is now available in the predata ACL (oversight).
4223 8. Tidy the search cache before the fork to do a delivery from a message
4224 received from the command line. Otherwise the child will trigger a lookup
4225 failure and thereby defer the delivery if it tries to use (for example) a
4226 cached ldap connection that the parent has called unbind on.
4228 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4229 of $address_data from the recipient verification was clobbered by the
4230 sender verification.
4232 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4233 was its contents. (It was OK if the option was not defined at all.)
4235 11. A "Completed" log line is now written for messages that are removed from
4236 the spool by the -Mrm option.
4238 12. $host_address is now set to the target address during the checking of
4239 ignore_target_hosts.
4241 13. When checking ignore_target_hosts for an ipliteral router, no host name was
4242 being passed; this would have caused $sender_host_name to have been used if
4243 matching the list had actually called for a host name (not very likely,
4244 since this list is usually IP addresses). A host name is now passed as
4247 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4248 code that specifies a non-restarting handler (typically sigaction() in
4249 modern systems) in an attempt to fix a rare and obscure crash bug.
4251 15. Narrowed the window for a race in the daemon that could cause it to ignore
4252 SIGCHLD signals. This is not a major problem, because they are used only to
4253 wake it up if nothing else does.
4255 16. A malformed maildirsize file could cause Exim to calculate negative values
4256 for the mailbox size or file count. Odd effects could occur as a result.
4257 The maildirsize information is now recalculated if the size or filecount
4260 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4261 support for a long time. Removed HAVE_SYS_VFS_H.
4263 18. Updated exipick to current release from John Jetmore.
4265 19. Allow an empty sender to be matched against a lookup in an address list.
4266 Previously the only cases considered were a regular expression, or an
4269 20. Exim went into a mad DNS lookup loop when doing a callout where the
4270 host was specified on the transport, if the DNS lookup yielded more than
4273 21. The RFC2047 encoding function was originally intended for short strings
4274 such as real names; it was not keeping to the 75-character limit for
4275 encoded words that the RFC imposes. It now respects the limit, and
4276 generates multiple encoded words if necessary. To be on the safe side, I
4277 have increased the buffer size for the ${rfc2047: expansion operator from
4280 22. Failure to deliver a bounce message always caused it to be frozen, even if
4281 there was an errors_to setting on the router. The errors_to setting is now
4284 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4285 canonical form (fully expanded) before being placed in
4286 $sender_host_address.
4288 24. Updated eximstats to version 1.33
4290 25. Include certificate and key file names in error message when GnuTLS fails
4291 to set them up, because the GnuTLS error message doesn't include the name
4292 of the failing file when there is a problem reading it.
4294 26. Expand error message when OpenSSL has problems setting up cert/key files.
4297 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
4298 (this can affect the format of dates).
4300 28. exim_tidydb, when checking for the continued existence of a message for
4301 which it has found a message-specific retry record, was not finding
4302 messages that were in split spool directories. Consequently, it was
4303 deleting retry records that should have stayed in existence.
4305 29. eximstats updated to version 1.35
4306 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
4307 1.35 - bugfix such that pie charts by volume are generated correctly
4309 30. The SPA authentication driver was not abandoning authentication and moving
4310 on to the next authenticator when an expansion was forced to fail,
4311 contradicting the general specification for all authenticators. Instead it
4312 was generating a temporary error. It now behaves as specified.
4314 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
4315 (the order specifies the preference for clients). The order is now AES256,
4316 AES128, 3DES, ARCFOUR128.
4318 31. Small patch to Sieve code - explicitly set From: when generating an
4321 32. Exim crashed if a remote delivery caused a very long error message to be
4322 recorded - for instance if somebody sent an entire SpamAssassin report back
4323 as a large number of 550 error lines. This bug was coincidentally fixed by
4324 increasing the size of one of Exim's internal buffers (big_buffer) that
4325 happened as part of the Exiscan merge. However, to be on the safe side, I
4326 have made the code more robust (and fixed the comments that describe what
4329 33. Some experimental protocols are using DNS PTR records for new purposes. The
4330 keys for these records are domain names, not reversed IP addresses. The
4331 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4332 leaves it alone. Component reversal etc. now happens only for IP addresses.
4335 34. The host_aton() function is supposed to be passed a string that is known
4336 to be a valid IP address. However, in the case of IPv6 addresses, it was
4337 not checking this. This is a hostage to fortune. Exim now panics and dies
4338 if the condition is not met. A case was found where this could be provoked
4339 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4340 components; fortuitously, this particular loophole had already been fixed
4341 by change 4.50/55 or 4.44/33 above.
4343 If there are any other similar loopholes, the new check in host_aton()
4344 itself should stop them being exploited. The report I received stated that
4345 data on the command line could provoke the exploit when Exim was running as
4346 exim, but did not say which command line option was involved. All I could
4347 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4348 running as the user.
4351 35. There was a buffer overflow vulnerability in the SPA authentication code
4352 (which came originally from the Samba project). I have added a test to the
4353 spa_base64_to_bits() function which I hope fixes it.
4356 36. The daemon start-up calls getloadavg() while still root for those OS that
4357 need the first call to be done as root, but it missed one case: when
4358 deliver_queue_load_max is set with deliver_drop_privilege. This is
4359 necessary for the benefit of the queue runner, because there is no re-exec
4360 when deliver_drop_privilege is set.
4362 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4363 was in use, was not putting the data itself into the right store pool;
4364 consequently, it could be overwritten for a subsequent message in the same
4365 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4368 38. Sometimes the final signoff response after QUIT could fail to get
4369 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4370 < 0 before doing a fflush(). This bug looks as though it goes back to the
4371 introduction of TLS in release 3.20, but "sometimes" must have been rare
4372 because the tests only now provoked it.
4378 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
4379 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
4380 Now it returns an integer. A number of calls were still expecting a T/F
4381 return. Fortuitously, in all cases, the tests worked in OK situations,
4382 which is the norm. However, things would have gone wrong on any write
4383 failures on the smtp file descriptor. This function is used when sending
4384 messages over SMTP and also when doing verify callouts.
4386 2. When Exim is called to do synchronous delivery of a locally submitted
4387 message (the -odf or -odi options), it no longer closes stderr before doing
4390 3. Implemented the mua_wrapper option.
4392 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
4394 5. Implemented the functions header_remove(), header_testname(),
4395 header_add_at_position(), and receive_remove_recipient(), and exported them
4398 6. If an ACL "warn" statement specified the addition of headers, Exim already
4399 inserted X-ACL-Warn: at the start if there was no header name. However, it
4400 was not making this test for the second and subsequent header lines if
4401 there were newlines in the string. This meant that an invalid header could
4402 be inserted if Exim was badly configured.
4404 7. Allow an ACL "warn" statement to add header lines at the start or after all
4405 the Received: headers, as well as at the end.
4407 8. Added the rcpt_4xx retry error code.
4409 9. Added postmaster_mailfrom=xxx to callout verification option.
4411 10. Added mailfrom=xxxx to the callout verification option, for verify=
4414 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
4415 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
4417 12. Inserted some casts to stop certain compilers warning when using pointer
4418 differences as field lengths or precisions in printf-type calls (mostly
4419 affecting debugging statements).
4421 13. Added optional readline() support for -be (dynamically loaded).
4423 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
4424 same clock tick as a message's arrival, so that its received time was the
4425 same as the "first fail" time on the retry record, and that message
4426 remained on the queue past the ultimate address timeout, every queue runner
4427 would try a delivery (because it was past the ultimate address timeout) but
4428 after another failure, the ultimate address timeout, which should have then
4429 bounced the address, did not kick in. This was a "< instead of <=" error;
4430 in most cases the first failure would have been in the next clock tick
4431 after the received time, and all would be well.
4433 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
4434 being recognized when the domain list was tested by the match_domain
4435 condition in an expansion string.
4437 16. Added the ${str2b64: operator.
4439 17. Exim was always calling setrlimit() to set a large limit for the number of
4440 processes, without checking whether the existing limit was already
4441 adequate. (It did check for the limit on file descriptors.) Furthermore,
4442 errors from getrlimit() and setrlimit() were being ignored. Now they are
4443 logged to the main and panic logs, but Exim does carry on, to try to do its
4444 job under whatever limits there are.
4446 18. Imported PCRE 5.0.
4448 19. Trivial typo in log message " temporarily refused connection" (the leading
4451 20. If the log selector return_path_on_delivery was set and an address was
4452 redirected to /dev/null, the delivery process crashed because it assumed
4453 that a return path would always be set for a "successful" delivery. In this
4454 case, the whole delivery is bypassed as an optimization, and therefore no
4457 21. Internal re-arrangement: the function for sending a challenge and reading
4458 a response while authentication was assuming a zero-terminated challenge
4459 string. It's now changed to take a pointer and a length, to allow for
4460 binary data in such strings.
4462 22. Added the cyrus_sasl authenticator (code supplied by MBM).
4464 23. Exim was not respecting finduser_retries when seeking the login of the
4465 uid under which it was called; it was always trying 10 times. (The default
4466 setting of finduser_retries is zero.) Also, it was sleeping after the final
4467 failure, which is pointless.
4469 24. Implemented tls_on_connect_ports.
4471 25. Implemented acl_smtp_predata.
4473 26. If the domain in control=submission is set empty, Exim assumes that the
4474 authenticated id is a complete email address when it generates From: or
4475 Sender: header lines.
4477 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
4478 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
4479 chown and chgrp in /bin and hostname in /usr/bin.
4481 28. Exim was keeping the "process log" file open after each use, just as it
4482 does for the main log. This opens the possibility of it remaining open for
4483 long periods when the USR1 signal hits a daemon. Occasional processlog
4484 errors were reported, that could have been caused by this. Anyway, it seems
4485 much more sensible not to leave this file open at all, so that is what now
4488 29. The long-running daemon process does not normally write to the log once it
4489 has entered its main loop, and it closes the log before doing so. This is
4490 so that log files can straightforwardly be renamed and moved. However,
4491 there are a couple of unusual error situations where the daemon does write
4492 log entries, and I had neglected to close the log afterwards.
4494 30. The text of an SMTP error response that was received during a remote
4495 delivery was being truncated at 512 bytes. This is too short for some of
4496 the long messages that one sometimes sees. I've increased the limit to
4499 31. It is now possible to make retry rules that apply only when a message has a
4500 specific sender, in particular, an empty sender.
4502 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
4503 it possible to be selective about when SMTP synchronization is enforced.
4505 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
4507 32. Implemented hosts_connection_nolog.
4509 33. Added an ACL for QUIT.
4511 34. Setting "delay_warning=" to disable warnings was not working; it gave a
4514 35. Added mailbox_size and mailbox_filecount to appendfile.
4516 36. Added control = no_multiline_responses to ACLs.
4518 37. There was a bug in the logic of the code that waits for the clock to tick
4519 in the case where the clock went backwards by a substantial amount such
4520 that the microsecond fraction of "now" was more than the microsecond
4521 fraction of "then" (but the whole seconds number was less).
4523 38. Added support for the libradius Radius client library this is found on
4524 FreeBSD (previously only the radiusclient library was supported).
4530 1. When certain lookups returned multiple values in the form name=value, the
4531 quoting of the values was not always being done properly. Specifically:
4532 (a) If the value started with a double quote, but contained no whitespace,
4534 (b) If the value contained whitespace other than a space character (i.e.
4535 tabs or newlines or carriage returns) it was not quoted.
4536 This fix has been applied to the mysql and pgsql lookups by writing a
4537 separate quoting function and calling it from the lookup code. The fix
4538 should probably also be applied to nisplus, ibase and oracle lookups, but
4539 since I cannot test any of those, I have not disturbed their existing code.
4541 2. A hit in the callout cache for a specific address caused a log line with no
4542 reason for rejecting RCPT. Now it says "Previous (cached) callout
4543 verification failure".
4545 3. There was an off-by-one bug in the queryprogram router. An over-long
4546 return line was truncated at 256 instead of 255 characters, thereby
4547 overflowing its buffer with the terminating zero. As well as fixing this, I
4548 have increased the buffer size to 1024 (and made a note to document this).
4550 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
4551 when Exim is waiting for an SMTP response from a remote server, Exim
4552 restarts its select() call on the socket, thereby resetting its timeout.
4553 This is not a problem when such interrupts are rare. Somebody set up a cron
4554 job to run exiwhat every 2 minutes, which is less than the normal select()
4555 timeout (5 or 10 minutes). This meant that the select() timeout never
4556 kicked in because it was always reset. I have fixed this by comparing the
4557 time when an interrupt arrives with the time at the start of the first call
4558 to select(). If more time than the timeout has elapsed, the interrupt is
4559 treated as a timeout.
4561 5. Some internal re-factoring in preparation for the addition of Sieve
4562 extensions (by MH). In particular, the "personal" test is moved to a
4563 separate function, and given an option for scanning Cc: and Bcc: (which is
4564 not set for Exim filters).
4566 6. When Exim created an email address using the login of the caller as the
4567 local part (e.g. when creating a From: or Sender: header line), it was not
4568 quoting the local part when it contained special characters such as @.
4570 7. Installed new OpenBSD configuration files.
4572 8. Reworded some messages for syntax errors in "and" and "or" conditions to
4573 try to make them clearer.
4575 9. Callout options, other than the timeout value, were being ignored when
4576 verifying sender addresses in header lines. For example, when using
4578 verify = header_sender/callout=no_cache
4580 the cache was (incorrectly) being used.
4582 10. Added a missing instance of ${EXE} to the exim_install script; this affects
4583 only the Cygwin environment.
4585 11. When return_path_on_delivery was set as a log selector, if different remote
4586 addresses in the same message used different return paths and parallel
4587 remote delivery occurred, the wrong values would sometimes be logged.
4588 (Whenever a remote delivery process finished, the return path value from
4589 the most recently started remote delivery process was logged.)
4591 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
4592 lines when AUTH and/or TLS are in use. This is the "received protocol"
4593 field. Exim used to use "asmtp" for authenticated SMTP, without any
4594 indication (in the protocol name) for TLS use. Now it follows the RFC and
4595 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
4596 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
4597 names appear in log lines as well as in Received: header lines.
4599 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
4600 extensions, and comparison tests, and to fix some bugs.
4602 14. Changes to the "personal" filter test:
4604 (1) The test was buggy in that it was just doing the equivalent of
4605 "contains" tests on header lines. For example, if a user's address was
4606 anne@some.where, the "personal" test would incorrectly be true for
4608 To: susanne@some.where
4610 This test is now done by extracting each address from the header in turn,
4611 and checking the entire address. Other tests that are part of "personal"
4612 are now done using regular expressions (for example, to check local parts
4613 of addresses in From: header lines).
4615 (2) The list of non-personal local parts in From: addresses has been
4616 extended to include "listserv", "majordomo", "*-request", and "owner-*",
4617 taken from the Sieve specification recommendations.
4619 (3) If the message contains any header line starting with "List-" it is
4620 treated as non-personal.
4622 (4) The test for "circular" in the Subject: header line has been removed
4623 because it now seems ill-conceived.
4625 15. Minor typos in src/EDITME comments corrected.
4627 16. Installed latest exipick from John Jetmore.
4629 17. If headers_add on a router specified a text string that was too long for
4630 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
4631 of string_sprintf() is now avoided.
4633 18. $message_body_size was not set (it was always zero) when running the DATA
4634 ACL and the local_scan() function.
4636 19. For the "mail" command in an Exim filter, no default was being set for
4637 the once_repeat time, causing a random time value to be used if "once" was
4638 specified. (If the value happened to be <= 0, no repeat happened.) The
4639 default is now 0s, meaning "never repeat". The "vacation" command was OK
4640 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
4641 (I found it when inspecting the code).
4643 20. There is now an overall timeout for performing a callout verification. It
4644 defaults to 4 times the callout timeout, which applies to individual SMTP
4645 commands during the callout. The overall timeout applies when there is more
4646 than one host that can be tried. The timeout is checked before trying the
4647 next host. This prevents very long delays if there are a large number of
4648 hosts and all are timing out (e.g. when the network connections are timing
4649 out). The value of the overall timeout can be changed by specifying an
4650 additional sub-option for "callout", called "maxwait". For example:
4652 verify = sender/callout=5s,maxwait=20s
4654 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
4655 to the end before writing, but this should make it even safer).
4657 22. Exim was forgetting that it had advertised PIPELINING for the second and
4658 subsequent messages on an SMTP connection. It was also not resetting its
4659 memory on STARTTLS and an internal HELO.
4661 23. When Exim logs an SMTP synchronization error within a session, it now
4662 records whether PIPELINING has been advertised or not.
4664 24. Added 3 instances of "(long int)" casts to time_t variables that were being
4665 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
4666 rather than long int.
4668 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
4670 26. Added the never_mail option to autoreply.
4676 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
4677 crash if the getsockname() call failed; this can happen if a connection is
4678 closed very soon after it is established. The problem was simply in the
4679 order in which certain operations were done, causing Exim to try to write
4680 to the SMTP stream before it had set up the file descriptor. The bug has
4681 been fixed by making things happen in the correct order.
4687 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
4688 before the connection was closed, thus losing the rejection response.
4690 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
4691 some early Solaris releases, but causes trouble in current releases where
4692 socklen_t is defined.
4694 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
4697 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
4700 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
4701 that releasing the top of it at the end releases what was used for sub-
4702 expansions (unless the block got too big). However, discard this block if
4703 the first thing is a variable or header, so that we can use its block when
4704 it is dynamic (useful for very large $message_headers, for example).
4706 6. Lookups now cache *every* query, not just the most recent. A new, separate
4707 store pool is used for this. It can be recovered when all lookup caches are
4708 flushed. Lookups now release memory at the end of their result strings.
4709 This has involved some general refactoring of the lookup sources.
4711 7. Some code has been added to the store_xxx() functions to reduce the amount
4712 of flapping under certain conditions.
4714 8. log_incoming_interface used to affect only the <= reception log lines. Now
4715 it causes the local interface and port to be added to several more SMTP log
4716 lines, for example "SMTP connection from", and rejection lines.
4718 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
4720 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
4722 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
4723 could be overwritten at the end of the current message (or the start of a
4724 new message if it was set in a HELO ACL). The value is now preserved for
4725 the duration of the SMTP connection.
4727 12. If a transport had a headers_rewrite setting, and a matching header line
4728 contained an unqualified address, that address was qualified, even if it
4729 did not match any rewriting rules. The underlying bug was that the values
4730 of the flags that permit the existence of unqualified sender and recipient
4731 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
4732 non-local messages, and by -bnq for local messages) were not being
4733 preserved with the message after it was received.
4735 13. When Exim was logging an SMTP synchronization error, it could sometimes log
4736 "next input=" as part of the text comprising the host identity instead of
4737 the correct text. The code was using the same buffer for two different
4738 strings. However, depending on which order the printing function evaluated
4739 its arguments, the bug did not always show up. Under Linux, for example, my
4740 test suite worked just fine.
4742 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
4743 doesn't work with some older versions of Perl. It has been changed to "my",
4744 which in any case is probably the better facility to use.
4746 15. A really picky compiler found some instances of statements for creating
4747 error messages that either had too many or two few arguments for the format
4750 16. The size of the buffer for calls to the DNS resolver has been increased
4751 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
4752 for addresses that have a lot of PTR records. This alleviates a problem; it
4753 does not fully solve it.
4755 17. A dnsdb lookup for PTR records that receives more data than will fit in the
4756 buffer now truncates the list and logs the incident, which is the same
4757 action as happens when Exim is looking up a host name and its aliases.
4758 Previously in this situation something unpredictable would happen;
4759 sometimes it was "internal error: store_reset failed".
4761 18. If a server dropped the connection unexpectedly when an Exim client was
4762 using GnuTLS and trying to read a response, the client delivery process
4763 crashed while trying to generate an error log message.
4765 19. If a "warn" verb in an ACL added multiple headers to a message in a single
4766 string, for example:
4768 warn message = H1: something\nH2: something
4770 the text was added as a single header line from Exim's point of view
4771 though it ended up OK in the delivered message. However, searching for the
4772 second and subsequent header lines using $h_h2: did not work. This has been
4773 fixed. Similarly, if a system filter added multiple headers in this way,
4774 the routers could not see them.
4776 20. Expanded the error message when iplsearch is called with an invalid key to
4777 suggest using net-iplsearch in a host list.
4779 21. When running tests using -bh, any delays imposed by "delay" modifiers in
4780 ACLs are no longer actually imposed (and a message to that effect is
4783 22. If a "gecos" field in a passwd entry contained escaped characters, in
4784 particular, if it contained a \" sequence, Exim got it wrong when building
4785 a From: or a Sender: header from that name. A second bug also caused
4786 incorrect handling when an unquoted " was present following a character
4787 that needed quoting.
4789 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
4790 was not being matched caselessly.
4792 24. Arranged for all hyphens in the exim.8 source to be escaped with
4795 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
4796 properly. Recipient callout cache records were still being keyed to include
4797 the sender, even when use_sender was set false. This led to far more
4798 callouts that were necessary. The sender is no longer included in the key
4799 when use_sender is false.
4801 26. Added "control = submission" modifier to ACLs.
4803 27. Added the ${base62d: operator to decode base 62 numbers.
4805 28. dnsdb lookups can now access SRV records.
4807 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
4808 the configuration file.
4810 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
4811 (-v) mode. This makes the output for a verbose queue run more intelligible.
4813 31. Added a use_postmaster feature to recipient callouts.
4815 32. Added the $body_zerocount variable, containing the number of binary zero
4816 bytes in the message body.
4818 33. The time of last modification of the "new" subdirectory is now used as the
4819 "mailbox time last read" when there is a quota error for a maildir
4822 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
4824 35. Added +ignore_unknown as a special item in host lists.
4826 36. Code for decoding IPv6 addresses in host lists is now included, even if
4827 IPv6 support is not being compiled. This fixes a bug in which an IPv6
4828 address was recognized as an IP address, but was then not correctly decoded
4829 into binary, causing unexpected and incorrect effects when compared with
4836 1. Very minor rewording of debugging text in manualroute to say "list of
4837 hosts" instead of "hostlist".
4839 2. If verify=header_syntax was set, and a header line with an unqualified
4840 address (no domain) and a large number of spaces between the end of the
4841 name and the colon was received, the reception process suffered a buffer
4842 overflow, and (when I tested it) crashed. This was caused by some obsolete
4843 code that should have been removed. The fix is to remove it!
4845 3. When running in the test harness, delay a bit after writing a bounce
4846 message to get a bit more predictability in the log output.
4848 4. Added a call to search_tidyup() just before forking a reception process. In
4849 theory, someone could use a lookup in the expansion of smtp_accept_max_
4850 per_host which, without the tidyup, could leave open a database connection.
4852 5. Added the variables $recipient_data and $sender_data which get set from a
4853 lookup success in an ACL "recipients" or "senders" condition, or a router
4854 "senders" option, similar to $domain_data and $local_part_data.
4856 6. Moved the writing of debug_print from before to after the "senders" test
4859 7. Change 4.31/66 (moving the time when the Received: is generated) caused
4860 problems for message scanning, either using a data ACL, or using
4861 local_scan() because the Received: header was not generated till after they
4862 were called (in order to set the time as the time of reception completion).
4863 I have revised the way this works. The header is now generated after the
4864 body is received, but before the ACL or local_scan() are called. After they
4865 are run, the timestamp in the header is updated.
4871 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
4872 before starting a queue runner without re-exec. This happened only when
4873 deliver_drop_privilege was set or when the Exim user was set to root. The
4874 effect of the bug was that timeouts during subsequent deliveries caused
4875 crashes instead of being properly handled. The handler is now left at its
4876 default (and expected) setting.
4878 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
4879 message, again when deliver_drop_privilege is set or Exim is run as root.
4880 The bug described in (1) was not present in this case, but the tidying up
4881 of the other signals was missing. I have made the two cases consistent.
4883 3. The ignore_target_hosts setting on a manualroute router was being ignored
4884 for hosts that were looked up using the /MX notation.
4886 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
4889 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
4890 operated on the sender address. After changing the $sender_address to <>
4891 for the sender address verify, Exim was re-instated it as the original
4892 (before rewriting) address, but remembering that it had rewritten it, so it
4893 wasn't rewriting it again. This bug also had the effect of breaking the
4894 sender address verification caching when the sender address was rewritten.
4896 6. The ignore_target_hosts option was being ignored by the ipliteral router.
4897 This has been changed so that if the ip literal address matches
4898 ignore_target_hosts, the router declines.
4900 7. Added expansion conditions match_domain, match_address, and match_local_
4901 part (NOT match_host).
4903 8. The placeholder for the Received: header didn't have a length field set.
4905 9. Added code to Exim itself and to exim_lock to test for a specific race
4906 condition that could lead to file corruption when using MBX delivery. The
4907 issue is with the lockfile that is created in /tmp. If this file is removed
4908 after a process has opened it but before that process has acquired a lock,
4909 there is the potential for a second process to recreate the file and also
4910 acquire a lock. This could lead to two Exim processes writing to the file
4911 at the same time. The added code performs the same test as UW imapd; it
4912 checks after acquiring the lock that its file descriptor still refers to
4913 the same named file.
4915 10. The buffer for building added header lines was of fixed size, 8192 bytes.
4916 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
4919 11. Added the smtp_active_hostname option. If used, this will typically be made
4920 to depend on the incoming interface address. Because $interface_address is
4921 not set up until the daemon has forked a reception process, error responses
4922 that can happen earlier (such as "too many connections") no longer contain
4925 12. If an expansion in a condition on a "warn" statement fails because a lookup
4926 defers, the "warn" statement is abandoned, and the next ACL statement is
4927 processed. Previously this caused the whole ACL to be aborted.
4929 13. Added the iplsearch lookup type.
4931 14. Added ident_timeout as a log selector.
4933 15. Added tls_certificate_verified as a log selector.
4935 16. Added a global option tls_require_ciphers (compare the smtp transport
4936 option of the same name). This controls incoming TLS connections.
4938 17. I finally figured out how to make tls_require_ciphers do a similar thing
4939 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
4940 before starting the TLS session.
4942 18. Tabs are now shown as \t in -bP output.
4944 19. If the log selector return_path_on_delivery was set, Exim crashed when
4945 bouncing a message because it had too many Received: header lines.
4947 20. If two routers both had headers_remove settings, and the first one included
4948 a superfluous trailing colon, the final name in the first list and the
4949 first name in the second list were incorrectly joined into one item (with a
4950 colon in the middle).
4956 1. Added -C and -D options to the exinext utility, mainly to make it easier
4957 to include in the automated testing, but these could be helpful when
4958 multiple configurations are in use.
4960 2. The exinext utility was not formatting the output nicely when there was
4961 an alternate port involved in the retry record key, nor when there was a
4962 message id as well (for retries that were specific to a specific message
4963 and a specific host). It was also confused by IPv6 addresses, because of
4964 the additional colons they contain. I have fixed the IPv4 problem, and
4965 patched it up to do a reasonable job for IPv6.
4967 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
4968 delivery, the log line now contains "pipelined" if PIPELINING was used.
4970 4. An SMTP transport process used to panic and die if the bind() call to set
4971 an explicit outgoing interface failed. This has been changed; it is now
4972 treated in the same way as a connect() failure.
4974 5. A reference to $sender_host_name in the part of a conditional expansion
4975 that was being skipped was still causing a DNS lookup. This no longer
4978 6. The def: expansion condition was not recognizing references to header lines
4979 that used bh_ and bheader_.
4981 7. Added the _cache feature to named lists.
4983 8. The code for checking quota_filecount in the appendfile transport was
4984 allowing one more file than it should have been.
4986 9. For compatibility with Sendmail, the command line option
4994 and sets the incoming protocol and host name (for trusted callers). The
4995 host name and its colon can be omitted when only the protocol is to be set.
4996 Note the Exim already has two private options, -pd and -ps, that refer to
4997 embedded Perl. It is therefore impossible to set a protocol value of "d" or
4998 "s", but I don't think that's a major issue.
5000 10. A number of refactoring changes to the code, none of which should affect
5003 (a) The number of logging options was getting close to filling up the
5004 32-bit word that was used as a bit map. I have split them into two classes:
5005 those that are passed in the argument to log_write(), and those that are
5006 only ever tested independently outside of that function. These are now in
5007 separate 32-bit words, so there is plenty of room for expansion again.
5008 There is no change in the user interface or the logging behaviour.
5010 (b) When building, for example, log lines, the code previously used a
5011 macro that called string_cat() twice, in order to add two strings. This is
5012 not really sufficiently general. Furthermore, there was one instance where
5013 it was actually wrong because one of the argument was used twice, and in
5014 one call a function was used. (As it happened, calling the function twice
5015 did not affect the overall behaviour.) The macro has been replaced by a
5016 function that can join an arbitrary number of extra strings onto a growing
5019 (c) The code for expansion conditions now uses a table and a binary chop
5020 instead of a serial search (which was left over from when there were very
5021 few conditions). Also, it now recognizes conditions like "pam" even when
5022 the relevant support is not compiled in: a suitably worded error message is
5023 given if an attempt is made to use such a condition.
5025 11. Added ${time_interval:xxxxx}.
5027 12. A bug was causing one of the ddress fields not to be passed back correctly
5028 from remote delivery subprocesses. The field in question was not being
5029 subsequently used, so this caused to problems in practice.
5031 13. Added new log selectors queue_time and deliver_time.
5033 14. Might have fixed a bug in maildirsizefile handling that threw up
5034 "unexpected character" debug warnings, and recalculated the data
5035 unnecessarily. In any case, I expanded the warning message to give more
5038 15. Added the message "Restricted characters in address" to the statements in
5039 the default ACL that block characters like @ and % in local parts.
5041 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
5042 Three changes have been made:
5044 (a) There was a serious bug; a negative response to MAIL caused the whole
5045 recipient domain to be cached as invalid, thereby blocking all messages
5046 to all local parts at the same domain, from all senders. This bug has
5047 been fixed. The domain is no longer cached after a negative response to
5048 MAIL if the sender used is not empty.
5050 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
5053 (c) A new callout option, "use_sender" has been added for people who want
5054 the modified behaviour.
5060 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
5063 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
5064 indeed breaks things for older releases.
5066 3. Added additional logging to the case where there is a problem reading data
5067 from a filter that is running in a subprocess using a pipe, in order to
5068 try to track down a specific problem.
5070 4. Testing facility fudge: when running in the test harness and attempting
5071 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
5072 getting "No route to host". Convert this to a timeout.
5074 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
5077 6. Some OS don't have socklen_t but use size_t instead. This affects the
5078 fifth argument of getsockopt() amongst other things. This is now
5079 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
5080 can be set for individual OS. I have set it for SunOS5, OSF1, and
5081 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
5082 some earlier ones do not.
5084 7. Change 4.30/15 was not doing the test caselessly.
5086 8. The standard form for an IPv6 address literal was being rejected by address
5087 parsing in, for example, MAIL and RCPT commands. An example of this kind of
5088 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5089 this, as well as the form without the "IPv6" on the front (but only when
5090 address literals are enabled, of course).
5092 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
5094 10. Exim crashed if a message with an empty sender address specified by -f
5095 encountered a router with an errors_to setting. This could be provoked only
5096 by a command such as
5100 where an empty string was supplied; "<>" did not hit this bug.
5102 11. Installed PCRE release 4.5.
5104 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
5105 remained set. It is now erased.
5107 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
5108 times from message ids (which are base 36 rather than the normal 62).
5110 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
5111 were being counted as actual protocol errors, and logged if the log
5112 selector +smtp_protocol_error was set. One cannot be perfect in this test,
5113 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
5114 and DATA following a set of rejected RCPTs do not count as protocol errors.
5115 In other words, Exim assumes they were pipelined, though this may not
5116 actually be the case. Of course, in all cases the client gets an
5117 appropriate error code.
5119 15. If a lookup fails in an ACL condition, a message about the failure may
5120 be available; it is used if testing the ACL cannot continue, because most
5121 such messages specify what the cause of the deferral is. However, some
5122 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
5123 that caused an old message to be retained and used if a later statement
5124 caused a defer, replacing the real cause of the deferral.
5126 16. If an IP address had so many PTR records that the DNS lookup buffer
5127 was not large enough to hold them, Exim could crash while trying to process
5128 the truncated data. It now detects and logs this case.
5130 17. Further to 4.21/58, another change has been made: if (and only if) the
5131 first line of a message (the first header line) ends with CRLF, a bare LF
5132 in a subsequent header line has a space inserted after it, so as not to
5133 terminate the header.
5135 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
5136 unnecessarily, used atoi() instead of strtol(), and didn't check the
5137 termination when getting file sizes from file names by regex.
5139 19. Completely re-implemented the support for maildirsize files, in the light
5140 of a number of problems with the previous contributed implementation
5141 (4.30/29). In particular:
5143 . If the quota is zero, the maildirsize file is maintained, but no quota is
5146 . If the maildir directory does not exist, it is created before any attempt
5147 to write a maildirsize file.
5149 . The quota value in the file is just a cache; if the quota is changed in
5150 the transport, the new value overrides.
5152 . A regular expression is available for excluding directories from the
5155 20. The autoreply transport checks the characters in options that define the
5156 message's headers; it allows continued headers, but it was checking with
5157 isspace() after an embedded newline instead of explicitly looking for a
5160 21. If all the "regular" hosts to which an address was routed had passed their
5161 expiry times, and had not reached their retry times, the address was
5162 bounced, even if fallback hosts were defined. Now Exim should go on to try
5165 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
5166 equivalent code in the SMTP transport. Some hosts send humungous responses
5167 to HELO/EHLO, more than 1024 it seems.
5169 23. Refactoring: code in filter.c used (void *) for "any old type" but this
5170 gives compiler warnings in some environments. I've now done it "properly",
5173 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
5174 (because of problems with the built-in one) was declared to return uschar *
5175 instead of char *, causing compiler failure.
5177 25. Fixed a file descriptor leak when processing alias/forward files.
5179 26. Fixed a minor format string issue in dbfn.c.
5181 27. Typo in exim.c: ("dmbnz" for "dbmnz").
5183 28. If a filter file refered to $h_xxx or $message_headers, and the headers
5184 contained RFC 2047 "words", Exim's memory could, under certain conditions,
5187 29. When a sender address is verified, it is cached, to save repeating the test
5188 when there is more than one recipient in a message. However, when the
5189 verification involves a callout, it is possible for different callout
5190 options to be set for different recipients. It is too complicated to keep
5191 track of this in the cache, so now Exim always runs a verification when a
5192 callout is required, relying on the callout cache for the optimization.
5193 The overhead is duplication of the address routing, but this should not be
5196 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
5197 to be verified with callout=postmaster, and the main callout worked but the
5198 postmaster check failed, the verification correctly failed. However, if a
5199 subsequent RCPT command asked for sender verification *without* the
5200 postmaster check, incorrect caching caused this verification also to fail,
5203 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
5204 it was not caching the DNS options (qualify_single, search_parents) that
5205 were used when the lookup failed. A subsequent lookup with different
5206 options therefore always gave the same answer, though there were cases
5207 where it should not have. (Example: a "domains = !$mx_any" option on a
5208 dnslookup router: the "domains" option is always processed without any
5209 widening, but the router might have qualify_single set.) Now Exim uses the
5210 cached value only when the same options are set.
5212 32. Added John Jetmore's "exipick" utility to the distribution.
5214 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
5215 than a timeout (e.g. a certificate is required, and is not provided), an
5216 Exim server now closes the connection immediately. Previously it waited for
5217 the client to close - but if the client is SSL, it seems that they each
5218 wait for each other, leading to a delay before one of them times out.
5220 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
5221 maintained 0.8.x compatibility because I don't think many are using it, and
5222 it is clearly obsolete.
5224 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
5227 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
5228 client certificate was expired. A simple patch fixes this, though I don't
5229 understand the full logic of why the verify callback is called multiple
5232 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
5233 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
5234 which causes problems with some clients (such as the Certicom SSL Plus
5235 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
5236 disables the coutermeasure allowing Eudora to connect."
5238 38. Exim was not checking that a write() to a log file succeeded. This could
5239 lead to Bad Things if a log got too big, in particular if it hit a file
5240 size limit. Exim now panics and dies if it cannot write to a log file, just
5241 as it does if it cannot open a log file.
5243 39. Modified OS/Makefile-Linux so that it now contains
5245 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
5247 The two -D definitions ensure that Exim is compiled with large file
5248 support, which makes it possible to handle log files that are bigger than
5251 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
5252 instance) a domain was checked against a named list that involved a lookup,
5253 causing $domain_data to be set, then another domain was checked against the
5254 same list, then the first domain was re-checked, the value of $domain_data
5255 after the final check could be wrong. In particular, if the second check
5256 failed, it could be set empty. This bug probably also applied to
5259 41. The strip_trailing_dot option was not being applied to the address given
5260 with the -f command-line option.
5262 42. The code for reading a message's header from the spool was incrementing
5263 $received_count, but never initializing it. This meant that the value was
5264 incorrect (doubled) while delivering a message in the same process in which
5265 it was received. In the most common configuration of Exim, this never
5266 happens - a fresh exec is done - but it can happen when
5267 deliver_drop_privilege is set.
5269 43. When Exim logs an SMTP synchronization error - client data sent too soon -
5270 it now includes up to 150 characters of the unexpected data in the log
5273 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
5274 and building data strings. The size of both of these buffers was 10 000
5275 bytes - far larger than anybody would *ever* want, thought I. Needless to
5276 say, somebody hit the limit. I have increased the maximum line length to
5277 20 000 and the maximum data length of concatenated lines to 100 000. I have
5278 also fixed two bugs, because there was no checking on these buffers. Tsk,
5279 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
5280 buffer is too small.
5282 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
5283 lsearch lookups. Now it does.
5285 46. When parsing a route_list item in a manualroute router, a fixed-length
5286 buffer was used for the list of hosts. I made this 1024 bytes long,
5287 thinking that nobody would ever have a list of hosts that long. Wrong.
5288 Somebody had a whole pile of complicated expansion conditions, and the
5289 string was silently truncated, leading to an expansion error. It turns out
5290 that it is easier to change to an unlimited length (owing to other changes
5291 that have happened since this code was originally written) than to build
5292 structure for giving a limitation error. The length of the item that
5293 expands into the list of hosts is now unlimited.
5295 47. The lsearch lookup could not handle data where the length of text line was
5296 more than 4095 characters. Such lines were truncated, leading to shortened
5297 data being returned. It should now handle lines of any length.
5299 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
5300 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
5303 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
5304 replacing strings such as BIN_DIRECTORY in a source file: the replacement
5305 no longer happens in comment lines. A list of replacements is now placed
5306 at the head of all of the source files, except those whose only change is
5307 to replace PERL_COMMAND in the very first #! line.
5309 50. Replaced the slow insertion sort in queue.c, for sorting the list of
5310 messages on the queue, with a bottom-up merge sort, using code contributed
5311 by Michael Haardt. This should make operations like -bp somewhat faster on
5312 large queues. It won't affect queue runners, except when queue_run_in_order
5315 51. Installed eximstats 1.31 in the distribution.
5317 52. Added support for SRV lookups to the dnslookup router.
5319 53. If an ACL referred to $message_body or $message_body_end, the value was not
5320 reset for any messages that followed in the same SMTP session.
5322 54. The store-handling optimization for building very long strings was not
5323 differentiating between the different store pools. I don't think this
5324 actually made any difference in practice, but I've tidied it.
5326 55. While running the routers to verify a sender address, $sender_address
5327 was still set to the sender address. This is wrong, because when routing to
5328 send a bounce to the sender, it would be empty. Therefore, I have changed
5329 it so that, while verifying a sender address, $sender_address is set to <>.
5330 (There is no change to what happens when verifying a recipient address.)
5332 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
5333 target A or AAAA records (if not already returned) without resetting the
5334 qualify_single or search_parents options of the DNS resolver. These are
5335 inappropriate in this case because the targets of MX and SRV records must
5336 be FQDNs. A broken DNS record could cause trouble if it happened to have a
5337 target that, when qualified, matched something in the local domain. These
5338 two options are now turned off when doing these lookups.
5340 57. It seems that at least some releases of Reiserfs (which does not have the
5341 concept of a fixed number of inodes) returns zero and not -1 for the
5342 number of available inodes. This interacted badly with check_spool_inodes,
5343 which assumed that -1 was the "no such thing" setting. What I have done is
5344 to check that the total number of inodes is greater than zero before doing
5345 the test of how many are available.
5347 58. When a "warn" ACL statement has a log_message modifier, the message is
5348 remembered, and not repeated. This is to avoid a lot of repetition when a
5349 message has many recipients that cause the same warning to be written.
5350 Howewer, Exim was preserving the list of already written lines for an
5351 entire SMTP session, which doesn't seem right. The memory is now reset if a
5352 new message is started.
5354 59. The "rewrite" debugging flag was not showing the result of rewriting in the
5355 debugging output unless log_rewrite was also set.
5357 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
5358 of (int)(handle) when we know that handle contains (void *)(-1).
5360 61. The Exim daemon panic-logs an error return when it closes the incoming
5361 connection. However "connection reset by peer" seems to be common, and
5362 isn't really an error worthy of noting specially, so that particular error
5365 62. When Exim is trying to find all the local interfaces, it used to panic and
5366 die if the ioctl to get the interface flags failed. However, it seems that
5367 on at least one OS (Solaris 9) it is possible to have an interface that is
5368 included in the list of interfaces, but for which you get a failure error
5369 for this call. This happens when the interface is not "plumbed" into a
5370 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
5371 failure of the "get flags" call assumes that the interface is down.
5373 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
5374 makes life easier for people who are doing arithmetic on fields extracted
5375 from dates, where you often get leading zeros that should not be
5376 interpreted as octal.
5378 64. Added qualify_domain to the redirect router, to override the global
5381 65. If a pathologically long header line contained very many addresses (the
5382 report of this problem mentioned 10 000) and each of them was rewritten,
5383 Exim could use up a very large amount of memory. (It kept on making new
5384 copies of the header line as it rewrote, and never released the old ones.)
5385 At the expense of a bit more processing, the header rewriting function has
5386 been changed so that it no longer eats memory in this way.
5388 66. The generation of the Received: header has been moved from the time that a
5389 message starts to be received, to the time that it finishes. The timestamp
5390 in the Received: header should now be very close to that of the <= log
5391 line. There are two side-effects of this change:
5393 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
5394 logged header lines no longer include the local Received: line, because
5395 it has not yet been created. The same applies to a copy of the message
5396 that is returned to a non-SMTP sender when a message is rejected.
5398 (b) When a filter file is tested using -bf, no additional Received: header
5399 is added to the test message. After some thought, I decided that this
5402 This change does not affect the value of $received_for. It is still set
5403 after address rewriting, but before local_scan() is called.
5405 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
5407 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
5408 gave an unhelpful panic error message, and a defer error. I have managed to
5409 change this behaviour so that it now rejects any supplied certificate,
5410 which seems right, as the list of acceptable certificates is empty.
5412 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
5413 gave an unhelpful defer error. I have not managed to make this reject any
5414 supplied certificates, but the error message it gives is "no certificate
5415 supplied", which is not helpful.
5417 70. exigrep's output now also includes lines that are not associated with any
5418 message, but which match the given pattern. Implemented by a patch from
5419 Martin Sluka, which also tidied up the Perl a bit.
5421 71. Recipient callout verification, like sender verification, was using <> in
5422 the MAIL FROM command. This isn't really the right thing, since the actual
5423 sender may affect whether the remote host accepts the recipient or not. I
5424 have changed it to use the actual sender in the callout; this means that
5425 the cache record is now keyed on a recipient/sender pair, not just the
5426 recipient address. There doesn't seem to be a real danger of callout loops,
5427 since a callout by the remote host to check the sender would use <>.
5428 [SEE ABOVE: changed after hitting problems.]
5430 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
5431 temporary errors. However, in the case of such a code being given after
5432 the end of a data transmission (i.e. after ".") Exim was failing to write
5433 a retry record for the message. (Yes, there was some broken host that was
5434 actually sending 8xx at this point.)
5436 73. An unknown lookup type in a host list could cause Exim to panic-die when
5437 the list was checked. (An example that provoked this was putting <; in the
5438 middle of a list instead of at the start.) If this happened during a DATA
5439 ACL check, a -D file could be left lying around. This kind of configuration
5440 error no longer causes Exim to die; instead it causes a defer errror. The
5441 incident is still logged to the main and panic logs.
5443 74. Buglet left over from Exim 3 conversion. The message "too many messages
5444 in one connection" was written to the rejectlog but not the mainlog, except
5445 when address rewriting (yes!) was being logged.
5447 75. Added write_rejectlog option.
5449 76. When a system filter was run not as root (that is, when system_filter_user
5450 was set), the values of the $n variables were not being returned to the
5451 main process; thus, they were not subsequently available in the $sn
5454 77. Added +return_path_on_delivery log selector.
5456 78. A connection timeout was being treated differently from recipients deferred
5457 when testing hosts_max_try with a message that was older than the host's
5458 retry timeout. (The host should not be counted, thus allowing all hosts to
5459 be tried at least once before bouncing.) This may have been the cause of an
5460 occasionally reported bug whereby a message would remain on the queue
5461 longer than the retry timeout, but would be bounced if a delivery was
5462 forced. I say "may" because I never totally pinned down the problem;
5463 setting up timeout/retry tests is difficult. See also the next item.
5465 79. The ultimate address timeout was not being applied to errors that involved
5466 a combination of host plus message (for example, a timeout on a MAIL
5467 command). When an address resolved to a number of possible hosts, and they
5468 were not all tried for each delivery (e.g. because of hosts_max_try), a
5469 message could remain on the queue longer than the retry timeout.
5471 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
5474 81. Fixed an obscure SMTP outgoing bug which required at least the following
5475 conditions: (a) there was another message waiting for the same server;
5476 (b) the server returned 5xx to all RCPT commands in the first message so
5477 that the message was not completed; (c) the server dropped the connection
5478 or gave a negative response to the RSET that Exim sends to abort the
5479 transaction. The observed case was a dropped connection after DATA that had
5480 been sent in pipelining mode. That is, the server had advertised PIPELINING
5481 but was not implementing it correctly. The effect of the bug was incorrect
5482 behaviour, such as trying another host, and this could lead to a crash.
5488 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
5489 and daemon.c were passed as pointers to ints; they should have been
5490 pointers to socklen_t variables (which are typically unsigned ints).
5492 2. Some signed/unsigned type warnings in the os.c file for Linux have been
5495 3. Fixed a really odd bug that affected only the testing scheme; patching a
5496 certain fixed string in the binary changed the value of another string that
5497 happened to be identical to the end of the original first string.
5499 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
5500 name", it returns that address as the IP address. On some operating
5501 systems (e.g. Solaris), it also passes back the IP address string as the
5502 "host name". However, on others (e.g. Linux), it passes back an empty
5503 string. Exim wasn't checking for this, and was changing the host name to an
5504 empty string, assuming it had been canonicized.
5506 5. Although rare, it is permitted to have more than one PTR record for a given
5507 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
5508 all the names associated with an address, because they do in Solaris.
5509 However, it seems that they do not in Linux for data that comes from the
5510 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
5511 I found this out when I moved to a new Linux workstation and tried to run
5512 the Exim test suite.
5514 To get round this problem I have changed the code so that it now does its
5515 own call to the DNS to look up PTR records when searching for a host name.
5516 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
5517 addresses that are only in /etc/hosts are still found.
5519 This behaviour is, however, controlled by an option called host_lookup_
5520 order, which defaults to "bydns:byaddr". If people want to use the other
5521 order, or indeed, just use one or the other means of lookup, they can
5522 specify it in this variable.
5524 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
5525 some operating systems, this comes back from gethostbyaddr() as an empty
5526 string, and this is what Exim used to test for. However, it seems that in
5527 other systems, "." is yielded. Exim now tests for this case too.
5529 7. The values of check_spool_space and check_log_space are now held internally
5530 as a number of kilobytes instead of an absolute number of bytes. If a
5531 numbers is specified without 'K' or 'M', it is rounded up to the nearest
5532 kilobyte. This means that much larger values can be stored.
5534 8. Exim monitor: an attempt to get the action menu when not actually pointing
5535 at a message produces an empty menu entitled "No message selected". This
5536 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
5537 no entries in it ("Shell widget menu has zero width and/or height"). So I
5538 have added a single, blank menu entry in this case.
5540 9. Added ${quote_local_part.
5542 10. MIME decoding is now applied to the contents of Subject: header lines when
5545 11. Now that a reference to $sender_host_address automatically causes a reverse
5546 lookup to occur if necessary (4.13/18), there is no need to arrange for a
5547 host lookup before query-style lookups in lists that might use this
5548 variable. This has therefore been abolished, and the "net-" prefix is no
5549 longer necessary for query-style lookups.
5551 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
5552 have been a typo for LFLAGS, so it has been changed.
5554 13. The install script calls Exim with "-C /dev/null" in order to find the
5555 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
5556 to be output. Howeve, since Exim outputs its version number before the
5557 error, it didn't break the script. It just looked ugly. I fixed this by
5558 always allowing "-C /dev/null" if the caller is root.
5560 14. Ignore overlarge ACL variable number when reading spool file - insurance
5561 against a later release with more variables having written the file.
5563 15. The standard form for an IPv6 address literal was being rejected by EHLO.
5564 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5565 this, as well as the form without the "IPv6" on the front.
5567 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
5568 OS/Makefile-Darwin file.
5570 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
5571 only to LDAP libraries that do not have LDAP_OPT_DEREF.
5573 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
5574 variable. However, some OS use ints for inodes. I've added cast to long int
5575 to get rid of the compiler warning.
5577 19. I had forgotten to lock out "/../" in configuration file names when
5578 ALT_CONFIG_PREFIX was set.
5580 20. Routers used for verification do not need to specify transports. However,
5581 if such a router generated a host list, and callout was configured, Exim
5582 crashed, because it could not find a port number from the (non-existent)
5583 transport. It now assumes port 25 in this circumstance.
5585 21. Added the -t option to exigrep.
5587 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
5588 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
5589 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
5590 them is set, LOOKUP_LSEARCH is forced.
5592 23. "exim -bV" now outputs a list of lookups that are included in the binary.
5594 24. Added sender and host information to the "rejected by local_scan()" log
5595 line; previously there was no indication of these.
5597 25. Added .include_if_exists.
5599 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
5600 Linux. It turns out that not all file systems support this. Apparently some
5601 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
5602 do it.) To cope with this, the error EINVAL, which means that sync-ing is
5603 not supported on the file descriptor, is now ignored when Exim is trying to
5604 sync a directory. This applies only to Linux.
5606 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
5608 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
5609 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
5612 29. Added support for maildirsize files from supplied patch (modified a bit).
5614 30. The use of :fail: followed by an empty string could lead Exim to respond to
5615 sender verification failures with (e.g.):
5617 550 Verification failed for <xxx>
5618 550 Sender verify failed
5620 where the first response line was missing the '-' that indicates it is not
5621 the final line of the response.
5623 31. The loop for finding the name of the user that called Exim had a hardwired
5624 limit of 10; it now uses the value of finduser_retries, which is used for
5625 all other user lookups.
5627 32. Added $received_count variable, available in data and not_smtp ACLs, and at
5630 33. Exim was neglecting to zero errno before one call of strtol() when
5631 expanding a string and expecting an integer value. On some systems this
5632 resulted in spurious "integer overflow" errors. Also, it was casting the
5633 result into an int without checking.
5635 34. Testing for a connection timeout using "timeout_connect" in the retry rules
5636 did not work. The code looks as if it has *never* worked, though it appears
5637 to have been documented since at least releast 1.62. I have made it work.
5639 35. The "timeout_DNS" error in retry rules, also documented since at least
5640 1.62, also never worked. As it isn't clear exactly what this means, and
5641 clearly it isn't a major issue, I have abolished the feature by treating it
5642 as "timeout", and writing a warning to the main and panic logs.
5644 36. The display of retry rules for -brt wasn't always showing the error code
5647 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
5648 timeout_connect_A, timeout_connect_MX.
5650 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
5651 to the empty sender.
5653 39. The daemon was not analysing the content of -oX till after it had closed
5654 stderr and disconnected from the controlling terminal. This meant that any
5655 syntax errors were only noted on the panic log, and the return code from
5656 the command was 0. By re-arranging the code a little, I've made the
5657 decoding happen first, so such errors now appear on stderr, and the return
5658 code is 1. However, the actual setting up of the sockets still happens in
5659 the disconnected process, so errors there are still only recorded on the
5662 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
5663 connections (as happens on some IP stacks) was logged at start up time as
5664 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
5665 it from "IPv6 and IPv4", which means that two separate sockets are being
5668 41. The debug output for gethostbyname2() or getipnodebyname() failures now
5669 says whether AF_INET or AF_INET6 was passed as an argument.
5671 42. Exiwhat output was messed up when time zones were included in log
5674 43. Exiwhat now gives more information about the daemon's listening ports,
5675 and whether -tls-on-connect was used.
5677 44. The "port" option of the smtp transport is now expanded.
5679 45. A "message" modifier in a "warn" statement in a non-message ACL was being
5680 silently ignored. Now an error message is written to the main and panic
5683 46. There's a new ACL modifier called "logwrite" which writes to a log file
5684 as soon as it is encountered.
5686 47. Added $local_user_uid and $local_user_gid at routing time.
5688 48. Exim crashed when trying to verify a sender address that was being
5691 49. Exim was recognizing only a space character after ".include". It now also
5692 recognizes a tab character.
5694 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
5695 extracting the relevant information from the specification. The man page no
5696 longer contains scrambled data for the -d option, and I've added a section
5697 at the front about calling Exim under different names.
5699 51. Added "extra_headers" argument to the "mail" command in filter files.
5701 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
5704 53. Installed eximstats 1.29.
5706 54. Added transport_filter_timeout as a generic transport option.
5708 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
5709 Cc: header lines. This was required by RFC 822, but it not required by RFC
5712 56. Exim used to add From:, Date:, and Message-Id: header lines to any
5713 incoming messages that did not have them. Now it does so only if the
5714 message originates locally, that is, if there is no associated remote host
5715 address. When Resent- header lines are present, this applies to the Resent-
5716 lines rather than the non-Resent- lines.
5718 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
5719 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
5721 58. Messages for configuration errors now include the name of the main
5722 configuration file - useful now that there may be more than one file in a
5723 list (.included file names were always shown).
5725 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
5726 for those rare installations that do not start the daemon as root or run it
5727 setuid root. I've cut out the call to initgroups() if the daemon is not
5730 60. The Exim user and group can now be bound into the binary as text strings
5731 that are looked up at the start of Exim's processing.
5733 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
5735 62. Added $mailstore_basename variable.
5737 63. Installed patch to sieve.c from Michael Haardt.
5739 64. When Exim failed to open the panic log after failing to open the main log,
5740 the original message it was trying to log was written to stderr and debug
5741 output, but if they were not available (the usual case in production), it
5742 was lost. Now it is written to syslog before the two lines that record the
5743 failures to open the logs.
5745 65. Users' Exim filters run in subprocesses under the user's uid. It is
5746 possible for a "deliver" command or an alias in a "personal" command to
5747 provoke an address rewrite. If logging of address rewriting is configured,
5748 this fails because the process is not running as root or exim. There may be
5749 a better way of dealing with this, but for the moment (because 4.30 needs
5750 to be released), I have disabled address rewrite logging when running a
5751 filter in a non-root, non-exim process.
5757 1. The buildconfig auxiliary program wasn't quoting the value set for
5758 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
5759 not defined. This bug was masked in 4.22 by the effect that was fixed in
5762 2. Some messages that were rejected after a message id was allocated were
5763 shown as "incomplete" by exigrep. It no longer does this for messages that
5764 are rejected by local_scan() or the DATA or non-SMTP ACLs.
5766 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
5767 have allow_domain_literals set, the ID did not get logged in the <= line.
5768 Domain literals are now always recognized in Message-ID: header lines.
5770 4. The first argument for a ${extract expansion item is the key name or field
5771 number. Leading and trailing spaces in this item were not being ignored,
5772 causing some misleading effects.
5774 5. When deliver_drop_privilege was set, single queue runner processes started
5775 manually (i.e. by the command "exim -q") or by the daemon (which uses the
5776 same command in the process it spins off) were not dropping privilege.
5778 6. When the daemon running as "exim" started a queue runner, it always
5779 re-executed Exim in the spun-off process. This is a waste of effort when
5780 deliver_drop_privilege is set. The new process now just calls the
5781 queue-runner function directly.
5787 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
5790 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
5791 ignored. Though the use of -oP was forcing the writing of a pid file, it
5792 was always written to the default place.
5794 3. If the message "no IP address found for host xxxx" is generated during
5795 incoming verification, it is now followed by identification of the incoming
5796 connection (so you can more easily find what provoked it).
5798 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
5800 5. Added some features to "harden" Exim a bit more against certain attacks:
5802 (a) There is now a build-time option called FIXED_NEVER_USERS that can
5803 be put in Local/Makefile. This is like the never_users runtime option,
5804 but it cannot be overridden. The default setting is "root".
5806 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
5807 prefix string with which any file named in a -C command line option
5810 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
5811 is retained for -C and -D only if the caller of Exim is root. Without
5812 it, the exim user may also use -C and -D and retain privilege.
5814 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
5815 command line option is disabled.
5817 6. Macro names set by the -D option must start with an upper case letter, just
5818 like macro names defined in the configuration file.
5820 7. Added "dereference=" facility to LDAP.
5822 8. Two instances of the typo "uknown" in the source files are fixed.
5824 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
5825 the Configure-Makefile script screwed up while processing it.
5827 10. Incorporated PCRE 4.4.
5829 11. The SMTP synchronization check was not operating right at the start of an
5830 SMTP session. For example, it could not catch a HELO sent before the client
5831 waited for the greeting. There is now a check for outstanding input at the
5832 point when the greeting is written. Because of the duplex, asynchronous
5833 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
5834 way, but not yet received, when the check is performed.
5836 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
5837 on TCP/IP sockets, because this apparently causes some broken clients to
5840 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
5841 unchanged) from the Cygwin maintainer.
5843 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
5844 was supported instead of testing for mbx. Effectively a typo.
5846 15. The spa authenticator server code was not checking that the input it
5847 received was valid base64.
5849 16. The debug output line for the "set" modifier in ACLs was not showing the
5850 name of the variable that was being set.
5852 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
5854 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
5855 lookup on its own. Something else was needed to trigger the lookup. For
5856 example, a match in host_lookup or the need for a host name in a host list.
5857 Now, if $sender_host_name is referenced and the host name has not yet been
5858 looked up, a lookup is performed. If the lookup fails, the variable remains
5859 empty, and $host_lookup_failed is set to "1".
5861 19. Added "eqi" as a case-independent comparison operator.
5863 20. The saslauthd authentication condition could segfault if neither service
5864 nor realm was specified.
5866 21. If an overflowing value such as "2048M" was set for message_size_limit, the
5867 error message that was logged was misleading, and incoming SMTP
5868 connections were dropped. The message is now more accurate, and temporary
5869 errors are given to SMTP connections.
5871 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
5872 (except RSET) with a 421 error, until QUIT is received. However, it was
5873 failing to send a response to QUIT.
5875 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
5876 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
5877 ACL is now run after the helo_try_verify_hosts code.
5879 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
5880 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
5881 case-independent, so other case variants are also recognized). Apparently
5882 some systems use these upper case variants.
5884 25. If more than two messages were waiting for the same host, and a transport
5885 filter was specified for the transport, Exim sent two messages over the
5886 same TCP/IP connection, and then failed with "socket operation on non-
5887 socket" when it tried to send the third.
5889 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
5891 27. The extern definition of crypt16() in expand.c was not being excluded when
5892 the OS had its own crypt16() function.
5894 28. Added bounce_return_body as a new option, and bounce_return_size_limit
5895 as a preferred synonym for return_size_limit, both as an option and as an
5898 29. Added LIBS=-liconv to OS/Makefile-OSF1.
5900 30. Changed the default configuration ACL to relax the local part checking rule
5901 for addresses that are not in any local domains. For these addresses,
5902 slashes and pipe symbols are allowed within local parts, but the sequence
5903 /../ is explicitly forbidden.
5905 31. SPA server authentication was not clearing the challenge buffer before
5908 32. log_message in a "warn" ACL statement was writing to the reject log as
5909 well as to the main log, which contradicts the documentation and doesn't
5910 seem right (because no rejection is happening). So I have stopped it.
5912 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
5913 However, I am unable to do any testing of this.
5915 34. Fixed an infelicity in the appendfile transport. When checking directories
5916 for a mailbox, to see if any needed to be created, it was accidentally
5917 using path names with one or more superfluous leading slashes; tracing
5918 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
5920 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
5921 contents are added to the log line that is written for every discarded
5922 recipient. (Previously a log_message setting was ignored.)
5924 36. The ${quote: operator now quotes the string if it is empty.
5926 37. The install script runs exim in order to find its version number. If for
5927 some reason other than non-existence or emptiness, which it checks, it
5928 could not run './exim', it was installing it with an empty version number,
5929 i.e. as "exim-". This error state is now caught, and the installation is
5932 38. An argument was missing from the function that creates an error message
5933 when Exim fails to connect to the socket for saslauthd authentication.
5934 This could cause Exim to crash, or give a corrupted message.
5936 39. Added isip, isip4, and isip6 to ${if conditions.
5938 40. The ACL variables $acl_xx are now saved with the message, and can be
5939 accessed later in routers, transports, and filters.
5941 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
5942 strings in the file are not string-expanded.
5944 42. If a MAIL command specified a SIZE value that was too large to fit into an
5945 int variable, the check against message_size_limit failed. Such values are
5946 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
5947 day this will have to be increased, but I don't think I want to be around
5948 when emails are that large.
5955 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
5956 iconv() is not standard in FreeBSD.
5958 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
5959 IPv6 enabled. The observed symptom was a segmentation fault on return from
5960 the function os_common_find_running_interfaces() in src/os.c.
5962 3. In the check_special_case() function in daemon.c I had used "errno" as an
5963 argument name, which causes warnings on some systems. This was basically a
5964 typo, since it was named "eno" in the comments!
5966 4. The code that waits for the clock to tick (at a resolution of some fraction
5967 of a second) so as to ensure message-id uniqueness was always waiting for
5968 at least one whole tick, when it could have waited for less. [This is
5969 almost certainly not relevant at current processor speeds, where it is
5970 unlikely to ever wait at all. But we try to future-proof.]
5972 5. The function that sleeps for a time interval that includes fractions of a
5973 second contained a race. It did not block SIGALRM between setting the
5974 timer, and suspending (a couple of lines later). If the interval was short
5975 and the sigsuspend() was delayed until after it had expired, the suspension
5976 never ended. On busy systems this could lead to processes getting stuck for
5979 6. Some uncommon configurations may cause a lookup to happen in a queue runner
5980 process, before it forks any delivery processes. The open lookup caching
5981 mechanism meant that the open file or database connection was passed into
5982 the delivery process. The problem was that delivery processes always tidy
5983 up cached lookup data. This could cause a problem for the next delivery
5984 process started by the queue runner, because the external queue runner
5985 process does not know about the closure. So the next delivery process
5986 still has data in the lookup cache. In the case of a file lookup, there was
5987 no problem because closing a file descriptor in a subprocess doesn't affect
5988 the parent. However, if the lookup was caching a connection to a database,
5989 the connection was closed, and the second delivery process was likely to
5990 see errors such as "PGSQL: query failed: server closed the connection
5991 unexpectedly". The problem has been fixed by closing all cached lookups
5992 in a queue runner before running a delivery process.
5994 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
5995 seem to have the "const" qualifier which it has on other OS. I've
5998 8. Change 4.21/2 was too strict. It is only if there are two authenticators
5999 *of the same type* (client or server) with the same public name that an
6000 error should be diagnosed.
6002 9. When Exim looked up a host name for an IP address, but failed to find the
6003 original IP address when looking up the host name (a safety check), it
6004 output the message "<ip address> does not match any IP for NULL", which was
6005 confusing, to say the least. The bug was that the host name should have
6006 appeared instead of "NULL".
6008 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
6009 user that is built into the binary, and the -C or -D options is used, root
6010 privilege is dropped before the configuration file is read. In addition,
6011 logging is switched to stderr instead of the normal log files. If the
6012 configuration then re-defines the Exim user, the unprivileged environment
6013 is probably not what is expected, so Exim logs a panic warning message (but
6016 However, if deliver_drop_privilege is set, the unprivileged state may well
6017 be exactly what is intended, so the warning has been cut out in that case,
6018 and Exim is allowed to try to write to its normal log files.
6024 1. smtp_return_error_details was not giving details for temporary sender
6025 or receiver verification errors.
6027 2. Diagnose a configuration error if two authenticators have the same public
6030 3. Exim used not to create the message log file for a message until the first
6031 delivery attempt. This could be confusing when incoming messages were held
6032 for policy or load reasons. The message log file is now created at the time
6033 the message is received, and an initial "Received" line is written to it.
6035 4. The automatically generated man page for command line options had a minor
6036 bug that caused no ill effects; however, a more serious problem was that
6037 the procedure for building the man page automatically didn't always
6038 operate. Consequently, release 4.20 contains an out-of-date version. This
6039 shouldn't happen again.
6041 5. When building Exim with embedded Perl support, the script that builds the
6042 Makefile was calling 'perl' to find its compile-time parameters, ignoring
6043 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
6045 6. The freeze_tell option was not being used for messages that were frozen on
6046 arrival, either by an ACL or by local_scan().
6048 7. Added the smtp_incomplete_transaction log selector.
6050 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
6051 was accepting AUTH without a new EHLO.
6053 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
6054 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
6057 10. Logging of TCP/IP connections (when configured) now happens in the main
6058 daemon process instead of the child process, so that the TCP/IP connection
6059 count is more accurate (but it can never be perfect).
6061 11. The use of "drop" in a nested ACL was not being handled correctly in the
6062 outer ACL. Now, if condition failure induced by the nested "drop" causes
6063 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
6064 or "require"), the connection is dropped.
6066 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
6067 that yield "discard" can now be used with an "accept" or a "discard" verb,
6068 but an error is generated for any others (because I can't see a useful way
6069 to define what should happen).
6071 13. When an ACL is read dynamically from a file (or anywhere else), the lines
6072 are now processed in the same way as lines in the Exim configuration file.
6073 In particular, continuation lines are supported.
6075 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
6077 15. Added -ti meaning -t -i.
6079 16. Check for letters, digits, hyphens, and dots in the names of dnslist
6080 domains, and warn by logging if others are found.
6082 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
6083 returned from GIFCONF when Exim is trying to find the list of interfaces on
6084 a host. The code in os.c has been modified to copy each ifreq to an aligned
6085 structure in all cases.
6087 Also, in some cases, the returned ifreq's were being copied to a 'struct
6088 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
6089 means the last couple of bytes of an IPv6 address could be chopped if the
6090 ifreq contained only a normal sockaddr (14 bytes storage).
6092 18. Named domain lists were not supported in the hosts_treat_as_local option.
6093 An entry such as +xxxx was not recognized, and was treated as a literal
6096 19. Ensure that header lines added by a DATA ACL are included in the reject log
6097 if the ACL subsequently rejects the message.
6099 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
6100 MD5 (which is deprecated).
6102 21. When testing a filter file using -bf, Exim was writing a message when it
6103 took the sender from a "From " line in the message, but it was not doing so
6104 when it took $return_path from a Return-Path: header line. It now does.
6106 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
6107 with a valid header line field name (a series of printing characters
6108 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
6110 23. Changed "disc" in the source to "disk" to conform to the documentation and
6111 the book and for uniformity.
6113 24. Ignore Sendmail's -Ooption=value command line item.
6115 25. When execve() failed while trying to run a command in a pipe transport,
6116 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
6117 could be confused with a return value of 69 from the command itself. This
6118 has been changed to 127, the value the shell returns if it is asked to run
6119 a non-existent command. The wording for the related log line suggests a
6120 non-existent command as the problem.
6122 26. If received_header_text expands to an empty string, do not add a Received:
6123 header line to the message. (Well, it adds a token one on the spool, but
6124 marks it "old" so that it doesn't get used or transmitted.)
6126 27. Installed eximstats 1.28 (addition of -nt option).
6128 28. There was no check for failure on the call to getsockname() in the daemon
6129 code. This can fail if there is a shortage of resources on the system, with
6130 ENOMEM, for example. A temporary error is now given on failure.
6132 29. Contrary to the C standard, it seems that in some environments, the
6133 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
6134 program. Exim now does this explicitly; it affects the formatting of
6135 timestamps using strftime().
6137 30. If exiqsumm was given junk data, it threw up some uninitialized variable
6138 complaints. I've now initialized all the variables, to avoid this.
6140 32. Header lines added by a system filter were not being "seen" during
6141 transport-time rewrites.
6143 33. The info_callback() function passed to OpenSSL is set up with type void
6144 (*)(SSL *, int, int), as described somewhere. However, when calling the
6145 function (actually a macro) that sets it up, the type void(*)() is
6146 expected. I've put in a cast to prevent warnings from picky compilers.
6148 34. If a DNS black list lookup found a CNAME record, but there were no A
6149 records associated with the domain it pointed at, Exim crashed.
6151 35. If a DNS black list lookup returned more than one A record, Exim ignored
6152 all but the first. It now scans all returned addresses if a particular IP
6153 value is being sought. In this situation, the contents of the
6154 $dnslist_value variable are a list of all the addresses, separated by a
6157 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
6158 to accept a host name and all its aliases if the forward lookup for any of
6159 them yielded the IP address of the incoming connection. Now it accepts only
6160 those names whose forward lookup yields the correct IP address. Any other
6161 names are discarded. This closes a loophole whereby a rogue DNS
6162 administrator could create reverse DNS records to break through a
6163 wildcarded host restriction in an ACL.
6165 37. If a user filter or a system filter that ran in a subprocess used any of
6166 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
6167 the wrong values were passed to the pipe command ($thisaddress had the
6168 value of $0, $0 had the value of $1, etc). This bug was introduced by
6169 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
6171 38. Improved the line breaking for long SMTP error messages from ACLs.
6172 Previously, if there was no break point between 40 and 75 characters, Exim
6173 left the rest of the message alone. Two changes have been made: (a) I've
6174 reduced the minimum length to 35 characters; (b) if it can't find a break
6175 point between 35 and 75 characters, it looks ahead and uses the first one
6176 that it finds. This may give the occasional overlong line, but at least the
6177 remaining text gets split now.
6179 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
6180 file descriptors might be low, and that setting 1000 would always raise it.
6181 It turns out that in some environments, the limit is already over 1000 and
6182 that lowering it causes trouble. So now Exim takes care not to decrease it.
6184 40. When delivering a message, the value of $return_path is set to $sender_
6185 address at the start of routing (routers may change the value). By an
6186 oversight, this default was not being set up when an address was tested by
6187 -bt or -bv, which affected the outcome if any router or filter referred to
6190 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
6191 log only when "message" or "log_message" are set. However, if one of the
6192 conditions was an address verification, or a call to a nested ACL, the
6193 messages generated by the underlying test were being passed through. This
6194 no longer happens. The underlying message is available in $acl_verify_
6195 message for both "message" and "log_message" expansions, so it can be
6196 passed through if needed.
6198 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
6199 new expansion $bh_ to give the encoded byte string without charset
6200 translation. Translation happens only if iconv() is available; HAVE_ICONV
6201 indicates this at build time. HEADERS_CHARSET gives the charset to
6202 translate to; headers_charset can change it in the configuration, and
6203 "headers charset" can change it in an individual filter file.
6205 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
6206 that creates RFC 2047 encoded "words" labels them as that charset instead
6207 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
6208 expansion operator; (ii) when Exim creates a From: line for a local
6209 message; (iii) when a header line is rewritten to include a "phrase" part.
6211 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
6212 buggy, causing it to skip the first lines of messages whose message ID
6213 ended in 'D'. This would not have bitten before Exim release 4.14, because
6214 message IDs were unlikely to end in 'D' before then. The effect was to have
6215 incorrect size information for certain domains.
6217 45. #include "config.h" was missing at the start of the crypt16.c module. This
6218 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
6221 46. If there was a timeout during a "random" callout check, Exim treated it as
6222 a failure of the random address, and carried on sending RSET and the real
6223 address. If the delay was just some slowness somewhere, the response to the
6224 original RCPT would be taken as a response to RSET and so on, causing
6225 mayhem of various kinds.
6227 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
6228 when I implemented it. It didn't allow for the fact that some option values
6229 may legitimatetly be negative (e.g. size_addition), and it didn't even do
6230 the right test for positive values.
6232 48. Domain names in DNS records are case-independent. Exim always looks them up
6233 in lower case. Some resolvers return domain names in exactly the case they
6234 appear in the zone file, that is, they may contain uppercase letters. Not
6235 all resolvers do this - some return always lower case. Exim was treating a
6236 change of case by a resolver as a change of domain, similar to a widening
6237 of a domain abbreviation. This triggered its re-routing code and so it was
6238 trying to route what was effectively the same domain again. This normally
6239 caused routing to fail (because the router wouldn't handle the domain
6240 twice). Now Exim checks for this case specially, and just changes the
6241 casing of the domain that it ultimately uses when it transmits the message
6244 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
6247 50. If a filter generated a file delivery with a non-absolute name (possible if
6248 no home directory exists for the router), the forbid_file option was not
6251 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
6252 the existing equality matching.
6254 52. Exim was using ints instead of ino_t variables in some places where it was
6255 dealing with inode numbers.
6257 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
6258 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
6259 called TMPDIR, and if it finds it is different, it changes its value.
6261 54. The smtp_printf() function is now made available to local_scan() so
6262 additional output lines can be written before returning. There is also an
6263 smtp_fflush() function to enable the detection of a dropped connection.
6264 The variables smtp_input and smtp_batched_input are exported to
6267 55. Changed the default runtime configuration: the message "Unknown user"
6268 has been removed from the ACL, and instead placed on the localuser router,
6269 using the cannot_route_message feature. This means that any verification
6270 failures that generate their own messages won't get overridden. Similarly,
6271 the "Unrouteable address" message that was in the ACL for unverifiable
6272 relay addresses has also been removed.
6274 56. Added hosts_avoid_esmtp to the smtp transport.
6276 57. The exicyclog script was not checking for the esoteric option
6277 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
6278 will work only if exicyclog is run under the appropriate euid.
6280 58. Following a discussion on the list, the rules by which Exim recognises line
6281 endings on incoming messages have been changed. The -dropcr and drop_cr
6282 options are now no-ops, retained only for backwards compatibility. The
6283 following line terminators are recognized: LF CRLF CR. However, special
6284 processing applies to CR:
6286 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
6287 nor a local message in the state where . is a terminator.
6289 (ii) If a bare CR is encountered in a header line, an extra space is added
6290 after the line terminator so as not to end the header. The reasoning
6291 behind this is that bare CRs in header lines are most likely either
6292 to be mistakes, or people trying to play silly games.
6294 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
6295 was being incorrectly given as 18 bytes larger than it should have been.
6296 This is a VOB (very old bug).
6298 60. This may never have affected anything current, but just in case it has:
6299 When the local host is found other than at the start of a list of hosts,
6300 the local host, those with the same MX, and any that follow, are discarded.
6301 When the list in question was part of a longer list of hosts, the following
6302 hosts (not currently being processed) were also being discarded. This no
6303 longer happens. I'm not sure if this situation could ever has previously
6306 61. Added the "/MX" feature to lists of hosts in the manualroute and query
6309 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
6310 header. This is something that is recommended in a new Internet Draft, and
6311 is something that is documented as being done by Sendmail. There are two
6312 possible values. For messages generated by the autoreply transport, Exim
6315 Auto-Submitted: auto-replied
6317 whereas for all other generated messages (e.g. bounces) it adds
6319 Auto-Submitted: auto-generated
6321 63. The "personal" condition in filters now includes a test for the
6322 Auto-Submitted: header. If it contains the string "auto-" the message it
6323 not considered personal.
6325 64. Added rcpt_include_affixes as a generic transport option.
6327 65. Added queue_only_override (default true).
6329 66. Added the syslog_duplication option.
6331 67. If what should have been the first header line of a message consisted of
6332 a space followed by a colon, Exim was mis-interpreting it as a header line.
6333 It isn't of course - it is syntactically invalid and should therefore be
6334 treated as the start of the message body. The misbehaviour could have
6335 caused a number of strange effects, including loss of data in subsequent
6336 header lines, and spool format errors.
6338 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
6339 client host had authenticated. This control can now be exercised by an ACL
6340 for more flexibility.
6342 69. By default, callouts do not happen when testing with -bh. There is now a
6343 variant, -bhc, which does actually run the callout code, including
6344 consulting and updating the callout cache.
6346 70. Added support for saslauthd authentication, courtesy of Alexander
6349 71. If statvfs() failed on the spool or log directories while checking their
6350 size for availability, Exim confusingly gave the error "space shortage".
6351 Furthermore, in debugging mode it crashed with a floating point exception.
6352 These checks are done if check_{spool,log}_{space,inodes} are set, and when
6353 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
6354 serious problem, Exim now writes to the main and panic logs when this
6355 happens, with details of the failure. It then refuses to accept the
6356 incoming message, giving the message "spool directory problem" or "log
6357 directory problem" with a 421 code for SMTP messages.
6359 72. When Exim is about to re-exec itself, it ensures that the file descriptors
6360 0, 1, and 2 exist, because some OS complain for execs without them (see
6361 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
6362 descriptors. However, the code omitted to check that the open succeeded,
6363 causing mysterious errors if for some reason the permissions on /dev/null
6364 got screwed. Now Exim writes a message to the main and panic logs, and
6365 bombs out if it can't open /dev/null.
6367 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
6368 interact so that it is all more flexible. It is supposed to remain
6369 backwards compatible. Also added extra_local_interfaces.
6371 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
6372 to bomb out with an assertion failure - to the client this appears as a
6373 connection drop. This problem occurs in the part of the code that was taken
6374 from the Samba project. Fortunately, the assertion is in a very simple
6375 function, so I have fixed this by reproducing the function inline in the
6376 one place where it is called, and arranging for authentication to fail
6377 instead of killing the process with assert().
6379 75. The SPA client code was not working when the server requested OEM rather
6380 than Unicode encoding.
6382 76. Added code to make require_files with a specific uid setting more usable in
6383 the case where statting the file as root fails - usually a non-root-mounted
6384 NFS file system. When this happens and the failure is EACCES, Exim now
6385 forks a subprocess and does the per-uid checking as the relevant uid.
6387 77. Added process_log_path.
6389 78. If log_file_path was not explicitly set, a setting of check_log_space or
6390 check_log_inodes was ignored.
6392 79. If a space check for the spool or log partitions fails, the incident is now
6393 logged. Of course, in the latter case the data may get lost...
6395 80. Added the %p formatting code to string_format() so that it can be used to
6396 print addresses in debug_print(). Adjusted all the address printing in the
6397 debugging in store.c to use %p rather than %d.
6399 81. There was a concern that a line of code in smtp_in.c could overflow a
6400 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
6401 initially expressed, the concern was not well-founded, because trailing
6402 spaces are removed early. However, if the trailing spaces were followed by
6403 a NULL, they did not get removed, so the overflow was possible. Two fixes
6406 (a) I re-wrote the offending code in a cleaner fashion.
6407 (b) If an incoming SMTP command contains a NULL character, it is rejected
6410 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
6411 runs initgroups(), so that if the Exim user is in any additional groups,
6412 they will be used during message reception.
6418 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / users / jgh / exim.git / blob