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.
64 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
67 JH/02 The smtp transport option "multi_domain" is now expanded.
69 JH/03 The smtp transport now requests PRDR by default, if the server offers
72 JH/04 Certificate name checking on server certificates, when exim is a client,
73 is now done by default. The transport option tls_verify_cert_hostnames
74 can be used to disable this per-host. The build option
75 EXPERIMENTAL_CERTNAMES is withdrawn.
77 JH/05 The value of the tls_verify_certificates smtp transport and main options
78 default to the word "system" to access the system default CA bundle.
79 For GnuTLS, only version 3.0.20 or later.
81 JH/06 Verification of the server certificate for a TLS connection is now tried
82 (but not required) by default. The verification status is now logged by
83 default, for both outbound TLS and client-certificate supplying inbound
86 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
89 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
90 Status Notification (bounce) messages are now MIME format per RFC 3464.
91 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
92 under the control of the dsn_advertise_hosts option, and routers may
93 have a dsn_lasthop option.
95 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
96 default, modifiable by a malware= option. The list separator for
97 the options can now be changed in the usual way. Bug 68.
99 JH/10 The smtp_receive_timeout main option is now expanded before use.
101 JH/11 The incoming_interface log option now also enables logging of the
102 local interface on delivery outgoing connections.
104 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
105 if the interface and destination host and port all match.
107 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
110 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
111 Patch from Andrew Lewis.
113 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
114 now supports optional time-restrictions, weighting, and priority
115 modifiers per server. Patch originally by <rommer@active.by>.
117 JH/16 The spamd_address main option now supports a mixed list of local
118 and remote servers. Remote servers can be IPv6 addresses, and
119 specify a port-range.
121 JH/17 Bug 68: The spamd_address main option now supports an optional
122 timeout value per server.
124 JH/18 Bug 1581: Router and transport options headers_add/remove can
125 now have the list separator specified.
127 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
130 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
133 JH/21 Support for the A6 type of dns record is withdrawn.
135 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
136 rather than the verbs used.
138 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
139 from 255 to 1024 chars.
141 JH/24 Verification callouts now attempt to use TLS by default.
143 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
144 are generic router options now. The defaults didn't change.
146 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
147 Original patch from Alexander Shikoff, worked over by JH.
149 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
150 files if ZCAT_COMMAND is not executable.
152 JH/26 Bug 1539: Add timout/retry options on dnsdb lookups.
154 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
156 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
157 Normally benign, it bites when the pair was led to by a CNAME;
158 modern usage is to not canoicalize the domain to a CNAME target
159 (and we were inconsistent anyway for A-only vs AAAA+A).
161 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
163 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
164 when evaluating $sender_host_dnssec.
166 JH/31 Check the HELO verification lookup for DNSSEC, adding new
167 $sender_helo_dnssec variable.
169 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
171 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
173 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
175 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
176 documented as working, but never had. Support all but $spam_report.
178 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
179 added for tls authenticator.
184 TL/01 When running the test suite, the README says that variables such as
185 no_msglog_check are global and can be placed anywhere in a specific
186 test's script, however it was observed that placement needed to be near
187 the beginning for it to behave that way. Changed the runtest perl
188 script to read through the entire script once to detect and set these
189 variables, reset to the beginning of the script, and then run through
190 the script parsing/test process like normal.
192 TL/02 The BSD's have an arc4random API. One of the functions to induce
193 adding randomness was arc4random_stir(), but it has been removed in
194 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
195 function when detected.
197 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
198 cause callback expansion.
200 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
201 syntax errors in an expansion can be treated as a string instead of
202 logging or causing an error, due to the internal use of bool_lax
203 instead of bool when processing it.
205 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
206 server certificates when making smtp deliveries.
208 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
210 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
212 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
214 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
215 Merged patch from Sebastian Wiedenroth.
217 JH/05 Fix results-pipe from transport process. Several recipients, combined
218 with certificate use, exposed issues where response data items split
219 over buffer boundaries were not parsed properly. This eventually
220 resulted in duplicates being sent. This issue only became common enough
221 to notice due to the introduction of conection certificate information,
222 the item size being so much larger. Found and fixed by Wolfgang Breyha.
224 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
225 size buffer was used, resulting in syntax errors when an expansion
228 JH/07 Add support for directories of certificates when compiled with a GnuTLS
229 version 3.3.6 or later.
231 JH/08 Rename the TPDA expermimental facility to Event Actions. The #ifdef
232 is EXPERIMENTAL_EVENT, the main-configuration and transport options
233 both become "event_action", the variables become $event_name, $event_data
234 and $event_defer_errno. There is a new variable $verify_mode, usable in
235 routers, transports and related events. The tls:cert event is now also
236 raised for inbound connections, if the main configuration event_action
239 TL/06 In test suite, disable OCSP for old versions of openssl which contained
240 early OCSP support, but no stapling (appears to be less than 1.0.0).
242 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
243 server certificate names available under the smtp transport option
244 "tls_verify_cert_hostname" now do not permit multi-component wildcard
247 JH/10 Time-related extraction expansions from certificates now use the main
248 option "timezone" setting for output formatting, and are consistent
249 between OpenSSL and GnuTLS compilations. Bug 1541.
251 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
252 encoded parameter in the incoming message. Bug 1558.
254 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
255 include certificate info, eximon was claiming there were spoolfile
258 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
260 JH/14 Log delivery-related information more consistently, using the sequence
261 "H=<name> [<ip>]" wherever possible.
263 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
264 are problematic for Debian distribution, omit them from the release
267 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
269 JH/16 Fix string representation of time values on 64bit time_t anchitectures.
272 JH/17 Fix a null-indirection in certextract expansions when a nondefault
273 output list separator was used.
278 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
279 checkers that were complaining about end of non-void function with no
282 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
283 This was a regression intruduced in 4.83 by another bugfix.
285 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
287 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
288 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
294 TF/01 Correctly close the server side of TLS when forking for delivery.
296 When a message was received over SMTP with TLS, Exim failed to clear up
297 the incoming connection properly after forking off the child process to
298 deliver the message. In some situations the subsequent outgoing
299 delivery connection happened to have the same fd number as the incoming
300 connection previously had. Exim would try to use TLS and fail, logging
301 a "Bad file descriptor" error.
303 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
304 utilities have not been installed.
306 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
307 temporary space as the ACL may create new global variables.
309 TL/01 LDAP support uses per connection or global context settings, depending
310 upon the detected version of the libraries at build time.
312 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
313 to extract and use the src ip:port in logging and expansions as if it
314 were a direct connection from the outside internet. PPv2 support was
315 updated based on HAProxy spec change in May 2014.
317 JH/02 Add ${listextract {number}{list}{success}{fail}}.
319 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
320 Properly escape header and check for NULL return.
322 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
325 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
327 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
328 characters in header names, implemented as a verify condition.
329 Contributed by Michael Fischer v. Mollard.
331 TL/05 Rename SPF condition results err_perm and err_temp to standardized
332 results permerror and temperror. Previous values are deprecated but
333 still accepted. In a future release, err_perm and err_temp will be
334 completely removed, which will be a backward incompatibility if the
335 ACL tests for either of these two old results. Patch contributed by
336 user bes-internal on the mailing list.
338 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
340 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
341 selectors, in both main and reject logs.
343 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
346 JH/07 Add malware type "sock" for talking to simple daemon.
348 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
350 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
351 routers/transports under cutthrough routing.
353 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
354 numbers. Touch up "bool" conditional to keep the same definition.
356 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
358 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
360 JH/12 Expand items in router/transport headers_add or headers_remove lists
361 individually rather than the list as a whole. Bug 1452.
363 Required for reasonable handling of multiple headers_ options when
364 they may be empty; requires that headers_remove items with embedded
365 colons must have them doubled (or the list-separator changed).
367 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
368 view the policy declared in the DMARC record. Currently, $dmarc_status
369 is a combined value of both the record presence and the result of the
372 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
374 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
375 dnslookup router and the smtp transport (applying to the forward
378 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
379 of ldap servers used for a specific lookup. Patch provided by Heiko
382 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
383 New variable $lookup_dnssec_authenticated for observability.
385 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
386 Patch submitted by Lars Timman.
388 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
390 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
391 Requires trusted mode and valid format message id, aborts otherwise.
392 Patch contributed by Heiko Schlichting.
394 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
395 certextract with support for various fields. Bug 1358.
397 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
398 is requested by default, modifiable by smtp transport option
401 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
402 operate on certificate variables to give certificate fingerprints
403 Also new ${sha256:cert_variable}.
405 JH/23 The PRDR feature is moved from being Experimental into the mainline.
407 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
408 Christian Aistleitner.
410 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
412 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
413 file. Patch from Wolfgang Breyha.
415 JH/25 Expand the coverage of the delivery $host and $host_address to
416 client authenticators run in verify callout. Bug 1476.
418 JH/26 Port service names are now accepted for tls_on_connect_ports, to
419 align with daemon_smtp_ports. Bug 72.
421 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
422 support and error reporting did not work properly.
424 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
425 and is readable. Patch from Andrew Colin Kissa.
427 TL/14 Enhance documentation of ${run expansion and how it parses the
428 commandline after expansion, particularly in the case when an
429 unquoted variable expansion results in an empty value.
431 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
433 PP/02 Fix internal collision of T_APL on systems which support RFC3123
434 by renaming away from it. Addresses GH issue 15, reported by
437 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
439 TL/15 SECURITY: prevent double expansion in math comparison functions
440 (can expand unsanitized data). Not remotely exploitable.
447 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
449 PP/02 Make -n do something, by making it not do something.
450 When combined with -bP, the name of an option is not output.
452 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
455 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
456 $sender_host_name and config options to manage this, and basic check
459 PP/05 DSCP support for outbound connections and control modifier for inbound.
461 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
462 (Only plugin which currently uses this is kerberos4, which nobody should
463 be using, but we should make it available and other future plugins might
464 conceivably use it, even though it would break NAT; stuff *should* be
465 using channel bindings instead).
467 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
468 name; added for Sendmail compatibility; requires admin caller.
469 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
470 just ignore it); requires trusted caller.
471 Also parse but ignore: -Ac -Am -X<logfile>
474 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
476 TL/02 Add +smtp_confirmation as a default logging option.
478 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
479 Patch by Magnus Holmgren from 2007-02-20.
481 TL/04 Bugzilla 1281 - Spec typo.
482 Bugzilla 1283 - Spec typo.
483 Bugzilla 1290 - Spec grammar fixes.
485 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
487 TL/06 Add Experimental DMARC support using libopendmarc libraries.
489 TL/07 Fix an out of order global option causing a segfault. Reported to dev
490 mailing list by by Dmitry Isaikin.
492 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
494 JH/02 Support "G" suffix to numbers in ${if comparisons.
496 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
498 NM/01 Bugzilla 1197 - Spec typo
499 Bugzilla 1196 - Spec examples corrections
501 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
503 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
504 gnutls_enable_pkcs11, but renamed to more accurately indicate its
507 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
508 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
510 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
511 "acl {{name}{arg}...}", and optional args on acl condition
514 JH/05 Permit multiple router/transport headers_add/remove lines.
516 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
518 JH/07 Avoid using a waiting database for a single-message-only transport.
519 Performance patch from Paul Fisher. Bugzilla 1262.
521 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
524 JH/09 Add $headers_added variable, with content from use of ACL modifier
525 add_header (but not yet added to the message). Bugzilla 199.
527 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
528 Pulled from Bugzilla 817 by Wolfgang Breyha.
530 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
532 (nb: this is the same fix as in Exim 4.80.1)
534 JH/11 Add A= logging on delivery lines, and a client_set_id option on
537 JH/12 Add optional authenticated_sender logging to A= and a log_selector
540 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
542 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
543 advertise SMTP AUTH mechanism to us, instead of a generic
544 protocol violation error. Also, make Exim more robust to bad
545 data from the Dovecot auth socket.
547 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
549 When a queue runner is handling a message, Exim first routes the
550 recipient addresses, during which it prunes them based on the retry
551 hints database. After that it attempts to deliver the message to
552 any remaining recipients. It then updates the hints database using
555 So if a recipient address works intermittently, it can get repeatedly
556 deferred at routing time. The retry hints record remains fresh so the
557 address never reaches the final cutoff time.
559 This is a fairly common occurrence when a user is bumping up against
560 their storage quota. Exim had some logic in its local delivery code
561 to deal with this. However it did not apply to per-recipient defers
562 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
564 This change adds a proper retry rule check during routing so that the
565 final cutoff time is checked against the message's age. We only do
566 this check if there is an address retry record and there is not a
567 domain retry record; this implies that previous attempts to handle
568 the address had the retry_use_local_parts option turned on. We use
569 this as an approximation for the destination being like a local
570 delivery, as in LMTP.
572 I suspect this new check makes the old local delivery cutoff check
573 redundant, but I have not verified this so I left the code in place.
575 TF/02 Correct gecos expansion when From: is a prefix of the username.
577 Test 0254 submits a message to Exim with the header
581 When I ran the test suite under the user fanf2, Exim expanded
582 the header to contain my full name, whereas it should have added
583 a Resent-Sender: header. It erroneously treats any prefix of the
584 username as equal to the username.
586 This change corrects that bug.
588 GF/01 DCC debug and logging tidyup
589 Error conditions log to paniclog rather than rejectlog.
590 Debug lines prefixed by "DCC: " to remove any ambiguity.
592 TF/03 Avoid unnecessary rebuilds of lookup-related code.
594 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
595 Bug spotted by Jeremy Harris; was flawed since initial commit.
596 Would have resulted in OCSP responses post-SNI triggering an Exim
597 NULL dereference and crash.
599 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
601 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
602 Bug detection, analysis and fix by Samuel Thibault.
603 Bugzilla 1331, Debian bug #698092.
605 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
607 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
608 Server implementation by Todd Lyons, client by JH.
609 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
610 config variable "prdr_enable" controls whether the server
611 advertises the facility. If the client requests PRDR a new
612 acl_data_smtp_prdr ACL is called once for each recipient, after
613 the body content is received and before the acl_smtp_data ACL.
614 The client is controlled by bolth of: a hosts_try_prdr option
615 on the smtp transport, and the server advertisement.
616 Default client logging of deliveries and rejections involving
617 PRDR are flagged with the string "PRDR".
619 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
620 fclose(). Diagnosis by Todd Lyons.
622 PP/17 Update configure.default to handle IPv6 localhost better.
623 Patch by Alain Williams (plus minor tweaks).
626 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
627 This is now consistent with GnuTLS, and is now documented: the
628 previous undocumented portable approach to treating the option as
629 unset was to force an expansion failure. That still works, and
630 an empty string is now equivalent.
632 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
633 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
634 not performing validation itself.
636 PP/20 Added force_command boolean option to pipe transport.
637 Patch from Nick Koston, of cPanel Inc.
639 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
642 TF/04 Added udpsend ACL modifer and hexquote expansion operator
644 PP/21 Fix eximon continuous updating with timestamped log-files.
645 Broken in a format-string cleanup in 4.80, missed when I repaired the
646 other false fix of the same issue.
647 Report and fix from Heiko Schlichting.
650 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
651 Report from Prashanth Katuri.
653 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
654 It's SecureTransport, so affects any MacOS clients which use the
655 system-integrated TLS libraries, including email clients.
657 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
658 using a MIME ACL for non-SMTP local injection.
659 Report and assistance in diagnosis by Warren Baker.
661 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
663 JH/16 Fix comparisons for 64b. Bugzilla 1385.
665 TL/09 Add expansion variable $authenticated_fail_id to keep track of
666 last id that failed so it may be referenced in subsequent ACL's.
668 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
671 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
672 ldap library initialization, allowing self-signed CA's to be
673 used. Also properly sets require_cert option later in code by
674 using NULL (global ldap config) instead of ldap handle (per
675 session). Bug diagnosis and testing by alxgomz.
677 TL/12 Enhanced documentation in the ratelimit.pl script provided in
678 the src/util/ subdirectory.
680 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
681 renamed to Transport Post Delivery Action by Jeremy Harris, as
684 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
685 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
686 redis_servers = needs to be configured which will be used by the redis
687 lookup. Patch from Warren Baker, of The Packet Hub.
689 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
691 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
692 hostname or reverse DNS when processing a host list. Used suggestions
693 from multiple comments on this bug.
695 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
697 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
698 Missed a few lines, added it to make the runtest require no keyboard
701 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
702 contains upper case chars. Make router use caseful_local_part.
704 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
705 support when GnuTLS has been built with p11-kit.
711 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
713 This, or similar/improved, will also be change PP/11 of 4.82.
719 PP/01 Handle short writes when writing local log-files.
720 In practice, only affects FreeBSD (8 onwards).
721 Bugzilla 1053, with thanks to Dmitry Isaikin.
723 NM/01 Bugzilla 949 - Documentation tweak
725 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
728 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
730 PP/02 Implemented gsasl authenticator.
732 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
734 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
735 `pkg-config foo` for cflags/libs.
737 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
738 with rest of GSASL and with heimdal_gssapi.
740 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
741 `pkg-config foo` for cflags/libs for the TLS implementation.
743 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
744 properties get this fed in as external SSF. A number of robustness
745 and debugging improvements to the cyrus_sasl authenticator.
747 PP/08 cyrus_sasl server now expands the server_realm option.
749 PP/09 Bugzilla 1214 - Log authentication information in reject log.
750 Patch by Jeremy Harris.
752 PP/10 Added dbmjz lookup type.
754 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
756 PP/12 MAIL args handles TAB as well as SP, for better interop with
757 non-compliant senders.
758 Analysis and variant patch by Todd Lyons.
760 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
761 Bug report from Lars Müller <lars@samba.org> (via SUSE),
762 Patch from Dirk Mueller <dmueller@suse.com>
764 PP/13 tls_peerdn now print-escaped for spool files.
765 Observed some $tls_peerdn in wild which contained \n, which resulted
766 in spool file corruption.
768 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
769 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
770 or write after TLS renegotiation, which otherwise led to messages
773 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
774 as a tracking header (ie: a signed header comes before the signature).
775 Patch from Wolfgang Breyha.
777 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
778 comma-sep list; embedded commas doubled.
780 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
782 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
784 Report and patch from Dmitry Banschikov.
786 PP/16 Removed "dont_insert_empty_fragments" fron "openssl_options".
787 Removed SSL_clear() after SSL_new() which led to protocol negotiation
788 failures. We appear to now support TLS1.1+ with Exim.
790 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
791 lets Exim select keys and certificates based upon TLS SNI from client.
792 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
793 before an outbound SMTP session. New log_selector, +tls_sni.
795 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
796 NULL dereference. Report and patch from Alun Jones.
798 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
799 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
800 Not seeing resolver debug output on NetBSD, but suspect this is a
801 resolver implementation change.
803 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
804 Left warnings. Added "eximon gdb" invocation mode.
806 PP/21 Defaulting "accept_8bitmime" to true, not false.
808 PP/22 Added -bw for inetd wait mode support.
810 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
811 locate the relevant includes and libraries. Made this the default.
813 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
814 Bugzilla 1246, report and most of solution from Tomasz Kusy.
816 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
817 This may cause build issues on older platforms.
819 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
820 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
821 gnutls_require_mac & gnutls_require_protocols (no longer supported).
822 Added SNI support via GnuTLS too.
823 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
825 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
827 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
828 Applied second patch from Janne, implementing suggestion to default
829 multiple-strings-in-record handling to match SPF spec.
831 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
833 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
834 read-only, out of scope).
835 Patch from Wolfgang Breyha, report from Stuart Northfield.
837 PP/29 Fix three issues highlighted by clang analyser static analysis.
838 Only crash-plausible issue would require the Cambridge-specific
839 iplookup router and a misconfiguration.
840 Report from Marcin Mirosław.
842 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
844 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
845 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
846 As part of this, removing so much warning spew let me fix some minor
847 real issues in debug logging.
849 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
850 assignment on my part. Fixed.
852 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
853 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
854 Janne Snabb (who went above and beyond: thank you).
856 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
857 string otherwise requires a connection and a bunch more work and it's
858 relatively easy to get wrong. Should also expose TLS library linkage
861 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
862 64-bit ${eval} (JH/03).
864 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
865 GNU libc to support some of the 64-bit stuff, should not lead to
866 conflicts. Defined before os.h is pulled in, so if a given platform
867 needs to override this, it can.
869 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
870 protection layer was required, which is not implemented.
871 Bugzilla 1254, patch from Wolfgang Breyha.
873 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
874 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
875 tls_dhparam take prime identifiers. Also unbreak combination of
876 OpenSSL+DH_params+TLSSNI.
878 PP/39 Disable SSLv2 by default in OpenSSL support.
884 PP/01 Solaris build fix for Oracle's LDAP libraries.
885 Bugzilla 1109, patch from Stephen Usher.
887 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
889 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
892 TF/02 Fix a couple more cases where we did not log the error message
893 when unlink() failed. See also change 4.74-TF/03.
895 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
896 lock up or crash if it happened to be inside a call to libc when it
897 got a SIGUSR1 from exiwhat.
899 The SIGUSR1 handler appends the current process status to the process
900 log which is later printed by exiwhat. It used to use the general
901 purpose logging code to do this, but several functions it calls are
902 not safe for signals.
904 The new output code in the SIGUSR1 handler is specific to the process
905 log, and simple enough that it's easy to inspect for signal safety.
906 Removing some special cases also simplifies the general logging code.
907 Removing the spurious timestamps from the process log simplifies
910 TF/04 Improved ratelimit ACL condition.
912 The /noupdate option has been deprecated in favour of /readonly which
913 has clearer semantics. The /leaky, /strict, and /readonly update modes
914 are mutually exclusive. The update mode is no longer included in the
915 database key; it just determines when the database is updated. (This
916 means that when you upgrde Exim will forget old rate measurements.)
918 Exim now checks that the per_* options are used with an update mode that
919 makes sense for the current ACL. For example, when Exim is processing a
920 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
921 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
922 must specify per_mail/readonly. If you omit the update mode it defaults to
923 /leaky where that makes sense (as before) or /readonly where required.
925 The /noupdate option is now undocumented but still supported for
926 backwards compatibility. It is equivalent to /readonly except that in
927 ACLs where /readonly is required you may specify /leaky/noupdate or
928 /strict/noupdate which are treated the same as /readonly.
930 A useful new feature is the /count= option. This is a generalization
931 of the per_byte option, so that you can measure the throughput of other
932 aggregate values. For example, the per_byte option is now equivalent
933 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
935 The per_rcpt option has been generalized using the /count= mechanism
936 (though it's more complicated than the per_byte equivalence). When it is
937 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
938 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
939 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
940 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
941 Note that using per_rcpt with a non-readonly update mode in more than
942 one ACL will cause the recipients to be double-counted. (The per_mail
943 and per_byte options don't have this problem.)
945 The handling of very low rates has changed slightly. If the computed rate
946 is less than the event's count (usually one) then this event is the first
947 after a long gap. In this case the rate is set to the same as this event's
948 count, so that the first message of a spam run is counted properly.
950 The major new feature is a mechanism for counting the rate of unique
951 events. The new per_addr option counts the number of different
952 recipients that someone has sent messages to in the last time period. It
953 behaves like per_rcpt if all the recipient addresses are different, but
954 duplicate recipient addresses do not increase the measured rate. Like
955 the /count= option this is a general mechanism, so the per_addr option
956 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
957 example, measure the rate that a client uses different sender addresses
958 with the options per_mail/unique=$sender_address. There are further
959 details in the main documentation.
961 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
963 TF/06 Removed a few PCRE remnants.
965 TF/07 Automatically extract Exim's version number from tags in the git
966 repository when doing development or release builds.
968 PP/02 Raise smtp_cmd_buffer_size to 16kB.
969 Bugzilla 879. Patch from Paul Fisher.
971 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
972 Heavily based on revision 40f9a89a from Simon Arlott's tree.
975 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
977 PP/05 Variable $av_failed, true if the AV scanner deferred.
978 Bugzilla 1078. Patch from John Horne.
980 PP/06 Stop make process more reliably on build failure.
981 Bugzilla 1087. Patch from Heiko Schlittermann.
983 PP/07 Make maildir_use_size_file an _expandable_ boolean.
984 Bugzilla 1089. Patch from Heiko Schlittermann.
986 PP/08 Handle ${run} returning more data than OS pipe buffer size.
987 Bugzilla 1131. Patch from Holger Weiß.
989 PP/09 Handle IPv6 addresses with SPF.
990 Bugzilla 860. Patch from Wolfgang Breyha.
992 PP/10 GnuTLS: support TLS 1.2 & 1.1.
994 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
997 PP/11 match_* no longer expand right-hand-side by default.
998 New compile-time build option, EXPAND_LISTMATCH_RHS.
999 New expansion conditions, "inlist", "inlisti".
1001 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
1003 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
1005 PP/14 fix log_write() format string regression from TF/03.
1006 Bugzilla 1152. Patch from Dmitry Isaikin.
1012 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
1014 PP/02 Harmonised TLS library version reporting; only show if debugging.
1015 Layout now matches that introduced for other libraries in 4.74 PP/03.
1017 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
1019 PP/04 New "dns_use_edns0" global option.
1021 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
1024 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
1025 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
1027 TK/01 Updated PolarSSL code to 0.14.2.
1028 Bugzilla 1097. Patch from Andreas Metzler.
1030 PP/07 Catch divide-by-zero in ${eval:...}.
1031 Fixes bugzilla 1102.
1033 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
1036 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
1037 format-string attack -- SECURITY: remote arbitrary code execution.
1039 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
1040 time unintentionally subject to list matching rules, letting the header
1041 cause arbitrary Exim lookups (of items which can occur in lists, *not*
1042 arbitrary string expansion). This allowed for information disclosure.
1044 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
1045 INT_MIN/-1 -- value coerced to INT_MAX.
1051 NM/01 Workround for PCRE version dependency in version reporting
1054 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
1055 This fixes portability to compilers other than gcc, notably
1056 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
1058 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
1059 makefiles for portability to HP-UX and POSIX correctness.
1061 PP/01 Permit LOOKUP_foo enabling on the make command-line.
1062 Also via indented variable definition in the Makefile.
1063 (Debugging by Oliver Heesakkers).
1065 PP/02 Restore caching of spamd results with expanded spamd_address.
1066 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
1068 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
1069 Improves build reliability. Fix from: Frank Elsner
1071 NM/02 Fix wide character breakage in the rfc2047 coding
1072 Fixes bug 1064. Patch from Andrey N. Oktyabrski
1074 NM/03 Allow underscore in dnslist lookups
1075 Fixes bug 1026. Patch from Graeme Fowler
1077 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
1078 Code patches from Adam Ciarcinski of NetBSD.
1080 NM/04 Fixed exiqgrep to cope with mailq missing size issue
1083 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
1084 is logged, to avoid truncation. Patch from John Horne.
1086 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
1087 Patch from Jakob Hirsch.
1089 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
1090 SQL string expansion failure details.
1091 Patch from Andrey Oktyabrski.
1093 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
1094 Patch from Simon Arlott.
1096 PP/09 New lookups functionality failed to compile on old gcc which rejects
1097 extern declarations in function scope.
1098 Patch from Oliver Fleischmann
1100 PP/10 Use sig_atomic_t for flags set from signal handlers.
1101 Check getgroups() return and improve debugging.
1102 Fixed developed for diagnosis in bug 927 (which turned out to be
1105 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
1106 Patch from Mark Zealey.
1108 PP/12 Bugzilla 1056: Improved spamd server selection.
1109 Patch from Mark Zealey.
1111 PP/13 Bugzilla 1086: Deal with maildir quota file races.
1112 Based on patch from Heiko Schlittermann.
1114 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
1115 Patch from Uwe Doering, sign-off by Michael Haardt.
1117 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
1118 variable declaration deep within a block. Bug and patch from
1121 PP/15 lookups-Makefile IRIX compatibilty coercion.
1123 PP/16 Make DISABLE_DKIM build knob functional.
1125 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
1126 Patch by Simon Arlott
1128 TF/03 Fix valgrind.h portability to C89 compilers that do not support
1129 variable argument macros. Our copy now differs from upstream.
1135 TF/01 Failure to get a lock on a hints database can have serious
1136 consequences so log it to the panic log.
1138 TF/02 Log LMTP confirmation messages in the same way as SMTP,
1139 controlled using the smtp_confirmation log selector.
1141 TF/03 Include the error message when we fail to unlink a spool file.
1143 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
1144 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
1145 for maintaining out-of-tree patches for some time.
1147 PP/01 Bugzilla 139: Documentation and portability issues.
1148 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
1149 Handle per-OS dynamic-module compilation flags.
1151 PP/02 Let /dev/null have normal permissions.
1152 The 4.73 fixes were a little too stringent and complained about the
1153 permissions on /dev/null. Exempt it from some checks.
1154 Reported by Andreas M. Kirchwitz.
1156 PP/03 Report version information for many libraries, including
1157 Exim version information for dynamically loaded libraries. Created
1158 version.h, now support a version extension string for distributors
1159 who patch heavily. Dynamic module ABI change.
1161 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
1162 privilege escalation vulnerability whereby the Exim run-time user
1163 can cause root to append content of the attacker's choosing to
1166 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
1169 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
1170 If dropping privileges for untrusted macros, we disabled normal logging
1171 on the basis that it would fail; for the Exim run-time user, this is not
1172 the case, and it resulted in successful deliveries going unlogged.
1173 Fixed. Reported by Andreas Metzler.
1179 PP/01 Date: & Message-Id: revert to normally being appended to a message,
1180 only prepend for the Resent-* case. Fixes regression introduced in
1181 Exim 4.70 by NM/22 for Bugzilla 607.
1183 PP/02 Include check_rfc2047_length in configure.default because we're seeing
1184 increasing numbers of administrators be bitten by this.
1186 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
1188 PP/03 Bugzilla 994: added openssl_options main configuration option.
1190 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
1192 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
1194 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
1196 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
1197 without a peer certificate, leading to a segfault because of an
1198 assumption that peers always have certificates. Be a little more
1199 paranoid. Problem reported by Martin Tscholak.
1201 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
1202 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
1203 NB: ClamAV planning to remove STREAM in "middle of 2010".
1204 CL also introduces -bmalware, various -d+acl logging additions and
1205 more caution in buffer sizes.
1207 PP/09 Implemented reverse_ip expansion operator.
1209 PP/10 Bugzilla 937: provide a "debug" ACL control.
1211 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
1213 PP/12 Bugzilla 973: Implement --version.
1215 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
1217 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
1219 PP/15 Bugzilla 816: support multiple condition rules on Routers.
1221 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
1222 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
1223 ignore trailing whitespace.
1225 JJ/02 prevent non-panic DKIM error from being sent to paniclog
1227 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
1230 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
1231 Notification from Dr Andrew Aitchison.
1233 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
1234 ExtendedDetectionInfo response format.
1235 Notification from John Horne.
1237 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
1240 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
1241 XSL and documented dependency on system catalogs, with examples of how
1244 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
1247 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
1248 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
1249 configuration file which is writeable by the Exim user or group.
1251 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
1252 of configuration files to cover files specified with the -C option if
1253 they are going to be used with root privileges, not just the default
1256 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
1257 option (effectively making it always true).
1259 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
1260 files to be used while preserving root privileges.
1262 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
1263 that rogue child processes cannot use them.
1265 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
1266 run-time user, instead of root.
1268 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
1269 Exim run-time user without dropping privileges.
1271 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
1272 result string, instead of calling string_vformat() twice with the same
1275 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
1276 for other users. Others should always drop root privileges if they use
1277 -C on the command line, even for a whitelisted configure file.
1279 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
1281 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
1287 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
1288 $data_path, and $header_path variables; fixed documentation bugs and
1291 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
1292 exipick to access non-standard spools, including the "frozen" queue
1295 NM/01 Bugzilla 965: Support mysql stored procedures.
1296 Patch from Alain Williams
1298 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
1300 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
1301 Patch from Andreas Metzler
1303 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
1304 Patch from Kirill Miazine
1306 NM/05 Bugzilla 671: Added umask to procmail example.
1308 JJ/03 installed exipick 20100323.0, fixing doc bug
1310 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
1311 directory. Notification and patch from Dan Rosenberg.
1313 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
1315 TK/02 Improve log output when DKIM signing operation fails.
1317 MH/01 Treat the transport option dkim_domain as a colon separated
1318 list, not as a single string, and sign the message with each element,
1319 omitting multiple occurences of the same signer.
1321 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
1322 Bugzilla 985, 986. Patch by Simon Arlott
1324 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
1325 Patch by Simon Arlott
1327 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
1328 MBX locking. Notification from Dan Rosenberg.
1334 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
1336 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
1338 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
1340 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
1342 NM/04 Bugzilla 915: Flag broken perl installation during build.
1348 TK/01 Added patch by Johannes Berg that expands the main option
1349 "spamd_address" if it starts with a dollar sign.
1351 TK/02 Write list of recipients to X-Envelope-Sender header when building
1352 the mbox-format spool file for content scanning (suggested by Jakob
1355 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
1356 (http://www.dcc-servers.net/) support via dccifd. Activated by
1357 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
1359 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
1360 by Mark Daniel Reidel <mr@df.eu>.
1362 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
1363 When building exim an external PCRE library is now needed -
1364 PCRE is a system library on the majority of modern systems.
1365 See entry on PCRE_LIBS in EDITME file.
1367 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
1368 conversation. Added nologin parameter to request.
1369 Patch contributed by Kirill Miazine.
1371 TF/01 Do not log submission mode rewrites if they do not change the address.
1373 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
1375 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
1376 log files in place. Contributed by Roberto Lima.
1378 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
1380 TF/03 Bugzilla 615: When checking the local_parts router precondition
1381 after a local_part_suffix or local_part_prefix option, Exim now
1382 does not use the address's named list lookup cache, since this
1383 contains cached lookups for the whole local part.
1385 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
1386 Robert Millan. Documentation is in experimental-spec.txt.
1388 TF/04 Bugzilla 668: Fix parallel build (make -j).
1390 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
1392 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
1393 Patch provided by Jan Srzednicki.
1395 TF/05 Leading white space used to be stripped from $spam_report which
1396 wrecked the formatting. Now it is preserved.
1398 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
1399 that they are available at delivery time.
1401 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
1403 TF/08 TLS error reporting now respects the incoming_interface and
1404 incoming_port log selectors.
1406 TF/09 Produce a more useful error message if an SMTP transport's hosts
1407 setting expands to an empty string.
1409 NM/06 Bugzilla 744: EXPN did not work under TLS.
1410 Patch provided by Phil Pennock.
1412 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
1413 Patch provided by Richard Godbee.
1415 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
1416 acl_smtp_notquit, added index entry.
1418 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
1419 Patch provided by Eugene Bujak.
1421 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
1422 accept(). Patch provided by Maxim Dounin.
1424 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
1425 Patch provided by Phil Pennock.
1427 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
1429 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
1430 Patch provided by Brad "anomie" Jorsch.
1432 NM/14 Bugzilla 622: Added timeout setting to transport filter.
1433 Patch provided by Dean Brooks.
1435 TK/05 Add native DKIM support (does not depend on external libraries).
1437 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
1438 Patch provided by Graeme Fowler.
1440 NM/16 Bugzilla 851: Documentation example syntax fix.
1442 NM/17 Changed NOTICE file to remove references to embedded PCRE.
1444 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
1447 NM/19 Bugzilla 745: TLS version reporting.
1448 Patch provided by Phil Pennock.
1450 NM/20 Bugzilla 167: bool: condition support.
1451 Patch provided by Phil Pennock.
1453 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
1454 clients. Patch provided by Phil Pennock.
1456 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
1457 Patch provided by Brad "anomie" Jorsch.
1459 NM/23 Bugzilla 687: Fix misparses in eximstats.
1460 Patch provided by Heiko Schlittermann.
1462 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
1463 Patch provided by Heiko Schlittermann.
1465 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
1466 plus update to original patch.
1468 NM/26 Bugzilla 799: Documentation correction for ratelimit.
1470 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
1471 Patch provided by David Brownlee.
1473 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
1475 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
1477 NM/30 Bugzilla 888: TLS documentation bugfixes.
1479 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
1481 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
1482 Unlike the original bugzilla I have changed all shell scripts in src tree.
1484 NM/33 Bugzilla 898: Transport filter timeout fix.
1485 Patch by Todd Rinaldo.
1487 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
1488 Patch by Serge Demonchaux.
1490 NM/35 Bugzilla 39: Base64 decode bug fixes.
1491 Patch by Jakob Hirsch.
1493 NM/36 Bugzilla 909: Correct connect() call in dcc code.
1495 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
1497 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
1499 NM/39 Bugzilla 911: Fixed MakeLinks build script.
1505 TK/01 Add preliminary DKIM support. Currently requires a forked version of
1506 ALT-N's libdkim that I have put here:
1507 http://duncanthrax.net/exim-experimental/
1509 Note to Michael Haardt: I had to rename some vars in sieve.c. They
1510 were called 'true' and it seems that C99 defines that as a reserved
1511 keyword to be used with 'bool' variable types. That means you could
1512 not include C99-style headers which use bools without triggering
1513 build errors in sieve.c.
1515 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
1516 as mailq or other aliases. Changed the --help handling significantly
1517 to do whats expected. exim_usage() emits usage/help information.
1519 SC/01 Added the -bylocaldomain option to eximstats.
1521 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
1523 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
1525 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
1531 PH/01 Another patch from the Sieve maintainer.
1533 PH/02 When an IPv6 address is converted to a string for single-key lookup
1534 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
1535 dots are used instead of colons so that keys in lsearch files need not
1536 contain colons. This was done some time before quoting was made available
1537 in lsearch files. However, iplsearch files do require colons in IPv6 keys
1538 (notated using the quote facility) so as to distinguish them from IPv4
1539 keys. This meant that lookups for IP addresses in host lists did not work
1540 for iplsearch lookups.
1542 This has been fixed by arranging for IPv6 addresses to be expressed with
1543 colons if the lookup type is iplsearch. This is not incompatible, because
1544 previously such lookups could never work.
1546 The situation is now rather anomolous, since one *can* have colons in
1547 ordinary lsearch keys. However, making the change in all cases is
1548 incompatible and would probably break a number of configurations.
1550 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
1553 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
1554 conversion specification without a maximum field width, thereby enabling
1555 a rogue spamd server to cause a buffer overflow. While nobody in their
1556 right mind would setup Exim to query an untrusted spamd server, an
1557 attacker that gains access to a server running spamd could potentially
1558 exploit this vulnerability to run arbitrary code as the Exim user.
1560 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
1561 $primary_hostname instead of what libspf2 thinks the hosts name is.
1563 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
1564 a directory entry by the name of the lookup key. Previously, if a
1565 symlink pointed to a non-existing file or a file in a directory that
1566 Exim lacked permissions to read, a lookup for a key matching that
1567 symlink would fail. Now it is enough that a matching directory entry
1568 exists, symlink or not. (Bugzilla 503.)
1570 PH/03 The body_linecount and body_zerocount variables are now exported in the
1573 PH/04 Added the $dnslist_matched variable.
1575 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
1576 This means they are set thereafter only if the connection becomes
1579 PH/06 Added the client_condition to authenticators so that some can be skipped
1580 by clients under certain conditions.
1582 PH/07 The error message for a badly-placed control=no_multiline_responses left
1583 "_responses" off the end of the name.
1585 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
1587 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
1588 (without spaces) instead of just copying the configuration text.
1590 PH/10 Added the /noupdate option to the ratelimit ACL condition.
1592 PH/11 Added $max_received_linelength.
1594 PH/12 Added +ignore_defer and +include_defer to host lists.
1596 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
1597 way in which PCRE > 7.0 is built.
1599 PH/14 Implemented queue_only_load_latch.
1601 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
1602 MAIL command. The effect was to mangle the value on 64-bit systems.
1604 PH/16 Another patch from the Sieve maintainer.
1606 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
1608 PH/18 If a system quota error occurred while trying to create the file for
1609 a maildir delivery, the message "Mailbox is full" was not appended to the
1610 bounce if the delivery eventually timed out. Change 4.67/27 below applied
1611 only to a quota excession during the actual writing of the file.
1613 PH/19 It seems that peer DN values may contain newlines (and other non-printing
1614 characters?) which causes problems in log lines. The DN values are now
1615 passed through string_printing() before being added to log lines.
1617 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
1618 and InterBase are left for another time.)
1620 PH/21 Added message_body_newlines option.
1622 PH/22 Guard against possible overflow in moan_check_errorcopy().
1624 PH/23 POSIX allows open() to be a macro; guard against that.
1626 PH/24 If the recipient of an error message contained an @ in the local part
1627 (suitably quoted, of course), incorrect values were put in $domain and
1628 $local_part during the evaluation of errors_copy.
1634 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
1635 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
1638 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
1639 issue a MAIL command.
1641 PH/02 In an ACL statement such as
1643 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
1645 if a client was not listed at all, or was listed with a value other than
1646 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
1647 the condition was not true (as it should be), so access was not denied.
1648 The bug was that the ! inversion was incorrectly passed on to the second
1649 item. This has been fixed.
1651 PH/03 Added additional dnslists conditions == and =& which are different from
1652 = and & when the dns lookup returns more than one IP address.
1654 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
1655 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
1657 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
1658 FSYNC, which compiles an option called disable_fsync that allows for
1659 bypassing fsync(). The documentation is heavily laced with warnings.
1661 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
1663 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
1664 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
1665 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
1666 including adding "make clean"; (3) Added -fPIC when compiling the test
1667 dynamically loaded module, to get rid of a warning.
1669 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
1670 message fails, move_frozen_messages = true and ignore_bounce_errors_after
1671 = 0s. The bug is otherwise harmless.
1673 PH/07 There was a bug in the dovecot authenticator such that the value of
1674 $auth1 could be overwritten, and so not correctly preserved, after a
1675 successful authentication. This usually meant that the value preserved by
1676 the server_setid option was incorrect.
1678 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
1680 PH/09 Installed PCRE release 7.0.
1682 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
1683 run for batched SMTP input. It is now run at the start of every message
1684 in the batch. While fixing this I discovered that the process information
1685 (output by running exiwhat) was not always getting set for -bs and -bS
1686 input. This is fixed, and it now also says "batched" for BSMTP.
1688 PH/11 Added control=no_pipelining.
1690 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
1691 patch, slightly modified), and move the expansion of helo_data till after
1692 the connection is made in the smtp transport (so it can use these
1695 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
1697 PH/14 Added log_selector = +pid.
1699 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
1701 PH/16 Add ${if forany and ${if forall.
1703 PH/17 Added dsn_from option to vary the From: line in DSNs.
1705 PH/18 Flush SMTP output before performing a callout, unless control =
1706 no_callout_flush is set.
1708 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
1709 was true (the default) a successful delivery failed to delete the retry
1710 item, thus causing premature timeout of the address. The bug is now
1713 PH/20 Added hosts_avoid_pipelining to the smtp transport.
1715 PH/21 Long custom messages for fakedefer and fakereject are now split up
1716 into multiline reponses in the same way that messages for "deny" and
1717 other ACL rejections are.
1719 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
1720 with slight modification.
1722 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
1723 draft, changing the syntax and factoring some duplicate code".
1725 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
1726 for deliveries of the second and subsequent messages over the same SMTP
1729 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
1730 ${reduce, with only minor "tidies".
1732 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
1734 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
1735 expansion side effects.
1737 PH/27 When a message times out after an over-quota error from an Exim-imposed
1738 quota, the bounce message says "mailbox is full". This message was not
1739 being given when it was a system quota that was exceeded. It now should
1742 MH/03 Made $recipients available in local_scan(). local_scan() already has
1743 better access to the recipient list through recipients_list[], but
1744 $recipients can be useful in postmaster-provided expansion strings.
1746 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
1747 in the case of a MAIL command with additional options following the
1748 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
1749 were accidentally chopped off.
1751 PH/29 SMTP synchronization checks are implemented when a command is read -
1752 there is a check that no more input is waiting when there shouldn't be
1753 any. However, for some commands, a delay in an ACL can mean that it is
1754 some time before the response is written. In this time, more input might
1755 arrive, invalidly. So now there are extra checks after an ACL has run for
1756 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
1757 pipelining has not been advertised.
1759 PH/30 MH's patch to allow iscntrl() characters to be list separators.
1761 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
1762 returned in the SMTP response when smtp_return_error_details was false.
1763 This has been fixed.
1765 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
1766 instead of the C I/O that was originally supplied, because problems were
1767 reported on Solaris.
1769 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
1770 Exim which did not show up earlier: it was assuming that a call to
1771 SSL_CTX_set_info_callback() might give an error value. In fact, there is
1772 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
1773 was a macro that became an assignment, so it seemed to work. This has
1774 changed to a proper function call with a void return, hence the compile
1775 error. Exim's code has been fixed.
1777 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
1780 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
1782 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
1784 PH/37 If a message is not accepted after it has had an id assigned (e.g.
1785 because it turns out to be too big or there is a timeout) there is no
1786 "Completed" line in the log. When some messages of this type were
1787 selected by exigrep, they were listed as "not completed". Others were
1788 picked up by some special patterns. I have improved the selection
1789 criteria to be more general.
1791 PH/38 The host_find_failed option in the manualroute router can now be set
1792 to "ignore", to completely ignore a host whose IP address cannot be
1793 found. If all hosts are ignored, the behaviour is controlled by the new
1794 host_all_ignored option.
1796 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
1797 homing or because of multiple MX records with /mx) generated more than
1798 one IP address, and the following item turned out to be the local host,
1799 all the secondary addresses of the first item were incorrectly removed
1800 from the list, along with the local host and any following hosts (which
1801 is what is supposed to happen).
1803 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
1804 whoever called Exim into the -H file. In the case of the daemon it was
1805 behaving confusingly. When first started, it used values for whoever
1806 started the daemon, but after a SIGHUP it used the Exim user (because it
1807 calls itself on a restart). I have changed the code so that it now always
1810 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
1811 message are rejected with the same error (e.g. no authentication or bad
1812 sender address), and a DATA command is nevertheless sent (as can happen
1813 with PIPELINING or a stupid MUA), the error message that was given to the
1814 RCPT commands is included in the rejection of the DATA command. This is
1815 intended to be helpful for MUAs that show only the final error to their
1818 PH/42 Another patch from the Sieve maintainer.
1820 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
1821 Eximstats - Fixed some broken HTML links and added missing column headers
1823 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
1824 columns for Rejects, Temp Rejects, Ham, and Spam rows.
1826 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
1828 PH/43 Yet another patch from the Sieve maintainer.
1830 PH/44 I found a way to check for a TCP/IP connection going away before sending
1831 the response to the final '.' that terminates a message, but only in the
1832 case where the client has not sent further data following the '.'
1833 (unfortunately, this is allowed). However, in many cases there won't be
1834 any further data because there won't be any more messages to send. A call
1835 to select() can be used: if it shows that the input is "ready", there is
1836 either input waiting, or the socket has been closed. An attempt to read
1837 the next input character can distinguish the two cases. Previously, Exim
1838 would have sent an OK response which the client would never have see.
1839 This could lead to message repetition. This fix should cure that, at
1840 least in a lot of common cases.
1842 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
1843 advertised in response to EHLO.
1849 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
1850 fixed by 4.65/MH/01 (is this a record?) are fixed:
1852 (i) An empty string was always treated as zero by the numeric comparison
1853 operators. This behaviour has been restored.
1855 (ii) It is documented that the numeric comparison operators always treat
1856 their arguments as decimal numbers. This was broken in that numbers
1857 starting with 0 were being interpreted as octal.
1859 While fixing these problems I realized that there was another issue that
1860 hadn't been noticed. Values of message_size_limit (both the global option
1861 and the transport option) were treated as octal if they started with 0.
1862 The documentation was vague. These values are now always treated as
1863 decimal, and I will make that clear in the documentation.
1869 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
1870 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
1873 MH/01 Don't check that the operands of numeric comparison operators are
1874 integers when their expansion is in "skipping" mode (fixes bug
1875 introduced by 4.64-PH/07).
1877 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
1878 child addresses, Exim now panics and dies. Previously, because the count
1879 is held in a short int, deliveries were likely to be lost. As such a
1880 large number of recipients for a single message is ridiculous
1881 (performance will be very, very poor), I have chosen to impose a limit
1882 rather than extend the field.
1888 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
1889 leftover -K file (the existence of which was triggered by #402).
1890 While we were at it, introduced process PID as part of the -K
1891 filename. This should rule out race conditions when creating
1894 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
1895 processing considerably. Previous code took too long for large mails,
1896 triggering a timeout which in turn triggers #401.
1898 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
1899 in the DK code in transports.c. sendfile() is not really portable,
1900 hence the _LINUX specificness.
1902 TF/01 In the add_headers option to the mail command in an Exim filter,
1903 there was a bug that Exim would claim a syntax error in any
1904 header after the first one which had an odd number of characters
1907 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
1908 callout verification, Exim cached a "reject" for the entire domain. This
1909 is correct for most verifications, but it is not correct for a recipient
1910 verification with use_sender or use_postmaster set, because in that case
1911 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
1912 case of MAIL FROM:<> rejection from other early rejections (e.g.
1913 rejection of HELO). When verifying a recipient using a non-null MAIL
1914 address, the cache is ignored if it shows MAIL FROM:<> rejection.
1915 Whatever the result of the callout, the value of the domain cache is
1916 left unchanged (for any other kind of callout, getting as far as trying
1917 RCPT means that the domain itself is ok).
1919 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
1922 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
1923 manifest itself as EPIPE rather than ECONNECT. When tidying away a
1924 session, the daemon ignores ECONNECT errors and logs others; it now
1925 ignores EPIPE as well.
1927 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
1928 (quoted-printable decoding).
1930 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
1931 later the small subsequent patch to fix an introduced bug.
1933 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
1935 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
1937 PH/08 An error is now given if message_size_limit is specified negative.
1939 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
1940 to be given (somewhat) arbitrary names.
1942 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
1945 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
1946 miscellaneous code fixes
1948 PH/10 Added the log_reject_target ACL modifier to specify where to log
1951 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
1952 hostname. This is wrong, because it relates to the incoming message (and
1953 probably the interface on which it is arriving) and not to the outgoing
1954 callout (which could be using a different interface). This has been
1955 changed to use the value of the helo_data option from the smtp transport
1956 instead - this is what is used when a message is actually being sent. If
1957 there is no remote transport (possible with a router that sets up host
1958 addresses), $smtp_active_hostname is used.
1960 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
1961 tweaks were necessary in order to get it to work (see also 21 below):
1962 (a) The code assumed that strncpy() returns a negative number on buffer
1963 overflow, which isn't the case. Replaced with Exim's string_format()
1965 (b) There were several signed/unsigned issues. I just did the minimum
1966 hacking in of casts. There is scope for a larger refactoring.
1967 (c) The code used strcasecmp() which is not a standard C function.
1968 Replaced with Exim's strcmpic() function.
1969 (d) The code set only $1; it now sets $auth1 as well.
1970 (e) A simple test gave the error "authentication client didn't specify
1971 service in request". It would seem that Dovecot has changed its
1972 interface. Fortunately there's a specification; I followed it and
1973 changed what the client sends and it appears to be working now.
1975 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
1978 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
1979 address A is aliased to B and C, where B exists and C does not. Without
1980 -v the output is "A verified" because verification stops after a
1981 successful redirection if more than one address is generated. However,
1982 with -v the child addresses are also verified. Exim was outputting "A
1983 failed to verify" and then showing the successful verification for C,
1984 with its parentage. It now outputs "B failed to verify", showing B's
1985 parentage before showing the successful verification of C.
1987 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
1988 look up a TXT record in a specific list after matching in a combined
1991 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
1992 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
1993 they consult the DNS. I had assumed they would set it the way they
1994 wanted; and indeed my experiments on Linux seem to show that in some
1995 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
1996 To be on the safe side, however, I have now made the interface to
1997 host_find_byname() similar to host_find_bydns(), with an argument
1998 containing the DNS resolver options. The host_find_byname() function now
1999 sets these options at its start, just as host_find_bydns() does. The smtp
2000 transport options dns_qualify_single and dns_search_parents are passed to
2001 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
2002 of host_find_byname() use the default settings of RES_DEFNAMES
2003 (qualify_single) but not RES_DNSRCH (search_parents).
2005 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
2006 spool_read_header() do less string testing, by means of a preliminary
2007 switch on the second character of optional "-foo" lines. (This is
2008 overdue, caused by the large number of possibilities that now exist.
2009 Originally there were few.) While I was there, I also converted the
2010 str(n)cmp tests so they don't re-test the leading "-" and the first
2011 character, in the hope this might squeeze out yet more improvement.
2013 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
2014 flag allowing group syntax was set by the header_syntax check but not
2015 turned off, possible causing trouble later; (2) The flag was not being
2016 set at all for the header_verify test, causing "group"-style headers to
2017 be rejected. I have now set it in this case, and also caused header_
2018 verify to ignore an empty address taken from a group. While doing this, I
2019 came across some other cases where the code for allowing group syntax
2020 while scanning a header line wasn't quite right (mostly, not resetting
2021 the flag correctly in the right place). These bugs could have caused
2022 trouble for malformed header lines. I hope it is now all correct.
2024 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
2025 with the "reply" argument non-NULL. The code, however (which originally
2026 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
2027 but it didn't always do it. This confused somebody who was copying the
2028 code for some other use. I have removed all the tests.
2030 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
2031 feature that was used to support insecure browsers during the U.S. crypto
2032 embargo. It requires special client support, and Exim is probably the
2033 only MTA that supported it -- and would never use it because real RSA is
2034 always available. This code has been removed, because it had the bad
2035 effect of slowing Exim down by computing (never used) parameters for the
2036 RSA_EXPORT functionality.
2038 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
2039 authenticator to fail if there's a tab character in the incoming data
2040 (there should never be unless someone is messing about, as it's supposed
2041 to be base64-encoded). Also added, on Timo's advice, the "secured" option
2042 if the connection is using TLS or if the remote IP is the same as the
2043 local IP, and the "valid-client-cert option" if a client certificate has
2046 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
2047 authenticators. This can be used for authorization after authentication
2048 succeeds. (In the case of plaintext, it servers for both authentication
2051 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
2052 if any retry times were supplied.
2054 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
2055 connection, where there is no client IP address to check. In this
2056 situation, the verify now always succeeds.
2058 PH/25 Applied John Jetmore's -Mset patch.
2060 PH/26 Added -bem to be like -Mset, but loading a message from a file.
2062 PH/27 In a string expansion for a processed (not raw) header when multiple
2063 headers of the same name were present, leading whitespace was being
2064 removed from all of them, but trailing whitespace was being removed only
2065 from the last one. Now trailing whitespace is removed from each header
2066 before concatenation. Completely empty headers in a concatenation (as
2067 before) are ignored.
2069 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
2070 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
2072 PH/29 [Removed. This was a change that I later backed out, and forgot to
2073 correct the ChangeLog entry (that I had efficiently created) before
2074 committing the later change.]
2076 PH/30 Exim was sometimes attempting to deliver messages that had suffered
2077 address errors (4xx response to RCPT) over the same connection as other
2078 messages routed to the same hosts. Such deliveries are always "forced",
2079 so retry times are not inspected. This resulted in far too many retries
2080 for the affected addresses. The effect occurred only when there were more
2081 hosts than the hosts_max_try setting in the smtp transport when it had
2082 the 4xx errors. Those hosts that it had tried were not added to the list
2083 of hosts for which the message was waiting, so if all were tried, there
2084 was no problem. Two fixes have been applied:
2086 (i) If there are any address or message errors in an SMTP delivery, none
2087 of the hosts (tried or untried) are now added to the list of hosts
2088 for which the message is waiting, so the message should not be a
2089 candidate for sending over the same connection that was used for a
2090 successful delivery of some other message. This seems entirely
2091 reasonable: after all the message is NOT "waiting for some host".
2092 This is so "obvious" that I'm not sure why it wasn't done
2093 previously. Hope I haven't missed anything, but it can't do any
2094 harm, as the worst effect is to miss an optimization.
2096 (ii) If, despite (i), such a delivery is accidentally attempted, the
2097 routing retry time is respected, so at least it doesn't keep
2098 hammering the server.
2100 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
2101 in ${readsocket because some servers need this prod.
2103 PH/32 Added some extra debug output when updating a wait-xxx database.
2105 PH/33 The hint "could be header name not terminated by colon", which has been
2106 given for certain expansion errors for a long time, was not being given
2107 for the ${if def:h_colon_omitted{... case.
2109 PH/34 The spec says: "With one important exception, whenever a domain list is
2110 being scanned, $domain contains the subject domain." There was at least
2111 one case where this was not true.
2113 PH/35 The error "getsockname() failed: connection reset by peer" was being
2114 written to the panic log as well as the main log, but it isn't really
2115 panic-worthy as it just means the connection died rather early on. I have
2116 removed the panic log writing for the ECONNRESET error when getsockname()
2119 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
2120 runs only) independently of the message's sender address. This meant
2121 that, if the 4xx error was in fact related to the sender, a different
2122 message to the same recipient with a different sender could confuse
2123 things. In particualar, this can happen when sending to a greylisting
2124 server, but other circumstances could also provoke similar problems.
2125 I have changed the default so that the retry time for these errors is now
2126 based a combination of the sender and recipient addresses. This change
2127 can be overridden by setting address_retry_include_sender=false in the
2130 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
2131 remote server are returned as part of bounce messages. This was not
2132 happening for LMTP over a pipe (the lmtp transport), but now it is the
2133 same for both kinds of LMTP.
2135 PH/38 Despite being documented as not happening, Exim was rewriting addresses
2136 in header lines that were in fact CNAMEs. This is no longer the case.
2138 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
2139 and queue runs started by the daemon processed all messages. This has
2140 been fixed so that -R and -S can now usefully be given with -q<time>.
2142 PH/40 Import PCRE release 6.7 (fixes some bugs).
2144 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
2146 PH/42 Give an error if -q is specified more than once.
2148 PH/43 Renamed the variables $interface_address and $interface_port as
2149 $received_ip_address and $received_port, to make it clear that these
2150 values apply to message reception, and not to the outgoing interface when
2151 a message is delivered. (The old names remain recognized, of course.)
2153 PH/44 There was no timeout on the connect() call when using a Unix domain
2154 socket in the ${readsocket expansion. There now is.
2156 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
2157 be meaningful with "accept".
2159 SC/01 Eximstats V1.43
2160 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
2162 SC/02 Eximstats V1.44
2163 Use a glob alias rather than an array ref in the generated
2164 parser. This improves both readability and performance.
2166 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
2167 Collect SpamAssassin and rejection statistics.
2168 Don't display local sender or destination tables unless
2169 there is data to show.
2170 Added average volumes into the top table text output.
2172 SC/04 Eximstats V1.46
2173 Collect data on the number of addresses (recipients)
2174 as well as the number of messages.
2176 SC/05 Eximstats V1.47
2177 Added 'Message too big' to the list of mail rejection
2178 reasons (thanks to Marco Gaiarin).
2180 SC/06 Eximstats V1.48
2181 Mainlog lines which have GMT offsets and are too short to
2182 have a flag are now skipped.
2184 SC/07 Eximstats V1.49 (Alain Williams)
2185 Added the -emptyok flag.
2187 SC/08 Eximstats V1.50
2188 Fixes for obtaining the IP address from reject messages.
2190 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
2191 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
2192 whitesspace changes from 4.64-PH/27
2194 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
2197 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
2198 are found, allow negative numbers in numeric criteria)
2200 JJ/06 exipick.20061117.2, added new $message_body_missing variable
2202 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
2203 to match changes made in 4.64-PH/43
2205 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
2207 PH/47 Put in an explicit test for a DNS lookup of an address record where the
2208 "domain" is actually an IP address, and force a failure. This locks out
2209 those revolvers/nameservers that support "A-for-A" lookups, in
2210 contravention of the specifications.
2212 PH/48 When a host name was looked up from an IP address, and the subsequent
2213 forward lookup of the name timed out, the host name was left in
2214 $sender_host_name, contrary to the specification.
2216 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
2217 restricted to single-key lookups, Exim was not diagnosing an error if
2218 * or *@ was used with a query-style lookup.
2220 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
2222 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
2223 long ago, but noone interested enough thought of it. Let's just say that
2224 the "1.1" means that there are some new functions that weren't there at
2225 some point in the past.
2227 PH/51 Error processing for expansion failure of helo_data from an smtp
2228 transport during callout processing was broken.
2230 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
2231 tested/used via the -bh/-bhc/-bs options.
2233 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
2234 bug, fixed in subsequent PCRE releases).
2236 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
2237 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
2239 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
2245 SC/01 Use a glob alias rather than an array ref in eximstats generated
2246 parser. This improves both readability and performance.
2248 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
2249 Don't display local sender or destination tables in eximstats unless
2250 there is data to show.
2251 Added average volumes into the eximstats top table text output.
2253 SC/03 Collect data on the number of addresses (recipients) as well
2254 as the number of messages in eximstats.
2256 TF/01 Correct an error in the documentation for the redirect router. Exim
2257 does (usually) call initgroups() when daemonizing.
2259 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
2260 with consistent privilege compared to when running as a daemon.
2262 TF/03 Note in the spec that $authenticated_id is not set for local
2263 submissions from trusted users.
2265 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
2266 Thanks to Dean Brooks <dean@iglou.com> for the patch.
2268 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
2269 by adding some example configuration directives to the default
2270 configuration file. A little bit of work is required to uncomment the
2271 directives and define how usernames and passwords are checked, but
2272 there is now a framework to start from.
2274 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
2275 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
2276 without this. I don't know how relevant this is to other LDAP libraries.
2278 PH/02 Add the verb name to the "unknown ACL verb" error.
2280 PH/03 Magnus Holmgren's patch for filter_prepend_home.
2282 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
2284 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
2285 directory not expanded when it should be if an expanded home directory
2286 was set for the address (which is overridden by the transport).
2288 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
2291 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
2292 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
2293 because it is too late at that time, and has no effect.
2295 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
2296 security issue with \' (bugzilla #107). I could not use the
2297 PQescapeStringConn() function, because it needs a PGconn value as one of
2300 PH/08 When testing addresses using -bt, indicate those final addresses that
2301 are duplicates that would not cause an additional delivery. At least one
2302 person was confused, thinking that -bt output corresponded to deliveries.
2303 (Suppressing duplicates isn't a good idea as you lose the information
2304 about possibly different redirections that led to the duplicates.)
2306 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
2307 systems where poll() doesn't work, in particular OS X.
2309 PH/10 Added more information to debugging output for retry time not reached.
2311 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
2312 operations in malware.c.
2314 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
2317 PH/13 If write_rejectlog was set false when logging was sent to syslog with
2318 syslog_duplication set false, log lines that would normally be written
2319 both the the main log and to the reject log were not written to syslog at
2322 PH/14 In the default configuration, change the use of "message" in ACL warn
2323 statements to "add_header".
2325 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
2326 not followed by a command (e.g. "seen endif").
2328 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
2329 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
2332 PH/17 Added extra conditions to the default value of delay_warning_condition
2336 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
2337 { match{$h_precedence:}{(?i)bulk|list|junk} } \
2338 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
2341 The Auto-Submitted: and various List- headers are standardised, whereas I
2342 don't think Precedence: ever was.
2344 PH/18 Refactored debugging code in route_finduser() to show more information,
2345 in particular, the error code if getpwnam() issues one.
2347 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
2348 This is apparently needed in addition to the PH/07 change above to avoid
2349 any possible encoding problems.
2351 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
2352 but not after initializing Perl.
2354 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
2355 output them only if debugging. By default they are written stderr,
2356 apparently, which is not desirable.
2358 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
2361 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
2364 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
2366 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
2367 authenticated or an ident call has been made. Suppress the default
2368 values for $authenticated_id and $authenticated_sender (but permit -oMai
2369 and -oMas) when testing with -bh.
2371 PH/24 Re-jigged the order of the tests in the default configuration so that the
2372 tests for valid domains and recipients precede the DNS black list and CSA
2373 tests, on the grounds that those ones are more expensive.
2375 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
2376 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
2377 This bug exists in every version of Exim that I still have, right back to
2380 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
2381 However, an attempt to turn on case-sensitivity in a regex key by
2382 including (?-i) didn't work because the subject string was already
2383 lowercased, and the effects were non-intuitive. It turns out that a
2384 one-line patch can be used to allow (?-i) to work as expected.
2390 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
2391 other effects) broke the use of negated acl sub-conditions.
2393 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
2396 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
2397 "Deny" causes Exim to reject the incoming connection with a 554 error.
2398 Unfortunately, if there is a major crisis, such as a disk failure,
2399 tcp-wrappers gives "deny", whereas what one would like would be some
2400 kind of temporary error. A kludge has been added to help with this.
2401 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
2402 554 error is used if errno is still zero or contains ENOENT (which occurs
2403 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
2406 PH/03 Add -lutil to the default FreeBSD LIBS setting.
2408 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
2409 errors. Otherwise a message that provokes a temporary error (when other
2410 messages do not) can cause a whole host to time out.
2412 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
2413 addresses were routed directly to files or pipes from a redirect router.
2414 File deliveries just didn't batch; pipe deliveries might have suffered
2417 PH/06 A failure to get a lock for a hints database would erroneously always say
2418 "Failed to get write lock", even when it was really a read lock.
2420 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
2421 of 0600. This has been changed to use the value of the lockfile_mode
2422 option (which defaults to 0600).
2424 PH/08 Applied small patch from the Sieve maintainer.
2426 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
2427 folder from quota calculations, a direct delivery into this folder messed
2428 up the contents of the maildirsize file. This was because the regex was
2429 used only to exclude .Trash (or whatever) when the size of the mailbox
2430 was calculated. There was no check that a delivery was happening into an
2431 excluded directory. This bug has been fixed by ignoring all quota
2432 processing for deliveries into excluded directories.
2434 PH/10 Added the maildirfolder_create_regex option to appendfile.
2440 PH/01 The code for finding all the local interface addresses on a FreeBSD
2441 system running IPv6 was broken. This may well have applied to all BSD
2442 systems, as well as to others that have similar system calls. The broken
2443 code found IPv4 interfaces correctly, but gave incorrect values for the
2444 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
2445 that it would not match correctly against @[] and not recognize the IPv6
2448 PH/02 The ipliteral router was not recognizing addresses of the form user@
2449 [ipv6:....] because it didn't know about the "ipv6:" prefix.
2451 PH/03 Added disable_ipv6.
2453 PH/04 Changed $reply_address to use the raw form of the headers instead of the
2454 decoded form, because it is most often used to construct To: headers
2455 lines in autoreplies, and the decoded form may well be syntactically
2456 invalid. However, $reply_address has leading white space removed, and all
2457 newlines turned into spaces so that the autoreply transport does not
2460 PH/05 If group was specified without a user on a router, and no group or user
2461 was specified on a transport, the group from the router was ignored.
2463 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
2464 for visible compile-time settings that can be used to change these
2465 numbers, for those that want even more. Backwards compatibility with old
2466 spool files has been maintained. However, going back to a previous Exim
2467 release will lost any variables that are in spool files.
2469 PH/07 Two small changes when running in the test harness: increase delay when
2470 passing a TCP/IP connection to a new process, in case the original
2471 process has to generate a bounce, and remove special handling of
2472 127.0.0.2 (sic), which is no longer necessary.
2474 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
2475 be the same on different OS.
2477 PH/09 Moved a debug statement in filter processing to avoid a race problem when
2480 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
2481 whether --show-vars was specified or not
2483 JJ/02 exipick: Added support for new ACL variable spool format introduced
2486 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
2487 syntactically invalid From: or Reply-to: line, and a filter used this to
2488 generate an autoreply, and therefore failed to obtain an address for the
2489 autoreply, Exim could try to deliver to a non-existent relative file
2490 name, causing unrelated and misleading errors. What now happens is that
2491 it logs this as a hard delivery error, but does not attempt to create a
2494 PH/11 The exinext utility has a -C option for testing purposes, but although
2495 the given file was scanned by exinext itself; it wasn't being passed on
2496 when Exim was called.
2498 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
2499 an end-of-file indication when reading a command response.
2501 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
2502 compiled. In many other places in Exim, IPv6 addresses are always
2503 recognized, so I have changed this. It also means that IPv4 domain
2504 literals of the form [IPV4:n.n.n.n] are now always recognized.
2506 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
2507 used if the router is not running as root, for example, when verifying at
2508 ACL time, or when using -bh. The debugging output from this situation was
2509 non-existent - all you got was a failure to exec. I have made two
2512 (a) Failures to set uid/gid, the current directory, or a process leader
2513 in a subprocess such as that created by queryprogram now generate
2514 suitable debugging ouput when -d is set.
2516 (b) The queryprogram router detects when it is not running as root,
2517 outputs suitable debugging information if -d is set, and then runs
2518 the subprocess without attempting to change uid/gid.
2520 PH/15 Minor change to Makefile for building test_host (undocumented testing
2523 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
2524 additional section of a DNS packet that returns MX or SRV records.
2525 Instead, it always explicitly searches for A/AAAA records. This avoids
2526 major problems that occur when a DNS server includes only records of one
2527 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
2528 fixed another bug: if SRV records were looked up and the corresponding
2529 address records were *not* found in the additional section, the port
2530 values from the SRV records were lost.
2532 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
2533 using the correct key (the original address) when searching the retry
2534 rules in order to find which one to use for generating the retry hint.
2536 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
2537 adding the default one. Similarly, if it contains a Reply-To: header, the
2538 errors_reply_to option, if set, is not used.
2540 PH/19 When calculating a retry time, Exim used to measure the "time since
2541 failure" by looking at the "first failed" field in the retry record. Now
2542 it does not use this if it is later than than the arrival time of the
2543 message. Instead it uses the arrival time. This makes for better
2544 behaviour in cases where some deliveries succeed, thus re-setting the
2545 "first failed" field. An example is a quota failure for a huge message
2546 when small messages continue to be delivered. Without this change, the
2547 "time since failure" will always be short, possible causing more frequent
2548 delivery attempts for the huge message than are intended.
2549 [Note: This change was subsequently modified - see PH/04 for 4.62.]
2551 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
2552 $1, $2, $3) because the numerical variables can be reset during some
2553 expansion items (e.g. "match"), thereby losing the authentication data.
2555 PH/21 Make -bV show the size of off_t variables so that the test suite can
2556 decide whether to run tests for quotas > 2G.
2558 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
2559 mailbox_size, and mailbox_filecount in the appendfile transport. If a
2560 filecount value is greater than 2G or if a quota value is greater than 2G
2561 on a system where the size of off_t is not greater than 4, a panic error
2564 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
2565 never match. The debug and -bh output now contains an explicit error
2566 message indicating a malformed IPv4 address or mask.
2568 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
2569 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
2570 PH/23 above applies.
2572 PH/25 Do not write to syslog when running in the test harness. The only
2573 occasion when this arises is a failure to open the main or panic logs
2574 (for which there is an explicit test).
2576 PH/26 Added the /no_tell option to "control=freeze".
2578 PH/27 If a host name lookup failed very early in a connection, for example, if
2579 the IP address matched host_lookup and the reverse lookup yielded a name
2580 that did not have a forward lookup, an error message of the form "no IP
2581 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
2582 could be logged. Now it outputs the IP address instead of "NULL".
2584 PH/28 An enabling patch from MH: add new function child_open_exim2() which
2585 allows the sender and the authenticated sender to be set when
2586 submitting a message from within Exim. Since child_open_exim() is
2587 documented for local_scan(), the new function should be too.
2589 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
2590 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
2591 results in an empty string is now treated as unset.
2593 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
2595 PH/31 Added sender_verify_fail logging option.
2597 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
2598 needed by RFC 822 but not by RFC 2822 was commented out. I have now
2599 tidied the source and removed it altogether.
2601 PH/33 When a queue run was abandoned because the load average was too high, a
2602 log line was always written; now it is written only if the queue_run log
2603 selector is set. In addition, the log line for abandonment now contains
2604 information about the queue run such as the pid. This is always present
2605 in "start" and "stop" lines but was omitted from the "abandon" line.
2607 PH/34 Omit spaces between a header name and the colon in the error message that
2608 is given when verify = headers_syntax fails (if there are lots of them,
2609 the message gets confusing).
2611 PH/35 Change the default for dns_check_names_pattern to allow slashes within
2612 names, as there are now some PTR records that contain slashes. This check
2613 is only to protect against broken name servers that fall over on strange
2614 characters, so the fact that it applies to all lookups doesn't matter.
2616 PH/36 Now that the new test suite is complete, we can remove some of the
2617 special code in Exim that was needed for the old test suite. For example,
2618 sorting DNS records because real resolvers return them in an arbitrary
2619 order. The new test suite's fake resolver always returns records in the
2622 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
2623 bounces) except when queue_only is set, to avoid logging races between
2624 the different processes.
2626 PH/38 Panic-die if .include specifies a non-absolute path.
2628 PH/39 A tweak to the "H" retry rule from its user.
2630 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
2631 a label. They prevented compilation on older perls.
2633 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
2634 a warning to be raised on newish perls.
2636 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
2637 on queue. Changes to match documented behaviour of showing count of
2638 messages matching specified criteria.
2640 PH/40 Changed the default ident timeout from 30s to 5s.
2642 PH/41 Added support for the use of login_cap features, on those BSD systems
2643 that have them, for controlling the resources used by pipe deliveries.
2645 PH/42 The content-scanning code uses fopen() to create files in which to put
2646 message data. Previously it was not paying any attention to the mode of
2647 the files. Exim runs with umask(0) because the rest of the code creates
2648 files with open(), and sets the required mode explicitly. Thus, these
2649 files were ending up world-writeable. This was not a big issue, because,
2650 being within the spool directory, they were not world-accessible. I have
2651 created a function called modefopen, which takes an additional mode
2652 argument. It sets umask(777), creates the file, chmods it to the required
2653 mode, then resets the umask. All the relevant calls to fopen() in the
2654 content scanning code have been changed to use this function.
2656 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
2657 to 24 hours. This avoids potential overflow problems when processing G
2658 and H retry rules. I suspect nobody ever tinkers with this value.
2660 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
2662 PH/45 When the plaintext authenticator is running as a client, the server's
2663 challenges are checked to ensure they are valid base64 strings. By
2664 default, the authentication attempt is cancelled if an invalid string is
2665 received. Setting client_ignore_invalid_base64 true ignores these errors.
2666 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
2667 they are received. Thus, the responses can be made to depend on the
2668 challenges. If an invalid string is ignored, an empty string is placed in
2671 PH/46 Messages that are created by the autoreply transport now contains a
2672 References: header, in accordance with RFCs 2822 and 3834.
2674 PH/47 Added authenticated_sender_force to the smtp transport.
2676 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
2678 PH/49 Installed latest patch from the Sieve maintainer.
2680 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
2681 was also set, the appendfile transport was unnecessarily scanning a
2682 directory of message files (e.g. for maildir delivery) to find the count
2683 of files (along with the size), even though it did not need this
2684 information. It now does the scan only if it needs to find either the
2685 size of the count of files.
2687 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
2689 PH/52 Two bugs concerned with error handling when the smtp transport is
2692 (i) Exim was not creating retry information for temporary errors given
2693 for individual recipients after the DATA command when the smtp transport
2694 was used in LMTP mode. This meant that they could be retried too
2695 frequently, and not timed out correctly.
2697 (ii) Exim was setting the flag that allows error details to be returned
2698 for LMTP errors on RCPT commands, but not for LMTP errors for individual
2699 recipients that were returned after the DATA command.
2701 PH/53 This is related to PH/52, but is more general: for any failing address,
2702 when detailed error information was permitted to be returned to the
2703 sender, but the error was temporary, then after the final timeout, only
2704 "retry timeout exceeded" was returned. Now it returns the full error as
2705 well as "retry timeout exceeded".
2707 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
2708 do this, and (what is worse) MTAs that accept it.
2710 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
2711 will now be deprecated.
2713 PH/56 New os.c-cygwin from the Cygwin maintainer.
2715 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
2716 formats (previously only available in exim formats via -bpr, -bpru,
2717 and -bpra. Now also available in native and exiqgrep formats)
2719 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
2720 with very large, slow to parse queues
2722 JJ/08 exipick: added ! as generic prefix to negate any criteria format
2724 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
2726 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
2727 responses to authentication challenges, though it was showing the
2728 challenges; (ii) I've removed the CR characters from the debug output for
2731 PH/58 Allow for the insertion of a newline as well as a space when a string
2732 is turned into more than one encoded-word during RFC 2047 encoding. The
2733 Sieve code now uses this.
2735 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
2736 data_4xx, lost_connection, tls_required.
2738 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
2739 message was being sent as an SMTP response.
2741 PH/61 Add -l and -k options to exicyclog.
2743 PH/62 When verifying, if an address was redirected to one new address, so that
2744 verification continued, and the new address failed or deferred after
2745 having set something in $address_data, the value of $address_data was not
2746 passed back to the ACL. This was different to the case when no
2747 redirection occurred. The value is now passed back in both cases.
2749 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
2750 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
2751 use login_cap.h, so on its own it isn't the distinguishing feature. The
2752 new name refers directly to the setclassresources() function.
2754 PH/65 Added configuration files for NetBSD3.
2756 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
2758 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
2759 is preferred over IPv4.
2761 PH/68 The bounce_return_message and bounce_return_body options were not being
2762 honoured for bounces generated during the reception of non-SMTP messages.
2763 In particular, this applied to messages rejected by the ACL. This bug has
2764 been fixed. However, if bounce_return_message is true and bounce_return_
2765 body is false, the headers that are returned for a non-SMTP message
2766 include only those that have been read before the error was detected.
2767 (In the case of an ACL rejection, they have all been read.)
2769 PH/69 The HTML version of the specification is now built in a directory called
2770 spec_html instead of spec.html, because the latter looks like a path with
2771 a MIME-type, and this confuses some software.
2773 PH/70 Catch two compiler warnings in sieve.c.
2775 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
2776 function verify_get_ident() calls ip_connect() to connect a socket, but
2777 if the "connect()" function timed out, ip_connect() used to close the
2778 socket. However, verify_get_ident() also closes the socket later, and in
2779 between Exim writes to the log, which may get opened at this point. When
2780 the socket was closed in ip_connect(), the log could get the same file
2781 descriptor number as the socket. This naturally causes chaos. The fix is
2782 not to close the socket in ip_connect(); the socket should be closed by
2783 the function that creates it. There was only one place in the code where
2784 this was missing, in the iplookup router, which I don't think anybody now
2785 uses, but I've fixed it anyway.
2787 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
2788 well as to direct DNS lookups. Otherwise the handling of names in host
2789 lists is inconsistent and therefore confusing.
2795 PH/01 Two changes to the default runtime configuration:
2797 (1) Move the checks for relay_from_hosts and authenticated clients from
2798 after to before the (commented out) DNS black list checks.
2800 (2) Add control=submission to the relay_from_hosts and authenticated
2801 clients checks, on the grounds that messages accepted by these
2802 statements are most likely to be submissions.
2804 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
2806 (1) Generate an error if the third argument for the ${prvs expansion is
2809 (2) Treat a missing third argument of ${prvscheck as if it were an empty
2812 (3) Reset the variables that are obtained from the first argument of
2813 ${prvscheck and used in the second argument before leaving the code,
2814 because their memory is reclaimed, so using them afterwards may do
2817 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
2818 one (it's much easier than Tom thought :-).
2820 (5) Because of (4), we can now allow for the use of $prvscheck_result
2821 inside the third argument.
2823 PH/03 For some reason, the default setting of PATH when running a command from
2824 a pipe transport was just "/usr/bin". I have changed it to
2827 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
2828 anything to be listed in the output from -bV.
2830 PH/05 When a filter generated an autoreply, the entire To: header line was
2831 quoted in the delivery log line, like this:
2833 => >A.N.Other <ano@some.domain> <original@ddress> ...
2835 This has been changed so that it extracts the operative address. There
2836 may be more than one such address. If so, they are comma-separated, like
2839 => >ano@some.domain,ona@other.domain <original@ddress> ...
2841 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
2842 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
2843 not being looked up in the rDNS to get a host name, Exim was showing the
2844 IP address twice in Received: lines, even though the IP addresses were
2845 identical. For example:
2847 Received: from [1.2.3.4] (helo=[1.2.3.4])
2849 However, if the real host name was known, it was omitting the HELO data
2850 if it matched the actual IP address. This has been tidied up so that it
2851 doesn't show the same IP address twice.
2853 PH/07 When both +timestamp and +memory debugging was on, the value given by
2854 $tod_xxx expansions could be wrong, because the tod_stamp() function was
2855 called by the debug printing, thereby overwriting the timestamp buffer.
2856 Debugging no longer uses the tod_stamp() function when +timestamp is set.
2858 PH/08 When the original message was included in an autoreply transport, it
2859 always said "this is a copy of the message, including all the headers",
2860 even if body_only or headers_only was set. It now gives an appropriate
2863 PH/09 Applied a patch from the Sieve maintainer which:
2865 o fixes some comments
2866 o adds the (disabled) notify extension core
2867 o adds some debug output for the result of if/elsif tests
2868 o points to the current vacation draft in the documentation
2869 and documents the missing references header update
2873 o fixes a bug in processing the envelope test (when testing
2874 multiple envelope elements, the last element determinted the
2877 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
2878 Electronic Mail") by including:
2880 Auto-submitted: auto-generated
2882 in the messages that it generates (bounce messages and others, such as
2883 warnings). In the case of bounce messages for non-SMTP mesages, there was
2884 also a typo: it was using "Auto_submitted" (underscore instead of
2885 hyphen). Since every message generated by Exim is necessarily in response
2886 to another message, thes have all been changed to:
2888 Auto-Submitted: auto-replied
2890 in accordance with these statements in the RFC:
2892 The auto-replied keyword:
2894 - SHOULD be used on messages sent in direct response to another
2895 message by an automatic process,
2897 - MUST NOT be used on manually-generated messages,
2899 - MAY be used on Delivery Status Notifications (DSNs) and Message
2900 Disposition Notifications (MDNs),
2902 - MUST NOT be used on messages generated by automatic or periodic
2903 processes, except for messages which are automatic responses to
2906 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
2907 to the default Received: header definition.
2909 PH/12 Added log selector acl_warn_skipped (default on).
2911 PH/13 After a successful wildlsearch lookup, discard the values of numeric
2912 variables because (a) they are in the wrong storage pool and (b) even if
2913 they were copied, it wouldn't work properly because of the caching.
2915 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
2916 checking when decoding. Apparently there are clients that generate
2917 overlong encoded strings. Why am I not surprised?
2919 PH/15 If the first argument of "${if match_address" was not empty, but did not
2920 contain an "@" character, Exim crashed. Now it writes a panic log message
2921 and treats the condition as false.
2923 PH/16 In autoreply, treat an empty string for "once" the same as unset.
2925 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
2926 extension "envelope-auth". The code is finished and in agreement with
2927 other implementations, but there is no documentation so far and in fact,
2928 nobody wrote the draft yet. This extension is currently #undef'ed, thus
2929 not changing the active code.
2931 Print executed "if" and "elsif" statements when debugging is used. This
2932 helps a great deal to understand what a filter does.
2934 Document more things not specified clearly in RFC3028. I had all this
2935 sorted out, when out of a sudden new issues came to my mind. Oops."
2937 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
2940 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
2941 canonical form (as documented). However, after a host name lookup from
2942 the IP address, check_host() was doing a simple string comparison with
2943 addresses acquired from the DNS when checking that the found name did
2944 have the original IP as one of its addresses. Since any found IPv6
2945 addresses are likely to be in abbreviated form, the comparison could
2946 fail. Luckily, there already exists a function for doing the comparison
2947 by converting both addresses to binary, so now that is used instead of
2948 the text comparison.
2950 PH/20 There was another similar case to PH/19, when a complete host name was
2951 given in a host list; looking up its IP address could give an abbreviated
2952 form, whereas the current host's name might or might not be abbreviated.
2953 The same fix has been applied.
2959 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
2960 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
2963 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
2964 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
2966 PH/03 Typo: missing ".o" in src/pcre/Makefile.
2968 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
2969 header line, restrict the check to what is listed in RFCs 2369 and 2929.
2970 Also, for "Auto-Submitted", treat anything other than "no" as
2971 non-personal, in accordance with RFC 3834. (Previously it treated
2972 anything starting "auto-" as non-personal.)
2974 TF/01 The control=submission/name=... option had a problem with syntax
2975 errors if the name included a slash character. The /name= option
2976 now slurps the rest of the string, so it can include any characters
2977 but it must come last in the list of options (after /sender_retain
2980 PH/05 Some modifications to the interface to the fake nameserver for the new
2988 TK/01 Added the "success_on_redirect" address verification option. See
2989 NewStuff for rationale and an example.
2991 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
2993 PH/02 Patch to exigrep to allow it to work on syslog lines.
2995 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
2996 fread() to skip over the body file's header line, because in Cygwin the
2997 header line is locked and is inaccessible.
2999 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
3000 co-exist for some time) to make it clear that it is the Exim ID that is
3001 referenced, not the Message-ID: header line.
3003 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
3004 string_format() function, because snprintf() does not exist on all
3007 PH/06 The use of forbid_filter_existstest now also locks out the use of the
3008 ${stat: expansion item.
3010 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
3011 protocol synchronization error", to keep the pedants happy.
3013 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
3014 well as for IRIX systems, when gcc is being used. See the host.c source
3017 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
3019 PH/10 Named domain lists were not working if used in a queue_smtp_domains
3022 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
3023 transport and to the smtp transport in LMTP mode.
3025 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
3027 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
3028 run a filter in a subprocess. This could lead to confusion in subsequent
3029 lookups in the parent process. There should also be a search_tidyup() at
3030 the end of the subprocess.
3032 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
3033 only if the host matched helo_try_verify_hosts, which caused the
3034 verification to occur when the EHLO/HELO command was issued. The ACL just
3035 tested the remembered result. Now, if a previous verification attempt has
3036 not happened, "verify = helo" does it there and then.
3038 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
3040 TK/03 Fix log output including CR from clamd.
3042 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
3043 exist provoked a memory error which could cause a segfault.
3045 PH/15 Installed PCRE 6.2
3047 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
3049 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
3050 of the problem. Specifically, suggested +O2 rather than +O1 for the
3053 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
3055 PH/20 If a delivery was routed to a non-standard port by means of an SRV
3056 record, the port was not correctly logged when the outgoing_port log
3057 selector was set (it logged the transort's default port).
3059 PH/21 Added support for host-specific ports to manualroute, queryprogram,
3060 fallback_hosts, and "hosts" in the smtp transport.
3062 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
3063 host errors such as "Connection refused".
3065 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
3066 authentication with radiusclient 0.4.9:
3068 - Error returned from rc_read_config was caught wrongly
3069 - Username/password not passed on to radius server due to wrong length.
3071 The presumption is that some radiusclient API changes for 4.51/PH/17
3072 were not taken care of correctly. The code is still untested by me (my
3073 Linux distribution still has 0.3.2 of radiusclient), but it was
3074 contributed by a Radius user.
3076 PH/24 When doing a callout, the value of $domain wasn't set correctly when
3077 expanding the "port" option of the smtp transport.
3079 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
3080 while reading a MIME header. Thanks to Tom Hughes for a patch.
3082 PH/24 Include config.h inside local_scan.h so that configuration settings are
3085 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
3086 that in an ACL for RCPT (for example), you can examine exactly what was
3089 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
3090 commands, but it was not correctly comparing the address with the actual
3091 client host address. Thus, it would show the EHLO address in Received:
3092 header lines when this was not necessary.
3094 PH/27 Added the % operator to ${eval:}.
3096 PH/28 Exim tries to create and chdir to its spool directory when it starts;
3097 it should be ignoring failures (because with -C, for example, it has lost
3098 privilege). It wasn't ignoring creation failures other than "already
3101 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
3104 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
3105 because an input line was too long, either on its own, or by virtue of
3106 too many continuations, the temporary file was not being removed, and the
3107 return code was incorrect.
3109 PH/31 Missing "BOOL" in function definition in filtertest.c.
3111 PH/32 Applied Sieve patches from the maintainer.
3113 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
3115 PH/33 Added "verify = not_blind".
3117 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
3118 Local/Makefile (with some defaults set). These are used in built scripts
3119 such as exicyclog, but they have never been used in the exim_install
3120 script (though there are many overriding facilities there). I have
3121 arranged that the exim_install script now takes note of these two
3124 PH/35 Installed configuration files for Dragonfly.
3126 PH/36 When a locally submitted message by a trusted user did not contain a
3127 From: header, and the sender address was obtained from -f or from an SMTP
3128 MAIL command, and the trusted user did not use -F to supply a sender
3129 name, $originator_name was incorrectly used when constructing a From:
3130 header. Furthermore, $originator_name was used for submission mode
3131 messages from external hosts without From: headers in a similar way,
3132 which is clearly wrong.
3134 PH/37 Added control=suppress_local_fixups.
3136 PH/38 When log_selector = +received_sender was set, and the addition of the
3137 sender made the log line's construction buffer exactly full, or one byte
3138 less than full, an overflow happened when the terminating "\n" was
3141 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
3142 when the result of a list match is failure because a DNS lookup failed.
3144 PH/40 RM_COMMAND is now used in the building process.
3146 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
3147 the "build-* directories that it finds.
3149 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
3150 address in a domain literal was a prefix of an interface address.
3152 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
3153 when verifying a sender address, unless rewrite_headers is false.
3155 PH/44 Wrote a long comment about why errors_to addresses are verified as
3156 recipients, not senders.
3158 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
3159 the ratelimit ACL was added.
3161 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
3163 PH/46 Added extra information about PostgreSQL errors to the error string.
3165 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
3166 avoiding the need to install special zones in a real server. This is
3167 backwards compatible; if it can't find the fake resolver, it drops back.
3168 Thus, both old and new test suites can be run.
3170 TF/02 Added util/ratelimit.pl
3172 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
3173 clock is set back in time.
3175 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
3176 Candler <B.Candler@pobox.com>.
3178 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
3179 OK for addresses that are the result of redirections.
3181 PH/48 A number of further additions for the benefit of the new test suite,
3182 including a fake gethostbyname() that interfaces to the fake DNS resolver
3185 TF/06 The fix for widen_domains has also been applied to qualify_single and
3186 search_parents which are the other dnslookup options that can cause
3189 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
3192 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
3194 TF/07 Exim produced the error message "an SRV record indicated no SMTP
3195 service" if it encountered an MX record with an empty target hostname.
3196 The message is now "an MX or SRV record indicated no SMTP service".
3198 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
3199 if the DNS of the sending site is misconfigured. This is quite a
3200 common situation. This change restores the behaviour of treating a
3201 helo verification defer as a failure.
3203 PH/51 If self=fail was set on a router, the bounce message did not include the
3204 actual error message.
3210 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
3212 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
3213 command itself ended in error, the underlying message about the transport
3214 filter timeout was being overwritten with the pipe command error. Now the
3215 underlying error message should be appended to the second error message.
3217 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
3219 PH/02 Reduce the amount of output that "make" produces by default. Full output
3220 can still be requested.
3222 PH/03 The warning log line about a condition test deferring for a "warn" verb
3223 was being output only once per connection, rather than after each
3224 occurrence (because it was using the same function as for successful
3225 "warn" verbs). This seems wrong, so I have changed it.
3227 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
3228 it should not have, which might have caused a crash in the right
3229 circumstances, but probably never did.
3231 PH/04 Installed a modified version of Tony Finch's patch to make submission
3232 mode fix the return path as well as the Sender: header line, and to
3233 add a /name= option so that you can make the user's friendly name appear
3236 TF/03 Added the control = fakedefer ACL modifier.
3238 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
3239 Mark Lowes for thorough testing.
3241 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
3243 TK/03 Merged latest SRS patch from Miles Wilton.
3245 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
3246 with the definition in sysexits.h (which is #included earlier).
3247 Fortunately, Exim does not actually use EX_OK. The code used to try to
3248 preserve the sysexits.h value, by assumimg that macro definitions were
3249 scanned for macro replacements. I have been disabused of this notion,
3250 so now the code just undefines EX_OK before #including unistd.h.
3252 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
3253 in the smtp transport. When a block could not be written in a single
3254 write() function, the timeout was being re-applied to each part-write.
3255 This seems wrong - if the receiver was accepting one byte at a time it
3256 would take for ever. The timeout is now adjusted when this happens. It
3257 doesn't have to be particularly precise.
3259 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
3260 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
3262 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
3263 without a domain if the check to <postmaster@domain> fails.
3265 SC/01 Eximstats: added -xls and the ability to specify output files
3266 (patch written by Frank Heydlauf).
3268 SC/02 Eximstats: use FileHandles for outputing results.
3270 SC/03 Eximstats: allow any combination of xls, txt, and html output.
3272 SC/04 Eximstats: fixed display of large numbers with -nvr option
3274 SC/05 Eximstats: fixed merging of reports with empty tables.
3276 SC/06 Eximstats: added the -include_original_destination flag
3278 SC/07 Eximstats: removed tabs and trailing whitespace.
3280 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
3282 TK/06 MBOX spool code: Add real "From " MBOX separator line
3283 so the .eml file is really in mbox format (even though
3284 most programs do not really care). Patch from Alex Miller.
3286 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
3287 The latter is generated from $received_to and is only set if the
3288 message has one envelope recipient. SA can use these headers,
3289 obviously out-of-the-box. Patch from Alex Miller.
3291 PH/08 The ${def test on a variable was returning false if the variable's
3292 value was "0", contrary to what the specification has always said!
3293 The result should be true unless the variable is empty.
3295 PH/09 The syntax error of a character other than { following "${if
3296 def:variable_name" (after optional whitespace) was not being diagnosed.
3297 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
3298 accidental colon was present, for example, could give incorrect results.
3300 PH/10 Tidied the code in a number of places where the st_size field of a stat()
3301 result is used (not including appendfile, where other changes are about
3304 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
3305 This involved changing a lot of size variables from int to off_t. It
3306 should work with maildirs and everything.
3308 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
3309 spamd dying while we are connected to it.
3311 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
3314 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
3315 function for BDB 4.3.
3317 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
3319 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
3320 the sources for PCRE have been re-arranged and more files are now
3323 PH/15 The code I had for printing potentially long long variables in PH/11
3324 above was not the best (it lost precision). The length of off_t variables
3325 is now inspected at build time, and an appropriate printing format (%ld
3326 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
3327 to be "long long int" or "long int". This is needed for the internal
3328 formatting function string_vformat().
3330 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
3331 the configuration file to be ":syslog", then the script "guesses" where
3332 the logs files are, rather than using the compiled in default. In our
3333 case the guess is not the same as the compiled default, so the script
3334 suddenly stopped working when I started to use syslog. The patch checks
3335 to see if log_file_path is "". If so, it attempts to read it from exim
3336 with no configuration file to get the compiled in version, before it
3337 falls back to the previous guessing code."
3339 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
3340 implementing BATV in an Exim configuration. See NewStuff for the gory
3343 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
3344 Makefile that are specific to HP-UX.
3346 PH/18 If the "use_postmaster" option was set for a recipient callout together
3347 with the "random" option, the postmaster address was used as the MAIL
3348 FROM address for the random test, but not for the subsequent recipient
3349 test. It is now used for both.
3351 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
3352 patch removes a few documentation additions to RFC 3028, because the
3353 latest draft now contains them. It adds the new en;ascii-case comparator
3354 and a new error check for 8bit text in MIME parts. Comparator and
3355 require names are now matched exactly. I enabled the subaddress
3356 extension, but it is not well tested yet (read: it works for me)."
3358 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
3359 rework some of the code of TK/09 above to avoid the hardwired use of
3360 "%lld" and "long long". Replaced the call to snprintf() with a call to
3363 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
3364 records point to non-existent hosts", "retry timeout exceeded", and
3365 "retry time not reached for any host after a long failure period".
3367 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
3368 experimental DomainKeys support:
3370 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
3371 (2) On an error such as an illegally used "control", the wrong name for
3372 the control was given.
3374 These problems did NOT occur unless DomainKeys support was compiled.
3376 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
3378 PH/24 Added ${if match_ip condition.
3380 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
3381 databases so that it will be absolutely obvious if a crash occurs in the
3382 DB library. This is a regular occurrence (often caused by mis-matched
3385 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
3386 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
3387 picked up on a user's system that detects such things. There doesn't seem
3388 to be a gcc warning option for this - only an attribute that has to be
3389 put on the function's prototype. It seems that in Fedora Core 4 they have
3390 set this on a number of new functions. No doubt there will be more in due
3393 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
3394 specify a transport. However, if an address that was verified by such a
3395 router was the subject of a callout, Exim crashed because it tried to
3396 read the rcpt_include_affixes from the non-existent transport. Now it
3397 just assumes that the setting of that option is false. This bug was
3398 introduced by 4.51/PH/31.
3400 PH/28 Changed -d+all to exclude +memory, because that information is very
3401 rarely of interest, but it makes the output a lot bigger. People tend to
3402 do -d+all out of habit.
3404 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
3405 code in os-type was giving problems when libc.so lives in lib64, like on
3408 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
3409 aren't the modern standard, and it seems that some systems' include files
3410 don't always have them. Exim was already checking for some of the newer
3411 ones like T_AAAA, and defining it itself. I've added checks for all the
3412 record types that Exim uses.
3414 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
3415 not automatically generating a new one, as it is supposed to. This
3416 prevented TLS from working. If the file did exist, but contained invalid
3417 data, a new version was generated, as expected. It was only the case of a
3418 non-existent file that was broken.
3420 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
3421 with a change in libdomainkeys > 0.64.
3423 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
3424 from DNS. If the selector record carries the flag, it now has
3425 precedence over the domain-wide flag.
3427 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
3429 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
3430 the use of a port name that isn't defined in /etc/services) occurred, the
3431 message was deferred as in a normal delivery, and thus remained on the
3432 spool, instead of being failed because of the mua_wrapper setting. This
3433 is now fixed, and I tidied up some of the mua_wrapper messages at the
3436 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
3437 the messages in a hash of arrays rather than using individual hashes.
3438 This is a bit cleaner and results in dramatic memory savings, albeit
3439 at a slight CPU cost.
3441 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
3442 as requested by Marc Sherman.
3444 SC/10 Eximstats: added histograms for user specified patterns as requested
3447 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
3449 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
3450 fopen() in the content-scanning modules that did not already have it.
3456 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
3457 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
3459 TK/02 Fix ACL "control" statement not being available in MIME ACL.
3461 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
3463 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
3464 to test Sieve filters that use "vacation".
3466 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
3467 that changes the way the GnuTLS parameters are stored in the cache file.
3468 The new format can be generated externally. For backward compatibility,
3469 if the data in the cache doesn't make sense, Exim assumes it has read an
3470 old-format file, and it generates new data and writes a new file. This
3471 means that you can't go back to an older release without removing the
3474 PH/03 A redirect router that has both "unseen" and "one_time" set does not
3475 work if there are any delivery delays because "one_time" forces the
3476 parent to be marked "delivered", so its unseen clone is never tried
3477 again. For this reason, Exim now forbids the simultaneous setting of
3480 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
3481 redirected to themselves ("homonym" addresses). Read the long ChangeLog
3482 entry if you want to know the details. The fix, however, neglected to
3483 consider the case when local delivery batching is involved. The test for
3484 "previously delivered" was not happening when checking to see if an
3485 address could be batched with a previous (undelivered) one; under
3486 certain circumstances this could lead to multiple deliveries to the same
3489 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
3490 in its include files, and this causes problems building Exim.
3492 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
3493 header_syntax) but Exim was just ignoring anything given after a slash.
3494 In particular, this caused confusion with an attempt to use "verify =
3495 reverse_host_lookup/defer_ok". An error is now given when options are
3496 supplied for verify items that do not have them. (Maybe reverse_host_
3497 lookup should have a defer_ok option, but that's a different point.)
3499 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
3500 defined by RFC 821) to 2048, because there were problems with some AUTH
3501 commands, and RFC 1869 says the size should be increased for extended
3502 SMTP commands that take arguments.
3504 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
3507 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
3508 "unknown" error; now it says that the functionality isn't in the binary.
3510 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
3511 an address' error message when a string expansion fails (syntax or
3512 whatever). Otherwise the password may appear in the log. Following change
3513 PH/42 below, there is no longer a chance of it appearing in a bounce
3516 PH/11 Installed exipick version 20050225.0 from John Jetmore.
3518 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
3519 of its addresses was ever tried. (Bugzilla bug #2.)
3521 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
3522 the result incorrectly in the debug output. (It correctly added a newline
3523 to what was transported.)
3525 TF/01 Added $received_time.
3527 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
3528 commented out examples of how to interface to a virus scanner and to
3529 SpamAssassin. Also added commented examples of av_scanner and
3530 spamd_address settings.
3532 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
3533 and controls are allowed in which ACLs. There were a couple of minor
3534 errors. Some of the entries in the conditions table (which is a table of
3535 where they are NOT allowed) were getting very unwieldy; rewrote them as a
3536 negation of where the condition IS allowed.
3538 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
3540 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
3541 header file does not have a version number, so I've had to invent a new
3542 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
3543 API. The code is untested by me (my Linux distribution still has 0.3.2 of
3544 radiusclient), but it was contributed by a Radius user.
3546 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
3547 files or directories, for OpenSSL.
3549 PH/19 When an Exim process that is running as root has to create an Exim log
3550 file, it does so in a subprocess that runs as exim:exim so as to get the
3551 ownership right at creation (otherwise, other Exim processes might see
3552 the file with the wrong ownership). There was no test for failure of this
3553 fork() call, which would lead to the process getting stuck as it waited
3554 for a non-existent subprocess. Forks do occasionally fail when resources
3555 run out. I reviewed all the other calls to fork(); they all seem to check
3558 PH/20 When checking for unexpected SMTP input at connect time (before writing
3559 the banner), Exim was not dealing correctly with a non-positive return
3560 from the read() function. If the client had disconnected by this time,
3561 the result was a log entry for a synchronization error with an empty
3562 string after "input=" when read() returned zero. If read() returned -1
3563 (an event I could not check), uninitialized data bytes were printed.
3564 There were reports of junk text (parts of files, etc) appearing after
3567 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
3569 PH/22 Added support for macro redefinition, and (re)definition in between
3570 driver and ACL definitions.
3572 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
3573 forgetting to use the resulting value; it was using the unexpanded value.
3575 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
3576 hadn't been configured. The fix is from Juergen Kreileder, who
3577 understands it better than I do:
3579 "Here's what I see happening with three configured cyrus_sasl
3580 authenticators configured (plain, login, cram-md5):
3582 On startup auth_cyrus_sasl_init() gets called for each of these.
3583 This means three calls to sasl_listmech() without a specified mech_list.
3584 => SASL tests which mechs of all available mechs actually work
3585 => three warnings about OTP not working
3586 => the returned list contains: plain, login, cram-md5, digest-md5, ...
3588 With the patch, sasl_listmech() also gets called three times. But now
3589 SASL's mech_list option is set to the server_mech specified in the the
3590 authenticator. Or in other words, the answer from sasl_listmech()
3591 gets limited to just the mech you're testing for (which is different
3593 => the return list contains just 'plain' or 'login', 'cram-md5' or
3594 nothing depending on the value of ob->server_mech.
3596 I've just tested the patch: Authentication still works fine,
3597 unavailable mechs specified in the exim configuration are still
3598 caught, and the auth.log warnings about OTP are gone."
3600 PH/25 When debugging is enabled, the contents of the command line are added
3601 to the debugging output, even when log_selector=+arguments is not
3604 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
3605 answer is "GNU", and only if the return is "GNU/something" is the answer
3608 PH/27 $acl_verify_message is now set immediately after the failure of a
3609 verification in an ACL, and so is available in subsequent modifiers. In
3610 particular, the message can be preserved by coding like this:
3612 warn !verify = sender
3613 set acl_m0 = $acl_verify_message
3615 Previously, $acl_verify_message was set only while expanding "message"
3616 and "log_message" when a very denied access.
3618 PH/28 Modified OS/os.c-Linux with
3620 -#ifndef OS_LOAD_AVERAGE
3621 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
3623 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
3624 nomenclature these days.)
3626 PH/29 Installed patch from the Sieve maintainer that adds the options
3627 sieve_useraddress and sieve_subaddress to the redirect router.
3629 PH/30 In these circumstances:
3630 . Two addresses routed to the same list of hosts;
3631 . First host does not offer TLS;
3632 . First host accepts first address;
3633 . First host gives temporary error to second address;
3634 . Second host offers TLS and a TLS session is established;
3635 . Second host accepts second address.
3636 Exim incorrectly logged both deliveries with the TLS parameters (cipher
3637 and peerdn, if requested) that were in fact used only for the second
3640 PH/31 When doing a callout as part of verifying an address, Exim was not paying
3641 attention to any local part prefix or suffix that was matched by the
3642 router that accepted the address. It now behaves in the same way as it
3643 does for delivery: the affixes are removed from the local part unless
3644 rcpt_include_affixes is set on the transport.
3646 PH/32 Add the sender address, as F=<...>, to the log line when logging a
3647 timeout during the DATA phase of an incoming message.
3649 PH/33 Sieve envelope tests were broken for match types other than :is. I have
3650 applied a patch sanctioned by the Sieve maintainer.
3652 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
3653 the uid or gid is negative. A case of a negative gid caused this to be
3654 noticed. The fix allows for either to be negative.
3656 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
3657 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
3660 PH/36 The change PH/12 above was broken. Fixed it.
3662 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
3663 the grounds that routing the same address twice would always produce the
3664 same answer. This might have been true once, but it is certainly no
3665 longer true now. Routing a child address may depend on the previous
3666 routing that produced that child. Some complicated redirection strategies
3667 went wrong when messages had multiple recipients, and made Exim's
3668 behaviour dependent on the order in which the addresses were given.
3670 I have moved the duplicate checking until after the routing is complete.
3671 Exim scans the addresses that are assigned to local and remote
3672 transports, and removes any duplicates. This means that more work will be
3673 done, as duplicates will always all be routed, but duplicates are
3674 presumably rare, so I don't expect this is of any significance.
3676 For deliveries to pipes, files, and autoreplies, the duplicate checking
3677 still happens during the routing process, since they are not going to be
3680 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
3681 It corrects a timeout issue with spamd. This is Ian's comment: "The
3682 background is that sometimes spamd either never reads data from a
3683 connection it has accepted, or it never writes response data. The exiscan
3684 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
3685 blindly assumes that writes won't block so it may never time out."
3687 PH/39 Allow G after quota size as well as K and M.
3689 PH/40 The value set for $authenticated_id in an authenticator may not contain
3690 binary zeroes or newlines because the value is written to log lines and
3691 to spool files. There was no check on this. Now the value is run through
3692 the string_printing() function so that such characters are converted to
3693 printable escape sequences.
3695 PH/41 $message_linecount is a new variable that contains the total number of
3696 lines in the message. Compare $body_linecount, which is the count for the
3699 PH/42 Exim no longer gives details of delivery errors for specific addresses in
3700 bounce and delay warning messages, except in certain special cases, which
3703 (a) An SMTP error message from a remote host;
3704 (b) A message specified in a :fail: redirection;
3705 (c) A message specified in a "fail" command in a system filter;
3706 (d) A message specified in a FAIL return from the queryprogram router;
3707 (e) A message specified by the cannot_route_message router option.
3709 In these cases only, Exim does include the error details in bounce and
3710 warning messages. There are also a few cases where bland messages such
3711 as "unrouteable address" or "local delivery error" are given.
3713 PH/43 $value is now also set for the "else" part of a ${run expansion.
3715 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
3716 being worked on, but at least Exim now implements the latest version to
3719 PH/45 In a pipe transport, although a timeout while waiting for the pipe
3720 process to complete was treated as a delivery failure, a timeout while
3721 writing the message to the pipe was logged, but erroneously treated as a
3722 successful delivery. Such timeouts include transport filter timeouts. For
3723 consistency with the overall process timeout, these timeouts are now
3724 treated as errors, giving rise to delivery failures by default. However,
3725 there is now a new Boolean option for the pipe transport called
3726 timeout_defer, which, if set TRUE, converts the failures into defers for
3727 both kinds of timeout. A transport filter timeout is now identified in
3730 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
3731 systems where "make" and "gmake" are different, calling "gmake" at top
3732 level broke things. I've arranged for the value of $(MAKE) to be passed
3733 from the Makefile to this script so that it can call the same version of
3737 A note about Exim versions 4.44 and 4.50
3738 ----------------------------------------
3740 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
3741 changes of various kinds. As a consequence, a big documentation update was
3742 needed. This delayed the release for rather longer than seemed good, especially
3743 in the light of a couple of (minor) security issues. Therefore, the changes
3744 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
3745 release. So 4.44 and 4.50 are in effect two different branches that both start
3748 I have left the 4.50 change log unchanged; it contains all the changes since
3749 4.43. The change log for 4.44 is below; many of its items are identical to
3750 those for 4.50. This seems to be the most sensible way to preserve the
3751 historical information.
3757 1. Minor wording change to the doc/README.SIEVE file.
3759 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
3760 computation of the current number of files was incorrect.
3762 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
3763 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
3766 4. Give more explanation in the error message when the command for a transport
3767 filter fails to execute.
3769 5. There are several places where Exim runs a non-Exim command in a
3770 subprocess. The SIGUSR1 signal should be disabled for these processes. This
3771 was being done only for the command run by the queryprogram router. It is
3772 now done for all such subprocesses. The other cases are: ${run, transport
3773 filters, and the commands run by the lmtp and pipe transports.
3775 6. Added CONFIGURE_GROUP build-time option.
3777 7. Some older OS have a limit of 256 on the maximum number of file
3778 descriptors. Exim was using setrlimit() to set 1000 as a large value
3779 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
3780 systems. I've change it so that if it can't get 1000, it tries for 256.
3782 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
3783 was an oversight, and furthermore, ever since the addition of extra
3784 controls (e.g. 4.43/32), the checks on when to allow different forms of
3785 "control" were broken. There should now be diagnostics for all cases when a
3786 control that does not make sense is encountered.
3788 9. Added the /retain_sender option to "control=submission".
3790 10. $recipients is now available in the predata ACL (oversight).
3792 11. Tidy the search cache before the fork to do a delivery from a message
3793 received from the command line. Otherwise the child will trigger a lookup
3794 failure and thereby defer the delivery if it tries to use (for example) a
3795 cached ldap connection that the parent has called unbind on.
3797 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
3798 of $address_data from the recipient verification was clobbered by the
3799 sender verification.
3801 13. The value of address_data from a sender verification is now available in
3802 $sender_address_data in subsequent conditions in the ACL statement.
3804 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
3806 15. Added a new option "connect=<time>" to callout options, to set a different
3809 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
3810 was its contents. (It was OK if the option was not defined at all.)
3812 17. A "Completed" log line is now written for messages that are removed from
3813 the spool by the -Mrm option.
3815 18. New variables $sender_verify_failure and $recipient_verify_failure contain
3816 information about exactly what failed.
3818 19. Added -dd to debug only the daemon process.
3820 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
3821 handles timeouts, both on the server side and network timeouts. Renamed the
3822 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
3824 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
3825 It is now set to "smtps".
3827 22. $host_address is now set to the target address during the checking of
3828 ignore_target_hosts.
3830 23. When checking ignore_target_hosts for an ipliteral router, no host name was
3831 being passed; this would have caused $sender_host_name to have been used if
3832 matching the list had actually called for a host name (not very likely,
3833 since this list is usually IP addresses). A host name is now passed as
3836 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
3837 code that specifies a non-restarting handler (typically sigaction() in
3838 modern systems) in an attempt to fix a rare and obscure crash bug.
3840 25. Narrowed the window for a race in the daemon that could cause it to ignore
3841 SIGCHLD signals. This is not a major problem, because they are used only to
3842 wake it up if nothing else does.
3844 26. A malformed maildirsize file could cause Exim to calculate negative values
3845 for the mailbox size or file count. Odd effects could occur as a result.
3846 The maildirsize information is now recalculated if the size or filecount
3849 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
3850 support for a long time. Removed HAVE_SYS_VFS_H.
3852 28. Installed the latest version of exipick from John Jetmore.
3854 29. In an address list, if the pattern was not a regular expression, an empty
3855 subject address (from a bounce message) matched only if the pattern was an
3856 empty string. Non-empty patterns were not even tested. This was the wrong
3857 because it is perfectly reasonable to use an empty address as part of a
3858 database query. An empty address is now tested by patterns that are
3859 lookups. However, all the other forms of pattern expect the subject to
3860 contain a local part and a domain, and therefore, for them, an empty
3861 address still always fails if the pattern is not itself empty.
3863 30. Exim went into a mad DNS loop when attempting to do a callout where the
3864 host was specified on an smtp transport, and looking it up yielded more
3865 than one IP address.
3867 31. Re-factored the code for checking spool and log partition space into a
3868 function that finds that data and another that does the check. The former
3869 is then used to implement four new variables: $spool_space, $log_space,
3870 $spool_inodes, and $log_inodes.
3872 32. The RFC2047 encoding function was originally intended for short strings
3873 such as real names; it was not keeping to the 75-character limit for
3874 encoded words that the RFC imposes. It now respects the limit, and
3875 generates multiple encoded words if necessary. To be on the safe side, I
3876 have increased the buffer size for the ${rfc2047: expansion operator from
3879 33. It is now permitted to omit both strings after an "if" condition; if the
3880 condition is true, the result is "true". As before, when the second string
3881 is omitted, a false condition yields an empty string. This makes it less
3882 cumbersome to write custom ACL and router conditions.
3884 34. Failure to deliver a bounce message always caused it to be frozen, even if
3885 there was an errors_to setting on the router. The errors_to setting is now
3888 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
3889 canonical form (fully expanded) before being placed in
3890 $sender_host_address.
3892 36. The table in the code that translates DNS record types into text (T_A to
3893 "A" for instance) was missing entries for NS and CNAME. It is just possible
3894 that this could have caused confusion if both these types were looked up
3895 for the same domain, because the text type is used as part of Exim's
3896 per-process caching. But the chance of anyone hitting this buglet seems
3899 37. The dnsdb lookup has been extended in a number of ways.
3901 (1) There is a new type, "zns", which walks up the domain tree until it
3902 finds some nameserver records. It should be used with care.
3904 (2) There is a new type, "mxh", which is like "mx" except that it returns
3905 just the host names, not the priorities.
3907 (3) It is now possible to give a list of domains (or IP addresses) to be
3908 looked up. The behaviour when one of the lookups defers can be
3909 controlled by a keyword.
3911 (4) It is now possible to specify the separator character for use when
3912 multiple records are returned.
3914 38. The dnslists ACL condition has been extended: it is now possible to supply
3915 a list of IP addresses and/or domains to be looked up in a particular DNS
3918 39. Added log_selector=+queue_time_overall.
3920 40. When running the queue in the test harness, wait just a tad after forking a
3921 delivery process, to get repeatability of debugging output.
3923 41. Include certificate and key file names in error message when GnuTLS fails
3924 to set them up, because the GnuTLS error message doesn't include the name
3925 of the failing file when there is a problem reading it.
3927 42. Allow both -bf and -bF in the same test run.
3929 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
3931 44. The "Exiscan patch" is now merged into the mainline Exim source.
3933 45. Sometimes the final signoff response after QUIT could fail to get
3934 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
3935 < 0 before doing a fflush(). This bug looks as though it goes back to the
3936 introduction of TLS in release 3.20, but "sometimes" must have been rare
3937 because the tests only now provoked it.
3939 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
3940 (this can affect the format of dates).
3942 47. exim_tidydb, when checking for the continued existence of a message for
3943 which it has found a message-specific retry record, was not finding
3944 messages that were in split spool directories. Consequently, it was
3945 deleting retry records that should have stayed in existence.
3947 48. Steve fixed some bugs in eximstats.
3949 49. The SPA authentication driver was not abandoning authentication and moving
3950 on to the next authenticator when an expansion was forced to fail,
3951 contradicting the general specification for all authenticators. Instead it
3952 was generating a temporary error. It now behaves as specified.
3954 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
3955 (the order specifies the preference for clients). The order is now AES256,
3956 AES128, 3DES, ARCFOUR128.
3958 51. Small patch to Sieve code - explicitly set From: when generating an
3961 52. Exim crashed if a remote delivery caused a very long error message to be
3962 recorded - for instance if somebody sent an entire SpamAssassin report back
3963 as a large number of 550 error lines. This bug was coincidentally fixed by
3964 increasing the size of one of Exim's internal buffers (big_buffer) that
3965 happened as part of the Exiscan merge. However, to be on the safe side, I
3966 have made the code more robust (and fixed the comments that describe what
3969 53. Now that there can be additional text after "Completed" in log lines (if
3970 the queue_time_overall log selector is set), a one-byte patch to exigrep
3971 was needed to allow it to recognize "Completed" as not the last thing in
3974 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
3975 patch that reportedly fixes this has been added. I am not expert enough to
3976 create a test for it. This is what the patch creator wrote:
3978 "I found a little strange behaviour of ldap code when working with
3979 Windows 2003 AD Domain, where users was placed in more than one
3980 Organization Units. When I tried to give exim partial DN, the exit code
3981 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
3982 But simultaneously result of request was absolutely normal ldap result,
3983 so I produce this patch..."
3985 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
3986 REFERENCE, so I have modified the code to exclude the patch when that macro
3989 55. Some experimental protocols are using DNS PTR records for new purposes. The
3990 keys for these records are domain names, not reversed IP addresses. The
3991 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
3992 leaves it alone. Component reversal etc. now happens only for IP addresses.
3995 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
3997 57. Double the size of the debug message buffer (to 2048) so that more of very
3998 long debug lines gets shown.
4000 58. The exicyclog utility now does better if the number of log files to keep
4001 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
4003 59. Two changes related to the smtp_active_hostname option:
4005 (1) $smtp_active_hostname is now available as a variable.
4006 (2) The default for smtp_banner uses $smtp_active_hostname instead
4007 of $primary_hostname.
4009 60. The host_aton() function is supposed to be passed a string that is known
4010 to be a valid IP address. However, in the case of IPv6 addresses, it was
4011 not checking this. This is a hostage to fortune. Exim now panics and dies
4012 if the condition is not met. A case was found where this could be provoked
4013 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4014 components; fortuitously, this particular loophole had already been fixed
4015 by change 4.50/55 above.
4017 If there are any other similar loopholes, the new check in host_aton()
4018 itself should stop them being exploited. The report I received stated that
4019 data on the command line could provoke the exploit when Exim was running as
4020 exim, but did not say which command line option was involved. All I could
4021 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4022 running as the user.
4025 61. There was a buffer overflow vulnerability in the SPA authentication code
4026 (which came originally from the Samba project). I have added a test to the
4027 spa_base64_to_bits() function which I hope fixes it.
4030 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
4031 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
4033 63. The daemon start-up calls getloadavg() while still root for those OS that
4034 need the first call to be done as root, but it missed one case: when
4035 deliver_queue_load_max is set with deliver_drop_privilege. This is
4036 necessary for the benefit of the queue runner, because there is no re-exec
4037 when deliver_drop_privilege is set.
4039 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
4040 This has been fixed.
4042 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4043 was in use, was not putting the data itself into the right store pool;
4044 consequently, it could be overwritten for a subsequent message in the same
4045 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4048 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
4050 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
4051 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
4052 the function were treating the return as a boolean value, which happened to
4053 work because 0=false and not-0=true, but is not correct code.
4055 68. The host_aton() function was not handling scoped IPv6 addresses (those
4056 with, for example, "%eth0" on the end) correctly.
4058 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
4059 negated items (that is, ~something) in unsigned ints. Some compilers
4060 apparently mutter when there is no cast.
4062 70. If an address verification called from an ACL failed, and did not produce a
4063 user-specific message (i.e. there was only a "system" message), nothing was
4064 put in $acl_verify_message. In this situation, it now puts the system
4067 71. Change 4.23/11 added synchronization checking at the start of an SMTP
4068 session; change 4.31/43 added the unwanted input to the log line - except
4069 that it did not do this in the start of session case. It now does.
4071 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
4072 This is wrong and can cause the other end to generate a synchronization
4073 error if it is another Exim or anything else that does the synchronization
4074 check. A QUIT command is no longer sent after a timeout.
4076 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
4077 during host lookups.
4079 74. The defer_ok option of callout verification was not working if it was used
4080 when verifying addresses in header lines, that is, for this case:
4082 verify = header_sender/callout=defer_ok
4084 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
4085 those file descriptors could be used for SMTP connections. If anything
4086 wrote to stderr (the example that came up was "warn" in embedded Perl), it
4087 could be sent to the SMTP client, causing chaos. The daemon now opens
4088 stdin, stdout, and stderr to /dev/null when it puts itself into the
4091 76. Arrange for output from Perl's "warn" command to be written to Exim's main
4092 log by default. The user can override this with suitable Perl magic.
4094 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
4095 the log message when discard triggers, was not working for the DATA ACL or
4096 for the non-SMTP ACL.
4098 78. Error message wording change in sieve.c.
4100 79. If smtp_accept_max_per_host was set, the number of connections could be
4101 restricted to fewer than expected, because the daemon was trying to set up
4102 a new connection before checking whether the processes handling previous
4103 connections had finished. The check for completed processes is now done
4104 earlier. On busy systems, this bug wouldn't be noticed because something
4105 else would have woken the daemon, and it would have reaped the completed
4108 80. If a message was submitted locally by a user whose login name contained one
4109 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
4110 It caused a spool format error. I have fixed the spool reading code. A
4111 related problem was that the "from" clause in the Received: line became
4112 illegal because of the space(s). It is now covered by ${quote_local_part.
4114 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
4117 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
4119 83. Patch from Sieve maintainer to fix unterminated string problem in
4120 "vacation" handling.
4122 84. Some minor changes to the Linux configuration files to help with other
4123 OS variants using glibc.
4125 85. One more patch for Sieve to update vacation handling to latest spec.
4128 ----------------------------------------------------
4129 See the note above about the 4.44 and 4.50 releases.
4130 ----------------------------------------------------
4136 1. Change 4.43/35 introduced a bug that caused file counts to be
4137 incorrectly computed when quota_filecount was set in an appendfile
4140 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4141 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4144 3. Give more explanation in the error message when the command for a transport
4145 filter fails to execute.
4147 4. There are several places where Exim runs a non-Exim command in a
4148 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4149 was being done only for the command run by the queryprogram router. It is
4150 now done for all such subprocesses. The other cases are: ${run, transport
4151 filters, and the commands run by the lmtp and pipe transports.
4153 5. Some older OS have a limit of 256 on the maximum number of file
4154 descriptors. Exim was using setrlimit() to set 1000 as a large value
4155 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4156 systems. I've change it so that if it can't get 1000, it tries for 256.
4158 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4159 was an oversight, and furthermore, ever since the addition of extra
4160 controls (e.g. 4.43/32), the checks on when to allow different forms of
4161 "control" were broken. There should now be diagnostics for all cases when a
4162 control that does not make sense is encountered.
4164 7. $recipients is now available in the predata ACL (oversight).
4166 8. Tidy the search cache before the fork to do a delivery from a message
4167 received from the command line. Otherwise the child will trigger a lookup
4168 failure and thereby defer the delivery if it tries to use (for example) a
4169 cached ldap connection that the parent has called unbind on.
4171 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4172 of $address_data from the recipient verification was clobbered by the
4173 sender verification.
4175 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4176 was its contents. (It was OK if the option was not defined at all.)
4178 11. A "Completed" log line is now written for messages that are removed from
4179 the spool by the -Mrm option.
4181 12. $host_address is now set to the target address during the checking of
4182 ignore_target_hosts.
4184 13. When checking ignore_target_hosts for an ipliteral router, no host name was
4185 being passed; this would have caused $sender_host_name to have been used if
4186 matching the list had actually called for a host name (not very likely,
4187 since this list is usually IP addresses). A host name is now passed as
4190 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4191 code that specifies a non-restarting handler (typically sigaction() in
4192 modern systems) in an attempt to fix a rare and obscure crash bug.
4194 15. Narrowed the window for a race in the daemon that could cause it to ignore
4195 SIGCHLD signals. This is not a major problem, because they are used only to
4196 wake it up if nothing else does.
4198 16. A malformed maildirsize file could cause Exim to calculate negative values
4199 for the mailbox size or file count. Odd effects could occur as a result.
4200 The maildirsize information is now recalculated if the size or filecount
4203 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4204 support for a long time. Removed HAVE_SYS_VFS_H.
4206 18. Updated exipick to current release from John Jetmore.
4208 19. Allow an empty sender to be matched against a lookup in an address list.
4209 Previously the only cases considered were a regular expression, or an
4212 20. Exim went into a mad DNS lookup loop when doing a callout where the
4213 host was specified on the transport, if the DNS lookup yielded more than
4216 21. The RFC2047 encoding function was originally intended for short strings
4217 such as real names; it was not keeping to the 75-character limit for
4218 encoded words that the RFC imposes. It now respects the limit, and
4219 generates multiple encoded words if necessary. To be on the safe side, I
4220 have increased the buffer size for the ${rfc2047: expansion operator from
4223 22. Failure to deliver a bounce message always caused it to be frozen, even if
4224 there was an errors_to setting on the router. The errors_to setting is now
4227 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4228 canonical form (fully expanded) before being placed in
4229 $sender_host_address.
4231 24. Updated eximstats to version 1.33
4233 25. Include certificate and key file names in error message when GnuTLS fails
4234 to set them up, because the GnuTLS error message doesn't include the name
4235 of the failing file when there is a problem reading it.
4237 26. Expand error message when OpenSSL has problems setting up cert/key files.
4240 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
4241 (this can affect the format of dates).
4243 28. exim_tidydb, when checking for the continued existence of a message for
4244 which it has found a message-specific retry record, was not finding
4245 messages that were in split spool directories. Consequently, it was
4246 deleting retry records that should have stayed in existence.
4248 29. eximstats updated to version 1.35
4249 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
4250 1.35 - bugfix such that pie charts by volume are generated correctly
4252 30. The SPA authentication driver was not abandoning authentication and moving
4253 on to the next authenticator when an expansion was forced to fail,
4254 contradicting the general specification for all authenticators. Instead it
4255 was generating a temporary error. It now behaves as specified.
4257 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
4258 (the order specifies the preference for clients). The order is now AES256,
4259 AES128, 3DES, ARCFOUR128.
4261 31. Small patch to Sieve code - explicitly set From: when generating an
4264 32. Exim crashed if a remote delivery caused a very long error message to be
4265 recorded - for instance if somebody sent an entire SpamAssassin report back
4266 as a large number of 550 error lines. This bug was coincidentally fixed by
4267 increasing the size of one of Exim's internal buffers (big_buffer) that
4268 happened as part of the Exiscan merge. However, to be on the safe side, I
4269 have made the code more robust (and fixed the comments that describe what
4272 33. Some experimental protocols are using DNS PTR records for new purposes. The
4273 keys for these records are domain names, not reversed IP addresses. The
4274 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4275 leaves it alone. Component reversal etc. now happens only for IP addresses.
4278 34. The host_aton() function is supposed to be passed a string that is known
4279 to be a valid IP address. However, in the case of IPv6 addresses, it was
4280 not checking this. This is a hostage to fortune. Exim now panics and dies
4281 if the condition is not met. A case was found where this could be provoked
4282 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4283 components; fortuitously, this particular loophole had already been fixed
4284 by change 4.50/55 or 4.44/33 above.
4286 If there are any other similar loopholes, the new check in host_aton()
4287 itself should stop them being exploited. The report I received stated that
4288 data on the command line could provoke the exploit when Exim was running as
4289 exim, but did not say which command line option was involved. All I could
4290 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4291 running as the user.
4294 35. There was a buffer overflow vulnerability in the SPA authentication code
4295 (which came originally from the Samba project). I have added a test to the
4296 spa_base64_to_bits() function which I hope fixes it.
4299 36. The daemon start-up calls getloadavg() while still root for those OS that
4300 need the first call to be done as root, but it missed one case: when
4301 deliver_queue_load_max is set with deliver_drop_privilege. This is
4302 necessary for the benefit of the queue runner, because there is no re-exec
4303 when deliver_drop_privilege is set.
4305 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4306 was in use, was not putting the data itself into the right store pool;
4307 consequently, it could be overwritten for a subsequent message in the same
4308 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4311 38. Sometimes the final signoff response after QUIT could fail to get
4312 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4313 < 0 before doing a fflush(). This bug looks as though it goes back to the
4314 introduction of TLS in release 3.20, but "sometimes" must have been rare
4315 because the tests only now provoked it.
4321 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
4322 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
4323 Now it returns an integer. A number of calls were still expecting a T/F
4324 return. Fortuitously, in all cases, the tests worked in OK situations,
4325 which is the norm. However, things would have gone wrong on any write
4326 failures on the smtp file descriptor. This function is used when sending
4327 messages over SMTP and also when doing verify callouts.
4329 2. When Exim is called to do synchronous delivery of a locally submitted
4330 message (the -odf or -odi options), it no longer closes stderr before doing
4333 3. Implemented the mua_wrapper option.
4335 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
4337 5. Implemented the functions header_remove(), header_testname(),
4338 header_add_at_position(), and receive_remove_recipient(), and exported them
4341 6. If an ACL "warn" statement specified the addition of headers, Exim already
4342 inserted X-ACL-Warn: at the start if there was no header name. However, it
4343 was not making this test for the second and subsequent header lines if
4344 there were newlines in the string. This meant that an invalid header could
4345 be inserted if Exim was badly configured.
4347 7. Allow an ACL "warn" statement to add header lines at the start or after all
4348 the Received: headers, as well as at the end.
4350 8. Added the rcpt_4xx retry error code.
4352 9. Added postmaster_mailfrom=xxx to callout verification option.
4354 10. Added mailfrom=xxxx to the callout verification option, for verify=
4357 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
4358 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
4360 12. Inserted some casts to stop certain compilers warning when using pointer
4361 differences as field lengths or precisions in printf-type calls (mostly
4362 affecting debugging statements).
4364 13. Added optional readline() support for -be (dynamically loaded).
4366 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
4367 same clock tick as a message's arrival, so that its received time was the
4368 same as the "first fail" time on the retry record, and that message
4369 remained on the queue past the ultimate address timeout, every queue runner
4370 would try a delivery (because it was past the ultimate address timeout) but
4371 after another failure, the ultimate address timeout, which should have then
4372 bounced the address, did not kick in. This was a "< instead of <=" error;
4373 in most cases the first failure would have been in the next clock tick
4374 after the received time, and all would be well.
4376 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
4377 being recognized when the domain list was tested by the match_domain
4378 condition in an expansion string.
4380 16. Added the ${str2b64: operator.
4382 17. Exim was always calling setrlimit() to set a large limit for the number of
4383 processes, without checking whether the existing limit was already
4384 adequate. (It did check for the limit on file descriptors.) Furthermore,
4385 errors from getrlimit() and setrlimit() were being ignored. Now they are
4386 logged to the main and panic logs, but Exim does carry on, to try to do its
4387 job under whatever limits there are.
4389 18. Imported PCRE 5.0.
4391 19. Trivial typo in log message " temporarily refused connection" (the leading
4394 20. If the log selector return_path_on_delivery was set and an address was
4395 redirected to /dev/null, the delivery process crashed because it assumed
4396 that a return path would always be set for a "successful" delivery. In this
4397 case, the whole delivery is bypassed as an optimization, and therefore no
4400 21. Internal re-arrangement: the function for sending a challenge and reading
4401 a response while authentication was assuming a zero-terminated challenge
4402 string. It's now changed to take a pointer and a length, to allow for
4403 binary data in such strings.
4405 22. Added the cyrus_sasl authenticator (code supplied by MBM).
4407 23. Exim was not respecting finduser_retries when seeking the login of the
4408 uid under which it was called; it was always trying 10 times. (The default
4409 setting of finduser_retries is zero.) Also, it was sleeping after the final
4410 failure, which is pointless.
4412 24. Implemented tls_on_connect_ports.
4414 25. Implemented acl_smtp_predata.
4416 26. If the domain in control=submission is set empty, Exim assumes that the
4417 authenticated id is a complete email address when it generates From: or
4418 Sender: header lines.
4420 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
4421 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
4422 chown and chgrp in /bin and hostname in /usr/bin.
4424 28. Exim was keeping the "process log" file open after each use, just as it
4425 does for the main log. This opens the possibility of it remaining open for
4426 long periods when the USR1 signal hits a daemon. Occasional processlog
4427 errors were reported, that could have been caused by this. Anyway, it seems
4428 much more sensible not to leave this file open at all, so that is what now
4431 29. The long-running daemon process does not normally write to the log once it
4432 has entered its main loop, and it closes the log before doing so. This is
4433 so that log files can straightforwardly be renamed and moved. However,
4434 there are a couple of unusual error situations where the daemon does write
4435 log entries, and I had neglected to close the log afterwards.
4437 30. The text of an SMTP error response that was received during a remote
4438 delivery was being truncated at 512 bytes. This is too short for some of
4439 the long messages that one sometimes sees. I've increased the limit to
4442 31. It is now possible to make retry rules that apply only when a message has a
4443 specific sender, in particular, an empty sender.
4445 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
4446 it possible to be selective about when SMTP synchronization is enforced.
4448 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
4450 32. Implemented hosts_connection_nolog.
4452 33. Added an ACL for QUIT.
4454 34. Setting "delay_warning=" to disable warnings was not working; it gave a
4457 35. Added mailbox_size and mailbox_filecount to appendfile.
4459 36. Added control = no_multiline_responses to ACLs.
4461 37. There was a bug in the logic of the code that waits for the clock to tick
4462 in the case where the clock went backwards by a substantial amount such
4463 that the microsecond fraction of "now" was more than the microsecond
4464 fraction of "then" (but the whole seconds number was less).
4466 38. Added support for the libradius Radius client library this is found on
4467 FreeBSD (previously only the radiusclient library was supported).
4473 1. When certain lookups returned multiple values in the form name=value, the
4474 quoting of the values was not always being done properly. Specifically:
4475 (a) If the value started with a double quote, but contained no whitespace,
4477 (b) If the value contained whitespace other than a space character (i.e.
4478 tabs or newlines or carriage returns) it was not quoted.
4479 This fix has been applied to the mysql and pgsql lookups by writing a
4480 separate quoting function and calling it from the lookup code. The fix
4481 should probably also be applied to nisplus, ibase and oracle lookups, but
4482 since I cannot test any of those, I have not disturbed their existing code.
4484 2. A hit in the callout cache for a specific address caused a log line with no
4485 reason for rejecting RCPT. Now it says "Previous (cached) callout
4486 verification failure".
4488 3. There was an off-by-one bug in the queryprogram router. An over-long
4489 return line was truncated at 256 instead of 255 characters, thereby
4490 overflowing its buffer with the terminating zero. As well as fixing this, I
4491 have increased the buffer size to 1024 (and made a note to document this).
4493 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
4494 when Exim is waiting for an SMTP response from a remote server, Exim
4495 restarts its select() call on the socket, thereby resetting its timeout.
4496 This is not a problem when such interrupts are rare. Somebody set up a cron
4497 job to run exiwhat every 2 minutes, which is less than the normal select()
4498 timeout (5 or 10 minutes). This meant that the select() timeout never
4499 kicked in because it was always reset. I have fixed this by comparing the
4500 time when an interrupt arrives with the time at the start of the first call
4501 to select(). If more time than the timeout has elapsed, the interrupt is
4502 treated as a timeout.
4504 5. Some internal re-factoring in preparation for the addition of Sieve
4505 extensions (by MH). In particular, the "personal" test is moved to a
4506 separate function, and given an option for scanning Cc: and Bcc: (which is
4507 not set for Exim filters).
4509 6. When Exim created an email address using the login of the caller as the
4510 local part (e.g. when creating a From: or Sender: header line), it was not
4511 quoting the local part when it contained special characters such as @.
4513 7. Installed new OpenBSD configuration files.
4515 8. Reworded some messages for syntax errors in "and" and "or" conditions to
4516 try to make them clearer.
4518 9. Callout options, other than the timeout value, were being ignored when
4519 verifying sender addresses in header lines. For example, when using
4521 verify = header_sender/callout=no_cache
4523 the cache was (incorrectly) being used.
4525 10. Added a missing instance of ${EXE} to the exim_install script; this affects
4526 only the Cygwin environment.
4528 11. When return_path_on_delivery was set as a log selector, if different remote
4529 addresses in the same message used different return paths and parallel
4530 remote delivery occurred, the wrong values would sometimes be logged.
4531 (Whenever a remote delivery process finished, the return path value from
4532 the most recently started remote delivery process was logged.)
4534 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
4535 lines when AUTH and/or TLS are in use. This is the "received protocol"
4536 field. Exim used to use "asmtp" for authenticated SMTP, without any
4537 indication (in the protocol name) for TLS use. Now it follows the RFC and
4538 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
4539 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
4540 names appear in log lines as well as in Received: header lines.
4542 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
4543 extensions, and comparison tests, and to fix some bugs.
4545 14. Changes to the "personal" filter test:
4547 (1) The test was buggy in that it was just doing the equivalent of
4548 "contains" tests on header lines. For example, if a user's address was
4549 anne@some.where, the "personal" test would incorrectly be true for
4551 To: susanne@some.where
4553 This test is now done by extracting each address from the header in turn,
4554 and checking the entire address. Other tests that are part of "personal"
4555 are now done using regular expressions (for example, to check local parts
4556 of addresses in From: header lines).
4558 (2) The list of non-personal local parts in From: addresses has been
4559 extended to include "listserv", "majordomo", "*-request", and "owner-*",
4560 taken from the Sieve specification recommendations.
4562 (3) If the message contains any header line starting with "List-" it is
4563 treated as non-personal.
4565 (4) The test for "circular" in the Subject: header line has been removed
4566 because it now seems ill-conceived.
4568 15. Minor typos in src/EDITME comments corrected.
4570 16. Installed latest exipick from John Jetmore.
4572 17. If headers_add on a router specified a text string that was too long for
4573 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
4574 of string_sprintf() is now avoided.
4576 18. $message_body_size was not set (it was always zero) when running the DATA
4577 ACL and the local_scan() function.
4579 19. For the "mail" command in an Exim filter, no default was being set for
4580 the once_repeat time, causing a random time value to be used if "once" was
4581 specified. (If the value happened to be <= 0, no repeat happened.) The
4582 default is now 0s, meaning "never repeat". The "vacation" command was OK
4583 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
4584 (I found it when inspecting the code).
4586 20. There is now an overall timeout for performing a callout verification. It
4587 defaults to 4 times the callout timeout, which applies to individual SMTP
4588 commands during the callout. The overall timeout applies when there is more
4589 than one host that can be tried. The timeout is checked before trying the
4590 next host. This prevents very long delays if there are a large number of
4591 hosts and all are timing out (e.g. when the network connections are timing
4592 out). The value of the overall timeout can be changed by specifying an
4593 additional sub-option for "callout", called "maxwait". For example:
4595 verify = sender/callout=5s,maxwait=20s
4597 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
4598 to the end before writing, but this should make it even safer).
4600 22. Exim was forgetting that it had advertised PIPELINING for the second and
4601 subsequent messages on an SMTP connection. It was also not resetting its
4602 memory on STARTTLS and an internal HELO.
4604 23. When Exim logs an SMTP synchronization error within a session, it now
4605 records whether PIPELINING has been advertised or not.
4607 24. Added 3 instances of "(long int)" casts to time_t variables that were being
4608 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
4609 rather than long int.
4611 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
4613 26. Added the never_mail option to autoreply.
4619 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
4620 crash if the getsockname() call failed; this can happen if a connection is
4621 closed very soon after it is established. The problem was simply in the
4622 order in which certain operations were done, causing Exim to try to write
4623 to the SMTP stream before it had set up the file descriptor. The bug has
4624 been fixed by making things happen in the correct order.
4630 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
4631 before the connection was closed, thus losing the rejection response.
4633 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
4634 some early Solaris releases, but causes trouble in current releases where
4635 socklen_t is defined.
4637 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
4640 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
4643 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
4644 that releasing the top of it at the end releases what was used for sub-
4645 expansions (unless the block got too big). However, discard this block if
4646 the first thing is a variable or header, so that we can use its block when
4647 it is dynamic (useful for very large $message_headers, for example).
4649 6. Lookups now cache *every* query, not just the most recent. A new, separate
4650 store pool is used for this. It can be recovered when all lookup caches are
4651 flushed. Lookups now release memory at the end of their result strings.
4652 This has involved some general refactoring of the lookup sources.
4654 7. Some code has been added to the store_xxx() functions to reduce the amount
4655 of flapping under certain conditions.
4657 8. log_incoming_interface used to affect only the <= reception log lines. Now
4658 it causes the local interface and port to be added to several more SMTP log
4659 lines, for example "SMTP connection from", and rejection lines.
4661 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
4663 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
4665 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
4666 could be overwritten at the end of the current message (or the start of a
4667 new message if it was set in a HELO ACL). The value is now preserved for
4668 the duration of the SMTP connection.
4670 12. If a transport had a headers_rewrite setting, and a matching header line
4671 contained an unqualified address, that address was qualified, even if it
4672 did not match any rewriting rules. The underlying bug was that the values
4673 of the flags that permit the existence of unqualified sender and recipient
4674 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
4675 non-local messages, and by -bnq for local messages) were not being
4676 preserved with the message after it was received.
4678 13. When Exim was logging an SMTP synchronization error, it could sometimes log
4679 "next input=" as part of the text comprising the host identity instead of
4680 the correct text. The code was using the same buffer for two different
4681 strings. However, depending on which order the printing function evaluated
4682 its arguments, the bug did not always show up. Under Linux, for example, my
4683 test suite worked just fine.
4685 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
4686 doesn't work with some older versions of Perl. It has been changed to "my",
4687 which in any case is probably the better facility to use.
4689 15. A really picky compiler found some instances of statements for creating
4690 error messages that either had too many or two few arguments for the format
4693 16. The size of the buffer for calls to the DNS resolver has been increased
4694 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
4695 for addresses that have a lot of PTR records. This alleviates a problem; it
4696 does not fully solve it.
4698 17. A dnsdb lookup for PTR records that receives more data than will fit in the
4699 buffer now truncates the list and logs the incident, which is the same
4700 action as happens when Exim is looking up a host name and its aliases.
4701 Previously in this situation something unpredictable would happen;
4702 sometimes it was "internal error: store_reset failed".
4704 18. If a server dropped the connection unexpectedly when an Exim client was
4705 using GnuTLS and trying to read a response, the client delivery process
4706 crashed while trying to generate an error log message.
4708 19. If a "warn" verb in an ACL added multiple headers to a message in a single
4709 string, for example:
4711 warn message = H1: something\nH2: something
4713 the text was added as a single header line from Exim's point of view
4714 though it ended up OK in the delivered message. However, searching for the
4715 second and subsequent header lines using $h_h2: did not work. This has been
4716 fixed. Similarly, if a system filter added multiple headers in this way,
4717 the routers could not see them.
4719 20. Expanded the error message when iplsearch is called with an invalid key to
4720 suggest using net-iplsearch in a host list.
4722 21. When running tests using -bh, any delays imposed by "delay" modifiers in
4723 ACLs are no longer actually imposed (and a message to that effect is
4726 22. If a "gecos" field in a passwd entry contained escaped characters, in
4727 particular, if it contained a \" sequence, Exim got it wrong when building
4728 a From: or a Sender: header from that name. A second bug also caused
4729 incorrect handling when an unquoted " was present following a character
4730 that needed quoting.
4732 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
4733 was not being matched caselessly.
4735 24. Arranged for all hyphens in the exim.8 source to be escaped with
4738 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
4739 properly. Recipient callout cache records were still being keyed to include
4740 the sender, even when use_sender was set false. This led to far more
4741 callouts that were necessary. The sender is no longer included in the key
4742 when use_sender is false.
4744 26. Added "control = submission" modifier to ACLs.
4746 27. Added the ${base62d: operator to decode base 62 numbers.
4748 28. dnsdb lookups can now access SRV records.
4750 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
4751 the configuration file.
4753 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
4754 (-v) mode. This makes the output for a verbose queue run more intelligible.
4756 31. Added a use_postmaster feature to recipient callouts.
4758 32. Added the $body_zerocount variable, containing the number of binary zero
4759 bytes in the message body.
4761 33. The time of last modification of the "new" subdirectory is now used as the
4762 "mailbox time last read" when there is a quota error for a maildir
4765 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
4767 35. Added +ignore_unknown as a special item in host lists.
4769 36. Code for decoding IPv6 addresses in host lists is now included, even if
4770 IPv6 support is not being compiled. This fixes a bug in which an IPv6
4771 address was recognized as an IP address, but was then not correctly decoded
4772 into binary, causing unexpected and incorrect effects when compared with
4779 1. Very minor rewording of debugging text in manualroute to say "list of
4780 hosts" instead of "hostlist".
4782 2. If verify=header_syntax was set, and a header line with an unqualified
4783 address (no domain) and a large number of spaces between the end of the
4784 name and the colon was received, the reception process suffered a buffer
4785 overflow, and (when I tested it) crashed. This was caused by some obsolete
4786 code that should have been removed. The fix is to remove it!
4788 3. When running in the test harness, delay a bit after writing a bounce
4789 message to get a bit more predictability in the log output.
4791 4. Added a call to search_tidyup() just before forking a reception process. In
4792 theory, someone could use a lookup in the expansion of smtp_accept_max_
4793 per_host which, without the tidyup, could leave open a database connection.
4795 5. Added the variables $recipient_data and $sender_data which get set from a
4796 lookup success in an ACL "recipients" or "senders" condition, or a router
4797 "senders" option, similar to $domain_data and $local_part_data.
4799 6. Moved the writing of debug_print from before to after the "senders" test
4802 7. Change 4.31/66 (moving the time when the Received: is generated) caused
4803 problems for message scanning, either using a data ACL, or using
4804 local_scan() because the Received: header was not generated till after they
4805 were called (in order to set the time as the time of reception completion).
4806 I have revised the way this works. The header is now generated after the
4807 body is received, but before the ACL or local_scan() are called. After they
4808 are run, the timestamp in the header is updated.
4814 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
4815 before starting a queue runner without re-exec. This happened only when
4816 deliver_drop_privilege was set or when the Exim user was set to root. The
4817 effect of the bug was that timeouts during subsequent deliveries caused
4818 crashes instead of being properly handled. The handler is now left at its
4819 default (and expected) setting.
4821 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
4822 message, again when deliver_drop_privilege is set or Exim is run as root.
4823 The bug described in (1) was not present in this case, but the tidying up
4824 of the other signals was missing. I have made the two cases consistent.
4826 3. The ignore_target_hosts setting on a manualroute router was being ignored
4827 for hosts that were looked up using the /MX notation.
4829 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
4832 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
4833 operated on the sender address. After changing the $sender_address to <>
4834 for the sender address verify, Exim was re-instated it as the original
4835 (before rewriting) address, but remembering that it had rewritten it, so it
4836 wasn't rewriting it again. This bug also had the effect of breaking the
4837 sender address verification caching when the sender address was rewritten.
4839 6. The ignore_target_hosts option was being ignored by the ipliteral router.
4840 This has been changed so that if the ip literal address matches
4841 ignore_target_hosts, the router declines.
4843 7. Added expansion conditions match_domain, match_address, and match_local_
4844 part (NOT match_host).
4846 8. The placeholder for the Received: header didn't have a length field set.
4848 9. Added code to Exim itself and to exim_lock to test for a specific race
4849 condition that could lead to file corruption when using MBX delivery. The
4850 issue is with the lockfile that is created in /tmp. If this file is removed
4851 after a process has opened it but before that process has acquired a lock,
4852 there is the potential for a second process to recreate the file and also
4853 acquire a lock. This could lead to two Exim processes writing to the file
4854 at the same time. The added code performs the same test as UW imapd; it
4855 checks after acquiring the lock that its file descriptor still refers to
4856 the same named file.
4858 10. The buffer for building added header lines was of fixed size, 8192 bytes.
4859 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
4862 11. Added the smtp_active_hostname option. If used, this will typically be made
4863 to depend on the incoming interface address. Because $interface_address is
4864 not set up until the daemon has forked a reception process, error responses
4865 that can happen earlier (such as "too many connections") no longer contain
4868 12. If an expansion in a condition on a "warn" statement fails because a lookup
4869 defers, the "warn" statement is abandoned, and the next ACL statement is
4870 processed. Previously this caused the whole ACL to be aborted.
4872 13. Added the iplsearch lookup type.
4874 14. Added ident_timeout as a log selector.
4876 15. Added tls_certificate_verified as a log selector.
4878 16. Added a global option tls_require_ciphers (compare the smtp transport
4879 option of the same name). This controls incoming TLS connections.
4881 17. I finally figured out how to make tls_require_ciphers do a similar thing
4882 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
4883 before starting the TLS session.
4885 18. Tabs are now shown as \t in -bP output.
4887 19. If the log selector return_path_on_delivery was set, Exim crashed when
4888 bouncing a message because it had too many Received: header lines.
4890 20. If two routers both had headers_remove settings, and the first one included
4891 a superfluous trailing colon, the final name in the first list and the
4892 first name in the second list were incorrectly joined into one item (with a
4893 colon in the middle).
4899 1. Added -C and -D options to the exinext utility, mainly to make it easier
4900 to include in the automated testing, but these could be helpful when
4901 multiple configurations are in use.
4903 2. The exinext utility was not formatting the output nicely when there was
4904 an alternate port involved in the retry record key, nor when there was a
4905 message id as well (for retries that were specific to a specific message
4906 and a specific host). It was also confused by IPv6 addresses, because of
4907 the additional colons they contain. I have fixed the IPv4 problem, and
4908 patched it up to do a reasonable job for IPv6.
4910 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
4911 delivery, the log line now contains "pipelined" if PIPELINING was used.
4913 4. An SMTP transport process used to panic and die if the bind() call to set
4914 an explicit outgoing interface failed. This has been changed; it is now
4915 treated in the same way as a connect() failure.
4917 5. A reference to $sender_host_name in the part of a conditional expansion
4918 that was being skipped was still causing a DNS lookup. This no longer
4921 6. The def: expansion condition was not recognizing references to header lines
4922 that used bh_ and bheader_.
4924 7. Added the _cache feature to named lists.
4926 8. The code for checking quota_filecount in the appendfile transport was
4927 allowing one more file than it should have been.
4929 9. For compatibility with Sendmail, the command line option
4937 and sets the incoming protocol and host name (for trusted callers). The
4938 host name and its colon can be omitted when only the protocol is to be set.
4939 Note the Exim already has two private options, -pd and -ps, that refer to
4940 embedded Perl. It is therefore impossible to set a protocol value of "d" or
4941 "s", but I don't think that's a major issue.
4943 10. A number of refactoring changes to the code, none of which should affect
4946 (a) The number of logging options was getting close to filling up the
4947 32-bit word that was used as a bit map. I have split them into two classes:
4948 those that are passed in the argument to log_write(), and those that are
4949 only ever tested independently outside of that function. These are now in
4950 separate 32-bit words, so there is plenty of room for expansion again.
4951 There is no change in the user interface or the logging behaviour.
4953 (b) When building, for example, log lines, the code previously used a
4954 macro that called string_cat() twice, in order to add two strings. This is
4955 not really sufficiently general. Furthermore, there was one instance where
4956 it was actually wrong because one of the argument was used twice, and in
4957 one call a function was used. (As it happened, calling the function twice
4958 did not affect the overall behaviour.) The macro has been replaced by a
4959 function that can join an arbitrary number of extra strings onto a growing
4962 (c) The code for expansion conditions now uses a table and a binary chop
4963 instead of a serial search (which was left over from when there were very
4964 few conditions). Also, it now recognizes conditions like "pam" even when
4965 the relevant support is not compiled in: a suitably worded error message is
4966 given if an attempt is made to use such a condition.
4968 11. Added ${time_interval:xxxxx}.
4970 12. A bug was causing one of the ddress fields not to be passed back correctly
4971 from remote delivery subprocesses. The field in question was not being
4972 subsequently used, so this caused to problems in practice.
4974 13. Added new log selectors queue_time and deliver_time.
4976 14. Might have fixed a bug in maildirsizefile handling that threw up
4977 "unexpected character" debug warnings, and recalculated the data
4978 unnecessarily. In any case, I expanded the warning message to give more
4981 15. Added the message "Restricted characters in address" to the statements in
4982 the default ACL that block characters like @ and % in local parts.
4984 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
4985 Three changes have been made:
4987 (a) There was a serious bug; a negative response to MAIL caused the whole
4988 recipient domain to be cached as invalid, thereby blocking all messages
4989 to all local parts at the same domain, from all senders. This bug has
4990 been fixed. The domain is no longer cached after a negative response to
4991 MAIL if the sender used is not empty.
4993 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
4996 (c) A new callout option, "use_sender" has been added for people who want
4997 the modified behaviour.
5003 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
5006 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
5007 indeed breaks things for older releases.
5009 3. Added additional logging to the case where there is a problem reading data
5010 from a filter that is running in a subprocess using a pipe, in order to
5011 try to track down a specific problem.
5013 4. Testing facility fudge: when running in the test harness and attempting
5014 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
5015 getting "No route to host". Convert this to a timeout.
5017 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
5020 6. Some OS don't have socklen_t but use size_t instead. This affects the
5021 fifth argument of getsockopt() amongst other things. This is now
5022 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
5023 can be set for individual OS. I have set it for SunOS5, OSF1, and
5024 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
5025 some earlier ones do not.
5027 7. Change 4.30/15 was not doing the test caselessly.
5029 8. The standard form for an IPv6 address literal was being rejected by address
5030 parsing in, for example, MAIL and RCPT commands. An example of this kind of
5031 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5032 this, as well as the form without the "IPv6" on the front (but only when
5033 address literals are enabled, of course).
5035 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
5037 10. Exim crashed if a message with an empty sender address specified by -f
5038 encountered a router with an errors_to setting. This could be provoked only
5039 by a command such as
5043 where an empty string was supplied; "<>" did not hit this bug.
5045 11. Installed PCRE release 4.5.
5047 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
5048 remained set. It is now erased.
5050 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
5051 times from message ids (which are base 36 rather than the normal 62).
5053 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
5054 were being counted as actual protocol errors, and logged if the log
5055 selector +smtp_protocol_error was set. One cannot be perfect in this test,
5056 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
5057 and DATA following a set of rejected RCPTs do not count as protocol errors.
5058 In other words, Exim assumes they were pipelined, though this may not
5059 actually be the case. Of course, in all cases the client gets an
5060 appropriate error code.
5062 15. If a lookup fails in an ACL condition, a message about the failure may
5063 be available; it is used if testing the ACL cannot continue, because most
5064 such messages specify what the cause of the deferral is. However, some
5065 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
5066 that caused an old message to be retained and used if a later statement
5067 caused a defer, replacing the real cause of the deferral.
5069 16. If an IP address had so many PTR records that the DNS lookup buffer
5070 was not large enough to hold them, Exim could crash while trying to process
5071 the truncated data. It now detects and logs this case.
5073 17. Further to 4.21/58, another change has been made: if (and only if) the
5074 first line of a message (the first header line) ends with CRLF, a bare LF
5075 in a subsequent header line has a space inserted after it, so as not to
5076 terminate the header.
5078 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
5079 unnecessarily, used atoi() instead of strtol(), and didn't check the
5080 termination when getting file sizes from file names by regex.
5082 19. Completely re-implemented the support for maildirsize files, in the light
5083 of a number of problems with the previous contributed implementation
5084 (4.30/29). In particular:
5086 . If the quota is zero, the maildirsize file is maintained, but no quota is
5089 . If the maildir directory does not exist, it is created before any attempt
5090 to write a maildirsize file.
5092 . The quota value in the file is just a cache; if the quota is changed in
5093 the transport, the new value overrides.
5095 . A regular expression is available for excluding directories from the
5098 20. The autoreply transport checks the characters in options that define the
5099 message's headers; it allows continued headers, but it was checking with
5100 isspace() after an embedded newline instead of explicitly looking for a
5103 21. If all the "regular" hosts to which an address was routed had passed their
5104 expiry times, and had not reached their retry times, the address was
5105 bounced, even if fallback hosts were defined. Now Exim should go on to try
5108 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
5109 equivalent code in the SMTP transport. Some hosts send humungous responses
5110 to HELO/EHLO, more than 1024 it seems.
5112 23. Refactoring: code in filter.c used (void *) for "any old type" but this
5113 gives compiler warnings in some environments. I've now done it "properly",
5116 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
5117 (because of problems with the built-in one) was declared to return uschar *
5118 instead of char *, causing compiler failure.
5120 25. Fixed a file descriptor leak when processing alias/forward files.
5122 26. Fixed a minor format string issue in dbfn.c.
5124 27. Typo in exim.c: ("dmbnz" for "dbmnz").
5126 28. If a filter file refered to $h_xxx or $message_headers, and the headers
5127 contained RFC 2047 "words", Exim's memory could, under certain conditions,
5130 29. When a sender address is verified, it is cached, to save repeating the test
5131 when there is more than one recipient in a message. However, when the
5132 verification involves a callout, it is possible for different callout
5133 options to be set for different recipients. It is too complicated to keep
5134 track of this in the cache, so now Exim always runs a verification when a
5135 callout is required, relying on the callout cache for the optimization.
5136 The overhead is duplication of the address routing, but this should not be
5139 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
5140 to be verified with callout=postmaster, and the main callout worked but the
5141 postmaster check failed, the verification correctly failed. However, if a
5142 subsequent RCPT command asked for sender verification *without* the
5143 postmaster check, incorrect caching caused this verification also to fail,
5146 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
5147 it was not caching the DNS options (qualify_single, search_parents) that
5148 were used when the lookup failed. A subsequent lookup with different
5149 options therefore always gave the same answer, though there were cases
5150 where it should not have. (Example: a "domains = !$mx_any" option on a
5151 dnslookup router: the "domains" option is always processed without any
5152 widening, but the router might have qualify_single set.) Now Exim uses the
5153 cached value only when the same options are set.
5155 32. Added John Jetmore's "exipick" utility to the distribution.
5157 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
5158 than a timeout (e.g. a certificate is required, and is not provided), an
5159 Exim server now closes the connection immediately. Previously it waited for
5160 the client to close - but if the client is SSL, it seems that they each
5161 wait for each other, leading to a delay before one of them times out.
5163 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
5164 maintained 0.8.x compatibility because I don't think many are using it, and
5165 it is clearly obsolete.
5167 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
5170 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
5171 client certificate was expired. A simple patch fixes this, though I don't
5172 understand the full logic of why the verify callback is called multiple
5175 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
5176 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
5177 which causes problems with some clients (such as the Certicom SSL Plus
5178 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
5179 disables the coutermeasure allowing Eudora to connect."
5181 38. Exim was not checking that a write() to a log file succeeded. This could
5182 lead to Bad Things if a log got too big, in particular if it hit a file
5183 size limit. Exim now panics and dies if it cannot write to a log file, just
5184 as it does if it cannot open a log file.
5186 39. Modified OS/Makefile-Linux so that it now contains
5188 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
5190 The two -D definitions ensure that Exim is compiled with large file
5191 support, which makes it possible to handle log files that are bigger than
5194 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
5195 instance) a domain was checked against a named list that involved a lookup,
5196 causing $domain_data to be set, then another domain was checked against the
5197 same list, then the first domain was re-checked, the value of $domain_data
5198 after the final check could be wrong. In particular, if the second check
5199 failed, it could be set empty. This bug probably also applied to
5202 41. The strip_trailing_dot option was not being applied to the address given
5203 with the -f command-line option.
5205 42. The code for reading a message's header from the spool was incrementing
5206 $received_count, but never initializing it. This meant that the value was
5207 incorrect (doubled) while delivering a message in the same process in which
5208 it was received. In the most common configuration of Exim, this never
5209 happens - a fresh exec is done - but it can happen when
5210 deliver_drop_privilege is set.
5212 43. When Exim logs an SMTP synchronization error - client data sent too soon -
5213 it now includes up to 150 characters of the unexpected data in the log
5216 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
5217 and building data strings. The size of both of these buffers was 10 000
5218 bytes - far larger than anybody would *ever* want, thought I. Needless to
5219 say, somebody hit the limit. I have increased the maximum line length to
5220 20 000 and the maximum data length of concatenated lines to 100 000. I have
5221 also fixed two bugs, because there was no checking on these buffers. Tsk,
5222 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
5223 buffer is too small.
5225 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
5226 lsearch lookups. Now it does.
5228 46. When parsing a route_list item in a manualroute router, a fixed-length
5229 buffer was used for the list of hosts. I made this 1024 bytes long,
5230 thinking that nobody would ever have a list of hosts that long. Wrong.
5231 Somebody had a whole pile of complicated expansion conditions, and the
5232 string was silently truncated, leading to an expansion error. It turns out
5233 that it is easier to change to an unlimited length (owing to other changes
5234 that have happened since this code was originally written) than to build
5235 structure for giving a limitation error. The length of the item that
5236 expands into the list of hosts is now unlimited.
5238 47. The lsearch lookup could not handle data where the length of text line was
5239 more than 4095 characters. Such lines were truncated, leading to shortened
5240 data being returned. It should now handle lines of any length.
5242 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
5243 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
5246 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
5247 replacing strings such as BIN_DIRECTORY in a source file: the replacement
5248 no longer happens in comment lines. A list of replacements is now placed
5249 at the head of all of the source files, except those whose only change is
5250 to replace PERL_COMMAND in the very first #! line.
5252 50. Replaced the slow insertion sort in queue.c, for sorting the list of
5253 messages on the queue, with a bottom-up merge sort, using code contributed
5254 by Michael Haardt. This should make operations like -bp somewhat faster on
5255 large queues. It won't affect queue runners, except when queue_run_in_order
5258 51. Installed eximstats 1.31 in the distribution.
5260 52. Added support for SRV lookups to the dnslookup router.
5262 53. If an ACL referred to $message_body or $message_body_end, the value was not
5263 reset for any messages that followed in the same SMTP session.
5265 54. The store-handling optimization for building very long strings was not
5266 differentiating between the different store pools. I don't think this
5267 actually made any difference in practice, but I've tidied it.
5269 55. While running the routers to verify a sender address, $sender_address
5270 was still set to the sender address. This is wrong, because when routing to
5271 send a bounce to the sender, it would be empty. Therefore, I have changed
5272 it so that, while verifying a sender address, $sender_address is set to <>.
5273 (There is no change to what happens when verifying a recipient address.)
5275 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
5276 target A or AAAA records (if not already returned) without resetting the
5277 qualify_single or search_parents options of the DNS resolver. These are
5278 inappropriate in this case because the targets of MX and SRV records must
5279 be FQDNs. A broken DNS record could cause trouble if it happened to have a
5280 target that, when qualified, matched something in the local domain. These
5281 two options are now turned off when doing these lookups.
5283 57. It seems that at least some releases of Reiserfs (which does not have the
5284 concept of a fixed number of inodes) returns zero and not -1 for the
5285 number of available inodes. This interacted badly with check_spool_inodes,
5286 which assumed that -1 was the "no such thing" setting. What I have done is
5287 to check that the total number of inodes is greater than zero before doing
5288 the test of how many are available.
5290 58. When a "warn" ACL statement has a log_message modifier, the message is
5291 remembered, and not repeated. This is to avoid a lot of repetition when a
5292 message has many recipients that cause the same warning to be written.
5293 Howewer, Exim was preserving the list of already written lines for an
5294 entire SMTP session, which doesn't seem right. The memory is now reset if a
5295 new message is started.
5297 59. The "rewrite" debugging flag was not showing the result of rewriting in the
5298 debugging output unless log_rewrite was also set.
5300 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
5301 of (int)(handle) when we know that handle contains (void *)(-1).
5303 61. The Exim daemon panic-logs an error return when it closes the incoming
5304 connection. However "connection reset by peer" seems to be common, and
5305 isn't really an error worthy of noting specially, so that particular error
5308 62. When Exim is trying to find all the local interfaces, it used to panic and
5309 die if the ioctl to get the interface flags failed. However, it seems that
5310 on at least one OS (Solaris 9) it is possible to have an interface that is
5311 included in the list of interfaces, but for which you get a failure error
5312 for this call. This happens when the interface is not "plumbed" into a
5313 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
5314 failure of the "get flags" call assumes that the interface is down.
5316 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
5317 makes life easier for people who are doing arithmetic on fields extracted
5318 from dates, where you often get leading zeros that should not be
5319 interpreted as octal.
5321 64. Added qualify_domain to the redirect router, to override the global
5324 65. If a pathologically long header line contained very many addresses (the
5325 report of this problem mentioned 10 000) and each of them was rewritten,
5326 Exim could use up a very large amount of memory. (It kept on making new
5327 copies of the header line as it rewrote, and never released the old ones.)
5328 At the expense of a bit more processing, the header rewriting function has
5329 been changed so that it no longer eats memory in this way.
5331 66. The generation of the Received: header has been moved from the time that a
5332 message starts to be received, to the time that it finishes. The timestamp
5333 in the Received: header should now be very close to that of the <= log
5334 line. There are two side-effects of this change:
5336 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
5337 logged header lines no longer include the local Received: line, because
5338 it has not yet been created. The same applies to a copy of the message
5339 that is returned to a non-SMTP sender when a message is rejected.
5341 (b) When a filter file is tested using -bf, no additional Received: header
5342 is added to the test message. After some thought, I decided that this
5345 This change does not affect the value of $received_for. It is still set
5346 after address rewriting, but before local_scan() is called.
5348 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
5350 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
5351 gave an unhelpful panic error message, and a defer error. I have managed to
5352 change this behaviour so that it now rejects any supplied certificate,
5353 which seems right, as the list of acceptable certificates is empty.
5355 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
5356 gave an unhelpful defer error. I have not managed to make this reject any
5357 supplied certificates, but the error message it gives is "no certificate
5358 supplied", which is not helpful.
5360 70. exigrep's output now also includes lines that are not associated with any
5361 message, but which match the given pattern. Implemented by a patch from
5362 Martin Sluka, which also tidied up the Perl a bit.
5364 71. Recipient callout verification, like sender verification, was using <> in
5365 the MAIL FROM command. This isn't really the right thing, since the actual
5366 sender may affect whether the remote host accepts the recipient or not. I
5367 have changed it to use the actual sender in the callout; this means that
5368 the cache record is now keyed on a recipient/sender pair, not just the
5369 recipient address. There doesn't seem to be a real danger of callout loops,
5370 since a callout by the remote host to check the sender would use <>.
5371 [SEE ABOVE: changed after hitting problems.]
5373 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
5374 temporary errors. However, in the case of such a code being given after
5375 the end of a data transmission (i.e. after ".") Exim was failing to write
5376 a retry record for the message. (Yes, there was some broken host that was
5377 actually sending 8xx at this point.)
5379 73. An unknown lookup type in a host list could cause Exim to panic-die when
5380 the list was checked. (An example that provoked this was putting <; in the
5381 middle of a list instead of at the start.) If this happened during a DATA
5382 ACL check, a -D file could be left lying around. This kind of configuration
5383 error no longer causes Exim to die; instead it causes a defer errror. The
5384 incident is still logged to the main and panic logs.
5386 74. Buglet left over from Exim 3 conversion. The message "too many messages
5387 in one connection" was written to the rejectlog but not the mainlog, except
5388 when address rewriting (yes!) was being logged.
5390 75. Added write_rejectlog option.
5392 76. When a system filter was run not as root (that is, when system_filter_user
5393 was set), the values of the $n variables were not being returned to the
5394 main process; thus, they were not subsequently available in the $sn
5397 77. Added +return_path_on_delivery log selector.
5399 78. A connection timeout was being treated differently from recipients deferred
5400 when testing hosts_max_try with a message that was older than the host's
5401 retry timeout. (The host should not be counted, thus allowing all hosts to
5402 be tried at least once before bouncing.) This may have been the cause of an
5403 occasionally reported bug whereby a message would remain on the queue
5404 longer than the retry timeout, but would be bounced if a delivery was
5405 forced. I say "may" because I never totally pinned down the problem;
5406 setting up timeout/retry tests is difficult. See also the next item.
5408 79. The ultimate address timeout was not being applied to errors that involved
5409 a combination of host plus message (for example, a timeout on a MAIL
5410 command). When an address resolved to a number of possible hosts, and they
5411 were not all tried for each delivery (e.g. because of hosts_max_try), a
5412 message could remain on the queue longer than the retry timeout.
5414 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
5417 81. Fixed an obscure SMTP outgoing bug which required at least the following
5418 conditions: (a) there was another message waiting for the same server;
5419 (b) the server returned 5xx to all RCPT commands in the first message so
5420 that the message was not completed; (c) the server dropped the connection
5421 or gave a negative response to the RSET that Exim sends to abort the
5422 transaction. The observed case was a dropped connection after DATA that had
5423 been sent in pipelining mode. That is, the server had advertised PIPELINING
5424 but was not implementing it correctly. The effect of the bug was incorrect
5425 behaviour, such as trying another host, and this could lead to a crash.
5431 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
5432 and daemon.c were passed as pointers to ints; they should have been
5433 pointers to socklen_t variables (which are typically unsigned ints).
5435 2. Some signed/unsigned type warnings in the os.c file for Linux have been
5438 3. Fixed a really odd bug that affected only the testing scheme; patching a
5439 certain fixed string in the binary changed the value of another string that
5440 happened to be identical to the end of the original first string.
5442 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
5443 name", it returns that address as the IP address. On some operating
5444 systems (e.g. Solaris), it also passes back the IP address string as the
5445 "host name". However, on others (e.g. Linux), it passes back an empty
5446 string. Exim wasn't checking for this, and was changing the host name to an
5447 empty string, assuming it had been canonicized.
5449 5. Although rare, it is permitted to have more than one PTR record for a given
5450 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
5451 all the names associated with an address, because they do in Solaris.
5452 However, it seems that they do not in Linux for data that comes from the
5453 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
5454 I found this out when I moved to a new Linux workstation and tried to run
5455 the Exim test suite.
5457 To get round this problem I have changed the code so that it now does its
5458 own call to the DNS to look up PTR records when searching for a host name.
5459 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
5460 addresses that are only in /etc/hosts are still found.
5462 This behaviour is, however, controlled by an option called host_lookup_
5463 order, which defaults to "bydns:byaddr". If people want to use the other
5464 order, or indeed, just use one or the other means of lookup, they can
5465 specify it in this variable.
5467 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
5468 some operating systems, this comes back from gethostbyaddr() as an empty
5469 string, and this is what Exim used to test for. However, it seems that in
5470 other systems, "." is yielded. Exim now tests for this case too.
5472 7. The values of check_spool_space and check_log_space are now held internally
5473 as a number of kilobytes instead of an absolute number of bytes. If a
5474 numbers is specified without 'K' or 'M', it is rounded up to the nearest
5475 kilobyte. This means that much larger values can be stored.
5477 8. Exim monitor: an attempt to get the action menu when not actually pointing
5478 at a message produces an empty menu entitled "No message selected". This
5479 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
5480 no entries in it ("Shell widget menu has zero width and/or height"). So I
5481 have added a single, blank menu entry in this case.
5483 9. Added ${quote_local_part.
5485 10. MIME decoding is now applied to the contents of Subject: header lines when
5488 11. Now that a reference to $sender_host_address automatically causes a reverse
5489 lookup to occur if necessary (4.13/18), there is no need to arrange for a
5490 host lookup before query-style lookups in lists that might use this
5491 variable. This has therefore been abolished, and the "net-" prefix is no
5492 longer necessary for query-style lookups.
5494 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
5495 have been a typo for LFLAGS, so it has been changed.
5497 13. The install script calls Exim with "-C /dev/null" in order to find the
5498 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
5499 to be output. Howeve, since Exim outputs its version number before the
5500 error, it didn't break the script. It just looked ugly. I fixed this by
5501 always allowing "-C /dev/null" if the caller is root.
5503 14. Ignore overlarge ACL variable number when reading spool file - insurance
5504 against a later release with more variables having written the file.
5506 15. The standard form for an IPv6 address literal was being rejected by EHLO.
5507 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5508 this, as well as the form without the "IPv6" on the front.
5510 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
5511 OS/Makefile-Darwin file.
5513 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
5514 only to LDAP libraries that do not have LDAP_OPT_DEREF.
5516 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
5517 variable. However, some OS use ints for inodes. I've added cast to long int
5518 to get rid of the compiler warning.
5520 19. I had forgotten to lock out "/../" in configuration file names when
5521 ALT_CONFIG_PREFIX was set.
5523 20. Routers used for verification do not need to specify transports. However,
5524 if such a router generated a host list, and callout was configured, Exim
5525 crashed, because it could not find a port number from the (non-existent)
5526 transport. It now assumes port 25 in this circumstance.
5528 21. Added the -t option to exigrep.
5530 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
5531 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
5532 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
5533 them is set, LOOKUP_LSEARCH is forced.
5535 23. "exim -bV" now outputs a list of lookups that are included in the binary.
5537 24. Added sender and host information to the "rejected by local_scan()" log
5538 line; previously there was no indication of these.
5540 25. Added .include_if_exists.
5542 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
5543 Linux. It turns out that not all file systems support this. Apparently some
5544 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
5545 do it.) To cope with this, the error EINVAL, which means that sync-ing is
5546 not supported on the file descriptor, is now ignored when Exim is trying to
5547 sync a directory. This applies only to Linux.
5549 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
5551 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
5552 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
5555 29. Added support for maildirsize files from supplied patch (modified a bit).
5557 30. The use of :fail: followed by an empty string could lead Exim to respond to
5558 sender verification failures with (e.g.):
5560 550 Verification failed for <xxx>
5561 550 Sender verify failed
5563 where the first response line was missing the '-' that indicates it is not
5564 the final line of the response.
5566 31. The loop for finding the name of the user that called Exim had a hardwired
5567 limit of 10; it now uses the value of finduser_retries, which is used for
5568 all other user lookups.
5570 32. Added $received_count variable, available in data and not_smtp ACLs, and at
5573 33. Exim was neglecting to zero errno before one call of strtol() when
5574 expanding a string and expecting an integer value. On some systems this
5575 resulted in spurious "integer overflow" errors. Also, it was casting the
5576 result into an int without checking.
5578 34. Testing for a connection timeout using "timeout_connect" in the retry rules
5579 did not work. The code looks as if it has *never* worked, though it appears
5580 to have been documented since at least releast 1.62. I have made it work.
5582 35. The "timeout_DNS" error in retry rules, also documented since at least
5583 1.62, also never worked. As it isn't clear exactly what this means, and
5584 clearly it isn't a major issue, I have abolished the feature by treating it
5585 as "timeout", and writing a warning to the main and panic logs.
5587 36. The display of retry rules for -brt wasn't always showing the error code
5590 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
5591 timeout_connect_A, timeout_connect_MX.
5593 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
5594 to the empty sender.
5596 39. The daemon was not analysing the content of -oX till after it had closed
5597 stderr and disconnected from the controlling terminal. This meant that any
5598 syntax errors were only noted on the panic log, and the return code from
5599 the command was 0. By re-arranging the code a little, I've made the
5600 decoding happen first, so such errors now appear on stderr, and the return
5601 code is 1. However, the actual setting up of the sockets still happens in
5602 the disconnected process, so errors there are still only recorded on the
5605 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
5606 connections (as happens on some IP stacks) was logged at start up time as
5607 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
5608 it from "IPv6 and IPv4", which means that two separate sockets are being
5611 41. The debug output for gethostbyname2() or getipnodebyname() failures now
5612 says whether AF_INET or AF_INET6 was passed as an argument.
5614 42. Exiwhat output was messed up when time zones were included in log
5617 43. Exiwhat now gives more information about the daemon's listening ports,
5618 and whether -tls-on-connect was used.
5620 44. The "port" option of the smtp transport is now expanded.
5622 45. A "message" modifier in a "warn" statement in a non-message ACL was being
5623 silently ignored. Now an error message is written to the main and panic
5626 46. There's a new ACL modifier called "logwrite" which writes to a log file
5627 as soon as it is encountered.
5629 47. Added $local_user_uid and $local_user_gid at routing time.
5631 48. Exim crashed when trying to verify a sender address that was being
5634 49. Exim was recognizing only a space character after ".include". It now also
5635 recognizes a tab character.
5637 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
5638 extracting the relevant information from the specification. The man page no
5639 longer contains scrambled data for the -d option, and I've added a section
5640 at the front about calling Exim under different names.
5642 51. Added "extra_headers" argument to the "mail" command in filter files.
5644 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
5647 53. Installed eximstats 1.29.
5649 54. Added transport_filter_timeout as a generic transport option.
5651 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
5652 Cc: header lines. This was required by RFC 822, but it not required by RFC
5655 56. Exim used to add From:, Date:, and Message-Id: header lines to any
5656 incoming messages that did not have them. Now it does so only if the
5657 message originates locally, that is, if there is no associated remote host
5658 address. When Resent- header lines are present, this applies to the Resent-
5659 lines rather than the non-Resent- lines.
5661 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
5662 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
5664 58. Messages for configuration errors now include the name of the main
5665 configuration file - useful now that there may be more than one file in a
5666 list (.included file names were always shown).
5668 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
5669 for those rare installations that do not start the daemon as root or run it
5670 setuid root. I've cut out the call to initgroups() if the daemon is not
5673 60. The Exim user and group can now be bound into the binary as text strings
5674 that are looked up at the start of Exim's processing.
5676 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
5678 62. Added $mailstore_basename variable.
5680 63. Installed patch to sieve.c from Michael Haardt.
5682 64. When Exim failed to open the panic log after failing to open the main log,
5683 the original message it was trying to log was written to stderr and debug
5684 output, but if they were not available (the usual case in production), it
5685 was lost. Now it is written to syslog before the two lines that record the
5686 failures to open the logs.
5688 65. Users' Exim filters run in subprocesses under the user's uid. It is
5689 possible for a "deliver" command or an alias in a "personal" command to
5690 provoke an address rewrite. If logging of address rewriting is configured,
5691 this fails because the process is not running as root or exim. There may be
5692 a better way of dealing with this, but for the moment (because 4.30 needs
5693 to be released), I have disabled address rewrite logging when running a
5694 filter in a non-root, non-exim process.
5700 1. The buildconfig auxiliary program wasn't quoting the value set for
5701 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
5702 not defined. This bug was masked in 4.22 by the effect that was fixed in
5705 2. Some messages that were rejected after a message id was allocated were
5706 shown as "incomplete" by exigrep. It no longer does this for messages that
5707 are rejected by local_scan() or the DATA or non-SMTP ACLs.
5709 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
5710 have allow_domain_literals set, the ID did not get logged in the <= line.
5711 Domain literals are now always recognized in Message-ID: header lines.
5713 4. The first argument for a ${extract expansion item is the key name or field
5714 number. Leading and trailing spaces in this item were not being ignored,
5715 causing some misleading effects.
5717 5. When deliver_drop_privilege was set, single queue runner processes started
5718 manually (i.e. by the command "exim -q") or by the daemon (which uses the
5719 same command in the process it spins off) were not dropping privilege.
5721 6. When the daemon running as "exim" started a queue runner, it always
5722 re-executed Exim in the spun-off process. This is a waste of effort when
5723 deliver_drop_privilege is set. The new process now just calls the
5724 queue-runner function directly.
5730 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
5733 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
5734 ignored. Though the use of -oP was forcing the writing of a pid file, it
5735 was always written to the default place.
5737 3. If the message "no IP address found for host xxxx" is generated during
5738 incoming verification, it is now followed by identification of the incoming
5739 connection (so you can more easily find what provoked it).
5741 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
5743 5. Added some features to "harden" Exim a bit more against certain attacks:
5745 (a) There is now a build-time option called FIXED_NEVER_USERS that can
5746 be put in Local/Makefile. This is like the never_users runtime option,
5747 but it cannot be overridden. The default setting is "root".
5749 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
5750 prefix string with which any file named in a -C command line option
5753 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
5754 is retained for -C and -D only if the caller of Exim is root. Without
5755 it, the exim user may also use -C and -D and retain privilege.
5757 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
5758 command line option is disabled.
5760 6. Macro names set by the -D option must start with an upper case letter, just
5761 like macro names defined in the configuration file.
5763 7. Added "dereference=" facility to LDAP.
5765 8. Two instances of the typo "uknown" in the source files are fixed.
5767 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
5768 the Configure-Makefile script screwed up while processing it.
5770 10. Incorporated PCRE 4.4.
5772 11. The SMTP synchronization check was not operating right at the start of an
5773 SMTP session. For example, it could not catch a HELO sent before the client
5774 waited for the greeting. There is now a check for outstanding input at the
5775 point when the greeting is written. Because of the duplex, asynchronous
5776 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
5777 way, but not yet received, when the check is performed.
5779 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
5780 on TCP/IP sockets, because this apparently causes some broken clients to
5783 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
5784 unchanged) from the Cygwin maintainer.
5786 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
5787 was supported instead of testing for mbx. Effectively a typo.
5789 15. The spa authenticator server code was not checking that the input it
5790 received was valid base64.
5792 16. The debug output line for the "set" modifier in ACLs was not showing the
5793 name of the variable that was being set.
5795 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
5797 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
5798 lookup on its own. Something else was needed to trigger the lookup. For
5799 example, a match in host_lookup or the need for a host name in a host list.
5800 Now, if $sender_host_name is referenced and the host name has not yet been
5801 looked up, a lookup is performed. If the lookup fails, the variable remains
5802 empty, and $host_lookup_failed is set to "1".
5804 19. Added "eqi" as a case-independent comparison operator.
5806 20. The saslauthd authentication condition could segfault if neither service
5807 nor realm was specified.
5809 21. If an overflowing value such as "2048M" was set for message_size_limit, the
5810 error message that was logged was misleading, and incoming SMTP
5811 connections were dropped. The message is now more accurate, and temporary
5812 errors are given to SMTP connections.
5814 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
5815 (except RSET) with a 421 error, until QUIT is received. However, it was
5816 failing to send a response to QUIT.
5818 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
5819 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
5820 ACL is now run after the helo_try_verify_hosts code.
5822 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
5823 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
5824 case-independent, so other case variants are also recognized). Apparently
5825 some systems use these upper case variants.
5827 25. If more than two messages were waiting for the same host, and a transport
5828 filter was specified for the transport, Exim sent two messages over the
5829 same TCP/IP connection, and then failed with "socket operation on non-
5830 socket" when it tried to send the third.
5832 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
5834 27. The extern definition of crypt16() in expand.c was not being excluded when
5835 the OS had its own crypt16() function.
5837 28. Added bounce_return_body as a new option, and bounce_return_size_limit
5838 as a preferred synonym for return_size_limit, both as an option and as an
5841 29. Added LIBS=-liconv to OS/Makefile-OSF1.
5843 30. Changed the default configuration ACL to relax the local part checking rule
5844 for addresses that are not in any local domains. For these addresses,
5845 slashes and pipe symbols are allowed within local parts, but the sequence
5846 /../ is explicitly forbidden.
5848 31. SPA server authentication was not clearing the challenge buffer before
5851 32. log_message in a "warn" ACL statement was writing to the reject log as
5852 well as to the main log, which contradicts the documentation and doesn't
5853 seem right (because no rejection is happening). So I have stopped it.
5855 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
5856 However, I am unable to do any testing of this.
5858 34. Fixed an infelicity in the appendfile transport. When checking directories
5859 for a mailbox, to see if any needed to be created, it was accidentally
5860 using path names with one or more superfluous leading slashes; tracing
5861 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
5863 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
5864 contents are added to the log line that is written for every discarded
5865 recipient. (Previously a log_message setting was ignored.)
5867 36. The ${quote: operator now quotes the string if it is empty.
5869 37. The install script runs exim in order to find its version number. If for
5870 some reason other than non-existence or emptiness, which it checks, it
5871 could not run './exim', it was installing it with an empty version number,
5872 i.e. as "exim-". This error state is now caught, and the installation is
5875 38. An argument was missing from the function that creates an error message
5876 when Exim fails to connect to the socket for saslauthd authentication.
5877 This could cause Exim to crash, or give a corrupted message.
5879 39. Added isip, isip4, and isip6 to ${if conditions.
5881 40. The ACL variables $acl_xx are now saved with the message, and can be
5882 accessed later in routers, transports, and filters.
5884 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
5885 strings in the file are not string-expanded.
5887 42. If a MAIL command specified a SIZE value that was too large to fit into an
5888 int variable, the check against message_size_limit failed. Such values are
5889 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
5890 day this will have to be increased, but I don't think I want to be around
5891 when emails are that large.
5898 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
5899 iconv() is not standard in FreeBSD.
5901 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
5902 IPv6 enabled. The observed symptom was a segmentation fault on return from
5903 the function os_common_find_running_interfaces() in src/os.c.
5905 3. In the check_special_case() function in daemon.c I had used "errno" as an
5906 argument name, which causes warnings on some systems. This was basically a
5907 typo, since it was named "eno" in the comments!
5909 4. The code that waits for the clock to tick (at a resolution of some fraction
5910 of a second) so as to ensure message-id uniqueness was always waiting for
5911 at least one whole tick, when it could have waited for less. [This is
5912 almost certainly not relevant at current processor speeds, where it is
5913 unlikely to ever wait at all. But we try to future-proof.]
5915 5. The function that sleeps for a time interval that includes fractions of a
5916 second contained a race. It did not block SIGALRM between setting the
5917 timer, and suspending (a couple of lines later). If the interval was short
5918 and the sigsuspend() was delayed until after it had expired, the suspension
5919 never ended. On busy systems this could lead to processes getting stuck for
5922 6. Some uncommon configurations may cause a lookup to happen in a queue runner
5923 process, before it forks any delivery processes. The open lookup caching
5924 mechanism meant that the open file or database connection was passed into
5925 the delivery process. The problem was that delivery processes always tidy
5926 up cached lookup data. This could cause a problem for the next delivery
5927 process started by the queue runner, because the external queue runner
5928 process does not know about the closure. So the next delivery process
5929 still has data in the lookup cache. In the case of a file lookup, there was
5930 no problem because closing a file descriptor in a subprocess doesn't affect
5931 the parent. However, if the lookup was caching a connection to a database,
5932 the connection was closed, and the second delivery process was likely to
5933 see errors such as "PGSQL: query failed: server closed the connection
5934 unexpectedly". The problem has been fixed by closing all cached lookups
5935 in a queue runner before running a delivery process.
5937 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
5938 seem to have the "const" qualifier which it has on other OS. I've
5941 8. Change 4.21/2 was too strict. It is only if there are two authenticators
5942 *of the same type* (client or server) with the same public name that an
5943 error should be diagnosed.
5945 9. When Exim looked up a host name for an IP address, but failed to find the
5946 original IP address when looking up the host name (a safety check), it
5947 output the message "<ip address> does not match any IP for NULL", which was
5948 confusing, to say the least. The bug was that the host name should have
5949 appeared instead of "NULL".
5951 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
5952 user that is built into the binary, and the -C or -D options is used, root
5953 privilege is dropped before the configuration file is read. In addition,
5954 logging is switched to stderr instead of the normal log files. If the
5955 configuration then re-defines the Exim user, the unprivileged environment
5956 is probably not what is expected, so Exim logs a panic warning message (but
5959 However, if deliver_drop_privilege is set, the unprivileged state may well
5960 be exactly what is intended, so the warning has been cut out in that case,
5961 and Exim is allowed to try to write to its normal log files.
5967 1. smtp_return_error_details was not giving details for temporary sender
5968 or receiver verification errors.
5970 2. Diagnose a configuration error if two authenticators have the same public
5973 3. Exim used not to create the message log file for a message until the first
5974 delivery attempt. This could be confusing when incoming messages were held
5975 for policy or load reasons. The message log file is now created at the time
5976 the message is received, and an initial "Received" line is written to it.
5978 4. The automatically generated man page for command line options had a minor
5979 bug that caused no ill effects; however, a more serious problem was that
5980 the procedure for building the man page automatically didn't always
5981 operate. Consequently, release 4.20 contains an out-of-date version. This
5982 shouldn't happen again.
5984 5. When building Exim with embedded Perl support, the script that builds the
5985 Makefile was calling 'perl' to find its compile-time parameters, ignoring
5986 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
5988 6. The freeze_tell option was not being used for messages that were frozen on
5989 arrival, either by an ACL or by local_scan().
5991 7. Added the smtp_incomplete_transaction log selector.
5993 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
5994 was accepting AUTH without a new EHLO.
5996 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
5997 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
6000 10. Logging of TCP/IP connections (when configured) now happens in the main
6001 daemon process instead of the child process, so that the TCP/IP connection
6002 count is more accurate (but it can never be perfect).
6004 11. The use of "drop" in a nested ACL was not being handled correctly in the
6005 outer ACL. Now, if condition failure induced by the nested "drop" causes
6006 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
6007 or "require"), the connection is dropped.
6009 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
6010 that yield "discard" can now be used with an "accept" or a "discard" verb,
6011 but an error is generated for any others (because I can't see a useful way
6012 to define what should happen).
6014 13. When an ACL is read dynamically from a file (or anywhere else), the lines
6015 are now processed in the same way as lines in the Exim configuration file.
6016 In particular, continuation lines are supported.
6018 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
6020 15. Added -ti meaning -t -i.
6022 16. Check for letters, digits, hyphens, and dots in the names of dnslist
6023 domains, and warn by logging if others are found.
6025 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
6026 returned from GIFCONF when Exim is trying to find the list of interfaces on
6027 a host. The code in os.c has been modified to copy each ifreq to an aligned
6028 structure in all cases.
6030 Also, in some cases, the returned ifreq's were being copied to a 'struct
6031 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
6032 means the last couple of bytes of an IPv6 address could be chopped if the
6033 ifreq contained only a normal sockaddr (14 bytes storage).
6035 18. Named domain lists were not supported in the hosts_treat_as_local option.
6036 An entry such as +xxxx was not recognized, and was treated as a literal
6039 19. Ensure that header lines added by a DATA ACL are included in the reject log
6040 if the ACL subsequently rejects the message.
6042 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
6043 MD5 (which is deprecated).
6045 21. When testing a filter file using -bf, Exim was writing a message when it
6046 took the sender from a "From " line in the message, but it was not doing so
6047 when it took $return_path from a Return-Path: header line. It now does.
6049 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
6050 with a valid header line field name (a series of printing characters
6051 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
6053 23. Changed "disc" in the source to "disk" to conform to the documentation and
6054 the book and for uniformity.
6056 24. Ignore Sendmail's -Ooption=value command line item.
6058 25. When execve() failed while trying to run a command in a pipe transport,
6059 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
6060 could be confused with a return value of 69 from the command itself. This
6061 has been changed to 127, the value the shell returns if it is asked to run
6062 a non-existent command. The wording for the related log line suggests a
6063 non-existent command as the problem.
6065 26. If received_header_text expands to an empty string, do not add a Received:
6066 header line to the message. (Well, it adds a token one on the spool, but
6067 marks it "old" so that it doesn't get used or transmitted.)
6069 27. Installed eximstats 1.28 (addition of -nt option).
6071 28. There was no check for failure on the call to getsockname() in the daemon
6072 code. This can fail if there is a shortage of resources on the system, with
6073 ENOMEM, for example. A temporary error is now given on failure.
6075 29. Contrary to the C standard, it seems that in some environments, the
6076 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
6077 program. Exim now does this explicitly; it affects the formatting of
6078 timestamps using strftime().
6080 30. If exiqsumm was given junk data, it threw up some uninitialized variable
6081 complaints. I've now initialized all the variables, to avoid this.
6083 32. Header lines added by a system filter were not being "seen" during
6084 transport-time rewrites.
6086 33. The info_callback() function passed to OpenSSL is set up with type void
6087 (*)(SSL *, int, int), as described somewhere. However, when calling the
6088 function (actually a macro) that sets it up, the type void(*)() is
6089 expected. I've put in a cast to prevent warnings from picky compilers.
6091 34. If a DNS black list lookup found a CNAME record, but there were no A
6092 records associated with the domain it pointed at, Exim crashed.
6094 35. If a DNS black list lookup returned more than one A record, Exim ignored
6095 all but the first. It now scans all returned addresses if a particular IP
6096 value is being sought. In this situation, the contents of the
6097 $dnslist_value variable are a list of all the addresses, separated by a
6100 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
6101 to accept a host name and all its aliases if the forward lookup for any of
6102 them yielded the IP address of the incoming connection. Now it accepts only
6103 those names whose forward lookup yields the correct IP address. Any other
6104 names are discarded. This closes a loophole whereby a rogue DNS
6105 administrator could create reverse DNS records to break through a
6106 wildcarded host restriction in an ACL.
6108 37. If a user filter or a system filter that ran in a subprocess used any of
6109 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
6110 the wrong values were passed to the pipe command ($thisaddress had the
6111 value of $0, $0 had the value of $1, etc). This bug was introduced by
6112 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
6114 38. Improved the line breaking for long SMTP error messages from ACLs.
6115 Previously, if there was no break point between 40 and 75 characters, Exim
6116 left the rest of the message alone. Two changes have been made: (a) I've
6117 reduced the minimum length to 35 characters; (b) if it can't find a break
6118 point between 35 and 75 characters, it looks ahead and uses the first one
6119 that it finds. This may give the occasional overlong line, but at least the
6120 remaining text gets split now.
6122 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
6123 file descriptors might be low, and that setting 1000 would always raise it.
6124 It turns out that in some environments, the limit is already over 1000 and
6125 that lowering it causes trouble. So now Exim takes care not to decrease it.
6127 40. When delivering a message, the value of $return_path is set to $sender_
6128 address at the start of routing (routers may change the value). By an
6129 oversight, this default was not being set up when an address was tested by
6130 -bt or -bv, which affected the outcome if any router or filter referred to
6133 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
6134 log only when "message" or "log_message" are set. However, if one of the
6135 conditions was an address verification, or a call to a nested ACL, the
6136 messages generated by the underlying test were being passed through. This
6137 no longer happens. The underlying message is available in $acl_verify_
6138 message for both "message" and "log_message" expansions, so it can be
6139 passed through if needed.
6141 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
6142 new expansion $bh_ to give the encoded byte string without charset
6143 translation. Translation happens only if iconv() is available; HAVE_ICONV
6144 indicates this at build time. HEADERS_CHARSET gives the charset to
6145 translate to; headers_charset can change it in the configuration, and
6146 "headers charset" can change it in an individual filter file.
6148 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
6149 that creates RFC 2047 encoded "words" labels them as that charset instead
6150 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
6151 expansion operator; (ii) when Exim creates a From: line for a local
6152 message; (iii) when a header line is rewritten to include a "phrase" part.
6154 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
6155 buggy, causing it to skip the first lines of messages whose message ID
6156 ended in 'D'. This would not have bitten before Exim release 4.14, because
6157 message IDs were unlikely to end in 'D' before then. The effect was to have
6158 incorrect size information for certain domains.
6160 45. #include "config.h" was missing at the start of the crypt16.c module. This
6161 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
6164 46. If there was a timeout during a "random" callout check, Exim treated it as
6165 a failure of the random address, and carried on sending RSET and the real
6166 address. If the delay was just some slowness somewhere, the response to the
6167 original RCPT would be taken as a response to RSET and so on, causing
6168 mayhem of various kinds.
6170 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
6171 when I implemented it. It didn't allow for the fact that some option values
6172 may legitimatetly be negative (e.g. size_addition), and it didn't even do
6173 the right test for positive values.
6175 48. Domain names in DNS records are case-independent. Exim always looks them up
6176 in lower case. Some resolvers return domain names in exactly the case they
6177 appear in the zone file, that is, they may contain uppercase letters. Not
6178 all resolvers do this - some return always lower case. Exim was treating a
6179 change of case by a resolver as a change of domain, similar to a widening
6180 of a domain abbreviation. This triggered its re-routing code and so it was
6181 trying to route what was effectively the same domain again. This normally
6182 caused routing to fail (because the router wouldn't handle the domain
6183 twice). Now Exim checks for this case specially, and just changes the
6184 casing of the domain that it ultimately uses when it transmits the message
6187 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
6190 50. If a filter generated a file delivery with a non-absolute name (possible if
6191 no home directory exists for the router), the forbid_file option was not
6194 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
6195 the existing equality matching.
6197 52. Exim was using ints instead of ino_t variables in some places where it was
6198 dealing with inode numbers.
6200 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
6201 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
6202 called TMPDIR, and if it finds it is different, it changes its value.
6204 54. The smtp_printf() function is now made available to local_scan() so
6205 additional output lines can be written before returning. There is also an
6206 smtp_fflush() function to enable the detection of a dropped connection.
6207 The variables smtp_input and smtp_batched_input are exported to
6210 55. Changed the default runtime configuration: the message "Unknown user"
6211 has been removed from the ACL, and instead placed on the localuser router,
6212 using the cannot_route_message feature. This means that any verification
6213 failures that generate their own messages won't get overridden. Similarly,
6214 the "Unrouteable address" message that was in the ACL for unverifiable
6215 relay addresses has also been removed.
6217 56. Added hosts_avoid_esmtp to the smtp transport.
6219 57. The exicyclog script was not checking for the esoteric option
6220 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
6221 will work only if exicyclog is run under the appropriate euid.
6223 58. Following a discussion on the list, the rules by which Exim recognises line
6224 endings on incoming messages have been changed. The -dropcr and drop_cr
6225 options are now no-ops, retained only for backwards compatibility. The
6226 following line terminators are recognized: LF CRLF CR. However, special
6227 processing applies to CR:
6229 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
6230 nor a local message in the state where . is a terminator.
6232 (ii) If a bare CR is encountered in a header line, an extra space is added
6233 after the line terminator so as not to end the header. The reasoning
6234 behind this is that bare CRs in header lines are most likely either
6235 to be mistakes, or people trying to play silly games.
6237 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
6238 was being incorrectly given as 18 bytes larger than it should have been.
6239 This is a VOB (very old bug).
6241 60. This may never have affected anything current, but just in case it has:
6242 When the local host is found other than at the start of a list of hosts,
6243 the local host, those with the same MX, and any that follow, are discarded.
6244 When the list in question was part of a longer list of hosts, the following
6245 hosts (not currently being processed) were also being discarded. This no
6246 longer happens. I'm not sure if this situation could ever has previously
6249 61. Added the "/MX" feature to lists of hosts in the manualroute and query
6252 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
6253 header. This is something that is recommended in a new Internet Draft, and
6254 is something that is documented as being done by Sendmail. There are two
6255 possible values. For messages generated by the autoreply transport, Exim
6258 Auto-Submitted: auto-replied
6260 whereas for all other generated messages (e.g. bounces) it adds
6262 Auto-Submitted: auto-generated
6264 63. The "personal" condition in filters now includes a test for the
6265 Auto-Submitted: header. If it contains the string "auto-" the message it
6266 not considered personal.
6268 64. Added rcpt_include_affixes as a generic transport option.
6270 65. Added queue_only_override (default true).
6272 66. Added the syslog_duplication option.
6274 67. If what should have been the first header line of a message consisted of
6275 a space followed by a colon, Exim was mis-interpreting it as a header line.
6276 It isn't of course - it is syntactically invalid and should therefore be
6277 treated as the start of the message body. The misbehaviour could have
6278 caused a number of strange effects, including loss of data in subsequent
6279 header lines, and spool format errors.
6281 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
6282 client host had authenticated. This control can now be exercised by an ACL
6283 for more flexibility.
6285 69. By default, callouts do not happen when testing with -bh. There is now a
6286 variant, -bhc, which does actually run the callout code, including
6287 consulting and updating the callout cache.
6289 70. Added support for saslauthd authentication, courtesy of Alexander
6292 71. If statvfs() failed on the spool or log directories while checking their
6293 size for availability, Exim confusingly gave the error "space shortage".
6294 Furthermore, in debugging mode it crashed with a floating point exception.
6295 These checks are done if check_{spool,log}_{space,inodes} are set, and when
6296 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
6297 serious problem, Exim now writes to the main and panic logs when this
6298 happens, with details of the failure. It then refuses to accept the
6299 incoming message, giving the message "spool directory problem" or "log
6300 directory problem" with a 421 code for SMTP messages.
6302 72. When Exim is about to re-exec itself, it ensures that the file descriptors
6303 0, 1, and 2 exist, because some OS complain for execs without them (see
6304 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
6305 descriptors. However, the code omitted to check that the open succeeded,
6306 causing mysterious errors if for some reason the permissions on /dev/null
6307 got screwed. Now Exim writes a message to the main and panic logs, and
6308 bombs out if it can't open /dev/null.
6310 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
6311 interact so that it is all more flexible. It is supposed to remain
6312 backwards compatible. Also added extra_local_interfaces.
6314 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
6315 to bomb out with an assertion failure - to the client this appears as a
6316 connection drop. This problem occurs in the part of the code that was taken
6317 from the Samba project. Fortunately, the assertion is in a very simple
6318 function, so I have fixed this by reproducing the function inline in the
6319 one place where it is called, and arranging for authentication to fail
6320 instead of killing the process with assert().
6322 75. The SPA client code was not working when the server requested OEM rather
6323 than Unicode encoding.
6325 76. Added code to make require_files with a specific uid setting more usable in
6326 the case where statting the file as root fails - usually a non-root-mounted
6327 NFS file system. When this happens and the failure is EACCES, Exim now
6328 forks a subprocess and does the per-uid checking as the relevant uid.
6330 77. Added process_log_path.
6332 78. If log_file_path was not explicitly set, a setting of check_log_space or
6333 check_log_inodes was ignored.
6335 79. If a space check for the spool or log partitions fails, the incident is now
6336 logged. Of course, in the latter case the data may get lost...
6338 80. Added the %p formatting code to string_format() so that it can be used to
6339 print addresses in debug_print(). Adjusted all the address printing in the
6340 debugging in store.c to use %p rather than %d.
6342 81. There was a concern that a line of code in smtp_in.c could overflow a
6343 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
6344 initially expressed, the concern was not well-founded, because trailing
6345 spaces are removed early. However, if the trailing spaces were followed by
6346 a NULL, they did not get removed, so the overflow was possible. Two fixes
6349 (a) I re-wrote the offending code in a cleaner fashion.
6350 (b) If an incoming SMTP command contains a NULL character, it is rejected
6353 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
6354 runs initgroups(), so that if the Exim user is in any additional groups,
6355 they will be used during message reception.
6361 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / users / jgh / exim.git / blob