1 Change log file for Exim from version 4.21
2 -------------------------------------------
7 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
9 PP/02 Make -n do something, by making it not do something.
10 When combined with -bP, the name of an option is not output.
12 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
15 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
16 $sender_host_name and config options to manage this, and basic check
19 PP/05 DSCP support for outbound connections and control modifier for inbound.
21 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
22 (Only plugin which currently uses this is kerberos4, which nobody should
23 be using, but we should make it available and other future plugins might
24 conceivably use it, even though it would break NAT; stuff *should* be
25 using channel bindings instead).
27 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
28 name; added for Sendmail compatibility; requires admin caller.
29 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
30 just ignore it); requires trusted caller.
31 Also parse but ignore: -Ac -Am -X<logfile>
34 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
36 TL/02 Add +smtp_confirmation as a default logging option.
38 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
39 Patch by Magnus Holmgren from 2007-02-20.
41 TL/04 Bugzilla 1281 - Spec typo.
42 Bugzilla 1283 - Spec typo.
43 Bugzilla 1290 - Spec grammar fixes.
45 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
47 TL/06 Add Experimental DMARC support using libopendmarc libraries.
49 TL/07 Fix an out of order global option causing a segfault. Reported to dev
50 mailing list by by Dmitry Isaikin.
52 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
54 JH/02 Support "G" suffix to numbers in ${if comparisons.
56 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
58 NM/01 Bugzilla 1197 - Spec typo
59 Bugzilla 1196 - Spec examples corrections
61 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
63 PP/09 Add gnutls_enable_pkcs11 option.
65 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
66 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
68 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
69 "acl {{name}{arg}...}", and optional args on acl condition
72 JH/05 Permit multiple router/transport headers_add/remove lines.
74 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
76 JH/07 Avoid using a waiting database for a single-message-only transport.
77 Performance patch from Paul Fisher. Bugzilla 1262.
79 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
82 JH/09 Add $headers_added variable, with content from use of ACL modifier
83 add_header (but not yet added to the message). Bugzilla 199.
85 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
86 Pulled from Bugzilla 817 by Wolfgang Breyha.
88 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
90 (nb: this is the same fix as in Exim 4.80.1)
92 JH/11 Add A= logging on delivery lines, and a client_set_id option on
95 JH/12 Add optional authenticated_sender logging to A= and a log_selector
98 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
100 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
101 advertise SMTP AUTH mechanism to us, instead of a generic
102 protocol violation error. Also, make Exim more robust to bad
103 data from the Dovecot auth socket.
105 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
107 When a queue runner is handling a message, Exim first routes the
108 recipient addresses, during which it prunes them based on the retry
109 hints database. After that it attempts to deliver the message to
110 any remaining recipients. It then updates the hints database using
113 So if a recipient address works intermittently, it can get repeatedly
114 deferred at routing time. The retry hints record remains fresh so the
115 address never reaches the final cutoff time.
117 This is a fairly common occurrence when a user is bumping up against
118 their storage quota. Exim had some logic in its local delivery code
119 to deal with this. However it did not apply to per-recipient defers
120 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
122 This change adds a proper retry rule check during routing so that the
123 final cutoff time is checked against the message's age. We only do
124 this check if there is an address retry record and there is not a
125 domain retry record; this implies that previous attempts to handle
126 the address had the retry_use_local_parts option turned on. We use
127 this as an approximation for the destination being like a local
128 delivery, as in LMTP.
130 I suspect this new check makes the old local delivery cutoff check
131 redundant, but I have not verified this so I left the code in place.
133 TF/02 Correct gecos expansion when From: is a prefix of the username.
135 Test 0254 submits a message to Exim with the header
139 When I ran the test suite under the user fanf2, Exim expanded
140 the header to contain my full name, whereas it should have added
141 a Resent-Sender: header. It erroneously treats any prefix of the
142 username as equal to the username.
144 This change corrects that bug.
146 GF/01 DCC debug and logging tidyup
147 Error conditions log to paniclog rather than rejectlog.
148 Debug lines prefixed by "DCC: " to remove any ambiguity.
150 TF/03 Avoid unnecessary rebuilds of lookup-related code.
152 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
153 Bug spotted by Jeremy Harris; was flawed since initial commit.
154 Would have resulted in OCSP responses post-SNI triggering an Exim
155 NULL dereference and crash.
157 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
159 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
160 Bug detection, analysis and fix by Samuel Thibault.
161 Bugzilla 1331, Debian bug #698092.
163 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
165 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
166 Server implementation by Todd Lyons, client by JH.
167 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
168 config variable "prdr_enable" controls whether the server
169 advertises the facility. If the client requests PRDR a new
170 acl_data_smtp_prdr ACL is called once for each recipient, after
171 the body content is received and before the acl_smtp_data ACL.
172 The client is controlled by bolth of: a hosts_try_prdr option
173 on the smtp transport, and the server advertisement.
174 Default client logging of deliveries and rejections involving
175 PRDR are flagged with the string "PRDR".
177 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
178 fclose(). Diagnosis by Todd Lyons.
180 PP/17 Update configure.default to handle IPv6 localhost better.
181 Patch by Alain Williams (plus minor tweaks).
184 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
185 This is now consistent with GnuTLS, and is now documented: the
186 previous undocumented portable approach to treating the option as
187 unset was to force an expansion failure. That still works, and
188 an empty string is now equivalent.
190 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
191 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
192 not performing validation itself.
194 PP/20 Added force_command boolean option to pipe transport.
195 Patch from Nick Koston, of cPanel Inc.
197 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
200 TF/04 Added udpsend ACL modifer and hexquote expansion operator
202 PP/21 Fix eximon continuous updating with timestamped log-files.
203 Broken in a format-string cleanup in 4.80, missed when I repaired the
204 other false fix of the same issue.
205 Report and fix from Heiko Schlichting.
208 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
209 Report from Prashanth Katuri.
211 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
212 It's SecureTransport, so affects any MacOS clients which use the
213 system-integrated TLS libraries, including email clients.
215 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
216 using a MIME ACL for non-SMTP local injection.
217 Report and assistance in diagnosis by Warren Baker.
219 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
221 JH/16 Fix comparisons for 64b. Bugzilla 1385.
223 TL/09 Add expansion variable $authenticated_fail_id to keep track of
224 last id that failed so it may be referenced in subsequent ACL's.
226 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
229 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
230 ldap library initialization, allowing self-signed CA's to be
231 used. Also properly sets require_cert option later in code by
232 using NULL (global ldap config) instead of ldap handle (per
233 session). Bug diagnosis and testing by alxgomz.
235 TL/12 Enhanced documentation in the ratelimit.pl script provided in
236 the src/util/ subdirectory.
238 TL/13 Bug 1301 - Imported transport SQL logging patch from Axel Rau
239 renamed to Transport Post Delivery Action by Jeremy Harris, as
242 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
243 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
244 redis_servers = needs to be configured which will be used by the redis
245 lookup. Patch from Warren Baker, of The Packet Hub.
252 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
254 This, or similar/improved, will also be change PP/11 of 4.82.
260 PP/01 Handle short writes when writing local log-files.
261 In practice, only affects FreeBSD (8 onwards).
262 Bugzilla 1053, with thanks to Dmitry Isaikin.
264 NM/01 Bugzilla 949 - Documentation tweak
266 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
269 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
271 PP/02 Implemented gsasl authenticator.
273 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
275 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
276 `pkg-config foo` for cflags/libs.
278 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
279 with rest of GSASL and with heimdal_gssapi.
281 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
282 `pkg-config foo` for cflags/libs for the TLS implementation.
284 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
285 properties get this fed in as external SSF. A number of robustness
286 and debugging improvements to the cyrus_sasl authenticator.
288 PP/08 cyrus_sasl server now expands the server_realm option.
290 PP/09 Bugzilla 1214 - Log authentication information in reject log.
291 Patch by Jeremy Harris.
293 PP/10 Added dbmjz lookup type.
295 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
297 PP/12 MAIL args handles TAB as well as SP, for better interop with
298 non-compliant senders.
299 Analysis and variant patch by Todd Lyons.
301 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
302 Bug report from Lars Müller <lars@samba.org> (via SUSE),
303 Patch from Dirk Mueller <dmueller@suse.com>
305 PP/13 tls_peerdn now print-escaped for spool files.
306 Observed some $tls_peerdn in wild which contained \n, which resulted
307 in spool file corruption.
309 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
310 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
311 or write after TLS renegotiation, which otherwise led to messages
314 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
315 as a tracking header (ie: a signed header comes before the signature).
316 Patch from Wolfgang Breyha.
318 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
319 comma-sep list; embedded commas doubled.
321 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
323 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
325 Report and patch from Dmitry Banschikov.
327 PP/16 Removed "dont_insert_empty_fragments" fron "openssl_options".
328 Removed SSL_clear() after SSL_new() which led to protocol negotiation
329 failures. We appear to now support TLS1.1+ with Exim.
331 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
332 lets Exim select keys and certificates based upon TLS SNI from client.
333 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
334 before an outbound SMTP session. New log_selector, +tls_sni.
336 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
337 NULL dereference. Report and patch from Alun Jones.
339 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
340 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
341 Not seeing resolver debug output on NetBSD, but suspect this is a
342 resolver implementation change.
344 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
345 Left warnings. Added "eximon gdb" invocation mode.
347 PP/21 Defaulting "accept_8bitmime" to true, not false.
349 PP/22 Added -bw for inetd wait mode support.
351 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
352 locate the relevant includes and libraries. Made this the default.
354 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
355 Bugzilla 1246, report and most of solution from Tomasz Kusy.
357 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
358 This may cause build issues on older platforms.
360 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
361 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
362 gnutls_require_mac & gnutls_require_protocols (no longer supported).
363 Added SNI support via GnuTLS too.
364 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
366 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
368 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
369 Applied second patch from Janne, implementing suggestion to default
370 multiple-strings-in-record handling to match SPF spec.
372 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
374 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
375 read-only, out of scope).
376 Patch from Wolfgang Breyha, report from Stuart Northfield.
378 PP/29 Fix three issues highlighted by clang analyser static analysis.
379 Only crash-plausible issue would require the Cambridge-specific
380 iplookup router and a misconfiguration.
381 Report from Marcin Mirosław.
383 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
385 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
386 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
387 As part of this, removing so much warning spew let me fix some minor
388 real issues in debug logging.
390 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
391 assignment on my part. Fixed.
393 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
394 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
395 Janne Snabb (who went above and beyond: thank you).
397 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
398 string otherwise requires a connection and a bunch more work and it's
399 relatively easy to get wrong. Should also expose TLS library linkage
402 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
403 64-bit ${eval} (JH/03).
405 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
406 GNU libc to support some of the 64-bit stuff, should not lead to
407 conflicts. Defined before os.h is pulled in, so if a given platform
408 needs to override this, it can.
410 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
411 protection layer was required, which is not implemented.
412 Bugzilla 1254, patch from Wolfgang Breyha.
414 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
415 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
416 tls_dhparam take prime identifiers. Also unbreak combination of
417 OpenSSL+DH_params+TLSSNI.
419 PP/39 Disable SSLv2 by default in OpenSSL support.
425 PP/01 Solaris build fix for Oracle's LDAP libraries.
426 Bugzilla 1109, patch from Stephen Usher.
428 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
430 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
433 TF/02 Fix a couple more cases where we did not log the error message
434 when unlink() failed. See also change 4.74-TF/03.
436 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
437 lock up or crash if it happened to be inside a call to libc when it
438 got a SIGUSR1 from exiwhat.
440 The SIGUSR1 handler appends the current process status to the process
441 log which is later printed by exiwhat. It used to use the general
442 purpose logging code to do this, but several functions it calls are
443 not safe for signals.
445 The new output code in the SIGUSR1 handler is specific to the process
446 log, and simple enough that it's easy to inspect for signal safety.
447 Removing some special cases also simplifies the general logging code.
448 Removing the spurious timestamps from the process log simplifies
451 TF/04 Improved ratelimit ACL condition.
453 The /noupdate option has been deprecated in favour of /readonly which
454 has clearer semantics. The /leaky, /strict, and /readonly update modes
455 are mutually exclusive. The update mode is no longer included in the
456 database key; it just determines when the database is updated. (This
457 means that when you upgrde Exim will forget old rate measurements.)
459 Exim now checks that the per_* options are used with an update mode that
460 makes sense for the current ACL. For example, when Exim is processing a
461 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
462 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
463 must specify per_mail/readonly. If you omit the update mode it defaults to
464 /leaky where that makes sense (as before) or /readonly where required.
466 The /noupdate option is now undocumented but still supported for
467 backwards compatibility. It is equivalent to /readonly except that in
468 ACLs where /readonly is required you may specify /leaky/noupdate or
469 /strict/noupdate which are treated the same as /readonly.
471 A useful new feature is the /count= option. This is a generalization
472 of the per_byte option, so that you can measure the throughput of other
473 aggregate values. For example, the per_byte option is now equivalent
474 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
476 The per_rcpt option has been generalized using the /count= mechanism
477 (though it's more complicated than the per_byte equivalence). When it is
478 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
479 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
480 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
481 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
482 Note that using per_rcpt with a non-readonly update mode in more than
483 one ACL will cause the recipients to be double-counted. (The per_mail
484 and per_byte options don't have this problem.)
486 The handling of very low rates has changed slightly. If the computed rate
487 is less than the event's count (usually one) then this event is the first
488 after a long gap. In this case the rate is set to the same as this event's
489 count, so that the first message of a spam run is counted properly.
491 The major new feature is a mechanism for counting the rate of unique
492 events. The new per_addr option counts the number of different
493 recipients that someone has sent messages to in the last time period. It
494 behaves like per_rcpt if all the recipient addresses are different, but
495 duplicate recipient addresses do not increase the measured rate. Like
496 the /count= option this is a general mechanism, so the per_addr option
497 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
498 example, measure the rate that a client uses different sender addresses
499 with the options per_mail/unique=$sender_address. There are further
500 details in the main documentation.
502 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
504 TF/06 Removed a few PCRE remnants.
506 TF/07 Automatically extract Exim's version number from tags in the git
507 repository when doing development or release builds.
509 PP/02 Raise smtp_cmd_buffer_size to 16kB.
510 Bugzilla 879. Patch from Paul Fisher.
512 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
513 Heavily based on revision 40f9a89a from Simon Arlott's tree.
516 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
518 PP/05 Variable $av_failed, true if the AV scanner deferred.
519 Bugzilla 1078. Patch from John Horne.
521 PP/06 Stop make process more reliably on build failure.
522 Bugzilla 1087. Patch from Heiko Schlittermann.
524 PP/07 Make maildir_use_size_file an _expandable_ boolean.
525 Bugzilla 1089. Patch from Heiko Schlittermann.
527 PP/08 Handle ${run} returning more data than OS pipe buffer size.
528 Bugzilla 1131. Patch from Holger Weiß.
530 PP/09 Handle IPv6 addresses with SPF.
531 Bugzilla 860. Patch from Wolfgang Breyha.
533 PP/10 GnuTLS: support TLS 1.2 & 1.1.
535 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
538 PP/11 match_* no longer expand right-hand-side by default.
539 New compile-time build option, EXPAND_LISTMATCH_RHS.
540 New expansion conditions, "inlist", "inlisti".
542 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
544 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
546 PP/14 fix log_write() format string regression from TF/03.
547 Bugzilla 1152. Patch from Dmitry Isaikin.
553 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
555 PP/02 Harmonised TLS library version reporting; only show if debugging.
556 Layout now matches that introduced for other libraries in 4.74 PP/03.
558 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
560 PP/04 New "dns_use_edns0" global option.
562 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
565 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
566 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
568 TK/01 Updated PolarSSL code to 0.14.2.
569 Bugzilla 1097. Patch from Andreas Metzler.
571 PP/07 Catch divide-by-zero in ${eval:...}.
574 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
577 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
578 format-string attack -- SECURITY: remote arbitrary code execution.
580 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
581 time unintentionally subject to list matching rules, letting the header
582 cause arbitrary Exim lookups (of items which can occur in lists, *not*
583 arbitrary string expansion). This allowed for information disclosure.
585 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
586 INT_MIN/-1 -- value coerced to INT_MAX.
592 NM/01 Workround for PCRE version dependency in version reporting
595 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
596 This fixes portability to compilers other than gcc, notably
597 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
599 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
600 makefiles for portability to HP-UX and POSIX correctness.
602 PP/01 Permit LOOKUP_foo enabling on the make command-line.
603 Also via indented variable definition in the Makefile.
604 (Debugging by Oliver Heesakkers).
606 PP/02 Restore caching of spamd results with expanded spamd_address.
607 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
609 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
610 Improves build reliability. Fix from: Frank Elsner
612 NM/02 Fix wide character breakage in the rfc2047 coding
613 Fixes bug 1064. Patch from Andrey N. Oktyabrski
615 NM/03 Allow underscore in dnslist lookups
616 Fixes bug 1026. Patch from Graeme Fowler
618 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
619 Code patches from Adam Ciarcinski of NetBSD.
621 NM/04 Fixed exiqgrep to cope with mailq missing size issue
624 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
625 is logged, to avoid truncation. Patch from John Horne.
627 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
628 Patch from Jakob Hirsch.
630 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
631 SQL string expansion failure details.
632 Patch from Andrey Oktyabrski.
634 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
635 Patch from Simon Arlott.
637 PP/09 New lookups functionality failed to compile on old gcc which rejects
638 extern declarations in function scope.
639 Patch from Oliver Fleischmann
641 PP/10 Use sig_atomic_t for flags set from signal handlers.
642 Check getgroups() return and improve debugging.
643 Fixed developed for diagnosis in bug 927 (which turned out to be
646 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
647 Patch from Mark Zealey.
649 PP/12 Bugzilla 1056: Improved spamd server selection.
650 Patch from Mark Zealey.
652 PP/13 Bugzilla 1086: Deal with maildir quota file races.
653 Based on patch from Heiko Schlittermann.
655 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
656 Patch from Uwe Doering, sign-off by Michael Haardt.
658 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
659 variable declaration deep within a block. Bug and patch from
662 PP/15 lookups-Makefile IRIX compatibilty coercion.
664 PP/16 Make DISABLE_DKIM build knob functional.
666 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
667 Patch by Simon Arlott
669 TF/03 Fix valgrind.h portability to C89 compilers that do not support
670 variable argument macros. Our copy now differs from upstream.
676 TF/01 Failure to get a lock on a hints database can have serious
677 consequences so log it to the panic log.
679 TF/02 Log LMTP confirmation messages in the same way as SMTP,
680 controlled using the smtp_confirmation log selector.
682 TF/03 Include the error message when we fail to unlink a spool file.
684 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
685 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
686 for maintaining out-of-tree patches for some time.
688 PP/01 Bugzilla 139: Documentation and portability issues.
689 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
690 Handle per-OS dynamic-module compilation flags.
692 PP/02 Let /dev/null have normal permissions.
693 The 4.73 fixes were a little too stringent and complained about the
694 permissions on /dev/null. Exempt it from some checks.
695 Reported by Andreas M. Kirchwitz.
697 PP/03 Report version information for many libraries, including
698 Exim version information for dynamically loaded libraries. Created
699 version.h, now support a version extension string for distributors
700 who patch heavily. Dynamic module ABI change.
702 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
703 privilege escalation vulnerability whereby the Exim run-time user
704 can cause root to append content of the attacker's choosing to
707 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
710 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
711 If dropping privileges for untrusted macros, we disabled normal logging
712 on the basis that it would fail; for the Exim run-time user, this is not
713 the case, and it resulted in successful deliveries going unlogged.
714 Fixed. Reported by Andreas Metzler.
720 PP/01 Date: & Message-Id: revert to normally being appended to a message,
721 only prepend for the Resent-* case. Fixes regression introduced in
722 Exim 4.70 by NM/22 for Bugzilla 607.
724 PP/02 Include check_rfc2047_length in configure.default because we're seeing
725 increasing numbers of administrators be bitten by this.
727 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
729 PP/03 Bugzilla 994: added openssl_options main configuration option.
731 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
733 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
735 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
737 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
738 without a peer certificate, leading to a segfault because of an
739 assumption that peers always have certificates. Be a little more
740 paranoid. Problem reported by Martin Tscholak.
742 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
743 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
744 NB: ClamAV planning to remove STREAM in "middle of 2010".
745 CL also introduces -bmalware, various -d+acl logging additions and
746 more caution in buffer sizes.
748 PP/09 Implemented reverse_ip expansion operator.
750 PP/10 Bugzilla 937: provide a "debug" ACL control.
752 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
754 PP/12 Bugzilla 973: Implement --version.
756 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
758 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
760 PP/15 Bugzilla 816: support multiple condition rules on Routers.
762 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
763 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
764 ignore trailing whitespace.
766 JJ/02 prevent non-panic DKIM error from being sent to paniclog
768 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
771 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
772 Notification from Dr Andrew Aitchison.
774 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
775 ExtendedDetectionInfo response format.
776 Notification from John Horne.
778 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
781 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
782 XSL and documented dependency on system catalogs, with examples of how
785 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
788 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
789 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
790 configuration file which is writeable by the Exim user or group.
792 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
793 of configuration files to cover files specified with the -C option if
794 they are going to be used with root privileges, not just the default
797 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
798 option (effectively making it always true).
800 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
801 files to be used while preserving root privileges.
803 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
804 that rogue child processes cannot use them.
806 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
807 run-time user, instead of root.
809 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
810 Exim run-time user without dropping privileges.
812 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
813 result string, instead of calling string_vformat() twice with the same
816 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
817 for other users. Others should always drop root privileges if they use
818 -C on the command line, even for a whitelisted configure file.
820 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
822 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
828 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
829 $data_path, and $header_path variables; fixed documentation bugs and
832 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
833 exipick to access non-standard spools, including the "frozen" queue
836 NM/01 Bugzilla 965: Support mysql stored procedures.
837 Patch from Alain Williams
839 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
841 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
842 Patch from Andreas Metzler
844 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
845 Patch from Kirill Miazine
847 NM/05 Bugzilla 671: Added umask to procmail example.
849 JJ/03 installed exipick 20100323.0, fixing doc bug
851 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
852 directory. Notification and patch from Dan Rosenberg.
854 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
856 TK/02 Improve log output when DKIM signing operation fails.
858 MH/01 Treat the transport option dkim_domain as a colon separated
859 list, not as a single string, and sign the message with each element,
860 omitting multiple occurences of the same signer.
862 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
863 Bugzilla 985, 986. Patch by Simon Arlott
865 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
866 Patch by Simon Arlott
868 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
869 MBX locking. Notification from Dan Rosenberg.
875 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
877 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
879 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
881 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
883 NM/04 Bugzilla 915: Flag broken perl installation during build.
889 TK/01 Added patch by Johannes Berg that expands the main option
890 "spamd_address" if it starts with a dollar sign.
892 TK/02 Write list of recipients to X-Envelope-Sender header when building
893 the mbox-format spool file for content scanning (suggested by Jakob
896 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
897 (http://www.dcc-servers.net/) support via dccifd. Activated by
898 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
900 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
901 by Mark Daniel Reidel <mr@df.eu>.
903 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
904 When building exim an external PCRE library is now needed -
905 PCRE is a system library on the majority of modern systems.
906 See entry on PCRE_LIBS in EDITME file.
908 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
909 conversation. Added nologin parameter to request.
910 Patch contributed by Kirill Miazine.
912 TF/01 Do not log submission mode rewrites if they do not change the address.
914 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
916 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
917 log files in place. Contributed by Roberto Lima.
919 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
921 TF/03 Bugzilla 615: When checking the local_parts router precondition
922 after a local_part_suffix or local_part_prefix option, Exim now
923 does not use the address's named list lookup cache, since this
924 contains cached lookups for the whole local part.
926 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
927 Robert Millan. Documentation is in experimental-spec.txt.
929 TF/04 Bugzilla 668: Fix parallel build (make -j).
931 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
933 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
934 Patch provided by Jan Srzednicki.
936 TF/05 Leading white space used to be stripped from $spam_report which
937 wrecked the formatting. Now it is preserved.
939 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
940 that they are available at delivery time.
942 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
944 TF/08 TLS error reporting now respects the incoming_interface and
945 incoming_port log selectors.
947 TF/09 Produce a more useful error message if an SMTP transport's hosts
948 setting expands to an empty string.
950 NM/06 Bugzilla 744: EXPN did not work under TLS.
951 Patch provided by Phil Pennock.
953 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
954 Patch provided by Richard Godbee.
956 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
957 acl_smtp_notquit, added index entry.
959 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
960 Patch provided by Eugene Bujak.
962 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
963 accept(). Patch provided by Maxim Dounin.
965 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
966 Patch provided by Phil Pennock.
968 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
970 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
971 Patch provided by Brad "anomie" Jorsch.
973 NM/14 Bugzilla 622: Added timeout setting to transport filter.
974 Patch provided by Dean Brooks.
976 TK/05 Add native DKIM support (does not depend on external libraries).
978 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
979 Patch provided by Graeme Fowler.
981 NM/16 Bugzilla 851: Documentation example syntax fix.
983 NM/17 Changed NOTICE file to remove references to embedded PCRE.
985 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
988 NM/19 Bugzilla 745: TLS version reporting.
989 Patch provided by Phil Pennock.
991 NM/20 Bugzilla 167: bool: condition support.
992 Patch provided by Phil Pennock.
994 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
995 clients. Patch provided by Phil Pennock.
997 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
998 Patch provided by Brad "anomie" Jorsch.
1000 NM/23 Bugzilla 687: Fix misparses in eximstats.
1001 Patch provided by Heiko Schlittermann.
1003 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
1004 Patch provided by Heiko Schlittermann.
1006 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
1007 plus update to original patch.
1009 NM/26 Bugzilla 799: Documentation correction for ratelimit.
1011 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
1012 Patch provided by David Brownlee.
1014 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
1016 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
1018 NM/30 Bugzilla 888: TLS documentation bugfixes.
1020 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
1022 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
1023 Unlike the original bugzilla I have changed all shell scripts in src tree.
1025 NM/33 Bugzilla 898: Transport filter timeout fix.
1026 Patch by Todd Rinaldo.
1028 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
1029 Patch by Serge Demonchaux.
1031 NM/35 Bugzilla 39: Base64 decode bug fixes.
1032 Patch by Jakob Hirsch.
1034 NM/36 Bugzilla 909: Correct connect() call in dcc code.
1036 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
1038 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
1040 NM/39 Bugzilla 911: Fixed MakeLinks build script.
1046 TK/01 Add preliminary DKIM support. Currently requires a forked version of
1047 ALT-N's libdkim that I have put here:
1048 http://duncanthrax.net/exim-experimental/
1050 Note to Michael Haardt: I had to rename some vars in sieve.c. They
1051 were called 'true' and it seems that C99 defines that as a reserved
1052 keyword to be used with 'bool' variable types. That means you could
1053 not include C99-style headers which use bools without triggering
1054 build errors in sieve.c.
1056 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
1057 as mailq or other aliases. Changed the --help handling significantly
1058 to do whats expected. exim_usage() emits usage/help information.
1060 SC/01 Added the -bylocaldomain option to eximstats.
1062 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
1064 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
1066 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
1072 PH/01 Another patch from the Sieve maintainer.
1074 PH/02 When an IPv6 address is converted to a string for single-key lookup
1075 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
1076 dots are used instead of colons so that keys in lsearch files need not
1077 contain colons. This was done some time before quoting was made available
1078 in lsearch files. However, iplsearch files do require colons in IPv6 keys
1079 (notated using the quote facility) so as to distinguish them from IPv4
1080 keys. This meant that lookups for IP addresses in host lists did not work
1081 for iplsearch lookups.
1083 This has been fixed by arranging for IPv6 addresses to be expressed with
1084 colons if the lookup type is iplsearch. This is not incompatible, because
1085 previously such lookups could never work.
1087 The situation is now rather anomolous, since one *can* have colons in
1088 ordinary lsearch keys. However, making the change in all cases is
1089 incompatible and would probably break a number of configurations.
1091 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
1094 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
1095 conversion specification without a maximum field width, thereby enabling
1096 a rogue spamd server to cause a buffer overflow. While nobody in their
1097 right mind would setup Exim to query an untrusted spamd server, an
1098 attacker that gains access to a server running spamd could potentially
1099 exploit this vulnerability to run arbitrary code as the Exim user.
1101 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
1102 $primary_hostname instead of what libspf2 thinks the hosts name is.
1104 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
1105 a directory entry by the name of the lookup key. Previously, if a
1106 symlink pointed to a non-existing file or a file in a directory that
1107 Exim lacked permissions to read, a lookup for a key matching that
1108 symlink would fail. Now it is enough that a matching directory entry
1109 exists, symlink or not. (Bugzilla 503.)
1111 PH/03 The body_linecount and body_zerocount variables are now exported in the
1114 PH/04 Added the $dnslist_matched variable.
1116 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
1117 This means they are set thereafter only if the connection becomes
1120 PH/06 Added the client_condition to authenticators so that some can be skipped
1121 by clients under certain conditions.
1123 PH/07 The error message for a badly-placed control=no_multiline_responses left
1124 "_responses" off the end of the name.
1126 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
1128 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
1129 (without spaces) instead of just copying the configuration text.
1131 PH/10 Added the /noupdate option to the ratelimit ACL condition.
1133 PH/11 Added $max_received_linelength.
1135 PH/12 Added +ignore_defer and +include_defer to host lists.
1137 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
1138 way in which PCRE > 7.0 is built.
1140 PH/14 Implemented queue_only_load_latch.
1142 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
1143 MAIL command. The effect was to mangle the value on 64-bit systems.
1145 PH/16 Another patch from the Sieve maintainer.
1147 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
1149 PH/18 If a system quota error occurred while trying to create the file for
1150 a maildir delivery, the message "Mailbox is full" was not appended to the
1151 bounce if the delivery eventually timed out. Change 4.67/27 below applied
1152 only to a quota excession during the actual writing of the file.
1154 PH/19 It seems that peer DN values may contain newlines (and other non-printing
1155 characters?) which causes problems in log lines. The DN values are now
1156 passed through string_printing() before being added to log lines.
1158 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
1159 and InterBase are left for another time.)
1161 PH/21 Added message_body_newlines option.
1163 PH/22 Guard against possible overflow in moan_check_errorcopy().
1165 PH/23 POSIX allows open() to be a macro; guard against that.
1167 PH/24 If the recipient of an error message contained an @ in the local part
1168 (suitably quoted, of course), incorrect values were put in $domain and
1169 $local_part during the evaluation of errors_copy.
1175 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
1176 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
1179 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
1180 issue a MAIL command.
1182 PH/02 In an ACL statement such as
1184 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
1186 if a client was not listed at all, or was listed with a value other than
1187 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
1188 the condition was not true (as it should be), so access was not denied.
1189 The bug was that the ! inversion was incorrectly passed on to the second
1190 item. This has been fixed.
1192 PH/03 Added additional dnslists conditions == and =& which are different from
1193 = and & when the dns lookup returns more than one IP address.
1195 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
1196 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
1198 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
1199 FSYNC, which compiles an option called disable_fsync that allows for
1200 bypassing fsync(). The documentation is heavily laced with warnings.
1202 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
1204 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
1205 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
1206 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
1207 including adding "make clean"; (3) Added -fPIC when compiling the test
1208 dynamically loaded module, to get rid of a warning.
1210 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
1211 message fails, move_frozen_messages = true and ignore_bounce_errors_after
1212 = 0s. The bug is otherwise harmless.
1214 PH/07 There was a bug in the dovecot authenticator such that the value of
1215 $auth1 could be overwritten, and so not correctly preserved, after a
1216 successful authentication. This usually meant that the value preserved by
1217 the server_setid option was incorrect.
1219 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
1221 PH/09 Installed PCRE release 7.0.
1223 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
1224 run for batched SMTP input. It is now run at the start of every message
1225 in the batch. While fixing this I discovered that the process information
1226 (output by running exiwhat) was not always getting set for -bs and -bS
1227 input. This is fixed, and it now also says "batched" for BSMTP.
1229 PH/11 Added control=no_pipelining.
1231 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
1232 patch, slightly modified), and move the expansion of helo_data till after
1233 the connection is made in the smtp transport (so it can use these
1236 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
1238 PH/14 Added log_selector = +pid.
1240 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
1242 PH/16 Add ${if forany and ${if forall.
1244 PH/17 Added dsn_from option to vary the From: line in DSNs.
1246 PH/18 Flush SMTP output before performing a callout, unless control =
1247 no_callout_flush is set.
1249 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
1250 was true (the default) a successful delivery failed to delete the retry
1251 item, thus causing premature timeout of the address. The bug is now
1254 PH/20 Added hosts_avoid_pipelining to the smtp transport.
1256 PH/21 Long custom messages for fakedefer and fakereject are now split up
1257 into multiline reponses in the same way that messages for "deny" and
1258 other ACL rejections are.
1260 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
1261 with slight modification.
1263 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
1264 draft, changing the syntax and factoring some duplicate code".
1266 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
1267 for deliveries of the second and subsequent messages over the same SMTP
1270 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
1271 ${reduce, with only minor "tidies".
1273 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
1275 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
1276 expansion side effects.
1278 PH/27 When a message times out after an over-quota error from an Exim-imposed
1279 quota, the bounce message says "mailbox is full". This message was not
1280 being given when it was a system quota that was exceeded. It now should
1283 MH/03 Made $recipients available in local_scan(). local_scan() already has
1284 better access to the recipient list through recipients_list[], but
1285 $recipients can be useful in postmaster-provided expansion strings.
1287 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
1288 in the case of a MAIL command with additional options following the
1289 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
1290 were accidentally chopped off.
1292 PH/29 SMTP synchronization checks are implemented when a command is read -
1293 there is a check that no more input is waiting when there shouldn't be
1294 any. However, for some commands, a delay in an ACL can mean that it is
1295 some time before the response is written. In this time, more input might
1296 arrive, invalidly. So now there are extra checks after an ACL has run for
1297 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
1298 pipelining has not been advertised.
1300 PH/30 MH's patch to allow iscntrl() characters to be list separators.
1302 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
1303 returned in the SMTP response when smtp_return_error_details was false.
1304 This has been fixed.
1306 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
1307 instead of the C I/O that was originally supplied, because problems were
1308 reported on Solaris.
1310 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
1311 Exim which did not show up earlier: it was assuming that a call to
1312 SSL_CTX_set_info_callback() might give an error value. In fact, there is
1313 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
1314 was a macro that became an assignment, so it seemed to work. This has
1315 changed to a proper function call with a void return, hence the compile
1316 error. Exim's code has been fixed.
1318 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
1321 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
1323 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
1325 PH/37 If a message is not accepted after it has had an id assigned (e.g.
1326 because it turns out to be too big or there is a timeout) there is no
1327 "Completed" line in the log. When some messages of this type were
1328 selected by exigrep, they were listed as "not completed". Others were
1329 picked up by some special patterns. I have improved the selection
1330 criteria to be more general.
1332 PH/38 The host_find_failed option in the manualroute router can now be set
1333 to "ignore", to completely ignore a host whose IP address cannot be
1334 found. If all hosts are ignored, the behaviour is controlled by the new
1335 host_all_ignored option.
1337 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
1338 homing or because of multiple MX records with /mx) generated more than
1339 one IP address, and the following item turned out to be the local host,
1340 all the secondary addresses of the first item were incorrectly removed
1341 from the list, along with the local host and any following hosts (which
1342 is what is supposed to happen).
1344 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
1345 whoever called Exim into the -H file. In the case of the daemon it was
1346 behaving confusingly. When first started, it used values for whoever
1347 started the daemon, but after a SIGHUP it used the Exim user (because it
1348 calls itself on a restart). I have changed the code so that it now always
1351 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
1352 message are rejected with the same error (e.g. no authentication or bad
1353 sender address), and a DATA command is nevertheless sent (as can happen
1354 with PIPELINING or a stupid MUA), the error message that was given to the
1355 RCPT commands is included in the rejection of the DATA command. This is
1356 intended to be helpful for MUAs that show only the final error to their
1359 PH/42 Another patch from the Sieve maintainer.
1361 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
1362 Eximstats - Fixed some broken HTML links and added missing column headers
1364 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
1365 columns for Rejects, Temp Rejects, Ham, and Spam rows.
1367 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
1369 PH/43 Yet another patch from the Sieve maintainer.
1371 PH/44 I found a way to check for a TCP/IP connection going away before sending
1372 the response to the final '.' that terminates a message, but only in the
1373 case where the client has not sent further data following the '.'
1374 (unfortunately, this is allowed). However, in many cases there won't be
1375 any further data because there won't be any more messages to send. A call
1376 to select() can be used: if it shows that the input is "ready", there is
1377 either input waiting, or the socket has been closed. An attempt to read
1378 the next input character can distinguish the two cases. Previously, Exim
1379 would have sent an OK response which the client would never have see.
1380 This could lead to message repetition. This fix should cure that, at
1381 least in a lot of common cases.
1383 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
1384 advertised in response to EHLO.
1390 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
1391 fixed by 4.65/MH/01 (is this a record?) are fixed:
1393 (i) An empty string was always treated as zero by the numeric comparison
1394 operators. This behaviour has been restored.
1396 (ii) It is documented that the numeric comparison operators always treat
1397 their arguments as decimal numbers. This was broken in that numbers
1398 starting with 0 were being interpreted as octal.
1400 While fixing these problems I realized that there was another issue that
1401 hadn't been noticed. Values of message_size_limit (both the global option
1402 and the transport option) were treated as octal if they started with 0.
1403 The documentation was vague. These values are now always treated as
1404 decimal, and I will make that clear in the documentation.
1410 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
1411 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
1414 MH/01 Don't check that the operands of numeric comparison operators are
1415 integers when their expansion is in "skipping" mode (fixes bug
1416 introduced by 4.64-PH/07).
1418 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
1419 child addresses, Exim now panics and dies. Previously, because the count
1420 is held in a short int, deliveries were likely to be lost. As such a
1421 large number of recipients for a single message is ridiculous
1422 (performance will be very, very poor), I have chosen to impose a limit
1423 rather than extend the field.
1429 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
1430 leftover -K file (the existence of which was triggered by #402).
1431 While we were at it, introduced process PID as part of the -K
1432 filename. This should rule out race conditions when creating
1435 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
1436 processing considerably. Previous code took too long for large mails,
1437 triggering a timeout which in turn triggers #401.
1439 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
1440 in the DK code in transports.c. sendfile() is not really portable,
1441 hence the _LINUX specificness.
1443 TF/01 In the add_headers option to the mail command in an Exim filter,
1444 there was a bug that Exim would claim a syntax error in any
1445 header after the first one which had an odd number of characters
1448 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
1449 callout verification, Exim cached a "reject" for the entire domain. This
1450 is correct for most verifications, but it is not correct for a recipient
1451 verification with use_sender or use_postmaster set, because in that case
1452 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
1453 case of MAIL FROM:<> rejection from other early rejections (e.g.
1454 rejection of HELO). When verifying a recipient using a non-null MAIL
1455 address, the cache is ignored if it shows MAIL FROM:<> rejection.
1456 Whatever the result of the callout, the value of the domain cache is
1457 left unchanged (for any other kind of callout, getting as far as trying
1458 RCPT means that the domain itself is ok).
1460 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
1463 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
1464 manifest itself as EPIPE rather than ECONNECT. When tidying away a
1465 session, the daemon ignores ECONNECT errors and logs others; it now
1466 ignores EPIPE as well.
1468 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
1469 (quoted-printable decoding).
1471 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
1472 later the small subsequent patch to fix an introduced bug.
1474 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
1476 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
1478 PH/08 An error is now given if message_size_limit is specified negative.
1480 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
1481 to be given (somewhat) arbitrary names.
1483 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
1486 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
1487 miscellaneous code fixes
1489 PH/10 Added the log_reject_target ACL modifier to specify where to log
1492 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
1493 hostname. This is wrong, because it relates to the incoming message (and
1494 probably the interface on which it is arriving) and not to the outgoing
1495 callout (which could be using a different interface). This has been
1496 changed to use the value of the helo_data option from the smtp transport
1497 instead - this is what is used when a message is actually being sent. If
1498 there is no remote transport (possible with a router that sets up host
1499 addresses), $smtp_active_hostname is used.
1501 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
1502 tweaks were necessary in order to get it to work (see also 21 below):
1503 (a) The code assumed that strncpy() returns a negative number on buffer
1504 overflow, which isn't the case. Replaced with Exim's string_format()
1506 (b) There were several signed/unsigned issues. I just did the minimum
1507 hacking in of casts. There is scope for a larger refactoring.
1508 (c) The code used strcasecmp() which is not a standard C function.
1509 Replaced with Exim's strcmpic() function.
1510 (d) The code set only $1; it now sets $auth1 as well.
1511 (e) A simple test gave the error "authentication client didn't specify
1512 service in request". It would seem that Dovecot has changed its
1513 interface. Fortunately there's a specification; I followed it and
1514 changed what the client sends and it appears to be working now.
1516 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
1519 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
1520 address A is aliased to B and C, where B exists and C does not. Without
1521 -v the output is "A verified" because verification stops after a
1522 successful redirection if more than one address is generated. However,
1523 with -v the child addresses are also verified. Exim was outputting "A
1524 failed to verify" and then showing the successful verification for C,
1525 with its parentage. It now outputs "B failed to verify", showing B's
1526 parentage before showing the successful verification of C.
1528 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
1529 look up a TXT record in a specific list after matching in a combined
1532 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
1533 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
1534 they consult the DNS. I had assumed they would set it the way they
1535 wanted; and indeed my experiments on Linux seem to show that in some
1536 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
1537 To be on the safe side, however, I have now made the interface to
1538 host_find_byname() similar to host_find_bydns(), with an argument
1539 containing the DNS resolver options. The host_find_byname() function now
1540 sets these options at its start, just as host_find_bydns() does. The smtp
1541 transport options dns_qualify_single and dns_search_parents are passed to
1542 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
1543 of host_find_byname() use the default settings of RES_DEFNAMES
1544 (qualify_single) but not RES_DNSRCH (search_parents).
1546 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
1547 spool_read_header() do less string testing, by means of a preliminary
1548 switch on the second character of optional "-foo" lines. (This is
1549 overdue, caused by the large number of possibilities that now exist.
1550 Originally there were few.) While I was there, I also converted the
1551 str(n)cmp tests so they don't re-test the leading "-" and the first
1552 character, in the hope this might squeeze out yet more improvement.
1554 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
1555 flag allowing group syntax was set by the header_syntax check but not
1556 turned off, possible causing trouble later; (2) The flag was not being
1557 set at all for the header_verify test, causing "group"-style headers to
1558 be rejected. I have now set it in this case, and also caused header_
1559 verify to ignore an empty address taken from a group. While doing this, I
1560 came across some other cases where the code for allowing group syntax
1561 while scanning a header line wasn't quite right (mostly, not resetting
1562 the flag correctly in the right place). These bugs could have caused
1563 trouble for malformed header lines. I hope it is now all correct.
1565 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
1566 with the "reply" argument non-NULL. The code, however (which originally
1567 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
1568 but it didn't always do it. This confused somebody who was copying the
1569 code for some other use. I have removed all the tests.
1571 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
1572 feature that was used to support insecure browsers during the U.S. crypto
1573 embargo. It requires special client support, and Exim is probably the
1574 only MTA that supported it -- and would never use it because real RSA is
1575 always available. This code has been removed, because it had the bad
1576 effect of slowing Exim down by computing (never used) parameters for the
1577 RSA_EXPORT functionality.
1579 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
1580 authenticator to fail if there's a tab character in the incoming data
1581 (there should never be unless someone is messing about, as it's supposed
1582 to be base64-encoded). Also added, on Timo's advice, the "secured" option
1583 if the connection is using TLS or if the remote IP is the same as the
1584 local IP, and the "valid-client-cert option" if a client certificate has
1587 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
1588 authenticators. This can be used for authorization after authentication
1589 succeeds. (In the case of plaintext, it servers for both authentication
1592 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
1593 if any retry times were supplied.
1595 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
1596 connection, where there is no client IP address to check. In this
1597 situation, the verify now always succeeds.
1599 PH/25 Applied John Jetmore's -Mset patch.
1601 PH/26 Added -bem to be like -Mset, but loading a message from a file.
1603 PH/27 In a string expansion for a processed (not raw) header when multiple
1604 headers of the same name were present, leading whitespace was being
1605 removed from all of them, but trailing whitespace was being removed only
1606 from the last one. Now trailing whitespace is removed from each header
1607 before concatenation. Completely empty headers in a concatenation (as
1608 before) are ignored.
1610 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
1611 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
1613 PH/29 [Removed. This was a change that I later backed out, and forgot to
1614 correct the ChangeLog entry (that I had efficiently created) before
1615 committing the later change.]
1617 PH/30 Exim was sometimes attempting to deliver messages that had suffered
1618 address errors (4xx response to RCPT) over the same connection as other
1619 messages routed to the same hosts. Such deliveries are always "forced",
1620 so retry times are not inspected. This resulted in far too many retries
1621 for the affected addresses. The effect occurred only when there were more
1622 hosts than the hosts_max_try setting in the smtp transport when it had
1623 the 4xx errors. Those hosts that it had tried were not added to the list
1624 of hosts for which the message was waiting, so if all were tried, there
1625 was no problem. Two fixes have been applied:
1627 (i) If there are any address or message errors in an SMTP delivery, none
1628 of the hosts (tried or untried) are now added to the list of hosts
1629 for which the message is waiting, so the message should not be a
1630 candidate for sending over the same connection that was used for a
1631 successful delivery of some other message. This seems entirely
1632 reasonable: after all the message is NOT "waiting for some host".
1633 This is so "obvious" that I'm not sure why it wasn't done
1634 previously. Hope I haven't missed anything, but it can't do any
1635 harm, as the worst effect is to miss an optimization.
1637 (ii) If, despite (i), such a delivery is accidentally attempted, the
1638 routing retry time is respected, so at least it doesn't keep
1639 hammering the server.
1641 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
1642 in ${readsocket because some servers need this prod.
1644 PH/32 Added some extra debug output when updating a wait-xxx database.
1646 PH/33 The hint "could be header name not terminated by colon", which has been
1647 given for certain expansion errors for a long time, was not being given
1648 for the ${if def:h_colon_omitted{... case.
1650 PH/34 The spec says: "With one important exception, whenever a domain list is
1651 being scanned, $domain contains the subject domain." There was at least
1652 one case where this was not true.
1654 PH/35 The error "getsockname() failed: connection reset by peer" was being
1655 written to the panic log as well as the main log, but it isn't really
1656 panic-worthy as it just means the connection died rather early on. I have
1657 removed the panic log writing for the ECONNRESET error when getsockname()
1660 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
1661 runs only) independently of the message's sender address. This meant
1662 that, if the 4xx error was in fact related to the sender, a different
1663 message to the same recipient with a different sender could confuse
1664 things. In particualar, this can happen when sending to a greylisting
1665 server, but other circumstances could also provoke similar problems.
1666 I have changed the default so that the retry time for these errors is now
1667 based a combination of the sender and recipient addresses. This change
1668 can be overridden by setting address_retry_include_sender=false in the
1671 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
1672 remote server are returned as part of bounce messages. This was not
1673 happening for LMTP over a pipe (the lmtp transport), but now it is the
1674 same for both kinds of LMTP.
1676 PH/38 Despite being documented as not happening, Exim was rewriting addresses
1677 in header lines that were in fact CNAMEs. This is no longer the case.
1679 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
1680 and queue runs started by the daemon processed all messages. This has
1681 been fixed so that -R and -S can now usefully be given with -q<time>.
1683 PH/40 Import PCRE release 6.7 (fixes some bugs).
1685 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
1687 PH/42 Give an error if -q is specified more than once.
1689 PH/43 Renamed the variables $interface_address and $interface_port as
1690 $received_ip_address and $received_port, to make it clear that these
1691 values apply to message reception, and not to the outgoing interface when
1692 a message is delivered. (The old names remain recognized, of course.)
1694 PH/44 There was no timeout on the connect() call when using a Unix domain
1695 socket in the ${readsocket expansion. There now is.
1697 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
1698 be meaningful with "accept".
1700 SC/01 Eximstats V1.43
1701 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
1703 SC/02 Eximstats V1.44
1704 Use a glob alias rather than an array ref in the generated
1705 parser. This improves both readability and performance.
1707 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
1708 Collect SpamAssassin and rejection statistics.
1709 Don't display local sender or destination tables unless
1710 there is data to show.
1711 Added average volumes into the top table text output.
1713 SC/04 Eximstats V1.46
1714 Collect data on the number of addresses (recipients)
1715 as well as the number of messages.
1717 SC/05 Eximstats V1.47
1718 Added 'Message too big' to the list of mail rejection
1719 reasons (thanks to Marco Gaiarin).
1721 SC/06 Eximstats V1.48
1722 Mainlog lines which have GMT offsets and are too short to
1723 have a flag are now skipped.
1725 SC/07 Eximstats V1.49 (Alain Williams)
1726 Added the -emptyok flag.
1728 SC/08 Eximstats V1.50
1729 Fixes for obtaining the IP address from reject messages.
1731 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
1732 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
1733 whitesspace changes from 4.64-PH/27
1735 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
1738 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
1739 are found, allow negative numbers in numeric criteria)
1741 JJ/06 exipick.20061117.2, added new $message_body_missing variable
1743 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
1744 to match changes made in 4.64-PH/43
1746 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
1748 PH/47 Put in an explicit test for a DNS lookup of an address record where the
1749 "domain" is actually an IP address, and force a failure. This locks out
1750 those revolvers/nameservers that support "A-for-A" lookups, in
1751 contravention of the specifications.
1753 PH/48 When a host name was looked up from an IP address, and the subsequent
1754 forward lookup of the name timed out, the host name was left in
1755 $sender_host_name, contrary to the specification.
1757 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
1758 restricted to single-key lookups, Exim was not diagnosing an error if
1759 * or *@ was used with a query-style lookup.
1761 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
1763 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
1764 long ago, but noone interested enough thought of it. Let's just say that
1765 the "1.1" means that there are some new functions that weren't there at
1766 some point in the past.
1768 PH/51 Error processing for expansion failure of helo_data from an smtp
1769 transport during callout processing was broken.
1771 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
1772 tested/used via the -bh/-bhc/-bs options.
1774 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
1775 bug, fixed in subsequent PCRE releases).
1777 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
1778 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
1780 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
1786 SC/01 Use a glob alias rather than an array ref in eximstats generated
1787 parser. This improves both readability and performance.
1789 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
1790 Don't display local sender or destination tables in eximstats unless
1791 there is data to show.
1792 Added average volumes into the eximstats top table text output.
1794 SC/03 Collect data on the number of addresses (recipients) as well
1795 as the number of messages in eximstats.
1797 TF/01 Correct an error in the documentation for the redirect router. Exim
1798 does (usually) call initgroups() when daemonizing.
1800 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
1801 with consistent privilege compared to when running as a daemon.
1803 TF/03 Note in the spec that $authenticated_id is not set for local
1804 submissions from trusted users.
1806 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
1807 Thanks to Dean Brooks <dean@iglou.com> for the patch.
1809 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
1810 by adding some example configuration directives to the default
1811 configuration file. A little bit of work is required to uncomment the
1812 directives and define how usernames and passwords are checked, but
1813 there is now a framework to start from.
1815 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
1816 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
1817 without this. I don't know how relevant this is to other LDAP libraries.
1819 PH/02 Add the verb name to the "unknown ACL verb" error.
1821 PH/03 Magnus Holmgren's patch for filter_prepend_home.
1823 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
1825 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
1826 directory not expanded when it should be if an expanded home directory
1827 was set for the address (which is overridden by the transport).
1829 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
1832 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
1833 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
1834 because it is too late at that time, and has no effect.
1836 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
1837 security issue with \' (bugzilla #107). I could not use the
1838 PQescapeStringConn() function, because it needs a PGconn value as one of
1841 PH/08 When testing addresses using -bt, indicate those final addresses that
1842 are duplicates that would not cause an additional delivery. At least one
1843 person was confused, thinking that -bt output corresponded to deliveries.
1844 (Suppressing duplicates isn't a good idea as you lose the information
1845 about possibly different redirections that led to the duplicates.)
1847 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
1848 systems where poll() doesn't work, in particular OS X.
1850 PH/10 Added more information to debugging output for retry time not reached.
1852 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
1853 operations in malware.c.
1855 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
1858 PH/13 If write_rejectlog was set false when logging was sent to syslog with
1859 syslog_duplication set false, log lines that would normally be written
1860 both the the main log and to the reject log were not written to syslog at
1863 PH/14 In the default configuration, change the use of "message" in ACL warn
1864 statements to "add_header".
1866 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
1867 not followed by a command (e.g. "seen endif").
1869 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
1870 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
1873 PH/17 Added extra conditions to the default value of delay_warning_condition
1877 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
1878 { match{$h_precedence:}{(?i)bulk|list|junk} } \
1879 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
1882 The Auto-Submitted: and various List- headers are standardised, whereas I
1883 don't think Precedence: ever was.
1885 PH/18 Refactored debugging code in route_finduser() to show more information,
1886 in particular, the error code if getpwnam() issues one.
1888 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
1889 This is apparently needed in addition to the PH/07 change above to avoid
1890 any possible encoding problems.
1892 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
1893 but not after initializing Perl.
1895 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
1896 output them only if debugging. By default they are written stderr,
1897 apparently, which is not desirable.
1899 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
1902 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
1905 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
1907 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
1908 authenticated or an ident call has been made. Suppress the default
1909 values for $authenticated_id and $authenticated_sender (but permit -oMai
1910 and -oMas) when testing with -bh.
1912 PH/24 Re-jigged the order of the tests in the default configuration so that the
1913 tests for valid domains and recipients precede the DNS black list and CSA
1914 tests, on the grounds that those ones are more expensive.
1916 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
1917 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
1918 This bug exists in every version of Exim that I still have, right back to
1921 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
1922 However, an attempt to turn on case-sensitivity in a regex key by
1923 including (?-i) didn't work because the subject string was already
1924 lowercased, and the effects were non-intuitive. It turns out that a
1925 one-line patch can be used to allow (?-i) to work as expected.
1931 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
1932 other effects) broke the use of negated acl sub-conditions.
1934 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
1937 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
1938 "Deny" causes Exim to reject the incoming connection with a 554 error.
1939 Unfortunately, if there is a major crisis, such as a disk failure,
1940 tcp-wrappers gives "deny", whereas what one would like would be some
1941 kind of temporary error. A kludge has been added to help with this.
1942 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
1943 554 error is used if errno is still zero or contains ENOENT (which occurs
1944 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
1947 PH/03 Add -lutil to the default FreeBSD LIBS setting.
1949 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
1950 errors. Otherwise a message that provokes a temporary error (when other
1951 messages do not) can cause a whole host to time out.
1953 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
1954 addresses were routed directly to files or pipes from a redirect router.
1955 File deliveries just didn't batch; pipe deliveries might have suffered
1958 PH/06 A failure to get a lock for a hints database would erroneously always say
1959 "Failed to get write lock", even when it was really a read lock.
1961 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
1962 of 0600. This has been changed to use the value of the lockfile_mode
1963 option (which defaults to 0600).
1965 PH/08 Applied small patch from the Sieve maintainer.
1967 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
1968 folder from quota calculations, a direct delivery into this folder messed
1969 up the contents of the maildirsize file. This was because the regex was
1970 used only to exclude .Trash (or whatever) when the size of the mailbox
1971 was calculated. There was no check that a delivery was happening into an
1972 excluded directory. This bug has been fixed by ignoring all quota
1973 processing for deliveries into excluded directories.
1975 PH/10 Added the maildirfolder_create_regex option to appendfile.
1981 PH/01 The code for finding all the local interface addresses on a FreeBSD
1982 system running IPv6 was broken. This may well have applied to all BSD
1983 systems, as well as to others that have similar system calls. The broken
1984 code found IPv4 interfaces correctly, but gave incorrect values for the
1985 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
1986 that it would not match correctly against @[] and not recognize the IPv6
1989 PH/02 The ipliteral router was not recognizing addresses of the form user@
1990 [ipv6:....] because it didn't know about the "ipv6:" prefix.
1992 PH/03 Added disable_ipv6.
1994 PH/04 Changed $reply_address to use the raw form of the headers instead of the
1995 decoded form, because it is most often used to construct To: headers
1996 lines in autoreplies, and the decoded form may well be syntactically
1997 invalid. However, $reply_address has leading white space removed, and all
1998 newlines turned into spaces so that the autoreply transport does not
2001 PH/05 If group was specified without a user on a router, and no group or user
2002 was specified on a transport, the group from the router was ignored.
2004 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
2005 for visible compile-time settings that can be used to change these
2006 numbers, for those that want even more. Backwards compatibility with old
2007 spool files has been maintained. However, going back to a previous Exim
2008 release will lost any variables that are in spool files.
2010 PH/07 Two small changes when running in the test harness: increase delay when
2011 passing a TCP/IP connection to a new process, in case the original
2012 process has to generate a bounce, and remove special handling of
2013 127.0.0.2 (sic), which is no longer necessary.
2015 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
2016 be the same on different OS.
2018 PH/09 Moved a debug statement in filter processing to avoid a race problem when
2021 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
2022 whether --show-vars was specified or not
2024 JJ/02 exipick: Added support for new ACL variable spool format introduced
2027 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
2028 syntactically invalid From: or Reply-to: line, and a filter used this to
2029 generate an autoreply, and therefore failed to obtain an address for the
2030 autoreply, Exim could try to deliver to a non-existent relative file
2031 name, causing unrelated and misleading errors. What now happens is that
2032 it logs this as a hard delivery error, but does not attempt to create a
2035 PH/11 The exinext utility has a -C option for testing purposes, but although
2036 the given file was scanned by exinext itself; it wasn't being passed on
2037 when Exim was called.
2039 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
2040 an end-of-file indication when reading a command response.
2042 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
2043 compiled. In many other places in Exim, IPv6 addresses are always
2044 recognized, so I have changed this. It also means that IPv4 domain
2045 literals of the form [IPV4:n.n.n.n] are now always recognized.
2047 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
2048 used if the router is not running as root, for example, when verifying at
2049 ACL time, or when using -bh. The debugging output from this situation was
2050 non-existent - all you got was a failure to exec. I have made two
2053 (a) Failures to set uid/gid, the current directory, or a process leader
2054 in a subprocess such as that created by queryprogram now generate
2055 suitable debugging ouput when -d is set.
2057 (b) The queryprogram router detects when it is not running as root,
2058 outputs suitable debugging information if -d is set, and then runs
2059 the subprocess without attempting to change uid/gid.
2061 PH/15 Minor change to Makefile for building test_host (undocumented testing
2064 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
2065 additional section of a DNS packet that returns MX or SRV records.
2066 Instead, it always explicitly searches for A/AAAA records. This avoids
2067 major problems that occur when a DNS server includes only records of one
2068 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
2069 fixed another bug: if SRV records were looked up and the corresponding
2070 address records were *not* found in the additional section, the port
2071 values from the SRV records were lost.
2073 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
2074 using the correct key (the original address) when searching the retry
2075 rules in order to find which one to use for generating the retry hint.
2077 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
2078 adding the default one. Similarly, if it contains a Reply-To: header, the
2079 errors_reply_to option, if set, is not used.
2081 PH/19 When calculating a retry time, Exim used to measure the "time since
2082 failure" by looking at the "first failed" field in the retry record. Now
2083 it does not use this if it is later than than the arrival time of the
2084 message. Instead it uses the arrival time. This makes for better
2085 behaviour in cases where some deliveries succeed, thus re-setting the
2086 "first failed" field. An example is a quota failure for a huge message
2087 when small messages continue to be delivered. Without this change, the
2088 "time since failure" will always be short, possible causing more frequent
2089 delivery attempts for the huge message than are intended.
2090 [Note: This change was subsequently modified - see PH/04 for 4.62.]
2092 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
2093 $1, $2, $3) because the numerical variables can be reset during some
2094 expansion items (e.g. "match"), thereby losing the authentication data.
2096 PH/21 Make -bV show the size of off_t variables so that the test suite can
2097 decide whether to run tests for quotas > 2G.
2099 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
2100 mailbox_size, and mailbox_filecount in the appendfile transport. If a
2101 filecount value is greater than 2G or if a quota value is greater than 2G
2102 on a system where the size of off_t is not greater than 4, a panic error
2105 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
2106 never match. The debug and -bh output now contains an explicit error
2107 message indicating a malformed IPv4 address or mask.
2109 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
2110 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
2111 PH/23 above applies.
2113 PH/25 Do not write to syslog when running in the test harness. The only
2114 occasion when this arises is a failure to open the main or panic logs
2115 (for which there is an explicit test).
2117 PH/26 Added the /no_tell option to "control=freeze".
2119 PH/27 If a host name lookup failed very early in a connection, for example, if
2120 the IP address matched host_lookup and the reverse lookup yielded a name
2121 that did not have a forward lookup, an error message of the form "no IP
2122 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
2123 could be logged. Now it outputs the IP address instead of "NULL".
2125 PH/28 An enabling patch from MH: add new function child_open_exim2() which
2126 allows the sender and the authenticated sender to be set when
2127 submitting a message from within Exim. Since child_open_exim() is
2128 documented for local_scan(), the new function should be too.
2130 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
2131 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
2132 results in an empty string is now treated as unset.
2134 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
2136 PH/31 Added sender_verify_fail logging option.
2138 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
2139 needed by RFC 822 but not by RFC 2822 was commented out. I have now
2140 tidied the source and removed it altogether.
2142 PH/33 When a queue run was abandoned because the load average was too high, a
2143 log line was always written; now it is written only if the queue_run log
2144 selector is set. In addition, the log line for abandonment now contains
2145 information about the queue run such as the pid. This is always present
2146 in "start" and "stop" lines but was omitted from the "abandon" line.
2148 PH/34 Omit spaces between a header name and the colon in the error message that
2149 is given when verify = headers_syntax fails (if there are lots of them,
2150 the message gets confusing).
2152 PH/35 Change the default for dns_check_names_pattern to allow slashes within
2153 names, as there are now some PTR records that contain slashes. This check
2154 is only to protect against broken name servers that fall over on strange
2155 characters, so the fact that it applies to all lookups doesn't matter.
2157 PH/36 Now that the new test suite is complete, we can remove some of the
2158 special code in Exim that was needed for the old test suite. For example,
2159 sorting DNS records because real resolvers return them in an arbitrary
2160 order. The new test suite's fake resolver always returns records in the
2163 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
2164 bounces) except when queue_only is set, to avoid logging races between
2165 the different processes.
2167 PH/38 Panic-die if .include specifies a non-absolute path.
2169 PH/39 A tweak to the "H" retry rule from its user.
2171 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
2172 a label. They prevented compilation on older perls.
2174 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
2175 a warning to be raised on newish perls.
2177 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
2178 on queue. Changes to match documented behaviour of showing count of
2179 messages matching specified criteria.
2181 PH/40 Changed the default ident timeout from 30s to 5s.
2183 PH/41 Added support for the use of login_cap features, on those BSD systems
2184 that have them, for controlling the resources used by pipe deliveries.
2186 PH/42 The content-scanning code uses fopen() to create files in which to put
2187 message data. Previously it was not paying any attention to the mode of
2188 the files. Exim runs with umask(0) because the rest of the code creates
2189 files with open(), and sets the required mode explicitly. Thus, these
2190 files were ending up world-writeable. This was not a big issue, because,
2191 being within the spool directory, they were not world-accessible. I have
2192 created a function called modefopen, which takes an additional mode
2193 argument. It sets umask(777), creates the file, chmods it to the required
2194 mode, then resets the umask. All the relevant calls to fopen() in the
2195 content scanning code have been changed to use this function.
2197 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
2198 to 24 hours. This avoids potential overflow problems when processing G
2199 and H retry rules. I suspect nobody ever tinkers with this value.
2201 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
2203 PH/45 When the plaintext authenticator is running as a client, the server's
2204 challenges are checked to ensure they are valid base64 strings. By
2205 default, the authentication attempt is cancelled if an invalid string is
2206 received. Setting client_ignore_invalid_base64 true ignores these errors.
2207 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
2208 they are received. Thus, the responses can be made to depend on the
2209 challenges. If an invalid string is ignored, an empty string is placed in
2212 PH/46 Messages that are created by the autoreply transport now contains a
2213 References: header, in accordance with RFCs 2822 and 3834.
2215 PH/47 Added authenticated_sender_force to the smtp transport.
2217 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
2219 PH/49 Installed latest patch from the Sieve maintainer.
2221 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
2222 was also set, the appendfile transport was unnecessarily scanning a
2223 directory of message files (e.g. for maildir delivery) to find the count
2224 of files (along with the size), even though it did not need this
2225 information. It now does the scan only if it needs to find either the
2226 size of the count of files.
2228 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
2230 PH/52 Two bugs concerned with error handling when the smtp transport is
2233 (i) Exim was not creating retry information for temporary errors given
2234 for individual recipients after the DATA command when the smtp transport
2235 was used in LMTP mode. This meant that they could be retried too
2236 frequently, and not timed out correctly.
2238 (ii) Exim was setting the flag that allows error details to be returned
2239 for LMTP errors on RCPT commands, but not for LMTP errors for individual
2240 recipients that were returned after the DATA command.
2242 PH/53 This is related to PH/52, but is more general: for any failing address,
2243 when detailed error information was permitted to be returned to the
2244 sender, but the error was temporary, then after the final timeout, only
2245 "retry timeout exceeded" was returned. Now it returns the full error as
2246 well as "retry timeout exceeded".
2248 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
2249 do this, and (what is worse) MTAs that accept it.
2251 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
2252 will now be deprecated.
2254 PH/56 New os.c-cygwin from the Cygwin maintainer.
2256 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
2257 formats (previously only available in exim formats via -bpr, -bpru,
2258 and -bpra. Now also available in native and exiqgrep formats)
2260 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
2261 with very large, slow to parse queues
2263 JJ/08 exipick: added ! as generic prefix to negate any criteria format
2265 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
2267 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
2268 responses to authentication challenges, though it was showing the
2269 challenges; (ii) I've removed the CR characters from the debug output for
2272 PH/58 Allow for the insertion of a newline as well as a space when a string
2273 is turned into more than one encoded-word during RFC 2047 encoding. The
2274 Sieve code now uses this.
2276 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
2277 data_4xx, lost_connection, tls_required.
2279 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
2280 message was being sent as an SMTP response.
2282 PH/61 Add -l and -k options to exicyclog.
2284 PH/62 When verifying, if an address was redirected to one new address, so that
2285 verification continued, and the new address failed or deferred after
2286 having set something in $address_data, the value of $address_data was not
2287 passed back to the ACL. This was different to the case when no
2288 redirection occurred. The value is now passed back in both cases.
2290 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
2291 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
2292 use login_cap.h, so on its own it isn't the distinguishing feature. The
2293 new name refers directly to the setclassresources() function.
2295 PH/65 Added configuration files for NetBSD3.
2297 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
2299 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
2300 is preferred over IPv4.
2302 PH/68 The bounce_return_message and bounce_return_body options were not being
2303 honoured for bounces generated during the reception of non-SMTP messages.
2304 In particular, this applied to messages rejected by the ACL. This bug has
2305 been fixed. However, if bounce_return_message is true and bounce_return_
2306 body is false, the headers that are returned for a non-SMTP message
2307 include only those that have been read before the error was detected.
2308 (In the case of an ACL rejection, they have all been read.)
2310 PH/69 The HTML version of the specification is now built in a directory called
2311 spec_html instead of spec.html, because the latter looks like a path with
2312 a MIME-type, and this confuses some software.
2314 PH/70 Catch two compiler warnings in sieve.c.
2316 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
2317 function verify_get_ident() calls ip_connect() to connect a socket, but
2318 if the "connect()" function timed out, ip_connect() used to close the
2319 socket. However, verify_get_ident() also closes the socket later, and in
2320 between Exim writes to the log, which may get opened at this point. When
2321 the socket was closed in ip_connect(), the log could get the same file
2322 descriptor number as the socket. This naturally causes chaos. The fix is
2323 not to close the socket in ip_connect(); the socket should be closed by
2324 the function that creates it. There was only one place in the code where
2325 this was missing, in the iplookup router, which I don't think anybody now
2326 uses, but I've fixed it anyway.
2328 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
2329 well as to direct DNS lookups. Otherwise the handling of names in host
2330 lists is inconsistent and therefore confusing.
2336 PH/01 Two changes to the default runtime configuration:
2338 (1) Move the checks for relay_from_hosts and authenticated clients from
2339 after to before the (commented out) DNS black list checks.
2341 (2) Add control=submission to the relay_from_hosts and authenticated
2342 clients checks, on the grounds that messages accepted by these
2343 statements are most likely to be submissions.
2345 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
2347 (1) Generate an error if the third argument for the ${prvs expansion is
2350 (2) Treat a missing third argument of ${prvscheck as if it were an empty
2353 (3) Reset the variables that are obtained from the first argument of
2354 ${prvscheck and used in the second argument before leaving the code,
2355 because their memory is reclaimed, so using them afterwards may do
2358 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
2359 one (it's much easier than Tom thought :-).
2361 (5) Because of (4), we can now allow for the use of $prvscheck_result
2362 inside the third argument.
2364 PH/03 For some reason, the default setting of PATH when running a command from
2365 a pipe transport was just "/usr/bin". I have changed it to
2368 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
2369 anything to be listed in the output from -bV.
2371 PH/05 When a filter generated an autoreply, the entire To: header line was
2372 quoted in the delivery log line, like this:
2374 => >A.N.Other <ano@some.domain> <original@ddress> ...
2376 This has been changed so that it extracts the operative address. There
2377 may be more than one such address. If so, they are comma-separated, like
2380 => >ano@some.domain,ona@other.domain <original@ddress> ...
2382 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
2383 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
2384 not being looked up in the rDNS to get a host name, Exim was showing the
2385 IP address twice in Received: lines, even though the IP addresses were
2386 identical. For example:
2388 Received: from [1.2.3.4] (helo=[1.2.3.4])
2390 However, if the real host name was known, it was omitting the HELO data
2391 if it matched the actual IP address. This has been tidied up so that it
2392 doesn't show the same IP address twice.
2394 PH/07 When both +timestamp and +memory debugging was on, the value given by
2395 $tod_xxx expansions could be wrong, because the tod_stamp() function was
2396 called by the debug printing, thereby overwriting the timestamp buffer.
2397 Debugging no longer uses the tod_stamp() function when +timestamp is set.
2399 PH/08 When the original message was included in an autoreply transport, it
2400 always said "this is a copy of the message, including all the headers",
2401 even if body_only or headers_only was set. It now gives an appropriate
2404 PH/09 Applied a patch from the Sieve maintainer which:
2406 o fixes some comments
2407 o adds the (disabled) notify extension core
2408 o adds some debug output for the result of if/elsif tests
2409 o points to the current vacation draft in the documentation
2410 and documents the missing references header update
2414 o fixes a bug in processing the envelope test (when testing
2415 multiple envelope elements, the last element determinted the
2418 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
2419 Electronic Mail") by including:
2421 Auto-submitted: auto-generated
2423 in the messages that it generates (bounce messages and others, such as
2424 warnings). In the case of bounce messages for non-SMTP mesages, there was
2425 also a typo: it was using "Auto_submitted" (underscore instead of
2426 hyphen). Since every message generated by Exim is necessarily in response
2427 to another message, thes have all been changed to:
2429 Auto-Submitted: auto-replied
2431 in accordance with these statements in the RFC:
2433 The auto-replied keyword:
2435 - SHOULD be used on messages sent in direct response to another
2436 message by an automatic process,
2438 - MUST NOT be used on manually-generated messages,
2440 - MAY be used on Delivery Status Notifications (DSNs) and Message
2441 Disposition Notifications (MDNs),
2443 - MUST NOT be used on messages generated by automatic or periodic
2444 processes, except for messages which are automatic responses to
2447 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
2448 to the default Received: header definition.
2450 PH/12 Added log selector acl_warn_skipped (default on).
2452 PH/13 After a successful wildlsearch lookup, discard the values of numeric
2453 variables because (a) they are in the wrong storage pool and (b) even if
2454 they were copied, it wouldn't work properly because of the caching.
2456 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
2457 checking when decoding. Apparently there are clients that generate
2458 overlong encoded strings. Why am I not surprised?
2460 PH/15 If the first argument of "${if match_address" was not empty, but did not
2461 contain an "@" character, Exim crashed. Now it writes a panic log message
2462 and treats the condition as false.
2464 PH/16 In autoreply, treat an empty string for "once" the same as unset.
2466 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
2467 extension "envelope-auth". The code is finished and in agreement with
2468 other implementations, but there is no documentation so far and in fact,
2469 nobody wrote the draft yet. This extension is currently #undef'ed, thus
2470 not changing the active code.
2472 Print executed "if" and "elsif" statements when debugging is used. This
2473 helps a great deal to understand what a filter does.
2475 Document more things not specified clearly in RFC3028. I had all this
2476 sorted out, when out of a sudden new issues came to my mind. Oops."
2478 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
2481 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
2482 canonical form (as documented). However, after a host name lookup from
2483 the IP address, check_host() was doing a simple string comparison with
2484 addresses acquired from the DNS when checking that the found name did
2485 have the original IP as one of its addresses. Since any found IPv6
2486 addresses are likely to be in abbreviated form, the comparison could
2487 fail. Luckily, there already exists a function for doing the comparison
2488 by converting both addresses to binary, so now that is used instead of
2489 the text comparison.
2491 PH/20 There was another similar case to PH/19, when a complete host name was
2492 given in a host list; looking up its IP address could give an abbreviated
2493 form, whereas the current host's name might or might not be abbreviated.
2494 The same fix has been applied.
2500 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
2501 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
2504 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
2505 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
2507 PH/03 Typo: missing ".o" in src/pcre/Makefile.
2509 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
2510 header line, restrict the check to what is listed in RFCs 2369 and 2929.
2511 Also, for "Auto-Submitted", treat anything other than "no" as
2512 non-personal, in accordance with RFC 3834. (Previously it treated
2513 anything starting "auto-" as non-personal.)
2515 TF/01 The control=submission/name=... option had a problem with syntax
2516 errors if the name included a slash character. The /name= option
2517 now slurps the rest of the string, so it can include any characters
2518 but it must come last in the list of options (after /sender_retain
2521 PH/05 Some modifications to the interface to the fake nameserver for the new
2529 TK/01 Added the "success_on_redirect" address verification option. See
2530 NewStuff for rationale and an example.
2532 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
2534 PH/02 Patch to exigrep to allow it to work on syslog lines.
2536 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
2537 fread() to skip over the body file's header line, because in Cygwin the
2538 header line is locked and is inaccessible.
2540 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
2541 co-exist for some time) to make it clear that it is the Exim ID that is
2542 referenced, not the Message-ID: header line.
2544 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
2545 string_format() function, because snprintf() does not exist on all
2548 PH/06 The use of forbid_filter_existstest now also locks out the use of the
2549 ${stat: expansion item.
2551 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
2552 protocol synchronization error", to keep the pedants happy.
2554 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
2555 well as for IRIX systems, when gcc is being used. See the host.c source
2558 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
2560 PH/10 Named domain lists were not working if used in a queue_smtp_domains
2563 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
2564 transport and to the smtp transport in LMTP mode.
2566 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
2568 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
2569 run a filter in a subprocess. This could lead to confusion in subsequent
2570 lookups in the parent process. There should also be a search_tidyup() at
2571 the end of the subprocess.
2573 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
2574 only if the host matched helo_try_verify_hosts, which caused the
2575 verification to occur when the EHLO/HELO command was issued. The ACL just
2576 tested the remembered result. Now, if a previous verification attempt has
2577 not happened, "verify = helo" does it there and then.
2579 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
2581 TK/03 Fix log output including CR from clamd.
2583 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
2584 exist provoked a memory error which could cause a segfault.
2586 PH/15 Installed PCRE 6.2
2588 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
2590 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
2591 of the problem. Specifically, suggested +O2 rather than +O1 for the
2594 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
2596 PH/20 If a delivery was routed to a non-standard port by means of an SRV
2597 record, the port was not correctly logged when the outgoing_port log
2598 selector was set (it logged the transort's default port).
2600 PH/21 Added support for host-specific ports to manualroute, queryprogram,
2601 fallback_hosts, and "hosts" in the smtp transport.
2603 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
2604 host errors such as "Connection refused".
2606 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
2607 authentication with radiusclient 0.4.9:
2609 - Error returned from rc_read_config was caught wrongly
2610 - Username/password not passed on to radius server due to wrong length.
2612 The presumption is that some radiusclient API changes for 4.51/PH/17
2613 were not taken care of correctly. The code is still untested by me (my
2614 Linux distribution still has 0.3.2 of radiusclient), but it was
2615 contributed by a Radius user.
2617 PH/24 When doing a callout, the value of $domain wasn't set correctly when
2618 expanding the "port" option of the smtp transport.
2620 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
2621 while reading a MIME header. Thanks to Tom Hughes for a patch.
2623 PH/24 Include config.h inside local_scan.h so that configuration settings are
2626 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
2627 that in an ACL for RCPT (for example), you can examine exactly what was
2630 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
2631 commands, but it was not correctly comparing the address with the actual
2632 client host address. Thus, it would show the EHLO address in Received:
2633 header lines when this was not necessary.
2635 PH/27 Added the % operator to ${eval:}.
2637 PH/28 Exim tries to create and chdir to its spool directory when it starts;
2638 it should be ignoring failures (because with -C, for example, it has lost
2639 privilege). It wasn't ignoring creation failures other than "already
2642 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
2645 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
2646 because an input line was too long, either on its own, or by virtue of
2647 too many continuations, the temporary file was not being removed, and the
2648 return code was incorrect.
2650 PH/31 Missing "BOOL" in function definition in filtertest.c.
2652 PH/32 Applied Sieve patches from the maintainer.
2654 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
2656 PH/33 Added "verify = not_blind".
2658 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
2659 Local/Makefile (with some defaults set). These are used in built scripts
2660 such as exicyclog, but they have never been used in the exim_install
2661 script (though there are many overriding facilities there). I have
2662 arranged that the exim_install script now takes note of these two
2665 PH/35 Installed configuration files for Dragonfly.
2667 PH/36 When a locally submitted message by a trusted user did not contain a
2668 From: header, and the sender address was obtained from -f or from an SMTP
2669 MAIL command, and the trusted user did not use -F to supply a sender
2670 name, $originator_name was incorrectly used when constructing a From:
2671 header. Furthermore, $originator_name was used for submission mode
2672 messages from external hosts without From: headers in a similar way,
2673 which is clearly wrong.
2675 PH/37 Added control=suppress_local_fixups.
2677 PH/38 When log_selector = +received_sender was set, and the addition of the
2678 sender made the log line's construction buffer exactly full, or one byte
2679 less than full, an overflow happened when the terminating "\n" was
2682 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
2683 when the result of a list match is failure because a DNS lookup failed.
2685 PH/40 RM_COMMAND is now used in the building process.
2687 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
2688 the "build-* directories that it finds.
2690 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
2691 address in a domain literal was a prefix of an interface address.
2693 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
2694 when verifying a sender address, unless rewrite_headers is false.
2696 PH/44 Wrote a long comment about why errors_to addresses are verified as
2697 recipients, not senders.
2699 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
2700 the ratelimit ACL was added.
2702 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
2704 PH/46 Added extra information about PostgreSQL errors to the error string.
2706 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
2707 avoiding the need to install special zones in a real server. This is
2708 backwards compatible; if it can't find the fake resolver, it drops back.
2709 Thus, both old and new test suites can be run.
2711 TF/02 Added util/ratelimit.pl
2713 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
2714 clock is set back in time.
2716 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
2717 Candler <B.Candler@pobox.com>.
2719 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
2720 OK for addresses that are the result of redirections.
2722 PH/48 A number of further additions for the benefit of the new test suite,
2723 including a fake gethostbyname() that interfaces to the fake DNS resolver
2726 TF/06 The fix for widen_domains has also been applied to qualify_single and
2727 search_parents which are the other dnslookup options that can cause
2730 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
2733 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
2735 TF/07 Exim produced the error message "an SRV record indicated no SMTP
2736 service" if it encountered an MX record with an empty target hostname.
2737 The message is now "an MX or SRV record indicated no SMTP service".
2739 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
2740 if the DNS of the sending site is misconfigured. This is quite a
2741 common situation. This change restores the behaviour of treating a
2742 helo verification defer as a failure.
2744 PH/51 If self=fail was set on a router, the bounce message did not include the
2745 actual error message.
2751 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
2753 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
2754 command itself ended in error, the underlying message about the transport
2755 filter timeout was being overwritten with the pipe command error. Now the
2756 underlying error message should be appended to the second error message.
2758 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
2760 PH/02 Reduce the amount of output that "make" produces by default. Full output
2761 can still be requested.
2763 PH/03 The warning log line about a condition test deferring for a "warn" verb
2764 was being output only once per connection, rather than after each
2765 occurrence (because it was using the same function as for successful
2766 "warn" verbs). This seems wrong, so I have changed it.
2768 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
2769 it should not have, which might have caused a crash in the right
2770 circumstances, but probably never did.
2772 PH/04 Installed a modified version of Tony Finch's patch to make submission
2773 mode fix the return path as well as the Sender: header line, and to
2774 add a /name= option so that you can make the user's friendly name appear
2777 TF/03 Added the control = fakedefer ACL modifier.
2779 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
2780 Mark Lowes for thorough testing.
2782 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
2784 TK/03 Merged latest SRS patch from Miles Wilton.
2786 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
2787 with the definition in sysexits.h (which is #included earlier).
2788 Fortunately, Exim does not actually use EX_OK. The code used to try to
2789 preserve the sysexits.h value, by assumimg that macro definitions were
2790 scanned for macro replacements. I have been disabused of this notion,
2791 so now the code just undefines EX_OK before #including unistd.h.
2793 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
2794 in the smtp transport. When a block could not be written in a single
2795 write() function, the timeout was being re-applied to each part-write.
2796 This seems wrong - if the receiver was accepting one byte at a time it
2797 would take for ever. The timeout is now adjusted when this happens. It
2798 doesn't have to be particularly precise.
2800 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
2801 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
2803 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
2804 without a domain if the check to <postmaster@domain> fails.
2806 SC/01 Eximstats: added -xls and the ability to specify output files
2807 (patch written by Frank Heydlauf).
2809 SC/02 Eximstats: use FileHandles for outputing results.
2811 SC/03 Eximstats: allow any combination of xls, txt, and html output.
2813 SC/04 Eximstats: fixed display of large numbers with -nvr option
2815 SC/05 Eximstats: fixed merging of reports with empty tables.
2817 SC/06 Eximstats: added the -include_original_destination flag
2819 SC/07 Eximstats: removed tabs and trailing whitespace.
2821 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
2823 TK/06 MBOX spool code: Add real "From " MBOX separator line
2824 so the .eml file is really in mbox format (even though
2825 most programs do not really care). Patch from Alex Miller.
2827 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
2828 The latter is generated from $received_to and is only set if the
2829 message has one envelope recipient. SA can use these headers,
2830 obviously out-of-the-box. Patch from Alex Miller.
2832 PH/08 The ${def test on a variable was returning false if the variable's
2833 value was "0", contrary to what the specification has always said!
2834 The result should be true unless the variable is empty.
2836 PH/09 The syntax error of a character other than { following "${if
2837 def:variable_name" (after optional whitespace) was not being diagnosed.
2838 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
2839 accidental colon was present, for example, could give incorrect results.
2841 PH/10 Tidied the code in a number of places where the st_size field of a stat()
2842 result is used (not including appendfile, where other changes are about
2845 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
2846 This involved changing a lot of size variables from int to off_t. It
2847 should work with maildirs and everything.
2849 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
2850 spamd dying while we are connected to it.
2852 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
2855 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
2856 function for BDB 4.3.
2858 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
2860 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
2861 the sources for PCRE have been re-arranged and more files are now
2864 PH/15 The code I had for printing potentially long long variables in PH/11
2865 above was not the best (it lost precision). The length of off_t variables
2866 is now inspected at build time, and an appropriate printing format (%ld
2867 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
2868 to be "long long int" or "long int". This is needed for the internal
2869 formatting function string_vformat().
2871 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
2872 the configuration file to be ":syslog", then the script "guesses" where
2873 the logs files are, rather than using the compiled in default. In our
2874 case the guess is not the same as the compiled default, so the script
2875 suddenly stopped working when I started to use syslog. The patch checks
2876 to see if log_file_path is "". If so, it attempts to read it from exim
2877 with no configuration file to get the compiled in version, before it
2878 falls back to the previous guessing code."
2880 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
2881 implementing BATV in an Exim configuration. See NewStuff for the gory
2884 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
2885 Makefile that are specific to HP-UX.
2887 PH/18 If the "use_postmaster" option was set for a recipient callout together
2888 with the "random" option, the postmaster address was used as the MAIL
2889 FROM address for the random test, but not for the subsequent recipient
2890 test. It is now used for both.
2892 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
2893 patch removes a few documentation additions to RFC 3028, because the
2894 latest draft now contains them. It adds the new en;ascii-case comparator
2895 and a new error check for 8bit text in MIME parts. Comparator and
2896 require names are now matched exactly. I enabled the subaddress
2897 extension, but it is not well tested yet (read: it works for me)."
2899 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
2900 rework some of the code of TK/09 above to avoid the hardwired use of
2901 "%lld" and "long long". Replaced the call to snprintf() with a call to
2904 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
2905 records point to non-existent hosts", "retry timeout exceeded", and
2906 "retry time not reached for any host after a long failure period".
2908 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
2909 experimental DomainKeys support:
2911 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
2912 (2) On an error such as an illegally used "control", the wrong name for
2913 the control was given.
2915 These problems did NOT occur unless DomainKeys support was compiled.
2917 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
2919 PH/24 Added ${if match_ip condition.
2921 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
2922 databases so that it will be absolutely obvious if a crash occurs in the
2923 DB library. This is a regular occurrence (often caused by mis-matched
2926 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
2927 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
2928 picked up on a user's system that detects such things. There doesn't seem
2929 to be a gcc warning option for this - only an attribute that has to be
2930 put on the function's prototype. It seems that in Fedora Core 4 they have
2931 set this on a number of new functions. No doubt there will be more in due
2934 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
2935 specify a transport. However, if an address that was verified by such a
2936 router was the subject of a callout, Exim crashed because it tried to
2937 read the rcpt_include_affixes from the non-existent transport. Now it
2938 just assumes that the setting of that option is false. This bug was
2939 introduced by 4.51/PH/31.
2941 PH/28 Changed -d+all to exclude +memory, because that information is very
2942 rarely of interest, but it makes the output a lot bigger. People tend to
2943 do -d+all out of habit.
2945 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
2946 code in os-type was giving problems when libc.so lives in lib64, like on
2949 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
2950 aren't the modern standard, and it seems that some systems' include files
2951 don't always have them. Exim was already checking for some of the newer
2952 ones like T_AAAA, and defining it itself. I've added checks for all the
2953 record types that Exim uses.
2955 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
2956 not automatically generating a new one, as it is supposed to. This
2957 prevented TLS from working. If the file did exist, but contained invalid
2958 data, a new version was generated, as expected. It was only the case of a
2959 non-existent file that was broken.
2961 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
2962 with a change in libdomainkeys > 0.64.
2964 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
2965 from DNS. If the selector record carries the flag, it now has
2966 precedence over the domain-wide flag.
2968 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
2970 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
2971 the use of a port name that isn't defined in /etc/services) occurred, the
2972 message was deferred as in a normal delivery, and thus remained on the
2973 spool, instead of being failed because of the mua_wrapper setting. This
2974 is now fixed, and I tidied up some of the mua_wrapper messages at the
2977 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
2978 the messages in a hash of arrays rather than using individual hashes.
2979 This is a bit cleaner and results in dramatic memory savings, albeit
2980 at a slight CPU cost.
2982 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
2983 as requested by Marc Sherman.
2985 SC/10 Eximstats: added histograms for user specified patterns as requested
2988 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
2990 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
2991 fopen() in the content-scanning modules that did not already have it.
2997 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
2998 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
3000 TK/02 Fix ACL "control" statement not being available in MIME ACL.
3002 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
3004 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
3005 to test Sieve filters that use "vacation".
3007 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
3008 that changes the way the GnuTLS parameters are stored in the cache file.
3009 The new format can be generated externally. For backward compatibility,
3010 if the data in the cache doesn't make sense, Exim assumes it has read an
3011 old-format file, and it generates new data and writes a new file. This
3012 means that you can't go back to an older release without removing the
3015 PH/03 A redirect router that has both "unseen" and "one_time" set does not
3016 work if there are any delivery delays because "one_time" forces the
3017 parent to be marked "delivered", so its unseen clone is never tried
3018 again. For this reason, Exim now forbids the simultaneous setting of
3021 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
3022 redirected to themselves ("homonym" addresses). Read the long ChangeLog
3023 entry if you want to know the details. The fix, however, neglected to
3024 consider the case when local delivery batching is involved. The test for
3025 "previously delivered" was not happening when checking to see if an
3026 address could be batched with a previous (undelivered) one; under
3027 certain circumstances this could lead to multiple deliveries to the same
3030 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
3031 in its include files, and this causes problems building Exim.
3033 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
3034 header_syntax) but Exim was just ignoring anything given after a slash.
3035 In particular, this caused confusion with an attempt to use "verify =
3036 reverse_host_lookup/defer_ok". An error is now given when options are
3037 supplied for verify items that do not have them. (Maybe reverse_host_
3038 lookup should have a defer_ok option, but that's a different point.)
3040 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
3041 defined by RFC 821) to 2048, because there were problems with some AUTH
3042 commands, and RFC 1869 says the size should be increased for extended
3043 SMTP commands that take arguments.
3045 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
3048 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
3049 "unknown" error; now it says that the functionality isn't in the binary.
3051 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
3052 an address' error message when a string expansion fails (syntax or
3053 whatever). Otherwise the password may appear in the log. Following change
3054 PH/42 below, there is no longer a chance of it appearing in a bounce
3057 PH/11 Installed exipick version 20050225.0 from John Jetmore.
3059 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
3060 of its addresses was ever tried. (Bugzilla bug #2.)
3062 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
3063 the result incorrectly in the debug output. (It correctly added a newline
3064 to what was transported.)
3066 TF/01 Added $received_time.
3068 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
3069 commented out examples of how to interface to a virus scanner and to
3070 SpamAssassin. Also added commented examples of av_scanner and
3071 spamd_address settings.
3073 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
3074 and controls are allowed in which ACLs. There were a couple of minor
3075 errors. Some of the entries in the conditions table (which is a table of
3076 where they are NOT allowed) were getting very unwieldy; rewrote them as a
3077 negation of where the condition IS allowed.
3079 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
3081 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
3082 header file does not have a version number, so I've had to invent a new
3083 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
3084 API. The code is untested by me (my Linux distribution still has 0.3.2 of
3085 radiusclient), but it was contributed by a Radius user.
3087 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
3088 files or directories, for OpenSSL.
3090 PH/19 When an Exim process that is running as root has to create an Exim log
3091 file, it does so in a subprocess that runs as exim:exim so as to get the
3092 ownership right at creation (otherwise, other Exim processes might see
3093 the file with the wrong ownership). There was no test for failure of this
3094 fork() call, which would lead to the process getting stuck as it waited
3095 for a non-existent subprocess. Forks do occasionally fail when resources
3096 run out. I reviewed all the other calls to fork(); they all seem to check
3099 PH/20 When checking for unexpected SMTP input at connect time (before writing
3100 the banner), Exim was not dealing correctly with a non-positive return
3101 from the read() function. If the client had disconnected by this time,
3102 the result was a log entry for a synchronization error with an empty
3103 string after "input=" when read() returned zero. If read() returned -1
3104 (an event I could not check), uninitialized data bytes were printed.
3105 There were reports of junk text (parts of files, etc) appearing after
3108 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
3110 PH/22 Added support for macro redefinition, and (re)definition in between
3111 driver and ACL definitions.
3113 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
3114 forgetting to use the resulting value; it was using the unexpanded value.
3116 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
3117 hadn't been configured. The fix is from Juergen Kreileder, who
3118 understands it better than I do:
3120 "Here's what I see happening with three configured cyrus_sasl
3121 authenticators configured (plain, login, cram-md5):
3123 On startup auth_cyrus_sasl_init() gets called for each of these.
3124 This means three calls to sasl_listmech() without a specified mech_list.
3125 => SASL tests which mechs of all available mechs actually work
3126 => three warnings about OTP not working
3127 => the returned list contains: plain, login, cram-md5, digest-md5, ...
3129 With the patch, sasl_listmech() also gets called three times. But now
3130 SASL's mech_list option is set to the server_mech specified in the the
3131 authenticator. Or in other words, the answer from sasl_listmech()
3132 gets limited to just the mech you're testing for (which is different
3134 => the return list contains just 'plain' or 'login', 'cram-md5' or
3135 nothing depending on the value of ob->server_mech.
3137 I've just tested the patch: Authentication still works fine,
3138 unavailable mechs specified in the exim configuration are still
3139 caught, and the auth.log warnings about OTP are gone."
3141 PH/25 When debugging is enabled, the contents of the command line are added
3142 to the debugging output, even when log_selector=+arguments is not
3145 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
3146 answer is "GNU", and only if the return is "GNU/something" is the answer
3149 PH/27 $acl_verify_message is now set immediately after the failure of a
3150 verification in an ACL, and so is available in subsequent modifiers. In
3151 particular, the message can be preserved by coding like this:
3153 warn !verify = sender
3154 set acl_m0 = $acl_verify_message
3156 Previously, $acl_verify_message was set only while expanding "message"
3157 and "log_message" when a very denied access.
3159 PH/28 Modified OS/os.c-Linux with
3161 -#ifndef OS_LOAD_AVERAGE
3162 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
3164 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
3165 nomenclature these days.)
3167 PH/29 Installed patch from the Sieve maintainer that adds the options
3168 sieve_useraddress and sieve_subaddress to the redirect router.
3170 PH/30 In these circumstances:
3171 . Two addresses routed to the same list of hosts;
3172 . First host does not offer TLS;
3173 . First host accepts first address;
3174 . First host gives temporary error to second address;
3175 . Second host offers TLS and a TLS session is established;
3176 . Second host accepts second address.
3177 Exim incorrectly logged both deliveries with the TLS parameters (cipher
3178 and peerdn, if requested) that were in fact used only for the second
3181 PH/31 When doing a callout as part of verifying an address, Exim was not paying
3182 attention to any local part prefix or suffix that was matched by the
3183 router that accepted the address. It now behaves in the same way as it
3184 does for delivery: the affixes are removed from the local part unless
3185 rcpt_include_affixes is set on the transport.
3187 PH/32 Add the sender address, as F=<...>, to the log line when logging a
3188 timeout during the DATA phase of an incoming message.
3190 PH/33 Sieve envelope tests were broken for match types other than :is. I have
3191 applied a patch sanctioned by the Sieve maintainer.
3193 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
3194 the uid or gid is negative. A case of a negative gid caused this to be
3195 noticed. The fix allows for either to be negative.
3197 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
3198 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
3201 PH/36 The change PH/12 above was broken. Fixed it.
3203 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
3204 the grounds that routing the same address twice would always produce the
3205 same answer. This might have been true once, but it is certainly no
3206 longer true now. Routing a child address may depend on the previous
3207 routing that produced that child. Some complicated redirection strategies
3208 went wrong when messages had multiple recipients, and made Exim's
3209 behaviour dependent on the order in which the addresses were given.
3211 I have moved the duplicate checking until after the routing is complete.
3212 Exim scans the addresses that are assigned to local and remote
3213 transports, and removes any duplicates. This means that more work will be
3214 done, as duplicates will always all be routed, but duplicates are
3215 presumably rare, so I don't expect this is of any significance.
3217 For deliveries to pipes, files, and autoreplies, the duplicate checking
3218 still happens during the routing process, since they are not going to be
3221 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
3222 It corrects a timeout issue with spamd. This is Ian's comment: "The
3223 background is that sometimes spamd either never reads data from a
3224 connection it has accepted, or it never writes response data. The exiscan
3225 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
3226 blindly assumes that writes won't block so it may never time out."
3228 PH/39 Allow G after quota size as well as K and M.
3230 PH/40 The value set for $authenticated_id in an authenticator may not contain
3231 binary zeroes or newlines because the value is written to log lines and
3232 to spool files. There was no check on this. Now the value is run through
3233 the string_printing() function so that such characters are converted to
3234 printable escape sequences.
3236 PH/41 $message_linecount is a new variable that contains the total number of
3237 lines in the message. Compare $body_linecount, which is the count for the
3240 PH/42 Exim no longer gives details of delivery errors for specific addresses in
3241 bounce and delay warning messages, except in certain special cases, which
3244 (a) An SMTP error message from a remote host;
3245 (b) A message specified in a :fail: redirection;
3246 (c) A message specified in a "fail" command in a system filter;
3247 (d) A message specified in a FAIL return from the queryprogram router;
3248 (e) A message specified by the cannot_route_message router option.
3250 In these cases only, Exim does include the error details in bounce and
3251 warning messages. There are also a few cases where bland messages such
3252 as "unrouteable address" or "local delivery error" are given.
3254 PH/43 $value is now also set for the "else" part of a ${run expansion.
3256 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
3257 being worked on, but at least Exim now implements the latest version to
3260 PH/45 In a pipe transport, although a timeout while waiting for the pipe
3261 process to complete was treated as a delivery failure, a timeout while
3262 writing the message to the pipe was logged, but erroneously treated as a
3263 successful delivery. Such timeouts include transport filter timeouts. For
3264 consistency with the overall process timeout, these timeouts are now
3265 treated as errors, giving rise to delivery failures by default. However,
3266 there is now a new Boolean option for the pipe transport called
3267 timeout_defer, which, if set TRUE, converts the failures into defers for
3268 both kinds of timeout. A transport filter timeout is now identified in
3271 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
3272 systems where "make" and "gmake" are different, calling "gmake" at top
3273 level broke things. I've arranged for the value of $(MAKE) to be passed
3274 from the Makefile to this script so that it can call the same version of
3278 A note about Exim versions 4.44 and 4.50
3279 ----------------------------------------
3281 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
3282 changes of various kinds. As a consequence, a big documentation update was
3283 needed. This delayed the release for rather longer than seemed good, especially
3284 in the light of a couple of (minor) security issues. Therefore, the changes
3285 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
3286 release. So 4.44 and 4.50 are in effect two different branches that both start
3289 I have left the 4.50 change log unchanged; it contains all the changes since
3290 4.43. The change log for 4.44 is below; many of its items are identical to
3291 those for 4.50. This seems to be the most sensible way to preserve the
3292 historical information.
3298 1. Minor wording change to the doc/README.SIEVE file.
3300 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
3301 computation of the current number of files was incorrect.
3303 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
3304 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
3307 4. Give more explanation in the error message when the command for a transport
3308 filter fails to execute.
3310 5. There are several places where Exim runs a non-Exim command in a
3311 subprocess. The SIGUSR1 signal should be disabled for these processes. This
3312 was being done only for the command run by the queryprogram router. It is
3313 now done for all such subprocesses. The other cases are: ${run, transport
3314 filters, and the commands run by the lmtp and pipe transports.
3316 6. Added CONFIGURE_GROUP build-time option.
3318 7. Some older OS have a limit of 256 on the maximum number of file
3319 descriptors. Exim was using setrlimit() to set 1000 as a large value
3320 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
3321 systems. I've change it so that if it can't get 1000, it tries for 256.
3323 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
3324 was an oversight, and furthermore, ever since the addition of extra
3325 controls (e.g. 4.43/32), the checks on when to allow different forms of
3326 "control" were broken. There should now be diagnostics for all cases when a
3327 control that does not make sense is encountered.
3329 9. Added the /retain_sender option to "control=submission".
3331 10. $recipients is now available in the predata ACL (oversight).
3333 11. Tidy the search cache before the fork to do a delivery from a message
3334 received from the command line. Otherwise the child will trigger a lookup
3335 failure and thereby defer the delivery if it tries to use (for example) a
3336 cached ldap connection that the parent has called unbind on.
3338 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
3339 of $address_data from the recipient verification was clobbered by the
3340 sender verification.
3342 13. The value of address_data from a sender verification is now available in
3343 $sender_address_data in subsequent conditions in the ACL statement.
3345 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
3347 15. Added a new option "connect=<time>" to callout options, to set a different
3350 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
3351 was its contents. (It was OK if the option was not defined at all.)
3353 17. A "Completed" log line is now written for messages that are removed from
3354 the spool by the -Mrm option.
3356 18. New variables $sender_verify_failure and $recipient_verify_failure contain
3357 information about exactly what failed.
3359 19. Added -dd to debug only the daemon process.
3361 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
3362 handles timeouts, both on the server side and network timeouts. Renamed the
3363 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
3365 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
3366 It is now set to "smtps".
3368 22. $host_address is now set to the target address during the checking of
3369 ignore_target_hosts.
3371 23. When checking ignore_target_hosts for an ipliteral router, no host name was
3372 being passed; this would have caused $sender_host_name to have been used if
3373 matching the list had actually called for a host name (not very likely,
3374 since this list is usually IP addresses). A host name is now passed as
3377 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
3378 code that specifies a non-restarting handler (typically sigaction() in
3379 modern systems) in an attempt to fix a rare and obscure crash bug.
3381 25. Narrowed the window for a race in the daemon that could cause it to ignore
3382 SIGCHLD signals. This is not a major problem, because they are used only to
3383 wake it up if nothing else does.
3385 26. A malformed maildirsize file could cause Exim to calculate negative values
3386 for the mailbox size or file count. Odd effects could occur as a result.
3387 The maildirsize information is now recalculated if the size or filecount
3390 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
3391 support for a long time. Removed HAVE_SYS_VFS_H.
3393 28. Installed the latest version of exipick from John Jetmore.
3395 29. In an address list, if the pattern was not a regular expression, an empty
3396 subject address (from a bounce message) matched only if the pattern was an
3397 empty string. Non-empty patterns were not even tested. This was the wrong
3398 because it is perfectly reasonable to use an empty address as part of a
3399 database query. An empty address is now tested by patterns that are
3400 lookups. However, all the other forms of pattern expect the subject to
3401 contain a local part and a domain, and therefore, for them, an empty
3402 address still always fails if the pattern is not itself empty.
3404 30. Exim went into a mad DNS loop when attempting to do a callout where the
3405 host was specified on an smtp transport, and looking it up yielded more
3406 than one IP address.
3408 31. Re-factored the code for checking spool and log partition space into a
3409 function that finds that data and another that does the check. The former
3410 is then used to implement four new variables: $spool_space, $log_space,
3411 $spool_inodes, and $log_inodes.
3413 32. The RFC2047 encoding function was originally intended for short strings
3414 such as real names; it was not keeping to the 75-character limit for
3415 encoded words that the RFC imposes. It now respects the limit, and
3416 generates multiple encoded words if necessary. To be on the safe side, I
3417 have increased the buffer size for the ${rfc2047: expansion operator from
3420 33. It is now permitted to omit both strings after an "if" condition; if the
3421 condition is true, the result is "true". As before, when the second string
3422 is omitted, a false condition yields an empty string. This makes it less
3423 cumbersome to write custom ACL and router conditions.
3425 34. Failure to deliver a bounce message always caused it to be frozen, even if
3426 there was an errors_to setting on the router. The errors_to setting is now
3429 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
3430 canonical form (fully expanded) before being placed in
3431 $sender_host_address.
3433 36. The table in the code that translates DNS record types into text (T_A to
3434 "A" for instance) was missing entries for NS and CNAME. It is just possible
3435 that this could have caused confusion if both these types were looked up
3436 for the same domain, because the text type is used as part of Exim's
3437 per-process caching. But the chance of anyone hitting this buglet seems
3440 37. The dnsdb lookup has been extended in a number of ways.
3442 (1) There is a new type, "zns", which walks up the domain tree until it
3443 finds some nameserver records. It should be used with care.
3445 (2) There is a new type, "mxh", which is like "mx" except that it returns
3446 just the host names, not the priorities.
3448 (3) It is now possible to give a list of domains (or IP addresses) to be
3449 looked up. The behaviour when one of the lookups defers can be
3450 controlled by a keyword.
3452 (4) It is now possible to specify the separator character for use when
3453 multiple records are returned.
3455 38. The dnslists ACL condition has been extended: it is now possible to supply
3456 a list of IP addresses and/or domains to be looked up in a particular DNS
3459 39. Added log_selector=+queue_time_overall.
3461 40. When running the queue in the test harness, wait just a tad after forking a
3462 delivery process, to get repeatability of debugging output.
3464 41. Include certificate and key file names in error message when GnuTLS fails
3465 to set them up, because the GnuTLS error message doesn't include the name
3466 of the failing file when there is a problem reading it.
3468 42. Allow both -bf and -bF in the same test run.
3470 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
3472 44. The "Exiscan patch" is now merged into the mainline Exim source.
3474 45. Sometimes the final signoff response after QUIT could fail to get
3475 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
3476 < 0 before doing a fflush(). This bug looks as though it goes back to the
3477 introduction of TLS in release 3.20, but "sometimes" must have been rare
3478 because the tests only now provoked it.
3480 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
3481 (this can affect the format of dates).
3483 47. exim_tidydb, when checking for the continued existence of a message for
3484 which it has found a message-specific retry record, was not finding
3485 messages that were in split spool directories. Consequently, it was
3486 deleting retry records that should have stayed in existence.
3488 48. Steve fixed some bugs in eximstats.
3490 49. The SPA authentication driver was not abandoning authentication and moving
3491 on to the next authenticator when an expansion was forced to fail,
3492 contradicting the general specification for all authenticators. Instead it
3493 was generating a temporary error. It now behaves as specified.
3495 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
3496 (the order specifies the preference for clients). The order is now AES256,
3497 AES128, 3DES, ARCFOUR128.
3499 51. Small patch to Sieve code - explicitly set From: when generating an
3502 52. Exim crashed if a remote delivery caused a very long error message to be
3503 recorded - for instance if somebody sent an entire SpamAssassin report back
3504 as a large number of 550 error lines. This bug was coincidentally fixed by
3505 increasing the size of one of Exim's internal buffers (big_buffer) that
3506 happened as part of the Exiscan merge. However, to be on the safe side, I
3507 have made the code more robust (and fixed the comments that describe what
3510 53. Now that there can be additional text after "Completed" in log lines (if
3511 the queue_time_overall log selector is set), a one-byte patch to exigrep
3512 was needed to allow it to recognize "Completed" as not the last thing in
3515 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
3516 patch that reportedly fixes this has been added. I am not expert enough to
3517 create a test for it. This is what the patch creator wrote:
3519 "I found a little strange behaviour of ldap code when working with
3520 Windows 2003 AD Domain, where users was placed in more than one
3521 Organization Units. When I tried to give exim partial DN, the exit code
3522 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
3523 But simultaneously result of request was absolutely normal ldap result,
3524 so I produce this patch..."
3526 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
3527 REFERENCE, so I have modified the code to exclude the patch when that macro
3530 55. Some experimental protocols are using DNS PTR records for new purposes. The
3531 keys for these records are domain names, not reversed IP addresses. The
3532 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
3533 leaves it alone. Component reversal etc. now happens only for IP addresses.
3536 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
3538 57. Double the size of the debug message buffer (to 2048) so that more of very
3539 long debug lines gets shown.
3541 58. The exicyclog utility now does better if the number of log files to keep
3542 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
3544 59. Two changes related to the smtp_active_hostname option:
3546 (1) $smtp_active_hostname is now available as a variable.
3547 (2) The default for smtp_banner uses $smtp_active_hostname instead
3548 of $primary_hostname.
3550 60. The host_aton() function is supposed to be passed a string that is known
3551 to be a valid IP address. However, in the case of IPv6 addresses, it was
3552 not checking this. This is a hostage to fortune. Exim now panics and dies
3553 if the condition is not met. A case was found where this could be provoked
3554 from a dnsdb PTR lookup with an IPv6 address that had more than 8
3555 components; fortuitously, this particular loophole had already been fixed
3556 by change 4.50/55 above.
3558 If there are any other similar loopholes, the new check in host_aton()
3559 itself should stop them being exploited. The report I received stated that
3560 data on the command line could provoke the exploit when Exim was running as
3561 exim, but did not say which command line option was involved. All I could
3562 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
3563 running as the user.
3566 61. There was a buffer overflow vulnerability in the SPA authentication code
3567 (which came originally from the Samba project). I have added a test to the
3568 spa_base64_to_bits() function which I hope fixes it.
3571 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
3572 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
3574 63. The daemon start-up calls getloadavg() while still root for those OS that
3575 need the first call to be done as root, but it missed one case: when
3576 deliver_queue_load_max is set with deliver_drop_privilege. This is
3577 necessary for the benefit of the queue runner, because there is no re-exec
3578 when deliver_drop_privilege is set.
3580 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
3581 This has been fixed.
3583 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
3584 was in use, was not putting the data itself into the right store pool;
3585 consequently, it could be overwritten for a subsequent message in the same
3586 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
3589 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
3591 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
3592 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
3593 the function were treating the return as a boolean value, which happened to
3594 work because 0=false and not-0=true, but is not correct code.
3596 68. The host_aton() function was not handling scoped IPv6 addresses (those
3597 with, for example, "%eth0" on the end) correctly.
3599 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
3600 negated items (that is, ~something) in unsigned ints. Some compilers
3601 apparently mutter when there is no cast.
3603 70. If an address verification called from an ACL failed, and did not produce a
3604 user-specific message (i.e. there was only a "system" message), nothing was
3605 put in $acl_verify_message. In this situation, it now puts the system
3608 71. Change 4.23/11 added synchronization checking at the start of an SMTP
3609 session; change 4.31/43 added the unwanted input to the log line - except
3610 that it did not do this in the start of session case. It now does.
3612 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
3613 This is wrong and can cause the other end to generate a synchronization
3614 error if it is another Exim or anything else that does the synchronization
3615 check. A QUIT command is no longer sent after a timeout.
3617 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
3618 during host lookups.
3620 74. The defer_ok option of callout verification was not working if it was used
3621 when verifying addresses in header lines, that is, for this case:
3623 verify = header_sender/callout=defer_ok
3625 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
3626 those file descriptors could be used for SMTP connections. If anything
3627 wrote to stderr (the example that came up was "warn" in embedded Perl), it
3628 could be sent to the SMTP client, causing chaos. The daemon now opens
3629 stdin, stdout, and stderr to /dev/null when it puts itself into the
3632 76. Arrange for output from Perl's "warn" command to be written to Exim's main
3633 log by default. The user can override this with suitable Perl magic.
3635 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
3636 the log message when discard triggers, was not working for the DATA ACL or
3637 for the non-SMTP ACL.
3639 78. Error message wording change in sieve.c.
3641 79. If smtp_accept_max_per_host was set, the number of connections could be
3642 restricted to fewer than expected, because the daemon was trying to set up
3643 a new connection before checking whether the processes handling previous
3644 connections had finished. The check for completed processes is now done
3645 earlier. On busy systems, this bug wouldn't be noticed because something
3646 else would have woken the daemon, and it would have reaped the completed
3649 80. If a message was submitted locally by a user whose login name contained one
3650 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
3651 It caused a spool format error. I have fixed the spool reading code. A
3652 related problem was that the "from" clause in the Received: line became
3653 illegal because of the space(s). It is now covered by ${quote_local_part.
3655 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
3658 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
3660 83. Patch from Sieve maintainer to fix unterminated string problem in
3661 "vacation" handling.
3663 84. Some minor changes to the Linux configuration files to help with other
3664 OS variants using glibc.
3666 85. One more patch for Sieve to update vacation handling to latest spec.
3669 ----------------------------------------------------
3670 See the note above about the 4.44 and 4.50 releases.
3671 ----------------------------------------------------
3677 1. Change 4.43/35 introduced a bug that caused file counts to be
3678 incorrectly computed when quota_filecount was set in an appendfile
3681 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
3682 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
3685 3. Give more explanation in the error message when the command for a transport
3686 filter fails to execute.
3688 4. There are several places where Exim runs a non-Exim command in a
3689 subprocess. The SIGUSR1 signal should be disabled for these processes. This
3690 was being done only for the command run by the queryprogram router. It is
3691 now done for all such subprocesses. The other cases are: ${run, transport
3692 filters, and the commands run by the lmtp and pipe transports.
3694 5. Some older OS have a limit of 256 on the maximum number of file
3695 descriptors. Exim was using setrlimit() to set 1000 as a large value
3696 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
3697 systems. I've change it so that if it can't get 1000, it tries for 256.
3699 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
3700 was an oversight, and furthermore, ever since the addition of extra
3701 controls (e.g. 4.43/32), the checks on when to allow different forms of
3702 "control" were broken. There should now be diagnostics for all cases when a
3703 control that does not make sense is encountered.
3705 7. $recipients is now available in the predata ACL (oversight).
3707 8. Tidy the search cache before the fork to do a delivery from a message
3708 received from the command line. Otherwise the child will trigger a lookup
3709 failure and thereby defer the delivery if it tries to use (for example) a
3710 cached ldap connection that the parent has called unbind on.
3712 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
3713 of $address_data from the recipient verification was clobbered by the
3714 sender verification.
3716 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
3717 was its contents. (It was OK if the option was not defined at all.)
3719 11. A "Completed" log line is now written for messages that are removed from
3720 the spool by the -Mrm option.
3722 12. $host_address is now set to the target address during the checking of
3723 ignore_target_hosts.
3725 13. When checking ignore_target_hosts for an ipliteral router, no host name was
3726 being passed; this would have caused $sender_host_name to have been used if
3727 matching the list had actually called for a host name (not very likely,
3728 since this list is usually IP addresses). A host name is now passed as
3731 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
3732 code that specifies a non-restarting handler (typically sigaction() in
3733 modern systems) in an attempt to fix a rare and obscure crash bug.
3735 15. Narrowed the window for a race in the daemon that could cause it to ignore
3736 SIGCHLD signals. This is not a major problem, because they are used only to
3737 wake it up if nothing else does.
3739 16. A malformed maildirsize file could cause Exim to calculate negative values
3740 for the mailbox size or file count. Odd effects could occur as a result.
3741 The maildirsize information is now recalculated if the size or filecount
3744 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
3745 support for a long time. Removed HAVE_SYS_VFS_H.
3747 18. Updated exipick to current release from John Jetmore.
3749 19. Allow an empty sender to be matched against a lookup in an address list.
3750 Previously the only cases considered were a regular expression, or an
3753 20. Exim went into a mad DNS lookup loop when doing a callout where the
3754 host was specified on the transport, if the DNS lookup yielded more than
3757 21. The RFC2047 encoding function was originally intended for short strings
3758 such as real names; it was not keeping to the 75-character limit for
3759 encoded words that the RFC imposes. It now respects the limit, and
3760 generates multiple encoded words if necessary. To be on the safe side, I
3761 have increased the buffer size for the ${rfc2047: expansion operator from
3764 22. Failure to deliver a bounce message always caused it to be frozen, even if
3765 there was an errors_to setting on the router. The errors_to setting is now
3768 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
3769 canonical form (fully expanded) before being placed in
3770 $sender_host_address.
3772 24. Updated eximstats to version 1.33
3774 25. Include certificate and key file names in error message when GnuTLS fails
3775 to set them up, because the GnuTLS error message doesn't include the name
3776 of the failing file when there is a problem reading it.
3778 26. Expand error message when OpenSSL has problems setting up cert/key files.
3781 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
3782 (this can affect the format of dates).
3784 28. exim_tidydb, when checking for the continued existence of a message for
3785 which it has found a message-specific retry record, was not finding
3786 messages that were in split spool directories. Consequently, it was
3787 deleting retry records that should have stayed in existence.
3789 29. eximstats updated to version 1.35
3790 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
3791 1.35 - bugfix such that pie charts by volume are generated correctly
3793 30. The SPA authentication driver was not abandoning authentication and moving
3794 on to the next authenticator when an expansion was forced to fail,
3795 contradicting the general specification for all authenticators. Instead it
3796 was generating a temporary error. It now behaves as specified.
3798 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
3799 (the order specifies the preference for clients). The order is now AES256,
3800 AES128, 3DES, ARCFOUR128.
3802 31. Small patch to Sieve code - explicitly set From: when generating an
3805 32. Exim crashed if a remote delivery caused a very long error message to be
3806 recorded - for instance if somebody sent an entire SpamAssassin report back
3807 as a large number of 550 error lines. This bug was coincidentally fixed by
3808 increasing the size of one of Exim's internal buffers (big_buffer) that
3809 happened as part of the Exiscan merge. However, to be on the safe side, I
3810 have made the code more robust (and fixed the comments that describe what
3813 33. Some experimental protocols are using DNS PTR records for new purposes. The
3814 keys for these records are domain names, not reversed IP addresses. The
3815 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
3816 leaves it alone. Component reversal etc. now happens only for IP addresses.
3819 34. The host_aton() function is supposed to be passed a string that is known
3820 to be a valid IP address. However, in the case of IPv6 addresses, it was
3821 not checking this. This is a hostage to fortune. Exim now panics and dies
3822 if the condition is not met. A case was found where this could be provoked
3823 from a dnsdb PTR lookup with an IPv6 address that had more than 8
3824 components; fortuitously, this particular loophole had already been fixed
3825 by change 4.50/55 or 4.44/33 above.
3827 If there are any other similar loopholes, the new check in host_aton()
3828 itself should stop them being exploited. The report I received stated that
3829 data on the command line could provoke the exploit when Exim was running as
3830 exim, but did not say which command line option was involved. All I could
3831 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
3832 running as the user.
3835 35. There was a buffer overflow vulnerability in the SPA authentication code
3836 (which came originally from the Samba project). I have added a test to the
3837 spa_base64_to_bits() function which I hope fixes it.
3840 36. The daemon start-up calls getloadavg() while still root for those OS that
3841 need the first call to be done as root, but it missed one case: when
3842 deliver_queue_load_max is set with deliver_drop_privilege. This is
3843 necessary for the benefit of the queue runner, because there is no re-exec
3844 when deliver_drop_privilege is set.
3846 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
3847 was in use, was not putting the data itself into the right store pool;
3848 consequently, it could be overwritten for a subsequent message in the same
3849 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
3852 38. Sometimes the final signoff response after QUIT could fail to get
3853 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
3854 < 0 before doing a fflush(). This bug looks as though it goes back to the
3855 introduction of TLS in release 3.20, but "sometimes" must have been rare
3856 because the tests only now provoked it.
3862 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
3863 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
3864 Now it returns an integer. A number of calls were still expecting a T/F
3865 return. Fortuitously, in all cases, the tests worked in OK situations,
3866 which is the norm. However, things would have gone wrong on any write
3867 failures on the smtp file descriptor. This function is used when sending
3868 messages over SMTP and also when doing verify callouts.
3870 2. When Exim is called to do synchronous delivery of a locally submitted
3871 message (the -odf or -odi options), it no longer closes stderr before doing
3874 3. Implemented the mua_wrapper option.
3876 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
3878 5. Implemented the functions header_remove(), header_testname(),
3879 header_add_at_position(), and receive_remove_recipient(), and exported them
3882 6. If an ACL "warn" statement specified the addition of headers, Exim already
3883 inserted X-ACL-Warn: at the start if there was no header name. However, it
3884 was not making this test for the second and subsequent header lines if
3885 there were newlines in the string. This meant that an invalid header could
3886 be inserted if Exim was badly configured.
3888 7. Allow an ACL "warn" statement to add header lines at the start or after all
3889 the Received: headers, as well as at the end.
3891 8. Added the rcpt_4xx retry error code.
3893 9. Added postmaster_mailfrom=xxx to callout verification option.
3895 10. Added mailfrom=xxxx to the callout verification option, for verify=
3898 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
3899 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
3901 12. Inserted some casts to stop certain compilers warning when using pointer
3902 differences as field lengths or precisions in printf-type calls (mostly
3903 affecting debugging statements).
3905 13. Added optional readline() support for -be (dynamically loaded).
3907 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
3908 same clock tick as a message's arrival, so that its received time was the
3909 same as the "first fail" time on the retry record, and that message
3910 remained on the queue past the ultimate address timeout, every queue runner
3911 would try a delivery (because it was past the ultimate address timeout) but
3912 after another failure, the ultimate address timeout, which should have then
3913 bounced the address, did not kick in. This was a "< instead of <=" error;
3914 in most cases the first failure would have been in the next clock tick
3915 after the received time, and all would be well.
3917 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
3918 being recognized when the domain list was tested by the match_domain
3919 condition in an expansion string.
3921 16. Added the ${str2b64: operator.
3923 17. Exim was always calling setrlimit() to set a large limit for the number of
3924 processes, without checking whether the existing limit was already
3925 adequate. (It did check for the limit on file descriptors.) Furthermore,
3926 errors from getrlimit() and setrlimit() were being ignored. Now they are
3927 logged to the main and panic logs, but Exim does carry on, to try to do its
3928 job under whatever limits there are.
3930 18. Imported PCRE 5.0.
3932 19. Trivial typo in log message " temporarily refused connection" (the leading
3935 20. If the log selector return_path_on_delivery was set and an address was
3936 redirected to /dev/null, the delivery process crashed because it assumed
3937 that a return path would always be set for a "successful" delivery. In this
3938 case, the whole delivery is bypassed as an optimization, and therefore no
3941 21. Internal re-arrangement: the function for sending a challenge and reading
3942 a response while authentication was assuming a zero-terminated challenge
3943 string. It's now changed to take a pointer and a length, to allow for
3944 binary data in such strings.
3946 22. Added the cyrus_sasl authenticator (code supplied by MBM).
3948 23. Exim was not respecting finduser_retries when seeking the login of the
3949 uid under which it was called; it was always trying 10 times. (The default
3950 setting of finduser_retries is zero.) Also, it was sleeping after the final
3951 failure, which is pointless.
3953 24. Implemented tls_on_connect_ports.
3955 25. Implemented acl_smtp_predata.
3957 26. If the domain in control=submission is set empty, Exim assumes that the
3958 authenticated id is a complete email address when it generates From: or
3959 Sender: header lines.
3961 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
3962 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
3963 chown and chgrp in /bin and hostname in /usr/bin.
3965 28. Exim was keeping the "process log" file open after each use, just as it
3966 does for the main log. This opens the possibility of it remaining open for
3967 long periods when the USR1 signal hits a daemon. Occasional processlog
3968 errors were reported, that could have been caused by this. Anyway, it seems
3969 much more sensible not to leave this file open at all, so that is what now
3972 29. The long-running daemon process does not normally write to the log once it
3973 has entered its main loop, and it closes the log before doing so. This is
3974 so that log files can straightforwardly be renamed and moved. However,
3975 there are a couple of unusual error situations where the daemon does write
3976 log entries, and I had neglected to close the log afterwards.
3978 30. The text of an SMTP error response that was received during a remote
3979 delivery was being truncated at 512 bytes. This is too short for some of
3980 the long messages that one sometimes sees. I've increased the limit to
3983 31. It is now possible to make retry rules that apply only when a message has a
3984 specific sender, in particular, an empty sender.
3986 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
3987 it possible to be selective about when SMTP synchronization is enforced.
3989 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
3991 32. Implemented hosts_connection_nolog.
3993 33. Added an ACL for QUIT.
3995 34. Setting "delay_warning=" to disable warnings was not working; it gave a
3998 35. Added mailbox_size and mailbox_filecount to appendfile.
4000 36. Added control = no_multiline_responses to ACLs.
4002 37. There was a bug in the logic of the code that waits for the clock to tick
4003 in the case where the clock went backwards by a substantial amount such
4004 that the microsecond fraction of "now" was more than the microsecond
4005 fraction of "then" (but the whole seconds number was less).
4007 38. Added support for the libradius Radius client library this is found on
4008 FreeBSD (previously only the radiusclient library was supported).
4014 1. When certain lookups returned multiple values in the form name=value, the
4015 quoting of the values was not always being done properly. Specifically:
4016 (a) If the value started with a double quote, but contained no whitespace,
4018 (b) If the value contained whitespace other than a space character (i.e.
4019 tabs or newlines or carriage returns) it was not quoted.
4020 This fix has been applied to the mysql and pgsql lookups by writing a
4021 separate quoting function and calling it from the lookup code. The fix
4022 should probably also be applied to nisplus, ibase and oracle lookups, but
4023 since I cannot test any of those, I have not disturbed their existing code.
4025 2. A hit in the callout cache for a specific address caused a log line with no
4026 reason for rejecting RCPT. Now it says "Previous (cached) callout
4027 verification failure".
4029 3. There was an off-by-one bug in the queryprogram router. An over-long
4030 return line was truncated at 256 instead of 255 characters, thereby
4031 overflowing its buffer with the terminating zero. As well as fixing this, I
4032 have increased the buffer size to 1024 (and made a note to document this).
4034 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
4035 when Exim is waiting for an SMTP response from a remote server, Exim
4036 restarts its select() call on the socket, thereby resetting its timeout.
4037 This is not a problem when such interrupts are rare. Somebody set up a cron
4038 job to run exiwhat every 2 minutes, which is less than the normal select()
4039 timeout (5 or 10 minutes). This meant that the select() timeout never
4040 kicked in because it was always reset. I have fixed this by comparing the
4041 time when an interrupt arrives with the time at the start of the first call
4042 to select(). If more time than the timeout has elapsed, the interrupt is
4043 treated as a timeout.
4045 5. Some internal re-factoring in preparation for the addition of Sieve
4046 extensions (by MH). In particular, the "personal" test is moved to a
4047 separate function, and given an option for scanning Cc: and Bcc: (which is
4048 not set for Exim filters).
4050 6. When Exim created an email address using the login of the caller as the
4051 local part (e.g. when creating a From: or Sender: header line), it was not
4052 quoting the local part when it contained special characters such as @.
4054 7. Installed new OpenBSD configuration files.
4056 8. Reworded some messages for syntax errors in "and" and "or" conditions to
4057 try to make them clearer.
4059 9. Callout options, other than the timeout value, were being ignored when
4060 verifying sender addresses in header lines. For example, when using
4062 verify = header_sender/callout=no_cache
4064 the cache was (incorrectly) being used.
4066 10. Added a missing instance of ${EXE} to the exim_install script; this affects
4067 only the Cygwin environment.
4069 11. When return_path_on_delivery was set as a log selector, if different remote
4070 addresses in the same message used different return paths and parallel
4071 remote delivery occurred, the wrong values would sometimes be logged.
4072 (Whenever a remote delivery process finished, the return path value from
4073 the most recently started remote delivery process was logged.)
4075 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
4076 lines when AUTH and/or TLS are in use. This is the "received protocol"
4077 field. Exim used to use "asmtp" for authenticated SMTP, without any
4078 indication (in the protocol name) for TLS use. Now it follows the RFC and
4079 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
4080 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
4081 names appear in log lines as well as in Received: header lines.
4083 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
4084 extensions, and comparison tests, and to fix some bugs.
4086 14. Changes to the "personal" filter test:
4088 (1) The test was buggy in that it was just doing the equivalent of
4089 "contains" tests on header lines. For example, if a user's address was
4090 anne@some.where, the "personal" test would incorrectly be true for
4092 To: susanne@some.where
4094 This test is now done by extracting each address from the header in turn,
4095 and checking the entire address. Other tests that are part of "personal"
4096 are now done using regular expressions (for example, to check local parts
4097 of addresses in From: header lines).
4099 (2) The list of non-personal local parts in From: addresses has been
4100 extended to include "listserv", "majordomo", "*-request", and "owner-*",
4101 taken from the Sieve specification recommendations.
4103 (3) If the message contains any header line starting with "List-" it is
4104 treated as non-personal.
4106 (4) The test for "circular" in the Subject: header line has been removed
4107 because it now seems ill-conceived.
4109 15. Minor typos in src/EDITME comments corrected.
4111 16. Installed latest exipick from John Jetmore.
4113 17. If headers_add on a router specified a text string that was too long for
4114 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
4115 of string_sprintf() is now avoided.
4117 18. $message_body_size was not set (it was always zero) when running the DATA
4118 ACL and the local_scan() function.
4120 19. For the "mail" command in an Exim filter, no default was being set for
4121 the once_repeat time, causing a random time value to be used if "once" was
4122 specified. (If the value happened to be <= 0, no repeat happened.) The
4123 default is now 0s, meaning "never repeat". The "vacation" command was OK
4124 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
4125 (I found it when inspecting the code).
4127 20. There is now an overall timeout for performing a callout verification. It
4128 defaults to 4 times the callout timeout, which applies to individual SMTP
4129 commands during the callout. The overall timeout applies when there is more
4130 than one host that can be tried. The timeout is checked before trying the
4131 next host. This prevents very long delays if there are a large number of
4132 hosts and all are timing out (e.g. when the network connections are timing
4133 out). The value of the overall timeout can be changed by specifying an
4134 additional sub-option for "callout", called "maxwait". For example:
4136 verify = sender/callout=5s,maxwait=20s
4138 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
4139 to the end before writing, but this should make it even safer).
4141 22. Exim was forgetting that it had advertised PIPELINING for the second and
4142 subsequent messages on an SMTP connection. It was also not resetting its
4143 memory on STARTTLS and an internal HELO.
4145 23. When Exim logs an SMTP synchronization error within a session, it now
4146 records whether PIPELINING has been advertised or not.
4148 24. Added 3 instances of "(long int)" casts to time_t variables that were being
4149 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
4150 rather than long int.
4152 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
4154 26. Added the never_mail option to autoreply.
4160 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
4161 crash if the getsockname() call failed; this can happen if a connection is
4162 closed very soon after it is established. The problem was simply in the
4163 order in which certain operations were done, causing Exim to try to write
4164 to the SMTP stream before it had set up the file descriptor. The bug has
4165 been fixed by making things happen in the correct order.
4171 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
4172 before the connection was closed, thus losing the rejection response.
4174 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
4175 some early Solaris releases, but causes trouble in current releases where
4176 socklen_t is defined.
4178 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
4181 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
4184 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
4185 that releasing the top of it at the end releases what was used for sub-
4186 expansions (unless the block got too big). However, discard this block if
4187 the first thing is a variable or header, so that we can use its block when
4188 it is dynamic (useful for very large $message_headers, for example).
4190 6. Lookups now cache *every* query, not just the most recent. A new, separate
4191 store pool is used for this. It can be recovered when all lookup caches are
4192 flushed. Lookups now release memory at the end of their result strings.
4193 This has involved some general refactoring of the lookup sources.
4195 7. Some code has been added to the store_xxx() functions to reduce the amount
4196 of flapping under certain conditions.
4198 8. log_incoming_interface used to affect only the <= reception log lines. Now
4199 it causes the local interface and port to be added to several more SMTP log
4200 lines, for example "SMTP connection from", and rejection lines.
4202 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
4204 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
4206 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
4207 could be overwritten at the end of the current message (or the start of a
4208 new message if it was set in a HELO ACL). The value is now preserved for
4209 the duration of the SMTP connection.
4211 12. If a transport had a headers_rewrite setting, and a matching header line
4212 contained an unqualified address, that address was qualified, even if it
4213 did not match any rewriting rules. The underlying bug was that the values
4214 of the flags that permit the existence of unqualified sender and recipient
4215 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
4216 non-local messages, and by -bnq for local messages) were not being
4217 preserved with the message after it was received.
4219 13. When Exim was logging an SMTP synchronization error, it could sometimes log
4220 "next input=" as part of the text comprising the host identity instead of
4221 the correct text. The code was using the same buffer for two different
4222 strings. However, depending on which order the printing function evaluated
4223 its arguments, the bug did not always show up. Under Linux, for example, my
4224 test suite worked just fine.
4226 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
4227 doesn't work with some older versions of Perl. It has been changed to "my",
4228 which in any case is probably the better facility to use.
4230 15. A really picky compiler found some instances of statements for creating
4231 error messages that either had too many or two few arguments for the format
4234 16. The size of the buffer for calls to the DNS resolver has been increased
4235 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
4236 for addresses that have a lot of PTR records. This alleviates a problem; it
4237 does not fully solve it.
4239 17. A dnsdb lookup for PTR records that receives more data than will fit in the
4240 buffer now truncates the list and logs the incident, which is the same
4241 action as happens when Exim is looking up a host name and its aliases.
4242 Previously in this situation something unpredictable would happen;
4243 sometimes it was "internal error: store_reset failed".
4245 18. If a server dropped the connection unexpectedly when an Exim client was
4246 using GnuTLS and trying to read a response, the client delivery process
4247 crashed while trying to generate an error log message.
4249 19. If a "warn" verb in an ACL added multiple headers to a message in a single
4250 string, for example:
4252 warn message = H1: something\nH2: something
4254 the text was added as a single header line from Exim's point of view
4255 though it ended up OK in the delivered message. However, searching for the
4256 second and subsequent header lines using $h_h2: did not work. This has been
4257 fixed. Similarly, if a system filter added multiple headers in this way,
4258 the routers could not see them.
4260 20. Expanded the error message when iplsearch is called with an invalid key to
4261 suggest using net-iplsearch in a host list.
4263 21. When running tests using -bh, any delays imposed by "delay" modifiers in
4264 ACLs are no longer actually imposed (and a message to that effect is
4267 22. If a "gecos" field in a passwd entry contained escaped characters, in
4268 particular, if it contained a \" sequence, Exim got it wrong when building
4269 a From: or a Sender: header from that name. A second bug also caused
4270 incorrect handling when an unquoted " was present following a character
4271 that needed quoting.
4273 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
4274 was not being matched caselessly.
4276 24. Arranged for all hyphens in the exim.8 source to be escaped with
4279 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
4280 properly. Recipient callout cache records were still being keyed to include
4281 the sender, even when use_sender was set false. This led to far more
4282 callouts that were necessary. The sender is no longer included in the key
4283 when use_sender is false.
4285 26. Added "control = submission" modifier to ACLs.
4287 27. Added the ${base62d: operator to decode base 62 numbers.
4289 28. dnsdb lookups can now access SRV records.
4291 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
4292 the configuration file.
4294 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
4295 (-v) mode. This makes the output for a verbose queue run more intelligible.
4297 31. Added a use_postmaster feature to recipient callouts.
4299 32. Added the $body_zerocount variable, containing the number of binary zero
4300 bytes in the message body.
4302 33. The time of last modification of the "new" subdirectory is now used as the
4303 "mailbox time last read" when there is a quota error for a maildir
4306 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
4308 35. Added +ignore_unknown as a special item in host lists.
4310 36. Code for decoding IPv6 addresses in host lists is now included, even if
4311 IPv6 support is not being compiled. This fixes a bug in which an IPv6
4312 address was recognized as an IP address, but was then not correctly decoded
4313 into binary, causing unexpected and incorrect effects when compared with
4320 1. Very minor rewording of debugging text in manualroute to say "list of
4321 hosts" instead of "hostlist".
4323 2. If verify=header_syntax was set, and a header line with an unqualified
4324 address (no domain) and a large number of spaces between the end of the
4325 name and the colon was received, the reception process suffered a buffer
4326 overflow, and (when I tested it) crashed. This was caused by some obsolete
4327 code that should have been removed. The fix is to remove it!
4329 3. When running in the test harness, delay a bit after writing a bounce
4330 message to get a bit more predictability in the log output.
4332 4. Added a call to search_tidyup() just before forking a reception process. In
4333 theory, someone could use a lookup in the expansion of smtp_accept_max_
4334 per_host which, without the tidyup, could leave open a database connection.
4336 5. Added the variables $recipient_data and $sender_data which get set from a
4337 lookup success in an ACL "recipients" or "senders" condition, or a router
4338 "senders" option, similar to $domain_data and $local_part_data.
4340 6. Moved the writing of debug_print from before to after the "senders" test
4343 7. Change 4.31/66 (moving the time when the Received: is generated) caused
4344 problems for message scanning, either using a data ACL, or using
4345 local_scan() because the Received: header was not generated till after they
4346 were called (in order to set the time as the time of reception completion).
4347 I have revised the way this works. The header is now generated after the
4348 body is received, but before the ACL or local_scan() are called. After they
4349 are run, the timestamp in the header is updated.
4355 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
4356 before starting a queue runner without re-exec. This happened only when
4357 deliver_drop_privilege was set or when the Exim user was set to root. The
4358 effect of the bug was that timeouts during subsequent deliveries caused
4359 crashes instead of being properly handled. The handler is now left at its
4360 default (and expected) setting.
4362 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
4363 message, again when deliver_drop_privilege is set or Exim is run as root.
4364 The bug described in (1) was not present in this case, but the tidying up
4365 of the other signals was missing. I have made the two cases consistent.
4367 3. The ignore_target_hosts setting on a manualroute router was being ignored
4368 for hosts that were looked up using the /MX notation.
4370 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
4373 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
4374 operated on the sender address. After changing the $sender_address to <>
4375 for the sender address verify, Exim was re-instated it as the original
4376 (before rewriting) address, but remembering that it had rewritten it, so it
4377 wasn't rewriting it again. This bug also had the effect of breaking the
4378 sender address verification caching when the sender address was rewritten.
4380 6. The ignore_target_hosts option was being ignored by the ipliteral router.
4381 This has been changed so that if the ip literal address matches
4382 ignore_target_hosts, the router declines.
4384 7. Added expansion conditions match_domain, match_address, and match_local_
4385 part (NOT match_host).
4387 8. The placeholder for the Received: header didn't have a length field set.
4389 9. Added code to Exim itself and to exim_lock to test for a specific race
4390 condition that could lead to file corruption when using MBX delivery. The
4391 issue is with the lockfile that is created in /tmp. If this file is removed
4392 after a process has opened it but before that process has acquired a lock,
4393 there is the potential for a second process to recreate the file and also
4394 acquire a lock. This could lead to two Exim processes writing to the file
4395 at the same time. The added code performs the same test as UW imapd; it
4396 checks after acquiring the lock that its file descriptor still refers to
4397 the same named file.
4399 10. The buffer for building added header lines was of fixed size, 8192 bytes.
4400 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
4403 11. Added the smtp_active_hostname option. If used, this will typically be made
4404 to depend on the incoming interface address. Because $interface_address is
4405 not set up until the daemon has forked a reception process, error responses
4406 that can happen earlier (such as "too many connections") no longer contain
4409 12. If an expansion in a condition on a "warn" statement fails because a lookup
4410 defers, the "warn" statement is abandoned, and the next ACL statement is
4411 processed. Previously this caused the whole ACL to be aborted.
4413 13. Added the iplsearch lookup type.
4415 14. Added ident_timeout as a log selector.
4417 15. Added tls_certificate_verified as a log selector.
4419 16. Added a global option tls_require_ciphers (compare the smtp transport
4420 option of the same name). This controls incoming TLS connections.
4422 17. I finally figured out how to make tls_require_ciphers do a similar thing
4423 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
4424 before starting the TLS session.
4426 18. Tabs are now shown as \t in -bP output.
4428 19. If the log selector return_path_on_delivery was set, Exim crashed when
4429 bouncing a message because it had too many Received: header lines.
4431 20. If two routers both had headers_remove settings, and the first one included
4432 a superfluous trailing colon, the final name in the first list and the
4433 first name in the second list were incorrectly joined into one item (with a
4434 colon in the middle).
4440 1. Added -C and -D options to the exinext utility, mainly to make it easier
4441 to include in the automated testing, but these could be helpful when
4442 multiple configurations are in use.
4444 2. The exinext utility was not formatting the output nicely when there was
4445 an alternate port involved in the retry record key, nor when there was a
4446 message id as well (for retries that were specific to a specific message
4447 and a specific host). It was also confused by IPv6 addresses, because of
4448 the additional colons they contain. I have fixed the IPv4 problem, and
4449 patched it up to do a reasonable job for IPv6.
4451 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
4452 delivery, the log line now contains "pipelined" if PIPELINING was used.
4454 4. An SMTP transport process used to panic and die if the bind() call to set
4455 an explicit outgoing interface failed. This has been changed; it is now
4456 treated in the same way as a connect() failure.
4458 5. A reference to $sender_host_name in the part of a conditional expansion
4459 that was being skipped was still causing a DNS lookup. This no longer
4462 6. The def: expansion condition was not recognizing references to header lines
4463 that used bh_ and bheader_.
4465 7. Added the _cache feature to named lists.
4467 8. The code for checking quota_filecount in the appendfile transport was
4468 allowing one more file than it should have been.
4470 9. For compatibility with Sendmail, the command line option
4478 and sets the incoming protocol and host name (for trusted callers). The
4479 host name and its colon can be omitted when only the protocol is to be set.
4480 Note the Exim already has two private options, -pd and -ps, that refer to
4481 embedded Perl. It is therefore impossible to set a protocol value of "d" or
4482 "s", but I don't think that's a major issue.
4484 10. A number of refactoring changes to the code, none of which should affect
4487 (a) The number of logging options was getting close to filling up the
4488 32-bit word that was used as a bit map. I have split them into two classes:
4489 those that are passed in the argument to log_write(), and those that are
4490 only ever tested independently outside of that function. These are now in
4491 separate 32-bit words, so there is plenty of room for expansion again.
4492 There is no change in the user interface or the logging behaviour.
4494 (b) When building, for example, log lines, the code previously used a
4495 macro that called string_cat() twice, in order to add two strings. This is
4496 not really sufficiently general. Furthermore, there was one instance where
4497 it was actually wrong because one of the argument was used twice, and in
4498 one call a function was used. (As it happened, calling the function twice
4499 did not affect the overall behaviour.) The macro has been replaced by a
4500 function that can join an arbitrary number of extra strings onto a growing
4503 (c) The code for expansion conditions now uses a table and a binary chop
4504 instead of a serial search (which was left over from when there were very
4505 few conditions). Also, it now recognizes conditions like "pam" even when
4506 the relevant support is not compiled in: a suitably worded error message is
4507 given if an attempt is made to use such a condition.
4509 11. Added ${time_interval:xxxxx}.
4511 12. A bug was causing one of the ddress fields not to be passed back correctly
4512 from remote delivery subprocesses. The field in question was not being
4513 subsequently used, so this caused to problems in practice.
4515 13. Added new log selectors queue_time and deliver_time.
4517 14. Might have fixed a bug in maildirsizefile handling that threw up
4518 "unexpected character" debug warnings, and recalculated the data
4519 unnecessarily. In any case, I expanded the warning message to give more
4522 15. Added the message "Restricted characters in address" to the statements in
4523 the default ACL that block characters like @ and % in local parts.
4525 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
4526 Three changes have been made:
4528 (a) There was a serious bug; a negative response to MAIL caused the whole
4529 recipient domain to be cached as invalid, thereby blocking all messages
4530 to all local parts at the same domain, from all senders. This bug has
4531 been fixed. The domain is no longer cached after a negative response to
4532 MAIL if the sender used is not empty.
4534 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
4537 (c) A new callout option, "use_sender" has been added for people who want
4538 the modified behaviour.
4544 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
4547 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
4548 indeed breaks things for older releases.
4550 3. Added additional logging to the case where there is a problem reading data
4551 from a filter that is running in a subprocess using a pipe, in order to
4552 try to track down a specific problem.
4554 4. Testing facility fudge: when running in the test harness and attempting
4555 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
4556 getting "No route to host". Convert this to a timeout.
4558 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
4561 6. Some OS don't have socklen_t but use size_t instead. This affects the
4562 fifth argument of getsockopt() amongst other things. This is now
4563 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
4564 can be set for individual OS. I have set it for SunOS5, OSF1, and
4565 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
4566 some earlier ones do not.
4568 7. Change 4.30/15 was not doing the test caselessly.
4570 8. The standard form for an IPv6 address literal was being rejected by address
4571 parsing in, for example, MAIL and RCPT commands. An example of this kind of
4572 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
4573 this, as well as the form without the "IPv6" on the front (but only when
4574 address literals are enabled, of course).
4576 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
4578 10. Exim crashed if a message with an empty sender address specified by -f
4579 encountered a router with an errors_to setting. This could be provoked only
4580 by a command such as
4584 where an empty string was supplied; "<>" did not hit this bug.
4586 11. Installed PCRE release 4.5.
4588 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
4589 remained set. It is now erased.
4591 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
4592 times from message ids (which are base 36 rather than the normal 62).
4594 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
4595 were being counted as actual protocol errors, and logged if the log
4596 selector +smtp_protocol_error was set. One cannot be perfect in this test,
4597 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
4598 and DATA following a set of rejected RCPTs do not count as protocol errors.
4599 In other words, Exim assumes they were pipelined, though this may not
4600 actually be the case. Of course, in all cases the client gets an
4601 appropriate error code.
4603 15. If a lookup fails in an ACL condition, a message about the failure may
4604 be available; it is used if testing the ACL cannot continue, because most
4605 such messages specify what the cause of the deferral is. However, some
4606 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
4607 that caused an old message to be retained and used if a later statement
4608 caused a defer, replacing the real cause of the deferral.
4610 16. If an IP address had so many PTR records that the DNS lookup buffer
4611 was not large enough to hold them, Exim could crash while trying to process
4612 the truncated data. It now detects and logs this case.
4614 17. Further to 4.21/58, another change has been made: if (and only if) the
4615 first line of a message (the first header line) ends with CRLF, a bare LF
4616 in a subsequent header line has a space inserted after it, so as not to
4617 terminate the header.
4619 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
4620 unnecessarily, used atoi() instead of strtol(), and didn't check the
4621 termination when getting file sizes from file names by regex.
4623 19. Completely re-implemented the support for maildirsize files, in the light
4624 of a number of problems with the previous contributed implementation
4625 (4.30/29). In particular:
4627 . If the quota is zero, the maildirsize file is maintained, but no quota is
4630 . If the maildir directory does not exist, it is created before any attempt
4631 to write a maildirsize file.
4633 . The quota value in the file is just a cache; if the quota is changed in
4634 the transport, the new value overrides.
4636 . A regular expression is available for excluding directories from the
4639 20. The autoreply transport checks the characters in options that define the
4640 message's headers; it allows continued headers, but it was checking with
4641 isspace() after an embedded newline instead of explicitly looking for a
4644 21. If all the "regular" hosts to which an address was routed had passed their
4645 expiry times, and had not reached their retry times, the address was
4646 bounced, even if fallback hosts were defined. Now Exim should go on to try
4649 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
4650 equivalent code in the SMTP transport. Some hosts send humungous responses
4651 to HELO/EHLO, more than 1024 it seems.
4653 23. Refactoring: code in filter.c used (void *) for "any old type" but this
4654 gives compiler warnings in some environments. I've now done it "properly",
4657 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
4658 (because of problems with the built-in one) was declared to return uschar *
4659 instead of char *, causing compiler failure.
4661 25. Fixed a file descriptor leak when processing alias/forward files.
4663 26. Fixed a minor format string issue in dbfn.c.
4665 27. Typo in exim.c: ("dmbnz" for "dbmnz").
4667 28. If a filter file refered to $h_xxx or $message_headers, and the headers
4668 contained RFC 2047 "words", Exim's memory could, under certain conditions,
4671 29. When a sender address is verified, it is cached, to save repeating the test
4672 when there is more than one recipient in a message. However, when the
4673 verification involves a callout, it is possible for different callout
4674 options to be set for different recipients. It is too complicated to keep
4675 track of this in the cache, so now Exim always runs a verification when a
4676 callout is required, relying on the callout cache for the optimization.
4677 The overhead is duplication of the address routing, but this should not be
4680 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
4681 to be verified with callout=postmaster, and the main callout worked but the
4682 postmaster check failed, the verification correctly failed. However, if a
4683 subsequent RCPT command asked for sender verification *without* the
4684 postmaster check, incorrect caching caused this verification also to fail,
4687 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
4688 it was not caching the DNS options (qualify_single, search_parents) that
4689 were used when the lookup failed. A subsequent lookup with different
4690 options therefore always gave the same answer, though there were cases
4691 where it should not have. (Example: a "domains = !$mx_any" option on a
4692 dnslookup router: the "domains" option is always processed without any
4693 widening, but the router might have qualify_single set.) Now Exim uses the
4694 cached value only when the same options are set.
4696 32. Added John Jetmore's "exipick" utility to the distribution.
4698 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
4699 than a timeout (e.g. a certificate is required, and is not provided), an
4700 Exim server now closes the connection immediately. Previously it waited for
4701 the client to close - but if the client is SSL, it seems that they each
4702 wait for each other, leading to a delay before one of them times out.
4704 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
4705 maintained 0.8.x compatibility because I don't think many are using it, and
4706 it is clearly obsolete.
4708 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
4711 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
4712 client certificate was expired. A simple patch fixes this, though I don't
4713 understand the full logic of why the verify callback is called multiple
4716 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
4717 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
4718 which causes problems with some clients (such as the Certicom SSL Plus
4719 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
4720 disables the coutermeasure allowing Eudora to connect."
4722 38. Exim was not checking that a write() to a log file succeeded. This could
4723 lead to Bad Things if a log got too big, in particular if it hit a file
4724 size limit. Exim now panics and dies if it cannot write to a log file, just
4725 as it does if it cannot open a log file.
4727 39. Modified OS/Makefile-Linux so that it now contains
4729 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
4731 The two -D definitions ensure that Exim is compiled with large file
4732 support, which makes it possible to handle log files that are bigger than
4735 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
4736 instance) a domain was checked against a named list that involved a lookup,
4737 causing $domain_data to be set, then another domain was checked against the
4738 same list, then the first domain was re-checked, the value of $domain_data
4739 after the final check could be wrong. In particular, if the second check
4740 failed, it could be set empty. This bug probably also applied to
4743 41. The strip_trailing_dot option was not being applied to the address given
4744 with the -f command-line option.
4746 42. The code for reading a message's header from the spool was incrementing
4747 $received_count, but never initializing it. This meant that the value was
4748 incorrect (doubled) while delivering a message in the same process in which
4749 it was received. In the most common configuration of Exim, this never
4750 happens - a fresh exec is done - but it can happen when
4751 deliver_drop_privilege is set.
4753 43. When Exim logs an SMTP synchronization error - client data sent too soon -
4754 it now includes up to 150 characters of the unexpected data in the log
4757 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
4758 and building data strings. The size of both of these buffers was 10 000
4759 bytes - far larger than anybody would *ever* want, thought I. Needless to
4760 say, somebody hit the limit. I have increased the maximum line length to
4761 20 000 and the maximum data length of concatenated lines to 100 000. I have
4762 also fixed two bugs, because there was no checking on these buffers. Tsk,
4763 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
4764 buffer is too small.
4766 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
4767 lsearch lookups. Now it does.
4769 46. When parsing a route_list item in a manualroute router, a fixed-length
4770 buffer was used for the list of hosts. I made this 1024 bytes long,
4771 thinking that nobody would ever have a list of hosts that long. Wrong.
4772 Somebody had a whole pile of complicated expansion conditions, and the
4773 string was silently truncated, leading to an expansion error. It turns out
4774 that it is easier to change to an unlimited length (owing to other changes
4775 that have happened since this code was originally written) than to build
4776 structure for giving a limitation error. The length of the item that
4777 expands into the list of hosts is now unlimited.
4779 47. The lsearch lookup could not handle data where the length of text line was
4780 more than 4095 characters. Such lines were truncated, leading to shortened
4781 data being returned. It should now handle lines of any length.
4783 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
4784 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
4787 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
4788 replacing strings such as BIN_DIRECTORY in a source file: the replacement
4789 no longer happens in comment lines. A list of replacements is now placed
4790 at the head of all of the source files, except those whose only change is
4791 to replace PERL_COMMAND in the very first #! line.
4793 50. Replaced the slow insertion sort in queue.c, for sorting the list of
4794 messages on the queue, with a bottom-up merge sort, using code contributed
4795 by Michael Haardt. This should make operations like -bp somewhat faster on
4796 large queues. It won't affect queue runners, except when queue_run_in_order
4799 51. Installed eximstats 1.31 in the distribution.
4801 52. Added support for SRV lookups to the dnslookup router.
4803 53. If an ACL referred to $message_body or $message_body_end, the value was not
4804 reset for any messages that followed in the same SMTP session.
4806 54. The store-handling optimization for building very long strings was not
4807 differentiating between the different store pools. I don't think this
4808 actually made any difference in practice, but I've tidied it.
4810 55. While running the routers to verify a sender address, $sender_address
4811 was still set to the sender address. This is wrong, because when routing to
4812 send a bounce to the sender, it would be empty. Therefore, I have changed
4813 it so that, while verifying a sender address, $sender_address is set to <>.
4814 (There is no change to what happens when verifying a recipient address.)
4816 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
4817 target A or AAAA records (if not already returned) without resetting the
4818 qualify_single or search_parents options of the DNS resolver. These are
4819 inappropriate in this case because the targets of MX and SRV records must
4820 be FQDNs. A broken DNS record could cause trouble if it happened to have a
4821 target that, when qualified, matched something in the local domain. These
4822 two options are now turned off when doing these lookups.
4824 57. It seems that at least some releases of Reiserfs (which does not have the
4825 concept of a fixed number of inodes) returns zero and not -1 for the
4826 number of available inodes. This interacted badly with check_spool_inodes,
4827 which assumed that -1 was the "no such thing" setting. What I have done is
4828 to check that the total number of inodes is greater than zero before doing
4829 the test of how many are available.
4831 58. When a "warn" ACL statement has a log_message modifier, the message is
4832 remembered, and not repeated. This is to avoid a lot of repetition when a
4833 message has many recipients that cause the same warning to be written.
4834 Howewer, Exim was preserving the list of already written lines for an
4835 entire SMTP session, which doesn't seem right. The memory is now reset if a
4836 new message is started.
4838 59. The "rewrite" debugging flag was not showing the result of rewriting in the
4839 debugging output unless log_rewrite was also set.
4841 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
4842 of (int)(handle) when we know that handle contains (void *)(-1).
4844 61. The Exim daemon panic-logs an error return when it closes the incoming
4845 connection. However "connection reset by peer" seems to be common, and
4846 isn't really an error worthy of noting specially, so that particular error
4849 62. When Exim is trying to find all the local interfaces, it used to panic and
4850 die if the ioctl to get the interface flags failed. However, it seems that
4851 on at least one OS (Solaris 9) it is possible to have an interface that is
4852 included in the list of interfaces, but for which you get a failure error
4853 for this call. This happens when the interface is not "plumbed" into a
4854 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
4855 failure of the "get flags" call assumes that the interface is down.
4857 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
4858 makes life easier for people who are doing arithmetic on fields extracted
4859 from dates, where you often get leading zeros that should not be
4860 interpreted as octal.
4862 64. Added qualify_domain to the redirect router, to override the global
4865 65. If a pathologically long header line contained very many addresses (the
4866 report of this problem mentioned 10 000) and each of them was rewritten,
4867 Exim could use up a very large amount of memory. (It kept on making new
4868 copies of the header line as it rewrote, and never released the old ones.)
4869 At the expense of a bit more processing, the header rewriting function has
4870 been changed so that it no longer eats memory in this way.
4872 66. The generation of the Received: header has been moved from the time that a
4873 message starts to be received, to the time that it finishes. The timestamp
4874 in the Received: header should now be very close to that of the <= log
4875 line. There are two side-effects of this change:
4877 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
4878 logged header lines no longer include the local Received: line, because
4879 it has not yet been created. The same applies to a copy of the message
4880 that is returned to a non-SMTP sender when a message is rejected.
4882 (b) When a filter file is tested using -bf, no additional Received: header
4883 is added to the test message. After some thought, I decided that this
4886 This change does not affect the value of $received_for. It is still set
4887 after address rewriting, but before local_scan() is called.
4889 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
4891 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
4892 gave an unhelpful panic error message, and a defer error. I have managed to
4893 change this behaviour so that it now rejects any supplied certificate,
4894 which seems right, as the list of acceptable certificates is empty.
4896 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
4897 gave an unhelpful defer error. I have not managed to make this reject any
4898 supplied certificates, but the error message it gives is "no certificate
4899 supplied", which is not helpful.
4901 70. exigrep's output now also includes lines that are not associated with any
4902 message, but which match the given pattern. Implemented by a patch from
4903 Martin Sluka, which also tidied up the Perl a bit.
4905 71. Recipient callout verification, like sender verification, was using <> in
4906 the MAIL FROM command. This isn't really the right thing, since the actual
4907 sender may affect whether the remote host accepts the recipient or not. I
4908 have changed it to use the actual sender in the callout; this means that
4909 the cache record is now keyed on a recipient/sender pair, not just the
4910 recipient address. There doesn't seem to be a real danger of callout loops,
4911 since a callout by the remote host to check the sender would use <>.
4912 [SEE ABOVE: changed after hitting problems.]
4914 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
4915 temporary errors. However, in the case of such a code being given after
4916 the end of a data transmission (i.e. after ".") Exim was failing to write
4917 a retry record for the message. (Yes, there was some broken host that was
4918 actually sending 8xx at this point.)
4920 73. An unknown lookup type in a host list could cause Exim to panic-die when
4921 the list was checked. (An example that provoked this was putting <; in the
4922 middle of a list instead of at the start.) If this happened during a DATA
4923 ACL check, a -D file could be left lying around. This kind of configuration
4924 error no longer causes Exim to die; instead it causes a defer errror. The
4925 incident is still logged to the main and panic logs.
4927 74. Buglet left over from Exim 3 conversion. The message "too many messages
4928 in one connection" was written to the rejectlog but not the mainlog, except
4929 when address rewriting (yes!) was being logged.
4931 75. Added write_rejectlog option.
4933 76. When a system filter was run not as root (that is, when system_filter_user
4934 was set), the values of the $n variables were not being returned to the
4935 main process; thus, they were not subsequently available in the $sn
4938 77. Added +return_path_on_delivery log selector.
4940 78. A connection timeout was being treated differently from recipients deferred
4941 when testing hosts_max_try with a message that was older than the host's
4942 retry timeout. (The host should not be counted, thus allowing all hosts to
4943 be tried at least once before bouncing.) This may have been the cause of an
4944 occasionally reported bug whereby a message would remain on the queue
4945 longer than the retry timeout, but would be bounced if a delivery was
4946 forced. I say "may" because I never totally pinned down the problem;
4947 setting up timeout/retry tests is difficult. See also the next item.
4949 79. The ultimate address timeout was not being applied to errors that involved
4950 a combination of host plus message (for example, a timeout on a MAIL
4951 command). When an address resolved to a number of possible hosts, and they
4952 were not all tried for each delivery (e.g. because of hosts_max_try), a
4953 message could remain on the queue longer than the retry timeout.
4955 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
4958 81. Fixed an obscure SMTP outgoing bug which required at least the following
4959 conditions: (a) there was another message waiting for the same server;
4960 (b) the server returned 5xx to all RCPT commands in the first message so
4961 that the message was not completed; (c) the server dropped the connection
4962 or gave a negative response to the RSET that Exim sends to abort the
4963 transaction. The observed case was a dropped connection after DATA that had
4964 been sent in pipelining mode. That is, the server had advertised PIPELINING
4965 but was not implementing it correctly. The effect of the bug was incorrect
4966 behaviour, such as trying another host, and this could lead to a crash.
4972 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
4973 and daemon.c were passed as pointers to ints; they should have been
4974 pointers to socklen_t variables (which are typically unsigned ints).
4976 2. Some signed/unsigned type warnings in the os.c file for Linux have been
4979 3. Fixed a really odd bug that affected only the testing scheme; patching a
4980 certain fixed string in the binary changed the value of another string that
4981 happened to be identical to the end of the original first string.
4983 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
4984 name", it returns that address as the IP address. On some operating
4985 systems (e.g. Solaris), it also passes back the IP address string as the
4986 "host name". However, on others (e.g. Linux), it passes back an empty
4987 string. Exim wasn't checking for this, and was changing the host name to an
4988 empty string, assuming it had been canonicized.
4990 5. Although rare, it is permitted to have more than one PTR record for a given
4991 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
4992 all the names associated with an address, because they do in Solaris.
4993 However, it seems that they do not in Linux for data that comes from the
4994 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
4995 I found this out when I moved to a new Linux workstation and tried to run
4996 the Exim test suite.
4998 To get round this problem I have changed the code so that it now does its
4999 own call to the DNS to look up PTR records when searching for a host name.
5000 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
5001 addresses that are only in /etc/hosts are still found.
5003 This behaviour is, however, controlled by an option called host_lookup_
5004 order, which defaults to "bydns:byaddr". If people want to use the other
5005 order, or indeed, just use one or the other means of lookup, they can
5006 specify it in this variable.
5008 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
5009 some operating systems, this comes back from gethostbyaddr() as an empty
5010 string, and this is what Exim used to test for. However, it seems that in
5011 other systems, "." is yielded. Exim now tests for this case too.
5013 7. The values of check_spool_space and check_log_space are now held internally
5014 as a number of kilobytes instead of an absolute number of bytes. If a
5015 numbers is specified without 'K' or 'M', it is rounded up to the nearest
5016 kilobyte. This means that much larger values can be stored.
5018 8. Exim monitor: an attempt to get the action menu when not actually pointing
5019 at a message produces an empty menu entitled "No message selected". This
5020 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
5021 no entries in it ("Shell widget menu has zero width and/or height"). So I
5022 have added a single, blank menu entry in this case.
5024 9. Added ${quote_local_part.
5026 10. MIME decoding is now applied to the contents of Subject: header lines when
5029 11. Now that a reference to $sender_host_address automatically causes a reverse
5030 lookup to occur if necessary (4.13/18), there is no need to arrange for a
5031 host lookup before query-style lookups in lists that might use this
5032 variable. This has therefore been abolished, and the "net-" prefix is no
5033 longer necessary for query-style lookups.
5035 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
5036 have been a typo for LFLAGS, so it has been changed.
5038 13. The install script calls Exim with "-C /dev/null" in order to find the
5039 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
5040 to be output. Howeve, since Exim outputs its version number before the
5041 error, it didn't break the script. It just looked ugly. I fixed this by
5042 always allowing "-C /dev/null" if the caller is root.
5044 14. Ignore overlarge ACL variable number when reading spool file - insurance
5045 against a later release with more variables having written the file.
5047 15. The standard form for an IPv6 address literal was being rejected by EHLO.
5048 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5049 this, as well as the form without the "IPv6" on the front.
5051 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
5052 OS/Makefile-Darwin file.
5054 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
5055 only to LDAP libraries that do not have LDAP_OPT_DEREF.
5057 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
5058 variable. However, some OS use ints for inodes. I've added cast to long int
5059 to get rid of the compiler warning.
5061 19. I had forgotten to lock out "/../" in configuration file names when
5062 ALT_CONFIG_PREFIX was set.
5064 20. Routers used for verification do not need to specify transports. However,
5065 if such a router generated a host list, and callout was configured, Exim
5066 crashed, because it could not find a port number from the (non-existent)
5067 transport. It now assumes port 25 in this circumstance.
5069 21. Added the -t option to exigrep.
5071 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
5072 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
5073 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
5074 them is set, LOOKUP_LSEARCH is forced.
5076 23. "exim -bV" now outputs a list of lookups that are included in the binary.
5078 24. Added sender and host information to the "rejected by local_scan()" log
5079 line; previously there was no indication of these.
5081 25. Added .include_if_exists.
5083 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
5084 Linux. It turns out that not all file systems support this. Apparently some
5085 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
5086 do it.) To cope with this, the error EINVAL, which means that sync-ing is
5087 not supported on the file descriptor, is now ignored when Exim is trying to
5088 sync a directory. This applies only to Linux.
5090 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
5092 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
5093 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
5096 29. Added support for maildirsize files from supplied patch (modified a bit).
5098 30. The use of :fail: followed by an empty string could lead Exim to respond to
5099 sender verification failures with (e.g.):
5101 550 Verification failed for <xxx>
5102 550 Sender verify failed
5104 where the first response line was missing the '-' that indicates it is not
5105 the final line of the response.
5107 31. The loop for finding the name of the user that called Exim had a hardwired
5108 limit of 10; it now uses the value of finduser_retries, which is used for
5109 all other user lookups.
5111 32. Added $received_count variable, available in data and not_smtp ACLs, and at
5114 33. Exim was neglecting to zero errno before one call of strtol() when
5115 expanding a string and expecting an integer value. On some systems this
5116 resulted in spurious "integer overflow" errors. Also, it was casting the
5117 result into an int without checking.
5119 34. Testing for a connection timeout using "timeout_connect" in the retry rules
5120 did not work. The code looks as if it has *never* worked, though it appears
5121 to have been documented since at least releast 1.62. I have made it work.
5123 35. The "timeout_DNS" error in retry rules, also documented since at least
5124 1.62, also never worked. As it isn't clear exactly what this means, and
5125 clearly it isn't a major issue, I have abolished the feature by treating it
5126 as "timeout", and writing a warning to the main and panic logs.
5128 36. The display of retry rules for -brt wasn't always showing the error code
5131 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
5132 timeout_connect_A, timeout_connect_MX.
5134 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
5135 to the empty sender.
5137 39. The daemon was not analysing the content of -oX till after it had closed
5138 stderr and disconnected from the controlling terminal. This meant that any
5139 syntax errors were only noted on the panic log, and the return code from
5140 the command was 0. By re-arranging the code a little, I've made the
5141 decoding happen first, so such errors now appear on stderr, and the return
5142 code is 1. However, the actual setting up of the sockets still happens in
5143 the disconnected process, so errors there are still only recorded on the
5146 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
5147 connections (as happens on some IP stacks) was logged at start up time as
5148 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
5149 it from "IPv6 and IPv4", which means that two separate sockets are being
5152 41. The debug output for gethostbyname2() or getipnodebyname() failures now
5153 says whether AF_INET or AF_INET6 was passed as an argument.
5155 42. Exiwhat output was messed up when time zones were included in log
5158 43. Exiwhat now gives more information about the daemon's listening ports,
5159 and whether -tls-on-connect was used.
5161 44. The "port" option of the smtp transport is now expanded.
5163 45. A "message" modifier in a "warn" statement in a non-message ACL was being
5164 silently ignored. Now an error message is written to the main and panic
5167 46. There's a new ACL modifier called "logwrite" which writes to a log file
5168 as soon as it is encountered.
5170 47. Added $local_user_uid and $local_user_gid at routing time.
5172 48. Exim crashed when trying to verify a sender address that was being
5175 49. Exim was recognizing only a space character after ".include". It now also
5176 recognizes a tab character.
5178 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
5179 extracting the relevant information from the specification. The man page no
5180 longer contains scrambled data for the -d option, and I've added a section
5181 at the front about calling Exim under different names.
5183 51. Added "extra_headers" argument to the "mail" command in filter files.
5185 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
5188 53. Installed eximstats 1.29.
5190 54. Added transport_filter_timeout as a generic transport option.
5192 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
5193 Cc: header lines. This was required by RFC 822, but it not required by RFC
5196 56. Exim used to add From:, Date:, and Message-Id: header lines to any
5197 incoming messages that did not have them. Now it does so only if the
5198 message originates locally, that is, if there is no associated remote host
5199 address. When Resent- header lines are present, this applies to the Resent-
5200 lines rather than the non-Resent- lines.
5202 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
5203 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
5205 58. Messages for configuration errors now include the name of the main
5206 configuration file - useful now that there may be more than one file in a
5207 list (.included file names were always shown).
5209 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
5210 for those rare installations that do not start the daemon as root or run it
5211 setuid root. I've cut out the call to initgroups() if the daemon is not
5214 60. The Exim user and group can now be bound into the binary as text strings
5215 that are looked up at the start of Exim's processing.
5217 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
5219 62. Added $mailstore_basename variable.
5221 63. Installed patch to sieve.c from Michael Haardt.
5223 64. When Exim failed to open the panic log after failing to open the main log,
5224 the original message it was trying to log was written to stderr and debug
5225 output, but if they were not available (the usual case in production), it
5226 was lost. Now it is written to syslog before the two lines that record the
5227 failures to open the logs.
5229 65. Users' Exim filters run in subprocesses under the user's uid. It is
5230 possible for a "deliver" command or an alias in a "personal" command to
5231 provoke an address rewrite. If logging of address rewriting is configured,
5232 this fails because the process is not running as root or exim. There may be
5233 a better way of dealing with this, but for the moment (because 4.30 needs
5234 to be released), I have disabled address rewrite logging when running a
5235 filter in a non-root, non-exim process.
5241 1. The buildconfig auxiliary program wasn't quoting the value set for
5242 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
5243 not defined. This bug was masked in 4.22 by the effect that was fixed in
5246 2. Some messages that were rejected after a message id was allocated were
5247 shown as "incomplete" by exigrep. It no longer does this for messages that
5248 are rejected by local_scan() or the DATA or non-SMTP ACLs.
5250 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
5251 have allow_domain_literals set, the ID did not get logged in the <= line.
5252 Domain literals are now always recognized in Message-ID: header lines.
5254 4. The first argument for a ${extract expansion item is the key name or field
5255 number. Leading and trailing spaces in this item were not being ignored,
5256 causing some misleading effects.
5258 5. When deliver_drop_privilege was set, single queue runner processes started
5259 manually (i.e. by the command "exim -q") or by the daemon (which uses the
5260 same command in the process it spins off) were not dropping privilege.
5262 6. When the daemon running as "exim" started a queue runner, it always
5263 re-executed Exim in the spun-off process. This is a waste of effort when
5264 deliver_drop_privilege is set. The new process now just calls the
5265 queue-runner function directly.
5271 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
5274 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
5275 ignored. Though the use of -oP was forcing the writing of a pid file, it
5276 was always written to the default place.
5278 3. If the message "no IP address found for host xxxx" is generated during
5279 incoming verification, it is now followed by identification of the incoming
5280 connection (so you can more easily find what provoked it).
5282 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
5284 5. Added some features to "harden" Exim a bit more against certain attacks:
5286 (a) There is now a build-time option called FIXED_NEVER_USERS that can
5287 be put in Local/Makefile. This is like the never_users runtime option,
5288 but it cannot be overridden. The default setting is "root".
5290 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
5291 prefix string with which any file named in a -C command line option
5294 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
5295 is retained for -C and -D only if the caller of Exim is root. Without
5296 it, the exim user may also use -C and -D and retain privilege.
5298 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
5299 command line option is disabled.
5301 6. Macro names set by the -D option must start with an upper case letter, just
5302 like macro names defined in the configuration file.
5304 7. Added "dereference=" facility to LDAP.
5306 8. Two instances of the typo "uknown" in the source files are fixed.
5308 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
5309 the Configure-Makefile script screwed up while processing it.
5311 10. Incorporated PCRE 4.4.
5313 11. The SMTP synchronization check was not operating right at the start of an
5314 SMTP session. For example, it could not catch a HELO sent before the client
5315 waited for the greeting. There is now a check for outstanding input at the
5316 point when the greeting is written. Because of the duplex, asynchronous
5317 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
5318 way, but not yet received, when the check is performed.
5320 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
5321 on TCP/IP sockets, because this apparently causes some broken clients to
5324 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
5325 unchanged) from the Cygwin maintainer.
5327 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
5328 was supported instead of testing for mbx. Effectively a typo.
5330 15. The spa authenticator server code was not checking that the input it
5331 received was valid base64.
5333 16. The debug output line for the "set" modifier in ACLs was not showing the
5334 name of the variable that was being set.
5336 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
5338 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
5339 lookup on its own. Something else was needed to trigger the lookup. For
5340 example, a match in host_lookup or the need for a host name in a host list.
5341 Now, if $sender_host_name is referenced and the host name has not yet been
5342 looked up, a lookup is performed. If the lookup fails, the variable remains
5343 empty, and $host_lookup_failed is set to "1".
5345 19. Added "eqi" as a case-independent comparison operator.
5347 20. The saslauthd authentication condition could segfault if neither service
5348 nor realm was specified.
5350 21. If an overflowing value such as "2048M" was set for message_size_limit, the
5351 error message that was logged was misleading, and incoming SMTP
5352 connections were dropped. The message is now more accurate, and temporary
5353 errors are given to SMTP connections.
5355 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
5356 (except RSET) with a 421 error, until QUIT is received. However, it was
5357 failing to send a response to QUIT.
5359 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
5360 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
5361 ACL is now run after the helo_try_verify_hosts code.
5363 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
5364 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
5365 case-independent, so other case variants are also recognized). Apparently
5366 some systems use these upper case variants.
5368 25. If more than two messages were waiting for the same host, and a transport
5369 filter was specified for the transport, Exim sent two messages over the
5370 same TCP/IP connection, and then failed with "socket operation on non-
5371 socket" when it tried to send the third.
5373 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
5375 27. The extern definition of crypt16() in expand.c was not being excluded when
5376 the OS had its own crypt16() function.
5378 28. Added bounce_return_body as a new option, and bounce_return_size_limit
5379 as a preferred synonym for return_size_limit, both as an option and as an
5382 29. Added LIBS=-liconv to OS/Makefile-OSF1.
5384 30. Changed the default configuration ACL to relax the local part checking rule
5385 for addresses that are not in any local domains. For these addresses,
5386 slashes and pipe symbols are allowed within local parts, but the sequence
5387 /../ is explicitly forbidden.
5389 31. SPA server authentication was not clearing the challenge buffer before
5392 32. log_message in a "warn" ACL statement was writing to the reject log as
5393 well as to the main log, which contradicts the documentation and doesn't
5394 seem right (because no rejection is happening). So I have stopped it.
5396 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
5397 However, I am unable to do any testing of this.
5399 34. Fixed an infelicity in the appendfile transport. When checking directories
5400 for a mailbox, to see if any needed to be created, it was accidentally
5401 using path names with one or more superfluous leading slashes; tracing
5402 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
5404 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
5405 contents are added to the log line that is written for every discarded
5406 recipient. (Previously a log_message setting was ignored.)
5408 36. The ${quote: operator now quotes the string if it is empty.
5410 37. The install script runs exim in order to find its version number. If for
5411 some reason other than non-existence or emptiness, which it checks, it
5412 could not run './exim', it was installing it with an empty version number,
5413 i.e. as "exim-". This error state is now caught, and the installation is
5416 38. An argument was missing from the function that creates an error message
5417 when Exim fails to connect to the socket for saslauthd authentication.
5418 This could cause Exim to crash, or give a corrupted message.
5420 39. Added isip, isip4, and isip6 to ${if conditions.
5422 40. The ACL variables $acl_xx are now saved with the message, and can be
5423 accessed later in routers, transports, and filters.
5425 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
5426 strings in the file are not string-expanded.
5428 42. If a MAIL command specified a SIZE value that was too large to fit into an
5429 int variable, the check against message_size_limit failed. Such values are
5430 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
5431 day this will have to be increased, but I don't think I want to be around
5432 when emails are that large.
5439 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
5440 iconv() is not standard in FreeBSD.
5442 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
5443 IPv6 enabled. The observed symptom was a segmentation fault on return from
5444 the function os_common_find_running_interfaces() in src/os.c.
5446 3. In the check_special_case() function in daemon.c I had used "errno" as an
5447 argument name, which causes warnings on some systems. This was basically a
5448 typo, since it was named "eno" in the comments!
5450 4. The code that waits for the clock to tick (at a resolution of some fraction
5451 of a second) so as to ensure message-id uniqueness was always waiting for
5452 at least one whole tick, when it could have waited for less. [This is
5453 almost certainly not relevant at current processor speeds, where it is
5454 unlikely to ever wait at all. But we try to future-proof.]
5456 5. The function that sleeps for a time interval that includes fractions of a
5457 second contained a race. It did not block SIGALRM between setting the
5458 timer, and suspending (a couple of lines later). If the interval was short
5459 and the sigsuspend() was delayed until after it had expired, the suspension
5460 never ended. On busy systems this could lead to processes getting stuck for
5463 6. Some uncommon configurations may cause a lookup to happen in a queue runner
5464 process, before it forks any delivery processes. The open lookup caching
5465 mechanism meant that the open file or database connection was passed into
5466 the delivery process. The problem was that delivery processes always tidy
5467 up cached lookup data. This could cause a problem for the next delivery
5468 process started by the queue runner, because the external queue runner
5469 process does not know about the closure. So the next delivery process
5470 still has data in the lookup cache. In the case of a file lookup, there was
5471 no problem because closing a file descriptor in a subprocess doesn't affect
5472 the parent. However, if the lookup was caching a connection to a database,
5473 the connection was closed, and the second delivery process was likely to
5474 see errors such as "PGSQL: query failed: server closed the connection
5475 unexpectedly". The problem has been fixed by closing all cached lookups
5476 in a queue runner before running a delivery process.
5478 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
5479 seem to have the "const" qualifier which it has on other OS. I've
5482 8. Change 4.21/2 was too strict. It is only if there are two authenticators
5483 *of the same type* (client or server) with the same public name that an
5484 error should be diagnosed.
5486 9. When Exim looked up a host name for an IP address, but failed to find the
5487 original IP address when looking up the host name (a safety check), it
5488 output the message "<ip address> does not match any IP for NULL", which was
5489 confusing, to say the least. The bug was that the host name should have
5490 appeared instead of "NULL".
5492 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
5493 user that is built into the binary, and the -C or -D options is used, root
5494 privilege is dropped before the configuration file is read. In addition,
5495 logging is switched to stderr instead of the normal log files. If the
5496 configuration then re-defines the Exim user, the unprivileged environment
5497 is probably not what is expected, so Exim logs a panic warning message (but
5500 However, if deliver_drop_privilege is set, the unprivileged state may well
5501 be exactly what is intended, so the warning has been cut out in that case,
5502 and Exim is allowed to try to write to its normal log files.
5508 1. smtp_return_error_details was not giving details for temporary sender
5509 or receiver verification errors.
5511 2. Diagnose a configuration error if two authenticators have the same public
5514 3. Exim used not to create the message log file for a message until the first
5515 delivery attempt. This could be confusing when incoming messages were held
5516 for policy or load reasons. The message log file is now created at the time
5517 the message is received, and an initial "Received" line is written to it.
5519 4. The automatically generated man page for command line options had a minor
5520 bug that caused no ill effects; however, a more serious problem was that
5521 the procedure for building the man page automatically didn't always
5522 operate. Consequently, release 4.20 contains an out-of-date version. This
5523 shouldn't happen again.
5525 5. When building Exim with embedded Perl support, the script that builds the
5526 Makefile was calling 'perl' to find its compile-time parameters, ignoring
5527 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
5529 6. The freeze_tell option was not being used for messages that were frozen on
5530 arrival, either by an ACL or by local_scan().
5532 7. Added the smtp_incomplete_transaction log selector.
5534 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
5535 was accepting AUTH without a new EHLO.
5537 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
5538 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
5541 10. Logging of TCP/IP connections (when configured) now happens in the main
5542 daemon process instead of the child process, so that the TCP/IP connection
5543 count is more accurate (but it can never be perfect).
5545 11. The use of "drop" in a nested ACL was not being handled correctly in the
5546 outer ACL. Now, if condition failure induced by the nested "drop" causes
5547 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
5548 or "require"), the connection is dropped.
5550 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
5551 that yield "discard" can now be used with an "accept" or a "discard" verb,
5552 but an error is generated for any others (because I can't see a useful way
5553 to define what should happen).
5555 13. When an ACL is read dynamically from a file (or anywhere else), the lines
5556 are now processed in the same way as lines in the Exim configuration file.
5557 In particular, continuation lines are supported.
5559 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
5561 15. Added -ti meaning -t -i.
5563 16. Check for letters, digits, hyphens, and dots in the names of dnslist
5564 domains, and warn by logging if others are found.
5566 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
5567 returned from GIFCONF when Exim is trying to find the list of interfaces on
5568 a host. The code in os.c has been modified to copy each ifreq to an aligned
5569 structure in all cases.
5571 Also, in some cases, the returned ifreq's were being copied to a 'struct
5572 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
5573 means the last couple of bytes of an IPv6 address could be chopped if the
5574 ifreq contained only a normal sockaddr (14 bytes storage).
5576 18. Named domain lists were not supported in the hosts_treat_as_local option.
5577 An entry such as +xxxx was not recognized, and was treated as a literal
5580 19. Ensure that header lines added by a DATA ACL are included in the reject log
5581 if the ACL subsequently rejects the message.
5583 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
5584 MD5 (which is deprecated).
5586 21. When testing a filter file using -bf, Exim was writing a message when it
5587 took the sender from a "From " line in the message, but it was not doing so
5588 when it took $return_path from a Return-Path: header line. It now does.
5590 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
5591 with a valid header line field name (a series of printing characters
5592 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
5594 23. Changed "disc" in the source to "disk" to conform to the documentation and
5595 the book and for uniformity.
5597 24. Ignore Sendmail's -Ooption=value command line item.
5599 25. When execve() failed while trying to run a command in a pipe transport,
5600 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
5601 could be confused with a return value of 69 from the command itself. This
5602 has been changed to 127, the value the shell returns if it is asked to run
5603 a non-existent command. The wording for the related log line suggests a
5604 non-existent command as the problem.
5606 26. If received_header_text expands to an empty string, do not add a Received:
5607 header line to the message. (Well, it adds a token one on the spool, but
5608 marks it "old" so that it doesn't get used or transmitted.)
5610 27. Installed eximstats 1.28 (addition of -nt option).
5612 28. There was no check for failure on the call to getsockname() in the daemon
5613 code. This can fail if there is a shortage of resources on the system, with
5614 ENOMEM, for example. A temporary error is now given on failure.
5616 29. Contrary to the C standard, it seems that in some environments, the
5617 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
5618 program. Exim now does this explicitly; it affects the formatting of
5619 timestamps using strftime().
5621 30. If exiqsumm was given junk data, it threw up some uninitialized variable
5622 complaints. I've now initialized all the variables, to avoid this.
5624 32. Header lines added by a system filter were not being "seen" during
5625 transport-time rewrites.
5627 33. The info_callback() function passed to OpenSSL is set up with type void
5628 (*)(SSL *, int, int), as described somewhere. However, when calling the
5629 function (actually a macro) that sets it up, the type void(*)() is
5630 expected. I've put in a cast to prevent warnings from picky compilers.
5632 34. If a DNS black list lookup found a CNAME record, but there were no A
5633 records associated with the domain it pointed at, Exim crashed.
5635 35. If a DNS black list lookup returned more than one A record, Exim ignored
5636 all but the first. It now scans all returned addresses if a particular IP
5637 value is being sought. In this situation, the contents of the
5638 $dnslist_value variable are a list of all the addresses, separated by a
5641 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
5642 to accept a host name and all its aliases if the forward lookup for any of
5643 them yielded the IP address of the incoming connection. Now it accepts only
5644 those names whose forward lookup yields the correct IP address. Any other
5645 names are discarded. This closes a loophole whereby a rogue DNS
5646 administrator could create reverse DNS records to break through a
5647 wildcarded host restriction in an ACL.
5649 37. If a user filter or a system filter that ran in a subprocess used any of
5650 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
5651 the wrong values were passed to the pipe command ($thisaddress had the
5652 value of $0, $0 had the value of $1, etc). This bug was introduced by
5653 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
5655 38. Improved the line breaking for long SMTP error messages from ACLs.
5656 Previously, if there was no break point between 40 and 75 characters, Exim
5657 left the rest of the message alone. Two changes have been made: (a) I've
5658 reduced the minimum length to 35 characters; (b) if it can't find a break
5659 point between 35 and 75 characters, it looks ahead and uses the first one
5660 that it finds. This may give the occasional overlong line, but at least the
5661 remaining text gets split now.
5663 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
5664 file descriptors might be low, and that setting 1000 would always raise it.
5665 It turns out that in some environments, the limit is already over 1000 and
5666 that lowering it causes trouble. So now Exim takes care not to decrease it.
5668 40. When delivering a message, the value of $return_path is set to $sender_
5669 address at the start of routing (routers may change the value). By an
5670 oversight, this default was not being set up when an address was tested by
5671 -bt or -bv, which affected the outcome if any router or filter referred to
5674 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
5675 log only when "message" or "log_message" are set. However, if one of the
5676 conditions was an address verification, or a call to a nested ACL, the
5677 messages generated by the underlying test were being passed through. This
5678 no longer happens. The underlying message is available in $acl_verify_
5679 message for both "message" and "log_message" expansions, so it can be
5680 passed through if needed.
5682 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
5683 new expansion $bh_ to give the encoded byte string without charset
5684 translation. Translation happens only if iconv() is available; HAVE_ICONV
5685 indicates this at build time. HEADERS_CHARSET gives the charset to
5686 translate to; headers_charset can change it in the configuration, and
5687 "headers charset" can change it in an individual filter file.
5689 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
5690 that creates RFC 2047 encoded "words" labels them as that charset instead
5691 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
5692 expansion operator; (ii) when Exim creates a From: line for a local
5693 message; (iii) when a header line is rewritten to include a "phrase" part.
5695 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
5696 buggy, causing it to skip the first lines of messages whose message ID
5697 ended in 'D'. This would not have bitten before Exim release 4.14, because
5698 message IDs were unlikely to end in 'D' before then. The effect was to have
5699 incorrect size information for certain domains.
5701 45. #include "config.h" was missing at the start of the crypt16.c module. This
5702 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
5705 46. If there was a timeout during a "random" callout check, Exim treated it as
5706 a failure of the random address, and carried on sending RSET and the real
5707 address. If the delay was just some slowness somewhere, the response to the
5708 original RCPT would be taken as a response to RSET and so on, causing
5709 mayhem of various kinds.
5711 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
5712 when I implemented it. It didn't allow for the fact that some option values
5713 may legitimatetly be negative (e.g. size_addition), and it didn't even do
5714 the right test for positive values.
5716 48. Domain names in DNS records are case-independent. Exim always looks them up
5717 in lower case. Some resolvers return domain names in exactly the case they
5718 appear in the zone file, that is, they may contain uppercase letters. Not
5719 all resolvers do this - some return always lower case. Exim was treating a
5720 change of case by a resolver as a change of domain, similar to a widening
5721 of a domain abbreviation. This triggered its re-routing code and so it was
5722 trying to route what was effectively the same domain again. This normally
5723 caused routing to fail (because the router wouldn't handle the domain
5724 twice). Now Exim checks for this case specially, and just changes the
5725 casing of the domain that it ultimately uses when it transmits the message
5728 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
5731 50. If a filter generated a file delivery with a non-absolute name (possible if
5732 no home directory exists for the router), the forbid_file option was not
5735 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
5736 the existing equality matching.
5738 52. Exim was using ints instead of ino_t variables in some places where it was
5739 dealing with inode numbers.
5741 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
5742 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
5743 called TMPDIR, and if it finds it is different, it changes its value.
5745 54. The smtp_printf() function is now made available to local_scan() so
5746 additional output lines can be written before returning. There is also an
5747 smtp_fflush() function to enable the detection of a dropped connection.
5748 The variables smtp_input and smtp_batched_input are exported to
5751 55. Changed the default runtime configuration: the message "Unknown user"
5752 has been removed from the ACL, and instead placed on the localuser router,
5753 using the cannot_route_message feature. This means that any verification
5754 failures that generate their own messages won't get overridden. Similarly,
5755 the "Unrouteable address" message that was in the ACL for unverifiable
5756 relay addresses has also been removed.
5758 56. Added hosts_avoid_esmtp to the smtp transport.
5760 57. The exicyclog script was not checking for the esoteric option
5761 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
5762 will work only if exicyclog is run under the appropriate euid.
5764 58. Following a discussion on the list, the rules by which Exim recognises line
5765 endings on incoming messages have been changed. The -dropcr and drop_cr
5766 options are now no-ops, retained only for backwards compatibility. The
5767 following line terminators are recognized: LF CRLF CR. However, special
5768 processing applies to CR:
5770 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
5771 nor a local message in the state where . is a terminator.
5773 (ii) If a bare CR is encountered in a header line, an extra space is added
5774 after the line terminator so as not to end the header. The reasoning
5775 behind this is that bare CRs in header lines are most likely either
5776 to be mistakes, or people trying to play silly games.
5778 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
5779 was being incorrectly given as 18 bytes larger than it should have been.
5780 This is a VOB (very old bug).
5782 60. This may never have affected anything current, but just in case it has:
5783 When the local host is found other than at the start of a list of hosts,
5784 the local host, those with the same MX, and any that follow, are discarded.
5785 When the list in question was part of a longer list of hosts, the following
5786 hosts (not currently being processed) were also being discarded. This no
5787 longer happens. I'm not sure if this situation could ever has previously
5790 61. Added the "/MX" feature to lists of hosts in the manualroute and query
5793 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
5794 header. This is something that is recommended in a new Internet Draft, and
5795 is something that is documented as being done by Sendmail. There are two
5796 possible values. For messages generated by the autoreply transport, Exim
5799 Auto-Submitted: auto-replied
5801 whereas for all other generated messages (e.g. bounces) it adds
5803 Auto-Submitted: auto-generated
5805 63. The "personal" condition in filters now includes a test for the
5806 Auto-Submitted: header. If it contains the string "auto-" the message it
5807 not considered personal.
5809 64. Added rcpt_include_affixes as a generic transport option.
5811 65. Added queue_only_override (default true).
5813 66. Added the syslog_duplication option.
5815 67. If what should have been the first header line of a message consisted of
5816 a space followed by a colon, Exim was mis-interpreting it as a header line.
5817 It isn't of course - it is syntactically invalid and should therefore be
5818 treated as the start of the message body. The misbehaviour could have
5819 caused a number of strange effects, including loss of data in subsequent
5820 header lines, and spool format errors.
5822 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
5823 client host had authenticated. This control can now be exercised by an ACL
5824 for more flexibility.
5826 69. By default, callouts do not happen when testing with -bh. There is now a
5827 variant, -bhc, which does actually run the callout code, including
5828 consulting and updating the callout cache.
5830 70. Added support for saslauthd authentication, courtesy of Alexander
5833 71. If statvfs() failed on the spool or log directories while checking their
5834 size for availability, Exim confusingly gave the error "space shortage".
5835 Furthermore, in debugging mode it crashed with a floating point exception.
5836 These checks are done if check_{spool,log}_{space,inodes} are set, and when
5837 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
5838 serious problem, Exim now writes to the main and panic logs when this
5839 happens, with details of the failure. It then refuses to accept the
5840 incoming message, giving the message "spool directory problem" or "log
5841 directory problem" with a 421 code for SMTP messages.
5843 72. When Exim is about to re-exec itself, it ensures that the file descriptors
5844 0, 1, and 2 exist, because some OS complain for execs without them (see
5845 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
5846 descriptors. However, the code omitted to check that the open succeeded,
5847 causing mysterious errors if for some reason the permissions on /dev/null
5848 got screwed. Now Exim writes a message to the main and panic logs, and
5849 bombs out if it can't open /dev/null.
5851 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
5852 interact so that it is all more flexible. It is supposed to remain
5853 backwards compatible. Also added extra_local_interfaces.
5855 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
5856 to bomb out with an assertion failure - to the client this appears as a
5857 connection drop. This problem occurs in the part of the code that was taken
5858 from the Samba project. Fortunately, the assertion is in a very simple
5859 function, so I have fixed this by reproducing the function inline in the
5860 one place where it is called, and arranging for authentication to fail
5861 instead of killing the process with assert().
5863 75. The SPA client code was not working when the server requested OEM rather
5864 than Unicode encoding.
5866 76. Added code to make require_files with a specific uid setting more usable in
5867 the case where statting the file as root fails - usually a non-root-mounted
5868 NFS file system. When this happens and the failure is EACCES, Exim now
5869 forks a subprocess and does the per-uid checking as the relevant uid.
5871 77. Added process_log_path.
5873 78. If log_file_path was not explicitly set, a setting of check_log_space or
5874 check_log_inodes was ignored.
5876 79. If a space check for the spool or log partitions fails, the incident is now
5877 logged. Of course, in the latter case the data may get lost...
5879 80. Added the %p formatting code to string_format() so that it can be used to
5880 print addresses in debug_print(). Adjusted all the address printing in the
5881 debugging in store.c to use %p rather than %d.
5883 81. There was a concern that a line of code in smtp_in.c could overflow a
5884 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
5885 initially expressed, the concern was not well-founded, because trailing
5886 spaces are removed early. However, if the trailing spaces were followed by
5887 a NULL, they did not get removed, so the overflow was possible. Two fixes
5890 (a) I re-wrote the offending code in a cleaner fashion.
5891 (b) If an incoming SMTP command contains a NULL character, it is rejected
5894 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
5895 runs initgroups(), so that if the Exim user is in any additional groups,
5896 they will be used during message reception.
5902 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob