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 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
49 JH/02 Support "G" suffix to numbers in ${if comparisons.
51 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
53 NM/01 Bugzilla 1197 - Spec typo
54 Bugzilla 1196 - Spec examples corrections
56 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
58 PP/09 Add gnutls_enable_pkcs11 option.
60 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
61 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
63 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
64 "acl {{name}{arg}...}", and optional args on acl condition
67 JH/05 Permit multiple router/transport headers_add/remove lines.
69 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
71 JH/07 Avoid using a waiting database for a single-message-only transport.
72 Performance patch from Paul Fisher. Bugzilla 1262.
74 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
77 JH/09 Add $headers_added variable, with content from use of ACL modifier
78 add_header (but not yet added to the message). Bugzilla 199.
80 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
81 Pulled from Bugzilla 817 by Wolfgang Breyha.
83 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
86 JH/11 Add A= logging on delivery lines, and a client_set_id option on
89 JH/12 Add optional authenticated_sender logging to A= and a log_selector
92 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
94 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
95 advertise SMTP AUTH mechanism to us, instead of a generic
96 protocol violation error. Also, make Exim more robust to bad
97 data from the Dovecot auth socket.
99 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
101 When a queue runner is handling a message, Exim first routes the
102 recipient addresses, during which it prunes them based on the retry
103 hints database. After that it attempts to deliver the message to
104 any remaining recipients. It then updates the hints database using
107 So if a recipient address works intermittently, it can get repeatedly
108 deferred at routing time. The retry hints record remains fresh so the
109 address never reaches the final cutoff time.
111 This is a fairly common occurrence when a user is bumping up against
112 their storage quota. Exim had some logic in its local delivery code
113 to deal with this. However it did not apply to per-recipient defers
114 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
116 This change adds a proper retry rule check during routing so that the
117 final cutoff time is checked against the message's age. We only do
118 this check if there is an address retry record and there is not a
119 domain retry record; this implies that previous attempts to handle
120 the address had the retry_use_local_parts option turned on. We use
121 this as an approximation for the destination being like a local
122 delivery, as in LMTP.
124 I suspect this new check makes the old local delivery cutoff check
125 redundant, but I have not verified this so I left the code in place.
127 TF/02 Correct gecos expansion when From: is a prefix of the username.
129 Test 0254 submits a message to Exim with the header
133 When I ran the test suite under the user fanf2, Exim expanded
134 the header to contain my full name, whereas it should have added
135 a Resent-Sender: header. It erroneously treats any prefix of the
136 username as equal to the username.
138 This change corrects that bug.
140 GF/01 DCC debug and logging tidyup
141 Error conditions log to paniclog rather than rejectlog.
142 Debug lines prefixed by "DCC: " to remove any ambiguity.
144 TF/03 Avoid unnecessary rebuilds of lookup-related code.
146 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
147 Bug spotted by Jeremy Harris; was flawed since initial commit.
148 Would have resulted in OCSP responses post-SNI triggering an Exim
149 NULL dereference and crash.
151 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
153 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
154 Bug detection, analysis and fix by Samuel Thibault.
155 Bugzilla 1331, Debian bug #698092.
161 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
163 This, or similar/improved, will also be change PP/11 of 4.82.
169 PP/01 Handle short writes when writing local log-files.
170 In practice, only affects FreeBSD (8 onwards).
171 Bugzilla 1053, with thanks to Dmitry Isaikin.
173 NM/01 Bugzilla 949 - Documentation tweak
175 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
178 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
180 PP/02 Implemented gsasl authenticator.
182 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
184 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
185 `pkg-config foo` for cflags/libs.
187 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
188 with rest of GSASL and with heimdal_gssapi.
190 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
191 `pkg-config foo` for cflags/libs for the TLS implementation.
193 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
194 properties get this fed in as external SSF. A number of robustness
195 and debugging improvements to the cyrus_sasl authenticator.
197 PP/08 cyrus_sasl server now expands the server_realm option.
199 PP/09 Bugzilla 1214 - Log authentication information in reject log.
200 Patch by Jeremy Harris.
202 PP/10 Added dbmjz lookup type.
204 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
206 PP/12 MAIL args handles TAB as well as SP, for better interop with
207 non-compliant senders.
208 Analysis and variant patch by Todd Lyons.
210 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
211 Bug report from Lars Müller <lars@samba.org> (via SUSE),
212 Patch from Dirk Mueller <dmueller@suse.com>
214 PP/13 tls_peerdn now print-escaped for spool files.
215 Observed some $tls_peerdn in wild which contained \n, which resulted
216 in spool file corruption.
218 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
219 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
220 or write after TLS renegotiation, which otherwise led to messages
223 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
224 as a tracking header (ie: a signed header comes before the signature).
225 Patch from Wolfgang Breyha.
227 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
228 comma-sep list; embedded commas doubled.
230 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
232 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
234 Report and patch from Dmitry Banschikov.
236 PP/16 Removed "dont_insert_empty_fragments" fron "openssl_options".
237 Removed SSL_clear() after SSL_new() which led to protocol negotiation
238 failures. We appear to now support TLS1.1+ with Exim.
240 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
241 lets Exim select keys and certificates based upon TLS SNI from client.
242 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
243 before an outbound SMTP session. New log_selector, +tls_sni.
245 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
246 NULL dereference. Report and patch from Alun Jones.
248 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
249 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
250 Not seeing resolver debug output on NetBSD, but suspect this is a
251 resolver implementation change.
253 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
254 Left warnings. Added "eximon gdb" invocation mode.
256 PP/21 Defaulting "accept_8bitmime" to true, not false.
258 PP/22 Added -bw for inetd wait mode support.
260 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
261 locate the relevant includes and libraries. Made this the default.
263 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
264 Bugzilla 1246, report and most of solution from Tomasz Kusy.
266 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
267 This may cause build issues on older platforms.
269 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
270 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
271 gnutls_require_mac & gnutls_require_protocols (no longer supported).
272 Added SNI support via GnuTLS too.
273 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
275 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
277 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
278 Applied second patch from Janne, implementing suggestion to default
279 multiple-strings-in-record handling to match SPF spec.
281 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
283 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
284 read-only, out of scope).
285 Patch from Wolfgang Breyha, report from Stuart Northfield.
287 PP/29 Fix three issues highlighted by clang analyser static analysis.
288 Only crash-plausible issue would require the Cambridge-specific
289 iplookup router and a misconfiguration.
290 Report from Marcin Mirosław.
292 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
294 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
295 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
296 As part of this, removing so much warning spew let me fix some minor
297 real issues in debug logging.
299 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
300 assignment on my part. Fixed.
302 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
303 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
304 Janne Snabb (who went above and beyond: thank you).
306 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
307 string otherwise requires a connection and a bunch more work and it's
308 relatively easy to get wrong. Should also expose TLS library linkage
311 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
312 64-bit ${eval} (JH/03).
314 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
315 GNU libc to support some of the 64-bit stuff, should not lead to
316 conflicts. Defined before os.h is pulled in, so if a given platform
317 needs to override this, it can.
319 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
320 protection layer was required, which is not implemented.
321 Bugzilla 1254, patch from Wolfgang Breyha.
323 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
324 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
325 tls_dhparam take prime identifiers. Also unbreak combination of
326 OpenSSL+DH_params+TLSSNI.
328 PP/39 Disable SSLv2 by default in OpenSSL support.
334 PP/01 Solaris build fix for Oracle's LDAP libraries.
335 Bugzilla 1109, patch from Stephen Usher.
337 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
339 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
342 TF/02 Fix a couple more cases where we did not log the error message
343 when unlink() failed. See also change 4.74-TF/03.
345 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
346 lock up or crash if it happened to be inside a call to libc when it
347 got a SIGUSR1 from exiwhat.
349 The SIGUSR1 handler appends the current process status to the process
350 log which is later printed by exiwhat. It used to use the general
351 purpose logging code to do this, but several functions it calls are
352 not safe for signals.
354 The new output code in the SIGUSR1 handler is specific to the process
355 log, and simple enough that it's easy to inspect for signal safety.
356 Removing some special cases also simplifies the general logging code.
357 Removing the spurious timestamps from the process log simplifies
360 TF/04 Improved ratelimit ACL condition.
362 The /noupdate option has been deprecated in favour of /readonly which
363 has clearer semantics. The /leaky, /strict, and /readonly update modes
364 are mutually exclusive. The update mode is no longer included in the
365 database key; it just determines when the database is updated. (This
366 means that when you upgrde Exim will forget old rate measurements.)
368 Exim now checks that the per_* options are used with an update mode that
369 makes sense for the current ACL. For example, when Exim is processing a
370 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
371 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
372 must specify per_mail/readonly. If you omit the update mode it defaults to
373 /leaky where that makes sense (as before) or /readonly where required.
375 The /noupdate option is now undocumented but still supported for
376 backwards compatibility. It is equivalent to /readonly except that in
377 ACLs where /readonly is required you may specify /leaky/noupdate or
378 /strict/noupdate which are treated the same as /readonly.
380 A useful new feature is the /count= option. This is a generalization
381 of the per_byte option, so that you can measure the throughput of other
382 aggregate values. For example, the per_byte option is now equivalent
383 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
385 The per_rcpt option has been generalized using the /count= mechanism
386 (though it's more complicated than the per_byte equivalence). When it is
387 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
388 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
389 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
390 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
391 Note that using per_rcpt with a non-readonly update mode in more than
392 one ACL will cause the recipients to be double-counted. (The per_mail
393 and per_byte options don't have this problem.)
395 The handling of very low rates has changed slightly. If the computed rate
396 is less than the event's count (usually one) then this event is the first
397 after a long gap. In this case the rate is set to the same as this event's
398 count, so that the first message of a spam run is counted properly.
400 The major new feature is a mechanism for counting the rate of unique
401 events. The new per_addr option counts the number of different
402 recipients that someone has sent messages to in the last time period. It
403 behaves like per_rcpt if all the recipient addresses are different, but
404 duplicate recipient addresses do not increase the measured rate. Like
405 the /count= option this is a general mechanism, so the per_addr option
406 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
407 example, measure the rate that a client uses different sender addresses
408 with the options per_mail/unique=$sender_address. There are further
409 details in the main documentation.
411 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
413 TF/06 Removed a few PCRE remnants.
415 TF/07 Automatically extract Exim's version number from tags in the git
416 repository when doing development or release builds.
418 PP/02 Raise smtp_cmd_buffer_size to 16kB.
419 Bugzilla 879. Patch from Paul Fisher.
421 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
422 Heavily based on revision 40f9a89a from Simon Arlott's tree.
425 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
427 PP/05 Variable $av_failed, true if the AV scanner deferred.
428 Bugzilla 1078. Patch from John Horne.
430 PP/06 Stop make process more reliably on build failure.
431 Bugzilla 1087. Patch from Heiko Schlittermann.
433 PP/07 Make maildir_use_size_file an _expandable_ boolean.
434 Bugzilla 1089. Patch from Heiko Schlittermann.
436 PP/08 Handle ${run} returning more data than OS pipe buffer size.
437 Bugzilla 1131. Patch from Holger Weiß.
439 PP/09 Handle IPv6 addresses with SPF.
440 Bugzilla 860. Patch from Wolfgang Breyha.
442 PP/10 GnuTLS: support TLS 1.2 & 1.1.
444 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
447 PP/11 match_* no longer expand right-hand-side by default.
448 New compile-time build option, EXPAND_LISTMATCH_RHS.
449 New expansion conditions, "inlist", "inlisti".
451 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
453 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
455 PP/14 fix log_write() format string regression from TF/03.
456 Bugzilla 1152. Patch from Dmitry Isaikin.
462 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
464 PP/02 Harmonised TLS library version reporting; only show if debugging.
465 Layout now matches that introduced for other libraries in 4.74 PP/03.
467 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
469 PP/04 New "dns_use_edns0" global option.
471 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
474 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
475 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
477 TK/01 Updated PolarSSL code to 0.14.2.
478 Bugzilla 1097. Patch from Andreas Metzler.
480 PP/07 Catch divide-by-zero in ${eval:...}.
483 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
486 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
487 format-string attack -- SECURITY: remote arbitrary code execution.
489 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
490 time unintentionally subject to list matching rules, letting the header
491 cause arbitrary Exim lookups (of items which can occur in lists, *not*
492 arbitrary string expansion). This allowed for information disclosure.
494 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
495 INT_MIN/-1 -- value coerced to INT_MAX.
501 NM/01 Workround for PCRE version dependency in version reporting
504 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
505 This fixes portability to compilers other than gcc, notably
506 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
508 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
509 makefiles for portability to HP-UX and POSIX correctness.
511 PP/01 Permit LOOKUP_foo enabling on the make command-line.
512 Also via indented variable definition in the Makefile.
513 (Debugging by Oliver Heesakkers).
515 PP/02 Restore caching of spamd results with expanded spamd_address.
516 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
518 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
519 Improves build reliability. Fix from: Frank Elsner
521 NM/02 Fix wide character breakage in the rfc2047 coding
522 Fixes bug 1064. Patch from Andrey N. Oktyabrski
524 NM/03 Allow underscore in dnslist lookups
525 Fixes bug 1026. Patch from Graeme Fowler
527 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
528 Code patches from Adam Ciarcinski of NetBSD.
530 NM/04 Fixed exiqgrep to cope with mailq missing size issue
533 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
534 is logged, to avoid truncation. Patch from John Horne.
536 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
537 Patch from Jakob Hirsch.
539 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
540 SQL string expansion failure details.
541 Patch from Andrey Oktyabrski.
543 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
544 Patch from Simon Arlott.
546 PP/09 New lookups functionality failed to compile on old gcc which rejects
547 extern declarations in function scope.
548 Patch from Oliver Fleischmann
550 PP/10 Use sig_atomic_t for flags set from signal handlers.
551 Check getgroups() return and improve debugging.
552 Fixed developed for diagnosis in bug 927 (which turned out to be
555 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
556 Patch from Mark Zealey.
558 PP/12 Bugzilla 1056: Improved spamd server selection.
559 Patch from Mark Zealey.
561 PP/13 Bugzilla 1086: Deal with maildir quota file races.
562 Based on patch from Heiko Schlittermann.
564 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
565 Patch from Uwe Doering, sign-off by Michael Haardt.
567 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
568 variable declaration deep within a block. Bug and patch from
571 PP/15 lookups-Makefile IRIX compatibilty coercion.
573 PP/16 Make DISABLE_DKIM build knob functional.
575 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
576 Patch by Simon Arlott
578 TF/03 Fix valgrind.h portability to C89 compilers that do not support
579 variable argument macros. Our copy now differs from upstream.
585 TF/01 Failure to get a lock on a hints database can have serious
586 consequences so log it to the panic log.
588 TF/02 Log LMTP confirmation messages in the same way as SMTP,
589 controlled using the smtp_confirmation log selector.
591 TF/03 Include the error message when we fail to unlink a spool file.
593 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
594 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
595 for maintaining out-of-tree patches for some time.
597 PP/01 Bugzilla 139: Documentation and portability issues.
598 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
599 Handle per-OS dynamic-module compilation flags.
601 PP/02 Let /dev/null have normal permissions.
602 The 4.73 fixes were a little too stringent and complained about the
603 permissions on /dev/null. Exempt it from some checks.
604 Reported by Andreas M. Kirchwitz.
606 PP/03 Report version information for many libraries, including
607 Exim version information for dynamically loaded libraries. Created
608 version.h, now support a version extension string for distributors
609 who patch heavily. Dynamic module ABI change.
611 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
612 privilege escalation vulnerability whereby the Exim run-time user
613 can cause root to append content of the attacker's choosing to
616 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
619 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
620 If dropping privileges for untrusted macros, we disabled normal logging
621 on the basis that it would fail; for the Exim run-time user, this is not
622 the case, and it resulted in successful deliveries going unlogged.
623 Fixed. Reported by Andreas Metzler.
629 PP/01 Date: & Message-Id: revert to normally being appended to a message,
630 only prepend for the Resent-* case. Fixes regression introduced in
631 Exim 4.70 by NM/22 for Bugzilla 607.
633 PP/02 Include check_rfc2047_length in configure.default because we're seeing
634 increasing numbers of administrators be bitten by this.
636 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
638 PP/03 Bugzilla 994: added openssl_options main configuration option.
640 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
642 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
644 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
646 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
647 without a peer certificate, leading to a segfault because of an
648 assumption that peers always have certificates. Be a little more
649 paranoid. Problem reported by Martin Tscholak.
651 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
652 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
653 NB: ClamAV planning to remove STREAM in "middle of 2010".
654 CL also introduces -bmalware, various -d+acl logging additions and
655 more caution in buffer sizes.
657 PP/09 Implemented reverse_ip expansion operator.
659 PP/10 Bugzilla 937: provide a "debug" ACL control.
661 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
663 PP/12 Bugzilla 973: Implement --version.
665 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
667 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
669 PP/15 Bugzilla 816: support multiple condition rules on Routers.
671 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
672 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
673 ignore trailing whitespace.
675 JJ/02 prevent non-panic DKIM error from being sent to paniclog
677 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
680 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
681 Notification from Dr Andrew Aitchison.
683 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
684 ExtendedDetectionInfo response format.
685 Notification from John Horne.
687 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
690 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
691 XSL and documented dependency on system catalogs, with examples of how
694 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
697 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
698 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
699 configuration file which is writeable by the Exim user or group.
701 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
702 of configuration files to cover files specified with the -C option if
703 they are going to be used with root privileges, not just the default
706 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
707 option (effectively making it always true).
709 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
710 files to be used while preserving root privileges.
712 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
713 that rogue child processes cannot use them.
715 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
716 run-time user, instead of root.
718 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
719 Exim run-time user without dropping privileges.
721 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
722 result string, instead of calling string_vformat() twice with the same
725 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
726 for other users. Others should always drop root privileges if they use
727 -C on the command line, even for a whitelisted configure file.
729 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
731 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
737 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
738 $data_path, and $header_path variables; fixed documentation bugs and
741 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
742 exipick to access non-standard spools, including the "frozen" queue
745 NM/01 Bugzilla 965: Support mysql stored procedures.
746 Patch from Alain Williams
748 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
750 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
751 Patch from Andreas Metzler
753 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
754 Patch from Kirill Miazine
756 NM/05 Bugzilla 671: Added umask to procmail example.
758 JJ/03 installed exipick 20100323.0, fixing doc bug
760 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
761 directory. Notification and patch from Dan Rosenberg.
763 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
765 TK/02 Improve log output when DKIM signing operation fails.
767 MH/01 Treat the transport option dkim_domain as a colon separated
768 list, not as a single string, and sign the message with each element,
769 omitting multiple occurences of the same signer.
771 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
772 Bugzilla 985, 986. Patch by Simon Arlott
774 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
775 Patch by Simon Arlott
777 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
778 MBX locking. Notification from Dan Rosenberg.
784 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
786 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
788 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
790 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
792 NM/04 Bugzilla 915: Flag broken perl installation during build.
798 TK/01 Added patch by Johannes Berg that expands the main option
799 "spamd_address" if it starts with a dollar sign.
801 TK/02 Write list of recipients to X-Envelope-Sender header when building
802 the mbox-format spool file for content scanning (suggested by Jakob
805 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
806 (http://www.dcc-servers.net/) support via dccifd. Activated by
807 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
809 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
810 by Mark Daniel Reidel <mr@df.eu>.
812 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
813 When building exim an external PCRE library is now needed -
814 PCRE is a system library on the majority of modern systems.
815 See entry on PCRE_LIBS in EDITME file.
817 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
818 conversation. Added nologin parameter to request.
819 Patch contributed by Kirill Miazine.
821 TF/01 Do not log submission mode rewrites if they do not change the address.
823 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
825 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
826 log files in place. Contributed by Roberto Lima.
828 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
830 TF/03 Bugzilla 615: When checking the local_parts router precondition
831 after a local_part_suffix or local_part_prefix option, Exim now
832 does not use the address's named list lookup cache, since this
833 contains cached lookups for the whole local part.
835 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
836 Robert Millan. Documentation is in experimental-spec.txt.
838 TF/04 Bugzilla 668: Fix parallel build (make -j).
840 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
842 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
843 Patch provided by Jan Srzednicki.
845 TF/05 Leading white space used to be stripped from $spam_report which
846 wrecked the formatting. Now it is preserved.
848 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
849 that they are available at delivery time.
851 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
853 TF/08 TLS error reporting now respects the incoming_interface and
854 incoming_port log selectors.
856 TF/09 Produce a more useful error message if an SMTP transport's hosts
857 setting expands to an empty string.
859 NM/06 Bugzilla 744: EXPN did not work under TLS.
860 Patch provided by Phil Pennock.
862 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
863 Patch provided by Richard Godbee.
865 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
866 acl_smtp_notquit, added index entry.
868 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
869 Patch provided by Eugene Bujak.
871 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
872 accept(). Patch provided by Maxim Dounin.
874 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
875 Patch provided by Phil Pennock.
877 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
879 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
880 Patch provided by Brad "anomie" Jorsch.
882 NM/14 Bugzilla 622: Added timeout setting to transport filter.
883 Patch provided by Dean Brooks.
885 TK/05 Add native DKIM support (does not depend on external libraries).
887 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
888 Patch provided by Graeme Fowler.
890 NM/16 Bugzilla 851: Documentation example syntax fix.
892 NM/17 Changed NOTICE file to remove references to embedded PCRE.
894 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
897 NM/19 Bugzilla 745: TLS version reporting.
898 Patch provided by Phil Pennock.
900 NM/20 Bugzilla 167: bool: condition support.
901 Patch provided by Phil Pennock.
903 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
904 clients. Patch provided by Phil Pennock.
906 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
907 Patch provided by Brad "anomie" Jorsch.
909 NM/23 Bugzilla 687: Fix misparses in eximstats.
910 Patch provided by Heiko Schlittermann.
912 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
913 Patch provided by Heiko Schlittermann.
915 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
916 plus update to original patch.
918 NM/26 Bugzilla 799: Documentation correction for ratelimit.
920 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
921 Patch provided by David Brownlee.
923 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
925 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
927 NM/30 Bugzilla 888: TLS documentation bugfixes.
929 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
931 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
932 Unlike the original bugzilla I have changed all shell scripts in src tree.
934 NM/33 Bugzilla 898: Transport filter timeout fix.
935 Patch by Todd Rinaldo.
937 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
938 Patch by Serge Demonchaux.
940 NM/35 Bugzilla 39: Base64 decode bug fixes.
941 Patch by Jakob Hirsch.
943 NM/36 Bugzilla 909: Correct connect() call in dcc code.
945 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
947 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
949 NM/39 Bugzilla 911: Fixed MakeLinks build script.
955 TK/01 Add preliminary DKIM support. Currently requires a forked version of
956 ALT-N's libdkim that I have put here:
957 http://duncanthrax.net/exim-experimental/
959 Note to Michael Haardt: I had to rename some vars in sieve.c. They
960 were called 'true' and it seems that C99 defines that as a reserved
961 keyword to be used with 'bool' variable types. That means you could
962 not include C99-style headers which use bools without triggering
963 build errors in sieve.c.
965 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
966 as mailq or other aliases. Changed the --help handling significantly
967 to do whats expected. exim_usage() emits usage/help information.
969 SC/01 Added the -bylocaldomain option to eximstats.
971 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
973 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
975 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
981 PH/01 Another patch from the Sieve maintainer.
983 PH/02 When an IPv6 address is converted to a string for single-key lookup
984 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
985 dots are used instead of colons so that keys in lsearch files need not
986 contain colons. This was done some time before quoting was made available
987 in lsearch files. However, iplsearch files do require colons in IPv6 keys
988 (notated using the quote facility) so as to distinguish them from IPv4
989 keys. This meant that lookups for IP addresses in host lists did not work
990 for iplsearch lookups.
992 This has been fixed by arranging for IPv6 addresses to be expressed with
993 colons if the lookup type is iplsearch. This is not incompatible, because
994 previously such lookups could never work.
996 The situation is now rather anomolous, since one *can* have colons in
997 ordinary lsearch keys. However, making the change in all cases is
998 incompatible and would probably break a number of configurations.
1000 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
1003 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
1004 conversion specification without a maximum field width, thereby enabling
1005 a rogue spamd server to cause a buffer overflow. While nobody in their
1006 right mind would setup Exim to query an untrusted spamd server, an
1007 attacker that gains access to a server running spamd could potentially
1008 exploit this vulnerability to run arbitrary code as the Exim user.
1010 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
1011 $primary_hostname instead of what libspf2 thinks the hosts name is.
1013 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
1014 a directory entry by the name of the lookup key. Previously, if a
1015 symlink pointed to a non-existing file or a file in a directory that
1016 Exim lacked permissions to read, a lookup for a key matching that
1017 symlink would fail. Now it is enough that a matching directory entry
1018 exists, symlink or not. (Bugzilla 503.)
1020 PH/03 The body_linecount and body_zerocount variables are now exported in the
1023 PH/04 Added the $dnslist_matched variable.
1025 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
1026 This means they are set thereafter only if the connection becomes
1029 PH/06 Added the client_condition to authenticators so that some can be skipped
1030 by clients under certain conditions.
1032 PH/07 The error message for a badly-placed control=no_multiline_responses left
1033 "_responses" off the end of the name.
1035 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
1037 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
1038 (without spaces) instead of just copying the configuration text.
1040 PH/10 Added the /noupdate option to the ratelimit ACL condition.
1042 PH/11 Added $max_received_linelength.
1044 PH/12 Added +ignore_defer and +include_defer to host lists.
1046 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
1047 way in which PCRE > 7.0 is built.
1049 PH/14 Implemented queue_only_load_latch.
1051 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
1052 MAIL command. The effect was to mangle the value on 64-bit systems.
1054 PH/16 Another patch from the Sieve maintainer.
1056 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
1058 PH/18 If a system quota error occurred while trying to create the file for
1059 a maildir delivery, the message "Mailbox is full" was not appended to the
1060 bounce if the delivery eventually timed out. Change 4.67/27 below applied
1061 only to a quota excession during the actual writing of the file.
1063 PH/19 It seems that peer DN values may contain newlines (and other non-printing
1064 characters?) which causes problems in log lines. The DN values are now
1065 passed through string_printing() before being added to log lines.
1067 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
1068 and InterBase are left for another time.)
1070 PH/21 Added message_body_newlines option.
1072 PH/22 Guard against possible overflow in moan_check_errorcopy().
1074 PH/23 POSIX allows open() to be a macro; guard against that.
1076 PH/24 If the recipient of an error message contained an @ in the local part
1077 (suitably quoted, of course), incorrect values were put in $domain and
1078 $local_part during the evaluation of errors_copy.
1084 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
1085 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
1088 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
1089 issue a MAIL command.
1091 PH/02 In an ACL statement such as
1093 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
1095 if a client was not listed at all, or was listed with a value other than
1096 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
1097 the condition was not true (as it should be), so access was not denied.
1098 The bug was that the ! inversion was incorrectly passed on to the second
1099 item. This has been fixed.
1101 PH/03 Added additional dnslists conditions == and =& which are different from
1102 = and & when the dns lookup returns more than one IP address.
1104 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
1105 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
1107 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
1108 FSYNC, which compiles an option called disable_fsync that allows for
1109 bypassing fsync(). The documentation is heavily laced with warnings.
1111 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
1113 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
1114 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
1115 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
1116 including adding "make clean"; (3) Added -fPIC when compiling the test
1117 dynamically loaded module, to get rid of a warning.
1119 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
1120 message fails, move_frozen_messages = true and ignore_bounce_errors_after
1121 = 0s. The bug is otherwise harmless.
1123 PH/07 There was a bug in the dovecot authenticator such that the value of
1124 $auth1 could be overwritten, and so not correctly preserved, after a
1125 successful authentication. This usually meant that the value preserved by
1126 the server_setid option was incorrect.
1128 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
1130 PH/09 Installed PCRE release 7.0.
1132 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
1133 run for batched SMTP input. It is now run at the start of every message
1134 in the batch. While fixing this I discovered that the process information
1135 (output by running exiwhat) was not always getting set for -bs and -bS
1136 input. This is fixed, and it now also says "batched" for BSMTP.
1138 PH/11 Added control=no_pipelining.
1140 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
1141 patch, slightly modified), and move the expansion of helo_data till after
1142 the connection is made in the smtp transport (so it can use these
1145 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
1147 PH/14 Added log_selector = +pid.
1149 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
1151 PH/16 Add ${if forany and ${if forall.
1153 PH/17 Added dsn_from option to vary the From: line in DSNs.
1155 PH/18 Flush SMTP output before performing a callout, unless control =
1156 no_callout_flush is set.
1158 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
1159 was true (the default) a successful delivery failed to delete the retry
1160 item, thus causing premature timeout of the address. The bug is now
1163 PH/20 Added hosts_avoid_pipelining to the smtp transport.
1165 PH/21 Long custom messages for fakedefer and fakereject are now split up
1166 into multiline reponses in the same way that messages for "deny" and
1167 other ACL rejections are.
1169 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
1170 with slight modification.
1172 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
1173 draft, changing the syntax and factoring some duplicate code".
1175 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
1176 for deliveries of the second and subsequent messages over the same SMTP
1179 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
1180 ${reduce, with only minor "tidies".
1182 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
1184 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
1185 expansion side effects.
1187 PH/27 When a message times out after an over-quota error from an Exim-imposed
1188 quota, the bounce message says "mailbox is full". This message was not
1189 being given when it was a system quota that was exceeded. It now should
1192 MH/03 Made $recipients available in local_scan(). local_scan() already has
1193 better access to the recipient list through recipients_list[], but
1194 $recipients can be useful in postmaster-provided expansion strings.
1196 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
1197 in the case of a MAIL command with additional options following the
1198 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
1199 were accidentally chopped off.
1201 PH/29 SMTP synchronization checks are implemented when a command is read -
1202 there is a check that no more input is waiting when there shouldn't be
1203 any. However, for some commands, a delay in an ACL can mean that it is
1204 some time before the response is written. In this time, more input might
1205 arrive, invalidly. So now there are extra checks after an ACL has run for
1206 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
1207 pipelining has not been advertised.
1209 PH/30 MH's patch to allow iscntrl() characters to be list separators.
1211 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
1212 returned in the SMTP response when smtp_return_error_details was false.
1213 This has been fixed.
1215 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
1216 instead of the C I/O that was originally supplied, because problems were
1217 reported on Solaris.
1219 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
1220 Exim which did not show up earlier: it was assuming that a call to
1221 SSL_CTX_set_info_callback() might give an error value. In fact, there is
1222 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
1223 was a macro that became an assignment, so it seemed to work. This has
1224 changed to a proper function call with a void return, hence the compile
1225 error. Exim's code has been fixed.
1227 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
1230 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
1232 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
1234 PH/37 If a message is not accepted after it has had an id assigned (e.g.
1235 because it turns out to be too big or there is a timeout) there is no
1236 "Completed" line in the log. When some messages of this type were
1237 selected by exigrep, they were listed as "not completed". Others were
1238 picked up by some special patterns. I have improved the selection
1239 criteria to be more general.
1241 PH/38 The host_find_failed option in the manualroute router can now be set
1242 to "ignore", to completely ignore a host whose IP address cannot be
1243 found. If all hosts are ignored, the behaviour is controlled by the new
1244 host_all_ignored option.
1246 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
1247 homing or because of multiple MX records with /mx) generated more than
1248 one IP address, and the following item turned out to be the local host,
1249 all the secondary addresses of the first item were incorrectly removed
1250 from the list, along with the local host and any following hosts (which
1251 is what is supposed to happen).
1253 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
1254 whoever called Exim into the -H file. In the case of the daemon it was
1255 behaving confusingly. When first started, it used values for whoever
1256 started the daemon, but after a SIGHUP it used the Exim user (because it
1257 calls itself on a restart). I have changed the code so that it now always
1260 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
1261 message are rejected with the same error (e.g. no authentication or bad
1262 sender address), and a DATA command is nevertheless sent (as can happen
1263 with PIPELINING or a stupid MUA), the error message that was given to the
1264 RCPT commands is included in the rejection of the DATA command. This is
1265 intended to be helpful for MUAs that show only the final error to their
1268 PH/42 Another patch from the Sieve maintainer.
1270 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
1271 Eximstats - Fixed some broken HTML links and added missing column headers
1273 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
1274 columns for Rejects, Temp Rejects, Ham, and Spam rows.
1276 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
1278 PH/43 Yet another patch from the Sieve maintainer.
1280 PH/44 I found a way to check for a TCP/IP connection going away before sending
1281 the response to the final '.' that terminates a message, but only in the
1282 case where the client has not sent further data following the '.'
1283 (unfortunately, this is allowed). However, in many cases there won't be
1284 any further data because there won't be any more messages to send. A call
1285 to select() can be used: if it shows that the input is "ready", there is
1286 either input waiting, or the socket has been closed. An attempt to read
1287 the next input character can distinguish the two cases. Previously, Exim
1288 would have sent an OK response which the client would never have see.
1289 This could lead to message repetition. This fix should cure that, at
1290 least in a lot of common cases.
1292 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
1293 advertised in response to EHLO.
1299 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
1300 fixed by 4.65/MH/01 (is this a record?) are fixed:
1302 (i) An empty string was always treated as zero by the numeric comparison
1303 operators. This behaviour has been restored.
1305 (ii) It is documented that the numeric comparison operators always treat
1306 their arguments as decimal numbers. This was broken in that numbers
1307 starting with 0 were being interpreted as octal.
1309 While fixing these problems I realized that there was another issue that
1310 hadn't been noticed. Values of message_size_limit (both the global option
1311 and the transport option) were treated as octal if they started with 0.
1312 The documentation was vague. These values are now always treated as
1313 decimal, and I will make that clear in the documentation.
1319 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
1320 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
1323 MH/01 Don't check that the operands of numeric comparison operators are
1324 integers when their expansion is in "skipping" mode (fixes bug
1325 introduced by 4.64-PH/07).
1327 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
1328 child addresses, Exim now panics and dies. Previously, because the count
1329 is held in a short int, deliveries were likely to be lost. As such a
1330 large number of recipients for a single message is ridiculous
1331 (performance will be very, very poor), I have chosen to impose a limit
1332 rather than extend the field.
1338 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
1339 leftover -K file (the existence of which was triggered by #402).
1340 While we were at it, introduced process PID as part of the -K
1341 filename. This should rule out race conditions when creating
1344 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
1345 processing considerably. Previous code took too long for large mails,
1346 triggering a timeout which in turn triggers #401.
1348 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
1349 in the DK code in transports.c. sendfile() is not really portable,
1350 hence the _LINUX specificness.
1352 TF/01 In the add_headers option to the mail command in an Exim filter,
1353 there was a bug that Exim would claim a syntax error in any
1354 header after the first one which had an odd number of characters
1357 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
1358 callout verification, Exim cached a "reject" for the entire domain. This
1359 is correct for most verifications, but it is not correct for a recipient
1360 verification with use_sender or use_postmaster set, because in that case
1361 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
1362 case of MAIL FROM:<> rejection from other early rejections (e.g.
1363 rejection of HELO). When verifying a recipient using a non-null MAIL
1364 address, the cache is ignored if it shows MAIL FROM:<> rejection.
1365 Whatever the result of the callout, the value of the domain cache is
1366 left unchanged (for any other kind of callout, getting as far as trying
1367 RCPT means that the domain itself is ok).
1369 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
1372 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
1373 manifest itself as EPIPE rather than ECONNECT. When tidying away a
1374 session, the daemon ignores ECONNECT errors and logs others; it now
1375 ignores EPIPE as well.
1377 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
1378 (quoted-printable decoding).
1380 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
1381 later the small subsequent patch to fix an introduced bug.
1383 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
1385 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
1387 PH/08 An error is now given if message_size_limit is specified negative.
1389 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
1390 to be given (somewhat) arbitrary names.
1392 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
1395 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
1396 miscellaneous code fixes
1398 PH/10 Added the log_reject_target ACL modifier to specify where to log
1401 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
1402 hostname. This is wrong, because it relates to the incoming message (and
1403 probably the interface on which it is arriving) and not to the outgoing
1404 callout (which could be using a different interface). This has been
1405 changed to use the value of the helo_data option from the smtp transport
1406 instead - this is what is used when a message is actually being sent. If
1407 there is no remote transport (possible with a router that sets up host
1408 addresses), $smtp_active_hostname is used.
1410 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
1411 tweaks were necessary in order to get it to work (see also 21 below):
1412 (a) The code assumed that strncpy() returns a negative number on buffer
1413 overflow, which isn't the case. Replaced with Exim's string_format()
1415 (b) There were several signed/unsigned issues. I just did the minimum
1416 hacking in of casts. There is scope for a larger refactoring.
1417 (c) The code used strcasecmp() which is not a standard C function.
1418 Replaced with Exim's strcmpic() function.
1419 (d) The code set only $1; it now sets $auth1 as well.
1420 (e) A simple test gave the error "authentication client didn't specify
1421 service in request". It would seem that Dovecot has changed its
1422 interface. Fortunately there's a specification; I followed it and
1423 changed what the client sends and it appears to be working now.
1425 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
1428 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
1429 address A is aliased to B and C, where B exists and C does not. Without
1430 -v the output is "A verified" because verification stops after a
1431 successful redirection if more than one address is generated. However,
1432 with -v the child addresses are also verified. Exim was outputting "A
1433 failed to verify" and then showing the successful verification for C,
1434 with its parentage. It now outputs "B failed to verify", showing B's
1435 parentage before showing the successful verification of C.
1437 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
1438 look up a TXT record in a specific list after matching in a combined
1441 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
1442 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
1443 they consult the DNS. I had assumed they would set it the way they
1444 wanted; and indeed my experiments on Linux seem to show that in some
1445 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
1446 To be on the safe side, however, I have now made the interface to
1447 host_find_byname() similar to host_find_bydns(), with an argument
1448 containing the DNS resolver options. The host_find_byname() function now
1449 sets these options at its start, just as host_find_bydns() does. The smtp
1450 transport options dns_qualify_single and dns_search_parents are passed to
1451 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
1452 of host_find_byname() use the default settings of RES_DEFNAMES
1453 (qualify_single) but not RES_DNSRCH (search_parents).
1455 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
1456 spool_read_header() do less string testing, by means of a preliminary
1457 switch on the second character of optional "-foo" lines. (This is
1458 overdue, caused by the large number of possibilities that now exist.
1459 Originally there were few.) While I was there, I also converted the
1460 str(n)cmp tests so they don't re-test the leading "-" and the first
1461 character, in the hope this might squeeze out yet more improvement.
1463 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
1464 flag allowing group syntax was set by the header_syntax check but not
1465 turned off, possible causing trouble later; (2) The flag was not being
1466 set at all for the header_verify test, causing "group"-style headers to
1467 be rejected. I have now set it in this case, and also caused header_
1468 verify to ignore an empty address taken from a group. While doing this, I
1469 came across some other cases where the code for allowing group syntax
1470 while scanning a header line wasn't quite right (mostly, not resetting
1471 the flag correctly in the right place). These bugs could have caused
1472 trouble for malformed header lines. I hope it is now all correct.
1474 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
1475 with the "reply" argument non-NULL. The code, however (which originally
1476 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
1477 but it didn't always do it. This confused somebody who was copying the
1478 code for some other use. I have removed all the tests.
1480 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
1481 feature that was used to support insecure browsers during the U.S. crypto
1482 embargo. It requires special client support, and Exim is probably the
1483 only MTA that supported it -- and would never use it because real RSA is
1484 always available. This code has been removed, because it had the bad
1485 effect of slowing Exim down by computing (never used) parameters for the
1486 RSA_EXPORT functionality.
1488 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
1489 authenticator to fail if there's a tab character in the incoming data
1490 (there should never be unless someone is messing about, as it's supposed
1491 to be base64-encoded). Also added, on Timo's advice, the "secured" option
1492 if the connection is using TLS or if the remote IP is the same as the
1493 local IP, and the "valid-client-cert option" if a client certificate has
1496 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
1497 authenticators. This can be used for authorization after authentication
1498 succeeds. (In the case of plaintext, it servers for both authentication
1501 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
1502 if any retry times were supplied.
1504 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
1505 connection, where there is no client IP address to check. In this
1506 situation, the verify now always succeeds.
1508 PH/25 Applied John Jetmore's -Mset patch.
1510 PH/26 Added -bem to be like -Mset, but loading a message from a file.
1512 PH/27 In a string expansion for a processed (not raw) header when multiple
1513 headers of the same name were present, leading whitespace was being
1514 removed from all of them, but trailing whitespace was being removed only
1515 from the last one. Now trailing whitespace is removed from each header
1516 before concatenation. Completely empty headers in a concatenation (as
1517 before) are ignored.
1519 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
1520 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
1522 PH/29 [Removed. This was a change that I later backed out, and forgot to
1523 correct the ChangeLog entry (that I had efficiently created) before
1524 committing the later change.]
1526 PH/30 Exim was sometimes attempting to deliver messages that had suffered
1527 address errors (4xx response to RCPT) over the same connection as other
1528 messages routed to the same hosts. Such deliveries are always "forced",
1529 so retry times are not inspected. This resulted in far too many retries
1530 for the affected addresses. The effect occurred only when there were more
1531 hosts than the hosts_max_try setting in the smtp transport when it had
1532 the 4xx errors. Those hosts that it had tried were not added to the list
1533 of hosts for which the message was waiting, so if all were tried, there
1534 was no problem. Two fixes have been applied:
1536 (i) If there are any address or message errors in an SMTP delivery, none
1537 of the hosts (tried or untried) are now added to the list of hosts
1538 for which the message is waiting, so the message should not be a
1539 candidate for sending over the same connection that was used for a
1540 successful delivery of some other message. This seems entirely
1541 reasonable: after all the message is NOT "waiting for some host".
1542 This is so "obvious" that I'm not sure why it wasn't done
1543 previously. Hope I haven't missed anything, but it can't do any
1544 harm, as the worst effect is to miss an optimization.
1546 (ii) If, despite (i), such a delivery is accidentally attempted, the
1547 routing retry time is respected, so at least it doesn't keep
1548 hammering the server.
1550 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
1551 in ${readsocket because some servers need this prod.
1553 PH/32 Added some extra debug output when updating a wait-xxx database.
1555 PH/33 The hint "could be header name not terminated by colon", which has been
1556 given for certain expansion errors for a long time, was not being given
1557 for the ${if def:h_colon_omitted{... case.
1559 PH/34 The spec says: "With one important exception, whenever a domain list is
1560 being scanned, $domain contains the subject domain." There was at least
1561 one case where this was not true.
1563 PH/35 The error "getsockname() failed: connection reset by peer" was being
1564 written to the panic log as well as the main log, but it isn't really
1565 panic-worthy as it just means the connection died rather early on. I have
1566 removed the panic log writing for the ECONNRESET error when getsockname()
1569 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
1570 runs only) independently of the message's sender address. This meant
1571 that, if the 4xx error was in fact related to the sender, a different
1572 message to the same recipient with a different sender could confuse
1573 things. In particualar, this can happen when sending to a greylisting
1574 server, but other circumstances could also provoke similar problems.
1575 I have changed the default so that the retry time for these errors is now
1576 based a combination of the sender and recipient addresses. This change
1577 can be overridden by setting address_retry_include_sender=false in the
1580 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
1581 remote server are returned as part of bounce messages. This was not
1582 happening for LMTP over a pipe (the lmtp transport), but now it is the
1583 same for both kinds of LMTP.
1585 PH/38 Despite being documented as not happening, Exim was rewriting addresses
1586 in header lines that were in fact CNAMEs. This is no longer the case.
1588 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
1589 and queue runs started by the daemon processed all messages. This has
1590 been fixed so that -R and -S can now usefully be given with -q<time>.
1592 PH/40 Import PCRE release 6.7 (fixes some bugs).
1594 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
1596 PH/42 Give an error if -q is specified more than once.
1598 PH/43 Renamed the variables $interface_address and $interface_port as
1599 $received_ip_address and $received_port, to make it clear that these
1600 values apply to message reception, and not to the outgoing interface when
1601 a message is delivered. (The old names remain recognized, of course.)
1603 PH/44 There was no timeout on the connect() call when using a Unix domain
1604 socket in the ${readsocket expansion. There now is.
1606 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
1607 be meaningful with "accept".
1609 SC/01 Eximstats V1.43
1610 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
1612 SC/02 Eximstats V1.44
1613 Use a glob alias rather than an array ref in the generated
1614 parser. This improves both readability and performance.
1616 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
1617 Collect SpamAssassin and rejection statistics.
1618 Don't display local sender or destination tables unless
1619 there is data to show.
1620 Added average volumes into the top table text output.
1622 SC/04 Eximstats V1.46
1623 Collect data on the number of addresses (recipients)
1624 as well as the number of messages.
1626 SC/05 Eximstats V1.47
1627 Added 'Message too big' to the list of mail rejection
1628 reasons (thanks to Marco Gaiarin).
1630 SC/06 Eximstats V1.48
1631 Mainlog lines which have GMT offsets and are too short to
1632 have a flag are now skipped.
1634 SC/07 Eximstats V1.49 (Alain Williams)
1635 Added the -emptyok flag.
1637 SC/08 Eximstats V1.50
1638 Fixes for obtaining the IP address from reject messages.
1640 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
1641 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
1642 whitesspace changes from 4.64-PH/27
1644 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
1647 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
1648 are found, allow negative numbers in numeric criteria)
1650 JJ/06 exipick.20061117.2, added new $message_body_missing variable
1652 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
1653 to match changes made in 4.64-PH/43
1655 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
1657 PH/47 Put in an explicit test for a DNS lookup of an address record where the
1658 "domain" is actually an IP address, and force a failure. This locks out
1659 those revolvers/nameservers that support "A-for-A" lookups, in
1660 contravention of the specifications.
1662 PH/48 When a host name was looked up from an IP address, and the subsequent
1663 forward lookup of the name timed out, the host name was left in
1664 $sender_host_name, contrary to the specification.
1666 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
1667 restricted to single-key lookups, Exim was not diagnosing an error if
1668 * or *@ was used with a query-style lookup.
1670 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
1672 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
1673 long ago, but noone interested enough thought of it. Let's just say that
1674 the "1.1" means that there are some new functions that weren't there at
1675 some point in the past.
1677 PH/51 Error processing for expansion failure of helo_data from an smtp
1678 transport during callout processing was broken.
1680 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
1681 tested/used via the -bh/-bhc/-bs options.
1683 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
1684 bug, fixed in subsequent PCRE releases).
1686 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
1687 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
1689 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
1695 SC/01 Use a glob alias rather than an array ref in eximstats generated
1696 parser. This improves both readability and performance.
1698 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
1699 Don't display local sender or destination tables in eximstats unless
1700 there is data to show.
1701 Added average volumes into the eximstats top table text output.
1703 SC/03 Collect data on the number of addresses (recipients) as well
1704 as the number of messages in eximstats.
1706 TF/01 Correct an error in the documentation for the redirect router. Exim
1707 does (usually) call initgroups() when daemonizing.
1709 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
1710 with consistent privilege compared to when running as a daemon.
1712 TF/03 Note in the spec that $authenticated_id is not set for local
1713 submissions from trusted users.
1715 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
1716 Thanks to Dean Brooks <dean@iglou.com> for the patch.
1718 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
1719 by adding some example configuration directives to the default
1720 configuration file. A little bit of work is required to uncomment the
1721 directives and define how usernames and passwords are checked, but
1722 there is now a framework to start from.
1724 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
1725 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
1726 without this. I don't know how relevant this is to other LDAP libraries.
1728 PH/02 Add the verb name to the "unknown ACL verb" error.
1730 PH/03 Magnus Holmgren's patch for filter_prepend_home.
1732 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
1734 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
1735 directory not expanded when it should be if an expanded home directory
1736 was set for the address (which is overridden by the transport).
1738 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
1741 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
1742 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
1743 because it is too late at that time, and has no effect.
1745 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
1746 security issue with \' (bugzilla #107). I could not use the
1747 PQescapeStringConn() function, because it needs a PGconn value as one of
1750 PH/08 When testing addresses using -bt, indicate those final addresses that
1751 are duplicates that would not cause an additional delivery. At least one
1752 person was confused, thinking that -bt output corresponded to deliveries.
1753 (Suppressing duplicates isn't a good idea as you lose the information
1754 about possibly different redirections that led to the duplicates.)
1756 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
1757 systems where poll() doesn't work, in particular OS X.
1759 PH/10 Added more information to debugging output for retry time not reached.
1761 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
1762 operations in malware.c.
1764 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
1767 PH/13 If write_rejectlog was set false when logging was sent to syslog with
1768 syslog_duplication set false, log lines that would normally be written
1769 both the the main log and to the reject log were not written to syslog at
1772 PH/14 In the default configuration, change the use of "message" in ACL warn
1773 statements to "add_header".
1775 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
1776 not followed by a command (e.g. "seen endif").
1778 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
1779 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
1782 PH/17 Added extra conditions to the default value of delay_warning_condition
1786 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
1787 { match{$h_precedence:}{(?i)bulk|list|junk} } \
1788 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
1791 The Auto-Submitted: and various List- headers are standardised, whereas I
1792 don't think Precedence: ever was.
1794 PH/18 Refactored debugging code in route_finduser() to show more information,
1795 in particular, the error code if getpwnam() issues one.
1797 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
1798 This is apparently needed in addition to the PH/07 change above to avoid
1799 any possible encoding problems.
1801 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
1802 but not after initializing Perl.
1804 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
1805 output them only if debugging. By default they are written stderr,
1806 apparently, which is not desirable.
1808 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
1811 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
1814 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
1816 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
1817 authenticated or an ident call has been made. Suppress the default
1818 values for $authenticated_id and $authenticated_sender (but permit -oMai
1819 and -oMas) when testing with -bh.
1821 PH/24 Re-jigged the order of the tests in the default configuration so that the
1822 tests for valid domains and recipients precede the DNS black list and CSA
1823 tests, on the grounds that those ones are more expensive.
1825 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
1826 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
1827 This bug exists in every version of Exim that I still have, right back to
1830 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
1831 However, an attempt to turn on case-sensitivity in a regex key by
1832 including (?-i) didn't work because the subject string was already
1833 lowercased, and the effects were non-intuitive. It turns out that a
1834 one-line patch can be used to allow (?-i) to work as expected.
1840 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
1841 other effects) broke the use of negated acl sub-conditions.
1843 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
1846 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
1847 "Deny" causes Exim to reject the incoming connection with a 554 error.
1848 Unfortunately, if there is a major crisis, such as a disk failure,
1849 tcp-wrappers gives "deny", whereas what one would like would be some
1850 kind of temporary error. A kludge has been added to help with this.
1851 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
1852 554 error is used if errno is still zero or contains ENOENT (which occurs
1853 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
1856 PH/03 Add -lutil to the default FreeBSD LIBS setting.
1858 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
1859 errors. Otherwise a message that provokes a temporary error (when other
1860 messages do not) can cause a whole host to time out.
1862 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
1863 addresses were routed directly to files or pipes from a redirect router.
1864 File deliveries just didn't batch; pipe deliveries might have suffered
1867 PH/06 A failure to get a lock for a hints database would erroneously always say
1868 "Failed to get write lock", even when it was really a read lock.
1870 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
1871 of 0600. This has been changed to use the value of the lockfile_mode
1872 option (which defaults to 0600).
1874 PH/08 Applied small patch from the Sieve maintainer.
1876 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
1877 folder from quota calculations, a direct delivery into this folder messed
1878 up the contents of the maildirsize file. This was because the regex was
1879 used only to exclude .Trash (or whatever) when the size of the mailbox
1880 was calculated. There was no check that a delivery was happening into an
1881 excluded directory. This bug has been fixed by ignoring all quota
1882 processing for deliveries into excluded directories.
1884 PH/10 Added the maildirfolder_create_regex option to appendfile.
1890 PH/01 The code for finding all the local interface addresses on a FreeBSD
1891 system running IPv6 was broken. This may well have applied to all BSD
1892 systems, as well as to others that have similar system calls. The broken
1893 code found IPv4 interfaces correctly, but gave incorrect values for the
1894 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
1895 that it would not match correctly against @[] and not recognize the IPv6
1898 PH/02 The ipliteral router was not recognizing addresses of the form user@
1899 [ipv6:....] because it didn't know about the "ipv6:" prefix.
1901 PH/03 Added disable_ipv6.
1903 PH/04 Changed $reply_address to use the raw form of the headers instead of the
1904 decoded form, because it is most often used to construct To: headers
1905 lines in autoreplies, and the decoded form may well be syntactically
1906 invalid. However, $reply_address has leading white space removed, and all
1907 newlines turned into spaces so that the autoreply transport does not
1910 PH/05 If group was specified without a user on a router, and no group or user
1911 was specified on a transport, the group from the router was ignored.
1913 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
1914 for visible compile-time settings that can be used to change these
1915 numbers, for those that want even more. Backwards compatibility with old
1916 spool files has been maintained. However, going back to a previous Exim
1917 release will lost any variables that are in spool files.
1919 PH/07 Two small changes when running in the test harness: increase delay when
1920 passing a TCP/IP connection to a new process, in case the original
1921 process has to generate a bounce, and remove special handling of
1922 127.0.0.2 (sic), which is no longer necessary.
1924 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
1925 be the same on different OS.
1927 PH/09 Moved a debug statement in filter processing to avoid a race problem when
1930 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
1931 whether --show-vars was specified or not
1933 JJ/02 exipick: Added support for new ACL variable spool format introduced
1936 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
1937 syntactically invalid From: or Reply-to: line, and a filter used this to
1938 generate an autoreply, and therefore failed to obtain an address for the
1939 autoreply, Exim could try to deliver to a non-existent relative file
1940 name, causing unrelated and misleading errors. What now happens is that
1941 it logs this as a hard delivery error, but does not attempt to create a
1944 PH/11 The exinext utility has a -C option for testing purposes, but although
1945 the given file was scanned by exinext itself; it wasn't being passed on
1946 when Exim was called.
1948 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
1949 an end-of-file indication when reading a command response.
1951 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
1952 compiled. In many other places in Exim, IPv6 addresses are always
1953 recognized, so I have changed this. It also means that IPv4 domain
1954 literals of the form [IPV4:n.n.n.n] are now always recognized.
1956 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
1957 used if the router is not running as root, for example, when verifying at
1958 ACL time, or when using -bh. The debugging output from this situation was
1959 non-existent - all you got was a failure to exec. I have made two
1962 (a) Failures to set uid/gid, the current directory, or a process leader
1963 in a subprocess such as that created by queryprogram now generate
1964 suitable debugging ouput when -d is set.
1966 (b) The queryprogram router detects when it is not running as root,
1967 outputs suitable debugging information if -d is set, and then runs
1968 the subprocess without attempting to change uid/gid.
1970 PH/15 Minor change to Makefile for building test_host (undocumented testing
1973 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
1974 additional section of a DNS packet that returns MX or SRV records.
1975 Instead, it always explicitly searches for A/AAAA records. This avoids
1976 major problems that occur when a DNS server includes only records of one
1977 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
1978 fixed another bug: if SRV records were looked up and the corresponding
1979 address records were *not* found in the additional section, the port
1980 values from the SRV records were lost.
1982 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
1983 using the correct key (the original address) when searching the retry
1984 rules in order to find which one to use for generating the retry hint.
1986 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
1987 adding the default one. Similarly, if it contains a Reply-To: header, the
1988 errors_reply_to option, if set, is not used.
1990 PH/19 When calculating a retry time, Exim used to measure the "time since
1991 failure" by looking at the "first failed" field in the retry record. Now
1992 it does not use this if it is later than than the arrival time of the
1993 message. Instead it uses the arrival time. This makes for better
1994 behaviour in cases where some deliveries succeed, thus re-setting the
1995 "first failed" field. An example is a quota failure for a huge message
1996 when small messages continue to be delivered. Without this change, the
1997 "time since failure" will always be short, possible causing more frequent
1998 delivery attempts for the huge message than are intended.
1999 [Note: This change was subsequently modified - see PH/04 for 4.62.]
2001 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
2002 $1, $2, $3) because the numerical variables can be reset during some
2003 expansion items (e.g. "match"), thereby losing the authentication data.
2005 PH/21 Make -bV show the size of off_t variables so that the test suite can
2006 decide whether to run tests for quotas > 2G.
2008 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
2009 mailbox_size, and mailbox_filecount in the appendfile transport. If a
2010 filecount value is greater than 2G or if a quota value is greater than 2G
2011 on a system where the size of off_t is not greater than 4, a panic error
2014 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
2015 never match. The debug and -bh output now contains an explicit error
2016 message indicating a malformed IPv4 address or mask.
2018 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
2019 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
2020 PH/23 above applies.
2022 PH/25 Do not write to syslog when running in the test harness. The only
2023 occasion when this arises is a failure to open the main or panic logs
2024 (for which there is an explicit test).
2026 PH/26 Added the /no_tell option to "control=freeze".
2028 PH/27 If a host name lookup failed very early in a connection, for example, if
2029 the IP address matched host_lookup and the reverse lookup yielded a name
2030 that did not have a forward lookup, an error message of the form "no IP
2031 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
2032 could be logged. Now it outputs the IP address instead of "NULL".
2034 PH/28 An enabling patch from MH: add new function child_open_exim2() which
2035 allows the sender and the authenticated sender to be set when
2036 submitting a message from within Exim. Since child_open_exim() is
2037 documented for local_scan(), the new function should be too.
2039 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
2040 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
2041 results in an empty string is now treated as unset.
2043 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
2045 PH/31 Added sender_verify_fail logging option.
2047 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
2048 needed by RFC 822 but not by RFC 2822 was commented out. I have now
2049 tidied the source and removed it altogether.
2051 PH/33 When a queue run was abandoned because the load average was too high, a
2052 log line was always written; now it is written only if the queue_run log
2053 selector is set. In addition, the log line for abandonment now contains
2054 information about the queue run such as the pid. This is always present
2055 in "start" and "stop" lines but was omitted from the "abandon" line.
2057 PH/34 Omit spaces between a header name and the colon in the error message that
2058 is given when verify = headers_syntax fails (if there are lots of them,
2059 the message gets confusing).
2061 PH/35 Change the default for dns_check_names_pattern to allow slashes within
2062 names, as there are now some PTR records that contain slashes. This check
2063 is only to protect against broken name servers that fall over on strange
2064 characters, so the fact that it applies to all lookups doesn't matter.
2066 PH/36 Now that the new test suite is complete, we can remove some of the
2067 special code in Exim that was needed for the old test suite. For example,
2068 sorting DNS records because real resolvers return them in an arbitrary
2069 order. The new test suite's fake resolver always returns records in the
2072 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
2073 bounces) except when queue_only is set, to avoid logging races between
2074 the different processes.
2076 PH/38 Panic-die if .include specifies a non-absolute path.
2078 PH/39 A tweak to the "H" retry rule from its user.
2080 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
2081 a label. They prevented compilation on older perls.
2083 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
2084 a warning to be raised on newish perls.
2086 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
2087 on queue. Changes to match documented behaviour of showing count of
2088 messages matching specified criteria.
2090 PH/40 Changed the default ident timeout from 30s to 5s.
2092 PH/41 Added support for the use of login_cap features, on those BSD systems
2093 that have them, for controlling the resources used by pipe deliveries.
2095 PH/42 The content-scanning code uses fopen() to create files in which to put
2096 message data. Previously it was not paying any attention to the mode of
2097 the files. Exim runs with umask(0) because the rest of the code creates
2098 files with open(), and sets the required mode explicitly. Thus, these
2099 files were ending up world-writeable. This was not a big issue, because,
2100 being within the spool directory, they were not world-accessible. I have
2101 created a function called modefopen, which takes an additional mode
2102 argument. It sets umask(777), creates the file, chmods it to the required
2103 mode, then resets the umask. All the relevant calls to fopen() in the
2104 content scanning code have been changed to use this function.
2106 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
2107 to 24 hours. This avoids potential overflow problems when processing G
2108 and H retry rules. I suspect nobody ever tinkers with this value.
2110 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
2112 PH/45 When the plaintext authenticator is running as a client, the server's
2113 challenges are checked to ensure they are valid base64 strings. By
2114 default, the authentication attempt is cancelled if an invalid string is
2115 received. Setting client_ignore_invalid_base64 true ignores these errors.
2116 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
2117 they are received. Thus, the responses can be made to depend on the
2118 challenges. If an invalid string is ignored, an empty string is placed in
2121 PH/46 Messages that are created by the autoreply transport now contains a
2122 References: header, in accordance with RFCs 2822 and 3834.
2124 PH/47 Added authenticated_sender_force to the smtp transport.
2126 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
2128 PH/49 Installed latest patch from the Sieve maintainer.
2130 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
2131 was also set, the appendfile transport was unnecessarily scanning a
2132 directory of message files (e.g. for maildir delivery) to find the count
2133 of files (along with the size), even though it did not need this
2134 information. It now does the scan only if it needs to find either the
2135 size of the count of files.
2137 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
2139 PH/52 Two bugs concerned with error handling when the smtp transport is
2142 (i) Exim was not creating retry information for temporary errors given
2143 for individual recipients after the DATA command when the smtp transport
2144 was used in LMTP mode. This meant that they could be retried too
2145 frequently, and not timed out correctly.
2147 (ii) Exim was setting the flag that allows error details to be returned
2148 for LMTP errors on RCPT commands, but not for LMTP errors for individual
2149 recipients that were returned after the DATA command.
2151 PH/53 This is related to PH/52, but is more general: for any failing address,
2152 when detailed error information was permitted to be returned to the
2153 sender, but the error was temporary, then after the final timeout, only
2154 "retry timeout exceeded" was returned. Now it returns the full error as
2155 well as "retry timeout exceeded".
2157 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
2158 do this, and (what is worse) MTAs that accept it.
2160 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
2161 will now be deprecated.
2163 PH/56 New os.c-cygwin from the Cygwin maintainer.
2165 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
2166 formats (previously only available in exim formats via -bpr, -bpru,
2167 and -bpra. Now also available in native and exiqgrep formats)
2169 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
2170 with very large, slow to parse queues
2172 JJ/08 exipick: added ! as generic prefix to negate any criteria format
2174 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
2176 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
2177 responses to authentication challenges, though it was showing the
2178 challenges; (ii) I've removed the CR characters from the debug output for
2181 PH/58 Allow for the insertion of a newline as well as a space when a string
2182 is turned into more than one encoded-word during RFC 2047 encoding. The
2183 Sieve code now uses this.
2185 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
2186 data_4xx, lost_connection, tls_required.
2188 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
2189 message was being sent as an SMTP response.
2191 PH/61 Add -l and -k options to exicyclog.
2193 PH/62 When verifying, if an address was redirected to one new address, so that
2194 verification continued, and the new address failed or deferred after
2195 having set something in $address_data, the value of $address_data was not
2196 passed back to the ACL. This was different to the case when no
2197 redirection occurred. The value is now passed back in both cases.
2199 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
2200 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
2201 use login_cap.h, so on its own it isn't the distinguishing feature. The
2202 new name refers directly to the setclassresources() function.
2204 PH/65 Added configuration files for NetBSD3.
2206 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
2208 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
2209 is preferred over IPv4.
2211 PH/68 The bounce_return_message and bounce_return_body options were not being
2212 honoured for bounces generated during the reception of non-SMTP messages.
2213 In particular, this applied to messages rejected by the ACL. This bug has
2214 been fixed. However, if bounce_return_message is true and bounce_return_
2215 body is false, the headers that are returned for a non-SMTP message
2216 include only those that have been read before the error was detected.
2217 (In the case of an ACL rejection, they have all been read.)
2219 PH/69 The HTML version of the specification is now built in a directory called
2220 spec_html instead of spec.html, because the latter looks like a path with
2221 a MIME-type, and this confuses some software.
2223 PH/70 Catch two compiler warnings in sieve.c.
2225 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
2226 function verify_get_ident() calls ip_connect() to connect a socket, but
2227 if the "connect()" function timed out, ip_connect() used to close the
2228 socket. However, verify_get_ident() also closes the socket later, and in
2229 between Exim writes to the log, which may get opened at this point. When
2230 the socket was closed in ip_connect(), the log could get the same file
2231 descriptor number as the socket. This naturally causes chaos. The fix is
2232 not to close the socket in ip_connect(); the socket should be closed by
2233 the function that creates it. There was only one place in the code where
2234 this was missing, in the iplookup router, which I don't think anybody now
2235 uses, but I've fixed it anyway.
2237 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
2238 well as to direct DNS lookups. Otherwise the handling of names in host
2239 lists is inconsistent and therefore confusing.
2245 PH/01 Two changes to the default runtime configuration:
2247 (1) Move the checks for relay_from_hosts and authenticated clients from
2248 after to before the (commented out) DNS black list checks.
2250 (2) Add control=submission to the relay_from_hosts and authenticated
2251 clients checks, on the grounds that messages accepted by these
2252 statements are most likely to be submissions.
2254 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
2256 (1) Generate an error if the third argument for the ${prvs expansion is
2259 (2) Treat a missing third argument of ${prvscheck as if it were an empty
2262 (3) Reset the variables that are obtained from the first argument of
2263 ${prvscheck and used in the second argument before leaving the code,
2264 because their memory is reclaimed, so using them afterwards may do
2267 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
2268 one (it's much easier than Tom thought :-).
2270 (5) Because of (4), we can now allow for the use of $prvscheck_result
2271 inside the third argument.
2273 PH/03 For some reason, the default setting of PATH when running a command from
2274 a pipe transport was just "/usr/bin". I have changed it to
2277 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
2278 anything to be listed in the output from -bV.
2280 PH/05 When a filter generated an autoreply, the entire To: header line was
2281 quoted in the delivery log line, like this:
2283 => >A.N.Other <ano@some.domain> <original@ddress> ...
2285 This has been changed so that it extracts the operative address. There
2286 may be more than one such address. If so, they are comma-separated, like
2289 => >ano@some.domain,ona@other.domain <original@ddress> ...
2291 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
2292 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
2293 not being looked up in the rDNS to get a host name, Exim was showing the
2294 IP address twice in Received: lines, even though the IP addresses were
2295 identical. For example:
2297 Received: from [1.2.3.4] (helo=[1.2.3.4])
2299 However, if the real host name was known, it was omitting the HELO data
2300 if it matched the actual IP address. This has been tidied up so that it
2301 doesn't show the same IP address twice.
2303 PH/07 When both +timestamp and +memory debugging was on, the value given by
2304 $tod_xxx expansions could be wrong, because the tod_stamp() function was
2305 called by the debug printing, thereby overwriting the timestamp buffer.
2306 Debugging no longer uses the tod_stamp() function when +timestamp is set.
2308 PH/08 When the original message was included in an autoreply transport, it
2309 always said "this is a copy of the message, including all the headers",
2310 even if body_only or headers_only was set. It now gives an appropriate
2313 PH/09 Applied a patch from the Sieve maintainer which:
2315 o fixes some comments
2316 o adds the (disabled) notify extension core
2317 o adds some debug output for the result of if/elsif tests
2318 o points to the current vacation draft in the documentation
2319 and documents the missing references header update
2323 o fixes a bug in processing the envelope test (when testing
2324 multiple envelope elements, the last element determinted the
2327 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
2328 Electronic Mail") by including:
2330 Auto-submitted: auto-generated
2332 in the messages that it generates (bounce messages and others, such as
2333 warnings). In the case of bounce messages for non-SMTP mesages, there was
2334 also a typo: it was using "Auto_submitted" (underscore instead of
2335 hyphen). Since every message generated by Exim is necessarily in response
2336 to another message, thes have all been changed to:
2338 Auto-Submitted: auto-replied
2340 in accordance with these statements in the RFC:
2342 The auto-replied keyword:
2344 - SHOULD be used on messages sent in direct response to another
2345 message by an automatic process,
2347 - MUST NOT be used on manually-generated messages,
2349 - MAY be used on Delivery Status Notifications (DSNs) and Message
2350 Disposition Notifications (MDNs),
2352 - MUST NOT be used on messages generated by automatic or periodic
2353 processes, except for messages which are automatic responses to
2356 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
2357 to the default Received: header definition.
2359 PH/12 Added log selector acl_warn_skipped (default on).
2361 PH/13 After a successful wildlsearch lookup, discard the values of numeric
2362 variables because (a) they are in the wrong storage pool and (b) even if
2363 they were copied, it wouldn't work properly because of the caching.
2365 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
2366 checking when decoding. Apparently there are clients that generate
2367 overlong encoded strings. Why am I not surprised?
2369 PH/15 If the first argument of "${if match_address" was not empty, but did not
2370 contain an "@" character, Exim crashed. Now it writes a panic log message
2371 and treats the condition as false.
2373 PH/16 In autoreply, treat an empty string for "once" the same as unset.
2375 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
2376 extension "envelope-auth". The code is finished and in agreement with
2377 other implementations, but there is no documentation so far and in fact,
2378 nobody wrote the draft yet. This extension is currently #undef'ed, thus
2379 not changing the active code.
2381 Print executed "if" and "elsif" statements when debugging is used. This
2382 helps a great deal to understand what a filter does.
2384 Document more things not specified clearly in RFC3028. I had all this
2385 sorted out, when out of a sudden new issues came to my mind. Oops."
2387 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
2390 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
2391 canonical form (as documented). However, after a host name lookup from
2392 the IP address, check_host() was doing a simple string comparison with
2393 addresses acquired from the DNS when checking that the found name did
2394 have the original IP as one of its addresses. Since any found IPv6
2395 addresses are likely to be in abbreviated form, the comparison could
2396 fail. Luckily, there already exists a function for doing the comparison
2397 by converting both addresses to binary, so now that is used instead of
2398 the text comparison.
2400 PH/20 There was another similar case to PH/19, when a complete host name was
2401 given in a host list; looking up its IP address could give an abbreviated
2402 form, whereas the current host's name might or might not be abbreviated.
2403 The same fix has been applied.
2409 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
2410 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
2413 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
2414 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
2416 PH/03 Typo: missing ".o" in src/pcre/Makefile.
2418 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
2419 header line, restrict the check to what is listed in RFCs 2369 and 2929.
2420 Also, for "Auto-Submitted", treat anything other than "no" as
2421 non-personal, in accordance with RFC 3834. (Previously it treated
2422 anything starting "auto-" as non-personal.)
2424 TF/01 The control=submission/name=... option had a problem with syntax
2425 errors if the name included a slash character. The /name= option
2426 now slurps the rest of the string, so it can include any characters
2427 but it must come last in the list of options (after /sender_retain
2430 PH/05 Some modifications to the interface to the fake nameserver for the new
2438 TK/01 Added the "success_on_redirect" address verification option. See
2439 NewStuff for rationale and an example.
2441 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
2443 PH/02 Patch to exigrep to allow it to work on syslog lines.
2445 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
2446 fread() to skip over the body file's header line, because in Cygwin the
2447 header line is locked and is inaccessible.
2449 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
2450 co-exist for some time) to make it clear that it is the Exim ID that is
2451 referenced, not the Message-ID: header line.
2453 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
2454 string_format() function, because snprintf() does not exist on all
2457 PH/06 The use of forbid_filter_existstest now also locks out the use of the
2458 ${stat: expansion item.
2460 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
2461 protocol synchronization error", to keep the pedants happy.
2463 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
2464 well as for IRIX systems, when gcc is being used. See the host.c source
2467 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
2469 PH/10 Named domain lists were not working if used in a queue_smtp_domains
2472 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
2473 transport and to the smtp transport in LMTP mode.
2475 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
2477 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
2478 run a filter in a subprocess. This could lead to confusion in subsequent
2479 lookups in the parent process. There should also be a search_tidyup() at
2480 the end of the subprocess.
2482 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
2483 only if the host matched helo_try_verify_hosts, which caused the
2484 verification to occur when the EHLO/HELO command was issued. The ACL just
2485 tested the remembered result. Now, if a previous verification attempt has
2486 not happened, "verify = helo" does it there and then.
2488 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
2490 TK/03 Fix log output including CR from clamd.
2492 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
2493 exist provoked a memory error which could cause a segfault.
2495 PH/15 Installed PCRE 6.2
2497 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
2499 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
2500 of the problem. Specifically, suggested +O2 rather than +O1 for the
2503 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
2505 PH/20 If a delivery was routed to a non-standard port by means of an SRV
2506 record, the port was not correctly logged when the outgoing_port log
2507 selector was set (it logged the transort's default port).
2509 PH/21 Added support for host-specific ports to manualroute, queryprogram,
2510 fallback_hosts, and "hosts" in the smtp transport.
2512 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
2513 host errors such as "Connection refused".
2515 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
2516 authentication with radiusclient 0.4.9:
2518 - Error returned from rc_read_config was caught wrongly
2519 - Username/password not passed on to radius server due to wrong length.
2521 The presumption is that some radiusclient API changes for 4.51/PH/17
2522 were not taken care of correctly. The code is still untested by me (my
2523 Linux distribution still has 0.3.2 of radiusclient), but it was
2524 contributed by a Radius user.
2526 PH/24 When doing a callout, the value of $domain wasn't set correctly when
2527 expanding the "port" option of the smtp transport.
2529 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
2530 while reading a MIME header. Thanks to Tom Hughes for a patch.
2532 PH/24 Include config.h inside local_scan.h so that configuration settings are
2535 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
2536 that in an ACL for RCPT (for example), you can examine exactly what was
2539 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
2540 commands, but it was not correctly comparing the address with the actual
2541 client host address. Thus, it would show the EHLO address in Received:
2542 header lines when this was not necessary.
2544 PH/27 Added the % operator to ${eval:}.
2546 PH/28 Exim tries to create and chdir to its spool directory when it starts;
2547 it should be ignoring failures (because with -C, for example, it has lost
2548 privilege). It wasn't ignoring creation failures other than "already
2551 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
2554 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
2555 because an input line was too long, either on its own, or by virtue of
2556 too many continuations, the temporary file was not being removed, and the
2557 return code was incorrect.
2559 PH/31 Missing "BOOL" in function definition in filtertest.c.
2561 PH/32 Applied Sieve patches from the maintainer.
2563 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
2565 PH/33 Added "verify = not_blind".
2567 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
2568 Local/Makefile (with some defaults set). These are used in built scripts
2569 such as exicyclog, but they have never been used in the exim_install
2570 script (though there are many overriding facilities there). I have
2571 arranged that the exim_install script now takes note of these two
2574 PH/35 Installed configuration files for Dragonfly.
2576 PH/36 When a locally submitted message by a trusted user did not contain a
2577 From: header, and the sender address was obtained from -f or from an SMTP
2578 MAIL command, and the trusted user did not use -F to supply a sender
2579 name, $originator_name was incorrectly used when constructing a From:
2580 header. Furthermore, $originator_name was used for submission mode
2581 messages from external hosts without From: headers in a similar way,
2582 which is clearly wrong.
2584 PH/37 Added control=suppress_local_fixups.
2586 PH/38 When log_selector = +received_sender was set, and the addition of the
2587 sender made the log line's construction buffer exactly full, or one byte
2588 less than full, an overflow happened when the terminating "\n" was
2591 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
2592 when the result of a list match is failure because a DNS lookup failed.
2594 PH/40 RM_COMMAND is now used in the building process.
2596 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
2597 the "build-* directories that it finds.
2599 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
2600 address in a domain literal was a prefix of an interface address.
2602 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
2603 when verifying a sender address, unless rewrite_headers is false.
2605 PH/44 Wrote a long comment about why errors_to addresses are verified as
2606 recipients, not senders.
2608 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
2609 the ratelimit ACL was added.
2611 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
2613 PH/46 Added extra information about PostgreSQL errors to the error string.
2615 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
2616 avoiding the need to install special zones in a real server. This is
2617 backwards compatible; if it can't find the fake resolver, it drops back.
2618 Thus, both old and new test suites can be run.
2620 TF/02 Added util/ratelimit.pl
2622 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
2623 clock is set back in time.
2625 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
2626 Candler <B.Candler@pobox.com>.
2628 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
2629 OK for addresses that are the result of redirections.
2631 PH/48 A number of further additions for the benefit of the new test suite,
2632 including a fake gethostbyname() that interfaces to the fake DNS resolver
2635 TF/06 The fix for widen_domains has also been applied to qualify_single and
2636 search_parents which are the other dnslookup options that can cause
2639 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
2642 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
2644 TF/07 Exim produced the error message "an SRV record indicated no SMTP
2645 service" if it encountered an MX record with an empty target hostname.
2646 The message is now "an MX or SRV record indicated no SMTP service".
2648 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
2649 if the DNS of the sending site is misconfigured. This is quite a
2650 common situation. This change restores the behaviour of treating a
2651 helo verification defer as a failure.
2653 PH/51 If self=fail was set on a router, the bounce message did not include the
2654 actual error message.
2660 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
2662 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
2663 command itself ended in error, the underlying message about the transport
2664 filter timeout was being overwritten with the pipe command error. Now the
2665 underlying error message should be appended to the second error message.
2667 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
2669 PH/02 Reduce the amount of output that "make" produces by default. Full output
2670 can still be requested.
2672 PH/03 The warning log line about a condition test deferring for a "warn" verb
2673 was being output only once per connection, rather than after each
2674 occurrence (because it was using the same function as for successful
2675 "warn" verbs). This seems wrong, so I have changed it.
2677 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
2678 it should not have, which might have caused a crash in the right
2679 circumstances, but probably never did.
2681 PH/04 Installed a modified version of Tony Finch's patch to make submission
2682 mode fix the return path as well as the Sender: header line, and to
2683 add a /name= option so that you can make the user's friendly name appear
2686 TF/03 Added the control = fakedefer ACL modifier.
2688 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
2689 Mark Lowes for thorough testing.
2691 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
2693 TK/03 Merged latest SRS patch from Miles Wilton.
2695 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
2696 with the definition in sysexits.h (which is #included earlier).
2697 Fortunately, Exim does not actually use EX_OK. The code used to try to
2698 preserve the sysexits.h value, by assumimg that macro definitions were
2699 scanned for macro replacements. I have been disabused of this notion,
2700 so now the code just undefines EX_OK before #including unistd.h.
2702 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
2703 in the smtp transport. When a block could not be written in a single
2704 write() function, the timeout was being re-applied to each part-write.
2705 This seems wrong - if the receiver was accepting one byte at a time it
2706 would take for ever. The timeout is now adjusted when this happens. It
2707 doesn't have to be particularly precise.
2709 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
2710 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
2712 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
2713 without a domain if the check to <postmaster@domain> fails.
2715 SC/01 Eximstats: added -xls and the ability to specify output files
2716 (patch written by Frank Heydlauf).
2718 SC/02 Eximstats: use FileHandles for outputing results.
2720 SC/03 Eximstats: allow any combination of xls, txt, and html output.
2722 SC/04 Eximstats: fixed display of large numbers with -nvr option
2724 SC/05 Eximstats: fixed merging of reports with empty tables.
2726 SC/06 Eximstats: added the -include_original_destination flag
2728 SC/07 Eximstats: removed tabs and trailing whitespace.
2730 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
2732 TK/06 MBOX spool code: Add real "From " MBOX separator line
2733 so the .eml file is really in mbox format (even though
2734 most programs do not really care). Patch from Alex Miller.
2736 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
2737 The latter is generated from $received_to and is only set if the
2738 message has one envelope recipient. SA can use these headers,
2739 obviously out-of-the-box. Patch from Alex Miller.
2741 PH/08 The ${def test on a variable was returning false if the variable's
2742 value was "0", contrary to what the specification has always said!
2743 The result should be true unless the variable is empty.
2745 PH/09 The syntax error of a character other than { following "${if
2746 def:variable_name" (after optional whitespace) was not being diagnosed.
2747 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
2748 accidental colon was present, for example, could give incorrect results.
2750 PH/10 Tidied the code in a number of places where the st_size field of a stat()
2751 result is used (not including appendfile, where other changes are about
2754 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
2755 This involved changing a lot of size variables from int to off_t. It
2756 should work with maildirs and everything.
2758 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
2759 spamd dying while we are connected to it.
2761 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
2764 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
2765 function for BDB 4.3.
2767 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
2769 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
2770 the sources for PCRE have been re-arranged and more files are now
2773 PH/15 The code I had for printing potentially long long variables in PH/11
2774 above was not the best (it lost precision). The length of off_t variables
2775 is now inspected at build time, and an appropriate printing format (%ld
2776 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
2777 to be "long long int" or "long int". This is needed for the internal
2778 formatting function string_vformat().
2780 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
2781 the configuration file to be ":syslog", then the script "guesses" where
2782 the logs files are, rather than using the compiled in default. In our
2783 case the guess is not the same as the compiled default, so the script
2784 suddenly stopped working when I started to use syslog. The patch checks
2785 to see if log_file_path is "". If so, it attempts to read it from exim
2786 with no configuration file to get the compiled in version, before it
2787 falls back to the previous guessing code."
2789 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
2790 implementing BATV in an Exim configuration. See NewStuff for the gory
2793 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
2794 Makefile that are specific to HP-UX.
2796 PH/18 If the "use_postmaster" option was set for a recipient callout together
2797 with the "random" option, the postmaster address was used as the MAIL
2798 FROM address for the random test, but not for the subsequent recipient
2799 test. It is now used for both.
2801 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
2802 patch removes a few documentation additions to RFC 3028, because the
2803 latest draft now contains them. It adds the new en;ascii-case comparator
2804 and a new error check for 8bit text in MIME parts. Comparator and
2805 require names are now matched exactly. I enabled the subaddress
2806 extension, but it is not well tested yet (read: it works for me)."
2808 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
2809 rework some of the code of TK/09 above to avoid the hardwired use of
2810 "%lld" and "long long". Replaced the call to snprintf() with a call to
2813 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
2814 records point to non-existent hosts", "retry timeout exceeded", and
2815 "retry time not reached for any host after a long failure period".
2817 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
2818 experimental DomainKeys support:
2820 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
2821 (2) On an error such as an illegally used "control", the wrong name for
2822 the control was given.
2824 These problems did NOT occur unless DomainKeys support was compiled.
2826 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
2828 PH/24 Added ${if match_ip condition.
2830 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
2831 databases so that it will be absolutely obvious if a crash occurs in the
2832 DB library. This is a regular occurrence (often caused by mis-matched
2835 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
2836 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
2837 picked up on a user's system that detects such things. There doesn't seem
2838 to be a gcc warning option for this - only an attribute that has to be
2839 put on the function's prototype. It seems that in Fedora Core 4 they have
2840 set this on a number of new functions. No doubt there will be more in due
2843 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
2844 specify a transport. However, if an address that was verified by such a
2845 router was the subject of a callout, Exim crashed because it tried to
2846 read the rcpt_include_affixes from the non-existent transport. Now it
2847 just assumes that the setting of that option is false. This bug was
2848 introduced by 4.51/PH/31.
2850 PH/28 Changed -d+all to exclude +memory, because that information is very
2851 rarely of interest, but it makes the output a lot bigger. People tend to
2852 do -d+all out of habit.
2854 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
2855 code in os-type was giving problems when libc.so lives in lib64, like on
2858 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
2859 aren't the modern standard, and it seems that some systems' include files
2860 don't always have them. Exim was already checking for some of the newer
2861 ones like T_AAAA, and defining it itself. I've added checks for all the
2862 record types that Exim uses.
2864 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
2865 not automatically generating a new one, as it is supposed to. This
2866 prevented TLS from working. If the file did exist, but contained invalid
2867 data, a new version was generated, as expected. It was only the case of a
2868 non-existent file that was broken.
2870 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
2871 with a change in libdomainkeys > 0.64.
2873 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
2874 from DNS. If the selector record carries the flag, it now has
2875 precedence over the domain-wide flag.
2877 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
2879 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
2880 the use of a port name that isn't defined in /etc/services) occurred, the
2881 message was deferred as in a normal delivery, and thus remained on the
2882 spool, instead of being failed because of the mua_wrapper setting. This
2883 is now fixed, and I tidied up some of the mua_wrapper messages at the
2886 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
2887 the messages in a hash of arrays rather than using individual hashes.
2888 This is a bit cleaner and results in dramatic memory savings, albeit
2889 at a slight CPU cost.
2891 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
2892 as requested by Marc Sherman.
2894 SC/10 Eximstats: added histograms for user specified patterns as requested
2897 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
2899 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
2900 fopen() in the content-scanning modules that did not already have it.
2906 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
2907 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
2909 TK/02 Fix ACL "control" statement not being available in MIME ACL.
2911 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
2913 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
2914 to test Sieve filters that use "vacation".
2916 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
2917 that changes the way the GnuTLS parameters are stored in the cache file.
2918 The new format can be generated externally. For backward compatibility,
2919 if the data in the cache doesn't make sense, Exim assumes it has read an
2920 old-format file, and it generates new data and writes a new file. This
2921 means that you can't go back to an older release without removing the
2924 PH/03 A redirect router that has both "unseen" and "one_time" set does not
2925 work if there are any delivery delays because "one_time" forces the
2926 parent to be marked "delivered", so its unseen clone is never tried
2927 again. For this reason, Exim now forbids the simultaneous setting of
2930 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
2931 redirected to themselves ("homonym" addresses). Read the long ChangeLog
2932 entry if you want to know the details. The fix, however, neglected to
2933 consider the case when local delivery batching is involved. The test for
2934 "previously delivered" was not happening when checking to see if an
2935 address could be batched with a previous (undelivered) one; under
2936 certain circumstances this could lead to multiple deliveries to the same
2939 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
2940 in its include files, and this causes problems building Exim.
2942 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
2943 header_syntax) but Exim was just ignoring anything given after a slash.
2944 In particular, this caused confusion with an attempt to use "verify =
2945 reverse_host_lookup/defer_ok". An error is now given when options are
2946 supplied for verify items that do not have them. (Maybe reverse_host_
2947 lookup should have a defer_ok option, but that's a different point.)
2949 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
2950 defined by RFC 821) to 2048, because there were problems with some AUTH
2951 commands, and RFC 1869 says the size should be increased for extended
2952 SMTP commands that take arguments.
2954 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
2957 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
2958 "unknown" error; now it says that the functionality isn't in the binary.
2960 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
2961 an address' error message when a string expansion fails (syntax or
2962 whatever). Otherwise the password may appear in the log. Following change
2963 PH/42 below, there is no longer a chance of it appearing in a bounce
2966 PH/11 Installed exipick version 20050225.0 from John Jetmore.
2968 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
2969 of its addresses was ever tried. (Bugzilla bug #2.)
2971 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
2972 the result incorrectly in the debug output. (It correctly added a newline
2973 to what was transported.)
2975 TF/01 Added $received_time.
2977 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
2978 commented out examples of how to interface to a virus scanner and to
2979 SpamAssassin. Also added commented examples of av_scanner and
2980 spamd_address settings.
2982 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
2983 and controls are allowed in which ACLs. There were a couple of minor
2984 errors. Some of the entries in the conditions table (which is a table of
2985 where they are NOT allowed) were getting very unwieldy; rewrote them as a
2986 negation of where the condition IS allowed.
2988 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
2990 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
2991 header file does not have a version number, so I've had to invent a new
2992 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
2993 API. The code is untested by me (my Linux distribution still has 0.3.2 of
2994 radiusclient), but it was contributed by a Radius user.
2996 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
2997 files or directories, for OpenSSL.
2999 PH/19 When an Exim process that is running as root has to create an Exim log
3000 file, it does so in a subprocess that runs as exim:exim so as to get the
3001 ownership right at creation (otherwise, other Exim processes might see
3002 the file with the wrong ownership). There was no test for failure of this
3003 fork() call, which would lead to the process getting stuck as it waited
3004 for a non-existent subprocess. Forks do occasionally fail when resources
3005 run out. I reviewed all the other calls to fork(); they all seem to check
3008 PH/20 When checking for unexpected SMTP input at connect time (before writing
3009 the banner), Exim was not dealing correctly with a non-positive return
3010 from the read() function. If the client had disconnected by this time,
3011 the result was a log entry for a synchronization error with an empty
3012 string after "input=" when read() returned zero. If read() returned -1
3013 (an event I could not check), uninitialized data bytes were printed.
3014 There were reports of junk text (parts of files, etc) appearing after
3017 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
3019 PH/22 Added support for macro redefinition, and (re)definition in between
3020 driver and ACL definitions.
3022 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
3023 forgetting to use the resulting value; it was using the unexpanded value.
3025 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
3026 hadn't been configured. The fix is from Juergen Kreileder, who
3027 understands it better than I do:
3029 "Here's what I see happening with three configured cyrus_sasl
3030 authenticators configured (plain, login, cram-md5):
3032 On startup auth_cyrus_sasl_init() gets called for each of these.
3033 This means three calls to sasl_listmech() without a specified mech_list.
3034 => SASL tests which mechs of all available mechs actually work
3035 => three warnings about OTP not working
3036 => the returned list contains: plain, login, cram-md5, digest-md5, ...
3038 With the patch, sasl_listmech() also gets called three times. But now
3039 SASL's mech_list option is set to the server_mech specified in the the
3040 authenticator. Or in other words, the answer from sasl_listmech()
3041 gets limited to just the mech you're testing for (which is different
3043 => the return list contains just 'plain' or 'login', 'cram-md5' or
3044 nothing depending on the value of ob->server_mech.
3046 I've just tested the patch: Authentication still works fine,
3047 unavailable mechs specified in the exim configuration are still
3048 caught, and the auth.log warnings about OTP are gone."
3050 PH/25 When debugging is enabled, the contents of the command line are added
3051 to the debugging output, even when log_selector=+arguments is not
3054 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
3055 answer is "GNU", and only if the return is "GNU/something" is the answer
3058 PH/27 $acl_verify_message is now set immediately after the failure of a
3059 verification in an ACL, and so is available in subsequent modifiers. In
3060 particular, the message can be preserved by coding like this:
3062 warn !verify = sender
3063 set acl_m0 = $acl_verify_message
3065 Previously, $acl_verify_message was set only while expanding "message"
3066 and "log_message" when a very denied access.
3068 PH/28 Modified OS/os.c-Linux with
3070 -#ifndef OS_LOAD_AVERAGE
3071 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
3073 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
3074 nomenclature these days.)
3076 PH/29 Installed patch from the Sieve maintainer that adds the options
3077 sieve_useraddress and sieve_subaddress to the redirect router.
3079 PH/30 In these circumstances:
3080 . Two addresses routed to the same list of hosts;
3081 . First host does not offer TLS;
3082 . First host accepts first address;
3083 . First host gives temporary error to second address;
3084 . Second host offers TLS and a TLS session is established;
3085 . Second host accepts second address.
3086 Exim incorrectly logged both deliveries with the TLS parameters (cipher
3087 and peerdn, if requested) that were in fact used only for the second
3090 PH/31 When doing a callout as part of verifying an address, Exim was not paying
3091 attention to any local part prefix or suffix that was matched by the
3092 router that accepted the address. It now behaves in the same way as it
3093 does for delivery: the affixes are removed from the local part unless
3094 rcpt_include_affixes is set on the transport.
3096 PH/32 Add the sender address, as F=<...>, to the log line when logging a
3097 timeout during the DATA phase of an incoming message.
3099 PH/33 Sieve envelope tests were broken for match types other than :is. I have
3100 applied a patch sanctioned by the Sieve maintainer.
3102 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
3103 the uid or gid is negative. A case of a negative gid caused this to be
3104 noticed. The fix allows for either to be negative.
3106 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
3107 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
3110 PH/36 The change PH/12 above was broken. Fixed it.
3112 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
3113 the grounds that routing the same address twice would always produce the
3114 same answer. This might have been true once, but it is certainly no
3115 longer true now. Routing a child address may depend on the previous
3116 routing that produced that child. Some complicated redirection strategies
3117 went wrong when messages had multiple recipients, and made Exim's
3118 behaviour dependent on the order in which the addresses were given.
3120 I have moved the duplicate checking until after the routing is complete.
3121 Exim scans the addresses that are assigned to local and remote
3122 transports, and removes any duplicates. This means that more work will be
3123 done, as duplicates will always all be routed, but duplicates are
3124 presumably rare, so I don't expect this is of any significance.
3126 For deliveries to pipes, files, and autoreplies, the duplicate checking
3127 still happens during the routing process, since they are not going to be
3130 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
3131 It corrects a timeout issue with spamd. This is Ian's comment: "The
3132 background is that sometimes spamd either never reads data from a
3133 connection it has accepted, or it never writes response data. The exiscan
3134 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
3135 blindly assumes that writes won't block so it may never time out."
3137 PH/39 Allow G after quota size as well as K and M.
3139 PH/40 The value set for $authenticated_id in an authenticator may not contain
3140 binary zeroes or newlines because the value is written to log lines and
3141 to spool files. There was no check on this. Now the value is run through
3142 the string_printing() function so that such characters are converted to
3143 printable escape sequences.
3145 PH/41 $message_linecount is a new variable that contains the total number of
3146 lines in the message. Compare $body_linecount, which is the count for the
3149 PH/42 Exim no longer gives details of delivery errors for specific addresses in
3150 bounce and delay warning messages, except in certain special cases, which
3153 (a) An SMTP error message from a remote host;
3154 (b) A message specified in a :fail: redirection;
3155 (c) A message specified in a "fail" command in a system filter;
3156 (d) A message specified in a FAIL return from the queryprogram router;
3157 (e) A message specified by the cannot_route_message router option.
3159 In these cases only, Exim does include the error details in bounce and
3160 warning messages. There are also a few cases where bland messages such
3161 as "unrouteable address" or "local delivery error" are given.
3163 PH/43 $value is now also set for the "else" part of a ${run expansion.
3165 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
3166 being worked on, but at least Exim now implements the latest version to
3169 PH/45 In a pipe transport, although a timeout while waiting for the pipe
3170 process to complete was treated as a delivery failure, a timeout while
3171 writing the message to the pipe was logged, but erroneously treated as a
3172 successful delivery. Such timeouts include transport filter timeouts. For
3173 consistency with the overall process timeout, these timeouts are now
3174 treated as errors, giving rise to delivery failures by default. However,
3175 there is now a new Boolean option for the pipe transport called
3176 timeout_defer, which, if set TRUE, converts the failures into defers for
3177 both kinds of timeout. A transport filter timeout is now identified in
3180 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
3181 systems where "make" and "gmake" are different, calling "gmake" at top
3182 level broke things. I've arranged for the value of $(MAKE) to be passed
3183 from the Makefile to this script so that it can call the same version of
3187 A note about Exim versions 4.44 and 4.50
3188 ----------------------------------------
3190 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
3191 changes of various kinds. As a consequence, a big documentation update was
3192 needed. This delayed the release for rather longer than seemed good, especially
3193 in the light of a couple of (minor) security issues. Therefore, the changes
3194 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
3195 release. So 4.44 and 4.50 are in effect two different branches that both start
3198 I have left the 4.50 change log unchanged; it contains all the changes since
3199 4.43. The change log for 4.44 is below; many of its items are identical to
3200 those for 4.50. This seems to be the most sensible way to preserve the
3201 historical information.
3207 1. Minor wording change to the doc/README.SIEVE file.
3209 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
3210 computation of the current number of files was incorrect.
3212 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
3213 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
3216 4. Give more explanation in the error message when the command for a transport
3217 filter fails to execute.
3219 5. There are several places where Exim runs a non-Exim command in a
3220 subprocess. The SIGUSR1 signal should be disabled for these processes. This
3221 was being done only for the command run by the queryprogram router. It is
3222 now done for all such subprocesses. The other cases are: ${run, transport
3223 filters, and the commands run by the lmtp and pipe transports.
3225 6. Added CONFIGURE_GROUP build-time option.
3227 7. Some older OS have a limit of 256 on the maximum number of file
3228 descriptors. Exim was using setrlimit() to set 1000 as a large value
3229 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
3230 systems. I've change it so that if it can't get 1000, it tries for 256.
3232 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
3233 was an oversight, and furthermore, ever since the addition of extra
3234 controls (e.g. 4.43/32), the checks on when to allow different forms of
3235 "control" were broken. There should now be diagnostics for all cases when a
3236 control that does not make sense is encountered.
3238 9. Added the /retain_sender option to "control=submission".
3240 10. $recipients is now available in the predata ACL (oversight).
3242 11. Tidy the search cache before the fork to do a delivery from a message
3243 received from the command line. Otherwise the child will trigger a lookup
3244 failure and thereby defer the delivery if it tries to use (for example) a
3245 cached ldap connection that the parent has called unbind on.
3247 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
3248 of $address_data from the recipient verification was clobbered by the
3249 sender verification.
3251 13. The value of address_data from a sender verification is now available in
3252 $sender_address_data in subsequent conditions in the ACL statement.
3254 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
3256 15. Added a new option "connect=<time>" to callout options, to set a different
3259 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
3260 was its contents. (It was OK if the option was not defined at all.)
3262 17. A "Completed" log line is now written for messages that are removed from
3263 the spool by the -Mrm option.
3265 18. New variables $sender_verify_failure and $recipient_verify_failure contain
3266 information about exactly what failed.
3268 19. Added -dd to debug only the daemon process.
3270 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
3271 handles timeouts, both on the server side and network timeouts. Renamed the
3272 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
3274 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
3275 It is now set to "smtps".
3277 22. $host_address is now set to the target address during the checking of
3278 ignore_target_hosts.
3280 23. When checking ignore_target_hosts for an ipliteral router, no host name was
3281 being passed; this would have caused $sender_host_name to have been used if
3282 matching the list had actually called for a host name (not very likely,
3283 since this list is usually IP addresses). A host name is now passed as
3286 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
3287 code that specifies a non-restarting handler (typically sigaction() in
3288 modern systems) in an attempt to fix a rare and obscure crash bug.
3290 25. Narrowed the window for a race in the daemon that could cause it to ignore
3291 SIGCHLD signals. This is not a major problem, because they are used only to
3292 wake it up if nothing else does.
3294 26. A malformed maildirsize file could cause Exim to calculate negative values
3295 for the mailbox size or file count. Odd effects could occur as a result.
3296 The maildirsize information is now recalculated if the size or filecount
3299 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
3300 support for a long time. Removed HAVE_SYS_VFS_H.
3302 28. Installed the latest version of exipick from John Jetmore.
3304 29. In an address list, if the pattern was not a regular expression, an empty
3305 subject address (from a bounce message) matched only if the pattern was an
3306 empty string. Non-empty patterns were not even tested. This was the wrong
3307 because it is perfectly reasonable to use an empty address as part of a
3308 database query. An empty address is now tested by patterns that are
3309 lookups. However, all the other forms of pattern expect the subject to
3310 contain a local part and a domain, and therefore, for them, an empty
3311 address still always fails if the pattern is not itself empty.
3313 30. Exim went into a mad DNS loop when attempting to do a callout where the
3314 host was specified on an smtp transport, and looking it up yielded more
3315 than one IP address.
3317 31. Re-factored the code for checking spool and log partition space into a
3318 function that finds that data and another that does the check. The former
3319 is then used to implement four new variables: $spool_space, $log_space,
3320 $spool_inodes, and $log_inodes.
3322 32. The RFC2047 encoding function was originally intended for short strings
3323 such as real names; it was not keeping to the 75-character limit for
3324 encoded words that the RFC imposes. It now respects the limit, and
3325 generates multiple encoded words if necessary. To be on the safe side, I
3326 have increased the buffer size for the ${rfc2047: expansion operator from
3329 33. It is now permitted to omit both strings after an "if" condition; if the
3330 condition is true, the result is "true". As before, when the second string
3331 is omitted, a false condition yields an empty string. This makes it less
3332 cumbersome to write custom ACL and router conditions.
3334 34. Failure to deliver a bounce message always caused it to be frozen, even if
3335 there was an errors_to setting on the router. The errors_to setting is now
3338 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
3339 canonical form (fully expanded) before being placed in
3340 $sender_host_address.
3342 36. The table in the code that translates DNS record types into text (T_A to
3343 "A" for instance) was missing entries for NS and CNAME. It is just possible
3344 that this could have caused confusion if both these types were looked up
3345 for the same domain, because the text type is used as part of Exim's
3346 per-process caching. But the chance of anyone hitting this buglet seems
3349 37. The dnsdb lookup has been extended in a number of ways.
3351 (1) There is a new type, "zns", which walks up the domain tree until it
3352 finds some nameserver records. It should be used with care.
3354 (2) There is a new type, "mxh", which is like "mx" except that it returns
3355 just the host names, not the priorities.
3357 (3) It is now possible to give a list of domains (or IP addresses) to be
3358 looked up. The behaviour when one of the lookups defers can be
3359 controlled by a keyword.
3361 (4) It is now possible to specify the separator character for use when
3362 multiple records are returned.
3364 38. The dnslists ACL condition has been extended: it is now possible to supply
3365 a list of IP addresses and/or domains to be looked up in a particular DNS
3368 39. Added log_selector=+queue_time_overall.
3370 40. When running the queue in the test harness, wait just a tad after forking a
3371 delivery process, to get repeatability of debugging output.
3373 41. Include certificate and key file names in error message when GnuTLS fails
3374 to set them up, because the GnuTLS error message doesn't include the name
3375 of the failing file when there is a problem reading it.
3377 42. Allow both -bf and -bF in the same test run.
3379 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
3381 44. The "Exiscan patch" is now merged into the mainline Exim source.
3383 45. Sometimes the final signoff response after QUIT could fail to get
3384 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
3385 < 0 before doing a fflush(). This bug looks as though it goes back to the
3386 introduction of TLS in release 3.20, but "sometimes" must have been rare
3387 because the tests only now provoked it.
3389 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
3390 (this can affect the format of dates).
3392 47. exim_tidydb, when checking for the continued existence of a message for
3393 which it has found a message-specific retry record, was not finding
3394 messages that were in split spool directories. Consequently, it was
3395 deleting retry records that should have stayed in existence.
3397 48. Steve fixed some bugs in eximstats.
3399 49. The SPA authentication driver was not abandoning authentication and moving
3400 on to the next authenticator when an expansion was forced to fail,
3401 contradicting the general specification for all authenticators. Instead it
3402 was generating a temporary error. It now behaves as specified.
3404 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
3405 (the order specifies the preference for clients). The order is now AES256,
3406 AES128, 3DES, ARCFOUR128.
3408 51. Small patch to Sieve code - explicitly set From: when generating an
3411 52. Exim crashed if a remote delivery caused a very long error message to be
3412 recorded - for instance if somebody sent an entire SpamAssassin report back
3413 as a large number of 550 error lines. This bug was coincidentally fixed by
3414 increasing the size of one of Exim's internal buffers (big_buffer) that
3415 happened as part of the Exiscan merge. However, to be on the safe side, I
3416 have made the code more robust (and fixed the comments that describe what
3419 53. Now that there can be additional text after "Completed" in log lines (if
3420 the queue_time_overall log selector is set), a one-byte patch to exigrep
3421 was needed to allow it to recognize "Completed" as not the last thing in
3424 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
3425 patch that reportedly fixes this has been added. I am not expert enough to
3426 create a test for it. This is what the patch creator wrote:
3428 "I found a little strange behaviour of ldap code when working with
3429 Windows 2003 AD Domain, where users was placed in more than one
3430 Organization Units. When I tried to give exim partial DN, the exit code
3431 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
3432 But simultaneously result of request was absolutely normal ldap result,
3433 so I produce this patch..."
3435 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
3436 REFERENCE, so I have modified the code to exclude the patch when that macro
3439 55. Some experimental protocols are using DNS PTR records for new purposes. The
3440 keys for these records are domain names, not reversed IP addresses. The
3441 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
3442 leaves it alone. Component reversal etc. now happens only for IP addresses.
3445 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
3447 57. Double the size of the debug message buffer (to 2048) so that more of very
3448 long debug lines gets shown.
3450 58. The exicyclog utility now does better if the number of log files to keep
3451 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
3453 59. Two changes related to the smtp_active_hostname option:
3455 (1) $smtp_active_hostname is now available as a variable.
3456 (2) The default for smtp_banner uses $smtp_active_hostname instead
3457 of $primary_hostname.
3459 60. The host_aton() function is supposed to be passed a string that is known
3460 to be a valid IP address. However, in the case of IPv6 addresses, it was
3461 not checking this. This is a hostage to fortune. Exim now panics and dies
3462 if the condition is not met. A case was found where this could be provoked
3463 from a dnsdb PTR lookup with an IPv6 address that had more than 8
3464 components; fortuitously, this particular loophole had already been fixed
3465 by change 4.50/55 above.
3467 If there are any other similar loopholes, the new check in host_aton()
3468 itself should stop them being exploited. The report I received stated that
3469 data on the command line could provoke the exploit when Exim was running as
3470 exim, but did not say which command line option was involved. All I could
3471 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
3472 running as the user.
3475 61. There was a buffer overflow vulnerability in the SPA authentication code
3476 (which came originally from the Samba project). I have added a test to the
3477 spa_base64_to_bits() function which I hope fixes it.
3480 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
3481 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
3483 63. The daemon start-up calls getloadavg() while still root for those OS that
3484 need the first call to be done as root, but it missed one case: when
3485 deliver_queue_load_max is set with deliver_drop_privilege. This is
3486 necessary for the benefit of the queue runner, because there is no re-exec
3487 when deliver_drop_privilege is set.
3489 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
3490 This has been fixed.
3492 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
3493 was in use, was not putting the data itself into the right store pool;
3494 consequently, it could be overwritten for a subsequent message in the same
3495 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
3498 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
3500 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
3501 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
3502 the function were treating the return as a boolean value, which happened to
3503 work because 0=false and not-0=true, but is not correct code.
3505 68. The host_aton() function was not handling scoped IPv6 addresses (those
3506 with, for example, "%eth0" on the end) correctly.
3508 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
3509 negated items (that is, ~something) in unsigned ints. Some compilers
3510 apparently mutter when there is no cast.
3512 70. If an address verification called from an ACL failed, and did not produce a
3513 user-specific message (i.e. there was only a "system" message), nothing was
3514 put in $acl_verify_message. In this situation, it now puts the system
3517 71. Change 4.23/11 added synchronization checking at the start of an SMTP
3518 session; change 4.31/43 added the unwanted input to the log line - except
3519 that it did not do this in the start of session case. It now does.
3521 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
3522 This is wrong and can cause the other end to generate a synchronization
3523 error if it is another Exim or anything else that does the synchronization
3524 check. A QUIT command is no longer sent after a timeout.
3526 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
3527 during host lookups.
3529 74. The defer_ok option of callout verification was not working if it was used
3530 when verifying addresses in header lines, that is, for this case:
3532 verify = header_sender/callout=defer_ok
3534 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
3535 those file descriptors could be used for SMTP connections. If anything
3536 wrote to stderr (the example that came up was "warn" in embedded Perl), it
3537 could be sent to the SMTP client, causing chaos. The daemon now opens
3538 stdin, stdout, and stderr to /dev/null when it puts itself into the
3541 76. Arrange for output from Perl's "warn" command to be written to Exim's main
3542 log by default. The user can override this with suitable Perl magic.
3544 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
3545 the log message when discard triggers, was not working for the DATA ACL or
3546 for the non-SMTP ACL.
3548 78. Error message wording change in sieve.c.
3550 79. If smtp_accept_max_per_host was set, the number of connections could be
3551 restricted to fewer than expected, because the daemon was trying to set up
3552 a new connection before checking whether the processes handling previous
3553 connections had finished. The check for completed processes is now done
3554 earlier. On busy systems, this bug wouldn't be noticed because something
3555 else would have woken the daemon, and it would have reaped the completed
3558 80. If a message was submitted locally by a user whose login name contained one
3559 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
3560 It caused a spool format error. I have fixed the spool reading code. A
3561 related problem was that the "from" clause in the Received: line became
3562 illegal because of the space(s). It is now covered by ${quote_local_part.
3564 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
3567 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
3569 83. Patch from Sieve maintainer to fix unterminated string problem in
3570 "vacation" handling.
3572 84. Some minor changes to the Linux configuration files to help with other
3573 OS variants using glibc.
3575 85. One more patch for Sieve to update vacation handling to latest spec.
3578 ----------------------------------------------------
3579 See the note above about the 4.44 and 4.50 releases.
3580 ----------------------------------------------------
3586 1. Change 4.43/35 introduced a bug that caused file counts to be
3587 incorrectly computed when quota_filecount was set in an appendfile
3590 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
3591 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
3594 3. Give more explanation in the error message when the command for a transport
3595 filter fails to execute.
3597 4. There are several places where Exim runs a non-Exim command in a
3598 subprocess. The SIGUSR1 signal should be disabled for these processes. This
3599 was being done only for the command run by the queryprogram router. It is
3600 now done for all such subprocesses. The other cases are: ${run, transport
3601 filters, and the commands run by the lmtp and pipe transports.
3603 5. Some older OS have a limit of 256 on the maximum number of file
3604 descriptors. Exim was using setrlimit() to set 1000 as a large value
3605 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
3606 systems. I've change it so that if it can't get 1000, it tries for 256.
3608 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
3609 was an oversight, and furthermore, ever since the addition of extra
3610 controls (e.g. 4.43/32), the checks on when to allow different forms of
3611 "control" were broken. There should now be diagnostics for all cases when a
3612 control that does not make sense is encountered.
3614 7. $recipients is now available in the predata ACL (oversight).
3616 8. Tidy the search cache before the fork to do a delivery from a message
3617 received from the command line. Otherwise the child will trigger a lookup
3618 failure and thereby defer the delivery if it tries to use (for example) a
3619 cached ldap connection that the parent has called unbind on.
3621 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
3622 of $address_data from the recipient verification was clobbered by the
3623 sender verification.
3625 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
3626 was its contents. (It was OK if the option was not defined at all.)
3628 11. A "Completed" log line is now written for messages that are removed from
3629 the spool by the -Mrm option.
3631 12. $host_address is now set to the target address during the checking of
3632 ignore_target_hosts.
3634 13. When checking ignore_target_hosts for an ipliteral router, no host name was
3635 being passed; this would have caused $sender_host_name to have been used if
3636 matching the list had actually called for a host name (not very likely,
3637 since this list is usually IP addresses). A host name is now passed as
3640 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
3641 code that specifies a non-restarting handler (typically sigaction() in
3642 modern systems) in an attempt to fix a rare and obscure crash bug.
3644 15. Narrowed the window for a race in the daemon that could cause it to ignore
3645 SIGCHLD signals. This is not a major problem, because they are used only to
3646 wake it up if nothing else does.
3648 16. A malformed maildirsize file could cause Exim to calculate negative values
3649 for the mailbox size or file count. Odd effects could occur as a result.
3650 The maildirsize information is now recalculated if the size or filecount
3653 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
3654 support for a long time. Removed HAVE_SYS_VFS_H.
3656 18. Updated exipick to current release from John Jetmore.
3658 19. Allow an empty sender to be matched against a lookup in an address list.
3659 Previously the only cases considered were a regular expression, or an
3662 20. Exim went into a mad DNS lookup loop when doing a callout where the
3663 host was specified on the transport, if the DNS lookup yielded more than
3666 21. The RFC2047 encoding function was originally intended for short strings
3667 such as real names; it was not keeping to the 75-character limit for
3668 encoded words that the RFC imposes. It now respects the limit, and
3669 generates multiple encoded words if necessary. To be on the safe side, I
3670 have increased the buffer size for the ${rfc2047: expansion operator from
3673 22. Failure to deliver a bounce message always caused it to be frozen, even if
3674 there was an errors_to setting on the router. The errors_to setting is now
3677 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
3678 canonical form (fully expanded) before being placed in
3679 $sender_host_address.
3681 24. Updated eximstats to version 1.33
3683 25. Include certificate and key file names in error message when GnuTLS fails
3684 to set them up, because the GnuTLS error message doesn't include the name
3685 of the failing file when there is a problem reading it.
3687 26. Expand error message when OpenSSL has problems setting up cert/key files.
3690 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
3691 (this can affect the format of dates).
3693 28. exim_tidydb, when checking for the continued existence of a message for
3694 which it has found a message-specific retry record, was not finding
3695 messages that were in split spool directories. Consequently, it was
3696 deleting retry records that should have stayed in existence.
3698 29. eximstats updated to version 1.35
3699 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
3700 1.35 - bugfix such that pie charts by volume are generated correctly
3702 30. The SPA authentication driver was not abandoning authentication and moving
3703 on to the next authenticator when an expansion was forced to fail,
3704 contradicting the general specification for all authenticators. Instead it
3705 was generating a temporary error. It now behaves as specified.
3707 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
3708 (the order specifies the preference for clients). The order is now AES256,
3709 AES128, 3DES, ARCFOUR128.
3711 31. Small patch to Sieve code - explicitly set From: when generating an
3714 32. Exim crashed if a remote delivery caused a very long error message to be
3715 recorded - for instance if somebody sent an entire SpamAssassin report back
3716 as a large number of 550 error lines. This bug was coincidentally fixed by
3717 increasing the size of one of Exim's internal buffers (big_buffer) that
3718 happened as part of the Exiscan merge. However, to be on the safe side, I
3719 have made the code more robust (and fixed the comments that describe what
3722 33. Some experimental protocols are using DNS PTR records for new purposes. The
3723 keys for these records are domain names, not reversed IP addresses. The
3724 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
3725 leaves it alone. Component reversal etc. now happens only for IP addresses.
3728 34. The host_aton() function is supposed to be passed a string that is known
3729 to be a valid IP address. However, in the case of IPv6 addresses, it was
3730 not checking this. This is a hostage to fortune. Exim now panics and dies
3731 if the condition is not met. A case was found where this could be provoked
3732 from a dnsdb PTR lookup with an IPv6 address that had more than 8
3733 components; fortuitously, this particular loophole had already been fixed
3734 by change 4.50/55 or 4.44/33 above.
3736 If there are any other similar loopholes, the new check in host_aton()
3737 itself should stop them being exploited. The report I received stated that
3738 data on the command line could provoke the exploit when Exim was running as
3739 exim, but did not say which command line option was involved. All I could
3740 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
3741 running as the user.
3744 35. There was a buffer overflow vulnerability in the SPA authentication code
3745 (which came originally from the Samba project). I have added a test to the
3746 spa_base64_to_bits() function which I hope fixes it.
3749 36. The daemon start-up calls getloadavg() while still root for those OS that
3750 need the first call to be done as root, but it missed one case: when
3751 deliver_queue_load_max is set with deliver_drop_privilege. This is
3752 necessary for the benefit of the queue runner, because there is no re-exec
3753 when deliver_drop_privilege is set.
3755 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
3756 was in use, was not putting the data itself into the right store pool;
3757 consequently, it could be overwritten for a subsequent message in the same
3758 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
3761 38. Sometimes the final signoff response after QUIT could fail to get
3762 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
3763 < 0 before doing a fflush(). This bug looks as though it goes back to the
3764 introduction of TLS in release 3.20, but "sometimes" must have been rare
3765 because the tests only now provoked it.
3771 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
3772 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
3773 Now it returns an integer. A number of calls were still expecting a T/F
3774 return. Fortuitously, in all cases, the tests worked in OK situations,
3775 which is the norm. However, things would have gone wrong on any write
3776 failures on the smtp file descriptor. This function is used when sending
3777 messages over SMTP and also when doing verify callouts.
3779 2. When Exim is called to do synchronous delivery of a locally submitted
3780 message (the -odf or -odi options), it no longer closes stderr before doing
3783 3. Implemented the mua_wrapper option.
3785 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
3787 5. Implemented the functions header_remove(), header_testname(),
3788 header_add_at_position(), and receive_remove_recipient(), and exported them
3791 6. If an ACL "warn" statement specified the addition of headers, Exim already
3792 inserted X-ACL-Warn: at the start if there was no header name. However, it
3793 was not making this test for the second and subsequent header lines if
3794 there were newlines in the string. This meant that an invalid header could
3795 be inserted if Exim was badly configured.
3797 7. Allow an ACL "warn" statement to add header lines at the start or after all
3798 the Received: headers, as well as at the end.
3800 8. Added the rcpt_4xx retry error code.
3802 9. Added postmaster_mailfrom=xxx to callout verification option.
3804 10. Added mailfrom=xxxx to the callout verification option, for verify=
3807 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
3808 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
3810 12. Inserted some casts to stop certain compilers warning when using pointer
3811 differences as field lengths or precisions in printf-type calls (mostly
3812 affecting debugging statements).
3814 13. Added optional readline() support for -be (dynamically loaded).
3816 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
3817 same clock tick as a message's arrival, so that its received time was the
3818 same as the "first fail" time on the retry record, and that message
3819 remained on the queue past the ultimate address timeout, every queue runner
3820 would try a delivery (because it was past the ultimate address timeout) but
3821 after another failure, the ultimate address timeout, which should have then
3822 bounced the address, did not kick in. This was a "< instead of <=" error;
3823 in most cases the first failure would have been in the next clock tick
3824 after the received time, and all would be well.
3826 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
3827 being recognized when the domain list was tested by the match_domain
3828 condition in an expansion string.
3830 16. Added the ${str2b64: operator.
3832 17. Exim was always calling setrlimit() to set a large limit for the number of
3833 processes, without checking whether the existing limit was already
3834 adequate. (It did check for the limit on file descriptors.) Furthermore,
3835 errors from getrlimit() and setrlimit() were being ignored. Now they are
3836 logged to the main and panic logs, but Exim does carry on, to try to do its
3837 job under whatever limits there are.
3839 18. Imported PCRE 5.0.
3841 19. Trivial typo in log message " temporarily refused connection" (the leading
3844 20. If the log selector return_path_on_delivery was set and an address was
3845 redirected to /dev/null, the delivery process crashed because it assumed
3846 that a return path would always be set for a "successful" delivery. In this
3847 case, the whole delivery is bypassed as an optimization, and therefore no
3850 21. Internal re-arrangement: the function for sending a challenge and reading
3851 a response while authentication was assuming a zero-terminated challenge
3852 string. It's now changed to take a pointer and a length, to allow for
3853 binary data in such strings.
3855 22. Added the cyrus_sasl authenticator (code supplied by MBM).
3857 23. Exim was not respecting finduser_retries when seeking the login of the
3858 uid under which it was called; it was always trying 10 times. (The default
3859 setting of finduser_retries is zero.) Also, it was sleeping after the final
3860 failure, which is pointless.
3862 24. Implemented tls_on_connect_ports.
3864 25. Implemented acl_smtp_predata.
3866 26. If the domain in control=submission is set empty, Exim assumes that the
3867 authenticated id is a complete email address when it generates From: or
3868 Sender: header lines.
3870 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
3871 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
3872 chown and chgrp in /bin and hostname in /usr/bin.
3874 28. Exim was keeping the "process log" file open after each use, just as it
3875 does for the main log. This opens the possibility of it remaining open for
3876 long periods when the USR1 signal hits a daemon. Occasional processlog
3877 errors were reported, that could have been caused by this. Anyway, it seems
3878 much more sensible not to leave this file open at all, so that is what now
3881 29. The long-running daemon process does not normally write to the log once it
3882 has entered its main loop, and it closes the log before doing so. This is
3883 so that log files can straightforwardly be renamed and moved. However,
3884 there are a couple of unusual error situations where the daemon does write
3885 log entries, and I had neglected to close the log afterwards.
3887 30. The text of an SMTP error response that was received during a remote
3888 delivery was being truncated at 512 bytes. This is too short for some of
3889 the long messages that one sometimes sees. I've increased the limit to
3892 31. It is now possible to make retry rules that apply only when a message has a
3893 specific sender, in particular, an empty sender.
3895 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
3896 it possible to be selective about when SMTP synchronization is enforced.
3898 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
3900 32. Implemented hosts_connection_nolog.
3902 33. Added an ACL for QUIT.
3904 34. Setting "delay_warning=" to disable warnings was not working; it gave a
3907 35. Added mailbox_size and mailbox_filecount to appendfile.
3909 36. Added control = no_multiline_responses to ACLs.
3911 37. There was a bug in the logic of the code that waits for the clock to tick
3912 in the case where the clock went backwards by a substantial amount such
3913 that the microsecond fraction of "now" was more than the microsecond
3914 fraction of "then" (but the whole seconds number was less).
3916 38. Added support for the libradius Radius client library this is found on
3917 FreeBSD (previously only the radiusclient library was supported).
3923 1. When certain lookups returned multiple values in the form name=value, the
3924 quoting of the values was not always being done properly. Specifically:
3925 (a) If the value started with a double quote, but contained no whitespace,
3927 (b) If the value contained whitespace other than a space character (i.e.
3928 tabs or newlines or carriage returns) it was not quoted.
3929 This fix has been applied to the mysql and pgsql lookups by writing a
3930 separate quoting function and calling it from the lookup code. The fix
3931 should probably also be applied to nisplus, ibase and oracle lookups, but
3932 since I cannot test any of those, I have not disturbed their existing code.
3934 2. A hit in the callout cache for a specific address caused a log line with no
3935 reason for rejecting RCPT. Now it says "Previous (cached) callout
3936 verification failure".
3938 3. There was an off-by-one bug in the queryprogram router. An over-long
3939 return line was truncated at 256 instead of 255 characters, thereby
3940 overflowing its buffer with the terminating zero. As well as fixing this, I
3941 have increased the buffer size to 1024 (and made a note to document this).
3943 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
3944 when Exim is waiting for an SMTP response from a remote server, Exim
3945 restarts its select() call on the socket, thereby resetting its timeout.
3946 This is not a problem when such interrupts are rare. Somebody set up a cron
3947 job to run exiwhat every 2 minutes, which is less than the normal select()
3948 timeout (5 or 10 minutes). This meant that the select() timeout never
3949 kicked in because it was always reset. I have fixed this by comparing the
3950 time when an interrupt arrives with the time at the start of the first call
3951 to select(). If more time than the timeout has elapsed, the interrupt is
3952 treated as a timeout.
3954 5. Some internal re-factoring in preparation for the addition of Sieve
3955 extensions (by MH). In particular, the "personal" test is moved to a
3956 separate function, and given an option for scanning Cc: and Bcc: (which is
3957 not set for Exim filters).
3959 6. When Exim created an email address using the login of the caller as the
3960 local part (e.g. when creating a From: or Sender: header line), it was not
3961 quoting the local part when it contained special characters such as @.
3963 7. Installed new OpenBSD configuration files.
3965 8. Reworded some messages for syntax errors in "and" and "or" conditions to
3966 try to make them clearer.
3968 9. Callout options, other than the timeout value, were being ignored when
3969 verifying sender addresses in header lines. For example, when using
3971 verify = header_sender/callout=no_cache
3973 the cache was (incorrectly) being used.
3975 10. Added a missing instance of ${EXE} to the exim_install script; this affects
3976 only the Cygwin environment.
3978 11. When return_path_on_delivery was set as a log selector, if different remote
3979 addresses in the same message used different return paths and parallel
3980 remote delivery occurred, the wrong values would sometimes be logged.
3981 (Whenever a remote delivery process finished, the return path value from
3982 the most recently started remote delivery process was logged.)
3984 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
3985 lines when AUTH and/or TLS are in use. This is the "received protocol"
3986 field. Exim used to use "asmtp" for authenticated SMTP, without any
3987 indication (in the protocol name) for TLS use. Now it follows the RFC and
3988 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
3989 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
3990 names appear in log lines as well as in Received: header lines.
3992 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
3993 extensions, and comparison tests, and to fix some bugs.
3995 14. Changes to the "personal" filter test:
3997 (1) The test was buggy in that it was just doing the equivalent of
3998 "contains" tests on header lines. For example, if a user's address was
3999 anne@some.where, the "personal" test would incorrectly be true for
4001 To: susanne@some.where
4003 This test is now done by extracting each address from the header in turn,
4004 and checking the entire address. Other tests that are part of "personal"
4005 are now done using regular expressions (for example, to check local parts
4006 of addresses in From: header lines).
4008 (2) The list of non-personal local parts in From: addresses has been
4009 extended to include "listserv", "majordomo", "*-request", and "owner-*",
4010 taken from the Sieve specification recommendations.
4012 (3) If the message contains any header line starting with "List-" it is
4013 treated as non-personal.
4015 (4) The test for "circular" in the Subject: header line has been removed
4016 because it now seems ill-conceived.
4018 15. Minor typos in src/EDITME comments corrected.
4020 16. Installed latest exipick from John Jetmore.
4022 17. If headers_add on a router specified a text string that was too long for
4023 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
4024 of string_sprintf() is now avoided.
4026 18. $message_body_size was not set (it was always zero) when running the DATA
4027 ACL and the local_scan() function.
4029 19. For the "mail" command in an Exim filter, no default was being set for
4030 the once_repeat time, causing a random time value to be used if "once" was
4031 specified. (If the value happened to be <= 0, no repeat happened.) The
4032 default is now 0s, meaning "never repeat". The "vacation" command was OK
4033 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
4034 (I found it when inspecting the code).
4036 20. There is now an overall timeout for performing a callout verification. It
4037 defaults to 4 times the callout timeout, which applies to individual SMTP
4038 commands during the callout. The overall timeout applies when there is more
4039 than one host that can be tried. The timeout is checked before trying the
4040 next host. This prevents very long delays if there are a large number of
4041 hosts and all are timing out (e.g. when the network connections are timing
4042 out). The value of the overall timeout can be changed by specifying an
4043 additional sub-option for "callout", called "maxwait". For example:
4045 verify = sender/callout=5s,maxwait=20s
4047 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
4048 to the end before writing, but this should make it even safer).
4050 22. Exim was forgetting that it had advertised PIPELINING for the second and
4051 subsequent messages on an SMTP connection. It was also not resetting its
4052 memory on STARTTLS and an internal HELO.
4054 23. When Exim logs an SMTP synchronization error within a session, it now
4055 records whether PIPELINING has been advertised or not.
4057 24. Added 3 instances of "(long int)" casts to time_t variables that were being
4058 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
4059 rather than long int.
4061 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
4063 26. Added the never_mail option to autoreply.
4069 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
4070 crash if the getsockname() call failed; this can happen if a connection is
4071 closed very soon after it is established. The problem was simply in the
4072 order in which certain operations were done, causing Exim to try to write
4073 to the SMTP stream before it had set up the file descriptor. The bug has
4074 been fixed by making things happen in the correct order.
4080 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
4081 before the connection was closed, thus losing the rejection response.
4083 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
4084 some early Solaris releases, but causes trouble in current releases where
4085 socklen_t is defined.
4087 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
4090 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
4093 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
4094 that releasing the top of it at the end releases what was used for sub-
4095 expansions (unless the block got too big). However, discard this block if
4096 the first thing is a variable or header, so that we can use its block when
4097 it is dynamic (useful for very large $message_headers, for example).
4099 6. Lookups now cache *every* query, not just the most recent. A new, separate
4100 store pool is used for this. It can be recovered when all lookup caches are
4101 flushed. Lookups now release memory at the end of their result strings.
4102 This has involved some general refactoring of the lookup sources.
4104 7. Some code has been added to the store_xxx() functions to reduce the amount
4105 of flapping under certain conditions.
4107 8. log_incoming_interface used to affect only the <= reception log lines. Now
4108 it causes the local interface and port to be added to several more SMTP log
4109 lines, for example "SMTP connection from", and rejection lines.
4111 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
4113 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
4115 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
4116 could be overwritten at the end of the current message (or the start of a
4117 new message if it was set in a HELO ACL). The value is now preserved for
4118 the duration of the SMTP connection.
4120 12. If a transport had a headers_rewrite setting, and a matching header line
4121 contained an unqualified address, that address was qualified, even if it
4122 did not match any rewriting rules. The underlying bug was that the values
4123 of the flags that permit the existence of unqualified sender and recipient
4124 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
4125 non-local messages, and by -bnq for local messages) were not being
4126 preserved with the message after it was received.
4128 13. When Exim was logging an SMTP synchronization error, it could sometimes log
4129 "next input=" as part of the text comprising the host identity instead of
4130 the correct text. The code was using the same buffer for two different
4131 strings. However, depending on which order the printing function evaluated
4132 its arguments, the bug did not always show up. Under Linux, for example, my
4133 test suite worked just fine.
4135 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
4136 doesn't work with some older versions of Perl. It has been changed to "my",
4137 which in any case is probably the better facility to use.
4139 15. A really picky compiler found some instances of statements for creating
4140 error messages that either had too many or two few arguments for the format
4143 16. The size of the buffer for calls to the DNS resolver has been increased
4144 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
4145 for addresses that have a lot of PTR records. This alleviates a problem; it
4146 does not fully solve it.
4148 17. A dnsdb lookup for PTR records that receives more data than will fit in the
4149 buffer now truncates the list and logs the incident, which is the same
4150 action as happens when Exim is looking up a host name and its aliases.
4151 Previously in this situation something unpredictable would happen;
4152 sometimes it was "internal error: store_reset failed".
4154 18. If a server dropped the connection unexpectedly when an Exim client was
4155 using GnuTLS and trying to read a response, the client delivery process
4156 crashed while trying to generate an error log message.
4158 19. If a "warn" verb in an ACL added multiple headers to a message in a single
4159 string, for example:
4161 warn message = H1: something\nH2: something
4163 the text was added as a single header line from Exim's point of view
4164 though it ended up OK in the delivered message. However, searching for the
4165 second and subsequent header lines using $h_h2: did not work. This has been
4166 fixed. Similarly, if a system filter added multiple headers in this way,
4167 the routers could not see them.
4169 20. Expanded the error message when iplsearch is called with an invalid key to
4170 suggest using net-iplsearch in a host list.
4172 21. When running tests using -bh, any delays imposed by "delay" modifiers in
4173 ACLs are no longer actually imposed (and a message to that effect is
4176 22. If a "gecos" field in a passwd entry contained escaped characters, in
4177 particular, if it contained a \" sequence, Exim got it wrong when building
4178 a From: or a Sender: header from that name. A second bug also caused
4179 incorrect handling when an unquoted " was present following a character
4180 that needed quoting.
4182 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
4183 was not being matched caselessly.
4185 24. Arranged for all hyphens in the exim.8 source to be escaped with
4188 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
4189 properly. Recipient callout cache records were still being keyed to include
4190 the sender, even when use_sender was set false. This led to far more
4191 callouts that were necessary. The sender is no longer included in the key
4192 when use_sender is false.
4194 26. Added "control = submission" modifier to ACLs.
4196 27. Added the ${base62d: operator to decode base 62 numbers.
4198 28. dnsdb lookups can now access SRV records.
4200 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
4201 the configuration file.
4203 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
4204 (-v) mode. This makes the output for a verbose queue run more intelligible.
4206 31. Added a use_postmaster feature to recipient callouts.
4208 32. Added the $body_zerocount variable, containing the number of binary zero
4209 bytes in the message body.
4211 33. The time of last modification of the "new" subdirectory is now used as the
4212 "mailbox time last read" when there is a quota error for a maildir
4215 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
4217 35. Added +ignore_unknown as a special item in host lists.
4219 36. Code for decoding IPv6 addresses in host lists is now included, even if
4220 IPv6 support is not being compiled. This fixes a bug in which an IPv6
4221 address was recognized as an IP address, but was then not correctly decoded
4222 into binary, causing unexpected and incorrect effects when compared with
4229 1. Very minor rewording of debugging text in manualroute to say "list of
4230 hosts" instead of "hostlist".
4232 2. If verify=header_syntax was set, and a header line with an unqualified
4233 address (no domain) and a large number of spaces between the end of the
4234 name and the colon was received, the reception process suffered a buffer
4235 overflow, and (when I tested it) crashed. This was caused by some obsolete
4236 code that should have been removed. The fix is to remove it!
4238 3. When running in the test harness, delay a bit after writing a bounce
4239 message to get a bit more predictability in the log output.
4241 4. Added a call to search_tidyup() just before forking a reception process. In
4242 theory, someone could use a lookup in the expansion of smtp_accept_max_
4243 per_host which, without the tidyup, could leave open a database connection.
4245 5. Added the variables $recipient_data and $sender_data which get set from a
4246 lookup success in an ACL "recipients" or "senders" condition, or a router
4247 "senders" option, similar to $domain_data and $local_part_data.
4249 6. Moved the writing of debug_print from before to after the "senders" test
4252 7. Change 4.31/66 (moving the time when the Received: is generated) caused
4253 problems for message scanning, either using a data ACL, or using
4254 local_scan() because the Received: header was not generated till after they
4255 were called (in order to set the time as the time of reception completion).
4256 I have revised the way this works. The header is now generated after the
4257 body is received, but before the ACL or local_scan() are called. After they
4258 are run, the timestamp in the header is updated.
4264 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
4265 before starting a queue runner without re-exec. This happened only when
4266 deliver_drop_privilege was set or when the Exim user was set to root. The
4267 effect of the bug was that timeouts during subsequent deliveries caused
4268 crashes instead of being properly handled. The handler is now left at its
4269 default (and expected) setting.
4271 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
4272 message, again when deliver_drop_privilege is set or Exim is run as root.
4273 The bug described in (1) was not present in this case, but the tidying up
4274 of the other signals was missing. I have made the two cases consistent.
4276 3. The ignore_target_hosts setting on a manualroute router was being ignored
4277 for hosts that were looked up using the /MX notation.
4279 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
4282 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
4283 operated on the sender address. After changing the $sender_address to <>
4284 for the sender address verify, Exim was re-instated it as the original
4285 (before rewriting) address, but remembering that it had rewritten it, so it
4286 wasn't rewriting it again. This bug also had the effect of breaking the
4287 sender address verification caching when the sender address was rewritten.
4289 6. The ignore_target_hosts option was being ignored by the ipliteral router.
4290 This has been changed so that if the ip literal address matches
4291 ignore_target_hosts, the router declines.
4293 7. Added expansion conditions match_domain, match_address, and match_local_
4294 part (NOT match_host).
4296 8. The placeholder for the Received: header didn't have a length field set.
4298 9. Added code to Exim itself and to exim_lock to test for a specific race
4299 condition that could lead to file corruption when using MBX delivery. The
4300 issue is with the lockfile that is created in /tmp. If this file is removed
4301 after a process has opened it but before that process has acquired a lock,
4302 there is the potential for a second process to recreate the file and also
4303 acquire a lock. This could lead to two Exim processes writing to the file
4304 at the same time. The added code performs the same test as UW imapd; it
4305 checks after acquiring the lock that its file descriptor still refers to
4306 the same named file.
4308 10. The buffer for building added header lines was of fixed size, 8192 bytes.
4309 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
4312 11. Added the smtp_active_hostname option. If used, this will typically be made
4313 to depend on the incoming interface address. Because $interface_address is
4314 not set up until the daemon has forked a reception process, error responses
4315 that can happen earlier (such as "too many connections") no longer contain
4318 12. If an expansion in a condition on a "warn" statement fails because a lookup
4319 defers, the "warn" statement is abandoned, and the next ACL statement is
4320 processed. Previously this caused the whole ACL to be aborted.
4322 13. Added the iplsearch lookup type.
4324 14. Added ident_timeout as a log selector.
4326 15. Added tls_certificate_verified as a log selector.
4328 16. Added a global option tls_require_ciphers (compare the smtp transport
4329 option of the same name). This controls incoming TLS connections.
4331 17. I finally figured out how to make tls_require_ciphers do a similar thing
4332 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
4333 before starting the TLS session.
4335 18. Tabs are now shown as \t in -bP output.
4337 19. If the log selector return_path_on_delivery was set, Exim crashed when
4338 bouncing a message because it had too many Received: header lines.
4340 20. If two routers both had headers_remove settings, and the first one included
4341 a superfluous trailing colon, the final name in the first list and the
4342 first name in the second list were incorrectly joined into one item (with a
4343 colon in the middle).
4349 1. Added -C and -D options to the exinext utility, mainly to make it easier
4350 to include in the automated testing, but these could be helpful when
4351 multiple configurations are in use.
4353 2. The exinext utility was not formatting the output nicely when there was
4354 an alternate port involved in the retry record key, nor when there was a
4355 message id as well (for retries that were specific to a specific message
4356 and a specific host). It was also confused by IPv6 addresses, because of
4357 the additional colons they contain. I have fixed the IPv4 problem, and
4358 patched it up to do a reasonable job for IPv6.
4360 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
4361 delivery, the log line now contains "pipelined" if PIPELINING was used.
4363 4. An SMTP transport process used to panic and die if the bind() call to set
4364 an explicit outgoing interface failed. This has been changed; it is now
4365 treated in the same way as a connect() failure.
4367 5. A reference to $sender_host_name in the part of a conditional expansion
4368 that was being skipped was still causing a DNS lookup. This no longer
4371 6. The def: expansion condition was not recognizing references to header lines
4372 that used bh_ and bheader_.
4374 7. Added the _cache feature to named lists.
4376 8. The code for checking quota_filecount in the appendfile transport was
4377 allowing one more file than it should have been.
4379 9. For compatibility with Sendmail, the command line option
4387 and sets the incoming protocol and host name (for trusted callers). The
4388 host name and its colon can be omitted when only the protocol is to be set.
4389 Note the Exim already has two private options, -pd and -ps, that refer to
4390 embedded Perl. It is therefore impossible to set a protocol value of "d" or
4391 "s", but I don't think that's a major issue.
4393 10. A number of refactoring changes to the code, none of which should affect
4396 (a) The number of logging options was getting close to filling up the
4397 32-bit word that was used as a bit map. I have split them into two classes:
4398 those that are passed in the argument to log_write(), and those that are
4399 only ever tested independently outside of that function. These are now in
4400 separate 32-bit words, so there is plenty of room for expansion again.
4401 There is no change in the user interface or the logging behaviour.
4403 (b) When building, for example, log lines, the code previously used a
4404 macro that called string_cat() twice, in order to add two strings. This is
4405 not really sufficiently general. Furthermore, there was one instance where
4406 it was actually wrong because one of the argument was used twice, and in
4407 one call a function was used. (As it happened, calling the function twice
4408 did not affect the overall behaviour.) The macro has been replaced by a
4409 function that can join an arbitrary number of extra strings onto a growing
4412 (c) The code for expansion conditions now uses a table and a binary chop
4413 instead of a serial search (which was left over from when there were very
4414 few conditions). Also, it now recognizes conditions like "pam" even when
4415 the relevant support is not compiled in: a suitably worded error message is
4416 given if an attempt is made to use such a condition.
4418 11. Added ${time_interval:xxxxx}.
4420 12. A bug was causing one of the ddress fields not to be passed back correctly
4421 from remote delivery subprocesses. The field in question was not being
4422 subsequently used, so this caused to problems in practice.
4424 13. Added new log selectors queue_time and deliver_time.
4426 14. Might have fixed a bug in maildirsizefile handling that threw up
4427 "unexpected character" debug warnings, and recalculated the data
4428 unnecessarily. In any case, I expanded the warning message to give more
4431 15. Added the message "Restricted characters in address" to the statements in
4432 the default ACL that block characters like @ and % in local parts.
4434 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
4435 Three changes have been made:
4437 (a) There was a serious bug; a negative response to MAIL caused the whole
4438 recipient domain to be cached as invalid, thereby blocking all messages
4439 to all local parts at the same domain, from all senders. This bug has
4440 been fixed. The domain is no longer cached after a negative response to
4441 MAIL if the sender used is not empty.
4443 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
4446 (c) A new callout option, "use_sender" has been added for people who want
4447 the modified behaviour.
4453 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
4456 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
4457 indeed breaks things for older releases.
4459 3. Added additional logging to the case where there is a problem reading data
4460 from a filter that is running in a subprocess using a pipe, in order to
4461 try to track down a specific problem.
4463 4. Testing facility fudge: when running in the test harness and attempting
4464 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
4465 getting "No route to host". Convert this to a timeout.
4467 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
4470 6. Some OS don't have socklen_t but use size_t instead. This affects the
4471 fifth argument of getsockopt() amongst other things. This is now
4472 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
4473 can be set for individual OS. I have set it for SunOS5, OSF1, and
4474 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
4475 some earlier ones do not.
4477 7. Change 4.30/15 was not doing the test caselessly.
4479 8. The standard form for an IPv6 address literal was being rejected by address
4480 parsing in, for example, MAIL and RCPT commands. An example of this kind of
4481 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
4482 this, as well as the form without the "IPv6" on the front (but only when
4483 address literals are enabled, of course).
4485 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
4487 10. Exim crashed if a message with an empty sender address specified by -f
4488 encountered a router with an errors_to setting. This could be provoked only
4489 by a command such as
4493 where an empty string was supplied; "<>" did not hit this bug.
4495 11. Installed PCRE release 4.5.
4497 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
4498 remained set. It is now erased.
4500 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
4501 times from message ids (which are base 36 rather than the normal 62).
4503 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
4504 were being counted as actual protocol errors, and logged if the log
4505 selector +smtp_protocol_error was set. One cannot be perfect in this test,
4506 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
4507 and DATA following a set of rejected RCPTs do not count as protocol errors.
4508 In other words, Exim assumes they were pipelined, though this may not
4509 actually be the case. Of course, in all cases the client gets an
4510 appropriate error code.
4512 15. If a lookup fails in an ACL condition, a message about the failure may
4513 be available; it is used if testing the ACL cannot continue, because most
4514 such messages specify what the cause of the deferral is. However, some
4515 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
4516 that caused an old message to be retained and used if a later statement
4517 caused a defer, replacing the real cause of the deferral.
4519 16. If an IP address had so many PTR records that the DNS lookup buffer
4520 was not large enough to hold them, Exim could crash while trying to process
4521 the truncated data. It now detects and logs this case.
4523 17. Further to 4.21/58, another change has been made: if (and only if) the
4524 first line of a message (the first header line) ends with CRLF, a bare LF
4525 in a subsequent header line has a space inserted after it, so as not to
4526 terminate the header.
4528 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
4529 unnecessarily, used atoi() instead of strtol(), and didn't check the
4530 termination when getting file sizes from file names by regex.
4532 19. Completely re-implemented the support for maildirsize files, in the light
4533 of a number of problems with the previous contributed implementation
4534 (4.30/29). In particular:
4536 . If the quota is zero, the maildirsize file is maintained, but no quota is
4539 . If the maildir directory does not exist, it is created before any attempt
4540 to write a maildirsize file.
4542 . The quota value in the file is just a cache; if the quota is changed in
4543 the transport, the new value overrides.
4545 . A regular expression is available for excluding directories from the
4548 20. The autoreply transport checks the characters in options that define the
4549 message's headers; it allows continued headers, but it was checking with
4550 isspace() after an embedded newline instead of explicitly looking for a
4553 21. If all the "regular" hosts to which an address was routed had passed their
4554 expiry times, and had not reached their retry times, the address was
4555 bounced, even if fallback hosts were defined. Now Exim should go on to try
4558 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
4559 equivalent code in the SMTP transport. Some hosts send humungous responses
4560 to HELO/EHLO, more than 1024 it seems.
4562 23. Refactoring: code in filter.c used (void *) for "any old type" but this
4563 gives compiler warnings in some environments. I've now done it "properly",
4566 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
4567 (because of problems with the built-in one) was declared to return uschar *
4568 instead of char *, causing compiler failure.
4570 25. Fixed a file descriptor leak when processing alias/forward files.
4572 26. Fixed a minor format string issue in dbfn.c.
4574 27. Typo in exim.c: ("dmbnz" for "dbmnz").
4576 28. If a filter file refered to $h_xxx or $message_headers, and the headers
4577 contained RFC 2047 "words", Exim's memory could, under certain conditions,
4580 29. When a sender address is verified, it is cached, to save repeating the test
4581 when there is more than one recipient in a message. However, when the
4582 verification involves a callout, it is possible for different callout
4583 options to be set for different recipients. It is too complicated to keep
4584 track of this in the cache, so now Exim always runs a verification when a
4585 callout is required, relying on the callout cache for the optimization.
4586 The overhead is duplication of the address routing, but this should not be
4589 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
4590 to be verified with callout=postmaster, and the main callout worked but the
4591 postmaster check failed, the verification correctly failed. However, if a
4592 subsequent RCPT command asked for sender verification *without* the
4593 postmaster check, incorrect caching caused this verification also to fail,
4596 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
4597 it was not caching the DNS options (qualify_single, search_parents) that
4598 were used when the lookup failed. A subsequent lookup with different
4599 options therefore always gave the same answer, though there were cases
4600 where it should not have. (Example: a "domains = !$mx_any" option on a
4601 dnslookup router: the "domains" option is always processed without any
4602 widening, but the router might have qualify_single set.) Now Exim uses the
4603 cached value only when the same options are set.
4605 32. Added John Jetmore's "exipick" utility to the distribution.
4607 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
4608 than a timeout (e.g. a certificate is required, and is not provided), an
4609 Exim server now closes the connection immediately. Previously it waited for
4610 the client to close - but if the client is SSL, it seems that they each
4611 wait for each other, leading to a delay before one of them times out.
4613 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
4614 maintained 0.8.x compatibility because I don't think many are using it, and
4615 it is clearly obsolete.
4617 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
4620 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
4621 client certificate was expired. A simple patch fixes this, though I don't
4622 understand the full logic of why the verify callback is called multiple
4625 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
4626 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
4627 which causes problems with some clients (such as the Certicom SSL Plus
4628 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
4629 disables the coutermeasure allowing Eudora to connect."
4631 38. Exim was not checking that a write() to a log file succeeded. This could
4632 lead to Bad Things if a log got too big, in particular if it hit a file
4633 size limit. Exim now panics and dies if it cannot write to a log file, just
4634 as it does if it cannot open a log file.
4636 39. Modified OS/Makefile-Linux so that it now contains
4638 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
4640 The two -D definitions ensure that Exim is compiled with large file
4641 support, which makes it possible to handle log files that are bigger than
4644 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
4645 instance) a domain was checked against a named list that involved a lookup,
4646 causing $domain_data to be set, then another domain was checked against the
4647 same list, then the first domain was re-checked, the value of $domain_data
4648 after the final check could be wrong. In particular, if the second check
4649 failed, it could be set empty. This bug probably also applied to
4652 41. The strip_trailing_dot option was not being applied to the address given
4653 with the -f command-line option.
4655 42. The code for reading a message's header from the spool was incrementing
4656 $received_count, but never initializing it. This meant that the value was
4657 incorrect (doubled) while delivering a message in the same process in which
4658 it was received. In the most common configuration of Exim, this never
4659 happens - a fresh exec is done - but it can happen when
4660 deliver_drop_privilege is set.
4662 43. When Exim logs an SMTP synchronization error - client data sent too soon -
4663 it now includes up to 150 characters of the unexpected data in the log
4666 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
4667 and building data strings. The size of both of these buffers was 10 000
4668 bytes - far larger than anybody would *ever* want, thought I. Needless to
4669 say, somebody hit the limit. I have increased the maximum line length to
4670 20 000 and the maximum data length of concatenated lines to 100 000. I have
4671 also fixed two bugs, because there was no checking on these buffers. Tsk,
4672 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
4673 buffer is too small.
4675 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
4676 lsearch lookups. Now it does.
4678 46. When parsing a route_list item in a manualroute router, a fixed-length
4679 buffer was used for the list of hosts. I made this 1024 bytes long,
4680 thinking that nobody would ever have a list of hosts that long. Wrong.
4681 Somebody had a whole pile of complicated expansion conditions, and the
4682 string was silently truncated, leading to an expansion error. It turns out
4683 that it is easier to change to an unlimited length (owing to other changes
4684 that have happened since this code was originally written) than to build
4685 structure for giving a limitation error. The length of the item that
4686 expands into the list of hosts is now unlimited.
4688 47. The lsearch lookup could not handle data where the length of text line was
4689 more than 4095 characters. Such lines were truncated, leading to shortened
4690 data being returned. It should now handle lines of any length.
4692 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
4693 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
4696 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
4697 replacing strings such as BIN_DIRECTORY in a source file: the replacement
4698 no longer happens in comment lines. A list of replacements is now placed
4699 at the head of all of the source files, except those whose only change is
4700 to replace PERL_COMMAND in the very first #! line.
4702 50. Replaced the slow insertion sort in queue.c, for sorting the list of
4703 messages on the queue, with a bottom-up merge sort, using code contributed
4704 by Michael Haardt. This should make operations like -bp somewhat faster on
4705 large queues. It won't affect queue runners, except when queue_run_in_order
4708 51. Installed eximstats 1.31 in the distribution.
4710 52. Added support for SRV lookups to the dnslookup router.
4712 53. If an ACL referred to $message_body or $message_body_end, the value was not
4713 reset for any messages that followed in the same SMTP session.
4715 54. The store-handling optimization for building very long strings was not
4716 differentiating between the different store pools. I don't think this
4717 actually made any difference in practice, but I've tidied it.
4719 55. While running the routers to verify a sender address, $sender_address
4720 was still set to the sender address. This is wrong, because when routing to
4721 send a bounce to the sender, it would be empty. Therefore, I have changed
4722 it so that, while verifying a sender address, $sender_address is set to <>.
4723 (There is no change to what happens when verifying a recipient address.)
4725 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
4726 target A or AAAA records (if not already returned) without resetting the
4727 qualify_single or search_parents options of the DNS resolver. These are
4728 inappropriate in this case because the targets of MX and SRV records must
4729 be FQDNs. A broken DNS record could cause trouble if it happened to have a
4730 target that, when qualified, matched something in the local domain. These
4731 two options are now turned off when doing these lookups.
4733 57. It seems that at least some releases of Reiserfs (which does not have the
4734 concept of a fixed number of inodes) returns zero and not -1 for the
4735 number of available inodes. This interacted badly with check_spool_inodes,
4736 which assumed that -1 was the "no such thing" setting. What I have done is
4737 to check that the total number of inodes is greater than zero before doing
4738 the test of how many are available.
4740 58. When a "warn" ACL statement has a log_message modifier, the message is
4741 remembered, and not repeated. This is to avoid a lot of repetition when a
4742 message has many recipients that cause the same warning to be written.
4743 Howewer, Exim was preserving the list of already written lines for an
4744 entire SMTP session, which doesn't seem right. The memory is now reset if a
4745 new message is started.
4747 59. The "rewrite" debugging flag was not showing the result of rewriting in the
4748 debugging output unless log_rewrite was also set.
4750 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
4751 of (int)(handle) when we know that handle contains (void *)(-1).
4753 61. The Exim daemon panic-logs an error return when it closes the incoming
4754 connection. However "connection reset by peer" seems to be common, and
4755 isn't really an error worthy of noting specially, so that particular error
4758 62. When Exim is trying to find all the local interfaces, it used to panic and
4759 die if the ioctl to get the interface flags failed. However, it seems that
4760 on at least one OS (Solaris 9) it is possible to have an interface that is
4761 included in the list of interfaces, but for which you get a failure error
4762 for this call. This happens when the interface is not "plumbed" into a
4763 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
4764 failure of the "get flags" call assumes that the interface is down.
4766 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
4767 makes life easier for people who are doing arithmetic on fields extracted
4768 from dates, where you often get leading zeros that should not be
4769 interpreted as octal.
4771 64. Added qualify_domain to the redirect router, to override the global
4774 65. If a pathologically long header line contained very many addresses (the
4775 report of this problem mentioned 10 000) and each of them was rewritten,
4776 Exim could use up a very large amount of memory. (It kept on making new
4777 copies of the header line as it rewrote, and never released the old ones.)
4778 At the expense of a bit more processing, the header rewriting function has
4779 been changed so that it no longer eats memory in this way.
4781 66. The generation of the Received: header has been moved from the time that a
4782 message starts to be received, to the time that it finishes. The timestamp
4783 in the Received: header should now be very close to that of the <= log
4784 line. There are two side-effects of this change:
4786 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
4787 logged header lines no longer include the local Received: line, because
4788 it has not yet been created. The same applies to a copy of the message
4789 that is returned to a non-SMTP sender when a message is rejected.
4791 (b) When a filter file is tested using -bf, no additional Received: header
4792 is added to the test message. After some thought, I decided that this
4795 This change does not affect the value of $received_for. It is still set
4796 after address rewriting, but before local_scan() is called.
4798 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
4800 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
4801 gave an unhelpful panic error message, and a defer error. I have managed to
4802 change this behaviour so that it now rejects any supplied certificate,
4803 which seems right, as the list of acceptable certificates is empty.
4805 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
4806 gave an unhelpful defer error. I have not managed to make this reject any
4807 supplied certificates, but the error message it gives is "no certificate
4808 supplied", which is not helpful.
4810 70. exigrep's output now also includes lines that are not associated with any
4811 message, but which match the given pattern. Implemented by a patch from
4812 Martin Sluka, which also tidied up the Perl a bit.
4814 71. Recipient callout verification, like sender verification, was using <> in
4815 the MAIL FROM command. This isn't really the right thing, since the actual
4816 sender may affect whether the remote host accepts the recipient or not. I
4817 have changed it to use the actual sender in the callout; this means that
4818 the cache record is now keyed on a recipient/sender pair, not just the
4819 recipient address. There doesn't seem to be a real danger of callout loops,
4820 since a callout by the remote host to check the sender would use <>.
4821 [SEE ABOVE: changed after hitting problems.]
4823 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
4824 temporary errors. However, in the case of such a code being given after
4825 the end of a data transmission (i.e. after ".") Exim was failing to write
4826 a retry record for the message. (Yes, there was some broken host that was
4827 actually sending 8xx at this point.)
4829 73. An unknown lookup type in a host list could cause Exim to panic-die when
4830 the list was checked. (An example that provoked this was putting <; in the
4831 middle of a list instead of at the start.) If this happened during a DATA
4832 ACL check, a -D file could be left lying around. This kind of configuration
4833 error no longer causes Exim to die; instead it causes a defer errror. The
4834 incident is still logged to the main and panic logs.
4836 74. Buglet left over from Exim 3 conversion. The message "too many messages
4837 in one connection" was written to the rejectlog but not the mainlog, except
4838 when address rewriting (yes!) was being logged.
4840 75. Added write_rejectlog option.
4842 76. When a system filter was run not as root (that is, when system_filter_user
4843 was set), the values of the $n variables were not being returned to the
4844 main process; thus, they were not subsequently available in the $sn
4847 77. Added +return_path_on_delivery log selector.
4849 78. A connection timeout was being treated differently from recipients deferred
4850 when testing hosts_max_try with a message that was older than the host's
4851 retry timeout. (The host should not be counted, thus allowing all hosts to
4852 be tried at least once before bouncing.) This may have been the cause of an
4853 occasionally reported bug whereby a message would remain on the queue
4854 longer than the retry timeout, but would be bounced if a delivery was
4855 forced. I say "may" because I never totally pinned down the problem;
4856 setting up timeout/retry tests is difficult. See also the next item.
4858 79. The ultimate address timeout was not being applied to errors that involved
4859 a combination of host plus message (for example, a timeout on a MAIL
4860 command). When an address resolved to a number of possible hosts, and they
4861 were not all tried for each delivery (e.g. because of hosts_max_try), a
4862 message could remain on the queue longer than the retry timeout.
4864 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
4867 81. Fixed an obscure SMTP outgoing bug which required at least the following
4868 conditions: (a) there was another message waiting for the same server;
4869 (b) the server returned 5xx to all RCPT commands in the first message so
4870 that the message was not completed; (c) the server dropped the connection
4871 or gave a negative response to the RSET that Exim sends to abort the
4872 transaction. The observed case was a dropped connection after DATA that had
4873 been sent in pipelining mode. That is, the server had advertised PIPELINING
4874 but was not implementing it correctly. The effect of the bug was incorrect
4875 behaviour, such as trying another host, and this could lead to a crash.
4881 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
4882 and daemon.c were passed as pointers to ints; they should have been
4883 pointers to socklen_t variables (which are typically unsigned ints).
4885 2. Some signed/unsigned type warnings in the os.c file for Linux have been
4888 3. Fixed a really odd bug that affected only the testing scheme; patching a
4889 certain fixed string in the binary changed the value of another string that
4890 happened to be identical to the end of the original first string.
4892 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
4893 name", it returns that address as the IP address. On some operating
4894 systems (e.g. Solaris), it also passes back the IP address string as the
4895 "host name". However, on others (e.g. Linux), it passes back an empty
4896 string. Exim wasn't checking for this, and was changing the host name to an
4897 empty string, assuming it had been canonicized.
4899 5. Although rare, it is permitted to have more than one PTR record for a given
4900 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
4901 all the names associated with an address, because they do in Solaris.
4902 However, it seems that they do not in Linux for data that comes from the
4903 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
4904 I found this out when I moved to a new Linux workstation and tried to run
4905 the Exim test suite.
4907 To get round this problem I have changed the code so that it now does its
4908 own call to the DNS to look up PTR records when searching for a host name.
4909 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
4910 addresses that are only in /etc/hosts are still found.
4912 This behaviour is, however, controlled by an option called host_lookup_
4913 order, which defaults to "bydns:byaddr". If people want to use the other
4914 order, or indeed, just use one or the other means of lookup, they can
4915 specify it in this variable.
4917 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
4918 some operating systems, this comes back from gethostbyaddr() as an empty
4919 string, and this is what Exim used to test for. However, it seems that in
4920 other systems, "." is yielded. Exim now tests for this case too.
4922 7. The values of check_spool_space and check_log_space are now held internally
4923 as a number of kilobytes instead of an absolute number of bytes. If a
4924 numbers is specified without 'K' or 'M', it is rounded up to the nearest
4925 kilobyte. This means that much larger values can be stored.
4927 8. Exim monitor: an attempt to get the action menu when not actually pointing
4928 at a message produces an empty menu entitled "No message selected". This
4929 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
4930 no entries in it ("Shell widget menu has zero width and/or height"). So I
4931 have added a single, blank menu entry in this case.
4933 9. Added ${quote_local_part.
4935 10. MIME decoding is now applied to the contents of Subject: header lines when
4938 11. Now that a reference to $sender_host_address automatically causes a reverse
4939 lookup to occur if necessary (4.13/18), there is no need to arrange for a
4940 host lookup before query-style lookups in lists that might use this
4941 variable. This has therefore been abolished, and the "net-" prefix is no
4942 longer necessary for query-style lookups.
4944 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
4945 have been a typo for LFLAGS, so it has been changed.
4947 13. The install script calls Exim with "-C /dev/null" in order to find the
4948 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
4949 to be output. Howeve, since Exim outputs its version number before the
4950 error, it didn't break the script. It just looked ugly. I fixed this by
4951 always allowing "-C /dev/null" if the caller is root.
4953 14. Ignore overlarge ACL variable number when reading spool file - insurance
4954 against a later release with more variables having written the file.
4956 15. The standard form for an IPv6 address literal was being rejected by EHLO.
4957 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
4958 this, as well as the form without the "IPv6" on the front.
4960 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
4961 OS/Makefile-Darwin file.
4963 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
4964 only to LDAP libraries that do not have LDAP_OPT_DEREF.
4966 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
4967 variable. However, some OS use ints for inodes. I've added cast to long int
4968 to get rid of the compiler warning.
4970 19. I had forgotten to lock out "/../" in configuration file names when
4971 ALT_CONFIG_PREFIX was set.
4973 20. Routers used for verification do not need to specify transports. However,
4974 if such a router generated a host list, and callout was configured, Exim
4975 crashed, because it could not find a port number from the (non-existent)
4976 transport. It now assumes port 25 in this circumstance.
4978 21. Added the -t option to exigrep.
4980 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
4981 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
4982 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
4983 them is set, LOOKUP_LSEARCH is forced.
4985 23. "exim -bV" now outputs a list of lookups that are included in the binary.
4987 24. Added sender and host information to the "rejected by local_scan()" log
4988 line; previously there was no indication of these.
4990 25. Added .include_if_exists.
4992 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
4993 Linux. It turns out that not all file systems support this. Apparently some
4994 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
4995 do it.) To cope with this, the error EINVAL, which means that sync-ing is
4996 not supported on the file descriptor, is now ignored when Exim is trying to
4997 sync a directory. This applies only to Linux.
4999 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
5001 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
5002 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
5005 29. Added support for maildirsize files from supplied patch (modified a bit).
5007 30. The use of :fail: followed by an empty string could lead Exim to respond to
5008 sender verification failures with (e.g.):
5010 550 Verification failed for <xxx>
5011 550 Sender verify failed
5013 where the first response line was missing the '-' that indicates it is not
5014 the final line of the response.
5016 31. The loop for finding the name of the user that called Exim had a hardwired
5017 limit of 10; it now uses the value of finduser_retries, which is used for
5018 all other user lookups.
5020 32. Added $received_count variable, available in data and not_smtp ACLs, and at
5023 33. Exim was neglecting to zero errno before one call of strtol() when
5024 expanding a string and expecting an integer value. On some systems this
5025 resulted in spurious "integer overflow" errors. Also, it was casting the
5026 result into an int without checking.
5028 34. Testing for a connection timeout using "timeout_connect" in the retry rules
5029 did not work. The code looks as if it has *never* worked, though it appears
5030 to have been documented since at least releast 1.62. I have made it work.
5032 35. The "timeout_DNS" error in retry rules, also documented since at least
5033 1.62, also never worked. As it isn't clear exactly what this means, and
5034 clearly it isn't a major issue, I have abolished the feature by treating it
5035 as "timeout", and writing a warning to the main and panic logs.
5037 36. The display of retry rules for -brt wasn't always showing the error code
5040 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
5041 timeout_connect_A, timeout_connect_MX.
5043 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
5044 to the empty sender.
5046 39. The daemon was not analysing the content of -oX till after it had closed
5047 stderr and disconnected from the controlling terminal. This meant that any
5048 syntax errors were only noted on the panic log, and the return code from
5049 the command was 0. By re-arranging the code a little, I've made the
5050 decoding happen first, so such errors now appear on stderr, and the return
5051 code is 1. However, the actual setting up of the sockets still happens in
5052 the disconnected process, so errors there are still only recorded on the
5055 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
5056 connections (as happens on some IP stacks) was logged at start up time as
5057 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
5058 it from "IPv6 and IPv4", which means that two separate sockets are being
5061 41. The debug output for gethostbyname2() or getipnodebyname() failures now
5062 says whether AF_INET or AF_INET6 was passed as an argument.
5064 42. Exiwhat output was messed up when time zones were included in log
5067 43. Exiwhat now gives more information about the daemon's listening ports,
5068 and whether -tls-on-connect was used.
5070 44. The "port" option of the smtp transport is now expanded.
5072 45. A "message" modifier in a "warn" statement in a non-message ACL was being
5073 silently ignored. Now an error message is written to the main and panic
5076 46. There's a new ACL modifier called "logwrite" which writes to a log file
5077 as soon as it is encountered.
5079 47. Added $local_user_uid and $local_user_gid at routing time.
5081 48. Exim crashed when trying to verify a sender address that was being
5084 49. Exim was recognizing only a space character after ".include". It now also
5085 recognizes a tab character.
5087 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
5088 extracting the relevant information from the specification. The man page no
5089 longer contains scrambled data for the -d option, and I've added a section
5090 at the front about calling Exim under different names.
5092 51. Added "extra_headers" argument to the "mail" command in filter files.
5094 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
5097 53. Installed eximstats 1.29.
5099 54. Added transport_filter_timeout as a generic transport option.
5101 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
5102 Cc: header lines. This was required by RFC 822, but it not required by RFC
5105 56. Exim used to add From:, Date:, and Message-Id: header lines to any
5106 incoming messages that did not have them. Now it does so only if the
5107 message originates locally, that is, if there is no associated remote host
5108 address. When Resent- header lines are present, this applies to the Resent-
5109 lines rather than the non-Resent- lines.
5111 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
5112 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
5114 58. Messages for configuration errors now include the name of the main
5115 configuration file - useful now that there may be more than one file in a
5116 list (.included file names were always shown).
5118 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
5119 for those rare installations that do not start the daemon as root or run it
5120 setuid root. I've cut out the call to initgroups() if the daemon is not
5123 60. The Exim user and group can now be bound into the binary as text strings
5124 that are looked up at the start of Exim's processing.
5126 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
5128 62. Added $mailstore_basename variable.
5130 63. Installed patch to sieve.c from Michael Haardt.
5132 64. When Exim failed to open the panic log after failing to open the main log,
5133 the original message it was trying to log was written to stderr and debug
5134 output, but if they were not available (the usual case in production), it
5135 was lost. Now it is written to syslog before the two lines that record the
5136 failures to open the logs.
5138 65. Users' Exim filters run in subprocesses under the user's uid. It is
5139 possible for a "deliver" command or an alias in a "personal" command to
5140 provoke an address rewrite. If logging of address rewriting is configured,
5141 this fails because the process is not running as root or exim. There may be
5142 a better way of dealing with this, but for the moment (because 4.30 needs
5143 to be released), I have disabled address rewrite logging when running a
5144 filter in a non-root, non-exim process.
5150 1. The buildconfig auxiliary program wasn't quoting the value set for
5151 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
5152 not defined. This bug was masked in 4.22 by the effect that was fixed in
5155 2. Some messages that were rejected after a message id was allocated were
5156 shown as "incomplete" by exigrep. It no longer does this for messages that
5157 are rejected by local_scan() or the DATA or non-SMTP ACLs.
5159 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
5160 have allow_domain_literals set, the ID did not get logged in the <= line.
5161 Domain literals are now always recognized in Message-ID: header lines.
5163 4. The first argument for a ${extract expansion item is the key name or field
5164 number. Leading and trailing spaces in this item were not being ignored,
5165 causing some misleading effects.
5167 5. When deliver_drop_privilege was set, single queue runner processes started
5168 manually (i.e. by the command "exim -q") or by the daemon (which uses the
5169 same command in the process it spins off) were not dropping privilege.
5171 6. When the daemon running as "exim" started a queue runner, it always
5172 re-executed Exim in the spun-off process. This is a waste of effort when
5173 deliver_drop_privilege is set. The new process now just calls the
5174 queue-runner function directly.
5180 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
5183 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
5184 ignored. Though the use of -oP was forcing the writing of a pid file, it
5185 was always written to the default place.
5187 3. If the message "no IP address found for host xxxx" is generated during
5188 incoming verification, it is now followed by identification of the incoming
5189 connection (so you can more easily find what provoked it).
5191 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
5193 5. Added some features to "harden" Exim a bit more against certain attacks:
5195 (a) There is now a build-time option called FIXED_NEVER_USERS that can
5196 be put in Local/Makefile. This is like the never_users runtime option,
5197 but it cannot be overridden. The default setting is "root".
5199 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
5200 prefix string with which any file named in a -C command line option
5203 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
5204 is retained for -C and -D only if the caller of Exim is root. Without
5205 it, the exim user may also use -C and -D and retain privilege.
5207 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
5208 command line option is disabled.
5210 6. Macro names set by the -D option must start with an upper case letter, just
5211 like macro names defined in the configuration file.
5213 7. Added "dereference=" facility to LDAP.
5215 8. Two instances of the typo "uknown" in the source files are fixed.
5217 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
5218 the Configure-Makefile script screwed up while processing it.
5220 10. Incorporated PCRE 4.4.
5222 11. The SMTP synchronization check was not operating right at the start of an
5223 SMTP session. For example, it could not catch a HELO sent before the client
5224 waited for the greeting. There is now a check for outstanding input at the
5225 point when the greeting is written. Because of the duplex, asynchronous
5226 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
5227 way, but not yet received, when the check is performed.
5229 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
5230 on TCP/IP sockets, because this apparently causes some broken clients to
5233 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
5234 unchanged) from the Cygwin maintainer.
5236 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
5237 was supported instead of testing for mbx. Effectively a typo.
5239 15. The spa authenticator server code was not checking that the input it
5240 received was valid base64.
5242 16. The debug output line for the "set" modifier in ACLs was not showing the
5243 name of the variable that was being set.
5245 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
5247 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
5248 lookup on its own. Something else was needed to trigger the lookup. For
5249 example, a match in host_lookup or the need for a host name in a host list.
5250 Now, if $sender_host_name is referenced and the host name has not yet been
5251 looked up, a lookup is performed. If the lookup fails, the variable remains
5252 empty, and $host_lookup_failed is set to "1".
5254 19. Added "eqi" as a case-independent comparison operator.
5256 20. The saslauthd authentication condition could segfault if neither service
5257 nor realm was specified.
5259 21. If an overflowing value such as "2048M" was set for message_size_limit, the
5260 error message that was logged was misleading, and incoming SMTP
5261 connections were dropped. The message is now more accurate, and temporary
5262 errors are given to SMTP connections.
5264 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
5265 (except RSET) with a 421 error, until QUIT is received. However, it was
5266 failing to send a response to QUIT.
5268 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
5269 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
5270 ACL is now run after the helo_try_verify_hosts code.
5272 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
5273 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
5274 case-independent, so other case variants are also recognized). Apparently
5275 some systems use these upper case variants.
5277 25. If more than two messages were waiting for the same host, and a transport
5278 filter was specified for the transport, Exim sent two messages over the
5279 same TCP/IP connection, and then failed with "socket operation on non-
5280 socket" when it tried to send the third.
5282 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
5284 27. The extern definition of crypt16() in expand.c was not being excluded when
5285 the OS had its own crypt16() function.
5287 28. Added bounce_return_body as a new option, and bounce_return_size_limit
5288 as a preferred synonym for return_size_limit, both as an option and as an
5291 29. Added LIBS=-liconv to OS/Makefile-OSF1.
5293 30. Changed the default configuration ACL to relax the local part checking rule
5294 for addresses that are not in any local domains. For these addresses,
5295 slashes and pipe symbols are allowed within local parts, but the sequence
5296 /../ is explicitly forbidden.
5298 31. SPA server authentication was not clearing the challenge buffer before
5301 32. log_message in a "warn" ACL statement was writing to the reject log as
5302 well as to the main log, which contradicts the documentation and doesn't
5303 seem right (because no rejection is happening). So I have stopped it.
5305 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
5306 However, I am unable to do any testing of this.
5308 34. Fixed an infelicity in the appendfile transport. When checking directories
5309 for a mailbox, to see if any needed to be created, it was accidentally
5310 using path names with one or more superfluous leading slashes; tracing
5311 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
5313 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
5314 contents are added to the log line that is written for every discarded
5315 recipient. (Previously a log_message setting was ignored.)
5317 36. The ${quote: operator now quotes the string if it is empty.
5319 37. The install script runs exim in order to find its version number. If for
5320 some reason other than non-existence or emptiness, which it checks, it
5321 could not run './exim', it was installing it with an empty version number,
5322 i.e. as "exim-". This error state is now caught, and the installation is
5325 38. An argument was missing from the function that creates an error message
5326 when Exim fails to connect to the socket for saslauthd authentication.
5327 This could cause Exim to crash, or give a corrupted message.
5329 39. Added isip, isip4, and isip6 to ${if conditions.
5331 40. The ACL variables $acl_xx are now saved with the message, and can be
5332 accessed later in routers, transports, and filters.
5334 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
5335 strings in the file are not string-expanded.
5337 42. If a MAIL command specified a SIZE value that was too large to fit into an
5338 int variable, the check against message_size_limit failed. Such values are
5339 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
5340 day this will have to be increased, but I don't think I want to be around
5341 when emails are that large.
5348 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
5349 iconv() is not standard in FreeBSD.
5351 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
5352 IPv6 enabled. The observed symptom was a segmentation fault on return from
5353 the function os_common_find_running_interfaces() in src/os.c.
5355 3. In the check_special_case() function in daemon.c I had used "errno" as an
5356 argument name, which causes warnings on some systems. This was basically a
5357 typo, since it was named "eno" in the comments!
5359 4. The code that waits for the clock to tick (at a resolution of some fraction
5360 of a second) so as to ensure message-id uniqueness was always waiting for
5361 at least one whole tick, when it could have waited for less. [This is
5362 almost certainly not relevant at current processor speeds, where it is
5363 unlikely to ever wait at all. But we try to future-proof.]
5365 5. The function that sleeps for a time interval that includes fractions of a
5366 second contained a race. It did not block SIGALRM between setting the
5367 timer, and suspending (a couple of lines later). If the interval was short
5368 and the sigsuspend() was delayed until after it had expired, the suspension
5369 never ended. On busy systems this could lead to processes getting stuck for
5372 6. Some uncommon configurations may cause a lookup to happen in a queue runner
5373 process, before it forks any delivery processes. The open lookup caching
5374 mechanism meant that the open file or database connection was passed into
5375 the delivery process. The problem was that delivery processes always tidy
5376 up cached lookup data. This could cause a problem for the next delivery
5377 process started by the queue runner, because the external queue runner
5378 process does not know about the closure. So the next delivery process
5379 still has data in the lookup cache. In the case of a file lookup, there was
5380 no problem because closing a file descriptor in a subprocess doesn't affect
5381 the parent. However, if the lookup was caching a connection to a database,
5382 the connection was closed, and the second delivery process was likely to
5383 see errors such as "PGSQL: query failed: server closed the connection
5384 unexpectedly". The problem has been fixed by closing all cached lookups
5385 in a queue runner before running a delivery process.
5387 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
5388 seem to have the "const" qualifier which it has on other OS. I've
5391 8. Change 4.21/2 was too strict. It is only if there are two authenticators
5392 *of the same type* (client or server) with the same public name that an
5393 error should be diagnosed.
5395 9. When Exim looked up a host name for an IP address, but failed to find the
5396 original IP address when looking up the host name (a safety check), it
5397 output the message "<ip address> does not match any IP for NULL", which was
5398 confusing, to say the least. The bug was that the host name should have
5399 appeared instead of "NULL".
5401 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
5402 user that is built into the binary, and the -C or -D options is used, root
5403 privilege is dropped before the configuration file is read. In addition,
5404 logging is switched to stderr instead of the normal log files. If the
5405 configuration then re-defines the Exim user, the unprivileged environment
5406 is probably not what is expected, so Exim logs a panic warning message (but
5409 However, if deliver_drop_privilege is set, the unprivileged state may well
5410 be exactly what is intended, so the warning has been cut out in that case,
5411 and Exim is allowed to try to write to its normal log files.
5417 1. smtp_return_error_details was not giving details for temporary sender
5418 or receiver verification errors.
5420 2. Diagnose a configuration error if two authenticators have the same public
5423 3. Exim used not to create the message log file for a message until the first
5424 delivery attempt. This could be confusing when incoming messages were held
5425 for policy or load reasons. The message log file is now created at the time
5426 the message is received, and an initial "Received" line is written to it.
5428 4. The automatically generated man page for command line options had a minor
5429 bug that caused no ill effects; however, a more serious problem was that
5430 the procedure for building the man page automatically didn't always
5431 operate. Consequently, release 4.20 contains an out-of-date version. This
5432 shouldn't happen again.
5434 5. When building Exim with embedded Perl support, the script that builds the
5435 Makefile was calling 'perl' to find its compile-time parameters, ignoring
5436 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
5438 6. The freeze_tell option was not being used for messages that were frozen on
5439 arrival, either by an ACL or by local_scan().
5441 7. Added the smtp_incomplete_transaction log selector.
5443 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
5444 was accepting AUTH without a new EHLO.
5446 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
5447 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
5450 10. Logging of TCP/IP connections (when configured) now happens in the main
5451 daemon process instead of the child process, so that the TCP/IP connection
5452 count is more accurate (but it can never be perfect).
5454 11. The use of "drop" in a nested ACL was not being handled correctly in the
5455 outer ACL. Now, if condition failure induced by the nested "drop" causes
5456 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
5457 or "require"), the connection is dropped.
5459 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
5460 that yield "discard" can now be used with an "accept" or a "discard" verb,
5461 but an error is generated for any others (because I can't see a useful way
5462 to define what should happen).
5464 13. When an ACL is read dynamically from a file (or anywhere else), the lines
5465 are now processed in the same way as lines in the Exim configuration file.
5466 In particular, continuation lines are supported.
5468 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
5470 15. Added -ti meaning -t -i.
5472 16. Check for letters, digits, hyphens, and dots in the names of dnslist
5473 domains, and warn by logging if others are found.
5475 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
5476 returned from GIFCONF when Exim is trying to find the list of interfaces on
5477 a host. The code in os.c has been modified to copy each ifreq to an aligned
5478 structure in all cases.
5480 Also, in some cases, the returned ifreq's were being copied to a 'struct
5481 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
5482 means the last couple of bytes of an IPv6 address could be chopped if the
5483 ifreq contained only a normal sockaddr (14 bytes storage).
5485 18. Named domain lists were not supported in the hosts_treat_as_local option.
5486 An entry such as +xxxx was not recognized, and was treated as a literal
5489 19. Ensure that header lines added by a DATA ACL are included in the reject log
5490 if the ACL subsequently rejects the message.
5492 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
5493 MD5 (which is deprecated).
5495 21. When testing a filter file using -bf, Exim was writing a message when it
5496 took the sender from a "From " line in the message, but it was not doing so
5497 when it took $return_path from a Return-Path: header line. It now does.
5499 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
5500 with a valid header line field name (a series of printing characters
5501 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
5503 23. Changed "disc" in the source to "disk" to conform to the documentation and
5504 the book and for uniformity.
5506 24. Ignore Sendmail's -Ooption=value command line item.
5508 25. When execve() failed while trying to run a command in a pipe transport,
5509 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
5510 could be confused with a return value of 69 from the command itself. This
5511 has been changed to 127, the value the shell returns if it is asked to run
5512 a non-existent command. The wording for the related log line suggests a
5513 non-existent command as the problem.
5515 26. If received_header_text expands to an empty string, do not add a Received:
5516 header line to the message. (Well, it adds a token one on the spool, but
5517 marks it "old" so that it doesn't get used or transmitted.)
5519 27. Installed eximstats 1.28 (addition of -nt option).
5521 28. There was no check for failure on the call to getsockname() in the daemon
5522 code. This can fail if there is a shortage of resources on the system, with
5523 ENOMEM, for example. A temporary error is now given on failure.
5525 29. Contrary to the C standard, it seems that in some environments, the
5526 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
5527 program. Exim now does this explicitly; it affects the formatting of
5528 timestamps using strftime().
5530 30. If exiqsumm was given junk data, it threw up some uninitialized variable
5531 complaints. I've now initialized all the variables, to avoid this.
5533 32. Header lines added by a system filter were not being "seen" during
5534 transport-time rewrites.
5536 33. The info_callback() function passed to OpenSSL is set up with type void
5537 (*)(SSL *, int, int), as described somewhere. However, when calling the
5538 function (actually a macro) that sets it up, the type void(*)() is
5539 expected. I've put in a cast to prevent warnings from picky compilers.
5541 34. If a DNS black list lookup found a CNAME record, but there were no A
5542 records associated with the domain it pointed at, Exim crashed.
5544 35. If a DNS black list lookup returned more than one A record, Exim ignored
5545 all but the first. It now scans all returned addresses if a particular IP
5546 value is being sought. In this situation, the contents of the
5547 $dnslist_value variable are a list of all the addresses, separated by a
5550 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
5551 to accept a host name and all its aliases if the forward lookup for any of
5552 them yielded the IP address of the incoming connection. Now it accepts only
5553 those names whose forward lookup yields the correct IP address. Any other
5554 names are discarded. This closes a loophole whereby a rogue DNS
5555 administrator could create reverse DNS records to break through a
5556 wildcarded host restriction in an ACL.
5558 37. If a user filter or a system filter that ran in a subprocess used any of
5559 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
5560 the wrong values were passed to the pipe command ($thisaddress had the
5561 value of $0, $0 had the value of $1, etc). This bug was introduced by
5562 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
5564 38. Improved the line breaking for long SMTP error messages from ACLs.
5565 Previously, if there was no break point between 40 and 75 characters, Exim
5566 left the rest of the message alone. Two changes have been made: (a) I've
5567 reduced the minimum length to 35 characters; (b) if it can't find a break
5568 point between 35 and 75 characters, it looks ahead and uses the first one
5569 that it finds. This may give the occasional overlong line, but at least the
5570 remaining text gets split now.
5572 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
5573 file descriptors might be low, and that setting 1000 would always raise it.
5574 It turns out that in some environments, the limit is already over 1000 and
5575 that lowering it causes trouble. So now Exim takes care not to decrease it.
5577 40. When delivering a message, the value of $return_path is set to $sender_
5578 address at the start of routing (routers may change the value). By an
5579 oversight, this default was not being set up when an address was tested by
5580 -bt or -bv, which affected the outcome if any router or filter referred to
5583 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
5584 log only when "message" or "log_message" are set. However, if one of the
5585 conditions was an address verification, or a call to a nested ACL, the
5586 messages generated by the underlying test were being passed through. This
5587 no longer happens. The underlying message is available in $acl_verify_
5588 message for both "message" and "log_message" expansions, so it can be
5589 passed through if needed.
5591 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
5592 new expansion $bh_ to give the encoded byte string without charset
5593 translation. Translation happens only if iconv() is available; HAVE_ICONV
5594 indicates this at build time. HEADERS_CHARSET gives the charset to
5595 translate to; headers_charset can change it in the configuration, and
5596 "headers charset" can change it in an individual filter file.
5598 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
5599 that creates RFC 2047 encoded "words" labels them as that charset instead
5600 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
5601 expansion operator; (ii) when Exim creates a From: line for a local
5602 message; (iii) when a header line is rewritten to include a "phrase" part.
5604 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
5605 buggy, causing it to skip the first lines of messages whose message ID
5606 ended in 'D'. This would not have bitten before Exim release 4.14, because
5607 message IDs were unlikely to end in 'D' before then. The effect was to have
5608 incorrect size information for certain domains.
5610 45. #include "config.h" was missing at the start of the crypt16.c module. This
5611 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
5614 46. If there was a timeout during a "random" callout check, Exim treated it as
5615 a failure of the random address, and carried on sending RSET and the real
5616 address. If the delay was just some slowness somewhere, the response to the
5617 original RCPT would be taken as a response to RSET and so on, causing
5618 mayhem of various kinds.
5620 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
5621 when I implemented it. It didn't allow for the fact that some option values
5622 may legitimatetly be negative (e.g. size_addition), and it didn't even do
5623 the right test for positive values.
5625 48. Domain names in DNS records are case-independent. Exim always looks them up
5626 in lower case. Some resolvers return domain names in exactly the case they
5627 appear in the zone file, that is, they may contain uppercase letters. Not
5628 all resolvers do this - some return always lower case. Exim was treating a
5629 change of case by a resolver as a change of domain, similar to a widening
5630 of a domain abbreviation. This triggered its re-routing code and so it was
5631 trying to route what was effectively the same domain again. This normally
5632 caused routing to fail (because the router wouldn't handle the domain
5633 twice). Now Exim checks for this case specially, and just changes the
5634 casing of the domain that it ultimately uses when it transmits the message
5637 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
5640 50. If a filter generated a file delivery with a non-absolute name (possible if
5641 no home directory exists for the router), the forbid_file option was not
5644 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
5645 the existing equality matching.
5647 52. Exim was using ints instead of ino_t variables in some places where it was
5648 dealing with inode numbers.
5650 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
5651 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
5652 called TMPDIR, and if it finds it is different, it changes its value.
5654 54. The smtp_printf() function is now made available to local_scan() so
5655 additional output lines can be written before returning. There is also an
5656 smtp_fflush() function to enable the detection of a dropped connection.
5657 The variables smtp_input and smtp_batched_input are exported to
5660 55. Changed the default runtime configuration: the message "Unknown user"
5661 has been removed from the ACL, and instead placed on the localuser router,
5662 using the cannot_route_message feature. This means that any verification
5663 failures that generate their own messages won't get overridden. Similarly,
5664 the "Unrouteable address" message that was in the ACL for unverifiable
5665 relay addresses has also been removed.
5667 56. Added hosts_avoid_esmtp to the smtp transport.
5669 57. The exicyclog script was not checking for the esoteric option
5670 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
5671 will work only if exicyclog is run under the appropriate euid.
5673 58. Following a discussion on the list, the rules by which Exim recognises line
5674 endings on incoming messages have been changed. The -dropcr and drop_cr
5675 options are now no-ops, retained only for backwards compatibility. The
5676 following line terminators are recognized: LF CRLF CR. However, special
5677 processing applies to CR:
5679 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
5680 nor a local message in the state where . is a terminator.
5682 (ii) If a bare CR is encountered in a header line, an extra space is added
5683 after the line terminator so as not to end the header. The reasoning
5684 behind this is that bare CRs in header lines are most likely either
5685 to be mistakes, or people trying to play silly games.
5687 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
5688 was being incorrectly given as 18 bytes larger than it should have been.
5689 This is a VOB (very old bug).
5691 60. This may never have affected anything current, but just in case it has:
5692 When the local host is found other than at the start of a list of hosts,
5693 the local host, those with the same MX, and any that follow, are discarded.
5694 When the list in question was part of a longer list of hosts, the following
5695 hosts (not currently being processed) were also being discarded. This no
5696 longer happens. I'm not sure if this situation could ever has previously
5699 61. Added the "/MX" feature to lists of hosts in the manualroute and query
5702 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
5703 header. This is something that is recommended in a new Internet Draft, and
5704 is something that is documented as being done by Sendmail. There are two
5705 possible values. For messages generated by the autoreply transport, Exim
5708 Auto-Submitted: auto-replied
5710 whereas for all other generated messages (e.g. bounces) it adds
5712 Auto-Submitted: auto-generated
5714 63. The "personal" condition in filters now includes a test for the
5715 Auto-Submitted: header. If it contains the string "auto-" the message it
5716 not considered personal.
5718 64. Added rcpt_include_affixes as a generic transport option.
5720 65. Added queue_only_override (default true).
5722 66. Added the syslog_duplication option.
5724 67. If what should have been the first header line of a message consisted of
5725 a space followed by a colon, Exim was mis-interpreting it as a header line.
5726 It isn't of course - it is syntactically invalid and should therefore be
5727 treated as the start of the message body. The misbehaviour could have
5728 caused a number of strange effects, including loss of data in subsequent
5729 header lines, and spool format errors.
5731 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
5732 client host had authenticated. This control can now be exercised by an ACL
5733 for more flexibility.
5735 69. By default, callouts do not happen when testing with -bh. There is now a
5736 variant, -bhc, which does actually run the callout code, including
5737 consulting and updating the callout cache.
5739 70. Added support for saslauthd authentication, courtesy of Alexander
5742 71. If statvfs() failed on the spool or log directories while checking their
5743 size for availability, Exim confusingly gave the error "space shortage".
5744 Furthermore, in debugging mode it crashed with a floating point exception.
5745 These checks are done if check_{spool,log}_{space,inodes} are set, and when
5746 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
5747 serious problem, Exim now writes to the main and panic logs when this
5748 happens, with details of the failure. It then refuses to accept the
5749 incoming message, giving the message "spool directory problem" or "log
5750 directory problem" with a 421 code for SMTP messages.
5752 72. When Exim is about to re-exec itself, it ensures that the file descriptors
5753 0, 1, and 2 exist, because some OS complain for execs without them (see
5754 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
5755 descriptors. However, the code omitted to check that the open succeeded,
5756 causing mysterious errors if for some reason the permissions on /dev/null
5757 got screwed. Now Exim writes a message to the main and panic logs, and
5758 bombs out if it can't open /dev/null.
5760 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
5761 interact so that it is all more flexible. It is supposed to remain
5762 backwards compatible. Also added extra_local_interfaces.
5764 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
5765 to bomb out with an assertion failure - to the client this appears as a
5766 connection drop. This problem occurs in the part of the code that was taken
5767 from the Samba project. Fortunately, the assertion is in a very simple
5768 function, so I have fixed this by reproducing the function inline in the
5769 one place where it is called, and arranging for authentication to fail
5770 instead of killing the process with assert().
5772 75. The SPA client code was not working when the server requested OEM rather
5773 than Unicode encoding.
5775 76. Added code to make require_files with a specific uid setting more usable in
5776 the case where statting the file as root fails - usually a non-root-mounted
5777 NFS file system. When this happens and the failure is EACCES, Exim now
5778 forks a subprocess and does the per-uid checking as the relevant uid.
5780 77. Added process_log_path.
5782 78. If log_file_path was not explicitly set, a setting of check_log_space or
5783 check_log_inodes was ignored.
5785 79. If a space check for the spool or log partitions fails, the incident is now
5786 logged. Of course, in the latter case the data may get lost...
5788 80. Added the %p formatting code to string_format() so that it can be used to
5789 print addresses in debug_print(). Adjusted all the address printing in the
5790 debugging in store.c to use %p rather than %d.
5792 81. There was a concern that a line of code in smtp_in.c could overflow a
5793 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
5794 initially expressed, the concern was not well-founded, because trailing
5795 spaces are removed early. However, if the trailing spaces were followed by
5796 a NULL, they did not get removed, so the overflow was possible. Two fixes
5799 (a) I re-wrote the offending code in a cleaner fashion.
5800 (b) If an incoming SMTP command contains a NULL character, it is rejected
5803 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
5804 runs initgroups(), so that if the Exim user is in any additional groups,
5805 they will be used during message reception.
5811 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob