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.
157 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
163 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
165 This, or similar/improved, will also be change PP/11 of 4.82.
171 PP/01 Handle short writes when writing local log-files.
172 In practice, only affects FreeBSD (8 onwards).
173 Bugzilla 1053, with thanks to Dmitry Isaikin.
175 NM/01 Bugzilla 949 - Documentation tweak
177 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
180 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
182 PP/02 Implemented gsasl authenticator.
184 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
186 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
187 `pkg-config foo` for cflags/libs.
189 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
190 with rest of GSASL and with heimdal_gssapi.
192 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
193 `pkg-config foo` for cflags/libs for the TLS implementation.
195 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
196 properties get this fed in as external SSF. A number of robustness
197 and debugging improvements to the cyrus_sasl authenticator.
199 PP/08 cyrus_sasl server now expands the server_realm option.
201 PP/09 Bugzilla 1214 - Log authentication information in reject log.
202 Patch by Jeremy Harris.
204 PP/10 Added dbmjz lookup type.
206 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
208 PP/12 MAIL args handles TAB as well as SP, for better interop with
209 non-compliant senders.
210 Analysis and variant patch by Todd Lyons.
212 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
213 Bug report from Lars Müller <lars@samba.org> (via SUSE),
214 Patch from Dirk Mueller <dmueller@suse.com>
216 PP/13 tls_peerdn now print-escaped for spool files.
217 Observed some $tls_peerdn in wild which contained \n, which resulted
218 in spool file corruption.
220 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
221 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
222 or write after TLS renegotiation, which otherwise led to messages
225 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
226 as a tracking header (ie: a signed header comes before the signature).
227 Patch from Wolfgang Breyha.
229 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
230 comma-sep list; embedded commas doubled.
232 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
234 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
236 Report and patch from Dmitry Banschikov.
238 PP/16 Removed "dont_insert_empty_fragments" fron "openssl_options".
239 Removed SSL_clear() after SSL_new() which led to protocol negotiation
240 failures. We appear to now support TLS1.1+ with Exim.
242 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
243 lets Exim select keys and certificates based upon TLS SNI from client.
244 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
245 before an outbound SMTP session. New log_selector, +tls_sni.
247 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
248 NULL dereference. Report and patch from Alun Jones.
250 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
251 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
252 Not seeing resolver debug output on NetBSD, but suspect this is a
253 resolver implementation change.
255 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
256 Left warnings. Added "eximon gdb" invocation mode.
258 PP/21 Defaulting "accept_8bitmime" to true, not false.
260 PP/22 Added -bw for inetd wait mode support.
262 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
263 locate the relevant includes and libraries. Made this the default.
265 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
266 Bugzilla 1246, report and most of solution from Tomasz Kusy.
268 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
269 This may cause build issues on older platforms.
271 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
272 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
273 gnutls_require_mac & gnutls_require_protocols (no longer supported).
274 Added SNI support via GnuTLS too.
275 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
277 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
279 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
280 Applied second patch from Janne, implementing suggestion to default
281 multiple-strings-in-record handling to match SPF spec.
283 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
285 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
286 read-only, out of scope).
287 Patch from Wolfgang Breyha, report from Stuart Northfield.
289 PP/29 Fix three issues highlighted by clang analyser static analysis.
290 Only crash-plausible issue would require the Cambridge-specific
291 iplookup router and a misconfiguration.
292 Report from Marcin Mirosław.
294 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
296 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
297 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
298 As part of this, removing so much warning spew let me fix some minor
299 real issues in debug logging.
301 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
302 assignment on my part. Fixed.
304 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
305 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
306 Janne Snabb (who went above and beyond: thank you).
308 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
309 string otherwise requires a connection and a bunch more work and it's
310 relatively easy to get wrong. Should also expose TLS library linkage
313 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
314 64-bit ${eval} (JH/03).
316 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
317 GNU libc to support some of the 64-bit stuff, should not lead to
318 conflicts. Defined before os.h is pulled in, so if a given platform
319 needs to override this, it can.
321 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
322 protection layer was required, which is not implemented.
323 Bugzilla 1254, patch from Wolfgang Breyha.
325 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
326 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
327 tls_dhparam take prime identifiers. Also unbreak combination of
328 OpenSSL+DH_params+TLSSNI.
330 PP/39 Disable SSLv2 by default in OpenSSL support.
336 PP/01 Solaris build fix for Oracle's LDAP libraries.
337 Bugzilla 1109, patch from Stephen Usher.
339 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
341 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
344 TF/02 Fix a couple more cases where we did not log the error message
345 when unlink() failed. See also change 4.74-TF/03.
347 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
348 lock up or crash if it happened to be inside a call to libc when it
349 got a SIGUSR1 from exiwhat.
351 The SIGUSR1 handler appends the current process status to the process
352 log which is later printed by exiwhat. It used to use the general
353 purpose logging code to do this, but several functions it calls are
354 not safe for signals.
356 The new output code in the SIGUSR1 handler is specific to the process
357 log, and simple enough that it's easy to inspect for signal safety.
358 Removing some special cases also simplifies the general logging code.
359 Removing the spurious timestamps from the process log simplifies
362 TF/04 Improved ratelimit ACL condition.
364 The /noupdate option has been deprecated in favour of /readonly which
365 has clearer semantics. The /leaky, /strict, and /readonly update modes
366 are mutually exclusive. The update mode is no longer included in the
367 database key; it just determines when the database is updated. (This
368 means that when you upgrde Exim will forget old rate measurements.)
370 Exim now checks that the per_* options are used with an update mode that
371 makes sense for the current ACL. For example, when Exim is processing a
372 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
373 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
374 must specify per_mail/readonly. If you omit the update mode it defaults to
375 /leaky where that makes sense (as before) or /readonly where required.
377 The /noupdate option is now undocumented but still supported for
378 backwards compatibility. It is equivalent to /readonly except that in
379 ACLs where /readonly is required you may specify /leaky/noupdate or
380 /strict/noupdate which are treated the same as /readonly.
382 A useful new feature is the /count= option. This is a generalization
383 of the per_byte option, so that you can measure the throughput of other
384 aggregate values. For example, the per_byte option is now equivalent
385 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
387 The per_rcpt option has been generalized using the /count= mechanism
388 (though it's more complicated than the per_byte equivalence). When it is
389 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
390 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
391 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
392 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
393 Note that using per_rcpt with a non-readonly update mode in more than
394 one ACL will cause the recipients to be double-counted. (The per_mail
395 and per_byte options don't have this problem.)
397 The handling of very low rates has changed slightly. If the computed rate
398 is less than the event's count (usually one) then this event is the first
399 after a long gap. In this case the rate is set to the same as this event's
400 count, so that the first message of a spam run is counted properly.
402 The major new feature is a mechanism for counting the rate of unique
403 events. The new per_addr option counts the number of different
404 recipients that someone has sent messages to in the last time period. It
405 behaves like per_rcpt if all the recipient addresses are different, but
406 duplicate recipient addresses do not increase the measured rate. Like
407 the /count= option this is a general mechanism, so the per_addr option
408 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
409 example, measure the rate that a client uses different sender addresses
410 with the options per_mail/unique=$sender_address. There are further
411 details in the main documentation.
413 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
415 TF/06 Removed a few PCRE remnants.
417 TF/07 Automatically extract Exim's version number from tags in the git
418 repository when doing development or release builds.
420 PP/02 Raise smtp_cmd_buffer_size to 16kB.
421 Bugzilla 879. Patch from Paul Fisher.
423 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
424 Heavily based on revision 40f9a89a from Simon Arlott's tree.
427 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
429 PP/05 Variable $av_failed, true if the AV scanner deferred.
430 Bugzilla 1078. Patch from John Horne.
432 PP/06 Stop make process more reliably on build failure.
433 Bugzilla 1087. Patch from Heiko Schlittermann.
435 PP/07 Make maildir_use_size_file an _expandable_ boolean.
436 Bugzilla 1089. Patch from Heiko Schlittermann.
438 PP/08 Handle ${run} returning more data than OS pipe buffer size.
439 Bugzilla 1131. Patch from Holger Weiß.
441 PP/09 Handle IPv6 addresses with SPF.
442 Bugzilla 860. Patch from Wolfgang Breyha.
444 PP/10 GnuTLS: support TLS 1.2 & 1.1.
446 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
449 PP/11 match_* no longer expand right-hand-side by default.
450 New compile-time build option, EXPAND_LISTMATCH_RHS.
451 New expansion conditions, "inlist", "inlisti".
453 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
455 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
457 PP/14 fix log_write() format string regression from TF/03.
458 Bugzilla 1152. Patch from Dmitry Isaikin.
464 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
466 PP/02 Harmonised TLS library version reporting; only show if debugging.
467 Layout now matches that introduced for other libraries in 4.74 PP/03.
469 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
471 PP/04 New "dns_use_edns0" global option.
473 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
476 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
477 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
479 TK/01 Updated PolarSSL code to 0.14.2.
480 Bugzilla 1097. Patch from Andreas Metzler.
482 PP/07 Catch divide-by-zero in ${eval:...}.
485 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
488 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
489 format-string attack -- SECURITY: remote arbitrary code execution.
491 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
492 time unintentionally subject to list matching rules, letting the header
493 cause arbitrary Exim lookups (of items which can occur in lists, *not*
494 arbitrary string expansion). This allowed for information disclosure.
496 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
497 INT_MIN/-1 -- value coerced to INT_MAX.
503 NM/01 Workround for PCRE version dependency in version reporting
506 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
507 This fixes portability to compilers other than gcc, notably
508 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
510 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
511 makefiles for portability to HP-UX and POSIX correctness.
513 PP/01 Permit LOOKUP_foo enabling on the make command-line.
514 Also via indented variable definition in the Makefile.
515 (Debugging by Oliver Heesakkers).
517 PP/02 Restore caching of spamd results with expanded spamd_address.
518 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
520 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
521 Improves build reliability. Fix from: Frank Elsner
523 NM/02 Fix wide character breakage in the rfc2047 coding
524 Fixes bug 1064. Patch from Andrey N. Oktyabrski
526 NM/03 Allow underscore in dnslist lookups
527 Fixes bug 1026. Patch from Graeme Fowler
529 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
530 Code patches from Adam Ciarcinski of NetBSD.
532 NM/04 Fixed exiqgrep to cope with mailq missing size issue
535 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
536 is logged, to avoid truncation. Patch from John Horne.
538 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
539 Patch from Jakob Hirsch.
541 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
542 SQL string expansion failure details.
543 Patch from Andrey Oktyabrski.
545 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
546 Patch from Simon Arlott.
548 PP/09 New lookups functionality failed to compile on old gcc which rejects
549 extern declarations in function scope.
550 Patch from Oliver Fleischmann
552 PP/10 Use sig_atomic_t for flags set from signal handlers.
553 Check getgroups() return and improve debugging.
554 Fixed developed for diagnosis in bug 927 (which turned out to be
557 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
558 Patch from Mark Zealey.
560 PP/12 Bugzilla 1056: Improved spamd server selection.
561 Patch from Mark Zealey.
563 PP/13 Bugzilla 1086: Deal with maildir quota file races.
564 Based on patch from Heiko Schlittermann.
566 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
567 Patch from Uwe Doering, sign-off by Michael Haardt.
569 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
570 variable declaration deep within a block. Bug and patch from
573 PP/15 lookups-Makefile IRIX compatibilty coercion.
575 PP/16 Make DISABLE_DKIM build knob functional.
577 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
578 Patch by Simon Arlott
580 TF/03 Fix valgrind.h portability to C89 compilers that do not support
581 variable argument macros. Our copy now differs from upstream.
587 TF/01 Failure to get a lock on a hints database can have serious
588 consequences so log it to the panic log.
590 TF/02 Log LMTP confirmation messages in the same way as SMTP,
591 controlled using the smtp_confirmation log selector.
593 TF/03 Include the error message when we fail to unlink a spool file.
595 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
596 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
597 for maintaining out-of-tree patches for some time.
599 PP/01 Bugzilla 139: Documentation and portability issues.
600 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
601 Handle per-OS dynamic-module compilation flags.
603 PP/02 Let /dev/null have normal permissions.
604 The 4.73 fixes were a little too stringent and complained about the
605 permissions on /dev/null. Exempt it from some checks.
606 Reported by Andreas M. Kirchwitz.
608 PP/03 Report version information for many libraries, including
609 Exim version information for dynamically loaded libraries. Created
610 version.h, now support a version extension string for distributors
611 who patch heavily. Dynamic module ABI change.
613 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
614 privilege escalation vulnerability whereby the Exim run-time user
615 can cause root to append content of the attacker's choosing to
618 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
621 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
622 If dropping privileges for untrusted macros, we disabled normal logging
623 on the basis that it would fail; for the Exim run-time user, this is not
624 the case, and it resulted in successful deliveries going unlogged.
625 Fixed. Reported by Andreas Metzler.
631 PP/01 Date: & Message-Id: revert to normally being appended to a message,
632 only prepend for the Resent-* case. Fixes regression introduced in
633 Exim 4.70 by NM/22 for Bugzilla 607.
635 PP/02 Include check_rfc2047_length in configure.default because we're seeing
636 increasing numbers of administrators be bitten by this.
638 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
640 PP/03 Bugzilla 994: added openssl_options main configuration option.
642 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
644 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
646 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
648 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
649 without a peer certificate, leading to a segfault because of an
650 assumption that peers always have certificates. Be a little more
651 paranoid. Problem reported by Martin Tscholak.
653 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
654 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
655 NB: ClamAV planning to remove STREAM in "middle of 2010".
656 CL also introduces -bmalware, various -d+acl logging additions and
657 more caution in buffer sizes.
659 PP/09 Implemented reverse_ip expansion operator.
661 PP/10 Bugzilla 937: provide a "debug" ACL control.
663 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
665 PP/12 Bugzilla 973: Implement --version.
667 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
669 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
671 PP/15 Bugzilla 816: support multiple condition rules on Routers.
673 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
674 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
675 ignore trailing whitespace.
677 JJ/02 prevent non-panic DKIM error from being sent to paniclog
679 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
682 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
683 Notification from Dr Andrew Aitchison.
685 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
686 ExtendedDetectionInfo response format.
687 Notification from John Horne.
689 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
692 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
693 XSL and documented dependency on system catalogs, with examples of how
696 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
699 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
700 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
701 configuration file which is writeable by the Exim user or group.
703 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
704 of configuration files to cover files specified with the -C option if
705 they are going to be used with root privileges, not just the default
708 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
709 option (effectively making it always true).
711 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
712 files to be used while preserving root privileges.
714 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
715 that rogue child processes cannot use them.
717 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
718 run-time user, instead of root.
720 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
721 Exim run-time user without dropping privileges.
723 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
724 result string, instead of calling string_vformat() twice with the same
727 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
728 for other users. Others should always drop root privileges if they use
729 -C on the command line, even for a whitelisted configure file.
731 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
733 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
739 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
740 $data_path, and $header_path variables; fixed documentation bugs and
743 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
744 exipick to access non-standard spools, including the "frozen" queue
747 NM/01 Bugzilla 965: Support mysql stored procedures.
748 Patch from Alain Williams
750 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
752 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
753 Patch from Andreas Metzler
755 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
756 Patch from Kirill Miazine
758 NM/05 Bugzilla 671: Added umask to procmail example.
760 JJ/03 installed exipick 20100323.0, fixing doc bug
762 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
763 directory. Notification and patch from Dan Rosenberg.
765 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
767 TK/02 Improve log output when DKIM signing operation fails.
769 MH/01 Treat the transport option dkim_domain as a colon separated
770 list, not as a single string, and sign the message with each element,
771 omitting multiple occurences of the same signer.
773 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
774 Bugzilla 985, 986. Patch by Simon Arlott
776 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
777 Patch by Simon Arlott
779 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
780 MBX locking. Notification from Dan Rosenberg.
786 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
788 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
790 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
792 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
794 NM/04 Bugzilla 915: Flag broken perl installation during build.
800 TK/01 Added patch by Johannes Berg that expands the main option
801 "spamd_address" if it starts with a dollar sign.
803 TK/02 Write list of recipients to X-Envelope-Sender header when building
804 the mbox-format spool file for content scanning (suggested by Jakob
807 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
808 (http://www.dcc-servers.net/) support via dccifd. Activated by
809 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
811 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
812 by Mark Daniel Reidel <mr@df.eu>.
814 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
815 When building exim an external PCRE library is now needed -
816 PCRE is a system library on the majority of modern systems.
817 See entry on PCRE_LIBS in EDITME file.
819 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
820 conversation. Added nologin parameter to request.
821 Patch contributed by Kirill Miazine.
823 TF/01 Do not log submission mode rewrites if they do not change the address.
825 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
827 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
828 log files in place. Contributed by Roberto Lima.
830 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
832 TF/03 Bugzilla 615: When checking the local_parts router precondition
833 after a local_part_suffix or local_part_prefix option, Exim now
834 does not use the address's named list lookup cache, since this
835 contains cached lookups for the whole local part.
837 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
838 Robert Millan. Documentation is in experimental-spec.txt.
840 TF/04 Bugzilla 668: Fix parallel build (make -j).
842 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
844 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
845 Patch provided by Jan Srzednicki.
847 TF/05 Leading white space used to be stripped from $spam_report which
848 wrecked the formatting. Now it is preserved.
850 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
851 that they are available at delivery time.
853 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
855 TF/08 TLS error reporting now respects the incoming_interface and
856 incoming_port log selectors.
858 TF/09 Produce a more useful error message if an SMTP transport's hosts
859 setting expands to an empty string.
861 NM/06 Bugzilla 744: EXPN did not work under TLS.
862 Patch provided by Phil Pennock.
864 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
865 Patch provided by Richard Godbee.
867 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
868 acl_smtp_notquit, added index entry.
870 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
871 Patch provided by Eugene Bujak.
873 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
874 accept(). Patch provided by Maxim Dounin.
876 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
877 Patch provided by Phil Pennock.
879 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
881 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
882 Patch provided by Brad "anomie" Jorsch.
884 NM/14 Bugzilla 622: Added timeout setting to transport filter.
885 Patch provided by Dean Brooks.
887 TK/05 Add native DKIM support (does not depend on external libraries).
889 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
890 Patch provided by Graeme Fowler.
892 NM/16 Bugzilla 851: Documentation example syntax fix.
894 NM/17 Changed NOTICE file to remove references to embedded PCRE.
896 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
899 NM/19 Bugzilla 745: TLS version reporting.
900 Patch provided by Phil Pennock.
902 NM/20 Bugzilla 167: bool: condition support.
903 Patch provided by Phil Pennock.
905 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
906 clients. Patch provided by Phil Pennock.
908 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
909 Patch provided by Brad "anomie" Jorsch.
911 NM/23 Bugzilla 687: Fix misparses in eximstats.
912 Patch provided by Heiko Schlittermann.
914 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
915 Patch provided by Heiko Schlittermann.
917 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
918 plus update to original patch.
920 NM/26 Bugzilla 799: Documentation correction for ratelimit.
922 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
923 Patch provided by David Brownlee.
925 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
927 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
929 NM/30 Bugzilla 888: TLS documentation bugfixes.
931 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
933 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
934 Unlike the original bugzilla I have changed all shell scripts in src tree.
936 NM/33 Bugzilla 898: Transport filter timeout fix.
937 Patch by Todd Rinaldo.
939 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
940 Patch by Serge Demonchaux.
942 NM/35 Bugzilla 39: Base64 decode bug fixes.
943 Patch by Jakob Hirsch.
945 NM/36 Bugzilla 909: Correct connect() call in dcc code.
947 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
949 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
951 NM/39 Bugzilla 911: Fixed MakeLinks build script.
957 TK/01 Add preliminary DKIM support. Currently requires a forked version of
958 ALT-N's libdkim that I have put here:
959 http://duncanthrax.net/exim-experimental/
961 Note to Michael Haardt: I had to rename some vars in sieve.c. They
962 were called 'true' and it seems that C99 defines that as a reserved
963 keyword to be used with 'bool' variable types. That means you could
964 not include C99-style headers which use bools without triggering
965 build errors in sieve.c.
967 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
968 as mailq or other aliases. Changed the --help handling significantly
969 to do whats expected. exim_usage() emits usage/help information.
971 SC/01 Added the -bylocaldomain option to eximstats.
973 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
975 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
977 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
983 PH/01 Another patch from the Sieve maintainer.
985 PH/02 When an IPv6 address is converted to a string for single-key lookup
986 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
987 dots are used instead of colons so that keys in lsearch files need not
988 contain colons. This was done some time before quoting was made available
989 in lsearch files. However, iplsearch files do require colons in IPv6 keys
990 (notated using the quote facility) so as to distinguish them from IPv4
991 keys. This meant that lookups for IP addresses in host lists did not work
992 for iplsearch lookups.
994 This has been fixed by arranging for IPv6 addresses to be expressed with
995 colons if the lookup type is iplsearch. This is not incompatible, because
996 previously such lookups could never work.
998 The situation is now rather anomolous, since one *can* have colons in
999 ordinary lsearch keys. However, making the change in all cases is
1000 incompatible and would probably break a number of configurations.
1002 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
1005 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
1006 conversion specification without a maximum field width, thereby enabling
1007 a rogue spamd server to cause a buffer overflow. While nobody in their
1008 right mind would setup Exim to query an untrusted spamd server, an
1009 attacker that gains access to a server running spamd could potentially
1010 exploit this vulnerability to run arbitrary code as the Exim user.
1012 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
1013 $primary_hostname instead of what libspf2 thinks the hosts name is.
1015 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
1016 a directory entry by the name of the lookup key. Previously, if a
1017 symlink pointed to a non-existing file or a file in a directory that
1018 Exim lacked permissions to read, a lookup for a key matching that
1019 symlink would fail. Now it is enough that a matching directory entry
1020 exists, symlink or not. (Bugzilla 503.)
1022 PH/03 The body_linecount and body_zerocount variables are now exported in the
1025 PH/04 Added the $dnslist_matched variable.
1027 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
1028 This means they are set thereafter only if the connection becomes
1031 PH/06 Added the client_condition to authenticators so that some can be skipped
1032 by clients under certain conditions.
1034 PH/07 The error message for a badly-placed control=no_multiline_responses left
1035 "_responses" off the end of the name.
1037 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
1039 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
1040 (without spaces) instead of just copying the configuration text.
1042 PH/10 Added the /noupdate option to the ratelimit ACL condition.
1044 PH/11 Added $max_received_linelength.
1046 PH/12 Added +ignore_defer and +include_defer to host lists.
1048 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
1049 way in which PCRE > 7.0 is built.
1051 PH/14 Implemented queue_only_load_latch.
1053 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
1054 MAIL command. The effect was to mangle the value on 64-bit systems.
1056 PH/16 Another patch from the Sieve maintainer.
1058 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
1060 PH/18 If a system quota error occurred while trying to create the file for
1061 a maildir delivery, the message "Mailbox is full" was not appended to the
1062 bounce if the delivery eventually timed out. Change 4.67/27 below applied
1063 only to a quota excession during the actual writing of the file.
1065 PH/19 It seems that peer DN values may contain newlines (and other non-printing
1066 characters?) which causes problems in log lines. The DN values are now
1067 passed through string_printing() before being added to log lines.
1069 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
1070 and InterBase are left for another time.)
1072 PH/21 Added message_body_newlines option.
1074 PH/22 Guard against possible overflow in moan_check_errorcopy().
1076 PH/23 POSIX allows open() to be a macro; guard against that.
1078 PH/24 If the recipient of an error message contained an @ in the local part
1079 (suitably quoted, of course), incorrect values were put in $domain and
1080 $local_part during the evaluation of errors_copy.
1086 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
1087 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
1090 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
1091 issue a MAIL command.
1093 PH/02 In an ACL statement such as
1095 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
1097 if a client was not listed at all, or was listed with a value other than
1098 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
1099 the condition was not true (as it should be), so access was not denied.
1100 The bug was that the ! inversion was incorrectly passed on to the second
1101 item. This has been fixed.
1103 PH/03 Added additional dnslists conditions == and =& which are different from
1104 = and & when the dns lookup returns more than one IP address.
1106 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
1107 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
1109 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
1110 FSYNC, which compiles an option called disable_fsync that allows for
1111 bypassing fsync(). The documentation is heavily laced with warnings.
1113 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
1115 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
1116 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
1117 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
1118 including adding "make clean"; (3) Added -fPIC when compiling the test
1119 dynamically loaded module, to get rid of a warning.
1121 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
1122 message fails, move_frozen_messages = true and ignore_bounce_errors_after
1123 = 0s. The bug is otherwise harmless.
1125 PH/07 There was a bug in the dovecot authenticator such that the value of
1126 $auth1 could be overwritten, and so not correctly preserved, after a
1127 successful authentication. This usually meant that the value preserved by
1128 the server_setid option was incorrect.
1130 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
1132 PH/09 Installed PCRE release 7.0.
1134 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
1135 run for batched SMTP input. It is now run at the start of every message
1136 in the batch. While fixing this I discovered that the process information
1137 (output by running exiwhat) was not always getting set for -bs and -bS
1138 input. This is fixed, and it now also says "batched" for BSMTP.
1140 PH/11 Added control=no_pipelining.
1142 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
1143 patch, slightly modified), and move the expansion of helo_data till after
1144 the connection is made in the smtp transport (so it can use these
1147 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
1149 PH/14 Added log_selector = +pid.
1151 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
1153 PH/16 Add ${if forany and ${if forall.
1155 PH/17 Added dsn_from option to vary the From: line in DSNs.
1157 PH/18 Flush SMTP output before performing a callout, unless control =
1158 no_callout_flush is set.
1160 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
1161 was true (the default) a successful delivery failed to delete the retry
1162 item, thus causing premature timeout of the address. The bug is now
1165 PH/20 Added hosts_avoid_pipelining to the smtp transport.
1167 PH/21 Long custom messages for fakedefer and fakereject are now split up
1168 into multiline reponses in the same way that messages for "deny" and
1169 other ACL rejections are.
1171 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
1172 with slight modification.
1174 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
1175 draft, changing the syntax and factoring some duplicate code".
1177 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
1178 for deliveries of the second and subsequent messages over the same SMTP
1181 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
1182 ${reduce, with only minor "tidies".
1184 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
1186 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
1187 expansion side effects.
1189 PH/27 When a message times out after an over-quota error from an Exim-imposed
1190 quota, the bounce message says "mailbox is full". This message was not
1191 being given when it was a system quota that was exceeded. It now should
1194 MH/03 Made $recipients available in local_scan(). local_scan() already has
1195 better access to the recipient list through recipients_list[], but
1196 $recipients can be useful in postmaster-provided expansion strings.
1198 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
1199 in the case of a MAIL command with additional options following the
1200 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
1201 were accidentally chopped off.
1203 PH/29 SMTP synchronization checks are implemented when a command is read -
1204 there is a check that no more input is waiting when there shouldn't be
1205 any. However, for some commands, a delay in an ACL can mean that it is
1206 some time before the response is written. In this time, more input might
1207 arrive, invalidly. So now there are extra checks after an ACL has run for
1208 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
1209 pipelining has not been advertised.
1211 PH/30 MH's patch to allow iscntrl() characters to be list separators.
1213 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
1214 returned in the SMTP response when smtp_return_error_details was false.
1215 This has been fixed.
1217 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
1218 instead of the C I/O that was originally supplied, because problems were
1219 reported on Solaris.
1221 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
1222 Exim which did not show up earlier: it was assuming that a call to
1223 SSL_CTX_set_info_callback() might give an error value. In fact, there is
1224 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
1225 was a macro that became an assignment, so it seemed to work. This has
1226 changed to a proper function call with a void return, hence the compile
1227 error. Exim's code has been fixed.
1229 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
1232 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
1234 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
1236 PH/37 If a message is not accepted after it has had an id assigned (e.g.
1237 because it turns out to be too big or there is a timeout) there is no
1238 "Completed" line in the log. When some messages of this type were
1239 selected by exigrep, they were listed as "not completed". Others were
1240 picked up by some special patterns. I have improved the selection
1241 criteria to be more general.
1243 PH/38 The host_find_failed option in the manualroute router can now be set
1244 to "ignore", to completely ignore a host whose IP address cannot be
1245 found. If all hosts are ignored, the behaviour is controlled by the new
1246 host_all_ignored option.
1248 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
1249 homing or because of multiple MX records with /mx) generated more than
1250 one IP address, and the following item turned out to be the local host,
1251 all the secondary addresses of the first item were incorrectly removed
1252 from the list, along with the local host and any following hosts (which
1253 is what is supposed to happen).
1255 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
1256 whoever called Exim into the -H file. In the case of the daemon it was
1257 behaving confusingly. When first started, it used values for whoever
1258 started the daemon, but after a SIGHUP it used the Exim user (because it
1259 calls itself on a restart). I have changed the code so that it now always
1262 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
1263 message are rejected with the same error (e.g. no authentication or bad
1264 sender address), and a DATA command is nevertheless sent (as can happen
1265 with PIPELINING or a stupid MUA), the error message that was given to the
1266 RCPT commands is included in the rejection of the DATA command. This is
1267 intended to be helpful for MUAs that show only the final error to their
1270 PH/42 Another patch from the Sieve maintainer.
1272 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
1273 Eximstats - Fixed some broken HTML links and added missing column headers
1275 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
1276 columns for Rejects, Temp Rejects, Ham, and Spam rows.
1278 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
1280 PH/43 Yet another patch from the Sieve maintainer.
1282 PH/44 I found a way to check for a TCP/IP connection going away before sending
1283 the response to the final '.' that terminates a message, but only in the
1284 case where the client has not sent further data following the '.'
1285 (unfortunately, this is allowed). However, in many cases there won't be
1286 any further data because there won't be any more messages to send. A call
1287 to select() can be used: if it shows that the input is "ready", there is
1288 either input waiting, or the socket has been closed. An attempt to read
1289 the next input character can distinguish the two cases. Previously, Exim
1290 would have sent an OK response which the client would never have see.
1291 This could lead to message repetition. This fix should cure that, at
1292 least in a lot of common cases.
1294 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
1295 advertised in response to EHLO.
1301 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
1302 fixed by 4.65/MH/01 (is this a record?) are fixed:
1304 (i) An empty string was always treated as zero by the numeric comparison
1305 operators. This behaviour has been restored.
1307 (ii) It is documented that the numeric comparison operators always treat
1308 their arguments as decimal numbers. This was broken in that numbers
1309 starting with 0 were being interpreted as octal.
1311 While fixing these problems I realized that there was another issue that
1312 hadn't been noticed. Values of message_size_limit (both the global option
1313 and the transport option) were treated as octal if they started with 0.
1314 The documentation was vague. These values are now always treated as
1315 decimal, and I will make that clear in the documentation.
1321 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
1322 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
1325 MH/01 Don't check that the operands of numeric comparison operators are
1326 integers when their expansion is in "skipping" mode (fixes bug
1327 introduced by 4.64-PH/07).
1329 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
1330 child addresses, Exim now panics and dies. Previously, because the count
1331 is held in a short int, deliveries were likely to be lost. As such a
1332 large number of recipients for a single message is ridiculous
1333 (performance will be very, very poor), I have chosen to impose a limit
1334 rather than extend the field.
1340 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
1341 leftover -K file (the existence of which was triggered by #402).
1342 While we were at it, introduced process PID as part of the -K
1343 filename. This should rule out race conditions when creating
1346 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
1347 processing considerably. Previous code took too long for large mails,
1348 triggering a timeout which in turn triggers #401.
1350 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
1351 in the DK code in transports.c. sendfile() is not really portable,
1352 hence the _LINUX specificness.
1354 TF/01 In the add_headers option to the mail command in an Exim filter,
1355 there was a bug that Exim would claim a syntax error in any
1356 header after the first one which had an odd number of characters
1359 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
1360 callout verification, Exim cached a "reject" for the entire domain. This
1361 is correct for most verifications, but it is not correct for a recipient
1362 verification with use_sender or use_postmaster set, because in that case
1363 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
1364 case of MAIL FROM:<> rejection from other early rejections (e.g.
1365 rejection of HELO). When verifying a recipient using a non-null MAIL
1366 address, the cache is ignored if it shows MAIL FROM:<> rejection.
1367 Whatever the result of the callout, the value of the domain cache is
1368 left unchanged (for any other kind of callout, getting as far as trying
1369 RCPT means that the domain itself is ok).
1371 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
1374 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
1375 manifest itself as EPIPE rather than ECONNECT. When tidying away a
1376 session, the daemon ignores ECONNECT errors and logs others; it now
1377 ignores EPIPE as well.
1379 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
1380 (quoted-printable decoding).
1382 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
1383 later the small subsequent patch to fix an introduced bug.
1385 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
1387 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
1389 PH/08 An error is now given if message_size_limit is specified negative.
1391 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
1392 to be given (somewhat) arbitrary names.
1394 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
1397 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
1398 miscellaneous code fixes
1400 PH/10 Added the log_reject_target ACL modifier to specify where to log
1403 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
1404 hostname. This is wrong, because it relates to the incoming message (and
1405 probably the interface on which it is arriving) and not to the outgoing
1406 callout (which could be using a different interface). This has been
1407 changed to use the value of the helo_data option from the smtp transport
1408 instead - this is what is used when a message is actually being sent. If
1409 there is no remote transport (possible with a router that sets up host
1410 addresses), $smtp_active_hostname is used.
1412 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
1413 tweaks were necessary in order to get it to work (see also 21 below):
1414 (a) The code assumed that strncpy() returns a negative number on buffer
1415 overflow, which isn't the case. Replaced with Exim's string_format()
1417 (b) There were several signed/unsigned issues. I just did the minimum
1418 hacking in of casts. There is scope for a larger refactoring.
1419 (c) The code used strcasecmp() which is not a standard C function.
1420 Replaced with Exim's strcmpic() function.
1421 (d) The code set only $1; it now sets $auth1 as well.
1422 (e) A simple test gave the error "authentication client didn't specify
1423 service in request". It would seem that Dovecot has changed its
1424 interface. Fortunately there's a specification; I followed it and
1425 changed what the client sends and it appears to be working now.
1427 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
1430 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
1431 address A is aliased to B and C, where B exists and C does not. Without
1432 -v the output is "A verified" because verification stops after a
1433 successful redirection if more than one address is generated. However,
1434 with -v the child addresses are also verified. Exim was outputting "A
1435 failed to verify" and then showing the successful verification for C,
1436 with its parentage. It now outputs "B failed to verify", showing B's
1437 parentage before showing the successful verification of C.
1439 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
1440 look up a TXT record in a specific list after matching in a combined
1443 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
1444 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
1445 they consult the DNS. I had assumed they would set it the way they
1446 wanted; and indeed my experiments on Linux seem to show that in some
1447 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
1448 To be on the safe side, however, I have now made the interface to
1449 host_find_byname() similar to host_find_bydns(), with an argument
1450 containing the DNS resolver options. The host_find_byname() function now
1451 sets these options at its start, just as host_find_bydns() does. The smtp
1452 transport options dns_qualify_single and dns_search_parents are passed to
1453 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
1454 of host_find_byname() use the default settings of RES_DEFNAMES
1455 (qualify_single) but not RES_DNSRCH (search_parents).
1457 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
1458 spool_read_header() do less string testing, by means of a preliminary
1459 switch on the second character of optional "-foo" lines. (This is
1460 overdue, caused by the large number of possibilities that now exist.
1461 Originally there were few.) While I was there, I also converted the
1462 str(n)cmp tests so they don't re-test the leading "-" and the first
1463 character, in the hope this might squeeze out yet more improvement.
1465 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
1466 flag allowing group syntax was set by the header_syntax check but not
1467 turned off, possible causing trouble later; (2) The flag was not being
1468 set at all for the header_verify test, causing "group"-style headers to
1469 be rejected. I have now set it in this case, and also caused header_
1470 verify to ignore an empty address taken from a group. While doing this, I
1471 came across some other cases where the code for allowing group syntax
1472 while scanning a header line wasn't quite right (mostly, not resetting
1473 the flag correctly in the right place). These bugs could have caused
1474 trouble for malformed header lines. I hope it is now all correct.
1476 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
1477 with the "reply" argument non-NULL. The code, however (which originally
1478 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
1479 but it didn't always do it. This confused somebody who was copying the
1480 code for some other use. I have removed all the tests.
1482 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
1483 feature that was used to support insecure browsers during the U.S. crypto
1484 embargo. It requires special client support, and Exim is probably the
1485 only MTA that supported it -- and would never use it because real RSA is
1486 always available. This code has been removed, because it had the bad
1487 effect of slowing Exim down by computing (never used) parameters for the
1488 RSA_EXPORT functionality.
1490 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
1491 authenticator to fail if there's a tab character in the incoming data
1492 (there should never be unless someone is messing about, as it's supposed
1493 to be base64-encoded). Also added, on Timo's advice, the "secured" option
1494 if the connection is using TLS or if the remote IP is the same as the
1495 local IP, and the "valid-client-cert option" if a client certificate has
1498 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
1499 authenticators. This can be used for authorization after authentication
1500 succeeds. (In the case of plaintext, it servers for both authentication
1503 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
1504 if any retry times were supplied.
1506 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
1507 connection, where there is no client IP address to check. In this
1508 situation, the verify now always succeeds.
1510 PH/25 Applied John Jetmore's -Mset patch.
1512 PH/26 Added -bem to be like -Mset, but loading a message from a file.
1514 PH/27 In a string expansion for a processed (not raw) header when multiple
1515 headers of the same name were present, leading whitespace was being
1516 removed from all of them, but trailing whitespace was being removed only
1517 from the last one. Now trailing whitespace is removed from each header
1518 before concatenation. Completely empty headers in a concatenation (as
1519 before) are ignored.
1521 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
1522 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
1524 PH/29 [Removed. This was a change that I later backed out, and forgot to
1525 correct the ChangeLog entry (that I had efficiently created) before
1526 committing the later change.]
1528 PH/30 Exim was sometimes attempting to deliver messages that had suffered
1529 address errors (4xx response to RCPT) over the same connection as other
1530 messages routed to the same hosts. Such deliveries are always "forced",
1531 so retry times are not inspected. This resulted in far too many retries
1532 for the affected addresses. The effect occurred only when there were more
1533 hosts than the hosts_max_try setting in the smtp transport when it had
1534 the 4xx errors. Those hosts that it had tried were not added to the list
1535 of hosts for which the message was waiting, so if all were tried, there
1536 was no problem. Two fixes have been applied:
1538 (i) If there are any address or message errors in an SMTP delivery, none
1539 of the hosts (tried or untried) are now added to the list of hosts
1540 for which the message is waiting, so the message should not be a
1541 candidate for sending over the same connection that was used for a
1542 successful delivery of some other message. This seems entirely
1543 reasonable: after all the message is NOT "waiting for some host".
1544 This is so "obvious" that I'm not sure why it wasn't done
1545 previously. Hope I haven't missed anything, but it can't do any
1546 harm, as the worst effect is to miss an optimization.
1548 (ii) If, despite (i), such a delivery is accidentally attempted, the
1549 routing retry time is respected, so at least it doesn't keep
1550 hammering the server.
1552 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
1553 in ${readsocket because some servers need this prod.
1555 PH/32 Added some extra debug output when updating a wait-xxx database.
1557 PH/33 The hint "could be header name not terminated by colon", which has been
1558 given for certain expansion errors for a long time, was not being given
1559 for the ${if def:h_colon_omitted{... case.
1561 PH/34 The spec says: "With one important exception, whenever a domain list is
1562 being scanned, $domain contains the subject domain." There was at least
1563 one case where this was not true.
1565 PH/35 The error "getsockname() failed: connection reset by peer" was being
1566 written to the panic log as well as the main log, but it isn't really
1567 panic-worthy as it just means the connection died rather early on. I have
1568 removed the panic log writing for the ECONNRESET error when getsockname()
1571 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
1572 runs only) independently of the message's sender address. This meant
1573 that, if the 4xx error was in fact related to the sender, a different
1574 message to the same recipient with a different sender could confuse
1575 things. In particualar, this can happen when sending to a greylisting
1576 server, but other circumstances could also provoke similar problems.
1577 I have changed the default so that the retry time for these errors is now
1578 based a combination of the sender and recipient addresses. This change
1579 can be overridden by setting address_retry_include_sender=false in the
1582 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
1583 remote server are returned as part of bounce messages. This was not
1584 happening for LMTP over a pipe (the lmtp transport), but now it is the
1585 same for both kinds of LMTP.
1587 PH/38 Despite being documented as not happening, Exim was rewriting addresses
1588 in header lines that were in fact CNAMEs. This is no longer the case.
1590 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
1591 and queue runs started by the daemon processed all messages. This has
1592 been fixed so that -R and -S can now usefully be given with -q<time>.
1594 PH/40 Import PCRE release 6.7 (fixes some bugs).
1596 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
1598 PH/42 Give an error if -q is specified more than once.
1600 PH/43 Renamed the variables $interface_address and $interface_port as
1601 $received_ip_address and $received_port, to make it clear that these
1602 values apply to message reception, and not to the outgoing interface when
1603 a message is delivered. (The old names remain recognized, of course.)
1605 PH/44 There was no timeout on the connect() call when using a Unix domain
1606 socket in the ${readsocket expansion. There now is.
1608 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
1609 be meaningful with "accept".
1611 SC/01 Eximstats V1.43
1612 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
1614 SC/02 Eximstats V1.44
1615 Use a glob alias rather than an array ref in the generated
1616 parser. This improves both readability and performance.
1618 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
1619 Collect SpamAssassin and rejection statistics.
1620 Don't display local sender or destination tables unless
1621 there is data to show.
1622 Added average volumes into the top table text output.
1624 SC/04 Eximstats V1.46
1625 Collect data on the number of addresses (recipients)
1626 as well as the number of messages.
1628 SC/05 Eximstats V1.47
1629 Added 'Message too big' to the list of mail rejection
1630 reasons (thanks to Marco Gaiarin).
1632 SC/06 Eximstats V1.48
1633 Mainlog lines which have GMT offsets and are too short to
1634 have a flag are now skipped.
1636 SC/07 Eximstats V1.49 (Alain Williams)
1637 Added the -emptyok flag.
1639 SC/08 Eximstats V1.50
1640 Fixes for obtaining the IP address from reject messages.
1642 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
1643 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
1644 whitesspace changes from 4.64-PH/27
1646 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
1649 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
1650 are found, allow negative numbers in numeric criteria)
1652 JJ/06 exipick.20061117.2, added new $message_body_missing variable
1654 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
1655 to match changes made in 4.64-PH/43
1657 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
1659 PH/47 Put in an explicit test for a DNS lookup of an address record where the
1660 "domain" is actually an IP address, and force a failure. This locks out
1661 those revolvers/nameservers that support "A-for-A" lookups, in
1662 contravention of the specifications.
1664 PH/48 When a host name was looked up from an IP address, and the subsequent
1665 forward lookup of the name timed out, the host name was left in
1666 $sender_host_name, contrary to the specification.
1668 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
1669 restricted to single-key lookups, Exim was not diagnosing an error if
1670 * or *@ was used with a query-style lookup.
1672 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
1674 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
1675 long ago, but noone interested enough thought of it. Let's just say that
1676 the "1.1" means that there are some new functions that weren't there at
1677 some point in the past.
1679 PH/51 Error processing for expansion failure of helo_data from an smtp
1680 transport during callout processing was broken.
1682 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
1683 tested/used via the -bh/-bhc/-bs options.
1685 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
1686 bug, fixed in subsequent PCRE releases).
1688 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
1689 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
1691 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
1697 SC/01 Use a glob alias rather than an array ref in eximstats generated
1698 parser. This improves both readability and performance.
1700 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
1701 Don't display local sender or destination tables in eximstats unless
1702 there is data to show.
1703 Added average volumes into the eximstats top table text output.
1705 SC/03 Collect data on the number of addresses (recipients) as well
1706 as the number of messages in eximstats.
1708 TF/01 Correct an error in the documentation for the redirect router. Exim
1709 does (usually) call initgroups() when daemonizing.
1711 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
1712 with consistent privilege compared to when running as a daemon.
1714 TF/03 Note in the spec that $authenticated_id is not set for local
1715 submissions from trusted users.
1717 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
1718 Thanks to Dean Brooks <dean@iglou.com> for the patch.
1720 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
1721 by adding some example configuration directives to the default
1722 configuration file. A little bit of work is required to uncomment the
1723 directives and define how usernames and passwords are checked, but
1724 there is now a framework to start from.
1726 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
1727 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
1728 without this. I don't know how relevant this is to other LDAP libraries.
1730 PH/02 Add the verb name to the "unknown ACL verb" error.
1732 PH/03 Magnus Holmgren's patch for filter_prepend_home.
1734 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
1736 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
1737 directory not expanded when it should be if an expanded home directory
1738 was set for the address (which is overridden by the transport).
1740 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
1743 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
1744 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
1745 because it is too late at that time, and has no effect.
1747 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
1748 security issue with \' (bugzilla #107). I could not use the
1749 PQescapeStringConn() function, because it needs a PGconn value as one of
1752 PH/08 When testing addresses using -bt, indicate those final addresses that
1753 are duplicates that would not cause an additional delivery. At least one
1754 person was confused, thinking that -bt output corresponded to deliveries.
1755 (Suppressing duplicates isn't a good idea as you lose the information
1756 about possibly different redirections that led to the duplicates.)
1758 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
1759 systems where poll() doesn't work, in particular OS X.
1761 PH/10 Added more information to debugging output for retry time not reached.
1763 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
1764 operations in malware.c.
1766 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
1769 PH/13 If write_rejectlog was set false when logging was sent to syslog with
1770 syslog_duplication set false, log lines that would normally be written
1771 both the the main log and to the reject log were not written to syslog at
1774 PH/14 In the default configuration, change the use of "message" in ACL warn
1775 statements to "add_header".
1777 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
1778 not followed by a command (e.g. "seen endif").
1780 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
1781 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
1784 PH/17 Added extra conditions to the default value of delay_warning_condition
1788 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
1789 { match{$h_precedence:}{(?i)bulk|list|junk} } \
1790 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
1793 The Auto-Submitted: and various List- headers are standardised, whereas I
1794 don't think Precedence: ever was.
1796 PH/18 Refactored debugging code in route_finduser() to show more information,
1797 in particular, the error code if getpwnam() issues one.
1799 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
1800 This is apparently needed in addition to the PH/07 change above to avoid
1801 any possible encoding problems.
1803 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
1804 but not after initializing Perl.
1806 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
1807 output them only if debugging. By default they are written stderr,
1808 apparently, which is not desirable.
1810 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
1813 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
1816 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
1818 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
1819 authenticated or an ident call has been made. Suppress the default
1820 values for $authenticated_id and $authenticated_sender (but permit -oMai
1821 and -oMas) when testing with -bh.
1823 PH/24 Re-jigged the order of the tests in the default configuration so that the
1824 tests for valid domains and recipients precede the DNS black list and CSA
1825 tests, on the grounds that those ones are more expensive.
1827 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
1828 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
1829 This bug exists in every version of Exim that I still have, right back to
1832 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
1833 However, an attempt to turn on case-sensitivity in a regex key by
1834 including (?-i) didn't work because the subject string was already
1835 lowercased, and the effects were non-intuitive. It turns out that a
1836 one-line patch can be used to allow (?-i) to work as expected.
1842 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
1843 other effects) broke the use of negated acl sub-conditions.
1845 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
1848 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
1849 "Deny" causes Exim to reject the incoming connection with a 554 error.
1850 Unfortunately, if there is a major crisis, such as a disk failure,
1851 tcp-wrappers gives "deny", whereas what one would like would be some
1852 kind of temporary error. A kludge has been added to help with this.
1853 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
1854 554 error is used if errno is still zero or contains ENOENT (which occurs
1855 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
1858 PH/03 Add -lutil to the default FreeBSD LIBS setting.
1860 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
1861 errors. Otherwise a message that provokes a temporary error (when other
1862 messages do not) can cause a whole host to time out.
1864 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
1865 addresses were routed directly to files or pipes from a redirect router.
1866 File deliveries just didn't batch; pipe deliveries might have suffered
1869 PH/06 A failure to get a lock for a hints database would erroneously always say
1870 "Failed to get write lock", even when it was really a read lock.
1872 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
1873 of 0600. This has been changed to use the value of the lockfile_mode
1874 option (which defaults to 0600).
1876 PH/08 Applied small patch from the Sieve maintainer.
1878 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
1879 folder from quota calculations, a direct delivery into this folder messed
1880 up the contents of the maildirsize file. This was because the regex was
1881 used only to exclude .Trash (or whatever) when the size of the mailbox
1882 was calculated. There was no check that a delivery was happening into an
1883 excluded directory. This bug has been fixed by ignoring all quota
1884 processing for deliveries into excluded directories.
1886 PH/10 Added the maildirfolder_create_regex option to appendfile.
1892 PH/01 The code for finding all the local interface addresses on a FreeBSD
1893 system running IPv6 was broken. This may well have applied to all BSD
1894 systems, as well as to others that have similar system calls. The broken
1895 code found IPv4 interfaces correctly, but gave incorrect values for the
1896 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
1897 that it would not match correctly against @[] and not recognize the IPv6
1900 PH/02 The ipliteral router was not recognizing addresses of the form user@
1901 [ipv6:....] because it didn't know about the "ipv6:" prefix.
1903 PH/03 Added disable_ipv6.
1905 PH/04 Changed $reply_address to use the raw form of the headers instead of the
1906 decoded form, because it is most often used to construct To: headers
1907 lines in autoreplies, and the decoded form may well be syntactically
1908 invalid. However, $reply_address has leading white space removed, and all
1909 newlines turned into spaces so that the autoreply transport does not
1912 PH/05 If group was specified without a user on a router, and no group or user
1913 was specified on a transport, the group from the router was ignored.
1915 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
1916 for visible compile-time settings that can be used to change these
1917 numbers, for those that want even more. Backwards compatibility with old
1918 spool files has been maintained. However, going back to a previous Exim
1919 release will lost any variables that are in spool files.
1921 PH/07 Two small changes when running in the test harness: increase delay when
1922 passing a TCP/IP connection to a new process, in case the original
1923 process has to generate a bounce, and remove special handling of
1924 127.0.0.2 (sic), which is no longer necessary.
1926 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
1927 be the same on different OS.
1929 PH/09 Moved a debug statement in filter processing to avoid a race problem when
1932 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
1933 whether --show-vars was specified or not
1935 JJ/02 exipick: Added support for new ACL variable spool format introduced
1938 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
1939 syntactically invalid From: or Reply-to: line, and a filter used this to
1940 generate an autoreply, and therefore failed to obtain an address for the
1941 autoreply, Exim could try to deliver to a non-existent relative file
1942 name, causing unrelated and misleading errors. What now happens is that
1943 it logs this as a hard delivery error, but does not attempt to create a
1946 PH/11 The exinext utility has a -C option for testing purposes, but although
1947 the given file was scanned by exinext itself; it wasn't being passed on
1948 when Exim was called.
1950 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
1951 an end-of-file indication when reading a command response.
1953 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
1954 compiled. In many other places in Exim, IPv6 addresses are always
1955 recognized, so I have changed this. It also means that IPv4 domain
1956 literals of the form [IPV4:n.n.n.n] are now always recognized.
1958 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
1959 used if the router is not running as root, for example, when verifying at
1960 ACL time, or when using -bh. The debugging output from this situation was
1961 non-existent - all you got was a failure to exec. I have made two
1964 (a) Failures to set uid/gid, the current directory, or a process leader
1965 in a subprocess such as that created by queryprogram now generate
1966 suitable debugging ouput when -d is set.
1968 (b) The queryprogram router detects when it is not running as root,
1969 outputs suitable debugging information if -d is set, and then runs
1970 the subprocess without attempting to change uid/gid.
1972 PH/15 Minor change to Makefile for building test_host (undocumented testing
1975 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
1976 additional section of a DNS packet that returns MX or SRV records.
1977 Instead, it always explicitly searches for A/AAAA records. This avoids
1978 major problems that occur when a DNS server includes only records of one
1979 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
1980 fixed another bug: if SRV records were looked up and the corresponding
1981 address records were *not* found in the additional section, the port
1982 values from the SRV records were lost.
1984 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
1985 using the correct key (the original address) when searching the retry
1986 rules in order to find which one to use for generating the retry hint.
1988 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
1989 adding the default one. Similarly, if it contains a Reply-To: header, the
1990 errors_reply_to option, if set, is not used.
1992 PH/19 When calculating a retry time, Exim used to measure the "time since
1993 failure" by looking at the "first failed" field in the retry record. Now
1994 it does not use this if it is later than than the arrival time of the
1995 message. Instead it uses the arrival time. This makes for better
1996 behaviour in cases where some deliveries succeed, thus re-setting the
1997 "first failed" field. An example is a quota failure for a huge message
1998 when small messages continue to be delivered. Without this change, the
1999 "time since failure" will always be short, possible causing more frequent
2000 delivery attempts for the huge message than are intended.
2001 [Note: This change was subsequently modified - see PH/04 for 4.62.]
2003 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
2004 $1, $2, $3) because the numerical variables can be reset during some
2005 expansion items (e.g. "match"), thereby losing the authentication data.
2007 PH/21 Make -bV show the size of off_t variables so that the test suite can
2008 decide whether to run tests for quotas > 2G.
2010 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
2011 mailbox_size, and mailbox_filecount in the appendfile transport. If a
2012 filecount value is greater than 2G or if a quota value is greater than 2G
2013 on a system where the size of off_t is not greater than 4, a panic error
2016 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
2017 never match. The debug and -bh output now contains an explicit error
2018 message indicating a malformed IPv4 address or mask.
2020 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
2021 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
2022 PH/23 above applies.
2024 PH/25 Do not write to syslog when running in the test harness. The only
2025 occasion when this arises is a failure to open the main or panic logs
2026 (for which there is an explicit test).
2028 PH/26 Added the /no_tell option to "control=freeze".
2030 PH/27 If a host name lookup failed very early in a connection, for example, if
2031 the IP address matched host_lookup and the reverse lookup yielded a name
2032 that did not have a forward lookup, an error message of the form "no IP
2033 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
2034 could be logged. Now it outputs the IP address instead of "NULL".
2036 PH/28 An enabling patch from MH: add new function child_open_exim2() which
2037 allows the sender and the authenticated sender to be set when
2038 submitting a message from within Exim. Since child_open_exim() is
2039 documented for local_scan(), the new function should be too.
2041 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
2042 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
2043 results in an empty string is now treated as unset.
2045 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
2047 PH/31 Added sender_verify_fail logging option.
2049 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
2050 needed by RFC 822 but not by RFC 2822 was commented out. I have now
2051 tidied the source and removed it altogether.
2053 PH/33 When a queue run was abandoned because the load average was too high, a
2054 log line was always written; now it is written only if the queue_run log
2055 selector is set. In addition, the log line for abandonment now contains
2056 information about the queue run such as the pid. This is always present
2057 in "start" and "stop" lines but was omitted from the "abandon" line.
2059 PH/34 Omit spaces between a header name and the colon in the error message that
2060 is given when verify = headers_syntax fails (if there are lots of them,
2061 the message gets confusing).
2063 PH/35 Change the default for dns_check_names_pattern to allow slashes within
2064 names, as there are now some PTR records that contain slashes. This check
2065 is only to protect against broken name servers that fall over on strange
2066 characters, so the fact that it applies to all lookups doesn't matter.
2068 PH/36 Now that the new test suite is complete, we can remove some of the
2069 special code in Exim that was needed for the old test suite. For example,
2070 sorting DNS records because real resolvers return them in an arbitrary
2071 order. The new test suite's fake resolver always returns records in the
2074 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
2075 bounces) except when queue_only is set, to avoid logging races between
2076 the different processes.
2078 PH/38 Panic-die if .include specifies a non-absolute path.
2080 PH/39 A tweak to the "H" retry rule from its user.
2082 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
2083 a label. They prevented compilation on older perls.
2085 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
2086 a warning to be raised on newish perls.
2088 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
2089 on queue. Changes to match documented behaviour of showing count of
2090 messages matching specified criteria.
2092 PH/40 Changed the default ident timeout from 30s to 5s.
2094 PH/41 Added support for the use of login_cap features, on those BSD systems
2095 that have them, for controlling the resources used by pipe deliveries.
2097 PH/42 The content-scanning code uses fopen() to create files in which to put
2098 message data. Previously it was not paying any attention to the mode of
2099 the files. Exim runs with umask(0) because the rest of the code creates
2100 files with open(), and sets the required mode explicitly. Thus, these
2101 files were ending up world-writeable. This was not a big issue, because,
2102 being within the spool directory, they were not world-accessible. I have
2103 created a function called modefopen, which takes an additional mode
2104 argument. It sets umask(777), creates the file, chmods it to the required
2105 mode, then resets the umask. All the relevant calls to fopen() in the
2106 content scanning code have been changed to use this function.
2108 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
2109 to 24 hours. This avoids potential overflow problems when processing G
2110 and H retry rules. I suspect nobody ever tinkers with this value.
2112 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
2114 PH/45 When the plaintext authenticator is running as a client, the server's
2115 challenges are checked to ensure they are valid base64 strings. By
2116 default, the authentication attempt is cancelled if an invalid string is
2117 received. Setting client_ignore_invalid_base64 true ignores these errors.
2118 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
2119 they are received. Thus, the responses can be made to depend on the
2120 challenges. If an invalid string is ignored, an empty string is placed in
2123 PH/46 Messages that are created by the autoreply transport now contains a
2124 References: header, in accordance with RFCs 2822 and 3834.
2126 PH/47 Added authenticated_sender_force to the smtp transport.
2128 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
2130 PH/49 Installed latest patch from the Sieve maintainer.
2132 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
2133 was also set, the appendfile transport was unnecessarily scanning a
2134 directory of message files (e.g. for maildir delivery) to find the count
2135 of files (along with the size), even though it did not need this
2136 information. It now does the scan only if it needs to find either the
2137 size of the count of files.
2139 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
2141 PH/52 Two bugs concerned with error handling when the smtp transport is
2144 (i) Exim was not creating retry information for temporary errors given
2145 for individual recipients after the DATA command when the smtp transport
2146 was used in LMTP mode. This meant that they could be retried too
2147 frequently, and not timed out correctly.
2149 (ii) Exim was setting the flag that allows error details to be returned
2150 for LMTP errors on RCPT commands, but not for LMTP errors for individual
2151 recipients that were returned after the DATA command.
2153 PH/53 This is related to PH/52, but is more general: for any failing address,
2154 when detailed error information was permitted to be returned to the
2155 sender, but the error was temporary, then after the final timeout, only
2156 "retry timeout exceeded" was returned. Now it returns the full error as
2157 well as "retry timeout exceeded".
2159 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
2160 do this, and (what is worse) MTAs that accept it.
2162 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
2163 will now be deprecated.
2165 PH/56 New os.c-cygwin from the Cygwin maintainer.
2167 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
2168 formats (previously only available in exim formats via -bpr, -bpru,
2169 and -bpra. Now also available in native and exiqgrep formats)
2171 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
2172 with very large, slow to parse queues
2174 JJ/08 exipick: added ! as generic prefix to negate any criteria format
2176 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
2178 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
2179 responses to authentication challenges, though it was showing the
2180 challenges; (ii) I've removed the CR characters from the debug output for
2183 PH/58 Allow for the insertion of a newline as well as a space when a string
2184 is turned into more than one encoded-word during RFC 2047 encoding. The
2185 Sieve code now uses this.
2187 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
2188 data_4xx, lost_connection, tls_required.
2190 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
2191 message was being sent as an SMTP response.
2193 PH/61 Add -l and -k options to exicyclog.
2195 PH/62 When verifying, if an address was redirected to one new address, so that
2196 verification continued, and the new address failed or deferred after
2197 having set something in $address_data, the value of $address_data was not
2198 passed back to the ACL. This was different to the case when no
2199 redirection occurred. The value is now passed back in both cases.
2201 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
2202 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
2203 use login_cap.h, so on its own it isn't the distinguishing feature. The
2204 new name refers directly to the setclassresources() function.
2206 PH/65 Added configuration files for NetBSD3.
2208 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
2210 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
2211 is preferred over IPv4.
2213 PH/68 The bounce_return_message and bounce_return_body options were not being
2214 honoured for bounces generated during the reception of non-SMTP messages.
2215 In particular, this applied to messages rejected by the ACL. This bug has
2216 been fixed. However, if bounce_return_message is true and bounce_return_
2217 body is false, the headers that are returned for a non-SMTP message
2218 include only those that have been read before the error was detected.
2219 (In the case of an ACL rejection, they have all been read.)
2221 PH/69 The HTML version of the specification is now built in a directory called
2222 spec_html instead of spec.html, because the latter looks like a path with
2223 a MIME-type, and this confuses some software.
2225 PH/70 Catch two compiler warnings in sieve.c.
2227 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
2228 function verify_get_ident() calls ip_connect() to connect a socket, but
2229 if the "connect()" function timed out, ip_connect() used to close the
2230 socket. However, verify_get_ident() also closes the socket later, and in
2231 between Exim writes to the log, which may get opened at this point. When
2232 the socket was closed in ip_connect(), the log could get the same file
2233 descriptor number as the socket. This naturally causes chaos. The fix is
2234 not to close the socket in ip_connect(); the socket should be closed by
2235 the function that creates it. There was only one place in the code where
2236 this was missing, in the iplookup router, which I don't think anybody now
2237 uses, but I've fixed it anyway.
2239 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
2240 well as to direct DNS lookups. Otherwise the handling of names in host
2241 lists is inconsistent and therefore confusing.
2247 PH/01 Two changes to the default runtime configuration:
2249 (1) Move the checks for relay_from_hosts and authenticated clients from
2250 after to before the (commented out) DNS black list checks.
2252 (2) Add control=submission to the relay_from_hosts and authenticated
2253 clients checks, on the grounds that messages accepted by these
2254 statements are most likely to be submissions.
2256 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
2258 (1) Generate an error if the third argument for the ${prvs expansion is
2261 (2) Treat a missing third argument of ${prvscheck as if it were an empty
2264 (3) Reset the variables that are obtained from the first argument of
2265 ${prvscheck and used in the second argument before leaving the code,
2266 because their memory is reclaimed, so using them afterwards may do
2269 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
2270 one (it's much easier than Tom thought :-).
2272 (5) Because of (4), we can now allow for the use of $prvscheck_result
2273 inside the third argument.
2275 PH/03 For some reason, the default setting of PATH when running a command from
2276 a pipe transport was just "/usr/bin". I have changed it to
2279 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
2280 anything to be listed in the output from -bV.
2282 PH/05 When a filter generated an autoreply, the entire To: header line was
2283 quoted in the delivery log line, like this:
2285 => >A.N.Other <ano@some.domain> <original@ddress> ...
2287 This has been changed so that it extracts the operative address. There
2288 may be more than one such address. If so, they are comma-separated, like
2291 => >ano@some.domain,ona@other.domain <original@ddress> ...
2293 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
2294 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
2295 not being looked up in the rDNS to get a host name, Exim was showing the
2296 IP address twice in Received: lines, even though the IP addresses were
2297 identical. For example:
2299 Received: from [1.2.3.4] (helo=[1.2.3.4])
2301 However, if the real host name was known, it was omitting the HELO data
2302 if it matched the actual IP address. This has been tidied up so that it
2303 doesn't show the same IP address twice.
2305 PH/07 When both +timestamp and +memory debugging was on, the value given by
2306 $tod_xxx expansions could be wrong, because the tod_stamp() function was
2307 called by the debug printing, thereby overwriting the timestamp buffer.
2308 Debugging no longer uses the tod_stamp() function when +timestamp is set.
2310 PH/08 When the original message was included in an autoreply transport, it
2311 always said "this is a copy of the message, including all the headers",
2312 even if body_only or headers_only was set. It now gives an appropriate
2315 PH/09 Applied a patch from the Sieve maintainer which:
2317 o fixes some comments
2318 o adds the (disabled) notify extension core
2319 o adds some debug output for the result of if/elsif tests
2320 o points to the current vacation draft in the documentation
2321 and documents the missing references header update
2325 o fixes a bug in processing the envelope test (when testing
2326 multiple envelope elements, the last element determinted the
2329 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
2330 Electronic Mail") by including:
2332 Auto-submitted: auto-generated
2334 in the messages that it generates (bounce messages and others, such as
2335 warnings). In the case of bounce messages for non-SMTP mesages, there was
2336 also a typo: it was using "Auto_submitted" (underscore instead of
2337 hyphen). Since every message generated by Exim is necessarily in response
2338 to another message, thes have all been changed to:
2340 Auto-Submitted: auto-replied
2342 in accordance with these statements in the RFC:
2344 The auto-replied keyword:
2346 - SHOULD be used on messages sent in direct response to another
2347 message by an automatic process,
2349 - MUST NOT be used on manually-generated messages,
2351 - MAY be used on Delivery Status Notifications (DSNs) and Message
2352 Disposition Notifications (MDNs),
2354 - MUST NOT be used on messages generated by automatic or periodic
2355 processes, except for messages which are automatic responses to
2358 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
2359 to the default Received: header definition.
2361 PH/12 Added log selector acl_warn_skipped (default on).
2363 PH/13 After a successful wildlsearch lookup, discard the values of numeric
2364 variables because (a) they are in the wrong storage pool and (b) even if
2365 they were copied, it wouldn't work properly because of the caching.
2367 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
2368 checking when decoding. Apparently there are clients that generate
2369 overlong encoded strings. Why am I not surprised?
2371 PH/15 If the first argument of "${if match_address" was not empty, but did not
2372 contain an "@" character, Exim crashed. Now it writes a panic log message
2373 and treats the condition as false.
2375 PH/16 In autoreply, treat an empty string for "once" the same as unset.
2377 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
2378 extension "envelope-auth". The code is finished and in agreement with
2379 other implementations, but there is no documentation so far and in fact,
2380 nobody wrote the draft yet. This extension is currently #undef'ed, thus
2381 not changing the active code.
2383 Print executed "if" and "elsif" statements when debugging is used. This
2384 helps a great deal to understand what a filter does.
2386 Document more things not specified clearly in RFC3028. I had all this
2387 sorted out, when out of a sudden new issues came to my mind. Oops."
2389 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
2392 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
2393 canonical form (as documented). However, after a host name lookup from
2394 the IP address, check_host() was doing a simple string comparison with
2395 addresses acquired from the DNS when checking that the found name did
2396 have the original IP as one of its addresses. Since any found IPv6
2397 addresses are likely to be in abbreviated form, the comparison could
2398 fail. Luckily, there already exists a function for doing the comparison
2399 by converting both addresses to binary, so now that is used instead of
2400 the text comparison.
2402 PH/20 There was another similar case to PH/19, when a complete host name was
2403 given in a host list; looking up its IP address could give an abbreviated
2404 form, whereas the current host's name might or might not be abbreviated.
2405 The same fix has been applied.
2411 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
2412 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
2415 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
2416 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
2418 PH/03 Typo: missing ".o" in src/pcre/Makefile.
2420 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
2421 header line, restrict the check to what is listed in RFCs 2369 and 2929.
2422 Also, for "Auto-Submitted", treat anything other than "no" as
2423 non-personal, in accordance with RFC 3834. (Previously it treated
2424 anything starting "auto-" as non-personal.)
2426 TF/01 The control=submission/name=... option had a problem with syntax
2427 errors if the name included a slash character. The /name= option
2428 now slurps the rest of the string, so it can include any characters
2429 but it must come last in the list of options (after /sender_retain
2432 PH/05 Some modifications to the interface to the fake nameserver for the new
2440 TK/01 Added the "success_on_redirect" address verification option. See
2441 NewStuff for rationale and an example.
2443 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
2445 PH/02 Patch to exigrep to allow it to work on syslog lines.
2447 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
2448 fread() to skip over the body file's header line, because in Cygwin the
2449 header line is locked and is inaccessible.
2451 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
2452 co-exist for some time) to make it clear that it is the Exim ID that is
2453 referenced, not the Message-ID: header line.
2455 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
2456 string_format() function, because snprintf() does not exist on all
2459 PH/06 The use of forbid_filter_existstest now also locks out the use of the
2460 ${stat: expansion item.
2462 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
2463 protocol synchronization error", to keep the pedants happy.
2465 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
2466 well as for IRIX systems, when gcc is being used. See the host.c source
2469 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
2471 PH/10 Named domain lists were not working if used in a queue_smtp_domains
2474 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
2475 transport and to the smtp transport in LMTP mode.
2477 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
2479 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
2480 run a filter in a subprocess. This could lead to confusion in subsequent
2481 lookups in the parent process. There should also be a search_tidyup() at
2482 the end of the subprocess.
2484 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
2485 only if the host matched helo_try_verify_hosts, which caused the
2486 verification to occur when the EHLO/HELO command was issued. The ACL just
2487 tested the remembered result. Now, if a previous verification attempt has
2488 not happened, "verify = helo" does it there and then.
2490 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
2492 TK/03 Fix log output including CR from clamd.
2494 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
2495 exist provoked a memory error which could cause a segfault.
2497 PH/15 Installed PCRE 6.2
2499 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
2501 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
2502 of the problem. Specifically, suggested +O2 rather than +O1 for the
2505 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
2507 PH/20 If a delivery was routed to a non-standard port by means of an SRV
2508 record, the port was not correctly logged when the outgoing_port log
2509 selector was set (it logged the transort's default port).
2511 PH/21 Added support for host-specific ports to manualroute, queryprogram,
2512 fallback_hosts, and "hosts" in the smtp transport.
2514 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
2515 host errors such as "Connection refused".
2517 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
2518 authentication with radiusclient 0.4.9:
2520 - Error returned from rc_read_config was caught wrongly
2521 - Username/password not passed on to radius server due to wrong length.
2523 The presumption is that some radiusclient API changes for 4.51/PH/17
2524 were not taken care of correctly. The code is still untested by me (my
2525 Linux distribution still has 0.3.2 of radiusclient), but it was
2526 contributed by a Radius user.
2528 PH/24 When doing a callout, the value of $domain wasn't set correctly when
2529 expanding the "port" option of the smtp transport.
2531 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
2532 while reading a MIME header. Thanks to Tom Hughes for a patch.
2534 PH/24 Include config.h inside local_scan.h so that configuration settings are
2537 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
2538 that in an ACL for RCPT (for example), you can examine exactly what was
2541 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
2542 commands, but it was not correctly comparing the address with the actual
2543 client host address. Thus, it would show the EHLO address in Received:
2544 header lines when this was not necessary.
2546 PH/27 Added the % operator to ${eval:}.
2548 PH/28 Exim tries to create and chdir to its spool directory when it starts;
2549 it should be ignoring failures (because with -C, for example, it has lost
2550 privilege). It wasn't ignoring creation failures other than "already
2553 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
2556 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
2557 because an input line was too long, either on its own, or by virtue of
2558 too many continuations, the temporary file was not being removed, and the
2559 return code was incorrect.
2561 PH/31 Missing "BOOL" in function definition in filtertest.c.
2563 PH/32 Applied Sieve patches from the maintainer.
2565 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
2567 PH/33 Added "verify = not_blind".
2569 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
2570 Local/Makefile (with some defaults set). These are used in built scripts
2571 such as exicyclog, but they have never been used in the exim_install
2572 script (though there are many overriding facilities there). I have
2573 arranged that the exim_install script now takes note of these two
2576 PH/35 Installed configuration files for Dragonfly.
2578 PH/36 When a locally submitted message by a trusted user did not contain a
2579 From: header, and the sender address was obtained from -f or from an SMTP
2580 MAIL command, and the trusted user did not use -F to supply a sender
2581 name, $originator_name was incorrectly used when constructing a From:
2582 header. Furthermore, $originator_name was used for submission mode
2583 messages from external hosts without From: headers in a similar way,
2584 which is clearly wrong.
2586 PH/37 Added control=suppress_local_fixups.
2588 PH/38 When log_selector = +received_sender was set, and the addition of the
2589 sender made the log line's construction buffer exactly full, or one byte
2590 less than full, an overflow happened when the terminating "\n" was
2593 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
2594 when the result of a list match is failure because a DNS lookup failed.
2596 PH/40 RM_COMMAND is now used in the building process.
2598 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
2599 the "build-* directories that it finds.
2601 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
2602 address in a domain literal was a prefix of an interface address.
2604 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
2605 when verifying a sender address, unless rewrite_headers is false.
2607 PH/44 Wrote a long comment about why errors_to addresses are verified as
2608 recipients, not senders.
2610 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
2611 the ratelimit ACL was added.
2613 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
2615 PH/46 Added extra information about PostgreSQL errors to the error string.
2617 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
2618 avoiding the need to install special zones in a real server. This is
2619 backwards compatible; if it can't find the fake resolver, it drops back.
2620 Thus, both old and new test suites can be run.
2622 TF/02 Added util/ratelimit.pl
2624 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
2625 clock is set back in time.
2627 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
2628 Candler <B.Candler@pobox.com>.
2630 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
2631 OK for addresses that are the result of redirections.
2633 PH/48 A number of further additions for the benefit of the new test suite,
2634 including a fake gethostbyname() that interfaces to the fake DNS resolver
2637 TF/06 The fix for widen_domains has also been applied to qualify_single and
2638 search_parents which are the other dnslookup options that can cause
2641 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
2644 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
2646 TF/07 Exim produced the error message "an SRV record indicated no SMTP
2647 service" if it encountered an MX record with an empty target hostname.
2648 The message is now "an MX or SRV record indicated no SMTP service".
2650 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
2651 if the DNS of the sending site is misconfigured. This is quite a
2652 common situation. This change restores the behaviour of treating a
2653 helo verification defer as a failure.
2655 PH/51 If self=fail was set on a router, the bounce message did not include the
2656 actual error message.
2662 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
2664 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
2665 command itself ended in error, the underlying message about the transport
2666 filter timeout was being overwritten with the pipe command error. Now the
2667 underlying error message should be appended to the second error message.
2669 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
2671 PH/02 Reduce the amount of output that "make" produces by default. Full output
2672 can still be requested.
2674 PH/03 The warning log line about a condition test deferring for a "warn" verb
2675 was being output only once per connection, rather than after each
2676 occurrence (because it was using the same function as for successful
2677 "warn" verbs). This seems wrong, so I have changed it.
2679 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
2680 it should not have, which might have caused a crash in the right
2681 circumstances, but probably never did.
2683 PH/04 Installed a modified version of Tony Finch's patch to make submission
2684 mode fix the return path as well as the Sender: header line, and to
2685 add a /name= option so that you can make the user's friendly name appear
2688 TF/03 Added the control = fakedefer ACL modifier.
2690 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
2691 Mark Lowes for thorough testing.
2693 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
2695 TK/03 Merged latest SRS patch from Miles Wilton.
2697 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
2698 with the definition in sysexits.h (which is #included earlier).
2699 Fortunately, Exim does not actually use EX_OK. The code used to try to
2700 preserve the sysexits.h value, by assumimg that macro definitions were
2701 scanned for macro replacements. I have been disabused of this notion,
2702 so now the code just undefines EX_OK before #including unistd.h.
2704 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
2705 in the smtp transport. When a block could not be written in a single
2706 write() function, the timeout was being re-applied to each part-write.
2707 This seems wrong - if the receiver was accepting one byte at a time it
2708 would take for ever. The timeout is now adjusted when this happens. It
2709 doesn't have to be particularly precise.
2711 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
2712 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
2714 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
2715 without a domain if the check to <postmaster@domain> fails.
2717 SC/01 Eximstats: added -xls and the ability to specify output files
2718 (patch written by Frank Heydlauf).
2720 SC/02 Eximstats: use FileHandles for outputing results.
2722 SC/03 Eximstats: allow any combination of xls, txt, and html output.
2724 SC/04 Eximstats: fixed display of large numbers with -nvr option
2726 SC/05 Eximstats: fixed merging of reports with empty tables.
2728 SC/06 Eximstats: added the -include_original_destination flag
2730 SC/07 Eximstats: removed tabs and trailing whitespace.
2732 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
2734 TK/06 MBOX spool code: Add real "From " MBOX separator line
2735 so the .eml file is really in mbox format (even though
2736 most programs do not really care). Patch from Alex Miller.
2738 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
2739 The latter is generated from $received_to and is only set if the
2740 message has one envelope recipient. SA can use these headers,
2741 obviously out-of-the-box. Patch from Alex Miller.
2743 PH/08 The ${def test on a variable was returning false if the variable's
2744 value was "0", contrary to what the specification has always said!
2745 The result should be true unless the variable is empty.
2747 PH/09 The syntax error of a character other than { following "${if
2748 def:variable_name" (after optional whitespace) was not being diagnosed.
2749 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
2750 accidental colon was present, for example, could give incorrect results.
2752 PH/10 Tidied the code in a number of places where the st_size field of a stat()
2753 result is used (not including appendfile, where other changes are about
2756 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
2757 This involved changing a lot of size variables from int to off_t. It
2758 should work with maildirs and everything.
2760 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
2761 spamd dying while we are connected to it.
2763 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
2766 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
2767 function for BDB 4.3.
2769 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
2771 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
2772 the sources for PCRE have been re-arranged and more files are now
2775 PH/15 The code I had for printing potentially long long variables in PH/11
2776 above was not the best (it lost precision). The length of off_t variables
2777 is now inspected at build time, and an appropriate printing format (%ld
2778 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
2779 to be "long long int" or "long int". This is needed for the internal
2780 formatting function string_vformat().
2782 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
2783 the configuration file to be ":syslog", then the script "guesses" where
2784 the logs files are, rather than using the compiled in default. In our
2785 case the guess is not the same as the compiled default, so the script
2786 suddenly stopped working when I started to use syslog. The patch checks
2787 to see if log_file_path is "". If so, it attempts to read it from exim
2788 with no configuration file to get the compiled in version, before it
2789 falls back to the previous guessing code."
2791 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
2792 implementing BATV in an Exim configuration. See NewStuff for the gory
2795 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
2796 Makefile that are specific to HP-UX.
2798 PH/18 If the "use_postmaster" option was set for a recipient callout together
2799 with the "random" option, the postmaster address was used as the MAIL
2800 FROM address for the random test, but not for the subsequent recipient
2801 test. It is now used for both.
2803 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
2804 patch removes a few documentation additions to RFC 3028, because the
2805 latest draft now contains them. It adds the new en;ascii-case comparator
2806 and a new error check for 8bit text in MIME parts. Comparator and
2807 require names are now matched exactly. I enabled the subaddress
2808 extension, but it is not well tested yet (read: it works for me)."
2810 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
2811 rework some of the code of TK/09 above to avoid the hardwired use of
2812 "%lld" and "long long". Replaced the call to snprintf() with a call to
2815 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
2816 records point to non-existent hosts", "retry timeout exceeded", and
2817 "retry time not reached for any host after a long failure period".
2819 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
2820 experimental DomainKeys support:
2822 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
2823 (2) On an error such as an illegally used "control", the wrong name for
2824 the control was given.
2826 These problems did NOT occur unless DomainKeys support was compiled.
2828 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
2830 PH/24 Added ${if match_ip condition.
2832 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
2833 databases so that it will be absolutely obvious if a crash occurs in the
2834 DB library. This is a regular occurrence (often caused by mis-matched
2837 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
2838 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
2839 picked up on a user's system that detects such things. There doesn't seem
2840 to be a gcc warning option for this - only an attribute that has to be
2841 put on the function's prototype. It seems that in Fedora Core 4 they have
2842 set this on a number of new functions. No doubt there will be more in due
2845 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
2846 specify a transport. However, if an address that was verified by such a
2847 router was the subject of a callout, Exim crashed because it tried to
2848 read the rcpt_include_affixes from the non-existent transport. Now it
2849 just assumes that the setting of that option is false. This bug was
2850 introduced by 4.51/PH/31.
2852 PH/28 Changed -d+all to exclude +memory, because that information is very
2853 rarely of interest, but it makes the output a lot bigger. People tend to
2854 do -d+all out of habit.
2856 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
2857 code in os-type was giving problems when libc.so lives in lib64, like on
2860 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
2861 aren't the modern standard, and it seems that some systems' include files
2862 don't always have them. Exim was already checking for some of the newer
2863 ones like T_AAAA, and defining it itself. I've added checks for all the
2864 record types that Exim uses.
2866 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
2867 not automatically generating a new one, as it is supposed to. This
2868 prevented TLS from working. If the file did exist, but contained invalid
2869 data, a new version was generated, as expected. It was only the case of a
2870 non-existent file that was broken.
2872 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
2873 with a change in libdomainkeys > 0.64.
2875 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
2876 from DNS. If the selector record carries the flag, it now has
2877 precedence over the domain-wide flag.
2879 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
2881 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
2882 the use of a port name that isn't defined in /etc/services) occurred, the
2883 message was deferred as in a normal delivery, and thus remained on the
2884 spool, instead of being failed because of the mua_wrapper setting. This
2885 is now fixed, and I tidied up some of the mua_wrapper messages at the
2888 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
2889 the messages in a hash of arrays rather than using individual hashes.
2890 This is a bit cleaner and results in dramatic memory savings, albeit
2891 at a slight CPU cost.
2893 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
2894 as requested by Marc Sherman.
2896 SC/10 Eximstats: added histograms for user specified patterns as requested
2899 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
2901 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
2902 fopen() in the content-scanning modules that did not already have it.
2908 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
2909 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
2911 TK/02 Fix ACL "control" statement not being available in MIME ACL.
2913 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
2915 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
2916 to test Sieve filters that use "vacation".
2918 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
2919 that changes the way the GnuTLS parameters are stored in the cache file.
2920 The new format can be generated externally. For backward compatibility,
2921 if the data in the cache doesn't make sense, Exim assumes it has read an
2922 old-format file, and it generates new data and writes a new file. This
2923 means that you can't go back to an older release without removing the
2926 PH/03 A redirect router that has both "unseen" and "one_time" set does not
2927 work if there are any delivery delays because "one_time" forces the
2928 parent to be marked "delivered", so its unseen clone is never tried
2929 again. For this reason, Exim now forbids the simultaneous setting of
2932 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
2933 redirected to themselves ("homonym" addresses). Read the long ChangeLog
2934 entry if you want to know the details. The fix, however, neglected to
2935 consider the case when local delivery batching is involved. The test for
2936 "previously delivered" was not happening when checking to see if an
2937 address could be batched with a previous (undelivered) one; under
2938 certain circumstances this could lead to multiple deliveries to the same
2941 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
2942 in its include files, and this causes problems building Exim.
2944 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
2945 header_syntax) but Exim was just ignoring anything given after a slash.
2946 In particular, this caused confusion with an attempt to use "verify =
2947 reverse_host_lookup/defer_ok". An error is now given when options are
2948 supplied for verify items that do not have them. (Maybe reverse_host_
2949 lookup should have a defer_ok option, but that's a different point.)
2951 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
2952 defined by RFC 821) to 2048, because there were problems with some AUTH
2953 commands, and RFC 1869 says the size should be increased for extended
2954 SMTP commands that take arguments.
2956 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
2959 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
2960 "unknown" error; now it says that the functionality isn't in the binary.
2962 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
2963 an address' error message when a string expansion fails (syntax or
2964 whatever). Otherwise the password may appear in the log. Following change
2965 PH/42 below, there is no longer a chance of it appearing in a bounce
2968 PH/11 Installed exipick version 20050225.0 from John Jetmore.
2970 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
2971 of its addresses was ever tried. (Bugzilla bug #2.)
2973 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
2974 the result incorrectly in the debug output. (It correctly added a newline
2975 to what was transported.)
2977 TF/01 Added $received_time.
2979 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
2980 commented out examples of how to interface to a virus scanner and to
2981 SpamAssassin. Also added commented examples of av_scanner and
2982 spamd_address settings.
2984 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
2985 and controls are allowed in which ACLs. There were a couple of minor
2986 errors. Some of the entries in the conditions table (which is a table of
2987 where they are NOT allowed) were getting very unwieldy; rewrote them as a
2988 negation of where the condition IS allowed.
2990 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
2992 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
2993 header file does not have a version number, so I've had to invent a new
2994 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
2995 API. The code is untested by me (my Linux distribution still has 0.3.2 of
2996 radiusclient), but it was contributed by a Radius user.
2998 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
2999 files or directories, for OpenSSL.
3001 PH/19 When an Exim process that is running as root has to create an Exim log
3002 file, it does so in a subprocess that runs as exim:exim so as to get the
3003 ownership right at creation (otherwise, other Exim processes might see
3004 the file with the wrong ownership). There was no test for failure of this
3005 fork() call, which would lead to the process getting stuck as it waited
3006 for a non-existent subprocess. Forks do occasionally fail when resources
3007 run out. I reviewed all the other calls to fork(); they all seem to check
3010 PH/20 When checking for unexpected SMTP input at connect time (before writing
3011 the banner), Exim was not dealing correctly with a non-positive return
3012 from the read() function. If the client had disconnected by this time,
3013 the result was a log entry for a synchronization error with an empty
3014 string after "input=" when read() returned zero. If read() returned -1
3015 (an event I could not check), uninitialized data bytes were printed.
3016 There were reports of junk text (parts of files, etc) appearing after
3019 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
3021 PH/22 Added support for macro redefinition, and (re)definition in between
3022 driver and ACL definitions.
3024 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
3025 forgetting to use the resulting value; it was using the unexpanded value.
3027 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
3028 hadn't been configured. The fix is from Juergen Kreileder, who
3029 understands it better than I do:
3031 "Here's what I see happening with three configured cyrus_sasl
3032 authenticators configured (plain, login, cram-md5):
3034 On startup auth_cyrus_sasl_init() gets called for each of these.
3035 This means three calls to sasl_listmech() without a specified mech_list.
3036 => SASL tests which mechs of all available mechs actually work
3037 => three warnings about OTP not working
3038 => the returned list contains: plain, login, cram-md5, digest-md5, ...
3040 With the patch, sasl_listmech() also gets called three times. But now
3041 SASL's mech_list option is set to the server_mech specified in the the
3042 authenticator. Or in other words, the answer from sasl_listmech()
3043 gets limited to just the mech you're testing for (which is different
3045 => the return list contains just 'plain' or 'login', 'cram-md5' or
3046 nothing depending on the value of ob->server_mech.
3048 I've just tested the patch: Authentication still works fine,
3049 unavailable mechs specified in the exim configuration are still
3050 caught, and the auth.log warnings about OTP are gone."
3052 PH/25 When debugging is enabled, the contents of the command line are added
3053 to the debugging output, even when log_selector=+arguments is not
3056 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
3057 answer is "GNU", and only if the return is "GNU/something" is the answer
3060 PH/27 $acl_verify_message is now set immediately after the failure of a
3061 verification in an ACL, and so is available in subsequent modifiers. In
3062 particular, the message can be preserved by coding like this:
3064 warn !verify = sender
3065 set acl_m0 = $acl_verify_message
3067 Previously, $acl_verify_message was set only while expanding "message"
3068 and "log_message" when a very denied access.
3070 PH/28 Modified OS/os.c-Linux with
3072 -#ifndef OS_LOAD_AVERAGE
3073 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
3075 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
3076 nomenclature these days.)
3078 PH/29 Installed patch from the Sieve maintainer that adds the options
3079 sieve_useraddress and sieve_subaddress to the redirect router.
3081 PH/30 In these circumstances:
3082 . Two addresses routed to the same list of hosts;
3083 . First host does not offer TLS;
3084 . First host accepts first address;
3085 . First host gives temporary error to second address;
3086 . Second host offers TLS and a TLS session is established;
3087 . Second host accepts second address.
3088 Exim incorrectly logged both deliveries with the TLS parameters (cipher
3089 and peerdn, if requested) that were in fact used only for the second
3092 PH/31 When doing a callout as part of verifying an address, Exim was not paying
3093 attention to any local part prefix or suffix that was matched by the
3094 router that accepted the address. It now behaves in the same way as it
3095 does for delivery: the affixes are removed from the local part unless
3096 rcpt_include_affixes is set on the transport.
3098 PH/32 Add the sender address, as F=<...>, to the log line when logging a
3099 timeout during the DATA phase of an incoming message.
3101 PH/33 Sieve envelope tests were broken for match types other than :is. I have
3102 applied a patch sanctioned by the Sieve maintainer.
3104 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
3105 the uid or gid is negative. A case of a negative gid caused this to be
3106 noticed. The fix allows for either to be negative.
3108 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
3109 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
3112 PH/36 The change PH/12 above was broken. Fixed it.
3114 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
3115 the grounds that routing the same address twice would always produce the
3116 same answer. This might have been true once, but it is certainly no
3117 longer true now. Routing a child address may depend on the previous
3118 routing that produced that child. Some complicated redirection strategies
3119 went wrong when messages had multiple recipients, and made Exim's
3120 behaviour dependent on the order in which the addresses were given.
3122 I have moved the duplicate checking until after the routing is complete.
3123 Exim scans the addresses that are assigned to local and remote
3124 transports, and removes any duplicates. This means that more work will be
3125 done, as duplicates will always all be routed, but duplicates are
3126 presumably rare, so I don't expect this is of any significance.
3128 For deliveries to pipes, files, and autoreplies, the duplicate checking
3129 still happens during the routing process, since they are not going to be
3132 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
3133 It corrects a timeout issue with spamd. This is Ian's comment: "The
3134 background is that sometimes spamd either never reads data from a
3135 connection it has accepted, or it never writes response data. The exiscan
3136 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
3137 blindly assumes that writes won't block so it may never time out."
3139 PH/39 Allow G after quota size as well as K and M.
3141 PH/40 The value set for $authenticated_id in an authenticator may not contain
3142 binary zeroes or newlines because the value is written to log lines and
3143 to spool files. There was no check on this. Now the value is run through
3144 the string_printing() function so that such characters are converted to
3145 printable escape sequences.
3147 PH/41 $message_linecount is a new variable that contains the total number of
3148 lines in the message. Compare $body_linecount, which is the count for the
3151 PH/42 Exim no longer gives details of delivery errors for specific addresses in
3152 bounce and delay warning messages, except in certain special cases, which
3155 (a) An SMTP error message from a remote host;
3156 (b) A message specified in a :fail: redirection;
3157 (c) A message specified in a "fail" command in a system filter;
3158 (d) A message specified in a FAIL return from the queryprogram router;
3159 (e) A message specified by the cannot_route_message router option.
3161 In these cases only, Exim does include the error details in bounce and
3162 warning messages. There are also a few cases where bland messages such
3163 as "unrouteable address" or "local delivery error" are given.
3165 PH/43 $value is now also set for the "else" part of a ${run expansion.
3167 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
3168 being worked on, but at least Exim now implements the latest version to
3171 PH/45 In a pipe transport, although a timeout while waiting for the pipe
3172 process to complete was treated as a delivery failure, a timeout while
3173 writing the message to the pipe was logged, but erroneously treated as a
3174 successful delivery. Such timeouts include transport filter timeouts. For
3175 consistency with the overall process timeout, these timeouts are now
3176 treated as errors, giving rise to delivery failures by default. However,
3177 there is now a new Boolean option for the pipe transport called
3178 timeout_defer, which, if set TRUE, converts the failures into defers for
3179 both kinds of timeout. A transport filter timeout is now identified in
3182 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
3183 systems where "make" and "gmake" are different, calling "gmake" at top
3184 level broke things. I've arranged for the value of $(MAKE) to be passed
3185 from the Makefile to this script so that it can call the same version of
3189 A note about Exim versions 4.44 and 4.50
3190 ----------------------------------------
3192 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
3193 changes of various kinds. As a consequence, a big documentation update was
3194 needed. This delayed the release for rather longer than seemed good, especially
3195 in the light of a couple of (minor) security issues. Therefore, the changes
3196 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
3197 release. So 4.44 and 4.50 are in effect two different branches that both start
3200 I have left the 4.50 change log unchanged; it contains all the changes since
3201 4.43. The change log for 4.44 is below; many of its items are identical to
3202 those for 4.50. This seems to be the most sensible way to preserve the
3203 historical information.
3209 1. Minor wording change to the doc/README.SIEVE file.
3211 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
3212 computation of the current number of files was incorrect.
3214 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
3215 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
3218 4. Give more explanation in the error message when the command for a transport
3219 filter fails to execute.
3221 5. There are several places where Exim runs a non-Exim command in a
3222 subprocess. The SIGUSR1 signal should be disabled for these processes. This
3223 was being done only for the command run by the queryprogram router. It is
3224 now done for all such subprocesses. The other cases are: ${run, transport
3225 filters, and the commands run by the lmtp and pipe transports.
3227 6. Added CONFIGURE_GROUP build-time option.
3229 7. Some older OS have a limit of 256 on the maximum number of file
3230 descriptors. Exim was using setrlimit() to set 1000 as a large value
3231 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
3232 systems. I've change it so that if it can't get 1000, it tries for 256.
3234 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
3235 was an oversight, and furthermore, ever since the addition of extra
3236 controls (e.g. 4.43/32), the checks on when to allow different forms of
3237 "control" were broken. There should now be diagnostics for all cases when a
3238 control that does not make sense is encountered.
3240 9. Added the /retain_sender option to "control=submission".
3242 10. $recipients is now available in the predata ACL (oversight).
3244 11. Tidy the search cache before the fork to do a delivery from a message
3245 received from the command line. Otherwise the child will trigger a lookup
3246 failure and thereby defer the delivery if it tries to use (for example) a
3247 cached ldap connection that the parent has called unbind on.
3249 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
3250 of $address_data from the recipient verification was clobbered by the
3251 sender verification.
3253 13. The value of address_data from a sender verification is now available in
3254 $sender_address_data in subsequent conditions in the ACL statement.
3256 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
3258 15. Added a new option "connect=<time>" to callout options, to set a different
3261 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
3262 was its contents. (It was OK if the option was not defined at all.)
3264 17. A "Completed" log line is now written for messages that are removed from
3265 the spool by the -Mrm option.
3267 18. New variables $sender_verify_failure and $recipient_verify_failure contain
3268 information about exactly what failed.
3270 19. Added -dd to debug only the daemon process.
3272 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
3273 handles timeouts, both on the server side and network timeouts. Renamed the
3274 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
3276 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
3277 It is now set to "smtps".
3279 22. $host_address is now set to the target address during the checking of
3280 ignore_target_hosts.
3282 23. When checking ignore_target_hosts for an ipliteral router, no host name was
3283 being passed; this would have caused $sender_host_name to have been used if
3284 matching the list had actually called for a host name (not very likely,
3285 since this list is usually IP addresses). A host name is now passed as
3288 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
3289 code that specifies a non-restarting handler (typically sigaction() in
3290 modern systems) in an attempt to fix a rare and obscure crash bug.
3292 25. Narrowed the window for a race in the daemon that could cause it to ignore
3293 SIGCHLD signals. This is not a major problem, because they are used only to
3294 wake it up if nothing else does.
3296 26. A malformed maildirsize file could cause Exim to calculate negative values
3297 for the mailbox size or file count. Odd effects could occur as a result.
3298 The maildirsize information is now recalculated if the size or filecount
3301 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
3302 support for a long time. Removed HAVE_SYS_VFS_H.
3304 28. Installed the latest version of exipick from John Jetmore.
3306 29. In an address list, if the pattern was not a regular expression, an empty
3307 subject address (from a bounce message) matched only if the pattern was an
3308 empty string. Non-empty patterns were not even tested. This was the wrong
3309 because it is perfectly reasonable to use an empty address as part of a
3310 database query. An empty address is now tested by patterns that are
3311 lookups. However, all the other forms of pattern expect the subject to
3312 contain a local part and a domain, and therefore, for them, an empty
3313 address still always fails if the pattern is not itself empty.
3315 30. Exim went into a mad DNS loop when attempting to do a callout where the
3316 host was specified on an smtp transport, and looking it up yielded more
3317 than one IP address.
3319 31. Re-factored the code for checking spool and log partition space into a
3320 function that finds that data and another that does the check. The former
3321 is then used to implement four new variables: $spool_space, $log_space,
3322 $spool_inodes, and $log_inodes.
3324 32. The RFC2047 encoding function was originally intended for short strings
3325 such as real names; it was not keeping to the 75-character limit for
3326 encoded words that the RFC imposes. It now respects the limit, and
3327 generates multiple encoded words if necessary. To be on the safe side, I
3328 have increased the buffer size for the ${rfc2047: expansion operator from
3331 33. It is now permitted to omit both strings after an "if" condition; if the
3332 condition is true, the result is "true". As before, when the second string
3333 is omitted, a false condition yields an empty string. This makes it less
3334 cumbersome to write custom ACL and router conditions.
3336 34. Failure to deliver a bounce message always caused it to be frozen, even if
3337 there was an errors_to setting on the router. The errors_to setting is now
3340 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
3341 canonical form (fully expanded) before being placed in
3342 $sender_host_address.
3344 36. The table in the code that translates DNS record types into text (T_A to
3345 "A" for instance) was missing entries for NS and CNAME. It is just possible
3346 that this could have caused confusion if both these types were looked up
3347 for the same domain, because the text type is used as part of Exim's
3348 per-process caching. But the chance of anyone hitting this buglet seems
3351 37. The dnsdb lookup has been extended in a number of ways.
3353 (1) There is a new type, "zns", which walks up the domain tree until it
3354 finds some nameserver records. It should be used with care.
3356 (2) There is a new type, "mxh", which is like "mx" except that it returns
3357 just the host names, not the priorities.
3359 (3) It is now possible to give a list of domains (or IP addresses) to be
3360 looked up. The behaviour when one of the lookups defers can be
3361 controlled by a keyword.
3363 (4) It is now possible to specify the separator character for use when
3364 multiple records are returned.
3366 38. The dnslists ACL condition has been extended: it is now possible to supply
3367 a list of IP addresses and/or domains to be looked up in a particular DNS
3370 39. Added log_selector=+queue_time_overall.
3372 40. When running the queue in the test harness, wait just a tad after forking a
3373 delivery process, to get repeatability of debugging output.
3375 41. Include certificate and key file names in error message when GnuTLS fails
3376 to set them up, because the GnuTLS error message doesn't include the name
3377 of the failing file when there is a problem reading it.
3379 42. Allow both -bf and -bF in the same test run.
3381 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
3383 44. The "Exiscan patch" is now merged into the mainline Exim source.
3385 45. Sometimes the final signoff response after QUIT could fail to get
3386 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
3387 < 0 before doing a fflush(). This bug looks as though it goes back to the
3388 introduction of TLS in release 3.20, but "sometimes" must have been rare
3389 because the tests only now provoked it.
3391 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
3392 (this can affect the format of dates).
3394 47. exim_tidydb, when checking for the continued existence of a message for
3395 which it has found a message-specific retry record, was not finding
3396 messages that were in split spool directories. Consequently, it was
3397 deleting retry records that should have stayed in existence.
3399 48. Steve fixed some bugs in eximstats.
3401 49. The SPA authentication driver was not abandoning authentication and moving
3402 on to the next authenticator when an expansion was forced to fail,
3403 contradicting the general specification for all authenticators. Instead it
3404 was generating a temporary error. It now behaves as specified.
3406 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
3407 (the order specifies the preference for clients). The order is now AES256,
3408 AES128, 3DES, ARCFOUR128.
3410 51. Small patch to Sieve code - explicitly set From: when generating an
3413 52. Exim crashed if a remote delivery caused a very long error message to be
3414 recorded - for instance if somebody sent an entire SpamAssassin report back
3415 as a large number of 550 error lines. This bug was coincidentally fixed by
3416 increasing the size of one of Exim's internal buffers (big_buffer) that
3417 happened as part of the Exiscan merge. However, to be on the safe side, I
3418 have made the code more robust (and fixed the comments that describe what
3421 53. Now that there can be additional text after "Completed" in log lines (if
3422 the queue_time_overall log selector is set), a one-byte patch to exigrep
3423 was needed to allow it to recognize "Completed" as not the last thing in
3426 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
3427 patch that reportedly fixes this has been added. I am not expert enough to
3428 create a test for it. This is what the patch creator wrote:
3430 "I found a little strange behaviour of ldap code when working with
3431 Windows 2003 AD Domain, where users was placed in more than one
3432 Organization Units. When I tried to give exim partial DN, the exit code
3433 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
3434 But simultaneously result of request was absolutely normal ldap result,
3435 so I produce this patch..."
3437 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
3438 REFERENCE, so I have modified the code to exclude the patch when that macro
3441 55. Some experimental protocols are using DNS PTR records for new purposes. The
3442 keys for these records are domain names, not reversed IP addresses. The
3443 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
3444 leaves it alone. Component reversal etc. now happens only for IP addresses.
3447 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
3449 57. Double the size of the debug message buffer (to 2048) so that more of very
3450 long debug lines gets shown.
3452 58. The exicyclog utility now does better if the number of log files to keep
3453 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
3455 59. Two changes related to the smtp_active_hostname option:
3457 (1) $smtp_active_hostname is now available as a variable.
3458 (2) The default for smtp_banner uses $smtp_active_hostname instead
3459 of $primary_hostname.
3461 60. The host_aton() function is supposed to be passed a string that is known
3462 to be a valid IP address. However, in the case of IPv6 addresses, it was
3463 not checking this. This is a hostage to fortune. Exim now panics and dies
3464 if the condition is not met. A case was found where this could be provoked
3465 from a dnsdb PTR lookup with an IPv6 address that had more than 8
3466 components; fortuitously, this particular loophole had already been fixed
3467 by change 4.50/55 above.
3469 If there are any other similar loopholes, the new check in host_aton()
3470 itself should stop them being exploited. The report I received stated that
3471 data on the command line could provoke the exploit when Exim was running as
3472 exim, but did not say which command line option was involved. All I could
3473 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
3474 running as the user.
3477 61. There was a buffer overflow vulnerability in the SPA authentication code
3478 (which came originally from the Samba project). I have added a test to the
3479 spa_base64_to_bits() function which I hope fixes it.
3482 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
3483 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
3485 63. The daemon start-up calls getloadavg() while still root for those OS that
3486 need the first call to be done as root, but it missed one case: when
3487 deliver_queue_load_max is set with deliver_drop_privilege. This is
3488 necessary for the benefit of the queue runner, because there is no re-exec
3489 when deliver_drop_privilege is set.
3491 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
3492 This has been fixed.
3494 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
3495 was in use, was not putting the data itself into the right store pool;
3496 consequently, it could be overwritten for a subsequent message in the same
3497 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
3500 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
3502 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
3503 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
3504 the function were treating the return as a boolean value, which happened to
3505 work because 0=false and not-0=true, but is not correct code.
3507 68. The host_aton() function was not handling scoped IPv6 addresses (those
3508 with, for example, "%eth0" on the end) correctly.
3510 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
3511 negated items (that is, ~something) in unsigned ints. Some compilers
3512 apparently mutter when there is no cast.
3514 70. If an address verification called from an ACL failed, and did not produce a
3515 user-specific message (i.e. there was only a "system" message), nothing was
3516 put in $acl_verify_message. In this situation, it now puts the system
3519 71. Change 4.23/11 added synchronization checking at the start of an SMTP
3520 session; change 4.31/43 added the unwanted input to the log line - except
3521 that it did not do this in the start of session case. It now does.
3523 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
3524 This is wrong and can cause the other end to generate a synchronization
3525 error if it is another Exim or anything else that does the synchronization
3526 check. A QUIT command is no longer sent after a timeout.
3528 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
3529 during host lookups.
3531 74. The defer_ok option of callout verification was not working if it was used
3532 when verifying addresses in header lines, that is, for this case:
3534 verify = header_sender/callout=defer_ok
3536 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
3537 those file descriptors could be used for SMTP connections. If anything
3538 wrote to stderr (the example that came up was "warn" in embedded Perl), it
3539 could be sent to the SMTP client, causing chaos. The daemon now opens
3540 stdin, stdout, and stderr to /dev/null when it puts itself into the
3543 76. Arrange for output from Perl's "warn" command to be written to Exim's main
3544 log by default. The user can override this with suitable Perl magic.
3546 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
3547 the log message when discard triggers, was not working for the DATA ACL or
3548 for the non-SMTP ACL.
3550 78. Error message wording change in sieve.c.
3552 79. If smtp_accept_max_per_host was set, the number of connections could be
3553 restricted to fewer than expected, because the daemon was trying to set up
3554 a new connection before checking whether the processes handling previous
3555 connections had finished. The check for completed processes is now done
3556 earlier. On busy systems, this bug wouldn't be noticed because something
3557 else would have woken the daemon, and it would have reaped the completed
3560 80. If a message was submitted locally by a user whose login name contained one
3561 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
3562 It caused a spool format error. I have fixed the spool reading code. A
3563 related problem was that the "from" clause in the Received: line became
3564 illegal because of the space(s). It is now covered by ${quote_local_part.
3566 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
3569 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
3571 83. Patch from Sieve maintainer to fix unterminated string problem in
3572 "vacation" handling.
3574 84. Some minor changes to the Linux configuration files to help with other
3575 OS variants using glibc.
3577 85. One more patch for Sieve to update vacation handling to latest spec.
3580 ----------------------------------------------------
3581 See the note above about the 4.44 and 4.50 releases.
3582 ----------------------------------------------------
3588 1. Change 4.43/35 introduced a bug that caused file counts to be
3589 incorrectly computed when quota_filecount was set in an appendfile
3592 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
3593 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
3596 3. Give more explanation in the error message when the command for a transport
3597 filter fails to execute.
3599 4. There are several places where Exim runs a non-Exim command in a
3600 subprocess. The SIGUSR1 signal should be disabled for these processes. This
3601 was being done only for the command run by the queryprogram router. It is
3602 now done for all such subprocesses. The other cases are: ${run, transport
3603 filters, and the commands run by the lmtp and pipe transports.
3605 5. Some older OS have a limit of 256 on the maximum number of file
3606 descriptors. Exim was using setrlimit() to set 1000 as a large value
3607 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
3608 systems. I've change it so that if it can't get 1000, it tries for 256.
3610 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
3611 was an oversight, and furthermore, ever since the addition of extra
3612 controls (e.g. 4.43/32), the checks on when to allow different forms of
3613 "control" were broken. There should now be diagnostics for all cases when a
3614 control that does not make sense is encountered.
3616 7. $recipients is now available in the predata ACL (oversight).
3618 8. Tidy the search cache before the fork to do a delivery from a message
3619 received from the command line. Otherwise the child will trigger a lookup
3620 failure and thereby defer the delivery if it tries to use (for example) a
3621 cached ldap connection that the parent has called unbind on.
3623 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
3624 of $address_data from the recipient verification was clobbered by the
3625 sender verification.
3627 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
3628 was its contents. (It was OK if the option was not defined at all.)
3630 11. A "Completed" log line is now written for messages that are removed from
3631 the spool by the -Mrm option.
3633 12. $host_address is now set to the target address during the checking of
3634 ignore_target_hosts.
3636 13. When checking ignore_target_hosts for an ipliteral router, no host name was
3637 being passed; this would have caused $sender_host_name to have been used if
3638 matching the list had actually called for a host name (not very likely,
3639 since this list is usually IP addresses). A host name is now passed as
3642 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
3643 code that specifies a non-restarting handler (typically sigaction() in
3644 modern systems) in an attempt to fix a rare and obscure crash bug.
3646 15. Narrowed the window for a race in the daemon that could cause it to ignore
3647 SIGCHLD signals. This is not a major problem, because they are used only to
3648 wake it up if nothing else does.
3650 16. A malformed maildirsize file could cause Exim to calculate negative values
3651 for the mailbox size or file count. Odd effects could occur as a result.
3652 The maildirsize information is now recalculated if the size or filecount
3655 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
3656 support for a long time. Removed HAVE_SYS_VFS_H.
3658 18. Updated exipick to current release from John Jetmore.
3660 19. Allow an empty sender to be matched against a lookup in an address list.
3661 Previously the only cases considered were a regular expression, or an
3664 20. Exim went into a mad DNS lookup loop when doing a callout where the
3665 host was specified on the transport, if the DNS lookup yielded more than
3668 21. The RFC2047 encoding function was originally intended for short strings
3669 such as real names; it was not keeping to the 75-character limit for
3670 encoded words that the RFC imposes. It now respects the limit, and
3671 generates multiple encoded words if necessary. To be on the safe side, I
3672 have increased the buffer size for the ${rfc2047: expansion operator from
3675 22. Failure to deliver a bounce message always caused it to be frozen, even if
3676 there was an errors_to setting on the router. The errors_to setting is now
3679 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
3680 canonical form (fully expanded) before being placed in
3681 $sender_host_address.
3683 24. Updated eximstats to version 1.33
3685 25. Include certificate and key file names in error message when GnuTLS fails
3686 to set them up, because the GnuTLS error message doesn't include the name
3687 of the failing file when there is a problem reading it.
3689 26. Expand error message when OpenSSL has problems setting up cert/key files.
3692 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
3693 (this can affect the format of dates).
3695 28. exim_tidydb, when checking for the continued existence of a message for
3696 which it has found a message-specific retry record, was not finding
3697 messages that were in split spool directories. Consequently, it was
3698 deleting retry records that should have stayed in existence.
3700 29. eximstats updated to version 1.35
3701 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
3702 1.35 - bugfix such that pie charts by volume are generated correctly
3704 30. The SPA authentication driver was not abandoning authentication and moving
3705 on to the next authenticator when an expansion was forced to fail,
3706 contradicting the general specification for all authenticators. Instead it
3707 was generating a temporary error. It now behaves as specified.
3709 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
3710 (the order specifies the preference for clients). The order is now AES256,
3711 AES128, 3DES, ARCFOUR128.
3713 31. Small patch to Sieve code - explicitly set From: when generating an
3716 32. Exim crashed if a remote delivery caused a very long error message to be
3717 recorded - for instance if somebody sent an entire SpamAssassin report back
3718 as a large number of 550 error lines. This bug was coincidentally fixed by
3719 increasing the size of one of Exim's internal buffers (big_buffer) that
3720 happened as part of the Exiscan merge. However, to be on the safe side, I
3721 have made the code more robust (and fixed the comments that describe what
3724 33. Some experimental protocols are using DNS PTR records for new purposes. The
3725 keys for these records are domain names, not reversed IP addresses. The
3726 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
3727 leaves it alone. Component reversal etc. now happens only for IP addresses.
3730 34. The host_aton() function is supposed to be passed a string that is known
3731 to be a valid IP address. However, in the case of IPv6 addresses, it was
3732 not checking this. This is a hostage to fortune. Exim now panics and dies
3733 if the condition is not met. A case was found where this could be provoked
3734 from a dnsdb PTR lookup with an IPv6 address that had more than 8
3735 components; fortuitously, this particular loophole had already been fixed
3736 by change 4.50/55 or 4.44/33 above.
3738 If there are any other similar loopholes, the new check in host_aton()
3739 itself should stop them being exploited. The report I received stated that
3740 data on the command line could provoke the exploit when Exim was running as
3741 exim, but did not say which command line option was involved. All I could
3742 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
3743 running as the user.
3746 35. There was a buffer overflow vulnerability in the SPA authentication code
3747 (which came originally from the Samba project). I have added a test to the
3748 spa_base64_to_bits() function which I hope fixes it.
3751 36. The daemon start-up calls getloadavg() while still root for those OS that
3752 need the first call to be done as root, but it missed one case: when
3753 deliver_queue_load_max is set with deliver_drop_privilege. This is
3754 necessary for the benefit of the queue runner, because there is no re-exec
3755 when deliver_drop_privilege is set.
3757 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
3758 was in use, was not putting the data itself into the right store pool;
3759 consequently, it could be overwritten for a subsequent message in the same
3760 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
3763 38. Sometimes the final signoff response after QUIT could fail to get
3764 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
3765 < 0 before doing a fflush(). This bug looks as though it goes back to the
3766 introduction of TLS in release 3.20, but "sometimes" must have been rare
3767 because the tests only now provoked it.
3773 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
3774 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
3775 Now it returns an integer. A number of calls were still expecting a T/F
3776 return. Fortuitously, in all cases, the tests worked in OK situations,
3777 which is the norm. However, things would have gone wrong on any write
3778 failures on the smtp file descriptor. This function is used when sending
3779 messages over SMTP and also when doing verify callouts.
3781 2. When Exim is called to do synchronous delivery of a locally submitted
3782 message (the -odf or -odi options), it no longer closes stderr before doing
3785 3. Implemented the mua_wrapper option.
3787 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
3789 5. Implemented the functions header_remove(), header_testname(),
3790 header_add_at_position(), and receive_remove_recipient(), and exported them
3793 6. If an ACL "warn" statement specified the addition of headers, Exim already
3794 inserted X-ACL-Warn: at the start if there was no header name. However, it
3795 was not making this test for the second and subsequent header lines if
3796 there were newlines in the string. This meant that an invalid header could
3797 be inserted if Exim was badly configured.
3799 7. Allow an ACL "warn" statement to add header lines at the start or after all
3800 the Received: headers, as well as at the end.
3802 8. Added the rcpt_4xx retry error code.
3804 9. Added postmaster_mailfrom=xxx to callout verification option.
3806 10. Added mailfrom=xxxx to the callout verification option, for verify=
3809 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
3810 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
3812 12. Inserted some casts to stop certain compilers warning when using pointer
3813 differences as field lengths or precisions in printf-type calls (mostly
3814 affecting debugging statements).
3816 13. Added optional readline() support for -be (dynamically loaded).
3818 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
3819 same clock tick as a message's arrival, so that its received time was the
3820 same as the "first fail" time on the retry record, and that message
3821 remained on the queue past the ultimate address timeout, every queue runner
3822 would try a delivery (because it was past the ultimate address timeout) but
3823 after another failure, the ultimate address timeout, which should have then
3824 bounced the address, did not kick in. This was a "< instead of <=" error;
3825 in most cases the first failure would have been in the next clock tick
3826 after the received time, and all would be well.
3828 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
3829 being recognized when the domain list was tested by the match_domain
3830 condition in an expansion string.
3832 16. Added the ${str2b64: operator.
3834 17. Exim was always calling setrlimit() to set a large limit for the number of
3835 processes, without checking whether the existing limit was already
3836 adequate. (It did check for the limit on file descriptors.) Furthermore,
3837 errors from getrlimit() and setrlimit() were being ignored. Now they are
3838 logged to the main and panic logs, but Exim does carry on, to try to do its
3839 job under whatever limits there are.
3841 18. Imported PCRE 5.0.
3843 19. Trivial typo in log message " temporarily refused connection" (the leading
3846 20. If the log selector return_path_on_delivery was set and an address was
3847 redirected to /dev/null, the delivery process crashed because it assumed
3848 that a return path would always be set for a "successful" delivery. In this
3849 case, the whole delivery is bypassed as an optimization, and therefore no
3852 21. Internal re-arrangement: the function for sending a challenge and reading
3853 a response while authentication was assuming a zero-terminated challenge
3854 string. It's now changed to take a pointer and a length, to allow for
3855 binary data in such strings.
3857 22. Added the cyrus_sasl authenticator (code supplied by MBM).
3859 23. Exim was not respecting finduser_retries when seeking the login of the
3860 uid under which it was called; it was always trying 10 times. (The default
3861 setting of finduser_retries is zero.) Also, it was sleeping after the final
3862 failure, which is pointless.
3864 24. Implemented tls_on_connect_ports.
3866 25. Implemented acl_smtp_predata.
3868 26. If the domain in control=submission is set empty, Exim assumes that the
3869 authenticated id is a complete email address when it generates From: or
3870 Sender: header lines.
3872 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
3873 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
3874 chown and chgrp in /bin and hostname in /usr/bin.
3876 28. Exim was keeping the "process log" file open after each use, just as it
3877 does for the main log. This opens the possibility of it remaining open for
3878 long periods when the USR1 signal hits a daemon. Occasional processlog
3879 errors were reported, that could have been caused by this. Anyway, it seems
3880 much more sensible not to leave this file open at all, so that is what now
3883 29. The long-running daemon process does not normally write to the log once it
3884 has entered its main loop, and it closes the log before doing so. This is
3885 so that log files can straightforwardly be renamed and moved. However,
3886 there are a couple of unusual error situations where the daemon does write
3887 log entries, and I had neglected to close the log afterwards.
3889 30. The text of an SMTP error response that was received during a remote
3890 delivery was being truncated at 512 bytes. This is too short for some of
3891 the long messages that one sometimes sees. I've increased the limit to
3894 31. It is now possible to make retry rules that apply only when a message has a
3895 specific sender, in particular, an empty sender.
3897 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
3898 it possible to be selective about when SMTP synchronization is enforced.
3900 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
3902 32. Implemented hosts_connection_nolog.
3904 33. Added an ACL for QUIT.
3906 34. Setting "delay_warning=" to disable warnings was not working; it gave a
3909 35. Added mailbox_size and mailbox_filecount to appendfile.
3911 36. Added control = no_multiline_responses to ACLs.
3913 37. There was a bug in the logic of the code that waits for the clock to tick
3914 in the case where the clock went backwards by a substantial amount such
3915 that the microsecond fraction of "now" was more than the microsecond
3916 fraction of "then" (but the whole seconds number was less).
3918 38. Added support for the libradius Radius client library this is found on
3919 FreeBSD (previously only the radiusclient library was supported).
3925 1. When certain lookups returned multiple values in the form name=value, the
3926 quoting of the values was not always being done properly. Specifically:
3927 (a) If the value started with a double quote, but contained no whitespace,
3929 (b) If the value contained whitespace other than a space character (i.e.
3930 tabs or newlines or carriage returns) it was not quoted.
3931 This fix has been applied to the mysql and pgsql lookups by writing a
3932 separate quoting function and calling it from the lookup code. The fix
3933 should probably also be applied to nisplus, ibase and oracle lookups, but
3934 since I cannot test any of those, I have not disturbed their existing code.
3936 2. A hit in the callout cache for a specific address caused a log line with no
3937 reason for rejecting RCPT. Now it says "Previous (cached) callout
3938 verification failure".
3940 3. There was an off-by-one bug in the queryprogram router. An over-long
3941 return line was truncated at 256 instead of 255 characters, thereby
3942 overflowing its buffer with the terminating zero. As well as fixing this, I
3943 have increased the buffer size to 1024 (and made a note to document this).
3945 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
3946 when Exim is waiting for an SMTP response from a remote server, Exim
3947 restarts its select() call on the socket, thereby resetting its timeout.
3948 This is not a problem when such interrupts are rare. Somebody set up a cron
3949 job to run exiwhat every 2 minutes, which is less than the normal select()
3950 timeout (5 or 10 minutes). This meant that the select() timeout never
3951 kicked in because it was always reset. I have fixed this by comparing the
3952 time when an interrupt arrives with the time at the start of the first call
3953 to select(). If more time than the timeout has elapsed, the interrupt is
3954 treated as a timeout.
3956 5. Some internal re-factoring in preparation for the addition of Sieve
3957 extensions (by MH). In particular, the "personal" test is moved to a
3958 separate function, and given an option for scanning Cc: and Bcc: (which is
3959 not set for Exim filters).
3961 6. When Exim created an email address using the login of the caller as the
3962 local part (e.g. when creating a From: or Sender: header line), it was not
3963 quoting the local part when it contained special characters such as @.
3965 7. Installed new OpenBSD configuration files.
3967 8. Reworded some messages for syntax errors in "and" and "or" conditions to
3968 try to make them clearer.
3970 9. Callout options, other than the timeout value, were being ignored when
3971 verifying sender addresses in header lines. For example, when using
3973 verify = header_sender/callout=no_cache
3975 the cache was (incorrectly) being used.
3977 10. Added a missing instance of ${EXE} to the exim_install script; this affects
3978 only the Cygwin environment.
3980 11. When return_path_on_delivery was set as a log selector, if different remote
3981 addresses in the same message used different return paths and parallel
3982 remote delivery occurred, the wrong values would sometimes be logged.
3983 (Whenever a remote delivery process finished, the return path value from
3984 the most recently started remote delivery process was logged.)
3986 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
3987 lines when AUTH and/or TLS are in use. This is the "received protocol"
3988 field. Exim used to use "asmtp" for authenticated SMTP, without any
3989 indication (in the protocol name) for TLS use. Now it follows the RFC and
3990 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
3991 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
3992 names appear in log lines as well as in Received: header lines.
3994 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
3995 extensions, and comparison tests, and to fix some bugs.
3997 14. Changes to the "personal" filter test:
3999 (1) The test was buggy in that it was just doing the equivalent of
4000 "contains" tests on header lines. For example, if a user's address was
4001 anne@some.where, the "personal" test would incorrectly be true for
4003 To: susanne@some.where
4005 This test is now done by extracting each address from the header in turn,
4006 and checking the entire address. Other tests that are part of "personal"
4007 are now done using regular expressions (for example, to check local parts
4008 of addresses in From: header lines).
4010 (2) The list of non-personal local parts in From: addresses has been
4011 extended to include "listserv", "majordomo", "*-request", and "owner-*",
4012 taken from the Sieve specification recommendations.
4014 (3) If the message contains any header line starting with "List-" it is
4015 treated as non-personal.
4017 (4) The test for "circular" in the Subject: header line has been removed
4018 because it now seems ill-conceived.
4020 15. Minor typos in src/EDITME comments corrected.
4022 16. Installed latest exipick from John Jetmore.
4024 17. If headers_add on a router specified a text string that was too long for
4025 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
4026 of string_sprintf() is now avoided.
4028 18. $message_body_size was not set (it was always zero) when running the DATA
4029 ACL and the local_scan() function.
4031 19. For the "mail" command in an Exim filter, no default was being set for
4032 the once_repeat time, causing a random time value to be used if "once" was
4033 specified. (If the value happened to be <= 0, no repeat happened.) The
4034 default is now 0s, meaning "never repeat". The "vacation" command was OK
4035 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
4036 (I found it when inspecting the code).
4038 20. There is now an overall timeout for performing a callout verification. It
4039 defaults to 4 times the callout timeout, which applies to individual SMTP
4040 commands during the callout. The overall timeout applies when there is more
4041 than one host that can be tried. The timeout is checked before trying the
4042 next host. This prevents very long delays if there are a large number of
4043 hosts and all are timing out (e.g. when the network connections are timing
4044 out). The value of the overall timeout can be changed by specifying an
4045 additional sub-option for "callout", called "maxwait". For example:
4047 verify = sender/callout=5s,maxwait=20s
4049 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
4050 to the end before writing, but this should make it even safer).
4052 22. Exim was forgetting that it had advertised PIPELINING for the second and
4053 subsequent messages on an SMTP connection. It was also not resetting its
4054 memory on STARTTLS and an internal HELO.
4056 23. When Exim logs an SMTP synchronization error within a session, it now
4057 records whether PIPELINING has been advertised or not.
4059 24. Added 3 instances of "(long int)" casts to time_t variables that were being
4060 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
4061 rather than long int.
4063 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
4065 26. Added the never_mail option to autoreply.
4071 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
4072 crash if the getsockname() call failed; this can happen if a connection is
4073 closed very soon after it is established. The problem was simply in the
4074 order in which certain operations were done, causing Exim to try to write
4075 to the SMTP stream before it had set up the file descriptor. The bug has
4076 been fixed by making things happen in the correct order.
4082 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
4083 before the connection was closed, thus losing the rejection response.
4085 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
4086 some early Solaris releases, but causes trouble in current releases where
4087 socklen_t is defined.
4089 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
4092 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
4095 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
4096 that releasing the top of it at the end releases what was used for sub-
4097 expansions (unless the block got too big). However, discard this block if
4098 the first thing is a variable or header, so that we can use its block when
4099 it is dynamic (useful for very large $message_headers, for example).
4101 6. Lookups now cache *every* query, not just the most recent. A new, separate
4102 store pool is used for this. It can be recovered when all lookup caches are
4103 flushed. Lookups now release memory at the end of their result strings.
4104 This has involved some general refactoring of the lookup sources.
4106 7. Some code has been added to the store_xxx() functions to reduce the amount
4107 of flapping under certain conditions.
4109 8. log_incoming_interface used to affect only the <= reception log lines. Now
4110 it causes the local interface and port to be added to several more SMTP log
4111 lines, for example "SMTP connection from", and rejection lines.
4113 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
4115 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
4117 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
4118 could be overwritten at the end of the current message (or the start of a
4119 new message if it was set in a HELO ACL). The value is now preserved for
4120 the duration of the SMTP connection.
4122 12. If a transport had a headers_rewrite setting, and a matching header line
4123 contained an unqualified address, that address was qualified, even if it
4124 did not match any rewriting rules. The underlying bug was that the values
4125 of the flags that permit the existence of unqualified sender and recipient
4126 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
4127 non-local messages, and by -bnq for local messages) were not being
4128 preserved with the message after it was received.
4130 13. When Exim was logging an SMTP synchronization error, it could sometimes log
4131 "next input=" as part of the text comprising the host identity instead of
4132 the correct text. The code was using the same buffer for two different
4133 strings. However, depending on which order the printing function evaluated
4134 its arguments, the bug did not always show up. Under Linux, for example, my
4135 test suite worked just fine.
4137 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
4138 doesn't work with some older versions of Perl. It has been changed to "my",
4139 which in any case is probably the better facility to use.
4141 15. A really picky compiler found some instances of statements for creating
4142 error messages that either had too many or two few arguments for the format
4145 16. The size of the buffer for calls to the DNS resolver has been increased
4146 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
4147 for addresses that have a lot of PTR records. This alleviates a problem; it
4148 does not fully solve it.
4150 17. A dnsdb lookup for PTR records that receives more data than will fit in the
4151 buffer now truncates the list and logs the incident, which is the same
4152 action as happens when Exim is looking up a host name and its aliases.
4153 Previously in this situation something unpredictable would happen;
4154 sometimes it was "internal error: store_reset failed".
4156 18. If a server dropped the connection unexpectedly when an Exim client was
4157 using GnuTLS and trying to read a response, the client delivery process
4158 crashed while trying to generate an error log message.
4160 19. If a "warn" verb in an ACL added multiple headers to a message in a single
4161 string, for example:
4163 warn message = H1: something\nH2: something
4165 the text was added as a single header line from Exim's point of view
4166 though it ended up OK in the delivered message. However, searching for the
4167 second and subsequent header lines using $h_h2: did not work. This has been
4168 fixed. Similarly, if a system filter added multiple headers in this way,
4169 the routers could not see them.
4171 20. Expanded the error message when iplsearch is called with an invalid key to
4172 suggest using net-iplsearch in a host list.
4174 21. When running tests using -bh, any delays imposed by "delay" modifiers in
4175 ACLs are no longer actually imposed (and a message to that effect is
4178 22. If a "gecos" field in a passwd entry contained escaped characters, in
4179 particular, if it contained a \" sequence, Exim got it wrong when building
4180 a From: or a Sender: header from that name. A second bug also caused
4181 incorrect handling when an unquoted " was present following a character
4182 that needed quoting.
4184 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
4185 was not being matched caselessly.
4187 24. Arranged for all hyphens in the exim.8 source to be escaped with
4190 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
4191 properly. Recipient callout cache records were still being keyed to include
4192 the sender, even when use_sender was set false. This led to far more
4193 callouts that were necessary. The sender is no longer included in the key
4194 when use_sender is false.
4196 26. Added "control = submission" modifier to ACLs.
4198 27. Added the ${base62d: operator to decode base 62 numbers.
4200 28. dnsdb lookups can now access SRV records.
4202 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
4203 the configuration file.
4205 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
4206 (-v) mode. This makes the output for a verbose queue run more intelligible.
4208 31. Added a use_postmaster feature to recipient callouts.
4210 32. Added the $body_zerocount variable, containing the number of binary zero
4211 bytes in the message body.
4213 33. The time of last modification of the "new" subdirectory is now used as the
4214 "mailbox time last read" when there is a quota error for a maildir
4217 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
4219 35. Added +ignore_unknown as a special item in host lists.
4221 36. Code for decoding IPv6 addresses in host lists is now included, even if
4222 IPv6 support is not being compiled. This fixes a bug in which an IPv6
4223 address was recognized as an IP address, but was then not correctly decoded
4224 into binary, causing unexpected and incorrect effects when compared with
4231 1. Very minor rewording of debugging text in manualroute to say "list of
4232 hosts" instead of "hostlist".
4234 2. If verify=header_syntax was set, and a header line with an unqualified
4235 address (no domain) and a large number of spaces between the end of the
4236 name and the colon was received, the reception process suffered a buffer
4237 overflow, and (when I tested it) crashed. This was caused by some obsolete
4238 code that should have been removed. The fix is to remove it!
4240 3. When running in the test harness, delay a bit after writing a bounce
4241 message to get a bit more predictability in the log output.
4243 4. Added a call to search_tidyup() just before forking a reception process. In
4244 theory, someone could use a lookup in the expansion of smtp_accept_max_
4245 per_host which, without the tidyup, could leave open a database connection.
4247 5. Added the variables $recipient_data and $sender_data which get set from a
4248 lookup success in an ACL "recipients" or "senders" condition, or a router
4249 "senders" option, similar to $domain_data and $local_part_data.
4251 6. Moved the writing of debug_print from before to after the "senders" test
4254 7. Change 4.31/66 (moving the time when the Received: is generated) caused
4255 problems for message scanning, either using a data ACL, or using
4256 local_scan() because the Received: header was not generated till after they
4257 were called (in order to set the time as the time of reception completion).
4258 I have revised the way this works. The header is now generated after the
4259 body is received, but before the ACL or local_scan() are called. After they
4260 are run, the timestamp in the header is updated.
4266 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
4267 before starting a queue runner without re-exec. This happened only when
4268 deliver_drop_privilege was set or when the Exim user was set to root. The
4269 effect of the bug was that timeouts during subsequent deliveries caused
4270 crashes instead of being properly handled. The handler is now left at its
4271 default (and expected) setting.
4273 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
4274 message, again when deliver_drop_privilege is set or Exim is run as root.
4275 The bug described in (1) was not present in this case, but the tidying up
4276 of the other signals was missing. I have made the two cases consistent.
4278 3. The ignore_target_hosts setting on a manualroute router was being ignored
4279 for hosts that were looked up using the /MX notation.
4281 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
4284 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
4285 operated on the sender address. After changing the $sender_address to <>
4286 for the sender address verify, Exim was re-instated it as the original
4287 (before rewriting) address, but remembering that it had rewritten it, so it
4288 wasn't rewriting it again. This bug also had the effect of breaking the
4289 sender address verification caching when the sender address was rewritten.
4291 6. The ignore_target_hosts option was being ignored by the ipliteral router.
4292 This has been changed so that if the ip literal address matches
4293 ignore_target_hosts, the router declines.
4295 7. Added expansion conditions match_domain, match_address, and match_local_
4296 part (NOT match_host).
4298 8. The placeholder for the Received: header didn't have a length field set.
4300 9. Added code to Exim itself and to exim_lock to test for a specific race
4301 condition that could lead to file corruption when using MBX delivery. The
4302 issue is with the lockfile that is created in /tmp. If this file is removed
4303 after a process has opened it but before that process has acquired a lock,
4304 there is the potential for a second process to recreate the file and also
4305 acquire a lock. This could lead to two Exim processes writing to the file
4306 at the same time. The added code performs the same test as UW imapd; it
4307 checks after acquiring the lock that its file descriptor still refers to
4308 the same named file.
4310 10. The buffer for building added header lines was of fixed size, 8192 bytes.
4311 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
4314 11. Added the smtp_active_hostname option. If used, this will typically be made
4315 to depend on the incoming interface address. Because $interface_address is
4316 not set up until the daemon has forked a reception process, error responses
4317 that can happen earlier (such as "too many connections") no longer contain
4320 12. If an expansion in a condition on a "warn" statement fails because a lookup
4321 defers, the "warn" statement is abandoned, and the next ACL statement is
4322 processed. Previously this caused the whole ACL to be aborted.
4324 13. Added the iplsearch lookup type.
4326 14. Added ident_timeout as a log selector.
4328 15. Added tls_certificate_verified as a log selector.
4330 16. Added a global option tls_require_ciphers (compare the smtp transport
4331 option of the same name). This controls incoming TLS connections.
4333 17. I finally figured out how to make tls_require_ciphers do a similar thing
4334 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
4335 before starting the TLS session.
4337 18. Tabs are now shown as \t in -bP output.
4339 19. If the log selector return_path_on_delivery was set, Exim crashed when
4340 bouncing a message because it had too many Received: header lines.
4342 20. If two routers both had headers_remove settings, and the first one included
4343 a superfluous trailing colon, the final name in the first list and the
4344 first name in the second list were incorrectly joined into one item (with a
4345 colon in the middle).
4351 1. Added -C and -D options to the exinext utility, mainly to make it easier
4352 to include in the automated testing, but these could be helpful when
4353 multiple configurations are in use.
4355 2. The exinext utility was not formatting the output nicely when there was
4356 an alternate port involved in the retry record key, nor when there was a
4357 message id as well (for retries that were specific to a specific message
4358 and a specific host). It was also confused by IPv6 addresses, because of
4359 the additional colons they contain. I have fixed the IPv4 problem, and
4360 patched it up to do a reasonable job for IPv6.
4362 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
4363 delivery, the log line now contains "pipelined" if PIPELINING was used.
4365 4. An SMTP transport process used to panic and die if the bind() call to set
4366 an explicit outgoing interface failed. This has been changed; it is now
4367 treated in the same way as a connect() failure.
4369 5. A reference to $sender_host_name in the part of a conditional expansion
4370 that was being skipped was still causing a DNS lookup. This no longer
4373 6. The def: expansion condition was not recognizing references to header lines
4374 that used bh_ and bheader_.
4376 7. Added the _cache feature to named lists.
4378 8. The code for checking quota_filecount in the appendfile transport was
4379 allowing one more file than it should have been.
4381 9. For compatibility with Sendmail, the command line option
4389 and sets the incoming protocol and host name (for trusted callers). The
4390 host name and its colon can be omitted when only the protocol is to be set.
4391 Note the Exim already has two private options, -pd and -ps, that refer to
4392 embedded Perl. It is therefore impossible to set a protocol value of "d" or
4393 "s", but I don't think that's a major issue.
4395 10. A number of refactoring changes to the code, none of which should affect
4398 (a) The number of logging options was getting close to filling up the
4399 32-bit word that was used as a bit map. I have split them into two classes:
4400 those that are passed in the argument to log_write(), and those that are
4401 only ever tested independently outside of that function. These are now in
4402 separate 32-bit words, so there is plenty of room for expansion again.
4403 There is no change in the user interface or the logging behaviour.
4405 (b) When building, for example, log lines, the code previously used a
4406 macro that called string_cat() twice, in order to add two strings. This is
4407 not really sufficiently general. Furthermore, there was one instance where
4408 it was actually wrong because one of the argument was used twice, and in
4409 one call a function was used. (As it happened, calling the function twice
4410 did not affect the overall behaviour.) The macro has been replaced by a
4411 function that can join an arbitrary number of extra strings onto a growing
4414 (c) The code for expansion conditions now uses a table and a binary chop
4415 instead of a serial search (which was left over from when there were very
4416 few conditions). Also, it now recognizes conditions like "pam" even when
4417 the relevant support is not compiled in: a suitably worded error message is
4418 given if an attempt is made to use such a condition.
4420 11. Added ${time_interval:xxxxx}.
4422 12. A bug was causing one of the ddress fields not to be passed back correctly
4423 from remote delivery subprocesses. The field in question was not being
4424 subsequently used, so this caused to problems in practice.
4426 13. Added new log selectors queue_time and deliver_time.
4428 14. Might have fixed a bug in maildirsizefile handling that threw up
4429 "unexpected character" debug warnings, and recalculated the data
4430 unnecessarily. In any case, I expanded the warning message to give more
4433 15. Added the message "Restricted characters in address" to the statements in
4434 the default ACL that block characters like @ and % in local parts.
4436 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
4437 Three changes have been made:
4439 (a) There was a serious bug; a negative response to MAIL caused the whole
4440 recipient domain to be cached as invalid, thereby blocking all messages
4441 to all local parts at the same domain, from all senders. This bug has
4442 been fixed. The domain is no longer cached after a negative response to
4443 MAIL if the sender used is not empty.
4445 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
4448 (c) A new callout option, "use_sender" has been added for people who want
4449 the modified behaviour.
4455 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
4458 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
4459 indeed breaks things for older releases.
4461 3. Added additional logging to the case where there is a problem reading data
4462 from a filter that is running in a subprocess using a pipe, in order to
4463 try to track down a specific problem.
4465 4. Testing facility fudge: when running in the test harness and attempting
4466 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
4467 getting "No route to host". Convert this to a timeout.
4469 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
4472 6. Some OS don't have socklen_t but use size_t instead. This affects the
4473 fifth argument of getsockopt() amongst other things. This is now
4474 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
4475 can be set for individual OS. I have set it for SunOS5, OSF1, and
4476 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
4477 some earlier ones do not.
4479 7. Change 4.30/15 was not doing the test caselessly.
4481 8. The standard form for an IPv6 address literal was being rejected by address
4482 parsing in, for example, MAIL and RCPT commands. An example of this kind of
4483 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
4484 this, as well as the form without the "IPv6" on the front (but only when
4485 address literals are enabled, of course).
4487 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
4489 10. Exim crashed if a message with an empty sender address specified by -f
4490 encountered a router with an errors_to setting. This could be provoked only
4491 by a command such as
4495 where an empty string was supplied; "<>" did not hit this bug.
4497 11. Installed PCRE release 4.5.
4499 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
4500 remained set. It is now erased.
4502 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
4503 times from message ids (which are base 36 rather than the normal 62).
4505 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
4506 were being counted as actual protocol errors, and logged if the log
4507 selector +smtp_protocol_error was set. One cannot be perfect in this test,
4508 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
4509 and DATA following a set of rejected RCPTs do not count as protocol errors.
4510 In other words, Exim assumes they were pipelined, though this may not
4511 actually be the case. Of course, in all cases the client gets an
4512 appropriate error code.
4514 15. If a lookup fails in an ACL condition, a message about the failure may
4515 be available; it is used if testing the ACL cannot continue, because most
4516 such messages specify what the cause of the deferral is. However, some
4517 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
4518 that caused an old message to be retained and used if a later statement
4519 caused a defer, replacing the real cause of the deferral.
4521 16. If an IP address had so many PTR records that the DNS lookup buffer
4522 was not large enough to hold them, Exim could crash while trying to process
4523 the truncated data. It now detects and logs this case.
4525 17. Further to 4.21/58, another change has been made: if (and only if) the
4526 first line of a message (the first header line) ends with CRLF, a bare LF
4527 in a subsequent header line has a space inserted after it, so as not to
4528 terminate the header.
4530 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
4531 unnecessarily, used atoi() instead of strtol(), and didn't check the
4532 termination when getting file sizes from file names by regex.
4534 19. Completely re-implemented the support for maildirsize files, in the light
4535 of a number of problems with the previous contributed implementation
4536 (4.30/29). In particular:
4538 . If the quota is zero, the maildirsize file is maintained, but no quota is
4541 . If the maildir directory does not exist, it is created before any attempt
4542 to write a maildirsize file.
4544 . The quota value in the file is just a cache; if the quota is changed in
4545 the transport, the new value overrides.
4547 . A regular expression is available for excluding directories from the
4550 20. The autoreply transport checks the characters in options that define the
4551 message's headers; it allows continued headers, but it was checking with
4552 isspace() after an embedded newline instead of explicitly looking for a
4555 21. If all the "regular" hosts to which an address was routed had passed their
4556 expiry times, and had not reached their retry times, the address was
4557 bounced, even if fallback hosts were defined. Now Exim should go on to try
4560 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
4561 equivalent code in the SMTP transport. Some hosts send humungous responses
4562 to HELO/EHLO, more than 1024 it seems.
4564 23. Refactoring: code in filter.c used (void *) for "any old type" but this
4565 gives compiler warnings in some environments. I've now done it "properly",
4568 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
4569 (because of problems with the built-in one) was declared to return uschar *
4570 instead of char *, causing compiler failure.
4572 25. Fixed a file descriptor leak when processing alias/forward files.
4574 26. Fixed a minor format string issue in dbfn.c.
4576 27. Typo in exim.c: ("dmbnz" for "dbmnz").
4578 28. If a filter file refered to $h_xxx or $message_headers, and the headers
4579 contained RFC 2047 "words", Exim's memory could, under certain conditions,
4582 29. When a sender address is verified, it is cached, to save repeating the test
4583 when there is more than one recipient in a message. However, when the
4584 verification involves a callout, it is possible for different callout
4585 options to be set for different recipients. It is too complicated to keep
4586 track of this in the cache, so now Exim always runs a verification when a
4587 callout is required, relying on the callout cache for the optimization.
4588 The overhead is duplication of the address routing, but this should not be
4591 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
4592 to be verified with callout=postmaster, and the main callout worked but the
4593 postmaster check failed, the verification correctly failed. However, if a
4594 subsequent RCPT command asked for sender verification *without* the
4595 postmaster check, incorrect caching caused this verification also to fail,
4598 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
4599 it was not caching the DNS options (qualify_single, search_parents) that
4600 were used when the lookup failed. A subsequent lookup with different
4601 options therefore always gave the same answer, though there were cases
4602 where it should not have. (Example: a "domains = !$mx_any" option on a
4603 dnslookup router: the "domains" option is always processed without any
4604 widening, but the router might have qualify_single set.) Now Exim uses the
4605 cached value only when the same options are set.
4607 32. Added John Jetmore's "exipick" utility to the distribution.
4609 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
4610 than a timeout (e.g. a certificate is required, and is not provided), an
4611 Exim server now closes the connection immediately. Previously it waited for
4612 the client to close - but if the client is SSL, it seems that they each
4613 wait for each other, leading to a delay before one of them times out.
4615 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
4616 maintained 0.8.x compatibility because I don't think many are using it, and
4617 it is clearly obsolete.
4619 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
4622 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
4623 client certificate was expired. A simple patch fixes this, though I don't
4624 understand the full logic of why the verify callback is called multiple
4627 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
4628 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
4629 which causes problems with some clients (such as the Certicom SSL Plus
4630 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
4631 disables the coutermeasure allowing Eudora to connect."
4633 38. Exim was not checking that a write() to a log file succeeded. This could
4634 lead to Bad Things if a log got too big, in particular if it hit a file
4635 size limit. Exim now panics and dies if it cannot write to a log file, just
4636 as it does if it cannot open a log file.
4638 39. Modified OS/Makefile-Linux so that it now contains
4640 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
4642 The two -D definitions ensure that Exim is compiled with large file
4643 support, which makes it possible to handle log files that are bigger than
4646 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
4647 instance) a domain was checked against a named list that involved a lookup,
4648 causing $domain_data to be set, then another domain was checked against the
4649 same list, then the first domain was re-checked, the value of $domain_data
4650 after the final check could be wrong. In particular, if the second check
4651 failed, it could be set empty. This bug probably also applied to
4654 41. The strip_trailing_dot option was not being applied to the address given
4655 with the -f command-line option.
4657 42. The code for reading a message's header from the spool was incrementing
4658 $received_count, but never initializing it. This meant that the value was
4659 incorrect (doubled) while delivering a message in the same process in which
4660 it was received. In the most common configuration of Exim, this never
4661 happens - a fresh exec is done - but it can happen when
4662 deliver_drop_privilege is set.
4664 43. When Exim logs an SMTP synchronization error - client data sent too soon -
4665 it now includes up to 150 characters of the unexpected data in the log
4668 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
4669 and building data strings. The size of both of these buffers was 10 000
4670 bytes - far larger than anybody would *ever* want, thought I. Needless to
4671 say, somebody hit the limit. I have increased the maximum line length to
4672 20 000 and the maximum data length of concatenated lines to 100 000. I have
4673 also fixed two bugs, because there was no checking on these buffers. Tsk,
4674 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
4675 buffer is too small.
4677 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
4678 lsearch lookups. Now it does.
4680 46. When parsing a route_list item in a manualroute router, a fixed-length
4681 buffer was used for the list of hosts. I made this 1024 bytes long,
4682 thinking that nobody would ever have a list of hosts that long. Wrong.
4683 Somebody had a whole pile of complicated expansion conditions, and the
4684 string was silently truncated, leading to an expansion error. It turns out
4685 that it is easier to change to an unlimited length (owing to other changes
4686 that have happened since this code was originally written) than to build
4687 structure for giving a limitation error. The length of the item that
4688 expands into the list of hosts is now unlimited.
4690 47. The lsearch lookup could not handle data where the length of text line was
4691 more than 4095 characters. Such lines were truncated, leading to shortened
4692 data being returned. It should now handle lines of any length.
4694 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
4695 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
4698 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
4699 replacing strings such as BIN_DIRECTORY in a source file: the replacement
4700 no longer happens in comment lines. A list of replacements is now placed
4701 at the head of all of the source files, except those whose only change is
4702 to replace PERL_COMMAND in the very first #! line.
4704 50. Replaced the slow insertion sort in queue.c, for sorting the list of
4705 messages on the queue, with a bottom-up merge sort, using code contributed
4706 by Michael Haardt. This should make operations like -bp somewhat faster on
4707 large queues. It won't affect queue runners, except when queue_run_in_order
4710 51. Installed eximstats 1.31 in the distribution.
4712 52. Added support for SRV lookups to the dnslookup router.
4714 53. If an ACL referred to $message_body or $message_body_end, the value was not
4715 reset for any messages that followed in the same SMTP session.
4717 54. The store-handling optimization for building very long strings was not
4718 differentiating between the different store pools. I don't think this
4719 actually made any difference in practice, but I've tidied it.
4721 55. While running the routers to verify a sender address, $sender_address
4722 was still set to the sender address. This is wrong, because when routing to
4723 send a bounce to the sender, it would be empty. Therefore, I have changed
4724 it so that, while verifying a sender address, $sender_address is set to <>.
4725 (There is no change to what happens when verifying a recipient address.)
4727 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
4728 target A or AAAA records (if not already returned) without resetting the
4729 qualify_single or search_parents options of the DNS resolver. These are
4730 inappropriate in this case because the targets of MX and SRV records must
4731 be FQDNs. A broken DNS record could cause trouble if it happened to have a
4732 target that, when qualified, matched something in the local domain. These
4733 two options are now turned off when doing these lookups.
4735 57. It seems that at least some releases of Reiserfs (which does not have the
4736 concept of a fixed number of inodes) returns zero and not -1 for the
4737 number of available inodes. This interacted badly with check_spool_inodes,
4738 which assumed that -1 was the "no such thing" setting. What I have done is
4739 to check that the total number of inodes is greater than zero before doing
4740 the test of how many are available.
4742 58. When a "warn" ACL statement has a log_message modifier, the message is
4743 remembered, and not repeated. This is to avoid a lot of repetition when a
4744 message has many recipients that cause the same warning to be written.
4745 Howewer, Exim was preserving the list of already written lines for an
4746 entire SMTP session, which doesn't seem right. The memory is now reset if a
4747 new message is started.
4749 59. The "rewrite" debugging flag was not showing the result of rewriting in the
4750 debugging output unless log_rewrite was also set.
4752 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
4753 of (int)(handle) when we know that handle contains (void *)(-1).
4755 61. The Exim daemon panic-logs an error return when it closes the incoming
4756 connection. However "connection reset by peer" seems to be common, and
4757 isn't really an error worthy of noting specially, so that particular error
4760 62. When Exim is trying to find all the local interfaces, it used to panic and
4761 die if the ioctl to get the interface flags failed. However, it seems that
4762 on at least one OS (Solaris 9) it is possible to have an interface that is
4763 included in the list of interfaces, but for which you get a failure error
4764 for this call. This happens when the interface is not "plumbed" into a
4765 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
4766 failure of the "get flags" call assumes that the interface is down.
4768 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
4769 makes life easier for people who are doing arithmetic on fields extracted
4770 from dates, where you often get leading zeros that should not be
4771 interpreted as octal.
4773 64. Added qualify_domain to the redirect router, to override the global
4776 65. If a pathologically long header line contained very many addresses (the
4777 report of this problem mentioned 10 000) and each of them was rewritten,
4778 Exim could use up a very large amount of memory. (It kept on making new
4779 copies of the header line as it rewrote, and never released the old ones.)
4780 At the expense of a bit more processing, the header rewriting function has
4781 been changed so that it no longer eats memory in this way.
4783 66. The generation of the Received: header has been moved from the time that a
4784 message starts to be received, to the time that it finishes. The timestamp
4785 in the Received: header should now be very close to that of the <= log
4786 line. There are two side-effects of this change:
4788 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
4789 logged header lines no longer include the local Received: line, because
4790 it has not yet been created. The same applies to a copy of the message
4791 that is returned to a non-SMTP sender when a message is rejected.
4793 (b) When a filter file is tested using -bf, no additional Received: header
4794 is added to the test message. After some thought, I decided that this
4797 This change does not affect the value of $received_for. It is still set
4798 after address rewriting, but before local_scan() is called.
4800 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
4802 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
4803 gave an unhelpful panic error message, and a defer error. I have managed to
4804 change this behaviour so that it now rejects any supplied certificate,
4805 which seems right, as the list of acceptable certificates is empty.
4807 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
4808 gave an unhelpful defer error. I have not managed to make this reject any
4809 supplied certificates, but the error message it gives is "no certificate
4810 supplied", which is not helpful.
4812 70. exigrep's output now also includes lines that are not associated with any
4813 message, but which match the given pattern. Implemented by a patch from
4814 Martin Sluka, which also tidied up the Perl a bit.
4816 71. Recipient callout verification, like sender verification, was using <> in
4817 the MAIL FROM command. This isn't really the right thing, since the actual
4818 sender may affect whether the remote host accepts the recipient or not. I
4819 have changed it to use the actual sender in the callout; this means that
4820 the cache record is now keyed on a recipient/sender pair, not just the
4821 recipient address. There doesn't seem to be a real danger of callout loops,
4822 since a callout by the remote host to check the sender would use <>.
4823 [SEE ABOVE: changed after hitting problems.]
4825 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
4826 temporary errors. However, in the case of such a code being given after
4827 the end of a data transmission (i.e. after ".") Exim was failing to write
4828 a retry record for the message. (Yes, there was some broken host that was
4829 actually sending 8xx at this point.)
4831 73. An unknown lookup type in a host list could cause Exim to panic-die when
4832 the list was checked. (An example that provoked this was putting <; in the
4833 middle of a list instead of at the start.) If this happened during a DATA
4834 ACL check, a -D file could be left lying around. This kind of configuration
4835 error no longer causes Exim to die; instead it causes a defer errror. The
4836 incident is still logged to the main and panic logs.
4838 74. Buglet left over from Exim 3 conversion. The message "too many messages
4839 in one connection" was written to the rejectlog but not the mainlog, except
4840 when address rewriting (yes!) was being logged.
4842 75. Added write_rejectlog option.
4844 76. When a system filter was run not as root (that is, when system_filter_user
4845 was set), the values of the $n variables were not being returned to the
4846 main process; thus, they were not subsequently available in the $sn
4849 77. Added +return_path_on_delivery log selector.
4851 78. A connection timeout was being treated differently from recipients deferred
4852 when testing hosts_max_try with a message that was older than the host's
4853 retry timeout. (The host should not be counted, thus allowing all hosts to
4854 be tried at least once before bouncing.) This may have been the cause of an
4855 occasionally reported bug whereby a message would remain on the queue
4856 longer than the retry timeout, but would be bounced if a delivery was
4857 forced. I say "may" because I never totally pinned down the problem;
4858 setting up timeout/retry tests is difficult. See also the next item.
4860 79. The ultimate address timeout was not being applied to errors that involved
4861 a combination of host plus message (for example, a timeout on a MAIL
4862 command). When an address resolved to a number of possible hosts, and they
4863 were not all tried for each delivery (e.g. because of hosts_max_try), a
4864 message could remain on the queue longer than the retry timeout.
4866 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
4869 81. Fixed an obscure SMTP outgoing bug which required at least the following
4870 conditions: (a) there was another message waiting for the same server;
4871 (b) the server returned 5xx to all RCPT commands in the first message so
4872 that the message was not completed; (c) the server dropped the connection
4873 or gave a negative response to the RSET that Exim sends to abort the
4874 transaction. The observed case was a dropped connection after DATA that had
4875 been sent in pipelining mode. That is, the server had advertised PIPELINING
4876 but was not implementing it correctly. The effect of the bug was incorrect
4877 behaviour, such as trying another host, and this could lead to a crash.
4883 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
4884 and daemon.c were passed as pointers to ints; they should have been
4885 pointers to socklen_t variables (which are typically unsigned ints).
4887 2. Some signed/unsigned type warnings in the os.c file for Linux have been
4890 3. Fixed a really odd bug that affected only the testing scheme; patching a
4891 certain fixed string in the binary changed the value of another string that
4892 happened to be identical to the end of the original first string.
4894 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
4895 name", it returns that address as the IP address. On some operating
4896 systems (e.g. Solaris), it also passes back the IP address string as the
4897 "host name". However, on others (e.g. Linux), it passes back an empty
4898 string. Exim wasn't checking for this, and was changing the host name to an
4899 empty string, assuming it had been canonicized.
4901 5. Although rare, it is permitted to have more than one PTR record for a given
4902 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
4903 all the names associated with an address, because they do in Solaris.
4904 However, it seems that they do not in Linux for data that comes from the
4905 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
4906 I found this out when I moved to a new Linux workstation and tried to run
4907 the Exim test suite.
4909 To get round this problem I have changed the code so that it now does its
4910 own call to the DNS to look up PTR records when searching for a host name.
4911 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
4912 addresses that are only in /etc/hosts are still found.
4914 This behaviour is, however, controlled by an option called host_lookup_
4915 order, which defaults to "bydns:byaddr". If people want to use the other
4916 order, or indeed, just use one or the other means of lookup, they can
4917 specify it in this variable.
4919 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
4920 some operating systems, this comes back from gethostbyaddr() as an empty
4921 string, and this is what Exim used to test for. However, it seems that in
4922 other systems, "." is yielded. Exim now tests for this case too.
4924 7. The values of check_spool_space and check_log_space are now held internally
4925 as a number of kilobytes instead of an absolute number of bytes. If a
4926 numbers is specified without 'K' or 'M', it is rounded up to the nearest
4927 kilobyte. This means that much larger values can be stored.
4929 8. Exim monitor: an attempt to get the action menu when not actually pointing
4930 at a message produces an empty menu entitled "No message selected". This
4931 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
4932 no entries in it ("Shell widget menu has zero width and/or height"). So I
4933 have added a single, blank menu entry in this case.
4935 9. Added ${quote_local_part.
4937 10. MIME decoding is now applied to the contents of Subject: header lines when
4940 11. Now that a reference to $sender_host_address automatically causes a reverse
4941 lookup to occur if necessary (4.13/18), there is no need to arrange for a
4942 host lookup before query-style lookups in lists that might use this
4943 variable. This has therefore been abolished, and the "net-" prefix is no
4944 longer necessary for query-style lookups.
4946 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
4947 have been a typo for LFLAGS, so it has been changed.
4949 13. The install script calls Exim with "-C /dev/null" in order to find the
4950 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
4951 to be output. Howeve, since Exim outputs its version number before the
4952 error, it didn't break the script. It just looked ugly. I fixed this by
4953 always allowing "-C /dev/null" if the caller is root.
4955 14. Ignore overlarge ACL variable number when reading spool file - insurance
4956 against a later release with more variables having written the file.
4958 15. The standard form for an IPv6 address literal was being rejected by EHLO.
4959 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
4960 this, as well as the form without the "IPv6" on the front.
4962 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
4963 OS/Makefile-Darwin file.
4965 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
4966 only to LDAP libraries that do not have LDAP_OPT_DEREF.
4968 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
4969 variable. However, some OS use ints for inodes. I've added cast to long int
4970 to get rid of the compiler warning.
4972 19. I had forgotten to lock out "/../" in configuration file names when
4973 ALT_CONFIG_PREFIX was set.
4975 20. Routers used for verification do not need to specify transports. However,
4976 if such a router generated a host list, and callout was configured, Exim
4977 crashed, because it could not find a port number from the (non-existent)
4978 transport. It now assumes port 25 in this circumstance.
4980 21. Added the -t option to exigrep.
4982 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
4983 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
4984 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
4985 them is set, LOOKUP_LSEARCH is forced.
4987 23. "exim -bV" now outputs a list of lookups that are included in the binary.
4989 24. Added sender and host information to the "rejected by local_scan()" log
4990 line; previously there was no indication of these.
4992 25. Added .include_if_exists.
4994 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
4995 Linux. It turns out that not all file systems support this. Apparently some
4996 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
4997 do it.) To cope with this, the error EINVAL, which means that sync-ing is
4998 not supported on the file descriptor, is now ignored when Exim is trying to
4999 sync a directory. This applies only to Linux.
5001 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
5003 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
5004 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
5007 29. Added support for maildirsize files from supplied patch (modified a bit).
5009 30. The use of :fail: followed by an empty string could lead Exim to respond to
5010 sender verification failures with (e.g.):
5012 550 Verification failed for <xxx>
5013 550 Sender verify failed
5015 where the first response line was missing the '-' that indicates it is not
5016 the final line of the response.
5018 31. The loop for finding the name of the user that called Exim had a hardwired
5019 limit of 10; it now uses the value of finduser_retries, which is used for
5020 all other user lookups.
5022 32. Added $received_count variable, available in data and not_smtp ACLs, and at
5025 33. Exim was neglecting to zero errno before one call of strtol() when
5026 expanding a string and expecting an integer value. On some systems this
5027 resulted in spurious "integer overflow" errors. Also, it was casting the
5028 result into an int without checking.
5030 34. Testing for a connection timeout using "timeout_connect" in the retry rules
5031 did not work. The code looks as if it has *never* worked, though it appears
5032 to have been documented since at least releast 1.62. I have made it work.
5034 35. The "timeout_DNS" error in retry rules, also documented since at least
5035 1.62, also never worked. As it isn't clear exactly what this means, and
5036 clearly it isn't a major issue, I have abolished the feature by treating it
5037 as "timeout", and writing a warning to the main and panic logs.
5039 36. The display of retry rules for -brt wasn't always showing the error code
5042 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
5043 timeout_connect_A, timeout_connect_MX.
5045 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
5046 to the empty sender.
5048 39. The daemon was not analysing the content of -oX till after it had closed
5049 stderr and disconnected from the controlling terminal. This meant that any
5050 syntax errors were only noted on the panic log, and the return code from
5051 the command was 0. By re-arranging the code a little, I've made the
5052 decoding happen first, so such errors now appear on stderr, and the return
5053 code is 1. However, the actual setting up of the sockets still happens in
5054 the disconnected process, so errors there are still only recorded on the
5057 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
5058 connections (as happens on some IP stacks) was logged at start up time as
5059 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
5060 it from "IPv6 and IPv4", which means that two separate sockets are being
5063 41. The debug output for gethostbyname2() or getipnodebyname() failures now
5064 says whether AF_INET or AF_INET6 was passed as an argument.
5066 42. Exiwhat output was messed up when time zones were included in log
5069 43. Exiwhat now gives more information about the daemon's listening ports,
5070 and whether -tls-on-connect was used.
5072 44. The "port" option of the smtp transport is now expanded.
5074 45. A "message" modifier in a "warn" statement in a non-message ACL was being
5075 silently ignored. Now an error message is written to the main and panic
5078 46. There's a new ACL modifier called "logwrite" which writes to a log file
5079 as soon as it is encountered.
5081 47. Added $local_user_uid and $local_user_gid at routing time.
5083 48. Exim crashed when trying to verify a sender address that was being
5086 49. Exim was recognizing only a space character after ".include". It now also
5087 recognizes a tab character.
5089 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
5090 extracting the relevant information from the specification. The man page no
5091 longer contains scrambled data for the -d option, and I've added a section
5092 at the front about calling Exim under different names.
5094 51. Added "extra_headers" argument to the "mail" command in filter files.
5096 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
5099 53. Installed eximstats 1.29.
5101 54. Added transport_filter_timeout as a generic transport option.
5103 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
5104 Cc: header lines. This was required by RFC 822, but it not required by RFC
5107 56. Exim used to add From:, Date:, and Message-Id: header lines to any
5108 incoming messages that did not have them. Now it does so only if the
5109 message originates locally, that is, if there is no associated remote host
5110 address. When Resent- header lines are present, this applies to the Resent-
5111 lines rather than the non-Resent- lines.
5113 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
5114 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
5116 58. Messages for configuration errors now include the name of the main
5117 configuration file - useful now that there may be more than one file in a
5118 list (.included file names were always shown).
5120 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
5121 for those rare installations that do not start the daemon as root or run it
5122 setuid root. I've cut out the call to initgroups() if the daemon is not
5125 60. The Exim user and group can now be bound into the binary as text strings
5126 that are looked up at the start of Exim's processing.
5128 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
5130 62. Added $mailstore_basename variable.
5132 63. Installed patch to sieve.c from Michael Haardt.
5134 64. When Exim failed to open the panic log after failing to open the main log,
5135 the original message it was trying to log was written to stderr and debug
5136 output, but if they were not available (the usual case in production), it
5137 was lost. Now it is written to syslog before the two lines that record the
5138 failures to open the logs.
5140 65. Users' Exim filters run in subprocesses under the user's uid. It is
5141 possible for a "deliver" command or an alias in a "personal" command to
5142 provoke an address rewrite. If logging of address rewriting is configured,
5143 this fails because the process is not running as root or exim. There may be
5144 a better way of dealing with this, but for the moment (because 4.30 needs
5145 to be released), I have disabled address rewrite logging when running a
5146 filter in a non-root, non-exim process.
5152 1. The buildconfig auxiliary program wasn't quoting the value set for
5153 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
5154 not defined. This bug was masked in 4.22 by the effect that was fixed in
5157 2. Some messages that were rejected after a message id was allocated were
5158 shown as "incomplete" by exigrep. It no longer does this for messages that
5159 are rejected by local_scan() or the DATA or non-SMTP ACLs.
5161 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
5162 have allow_domain_literals set, the ID did not get logged in the <= line.
5163 Domain literals are now always recognized in Message-ID: header lines.
5165 4. The first argument for a ${extract expansion item is the key name or field
5166 number. Leading and trailing spaces in this item were not being ignored,
5167 causing some misleading effects.
5169 5. When deliver_drop_privilege was set, single queue runner processes started
5170 manually (i.e. by the command "exim -q") or by the daemon (which uses the
5171 same command in the process it spins off) were not dropping privilege.
5173 6. When the daemon running as "exim" started a queue runner, it always
5174 re-executed Exim in the spun-off process. This is a waste of effort when
5175 deliver_drop_privilege is set. The new process now just calls the
5176 queue-runner function directly.
5182 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
5185 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
5186 ignored. Though the use of -oP was forcing the writing of a pid file, it
5187 was always written to the default place.
5189 3. If the message "no IP address found for host xxxx" is generated during
5190 incoming verification, it is now followed by identification of the incoming
5191 connection (so you can more easily find what provoked it).
5193 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
5195 5. Added some features to "harden" Exim a bit more against certain attacks:
5197 (a) There is now a build-time option called FIXED_NEVER_USERS that can
5198 be put in Local/Makefile. This is like the never_users runtime option,
5199 but it cannot be overridden. The default setting is "root".
5201 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
5202 prefix string with which any file named in a -C command line option
5205 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
5206 is retained for -C and -D only if the caller of Exim is root. Without
5207 it, the exim user may also use -C and -D and retain privilege.
5209 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
5210 command line option is disabled.
5212 6. Macro names set by the -D option must start with an upper case letter, just
5213 like macro names defined in the configuration file.
5215 7. Added "dereference=" facility to LDAP.
5217 8. Two instances of the typo "uknown" in the source files are fixed.
5219 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
5220 the Configure-Makefile script screwed up while processing it.
5222 10. Incorporated PCRE 4.4.
5224 11. The SMTP synchronization check was not operating right at the start of an
5225 SMTP session. For example, it could not catch a HELO sent before the client
5226 waited for the greeting. There is now a check for outstanding input at the
5227 point when the greeting is written. Because of the duplex, asynchronous
5228 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
5229 way, but not yet received, when the check is performed.
5231 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
5232 on TCP/IP sockets, because this apparently causes some broken clients to
5235 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
5236 unchanged) from the Cygwin maintainer.
5238 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
5239 was supported instead of testing for mbx. Effectively a typo.
5241 15. The spa authenticator server code was not checking that the input it
5242 received was valid base64.
5244 16. The debug output line for the "set" modifier in ACLs was not showing the
5245 name of the variable that was being set.
5247 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
5249 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
5250 lookup on its own. Something else was needed to trigger the lookup. For
5251 example, a match in host_lookup or the need for a host name in a host list.
5252 Now, if $sender_host_name is referenced and the host name has not yet been
5253 looked up, a lookup is performed. If the lookup fails, the variable remains
5254 empty, and $host_lookup_failed is set to "1".
5256 19. Added "eqi" as a case-independent comparison operator.
5258 20. The saslauthd authentication condition could segfault if neither service
5259 nor realm was specified.
5261 21. If an overflowing value such as "2048M" was set for message_size_limit, the
5262 error message that was logged was misleading, and incoming SMTP
5263 connections were dropped. The message is now more accurate, and temporary
5264 errors are given to SMTP connections.
5266 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
5267 (except RSET) with a 421 error, until QUIT is received. However, it was
5268 failing to send a response to QUIT.
5270 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
5271 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
5272 ACL is now run after the helo_try_verify_hosts code.
5274 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
5275 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
5276 case-independent, so other case variants are also recognized). Apparently
5277 some systems use these upper case variants.
5279 25. If more than two messages were waiting for the same host, and a transport
5280 filter was specified for the transport, Exim sent two messages over the
5281 same TCP/IP connection, and then failed with "socket operation on non-
5282 socket" when it tried to send the third.
5284 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
5286 27. The extern definition of crypt16() in expand.c was not being excluded when
5287 the OS had its own crypt16() function.
5289 28. Added bounce_return_body as a new option, and bounce_return_size_limit
5290 as a preferred synonym for return_size_limit, both as an option and as an
5293 29. Added LIBS=-liconv to OS/Makefile-OSF1.
5295 30. Changed the default configuration ACL to relax the local part checking rule
5296 for addresses that are not in any local domains. For these addresses,
5297 slashes and pipe symbols are allowed within local parts, but the sequence
5298 /../ is explicitly forbidden.
5300 31. SPA server authentication was not clearing the challenge buffer before
5303 32. log_message in a "warn" ACL statement was writing to the reject log as
5304 well as to the main log, which contradicts the documentation and doesn't
5305 seem right (because no rejection is happening). So I have stopped it.
5307 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
5308 However, I am unable to do any testing of this.
5310 34. Fixed an infelicity in the appendfile transport. When checking directories
5311 for a mailbox, to see if any needed to be created, it was accidentally
5312 using path names with one or more superfluous leading slashes; tracing
5313 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
5315 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
5316 contents are added to the log line that is written for every discarded
5317 recipient. (Previously a log_message setting was ignored.)
5319 36. The ${quote: operator now quotes the string if it is empty.
5321 37. The install script runs exim in order to find its version number. If for
5322 some reason other than non-existence or emptiness, which it checks, it
5323 could not run './exim', it was installing it with an empty version number,
5324 i.e. as "exim-". This error state is now caught, and the installation is
5327 38. An argument was missing from the function that creates an error message
5328 when Exim fails to connect to the socket for saslauthd authentication.
5329 This could cause Exim to crash, or give a corrupted message.
5331 39. Added isip, isip4, and isip6 to ${if conditions.
5333 40. The ACL variables $acl_xx are now saved with the message, and can be
5334 accessed later in routers, transports, and filters.
5336 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
5337 strings in the file are not string-expanded.
5339 42. If a MAIL command specified a SIZE value that was too large to fit into an
5340 int variable, the check against message_size_limit failed. Such values are
5341 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
5342 day this will have to be increased, but I don't think I want to be around
5343 when emails are that large.
5350 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
5351 iconv() is not standard in FreeBSD.
5353 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
5354 IPv6 enabled. The observed symptom was a segmentation fault on return from
5355 the function os_common_find_running_interfaces() in src/os.c.
5357 3. In the check_special_case() function in daemon.c I had used "errno" as an
5358 argument name, which causes warnings on some systems. This was basically a
5359 typo, since it was named "eno" in the comments!
5361 4. The code that waits for the clock to tick (at a resolution of some fraction
5362 of a second) so as to ensure message-id uniqueness was always waiting for
5363 at least one whole tick, when it could have waited for less. [This is
5364 almost certainly not relevant at current processor speeds, where it is
5365 unlikely to ever wait at all. But we try to future-proof.]
5367 5. The function that sleeps for a time interval that includes fractions of a
5368 second contained a race. It did not block SIGALRM between setting the
5369 timer, and suspending (a couple of lines later). If the interval was short
5370 and the sigsuspend() was delayed until after it had expired, the suspension
5371 never ended. On busy systems this could lead to processes getting stuck for
5374 6. Some uncommon configurations may cause a lookup to happen in a queue runner
5375 process, before it forks any delivery processes. The open lookup caching
5376 mechanism meant that the open file or database connection was passed into
5377 the delivery process. The problem was that delivery processes always tidy
5378 up cached lookup data. This could cause a problem for the next delivery
5379 process started by the queue runner, because the external queue runner
5380 process does not know about the closure. So the next delivery process
5381 still has data in the lookup cache. In the case of a file lookup, there was
5382 no problem because closing a file descriptor in a subprocess doesn't affect
5383 the parent. However, if the lookup was caching a connection to a database,
5384 the connection was closed, and the second delivery process was likely to
5385 see errors such as "PGSQL: query failed: server closed the connection
5386 unexpectedly". The problem has been fixed by closing all cached lookups
5387 in a queue runner before running a delivery process.
5389 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
5390 seem to have the "const" qualifier which it has on other OS. I've
5393 8. Change 4.21/2 was too strict. It is only if there are two authenticators
5394 *of the same type* (client or server) with the same public name that an
5395 error should be diagnosed.
5397 9. When Exim looked up a host name for an IP address, but failed to find the
5398 original IP address when looking up the host name (a safety check), it
5399 output the message "<ip address> does not match any IP for NULL", which was
5400 confusing, to say the least. The bug was that the host name should have
5401 appeared instead of "NULL".
5403 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
5404 user that is built into the binary, and the -C or -D options is used, root
5405 privilege is dropped before the configuration file is read. In addition,
5406 logging is switched to stderr instead of the normal log files. If the
5407 configuration then re-defines the Exim user, the unprivileged environment
5408 is probably not what is expected, so Exim logs a panic warning message (but
5411 However, if deliver_drop_privilege is set, the unprivileged state may well
5412 be exactly what is intended, so the warning has been cut out in that case,
5413 and Exim is allowed to try to write to its normal log files.
5419 1. smtp_return_error_details was not giving details for temporary sender
5420 or receiver verification errors.
5422 2. Diagnose a configuration error if two authenticators have the same public
5425 3. Exim used not to create the message log file for a message until the first
5426 delivery attempt. This could be confusing when incoming messages were held
5427 for policy or load reasons. The message log file is now created at the time
5428 the message is received, and an initial "Received" line is written to it.
5430 4. The automatically generated man page for command line options had a minor
5431 bug that caused no ill effects; however, a more serious problem was that
5432 the procedure for building the man page automatically didn't always
5433 operate. Consequently, release 4.20 contains an out-of-date version. This
5434 shouldn't happen again.
5436 5. When building Exim with embedded Perl support, the script that builds the
5437 Makefile was calling 'perl' to find its compile-time parameters, ignoring
5438 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
5440 6. The freeze_tell option was not being used for messages that were frozen on
5441 arrival, either by an ACL or by local_scan().
5443 7. Added the smtp_incomplete_transaction log selector.
5445 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
5446 was accepting AUTH without a new EHLO.
5448 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
5449 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
5452 10. Logging of TCP/IP connections (when configured) now happens in the main
5453 daemon process instead of the child process, so that the TCP/IP connection
5454 count is more accurate (but it can never be perfect).
5456 11. The use of "drop" in a nested ACL was not being handled correctly in the
5457 outer ACL. Now, if condition failure induced by the nested "drop" causes
5458 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
5459 or "require"), the connection is dropped.
5461 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
5462 that yield "discard" can now be used with an "accept" or a "discard" verb,
5463 but an error is generated for any others (because I can't see a useful way
5464 to define what should happen).
5466 13. When an ACL is read dynamically from a file (or anywhere else), the lines
5467 are now processed in the same way as lines in the Exim configuration file.
5468 In particular, continuation lines are supported.
5470 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
5472 15. Added -ti meaning -t -i.
5474 16. Check for letters, digits, hyphens, and dots in the names of dnslist
5475 domains, and warn by logging if others are found.
5477 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
5478 returned from GIFCONF when Exim is trying to find the list of interfaces on
5479 a host. The code in os.c has been modified to copy each ifreq to an aligned
5480 structure in all cases.
5482 Also, in some cases, the returned ifreq's were being copied to a 'struct
5483 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
5484 means the last couple of bytes of an IPv6 address could be chopped if the
5485 ifreq contained only a normal sockaddr (14 bytes storage).
5487 18. Named domain lists were not supported in the hosts_treat_as_local option.
5488 An entry such as +xxxx was not recognized, and was treated as a literal
5491 19. Ensure that header lines added by a DATA ACL are included in the reject log
5492 if the ACL subsequently rejects the message.
5494 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
5495 MD5 (which is deprecated).
5497 21. When testing a filter file using -bf, Exim was writing a message when it
5498 took the sender from a "From " line in the message, but it was not doing so
5499 when it took $return_path from a Return-Path: header line. It now does.
5501 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
5502 with a valid header line field name (a series of printing characters
5503 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
5505 23. Changed "disc" in the source to "disk" to conform to the documentation and
5506 the book and for uniformity.
5508 24. Ignore Sendmail's -Ooption=value command line item.
5510 25. When execve() failed while trying to run a command in a pipe transport,
5511 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
5512 could be confused with a return value of 69 from the command itself. This
5513 has been changed to 127, the value the shell returns if it is asked to run
5514 a non-existent command. The wording for the related log line suggests a
5515 non-existent command as the problem.
5517 26. If received_header_text expands to an empty string, do not add a Received:
5518 header line to the message. (Well, it adds a token one on the spool, but
5519 marks it "old" so that it doesn't get used or transmitted.)
5521 27. Installed eximstats 1.28 (addition of -nt option).
5523 28. There was no check for failure on the call to getsockname() in the daemon
5524 code. This can fail if there is a shortage of resources on the system, with
5525 ENOMEM, for example. A temporary error is now given on failure.
5527 29. Contrary to the C standard, it seems that in some environments, the
5528 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
5529 program. Exim now does this explicitly; it affects the formatting of
5530 timestamps using strftime().
5532 30. If exiqsumm was given junk data, it threw up some uninitialized variable
5533 complaints. I've now initialized all the variables, to avoid this.
5535 32. Header lines added by a system filter were not being "seen" during
5536 transport-time rewrites.
5538 33. The info_callback() function passed to OpenSSL is set up with type void
5539 (*)(SSL *, int, int), as described somewhere. However, when calling the
5540 function (actually a macro) that sets it up, the type void(*)() is
5541 expected. I've put in a cast to prevent warnings from picky compilers.
5543 34. If a DNS black list lookup found a CNAME record, but there were no A
5544 records associated with the domain it pointed at, Exim crashed.
5546 35. If a DNS black list lookup returned more than one A record, Exim ignored
5547 all but the first. It now scans all returned addresses if a particular IP
5548 value is being sought. In this situation, the contents of the
5549 $dnslist_value variable are a list of all the addresses, separated by a
5552 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
5553 to accept a host name and all its aliases if the forward lookup for any of
5554 them yielded the IP address of the incoming connection. Now it accepts only
5555 those names whose forward lookup yields the correct IP address. Any other
5556 names are discarded. This closes a loophole whereby a rogue DNS
5557 administrator could create reverse DNS records to break through a
5558 wildcarded host restriction in an ACL.
5560 37. If a user filter or a system filter that ran in a subprocess used any of
5561 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
5562 the wrong values were passed to the pipe command ($thisaddress had the
5563 value of $0, $0 had the value of $1, etc). This bug was introduced by
5564 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
5566 38. Improved the line breaking for long SMTP error messages from ACLs.
5567 Previously, if there was no break point between 40 and 75 characters, Exim
5568 left the rest of the message alone. Two changes have been made: (a) I've
5569 reduced the minimum length to 35 characters; (b) if it can't find a break
5570 point between 35 and 75 characters, it looks ahead and uses the first one
5571 that it finds. This may give the occasional overlong line, but at least the
5572 remaining text gets split now.
5574 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
5575 file descriptors might be low, and that setting 1000 would always raise it.
5576 It turns out that in some environments, the limit is already over 1000 and
5577 that lowering it causes trouble. So now Exim takes care not to decrease it.
5579 40. When delivering a message, the value of $return_path is set to $sender_
5580 address at the start of routing (routers may change the value). By an
5581 oversight, this default was not being set up when an address was tested by
5582 -bt or -bv, which affected the outcome if any router or filter referred to
5585 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
5586 log only when "message" or "log_message" are set. However, if one of the
5587 conditions was an address verification, or a call to a nested ACL, the
5588 messages generated by the underlying test were being passed through. This
5589 no longer happens. The underlying message is available in $acl_verify_
5590 message for both "message" and "log_message" expansions, so it can be
5591 passed through if needed.
5593 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
5594 new expansion $bh_ to give the encoded byte string without charset
5595 translation. Translation happens only if iconv() is available; HAVE_ICONV
5596 indicates this at build time. HEADERS_CHARSET gives the charset to
5597 translate to; headers_charset can change it in the configuration, and
5598 "headers charset" can change it in an individual filter file.
5600 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
5601 that creates RFC 2047 encoded "words" labels them as that charset instead
5602 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
5603 expansion operator; (ii) when Exim creates a From: line for a local
5604 message; (iii) when a header line is rewritten to include a "phrase" part.
5606 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
5607 buggy, causing it to skip the first lines of messages whose message ID
5608 ended in 'D'. This would not have bitten before Exim release 4.14, because
5609 message IDs were unlikely to end in 'D' before then. The effect was to have
5610 incorrect size information for certain domains.
5612 45. #include "config.h" was missing at the start of the crypt16.c module. This
5613 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
5616 46. If there was a timeout during a "random" callout check, Exim treated it as
5617 a failure of the random address, and carried on sending RSET and the real
5618 address. If the delay was just some slowness somewhere, the response to the
5619 original RCPT would be taken as a response to RSET and so on, causing
5620 mayhem of various kinds.
5622 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
5623 when I implemented it. It didn't allow for the fact that some option values
5624 may legitimatetly be negative (e.g. size_addition), and it didn't even do
5625 the right test for positive values.
5627 48. Domain names in DNS records are case-independent. Exim always looks them up
5628 in lower case. Some resolvers return domain names in exactly the case they
5629 appear in the zone file, that is, they may contain uppercase letters. Not
5630 all resolvers do this - some return always lower case. Exim was treating a
5631 change of case by a resolver as a change of domain, similar to a widening
5632 of a domain abbreviation. This triggered its re-routing code and so it was
5633 trying to route what was effectively the same domain again. This normally
5634 caused routing to fail (because the router wouldn't handle the domain
5635 twice). Now Exim checks for this case specially, and just changes the
5636 casing of the domain that it ultimately uses when it transmits the message
5639 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
5642 50. If a filter generated a file delivery with a non-absolute name (possible if
5643 no home directory exists for the router), the forbid_file option was not
5646 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
5647 the existing equality matching.
5649 52. Exim was using ints instead of ino_t variables in some places where it was
5650 dealing with inode numbers.
5652 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
5653 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
5654 called TMPDIR, and if it finds it is different, it changes its value.
5656 54. The smtp_printf() function is now made available to local_scan() so
5657 additional output lines can be written before returning. There is also an
5658 smtp_fflush() function to enable the detection of a dropped connection.
5659 The variables smtp_input and smtp_batched_input are exported to
5662 55. Changed the default runtime configuration: the message "Unknown user"
5663 has been removed from the ACL, and instead placed on the localuser router,
5664 using the cannot_route_message feature. This means that any verification
5665 failures that generate their own messages won't get overridden. Similarly,
5666 the "Unrouteable address" message that was in the ACL for unverifiable
5667 relay addresses has also been removed.
5669 56. Added hosts_avoid_esmtp to the smtp transport.
5671 57. The exicyclog script was not checking for the esoteric option
5672 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
5673 will work only if exicyclog is run under the appropriate euid.
5675 58. Following a discussion on the list, the rules by which Exim recognises line
5676 endings on incoming messages have been changed. The -dropcr and drop_cr
5677 options are now no-ops, retained only for backwards compatibility. The
5678 following line terminators are recognized: LF CRLF CR. However, special
5679 processing applies to CR:
5681 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
5682 nor a local message in the state where . is a terminator.
5684 (ii) If a bare CR is encountered in a header line, an extra space is added
5685 after the line terminator so as not to end the header. The reasoning
5686 behind this is that bare CRs in header lines are most likely either
5687 to be mistakes, or people trying to play silly games.
5689 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
5690 was being incorrectly given as 18 bytes larger than it should have been.
5691 This is a VOB (very old bug).
5693 60. This may never have affected anything current, but just in case it has:
5694 When the local host is found other than at the start of a list of hosts,
5695 the local host, those with the same MX, and any that follow, are discarded.
5696 When the list in question was part of a longer list of hosts, the following
5697 hosts (not currently being processed) were also being discarded. This no
5698 longer happens. I'm not sure if this situation could ever has previously
5701 61. Added the "/MX" feature to lists of hosts in the manualroute and query
5704 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
5705 header. This is something that is recommended in a new Internet Draft, and
5706 is something that is documented as being done by Sendmail. There are two
5707 possible values. For messages generated by the autoreply transport, Exim
5710 Auto-Submitted: auto-replied
5712 whereas for all other generated messages (e.g. bounces) it adds
5714 Auto-Submitted: auto-generated
5716 63. The "personal" condition in filters now includes a test for the
5717 Auto-Submitted: header. If it contains the string "auto-" the message it
5718 not considered personal.
5720 64. Added rcpt_include_affixes as a generic transport option.
5722 65. Added queue_only_override (default true).
5724 66. Added the syslog_duplication option.
5726 67. If what should have been the first header line of a message consisted of
5727 a space followed by a colon, Exim was mis-interpreting it as a header line.
5728 It isn't of course - it is syntactically invalid and should therefore be
5729 treated as the start of the message body. The misbehaviour could have
5730 caused a number of strange effects, including loss of data in subsequent
5731 header lines, and spool format errors.
5733 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
5734 client host had authenticated. This control can now be exercised by an ACL
5735 for more flexibility.
5737 69. By default, callouts do not happen when testing with -bh. There is now a
5738 variant, -bhc, which does actually run the callout code, including
5739 consulting and updating the callout cache.
5741 70. Added support for saslauthd authentication, courtesy of Alexander
5744 71. If statvfs() failed on the spool or log directories while checking their
5745 size for availability, Exim confusingly gave the error "space shortage".
5746 Furthermore, in debugging mode it crashed with a floating point exception.
5747 These checks are done if check_{spool,log}_{space,inodes} are set, and when
5748 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
5749 serious problem, Exim now writes to the main and panic logs when this
5750 happens, with details of the failure. It then refuses to accept the
5751 incoming message, giving the message "spool directory problem" or "log
5752 directory problem" with a 421 code for SMTP messages.
5754 72. When Exim is about to re-exec itself, it ensures that the file descriptors
5755 0, 1, and 2 exist, because some OS complain for execs without them (see
5756 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
5757 descriptors. However, the code omitted to check that the open succeeded,
5758 causing mysterious errors if for some reason the permissions on /dev/null
5759 got screwed. Now Exim writes a message to the main and panic logs, and
5760 bombs out if it can't open /dev/null.
5762 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
5763 interact so that it is all more flexible. It is supposed to remain
5764 backwards compatible. Also added extra_local_interfaces.
5766 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
5767 to bomb out with an assertion failure - to the client this appears as a
5768 connection drop. This problem occurs in the part of the code that was taken
5769 from the Samba project. Fortunately, the assertion is in a very simple
5770 function, so I have fixed this by reproducing the function inline in the
5771 one place where it is called, and arranging for authentication to fail
5772 instead of killing the process with assert().
5774 75. The SPA client code was not working when the server requested OEM rather
5775 than Unicode encoding.
5777 76. Added code to make require_files with a specific uid setting more usable in
5778 the case where statting the file as root fails - usually a non-root-mounted
5779 NFS file system. When this happens and the failure is EACCES, Exim now
5780 forks a subprocess and does the per-uid checking as the relevant uid.
5782 77. Added process_log_path.
5784 78. If log_file_path was not explicitly set, a setting of check_log_space or
5785 check_log_inodes was ignored.
5787 79. If a space check for the spool or log partitions fails, the incident is now
5788 logged. Of course, in the latter case the data may get lost...
5790 80. Added the %p formatting code to string_format() so that it can be used to
5791 print addresses in debug_print(). Adjusted all the address printing in the
5792 debugging in store.c to use %p rather than %d.
5794 81. There was a concern that a line of code in smtp_in.c could overflow a
5795 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
5796 initially expressed, the concern was not well-founded, because trailing
5797 spaces are removed early. However, if the trailing spaces were followed by
5798 a NULL, they did not get removed, so the overflow was possible. Two fixes
5801 (a) I re-wrote the offending code in a cleaner fashion.
5802 (b) If an incoming SMTP command contains a NULL character, it is rejected
5805 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
5806 runs initgroups(), so that if the Exim user is in any additional groups,
5807 they will be used during message reception.
5813 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob