1 Change log file for Exim from version 4.21
2 -------------------------------------------
7 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
9 PP/02 Make -n do something, by making it not do something.
10 When combined with -bP, the name of an option is not output.
12 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
15 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
16 $sender_host_name and config options to manage this, and basic check
19 PP/05 DSCP support for outbound connections and control modifier for inbound.
21 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
22 (Only plugin which currently uses this is kerberos4, which nobody should
23 be using, but we should make it available and other future plugins might
24 conceivably use it, even though it would break NAT; stuff *should* be
25 using channel bindings instead).
27 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
28 name; added for Sendmail compatibility; requires admin caller.
29 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
30 just ignore it); requires trusted caller.
31 Also parse but ignore: -Ac -Am -X<logfile>
34 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
36 TL/02 Add +smtp_confirmation as a default logging option.
38 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
39 Patch by Magnus Holmgren from 2007-02-20.
41 TL/04 Bugzilla 1281 - Spec typo.
42 Bugzilla 1283 - Spec typo.
43 Bugzilla 1290 - Spec grammar fixes.
45 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
47 TL/06 Add Experimental DMARC support using libopendmarc libraries.
49 TL/07 Fix an out of order global option causing a segfault. Reported to dev
50 mailing list by by Dmitry Isaikin.
52 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
54 JH/02 Support "G" suffix to numbers in ${if comparisons.
56 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
58 NM/01 Bugzilla 1197 - Spec typo
59 Bugzilla 1196 - Spec examples corrections
61 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
63 PP/09 Add gnutls_enable_pkcs11 option.
65 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
66 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
68 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
69 "acl {{name}{arg}...}", and optional args on acl condition
72 JH/05 Permit multiple router/transport headers_add/remove lines.
74 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
76 JH/07 Avoid using a waiting database for a single-message-only transport.
77 Performance patch from Paul Fisher. Bugzilla 1262.
79 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
82 JH/09 Add $headers_added variable, with content from use of ACL modifier
83 add_header (but not yet added to the message). Bugzilla 199.
85 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
86 Pulled from Bugzilla 817 by Wolfgang Breyha.
88 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
91 JH/11 Add A= logging on delivery lines, and a client_set_id option on
94 JH/12 Add optional authenticated_sender logging to A= and a log_selector
97 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
99 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
100 advertise SMTP AUTH mechanism to us, instead of a generic
101 protocol violation error. Also, make Exim more robust to bad
102 data from the Dovecot auth socket.
104 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
106 When a queue runner is handling a message, Exim first routes the
107 recipient addresses, during which it prunes them based on the retry
108 hints database. After that it attempts to deliver the message to
109 any remaining recipients. It then updates the hints database using
112 So if a recipient address works intermittently, it can get repeatedly
113 deferred at routing time. The retry hints record remains fresh so the
114 address never reaches the final cutoff time.
116 This is a fairly common occurrence when a user is bumping up against
117 their storage quota. Exim had some logic in its local delivery code
118 to deal with this. However it did not apply to per-recipient defers
119 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
121 This change adds a proper retry rule check during routing so that the
122 final cutoff time is checked against the message's age. We only do
123 this check if there is an address retry record and there is not a
124 domain retry record; this implies that previous attempts to handle
125 the address had the retry_use_local_parts option turned on. We use
126 this as an approximation for the destination being like a local
127 delivery, as in LMTP.
129 I suspect this new check makes the old local delivery cutoff check
130 redundant, but I have not verified this so I left the code in place.
132 TF/02 Correct gecos expansion when From: is a prefix of the username.
134 Test 0254 submits a message to Exim with the header
138 When I ran the test suite under the user fanf2, Exim expanded
139 the header to contain my full name, whereas it should have added
140 a Resent-Sender: header. It erroneously treats any prefix of the
141 username as equal to the username.
143 This change corrects that bug.
145 GF/01 DCC debug and logging tidyup
146 Error conditions log to paniclog rather than rejectlog.
147 Debug lines prefixed by "DCC: " to remove any ambiguity.
149 TF/03 Avoid unnecessary rebuilds of lookup-related code.
151 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
152 Bug spotted by Jeremy Harris; was flawed since initial commit.
153 Would have resulted in OCSP responses post-SNI triggering an Exim
154 NULL dereference and crash.
156 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
158 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
159 Bug detection, analysis and fix by Samuel Thibault.
160 Bugzilla 1331, Debian bug #698092.
162 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
164 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
165 Server implementation by Todd Lyons, client by JH.
166 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
167 config variable "prdr_enable" controls whether the server
168 advertises the facility. If the client requests PRDR a new
169 acl_data_smtp_prdr ACL is called once for each recipient, after
170 the body content is received and before the acl_smtp_data ACL.
171 The client is controlled by bolth of: a hosts_try_prdr option
172 on the smtp transport, and the server advertisement.
173 Default client logging of deliveries and rejections involving
174 PRDR are flagged with the string "PRDR".
176 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
177 fclose(). Diagnosis by Todd Lyons.
179 PP/17 Update configure.default to handle IPv6 localhost better.
180 Patch by Alain Williams (plus minor tweaks).
183 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
184 This is now consistent with GnuTLS, and is now documented: the
185 previous undocumented portable approach to treating the option as
186 unset was to force an expansion failure. That still works, and
187 an empty string is now equivalent.
189 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
190 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
191 not performing validation itself.
193 PP/20 Added force_command boolean option to pipe transport.
194 Patch from Nick Koston, of cPanel Inc.
196 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
199 PP/21 Fix eximon continuous updating with timestamped log-files.
200 Broken in a format-string cleanup in 4.80, missed when I repaired the
201 other false fix of the same issue.
202 Report and fix from Heiko Schlichting.
209 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
211 This, or similar/improved, will also be change PP/11 of 4.82.
217 PP/01 Handle short writes when writing local log-files.
218 In practice, only affects FreeBSD (8 onwards).
219 Bugzilla 1053, with thanks to Dmitry Isaikin.
221 NM/01 Bugzilla 949 - Documentation tweak
223 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
226 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
228 PP/02 Implemented gsasl authenticator.
230 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
232 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
233 `pkg-config foo` for cflags/libs.
235 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
236 with rest of GSASL and with heimdal_gssapi.
238 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
239 `pkg-config foo` for cflags/libs for the TLS implementation.
241 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
242 properties get this fed in as external SSF. A number of robustness
243 and debugging improvements to the cyrus_sasl authenticator.
245 PP/08 cyrus_sasl server now expands the server_realm option.
247 PP/09 Bugzilla 1214 - Log authentication information in reject log.
248 Patch by Jeremy Harris.
250 PP/10 Added dbmjz lookup type.
252 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
254 PP/12 MAIL args handles TAB as well as SP, for better interop with
255 non-compliant senders.
256 Analysis and variant patch by Todd Lyons.
258 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
259 Bug report from Lars Müller <lars@samba.org> (via SUSE),
260 Patch from Dirk Mueller <dmueller@suse.com>
262 PP/13 tls_peerdn now print-escaped for spool files.
263 Observed some $tls_peerdn in wild which contained \n, which resulted
264 in spool file corruption.
266 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
267 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
268 or write after TLS renegotiation, which otherwise led to messages
271 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
272 as a tracking header (ie: a signed header comes before the signature).
273 Patch from Wolfgang Breyha.
275 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
276 comma-sep list; embedded commas doubled.
278 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
280 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
282 Report and patch from Dmitry Banschikov.
284 PP/16 Removed "dont_insert_empty_fragments" fron "openssl_options".
285 Removed SSL_clear() after SSL_new() which led to protocol negotiation
286 failures. We appear to now support TLS1.1+ with Exim.
288 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
289 lets Exim select keys and certificates based upon TLS SNI from client.
290 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
291 before an outbound SMTP session. New log_selector, +tls_sni.
293 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
294 NULL dereference. Report and patch from Alun Jones.
296 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
297 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
298 Not seeing resolver debug output on NetBSD, but suspect this is a
299 resolver implementation change.
301 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
302 Left warnings. Added "eximon gdb" invocation mode.
304 PP/21 Defaulting "accept_8bitmime" to true, not false.
306 PP/22 Added -bw for inetd wait mode support.
308 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
309 locate the relevant includes and libraries. Made this the default.
311 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
312 Bugzilla 1246, report and most of solution from Tomasz Kusy.
314 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
315 This may cause build issues on older platforms.
317 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
318 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
319 gnutls_require_mac & gnutls_require_protocols (no longer supported).
320 Added SNI support via GnuTLS too.
321 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
323 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
325 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
326 Applied second patch from Janne, implementing suggestion to default
327 multiple-strings-in-record handling to match SPF spec.
329 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
331 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
332 read-only, out of scope).
333 Patch from Wolfgang Breyha, report from Stuart Northfield.
335 PP/29 Fix three issues highlighted by clang analyser static analysis.
336 Only crash-plausible issue would require the Cambridge-specific
337 iplookup router and a misconfiguration.
338 Report from Marcin Mirosław.
340 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
342 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
343 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
344 As part of this, removing so much warning spew let me fix some minor
345 real issues in debug logging.
347 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
348 assignment on my part. Fixed.
350 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
351 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
352 Janne Snabb (who went above and beyond: thank you).
354 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
355 string otherwise requires a connection and a bunch more work and it's
356 relatively easy to get wrong. Should also expose TLS library linkage
359 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
360 64-bit ${eval} (JH/03).
362 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
363 GNU libc to support some of the 64-bit stuff, should not lead to
364 conflicts. Defined before os.h is pulled in, so if a given platform
365 needs to override this, it can.
367 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
368 protection layer was required, which is not implemented.
369 Bugzilla 1254, patch from Wolfgang Breyha.
371 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
372 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
373 tls_dhparam take prime identifiers. Also unbreak combination of
374 OpenSSL+DH_params+TLSSNI.
376 PP/39 Disable SSLv2 by default in OpenSSL support.
382 PP/01 Solaris build fix for Oracle's LDAP libraries.
383 Bugzilla 1109, patch from Stephen Usher.
385 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
387 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
390 TF/02 Fix a couple more cases where we did not log the error message
391 when unlink() failed. See also change 4.74-TF/03.
393 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
394 lock up or crash if it happened to be inside a call to libc when it
395 got a SIGUSR1 from exiwhat.
397 The SIGUSR1 handler appends the current process status to the process
398 log which is later printed by exiwhat. It used to use the general
399 purpose logging code to do this, but several functions it calls are
400 not safe for signals.
402 The new output code in the SIGUSR1 handler is specific to the process
403 log, and simple enough that it's easy to inspect for signal safety.
404 Removing some special cases also simplifies the general logging code.
405 Removing the spurious timestamps from the process log simplifies
408 TF/04 Improved ratelimit ACL condition.
410 The /noupdate option has been deprecated in favour of /readonly which
411 has clearer semantics. The /leaky, /strict, and /readonly update modes
412 are mutually exclusive. The update mode is no longer included in the
413 database key; it just determines when the database is updated. (This
414 means that when you upgrde Exim will forget old rate measurements.)
416 Exim now checks that the per_* options are used with an update mode that
417 makes sense for the current ACL. For example, when Exim is processing a
418 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
419 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
420 must specify per_mail/readonly. If you omit the update mode it defaults to
421 /leaky where that makes sense (as before) or /readonly where required.
423 The /noupdate option is now undocumented but still supported for
424 backwards compatibility. It is equivalent to /readonly except that in
425 ACLs where /readonly is required you may specify /leaky/noupdate or
426 /strict/noupdate which are treated the same as /readonly.
428 A useful new feature is the /count= option. This is a generalization
429 of the per_byte option, so that you can measure the throughput of other
430 aggregate values. For example, the per_byte option is now equivalent
431 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
433 The per_rcpt option has been generalized using the /count= mechanism
434 (though it's more complicated than the per_byte equivalence). When it is
435 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
436 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
437 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
438 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
439 Note that using per_rcpt with a non-readonly update mode in more than
440 one ACL will cause the recipients to be double-counted. (The per_mail
441 and per_byte options don't have this problem.)
443 The handling of very low rates has changed slightly. If the computed rate
444 is less than the event's count (usually one) then this event is the first
445 after a long gap. In this case the rate is set to the same as this event's
446 count, so that the first message of a spam run is counted properly.
448 The major new feature is a mechanism for counting the rate of unique
449 events. The new per_addr option counts the number of different
450 recipients that someone has sent messages to in the last time period. It
451 behaves like per_rcpt if all the recipient addresses are different, but
452 duplicate recipient addresses do not increase the measured rate. Like
453 the /count= option this is a general mechanism, so the per_addr option
454 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
455 example, measure the rate that a client uses different sender addresses
456 with the options per_mail/unique=$sender_address. There are further
457 details in the main documentation.
459 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
461 TF/06 Removed a few PCRE remnants.
463 TF/07 Automatically extract Exim's version number from tags in the git
464 repository when doing development or release builds.
466 PP/02 Raise smtp_cmd_buffer_size to 16kB.
467 Bugzilla 879. Patch from Paul Fisher.
469 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
470 Heavily based on revision 40f9a89a from Simon Arlott's tree.
473 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
475 PP/05 Variable $av_failed, true if the AV scanner deferred.
476 Bugzilla 1078. Patch from John Horne.
478 PP/06 Stop make process more reliably on build failure.
479 Bugzilla 1087. Patch from Heiko Schlittermann.
481 PP/07 Make maildir_use_size_file an _expandable_ boolean.
482 Bugzilla 1089. Patch from Heiko Schlittermann.
484 PP/08 Handle ${run} returning more data than OS pipe buffer size.
485 Bugzilla 1131. Patch from Holger Weiß.
487 PP/09 Handle IPv6 addresses with SPF.
488 Bugzilla 860. Patch from Wolfgang Breyha.
490 PP/10 GnuTLS: support TLS 1.2 & 1.1.
492 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
495 PP/11 match_* no longer expand right-hand-side by default.
496 New compile-time build option, EXPAND_LISTMATCH_RHS.
497 New expansion conditions, "inlist", "inlisti".
499 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
501 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
503 PP/14 fix log_write() format string regression from TF/03.
504 Bugzilla 1152. Patch from Dmitry Isaikin.
510 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
512 PP/02 Harmonised TLS library version reporting; only show if debugging.
513 Layout now matches that introduced for other libraries in 4.74 PP/03.
515 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
517 PP/04 New "dns_use_edns0" global option.
519 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
522 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
523 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
525 TK/01 Updated PolarSSL code to 0.14.2.
526 Bugzilla 1097. Patch from Andreas Metzler.
528 PP/07 Catch divide-by-zero in ${eval:...}.
531 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
534 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
535 format-string attack -- SECURITY: remote arbitrary code execution.
537 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
538 time unintentionally subject to list matching rules, letting the header
539 cause arbitrary Exim lookups (of items which can occur in lists, *not*
540 arbitrary string expansion). This allowed for information disclosure.
542 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
543 INT_MIN/-1 -- value coerced to INT_MAX.
549 NM/01 Workround for PCRE version dependency in version reporting
552 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
553 This fixes portability to compilers other than gcc, notably
554 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
556 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
557 makefiles for portability to HP-UX and POSIX correctness.
559 PP/01 Permit LOOKUP_foo enabling on the make command-line.
560 Also via indented variable definition in the Makefile.
561 (Debugging by Oliver Heesakkers).
563 PP/02 Restore caching of spamd results with expanded spamd_address.
564 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
566 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
567 Improves build reliability. Fix from: Frank Elsner
569 NM/02 Fix wide character breakage in the rfc2047 coding
570 Fixes bug 1064. Patch from Andrey N. Oktyabrski
572 NM/03 Allow underscore in dnslist lookups
573 Fixes bug 1026. Patch from Graeme Fowler
575 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
576 Code patches from Adam Ciarcinski of NetBSD.
578 NM/04 Fixed exiqgrep to cope with mailq missing size issue
581 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
582 is logged, to avoid truncation. Patch from John Horne.
584 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
585 Patch from Jakob Hirsch.
587 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
588 SQL string expansion failure details.
589 Patch from Andrey Oktyabrski.
591 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
592 Patch from Simon Arlott.
594 PP/09 New lookups functionality failed to compile on old gcc which rejects
595 extern declarations in function scope.
596 Patch from Oliver Fleischmann
598 PP/10 Use sig_atomic_t for flags set from signal handlers.
599 Check getgroups() return and improve debugging.
600 Fixed developed for diagnosis in bug 927 (which turned out to be
603 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
604 Patch from Mark Zealey.
606 PP/12 Bugzilla 1056: Improved spamd server selection.
607 Patch from Mark Zealey.
609 PP/13 Bugzilla 1086: Deal with maildir quota file races.
610 Based on patch from Heiko Schlittermann.
612 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
613 Patch from Uwe Doering, sign-off by Michael Haardt.
615 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
616 variable declaration deep within a block. Bug and patch from
619 PP/15 lookups-Makefile IRIX compatibilty coercion.
621 PP/16 Make DISABLE_DKIM build knob functional.
623 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
624 Patch by Simon Arlott
626 TF/03 Fix valgrind.h portability to C89 compilers that do not support
627 variable argument macros. Our copy now differs from upstream.
633 TF/01 Failure to get a lock on a hints database can have serious
634 consequences so log it to the panic log.
636 TF/02 Log LMTP confirmation messages in the same way as SMTP,
637 controlled using the smtp_confirmation log selector.
639 TF/03 Include the error message when we fail to unlink a spool file.
641 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
642 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
643 for maintaining out-of-tree patches for some time.
645 PP/01 Bugzilla 139: Documentation and portability issues.
646 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
647 Handle per-OS dynamic-module compilation flags.
649 PP/02 Let /dev/null have normal permissions.
650 The 4.73 fixes were a little too stringent and complained about the
651 permissions on /dev/null. Exempt it from some checks.
652 Reported by Andreas M. Kirchwitz.
654 PP/03 Report version information for many libraries, including
655 Exim version information for dynamically loaded libraries. Created
656 version.h, now support a version extension string for distributors
657 who patch heavily. Dynamic module ABI change.
659 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
660 privilege escalation vulnerability whereby the Exim run-time user
661 can cause root to append content of the attacker's choosing to
664 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
667 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
668 If dropping privileges for untrusted macros, we disabled normal logging
669 on the basis that it would fail; for the Exim run-time user, this is not
670 the case, and it resulted in successful deliveries going unlogged.
671 Fixed. Reported by Andreas Metzler.
677 PP/01 Date: & Message-Id: revert to normally being appended to a message,
678 only prepend for the Resent-* case. Fixes regression introduced in
679 Exim 4.70 by NM/22 for Bugzilla 607.
681 PP/02 Include check_rfc2047_length in configure.default because we're seeing
682 increasing numbers of administrators be bitten by this.
684 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
686 PP/03 Bugzilla 994: added openssl_options main configuration option.
688 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
690 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
692 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
694 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
695 without a peer certificate, leading to a segfault because of an
696 assumption that peers always have certificates. Be a little more
697 paranoid. Problem reported by Martin Tscholak.
699 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
700 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
701 NB: ClamAV planning to remove STREAM in "middle of 2010".
702 CL also introduces -bmalware, various -d+acl logging additions and
703 more caution in buffer sizes.
705 PP/09 Implemented reverse_ip expansion operator.
707 PP/10 Bugzilla 937: provide a "debug" ACL control.
709 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
711 PP/12 Bugzilla 973: Implement --version.
713 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
715 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
717 PP/15 Bugzilla 816: support multiple condition rules on Routers.
719 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
720 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
721 ignore trailing whitespace.
723 JJ/02 prevent non-panic DKIM error from being sent to paniclog
725 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
728 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
729 Notification from Dr Andrew Aitchison.
731 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
732 ExtendedDetectionInfo response format.
733 Notification from John Horne.
735 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
738 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
739 XSL and documented dependency on system catalogs, with examples of how
742 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
745 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
746 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
747 configuration file which is writeable by the Exim user or group.
749 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
750 of configuration files to cover files specified with the -C option if
751 they are going to be used with root privileges, not just the default
754 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
755 option (effectively making it always true).
757 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
758 files to be used while preserving root privileges.
760 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
761 that rogue child processes cannot use them.
763 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
764 run-time user, instead of root.
766 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
767 Exim run-time user without dropping privileges.
769 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
770 result string, instead of calling string_vformat() twice with the same
773 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
774 for other users. Others should always drop root privileges if they use
775 -C on the command line, even for a whitelisted configure file.
777 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
779 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
785 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
786 $data_path, and $header_path variables; fixed documentation bugs and
789 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
790 exipick to access non-standard spools, including the "frozen" queue
793 NM/01 Bugzilla 965: Support mysql stored procedures.
794 Patch from Alain Williams
796 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
798 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
799 Patch from Andreas Metzler
801 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
802 Patch from Kirill Miazine
804 NM/05 Bugzilla 671: Added umask to procmail example.
806 JJ/03 installed exipick 20100323.0, fixing doc bug
808 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
809 directory. Notification and patch from Dan Rosenberg.
811 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
813 TK/02 Improve log output when DKIM signing operation fails.
815 MH/01 Treat the transport option dkim_domain as a colon separated
816 list, not as a single string, and sign the message with each element,
817 omitting multiple occurences of the same signer.
819 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
820 Bugzilla 985, 986. Patch by Simon Arlott
822 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
823 Patch by Simon Arlott
825 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
826 MBX locking. Notification from Dan Rosenberg.
832 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
834 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
836 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
838 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
840 NM/04 Bugzilla 915: Flag broken perl installation during build.
846 TK/01 Added patch by Johannes Berg that expands the main option
847 "spamd_address" if it starts with a dollar sign.
849 TK/02 Write list of recipients to X-Envelope-Sender header when building
850 the mbox-format spool file for content scanning (suggested by Jakob
853 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
854 (http://www.dcc-servers.net/) support via dccifd. Activated by
855 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
857 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
858 by Mark Daniel Reidel <mr@df.eu>.
860 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
861 When building exim an external PCRE library is now needed -
862 PCRE is a system library on the majority of modern systems.
863 See entry on PCRE_LIBS in EDITME file.
865 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
866 conversation. Added nologin parameter to request.
867 Patch contributed by Kirill Miazine.
869 TF/01 Do not log submission mode rewrites if they do not change the address.
871 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
873 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
874 log files in place. Contributed by Roberto Lima.
876 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
878 TF/03 Bugzilla 615: When checking the local_parts router precondition
879 after a local_part_suffix or local_part_prefix option, Exim now
880 does not use the address's named list lookup cache, since this
881 contains cached lookups for the whole local part.
883 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
884 Robert Millan. Documentation is in experimental-spec.txt.
886 TF/04 Bugzilla 668: Fix parallel build (make -j).
888 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
890 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
891 Patch provided by Jan Srzednicki.
893 TF/05 Leading white space used to be stripped from $spam_report which
894 wrecked the formatting. Now it is preserved.
896 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
897 that they are available at delivery time.
899 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
901 TF/08 TLS error reporting now respects the incoming_interface and
902 incoming_port log selectors.
904 TF/09 Produce a more useful error message if an SMTP transport's hosts
905 setting expands to an empty string.
907 NM/06 Bugzilla 744: EXPN did not work under TLS.
908 Patch provided by Phil Pennock.
910 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
911 Patch provided by Richard Godbee.
913 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
914 acl_smtp_notquit, added index entry.
916 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
917 Patch provided by Eugene Bujak.
919 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
920 accept(). Patch provided by Maxim Dounin.
922 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
923 Patch provided by Phil Pennock.
925 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
927 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
928 Patch provided by Brad "anomie" Jorsch.
930 NM/14 Bugzilla 622: Added timeout setting to transport filter.
931 Patch provided by Dean Brooks.
933 TK/05 Add native DKIM support (does not depend on external libraries).
935 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
936 Patch provided by Graeme Fowler.
938 NM/16 Bugzilla 851: Documentation example syntax fix.
940 NM/17 Changed NOTICE file to remove references to embedded PCRE.
942 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
945 NM/19 Bugzilla 745: TLS version reporting.
946 Patch provided by Phil Pennock.
948 NM/20 Bugzilla 167: bool: condition support.
949 Patch provided by Phil Pennock.
951 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
952 clients. Patch provided by Phil Pennock.
954 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
955 Patch provided by Brad "anomie" Jorsch.
957 NM/23 Bugzilla 687: Fix misparses in eximstats.
958 Patch provided by Heiko Schlittermann.
960 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
961 Patch provided by Heiko Schlittermann.
963 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
964 plus update to original patch.
966 NM/26 Bugzilla 799: Documentation correction for ratelimit.
968 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
969 Patch provided by David Brownlee.
971 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
973 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
975 NM/30 Bugzilla 888: TLS documentation bugfixes.
977 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
979 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
980 Unlike the original bugzilla I have changed all shell scripts in src tree.
982 NM/33 Bugzilla 898: Transport filter timeout fix.
983 Patch by Todd Rinaldo.
985 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
986 Patch by Serge Demonchaux.
988 NM/35 Bugzilla 39: Base64 decode bug fixes.
989 Patch by Jakob Hirsch.
991 NM/36 Bugzilla 909: Correct connect() call in dcc code.
993 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
995 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
997 NM/39 Bugzilla 911: Fixed MakeLinks build script.
1003 TK/01 Add preliminary DKIM support. Currently requires a forked version of
1004 ALT-N's libdkim that I have put here:
1005 http://duncanthrax.net/exim-experimental/
1007 Note to Michael Haardt: I had to rename some vars in sieve.c. They
1008 were called 'true' and it seems that C99 defines that as a reserved
1009 keyword to be used with 'bool' variable types. That means you could
1010 not include C99-style headers which use bools without triggering
1011 build errors in sieve.c.
1013 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
1014 as mailq or other aliases. Changed the --help handling significantly
1015 to do whats expected. exim_usage() emits usage/help information.
1017 SC/01 Added the -bylocaldomain option to eximstats.
1019 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
1021 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
1023 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
1029 PH/01 Another patch from the Sieve maintainer.
1031 PH/02 When an IPv6 address is converted to a string for single-key lookup
1032 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
1033 dots are used instead of colons so that keys in lsearch files need not
1034 contain colons. This was done some time before quoting was made available
1035 in lsearch files. However, iplsearch files do require colons in IPv6 keys
1036 (notated using the quote facility) so as to distinguish them from IPv4
1037 keys. This meant that lookups for IP addresses in host lists did not work
1038 for iplsearch lookups.
1040 This has been fixed by arranging for IPv6 addresses to be expressed with
1041 colons if the lookup type is iplsearch. This is not incompatible, because
1042 previously such lookups could never work.
1044 The situation is now rather anomolous, since one *can* have colons in
1045 ordinary lsearch keys. However, making the change in all cases is
1046 incompatible and would probably break a number of configurations.
1048 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
1051 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
1052 conversion specification without a maximum field width, thereby enabling
1053 a rogue spamd server to cause a buffer overflow. While nobody in their
1054 right mind would setup Exim to query an untrusted spamd server, an
1055 attacker that gains access to a server running spamd could potentially
1056 exploit this vulnerability to run arbitrary code as the Exim user.
1058 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
1059 $primary_hostname instead of what libspf2 thinks the hosts name is.
1061 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
1062 a directory entry by the name of the lookup key. Previously, if a
1063 symlink pointed to a non-existing file or a file in a directory that
1064 Exim lacked permissions to read, a lookup for a key matching that
1065 symlink would fail. Now it is enough that a matching directory entry
1066 exists, symlink or not. (Bugzilla 503.)
1068 PH/03 The body_linecount and body_zerocount variables are now exported in the
1071 PH/04 Added the $dnslist_matched variable.
1073 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
1074 This means they are set thereafter only if the connection becomes
1077 PH/06 Added the client_condition to authenticators so that some can be skipped
1078 by clients under certain conditions.
1080 PH/07 The error message for a badly-placed control=no_multiline_responses left
1081 "_responses" off the end of the name.
1083 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
1085 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
1086 (without spaces) instead of just copying the configuration text.
1088 PH/10 Added the /noupdate option to the ratelimit ACL condition.
1090 PH/11 Added $max_received_linelength.
1092 PH/12 Added +ignore_defer and +include_defer to host lists.
1094 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
1095 way in which PCRE > 7.0 is built.
1097 PH/14 Implemented queue_only_load_latch.
1099 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
1100 MAIL command. The effect was to mangle the value on 64-bit systems.
1102 PH/16 Another patch from the Sieve maintainer.
1104 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
1106 PH/18 If a system quota error occurred while trying to create the file for
1107 a maildir delivery, the message "Mailbox is full" was not appended to the
1108 bounce if the delivery eventually timed out. Change 4.67/27 below applied
1109 only to a quota excession during the actual writing of the file.
1111 PH/19 It seems that peer DN values may contain newlines (and other non-printing
1112 characters?) which causes problems in log lines. The DN values are now
1113 passed through string_printing() before being added to log lines.
1115 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
1116 and InterBase are left for another time.)
1118 PH/21 Added message_body_newlines option.
1120 PH/22 Guard against possible overflow in moan_check_errorcopy().
1122 PH/23 POSIX allows open() to be a macro; guard against that.
1124 PH/24 If the recipient of an error message contained an @ in the local part
1125 (suitably quoted, of course), incorrect values were put in $domain and
1126 $local_part during the evaluation of errors_copy.
1132 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
1133 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
1136 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
1137 issue a MAIL command.
1139 PH/02 In an ACL statement such as
1141 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
1143 if a client was not listed at all, or was listed with a value other than
1144 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
1145 the condition was not true (as it should be), so access was not denied.
1146 The bug was that the ! inversion was incorrectly passed on to the second
1147 item. This has been fixed.
1149 PH/03 Added additional dnslists conditions == and =& which are different from
1150 = and & when the dns lookup returns more than one IP address.
1152 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
1153 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
1155 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
1156 FSYNC, which compiles an option called disable_fsync that allows for
1157 bypassing fsync(). The documentation is heavily laced with warnings.
1159 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
1161 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
1162 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
1163 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
1164 including adding "make clean"; (3) Added -fPIC when compiling the test
1165 dynamically loaded module, to get rid of a warning.
1167 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
1168 message fails, move_frozen_messages = true and ignore_bounce_errors_after
1169 = 0s. The bug is otherwise harmless.
1171 PH/07 There was a bug in the dovecot authenticator such that the value of
1172 $auth1 could be overwritten, and so not correctly preserved, after a
1173 successful authentication. This usually meant that the value preserved by
1174 the server_setid option was incorrect.
1176 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
1178 PH/09 Installed PCRE release 7.0.
1180 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
1181 run for batched SMTP input. It is now run at the start of every message
1182 in the batch. While fixing this I discovered that the process information
1183 (output by running exiwhat) was not always getting set for -bs and -bS
1184 input. This is fixed, and it now also says "batched" for BSMTP.
1186 PH/11 Added control=no_pipelining.
1188 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
1189 patch, slightly modified), and move the expansion of helo_data till after
1190 the connection is made in the smtp transport (so it can use these
1193 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
1195 PH/14 Added log_selector = +pid.
1197 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
1199 PH/16 Add ${if forany and ${if forall.
1201 PH/17 Added dsn_from option to vary the From: line in DSNs.
1203 PH/18 Flush SMTP output before performing a callout, unless control =
1204 no_callout_flush is set.
1206 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
1207 was true (the default) a successful delivery failed to delete the retry
1208 item, thus causing premature timeout of the address. The bug is now
1211 PH/20 Added hosts_avoid_pipelining to the smtp transport.
1213 PH/21 Long custom messages for fakedefer and fakereject are now split up
1214 into multiline reponses in the same way that messages for "deny" and
1215 other ACL rejections are.
1217 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
1218 with slight modification.
1220 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
1221 draft, changing the syntax and factoring some duplicate code".
1223 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
1224 for deliveries of the second and subsequent messages over the same SMTP
1227 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
1228 ${reduce, with only minor "tidies".
1230 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
1232 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
1233 expansion side effects.
1235 PH/27 When a message times out after an over-quota error from an Exim-imposed
1236 quota, the bounce message says "mailbox is full". This message was not
1237 being given when it was a system quota that was exceeded. It now should
1240 MH/03 Made $recipients available in local_scan(). local_scan() already has
1241 better access to the recipient list through recipients_list[], but
1242 $recipients can be useful in postmaster-provided expansion strings.
1244 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
1245 in the case of a MAIL command with additional options following the
1246 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
1247 were accidentally chopped off.
1249 PH/29 SMTP synchronization checks are implemented when a command is read -
1250 there is a check that no more input is waiting when there shouldn't be
1251 any. However, for some commands, a delay in an ACL can mean that it is
1252 some time before the response is written. In this time, more input might
1253 arrive, invalidly. So now there are extra checks after an ACL has run for
1254 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
1255 pipelining has not been advertised.
1257 PH/30 MH's patch to allow iscntrl() characters to be list separators.
1259 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
1260 returned in the SMTP response when smtp_return_error_details was false.
1261 This has been fixed.
1263 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
1264 instead of the C I/O that was originally supplied, because problems were
1265 reported on Solaris.
1267 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
1268 Exim which did not show up earlier: it was assuming that a call to
1269 SSL_CTX_set_info_callback() might give an error value. In fact, there is
1270 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
1271 was a macro that became an assignment, so it seemed to work. This has
1272 changed to a proper function call with a void return, hence the compile
1273 error. Exim's code has been fixed.
1275 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
1278 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
1280 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
1282 PH/37 If a message is not accepted after it has had an id assigned (e.g.
1283 because it turns out to be too big or there is a timeout) there is no
1284 "Completed" line in the log. When some messages of this type were
1285 selected by exigrep, they were listed as "not completed". Others were
1286 picked up by some special patterns. I have improved the selection
1287 criteria to be more general.
1289 PH/38 The host_find_failed option in the manualroute router can now be set
1290 to "ignore", to completely ignore a host whose IP address cannot be
1291 found. If all hosts are ignored, the behaviour is controlled by the new
1292 host_all_ignored option.
1294 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
1295 homing or because of multiple MX records with /mx) generated more than
1296 one IP address, and the following item turned out to be the local host,
1297 all the secondary addresses of the first item were incorrectly removed
1298 from the list, along with the local host and any following hosts (which
1299 is what is supposed to happen).
1301 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
1302 whoever called Exim into the -H file. In the case of the daemon it was
1303 behaving confusingly. When first started, it used values for whoever
1304 started the daemon, but after a SIGHUP it used the Exim user (because it
1305 calls itself on a restart). I have changed the code so that it now always
1308 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
1309 message are rejected with the same error (e.g. no authentication or bad
1310 sender address), and a DATA command is nevertheless sent (as can happen
1311 with PIPELINING or a stupid MUA), the error message that was given to the
1312 RCPT commands is included in the rejection of the DATA command. This is
1313 intended to be helpful for MUAs that show only the final error to their
1316 PH/42 Another patch from the Sieve maintainer.
1318 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
1319 Eximstats - Fixed some broken HTML links and added missing column headers
1321 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
1322 columns for Rejects, Temp Rejects, Ham, and Spam rows.
1324 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
1326 PH/43 Yet another patch from the Sieve maintainer.
1328 PH/44 I found a way to check for a TCP/IP connection going away before sending
1329 the response to the final '.' that terminates a message, but only in the
1330 case where the client has not sent further data following the '.'
1331 (unfortunately, this is allowed). However, in many cases there won't be
1332 any further data because there won't be any more messages to send. A call
1333 to select() can be used: if it shows that the input is "ready", there is
1334 either input waiting, or the socket has been closed. An attempt to read
1335 the next input character can distinguish the two cases. Previously, Exim
1336 would have sent an OK response which the client would never have see.
1337 This could lead to message repetition. This fix should cure that, at
1338 least in a lot of common cases.
1340 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
1341 advertised in response to EHLO.
1347 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
1348 fixed by 4.65/MH/01 (is this a record?) are fixed:
1350 (i) An empty string was always treated as zero by the numeric comparison
1351 operators. This behaviour has been restored.
1353 (ii) It is documented that the numeric comparison operators always treat
1354 their arguments as decimal numbers. This was broken in that numbers
1355 starting with 0 were being interpreted as octal.
1357 While fixing these problems I realized that there was another issue that
1358 hadn't been noticed. Values of message_size_limit (both the global option
1359 and the transport option) were treated as octal if they started with 0.
1360 The documentation was vague. These values are now always treated as
1361 decimal, and I will make that clear in the documentation.
1367 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
1368 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
1371 MH/01 Don't check that the operands of numeric comparison operators are
1372 integers when their expansion is in "skipping" mode (fixes bug
1373 introduced by 4.64-PH/07).
1375 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
1376 child addresses, Exim now panics and dies. Previously, because the count
1377 is held in a short int, deliveries were likely to be lost. As such a
1378 large number of recipients for a single message is ridiculous
1379 (performance will be very, very poor), I have chosen to impose a limit
1380 rather than extend the field.
1386 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
1387 leftover -K file (the existence of which was triggered by #402).
1388 While we were at it, introduced process PID as part of the -K
1389 filename. This should rule out race conditions when creating
1392 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
1393 processing considerably. Previous code took too long for large mails,
1394 triggering a timeout which in turn triggers #401.
1396 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
1397 in the DK code in transports.c. sendfile() is not really portable,
1398 hence the _LINUX specificness.
1400 TF/01 In the add_headers option to the mail command in an Exim filter,
1401 there was a bug that Exim would claim a syntax error in any
1402 header after the first one which had an odd number of characters
1405 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
1406 callout verification, Exim cached a "reject" for the entire domain. This
1407 is correct for most verifications, but it is not correct for a recipient
1408 verification with use_sender or use_postmaster set, because in that case
1409 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
1410 case of MAIL FROM:<> rejection from other early rejections (e.g.
1411 rejection of HELO). When verifying a recipient using a non-null MAIL
1412 address, the cache is ignored if it shows MAIL FROM:<> rejection.
1413 Whatever the result of the callout, the value of the domain cache is
1414 left unchanged (for any other kind of callout, getting as far as trying
1415 RCPT means that the domain itself is ok).
1417 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
1420 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
1421 manifest itself as EPIPE rather than ECONNECT. When tidying away a
1422 session, the daemon ignores ECONNECT errors and logs others; it now
1423 ignores EPIPE as well.
1425 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
1426 (quoted-printable decoding).
1428 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
1429 later the small subsequent patch to fix an introduced bug.
1431 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
1433 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
1435 PH/08 An error is now given if message_size_limit is specified negative.
1437 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
1438 to be given (somewhat) arbitrary names.
1440 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
1443 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
1444 miscellaneous code fixes
1446 PH/10 Added the log_reject_target ACL modifier to specify where to log
1449 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
1450 hostname. This is wrong, because it relates to the incoming message (and
1451 probably the interface on which it is arriving) and not to the outgoing
1452 callout (which could be using a different interface). This has been
1453 changed to use the value of the helo_data option from the smtp transport
1454 instead - this is what is used when a message is actually being sent. If
1455 there is no remote transport (possible with a router that sets up host
1456 addresses), $smtp_active_hostname is used.
1458 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
1459 tweaks were necessary in order to get it to work (see also 21 below):
1460 (a) The code assumed that strncpy() returns a negative number on buffer
1461 overflow, which isn't the case. Replaced with Exim's string_format()
1463 (b) There were several signed/unsigned issues. I just did the minimum
1464 hacking in of casts. There is scope for a larger refactoring.
1465 (c) The code used strcasecmp() which is not a standard C function.
1466 Replaced with Exim's strcmpic() function.
1467 (d) The code set only $1; it now sets $auth1 as well.
1468 (e) A simple test gave the error "authentication client didn't specify
1469 service in request". It would seem that Dovecot has changed its
1470 interface. Fortunately there's a specification; I followed it and
1471 changed what the client sends and it appears to be working now.
1473 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
1476 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
1477 address A is aliased to B and C, where B exists and C does not. Without
1478 -v the output is "A verified" because verification stops after a
1479 successful redirection if more than one address is generated. However,
1480 with -v the child addresses are also verified. Exim was outputting "A
1481 failed to verify" and then showing the successful verification for C,
1482 with its parentage. It now outputs "B failed to verify", showing B's
1483 parentage before showing the successful verification of C.
1485 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
1486 look up a TXT record in a specific list after matching in a combined
1489 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
1490 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
1491 they consult the DNS. I had assumed they would set it the way they
1492 wanted; and indeed my experiments on Linux seem to show that in some
1493 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
1494 To be on the safe side, however, I have now made the interface to
1495 host_find_byname() similar to host_find_bydns(), with an argument
1496 containing the DNS resolver options. The host_find_byname() function now
1497 sets these options at its start, just as host_find_bydns() does. The smtp
1498 transport options dns_qualify_single and dns_search_parents are passed to
1499 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
1500 of host_find_byname() use the default settings of RES_DEFNAMES
1501 (qualify_single) but not RES_DNSRCH (search_parents).
1503 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
1504 spool_read_header() do less string testing, by means of a preliminary
1505 switch on the second character of optional "-foo" lines. (This is
1506 overdue, caused by the large number of possibilities that now exist.
1507 Originally there were few.) While I was there, I also converted the
1508 str(n)cmp tests so they don't re-test the leading "-" and the first
1509 character, in the hope this might squeeze out yet more improvement.
1511 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
1512 flag allowing group syntax was set by the header_syntax check but not
1513 turned off, possible causing trouble later; (2) The flag was not being
1514 set at all for the header_verify test, causing "group"-style headers to
1515 be rejected. I have now set it in this case, and also caused header_
1516 verify to ignore an empty address taken from a group. While doing this, I
1517 came across some other cases where the code for allowing group syntax
1518 while scanning a header line wasn't quite right (mostly, not resetting
1519 the flag correctly in the right place). These bugs could have caused
1520 trouble for malformed header lines. I hope it is now all correct.
1522 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
1523 with the "reply" argument non-NULL. The code, however (which originally
1524 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
1525 but it didn't always do it. This confused somebody who was copying the
1526 code for some other use. I have removed all the tests.
1528 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
1529 feature that was used to support insecure browsers during the U.S. crypto
1530 embargo. It requires special client support, and Exim is probably the
1531 only MTA that supported it -- and would never use it because real RSA is
1532 always available. This code has been removed, because it had the bad
1533 effect of slowing Exim down by computing (never used) parameters for the
1534 RSA_EXPORT functionality.
1536 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
1537 authenticator to fail if there's a tab character in the incoming data
1538 (there should never be unless someone is messing about, as it's supposed
1539 to be base64-encoded). Also added, on Timo's advice, the "secured" option
1540 if the connection is using TLS or if the remote IP is the same as the
1541 local IP, and the "valid-client-cert option" if a client certificate has
1544 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
1545 authenticators. This can be used for authorization after authentication
1546 succeeds. (In the case of plaintext, it servers for both authentication
1549 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
1550 if any retry times were supplied.
1552 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
1553 connection, where there is no client IP address to check. In this
1554 situation, the verify now always succeeds.
1556 PH/25 Applied John Jetmore's -Mset patch.
1558 PH/26 Added -bem to be like -Mset, but loading a message from a file.
1560 PH/27 In a string expansion for a processed (not raw) header when multiple
1561 headers of the same name were present, leading whitespace was being
1562 removed from all of them, but trailing whitespace was being removed only
1563 from the last one. Now trailing whitespace is removed from each header
1564 before concatenation. Completely empty headers in a concatenation (as
1565 before) are ignored.
1567 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
1568 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
1570 PH/29 [Removed. This was a change that I later backed out, and forgot to
1571 correct the ChangeLog entry (that I had efficiently created) before
1572 committing the later change.]
1574 PH/30 Exim was sometimes attempting to deliver messages that had suffered
1575 address errors (4xx response to RCPT) over the same connection as other
1576 messages routed to the same hosts. Such deliveries are always "forced",
1577 so retry times are not inspected. This resulted in far too many retries
1578 for the affected addresses. The effect occurred only when there were more
1579 hosts than the hosts_max_try setting in the smtp transport when it had
1580 the 4xx errors. Those hosts that it had tried were not added to the list
1581 of hosts for which the message was waiting, so if all were tried, there
1582 was no problem. Two fixes have been applied:
1584 (i) If there are any address or message errors in an SMTP delivery, none
1585 of the hosts (tried or untried) are now added to the list of hosts
1586 for which the message is waiting, so the message should not be a
1587 candidate for sending over the same connection that was used for a
1588 successful delivery of some other message. This seems entirely
1589 reasonable: after all the message is NOT "waiting for some host".
1590 This is so "obvious" that I'm not sure why it wasn't done
1591 previously. Hope I haven't missed anything, but it can't do any
1592 harm, as the worst effect is to miss an optimization.
1594 (ii) If, despite (i), such a delivery is accidentally attempted, the
1595 routing retry time is respected, so at least it doesn't keep
1596 hammering the server.
1598 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
1599 in ${readsocket because some servers need this prod.
1601 PH/32 Added some extra debug output when updating a wait-xxx database.
1603 PH/33 The hint "could be header name not terminated by colon", which has been
1604 given for certain expansion errors for a long time, was not being given
1605 for the ${if def:h_colon_omitted{... case.
1607 PH/34 The spec says: "With one important exception, whenever a domain list is
1608 being scanned, $domain contains the subject domain." There was at least
1609 one case where this was not true.
1611 PH/35 The error "getsockname() failed: connection reset by peer" was being
1612 written to the panic log as well as the main log, but it isn't really
1613 panic-worthy as it just means the connection died rather early on. I have
1614 removed the panic log writing for the ECONNRESET error when getsockname()
1617 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
1618 runs only) independently of the message's sender address. This meant
1619 that, if the 4xx error was in fact related to the sender, a different
1620 message to the same recipient with a different sender could confuse
1621 things. In particualar, this can happen when sending to a greylisting
1622 server, but other circumstances could also provoke similar problems.
1623 I have changed the default so that the retry time for these errors is now
1624 based a combination of the sender and recipient addresses. This change
1625 can be overridden by setting address_retry_include_sender=false in the
1628 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
1629 remote server are returned as part of bounce messages. This was not
1630 happening for LMTP over a pipe (the lmtp transport), but now it is the
1631 same for both kinds of LMTP.
1633 PH/38 Despite being documented as not happening, Exim was rewriting addresses
1634 in header lines that were in fact CNAMEs. This is no longer the case.
1636 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
1637 and queue runs started by the daemon processed all messages. This has
1638 been fixed so that -R and -S can now usefully be given with -q<time>.
1640 PH/40 Import PCRE release 6.7 (fixes some bugs).
1642 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
1644 PH/42 Give an error if -q is specified more than once.
1646 PH/43 Renamed the variables $interface_address and $interface_port as
1647 $received_ip_address and $received_port, to make it clear that these
1648 values apply to message reception, and not to the outgoing interface when
1649 a message is delivered. (The old names remain recognized, of course.)
1651 PH/44 There was no timeout on the connect() call when using a Unix domain
1652 socket in the ${readsocket expansion. There now is.
1654 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
1655 be meaningful with "accept".
1657 SC/01 Eximstats V1.43
1658 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
1660 SC/02 Eximstats V1.44
1661 Use a glob alias rather than an array ref in the generated
1662 parser. This improves both readability and performance.
1664 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
1665 Collect SpamAssassin and rejection statistics.
1666 Don't display local sender or destination tables unless
1667 there is data to show.
1668 Added average volumes into the top table text output.
1670 SC/04 Eximstats V1.46
1671 Collect data on the number of addresses (recipients)
1672 as well as the number of messages.
1674 SC/05 Eximstats V1.47
1675 Added 'Message too big' to the list of mail rejection
1676 reasons (thanks to Marco Gaiarin).
1678 SC/06 Eximstats V1.48
1679 Mainlog lines which have GMT offsets and are too short to
1680 have a flag are now skipped.
1682 SC/07 Eximstats V1.49 (Alain Williams)
1683 Added the -emptyok flag.
1685 SC/08 Eximstats V1.50
1686 Fixes for obtaining the IP address from reject messages.
1688 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
1689 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
1690 whitesspace changes from 4.64-PH/27
1692 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
1695 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
1696 are found, allow negative numbers in numeric criteria)
1698 JJ/06 exipick.20061117.2, added new $message_body_missing variable
1700 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
1701 to match changes made in 4.64-PH/43
1703 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
1705 PH/47 Put in an explicit test for a DNS lookup of an address record where the
1706 "domain" is actually an IP address, and force a failure. This locks out
1707 those revolvers/nameservers that support "A-for-A" lookups, in
1708 contravention of the specifications.
1710 PH/48 When a host name was looked up from an IP address, and the subsequent
1711 forward lookup of the name timed out, the host name was left in
1712 $sender_host_name, contrary to the specification.
1714 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
1715 restricted to single-key lookups, Exim was not diagnosing an error if
1716 * or *@ was used with a query-style lookup.
1718 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
1720 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
1721 long ago, but noone interested enough thought of it. Let's just say that
1722 the "1.1" means that there are some new functions that weren't there at
1723 some point in the past.
1725 PH/51 Error processing for expansion failure of helo_data from an smtp
1726 transport during callout processing was broken.
1728 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
1729 tested/used via the -bh/-bhc/-bs options.
1731 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
1732 bug, fixed in subsequent PCRE releases).
1734 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
1735 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
1737 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
1743 SC/01 Use a glob alias rather than an array ref in eximstats generated
1744 parser. This improves both readability and performance.
1746 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
1747 Don't display local sender or destination tables in eximstats unless
1748 there is data to show.
1749 Added average volumes into the eximstats top table text output.
1751 SC/03 Collect data on the number of addresses (recipients) as well
1752 as the number of messages in eximstats.
1754 TF/01 Correct an error in the documentation for the redirect router. Exim
1755 does (usually) call initgroups() when daemonizing.
1757 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
1758 with consistent privilege compared to when running as a daemon.
1760 TF/03 Note in the spec that $authenticated_id is not set for local
1761 submissions from trusted users.
1763 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
1764 Thanks to Dean Brooks <dean@iglou.com> for the patch.
1766 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
1767 by adding some example configuration directives to the default
1768 configuration file. A little bit of work is required to uncomment the
1769 directives and define how usernames and passwords are checked, but
1770 there is now a framework to start from.
1772 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
1773 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
1774 without this. I don't know how relevant this is to other LDAP libraries.
1776 PH/02 Add the verb name to the "unknown ACL verb" error.
1778 PH/03 Magnus Holmgren's patch for filter_prepend_home.
1780 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
1782 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
1783 directory not expanded when it should be if an expanded home directory
1784 was set for the address (which is overridden by the transport).
1786 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
1789 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
1790 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
1791 because it is too late at that time, and has no effect.
1793 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
1794 security issue with \' (bugzilla #107). I could not use the
1795 PQescapeStringConn() function, because it needs a PGconn value as one of
1798 PH/08 When testing addresses using -bt, indicate those final addresses that
1799 are duplicates that would not cause an additional delivery. At least one
1800 person was confused, thinking that -bt output corresponded to deliveries.
1801 (Suppressing duplicates isn't a good idea as you lose the information
1802 about possibly different redirections that led to the duplicates.)
1804 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
1805 systems where poll() doesn't work, in particular OS X.
1807 PH/10 Added more information to debugging output for retry time not reached.
1809 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
1810 operations in malware.c.
1812 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
1815 PH/13 If write_rejectlog was set false when logging was sent to syslog with
1816 syslog_duplication set false, log lines that would normally be written
1817 both the the main log and to the reject log were not written to syslog at
1820 PH/14 In the default configuration, change the use of "message" in ACL warn
1821 statements to "add_header".
1823 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
1824 not followed by a command (e.g. "seen endif").
1826 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
1827 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
1830 PH/17 Added extra conditions to the default value of delay_warning_condition
1834 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
1835 { match{$h_precedence:}{(?i)bulk|list|junk} } \
1836 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
1839 The Auto-Submitted: and various List- headers are standardised, whereas I
1840 don't think Precedence: ever was.
1842 PH/18 Refactored debugging code in route_finduser() to show more information,
1843 in particular, the error code if getpwnam() issues one.
1845 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
1846 This is apparently needed in addition to the PH/07 change above to avoid
1847 any possible encoding problems.
1849 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
1850 but not after initializing Perl.
1852 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
1853 output them only if debugging. By default they are written stderr,
1854 apparently, which is not desirable.
1856 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
1859 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
1862 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
1864 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
1865 authenticated or an ident call has been made. Suppress the default
1866 values for $authenticated_id and $authenticated_sender (but permit -oMai
1867 and -oMas) when testing with -bh.
1869 PH/24 Re-jigged the order of the tests in the default configuration so that the
1870 tests for valid domains and recipients precede the DNS black list and CSA
1871 tests, on the grounds that those ones are more expensive.
1873 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
1874 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
1875 This bug exists in every version of Exim that I still have, right back to
1878 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
1879 However, an attempt to turn on case-sensitivity in a regex key by
1880 including (?-i) didn't work because the subject string was already
1881 lowercased, and the effects were non-intuitive. It turns out that a
1882 one-line patch can be used to allow (?-i) to work as expected.
1888 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
1889 other effects) broke the use of negated acl sub-conditions.
1891 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
1894 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
1895 "Deny" causes Exim to reject the incoming connection with a 554 error.
1896 Unfortunately, if there is a major crisis, such as a disk failure,
1897 tcp-wrappers gives "deny", whereas what one would like would be some
1898 kind of temporary error. A kludge has been added to help with this.
1899 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
1900 554 error is used if errno is still zero or contains ENOENT (which occurs
1901 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
1904 PH/03 Add -lutil to the default FreeBSD LIBS setting.
1906 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
1907 errors. Otherwise a message that provokes a temporary error (when other
1908 messages do not) can cause a whole host to time out.
1910 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
1911 addresses were routed directly to files or pipes from a redirect router.
1912 File deliveries just didn't batch; pipe deliveries might have suffered
1915 PH/06 A failure to get a lock for a hints database would erroneously always say
1916 "Failed to get write lock", even when it was really a read lock.
1918 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
1919 of 0600. This has been changed to use the value of the lockfile_mode
1920 option (which defaults to 0600).
1922 PH/08 Applied small patch from the Sieve maintainer.
1924 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
1925 folder from quota calculations, a direct delivery into this folder messed
1926 up the contents of the maildirsize file. This was because the regex was
1927 used only to exclude .Trash (or whatever) when the size of the mailbox
1928 was calculated. There was no check that a delivery was happening into an
1929 excluded directory. This bug has been fixed by ignoring all quota
1930 processing for deliveries into excluded directories.
1932 PH/10 Added the maildirfolder_create_regex option to appendfile.
1938 PH/01 The code for finding all the local interface addresses on a FreeBSD
1939 system running IPv6 was broken. This may well have applied to all BSD
1940 systems, as well as to others that have similar system calls. The broken
1941 code found IPv4 interfaces correctly, but gave incorrect values for the
1942 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
1943 that it would not match correctly against @[] and not recognize the IPv6
1946 PH/02 The ipliteral router was not recognizing addresses of the form user@
1947 [ipv6:....] because it didn't know about the "ipv6:" prefix.
1949 PH/03 Added disable_ipv6.
1951 PH/04 Changed $reply_address to use the raw form of the headers instead of the
1952 decoded form, because it is most often used to construct To: headers
1953 lines in autoreplies, and the decoded form may well be syntactically
1954 invalid. However, $reply_address has leading white space removed, and all
1955 newlines turned into spaces so that the autoreply transport does not
1958 PH/05 If group was specified without a user on a router, and no group or user
1959 was specified on a transport, the group from the router was ignored.
1961 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
1962 for visible compile-time settings that can be used to change these
1963 numbers, for those that want even more. Backwards compatibility with old
1964 spool files has been maintained. However, going back to a previous Exim
1965 release will lost any variables that are in spool files.
1967 PH/07 Two small changes when running in the test harness: increase delay when
1968 passing a TCP/IP connection to a new process, in case the original
1969 process has to generate a bounce, and remove special handling of
1970 127.0.0.2 (sic), which is no longer necessary.
1972 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
1973 be the same on different OS.
1975 PH/09 Moved a debug statement in filter processing to avoid a race problem when
1978 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
1979 whether --show-vars was specified or not
1981 JJ/02 exipick: Added support for new ACL variable spool format introduced
1984 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
1985 syntactically invalid From: or Reply-to: line, and a filter used this to
1986 generate an autoreply, and therefore failed to obtain an address for the
1987 autoreply, Exim could try to deliver to a non-existent relative file
1988 name, causing unrelated and misleading errors. What now happens is that
1989 it logs this as a hard delivery error, but does not attempt to create a
1992 PH/11 The exinext utility has a -C option for testing purposes, but although
1993 the given file was scanned by exinext itself; it wasn't being passed on
1994 when Exim was called.
1996 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
1997 an end-of-file indication when reading a command response.
1999 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
2000 compiled. In many other places in Exim, IPv6 addresses are always
2001 recognized, so I have changed this. It also means that IPv4 domain
2002 literals of the form [IPV4:n.n.n.n] are now always recognized.
2004 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
2005 used if the router is not running as root, for example, when verifying at
2006 ACL time, or when using -bh. The debugging output from this situation was
2007 non-existent - all you got was a failure to exec. I have made two
2010 (a) Failures to set uid/gid, the current directory, or a process leader
2011 in a subprocess such as that created by queryprogram now generate
2012 suitable debugging ouput when -d is set.
2014 (b) The queryprogram router detects when it is not running as root,
2015 outputs suitable debugging information if -d is set, and then runs
2016 the subprocess without attempting to change uid/gid.
2018 PH/15 Minor change to Makefile for building test_host (undocumented testing
2021 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
2022 additional section of a DNS packet that returns MX or SRV records.
2023 Instead, it always explicitly searches for A/AAAA records. This avoids
2024 major problems that occur when a DNS server includes only records of one
2025 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
2026 fixed another bug: if SRV records were looked up and the corresponding
2027 address records were *not* found in the additional section, the port
2028 values from the SRV records were lost.
2030 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
2031 using the correct key (the original address) when searching the retry
2032 rules in order to find which one to use for generating the retry hint.
2034 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
2035 adding the default one. Similarly, if it contains a Reply-To: header, the
2036 errors_reply_to option, if set, is not used.
2038 PH/19 When calculating a retry time, Exim used to measure the "time since
2039 failure" by looking at the "first failed" field in the retry record. Now
2040 it does not use this if it is later than than the arrival time of the
2041 message. Instead it uses the arrival time. This makes for better
2042 behaviour in cases where some deliveries succeed, thus re-setting the
2043 "first failed" field. An example is a quota failure for a huge message
2044 when small messages continue to be delivered. Without this change, the
2045 "time since failure" will always be short, possible causing more frequent
2046 delivery attempts for the huge message than are intended.
2047 [Note: This change was subsequently modified - see PH/04 for 4.62.]
2049 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
2050 $1, $2, $3) because the numerical variables can be reset during some
2051 expansion items (e.g. "match"), thereby losing the authentication data.
2053 PH/21 Make -bV show the size of off_t variables so that the test suite can
2054 decide whether to run tests for quotas > 2G.
2056 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
2057 mailbox_size, and mailbox_filecount in the appendfile transport. If a
2058 filecount value is greater than 2G or if a quota value is greater than 2G
2059 on a system where the size of off_t is not greater than 4, a panic error
2062 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
2063 never match. The debug and -bh output now contains an explicit error
2064 message indicating a malformed IPv4 address or mask.
2066 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
2067 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
2068 PH/23 above applies.
2070 PH/25 Do not write to syslog when running in the test harness. The only
2071 occasion when this arises is a failure to open the main or panic logs
2072 (for which there is an explicit test).
2074 PH/26 Added the /no_tell option to "control=freeze".
2076 PH/27 If a host name lookup failed very early in a connection, for example, if
2077 the IP address matched host_lookup and the reverse lookup yielded a name
2078 that did not have a forward lookup, an error message of the form "no IP
2079 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
2080 could be logged. Now it outputs the IP address instead of "NULL".
2082 PH/28 An enabling patch from MH: add new function child_open_exim2() which
2083 allows the sender and the authenticated sender to be set when
2084 submitting a message from within Exim. Since child_open_exim() is
2085 documented for local_scan(), the new function should be too.
2087 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
2088 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
2089 results in an empty string is now treated as unset.
2091 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
2093 PH/31 Added sender_verify_fail logging option.
2095 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
2096 needed by RFC 822 but not by RFC 2822 was commented out. I have now
2097 tidied the source and removed it altogether.
2099 PH/33 When a queue run was abandoned because the load average was too high, a
2100 log line was always written; now it is written only if the queue_run log
2101 selector is set. In addition, the log line for abandonment now contains
2102 information about the queue run such as the pid. This is always present
2103 in "start" and "stop" lines but was omitted from the "abandon" line.
2105 PH/34 Omit spaces between a header name and the colon in the error message that
2106 is given when verify = headers_syntax fails (if there are lots of them,
2107 the message gets confusing).
2109 PH/35 Change the default for dns_check_names_pattern to allow slashes within
2110 names, as there are now some PTR records that contain slashes. This check
2111 is only to protect against broken name servers that fall over on strange
2112 characters, so the fact that it applies to all lookups doesn't matter.
2114 PH/36 Now that the new test suite is complete, we can remove some of the
2115 special code in Exim that was needed for the old test suite. For example,
2116 sorting DNS records because real resolvers return them in an arbitrary
2117 order. The new test suite's fake resolver always returns records in the
2120 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
2121 bounces) except when queue_only is set, to avoid logging races between
2122 the different processes.
2124 PH/38 Panic-die if .include specifies a non-absolute path.
2126 PH/39 A tweak to the "H" retry rule from its user.
2128 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
2129 a label. They prevented compilation on older perls.
2131 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
2132 a warning to be raised on newish perls.
2134 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
2135 on queue. Changes to match documented behaviour of showing count of
2136 messages matching specified criteria.
2138 PH/40 Changed the default ident timeout from 30s to 5s.
2140 PH/41 Added support for the use of login_cap features, on those BSD systems
2141 that have them, for controlling the resources used by pipe deliveries.
2143 PH/42 The content-scanning code uses fopen() to create files in which to put
2144 message data. Previously it was not paying any attention to the mode of
2145 the files. Exim runs with umask(0) because the rest of the code creates
2146 files with open(), and sets the required mode explicitly. Thus, these
2147 files were ending up world-writeable. This was not a big issue, because,
2148 being within the spool directory, they were not world-accessible. I have
2149 created a function called modefopen, which takes an additional mode
2150 argument. It sets umask(777), creates the file, chmods it to the required
2151 mode, then resets the umask. All the relevant calls to fopen() in the
2152 content scanning code have been changed to use this function.
2154 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
2155 to 24 hours. This avoids potential overflow problems when processing G
2156 and H retry rules. I suspect nobody ever tinkers with this value.
2158 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
2160 PH/45 When the plaintext authenticator is running as a client, the server's
2161 challenges are checked to ensure they are valid base64 strings. By
2162 default, the authentication attempt is cancelled if an invalid string is
2163 received. Setting client_ignore_invalid_base64 true ignores these errors.
2164 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
2165 they are received. Thus, the responses can be made to depend on the
2166 challenges. If an invalid string is ignored, an empty string is placed in
2169 PH/46 Messages that are created by the autoreply transport now contains a
2170 References: header, in accordance with RFCs 2822 and 3834.
2172 PH/47 Added authenticated_sender_force to the smtp transport.
2174 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
2176 PH/49 Installed latest patch from the Sieve maintainer.
2178 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
2179 was also set, the appendfile transport was unnecessarily scanning a
2180 directory of message files (e.g. for maildir delivery) to find the count
2181 of files (along with the size), even though it did not need this
2182 information. It now does the scan only if it needs to find either the
2183 size of the count of files.
2185 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
2187 PH/52 Two bugs concerned with error handling when the smtp transport is
2190 (i) Exim was not creating retry information for temporary errors given
2191 for individual recipients after the DATA command when the smtp transport
2192 was used in LMTP mode. This meant that they could be retried too
2193 frequently, and not timed out correctly.
2195 (ii) Exim was setting the flag that allows error details to be returned
2196 for LMTP errors on RCPT commands, but not for LMTP errors for individual
2197 recipients that were returned after the DATA command.
2199 PH/53 This is related to PH/52, but is more general: for any failing address,
2200 when detailed error information was permitted to be returned to the
2201 sender, but the error was temporary, then after the final timeout, only
2202 "retry timeout exceeded" was returned. Now it returns the full error as
2203 well as "retry timeout exceeded".
2205 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
2206 do this, and (what is worse) MTAs that accept it.
2208 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
2209 will now be deprecated.
2211 PH/56 New os.c-cygwin from the Cygwin maintainer.
2213 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
2214 formats (previously only available in exim formats via -bpr, -bpru,
2215 and -bpra. Now also available in native and exiqgrep formats)
2217 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
2218 with very large, slow to parse queues
2220 JJ/08 exipick: added ! as generic prefix to negate any criteria format
2222 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
2224 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
2225 responses to authentication challenges, though it was showing the
2226 challenges; (ii) I've removed the CR characters from the debug output for
2229 PH/58 Allow for the insertion of a newline as well as a space when a string
2230 is turned into more than one encoded-word during RFC 2047 encoding. The
2231 Sieve code now uses this.
2233 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
2234 data_4xx, lost_connection, tls_required.
2236 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
2237 message was being sent as an SMTP response.
2239 PH/61 Add -l and -k options to exicyclog.
2241 PH/62 When verifying, if an address was redirected to one new address, so that
2242 verification continued, and the new address failed or deferred after
2243 having set something in $address_data, the value of $address_data was not
2244 passed back to the ACL. This was different to the case when no
2245 redirection occurred. The value is now passed back in both cases.
2247 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
2248 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
2249 use login_cap.h, so on its own it isn't the distinguishing feature. The
2250 new name refers directly to the setclassresources() function.
2252 PH/65 Added configuration files for NetBSD3.
2254 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
2256 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
2257 is preferred over IPv4.
2259 PH/68 The bounce_return_message and bounce_return_body options were not being
2260 honoured for bounces generated during the reception of non-SMTP messages.
2261 In particular, this applied to messages rejected by the ACL. This bug has
2262 been fixed. However, if bounce_return_message is true and bounce_return_
2263 body is false, the headers that are returned for a non-SMTP message
2264 include only those that have been read before the error was detected.
2265 (In the case of an ACL rejection, they have all been read.)
2267 PH/69 The HTML version of the specification is now built in a directory called
2268 spec_html instead of spec.html, because the latter looks like a path with
2269 a MIME-type, and this confuses some software.
2271 PH/70 Catch two compiler warnings in sieve.c.
2273 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
2274 function verify_get_ident() calls ip_connect() to connect a socket, but
2275 if the "connect()" function timed out, ip_connect() used to close the
2276 socket. However, verify_get_ident() also closes the socket later, and in
2277 between Exim writes to the log, which may get opened at this point. When
2278 the socket was closed in ip_connect(), the log could get the same file
2279 descriptor number as the socket. This naturally causes chaos. The fix is
2280 not to close the socket in ip_connect(); the socket should be closed by
2281 the function that creates it. There was only one place in the code where
2282 this was missing, in the iplookup router, which I don't think anybody now
2283 uses, but I've fixed it anyway.
2285 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
2286 well as to direct DNS lookups. Otherwise the handling of names in host
2287 lists is inconsistent and therefore confusing.
2293 PH/01 Two changes to the default runtime configuration:
2295 (1) Move the checks for relay_from_hosts and authenticated clients from
2296 after to before the (commented out) DNS black list checks.
2298 (2) Add control=submission to the relay_from_hosts and authenticated
2299 clients checks, on the grounds that messages accepted by these
2300 statements are most likely to be submissions.
2302 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
2304 (1) Generate an error if the third argument for the ${prvs expansion is
2307 (2) Treat a missing third argument of ${prvscheck as if it were an empty
2310 (3) Reset the variables that are obtained from the first argument of
2311 ${prvscheck and used in the second argument before leaving the code,
2312 because their memory is reclaimed, so using them afterwards may do
2315 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
2316 one (it's much easier than Tom thought :-).
2318 (5) Because of (4), we can now allow for the use of $prvscheck_result
2319 inside the third argument.
2321 PH/03 For some reason, the default setting of PATH when running a command from
2322 a pipe transport was just "/usr/bin". I have changed it to
2325 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
2326 anything to be listed in the output from -bV.
2328 PH/05 When a filter generated an autoreply, the entire To: header line was
2329 quoted in the delivery log line, like this:
2331 => >A.N.Other <ano@some.domain> <original@ddress> ...
2333 This has been changed so that it extracts the operative address. There
2334 may be more than one such address. If so, they are comma-separated, like
2337 => >ano@some.domain,ona@other.domain <original@ddress> ...
2339 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
2340 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
2341 not being looked up in the rDNS to get a host name, Exim was showing the
2342 IP address twice in Received: lines, even though the IP addresses were
2343 identical. For example:
2345 Received: from [1.2.3.4] (helo=[1.2.3.4])
2347 However, if the real host name was known, it was omitting the HELO data
2348 if it matched the actual IP address. This has been tidied up so that it
2349 doesn't show the same IP address twice.
2351 PH/07 When both +timestamp and +memory debugging was on, the value given by
2352 $tod_xxx expansions could be wrong, because the tod_stamp() function was
2353 called by the debug printing, thereby overwriting the timestamp buffer.
2354 Debugging no longer uses the tod_stamp() function when +timestamp is set.
2356 PH/08 When the original message was included in an autoreply transport, it
2357 always said "this is a copy of the message, including all the headers",
2358 even if body_only or headers_only was set. It now gives an appropriate
2361 PH/09 Applied a patch from the Sieve maintainer which:
2363 o fixes some comments
2364 o adds the (disabled) notify extension core
2365 o adds some debug output for the result of if/elsif tests
2366 o points to the current vacation draft in the documentation
2367 and documents the missing references header update
2371 o fixes a bug in processing the envelope test (when testing
2372 multiple envelope elements, the last element determinted the
2375 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
2376 Electronic Mail") by including:
2378 Auto-submitted: auto-generated
2380 in the messages that it generates (bounce messages and others, such as
2381 warnings). In the case of bounce messages for non-SMTP mesages, there was
2382 also a typo: it was using "Auto_submitted" (underscore instead of
2383 hyphen). Since every message generated by Exim is necessarily in response
2384 to another message, thes have all been changed to:
2386 Auto-Submitted: auto-replied
2388 in accordance with these statements in the RFC:
2390 The auto-replied keyword:
2392 - SHOULD be used on messages sent in direct response to another
2393 message by an automatic process,
2395 - MUST NOT be used on manually-generated messages,
2397 - MAY be used on Delivery Status Notifications (DSNs) and Message
2398 Disposition Notifications (MDNs),
2400 - MUST NOT be used on messages generated by automatic or periodic
2401 processes, except for messages which are automatic responses to
2404 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
2405 to the default Received: header definition.
2407 PH/12 Added log selector acl_warn_skipped (default on).
2409 PH/13 After a successful wildlsearch lookup, discard the values of numeric
2410 variables because (a) they are in the wrong storage pool and (b) even if
2411 they were copied, it wouldn't work properly because of the caching.
2413 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
2414 checking when decoding. Apparently there are clients that generate
2415 overlong encoded strings. Why am I not surprised?
2417 PH/15 If the first argument of "${if match_address" was not empty, but did not
2418 contain an "@" character, Exim crashed. Now it writes a panic log message
2419 and treats the condition as false.
2421 PH/16 In autoreply, treat an empty string for "once" the same as unset.
2423 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
2424 extension "envelope-auth". The code is finished and in agreement with
2425 other implementations, but there is no documentation so far and in fact,
2426 nobody wrote the draft yet. This extension is currently #undef'ed, thus
2427 not changing the active code.
2429 Print executed "if" and "elsif" statements when debugging is used. This
2430 helps a great deal to understand what a filter does.
2432 Document more things not specified clearly in RFC3028. I had all this
2433 sorted out, when out of a sudden new issues came to my mind. Oops."
2435 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
2438 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
2439 canonical form (as documented). However, after a host name lookup from
2440 the IP address, check_host() was doing a simple string comparison with
2441 addresses acquired from the DNS when checking that the found name did
2442 have the original IP as one of its addresses. Since any found IPv6
2443 addresses are likely to be in abbreviated form, the comparison could
2444 fail. Luckily, there already exists a function for doing the comparison
2445 by converting both addresses to binary, so now that is used instead of
2446 the text comparison.
2448 PH/20 There was another similar case to PH/19, when a complete host name was
2449 given in a host list; looking up its IP address could give an abbreviated
2450 form, whereas the current host's name might or might not be abbreviated.
2451 The same fix has been applied.
2457 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
2458 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
2461 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
2462 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
2464 PH/03 Typo: missing ".o" in src/pcre/Makefile.
2466 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
2467 header line, restrict the check to what is listed in RFCs 2369 and 2929.
2468 Also, for "Auto-Submitted", treat anything other than "no" as
2469 non-personal, in accordance with RFC 3834. (Previously it treated
2470 anything starting "auto-" as non-personal.)
2472 TF/01 The control=submission/name=... option had a problem with syntax
2473 errors if the name included a slash character. The /name= option
2474 now slurps the rest of the string, so it can include any characters
2475 but it must come last in the list of options (after /sender_retain
2478 PH/05 Some modifications to the interface to the fake nameserver for the new
2486 TK/01 Added the "success_on_redirect" address verification option. See
2487 NewStuff for rationale and an example.
2489 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
2491 PH/02 Patch to exigrep to allow it to work on syslog lines.
2493 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
2494 fread() to skip over the body file's header line, because in Cygwin the
2495 header line is locked and is inaccessible.
2497 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
2498 co-exist for some time) to make it clear that it is the Exim ID that is
2499 referenced, not the Message-ID: header line.
2501 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
2502 string_format() function, because snprintf() does not exist on all
2505 PH/06 The use of forbid_filter_existstest now also locks out the use of the
2506 ${stat: expansion item.
2508 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
2509 protocol synchronization error", to keep the pedants happy.
2511 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
2512 well as for IRIX systems, when gcc is being used. See the host.c source
2515 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
2517 PH/10 Named domain lists were not working if used in a queue_smtp_domains
2520 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
2521 transport and to the smtp transport in LMTP mode.
2523 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
2525 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
2526 run a filter in a subprocess. This could lead to confusion in subsequent
2527 lookups in the parent process. There should also be a search_tidyup() at
2528 the end of the subprocess.
2530 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
2531 only if the host matched helo_try_verify_hosts, which caused the
2532 verification to occur when the EHLO/HELO command was issued. The ACL just
2533 tested the remembered result. Now, if a previous verification attempt has
2534 not happened, "verify = helo" does it there and then.
2536 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
2538 TK/03 Fix log output including CR from clamd.
2540 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
2541 exist provoked a memory error which could cause a segfault.
2543 PH/15 Installed PCRE 6.2
2545 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
2547 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
2548 of the problem. Specifically, suggested +O2 rather than +O1 for the
2551 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
2553 PH/20 If a delivery was routed to a non-standard port by means of an SRV
2554 record, the port was not correctly logged when the outgoing_port log
2555 selector was set (it logged the transort's default port).
2557 PH/21 Added support for host-specific ports to manualroute, queryprogram,
2558 fallback_hosts, and "hosts" in the smtp transport.
2560 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
2561 host errors such as "Connection refused".
2563 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
2564 authentication with radiusclient 0.4.9:
2566 - Error returned from rc_read_config was caught wrongly
2567 - Username/password not passed on to radius server due to wrong length.
2569 The presumption is that some radiusclient API changes for 4.51/PH/17
2570 were not taken care of correctly. The code is still untested by me (my
2571 Linux distribution still has 0.3.2 of radiusclient), but it was
2572 contributed by a Radius user.
2574 PH/24 When doing a callout, the value of $domain wasn't set correctly when
2575 expanding the "port" option of the smtp transport.
2577 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
2578 while reading a MIME header. Thanks to Tom Hughes for a patch.
2580 PH/24 Include config.h inside local_scan.h so that configuration settings are
2583 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
2584 that in an ACL for RCPT (for example), you can examine exactly what was
2587 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
2588 commands, but it was not correctly comparing the address with the actual
2589 client host address. Thus, it would show the EHLO address in Received:
2590 header lines when this was not necessary.
2592 PH/27 Added the % operator to ${eval:}.
2594 PH/28 Exim tries to create and chdir to its spool directory when it starts;
2595 it should be ignoring failures (because with -C, for example, it has lost
2596 privilege). It wasn't ignoring creation failures other than "already
2599 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
2602 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
2603 because an input line was too long, either on its own, or by virtue of
2604 too many continuations, the temporary file was not being removed, and the
2605 return code was incorrect.
2607 PH/31 Missing "BOOL" in function definition in filtertest.c.
2609 PH/32 Applied Sieve patches from the maintainer.
2611 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
2613 PH/33 Added "verify = not_blind".
2615 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
2616 Local/Makefile (with some defaults set). These are used in built scripts
2617 such as exicyclog, but they have never been used in the exim_install
2618 script (though there are many overriding facilities there). I have
2619 arranged that the exim_install script now takes note of these two
2622 PH/35 Installed configuration files for Dragonfly.
2624 PH/36 When a locally submitted message by a trusted user did not contain a
2625 From: header, and the sender address was obtained from -f or from an SMTP
2626 MAIL command, and the trusted user did not use -F to supply a sender
2627 name, $originator_name was incorrectly used when constructing a From:
2628 header. Furthermore, $originator_name was used for submission mode
2629 messages from external hosts without From: headers in a similar way,
2630 which is clearly wrong.
2632 PH/37 Added control=suppress_local_fixups.
2634 PH/38 When log_selector = +received_sender was set, and the addition of the
2635 sender made the log line's construction buffer exactly full, or one byte
2636 less than full, an overflow happened when the terminating "\n" was
2639 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
2640 when the result of a list match is failure because a DNS lookup failed.
2642 PH/40 RM_COMMAND is now used in the building process.
2644 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
2645 the "build-* directories that it finds.
2647 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
2648 address in a domain literal was a prefix of an interface address.
2650 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
2651 when verifying a sender address, unless rewrite_headers is false.
2653 PH/44 Wrote a long comment about why errors_to addresses are verified as
2654 recipients, not senders.
2656 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
2657 the ratelimit ACL was added.
2659 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
2661 PH/46 Added extra information about PostgreSQL errors to the error string.
2663 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
2664 avoiding the need to install special zones in a real server. This is
2665 backwards compatible; if it can't find the fake resolver, it drops back.
2666 Thus, both old and new test suites can be run.
2668 TF/02 Added util/ratelimit.pl
2670 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
2671 clock is set back in time.
2673 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
2674 Candler <B.Candler@pobox.com>.
2676 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
2677 OK for addresses that are the result of redirections.
2679 PH/48 A number of further additions for the benefit of the new test suite,
2680 including a fake gethostbyname() that interfaces to the fake DNS resolver
2683 TF/06 The fix for widen_domains has also been applied to qualify_single and
2684 search_parents which are the other dnslookup options that can cause
2687 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
2690 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
2692 TF/07 Exim produced the error message "an SRV record indicated no SMTP
2693 service" if it encountered an MX record with an empty target hostname.
2694 The message is now "an MX or SRV record indicated no SMTP service".
2696 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
2697 if the DNS of the sending site is misconfigured. This is quite a
2698 common situation. This change restores the behaviour of treating a
2699 helo verification defer as a failure.
2701 PH/51 If self=fail was set on a router, the bounce message did not include the
2702 actual error message.
2708 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
2710 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
2711 command itself ended in error, the underlying message about the transport
2712 filter timeout was being overwritten with the pipe command error. Now the
2713 underlying error message should be appended to the second error message.
2715 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
2717 PH/02 Reduce the amount of output that "make" produces by default. Full output
2718 can still be requested.
2720 PH/03 The warning log line about a condition test deferring for a "warn" verb
2721 was being output only once per connection, rather than after each
2722 occurrence (because it was using the same function as for successful
2723 "warn" verbs). This seems wrong, so I have changed it.
2725 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
2726 it should not have, which might have caused a crash in the right
2727 circumstances, but probably never did.
2729 PH/04 Installed a modified version of Tony Finch's patch to make submission
2730 mode fix the return path as well as the Sender: header line, and to
2731 add a /name= option so that you can make the user's friendly name appear
2734 TF/03 Added the control = fakedefer ACL modifier.
2736 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
2737 Mark Lowes for thorough testing.
2739 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
2741 TK/03 Merged latest SRS patch from Miles Wilton.
2743 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
2744 with the definition in sysexits.h (which is #included earlier).
2745 Fortunately, Exim does not actually use EX_OK. The code used to try to
2746 preserve the sysexits.h value, by assumimg that macro definitions were
2747 scanned for macro replacements. I have been disabused of this notion,
2748 so now the code just undefines EX_OK before #including unistd.h.
2750 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
2751 in the smtp transport. When a block could not be written in a single
2752 write() function, the timeout was being re-applied to each part-write.
2753 This seems wrong - if the receiver was accepting one byte at a time it
2754 would take for ever. The timeout is now adjusted when this happens. It
2755 doesn't have to be particularly precise.
2757 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
2758 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
2760 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
2761 without a domain if the check to <postmaster@domain> fails.
2763 SC/01 Eximstats: added -xls and the ability to specify output files
2764 (patch written by Frank Heydlauf).
2766 SC/02 Eximstats: use FileHandles for outputing results.
2768 SC/03 Eximstats: allow any combination of xls, txt, and html output.
2770 SC/04 Eximstats: fixed display of large numbers with -nvr option
2772 SC/05 Eximstats: fixed merging of reports with empty tables.
2774 SC/06 Eximstats: added the -include_original_destination flag
2776 SC/07 Eximstats: removed tabs and trailing whitespace.
2778 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
2780 TK/06 MBOX spool code: Add real "From " MBOX separator line
2781 so the .eml file is really in mbox format (even though
2782 most programs do not really care). Patch from Alex Miller.
2784 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
2785 The latter is generated from $received_to and is only set if the
2786 message has one envelope recipient. SA can use these headers,
2787 obviously out-of-the-box. Patch from Alex Miller.
2789 PH/08 The ${def test on a variable was returning false if the variable's
2790 value was "0", contrary to what the specification has always said!
2791 The result should be true unless the variable is empty.
2793 PH/09 The syntax error of a character other than { following "${if
2794 def:variable_name" (after optional whitespace) was not being diagnosed.
2795 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
2796 accidental colon was present, for example, could give incorrect results.
2798 PH/10 Tidied the code in a number of places where the st_size field of a stat()
2799 result is used (not including appendfile, where other changes are about
2802 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
2803 This involved changing a lot of size variables from int to off_t. It
2804 should work with maildirs and everything.
2806 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
2807 spamd dying while we are connected to it.
2809 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
2812 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
2813 function for BDB 4.3.
2815 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
2817 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
2818 the sources for PCRE have been re-arranged and more files are now
2821 PH/15 The code I had for printing potentially long long variables in PH/11
2822 above was not the best (it lost precision). The length of off_t variables
2823 is now inspected at build time, and an appropriate printing format (%ld
2824 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
2825 to be "long long int" or "long int". This is needed for the internal
2826 formatting function string_vformat().
2828 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
2829 the configuration file to be ":syslog", then the script "guesses" where
2830 the logs files are, rather than using the compiled in default. In our
2831 case the guess is not the same as the compiled default, so the script
2832 suddenly stopped working when I started to use syslog. The patch checks
2833 to see if log_file_path is "". If so, it attempts to read it from exim
2834 with no configuration file to get the compiled in version, before it
2835 falls back to the previous guessing code."
2837 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
2838 implementing BATV in an Exim configuration. See NewStuff for the gory
2841 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
2842 Makefile that are specific to HP-UX.
2844 PH/18 If the "use_postmaster" option was set for a recipient callout together
2845 with the "random" option, the postmaster address was used as the MAIL
2846 FROM address for the random test, but not for the subsequent recipient
2847 test. It is now used for both.
2849 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
2850 patch removes a few documentation additions to RFC 3028, because the
2851 latest draft now contains them. It adds the new en;ascii-case comparator
2852 and a new error check for 8bit text in MIME parts. Comparator and
2853 require names are now matched exactly. I enabled the subaddress
2854 extension, but it is not well tested yet (read: it works for me)."
2856 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
2857 rework some of the code of TK/09 above to avoid the hardwired use of
2858 "%lld" and "long long". Replaced the call to snprintf() with a call to
2861 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
2862 records point to non-existent hosts", "retry timeout exceeded", and
2863 "retry time not reached for any host after a long failure period".
2865 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
2866 experimental DomainKeys support:
2868 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
2869 (2) On an error such as an illegally used "control", the wrong name for
2870 the control was given.
2872 These problems did NOT occur unless DomainKeys support was compiled.
2874 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
2876 PH/24 Added ${if match_ip condition.
2878 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
2879 databases so that it will be absolutely obvious if a crash occurs in the
2880 DB library. This is a regular occurrence (often caused by mis-matched
2883 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
2884 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
2885 picked up on a user's system that detects such things. There doesn't seem
2886 to be a gcc warning option for this - only an attribute that has to be
2887 put on the function's prototype. It seems that in Fedora Core 4 they have
2888 set this on a number of new functions. No doubt there will be more in due
2891 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
2892 specify a transport. However, if an address that was verified by such a
2893 router was the subject of a callout, Exim crashed because it tried to
2894 read the rcpt_include_affixes from the non-existent transport. Now it
2895 just assumes that the setting of that option is false. This bug was
2896 introduced by 4.51/PH/31.
2898 PH/28 Changed -d+all to exclude +memory, because that information is very
2899 rarely of interest, but it makes the output a lot bigger. People tend to
2900 do -d+all out of habit.
2902 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
2903 code in os-type was giving problems when libc.so lives in lib64, like on
2906 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
2907 aren't the modern standard, and it seems that some systems' include files
2908 don't always have them. Exim was already checking for some of the newer
2909 ones like T_AAAA, and defining it itself. I've added checks for all the
2910 record types that Exim uses.
2912 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
2913 not automatically generating a new one, as it is supposed to. This
2914 prevented TLS from working. If the file did exist, but contained invalid
2915 data, a new version was generated, as expected. It was only the case of a
2916 non-existent file that was broken.
2918 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
2919 with a change in libdomainkeys > 0.64.
2921 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
2922 from DNS. If the selector record carries the flag, it now has
2923 precedence over the domain-wide flag.
2925 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
2927 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
2928 the use of a port name that isn't defined in /etc/services) occurred, the
2929 message was deferred as in a normal delivery, and thus remained on the
2930 spool, instead of being failed because of the mua_wrapper setting. This
2931 is now fixed, and I tidied up some of the mua_wrapper messages at the
2934 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
2935 the messages in a hash of arrays rather than using individual hashes.
2936 This is a bit cleaner and results in dramatic memory savings, albeit
2937 at a slight CPU cost.
2939 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
2940 as requested by Marc Sherman.
2942 SC/10 Eximstats: added histograms for user specified patterns as requested
2945 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
2947 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
2948 fopen() in the content-scanning modules that did not already have it.
2954 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
2955 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
2957 TK/02 Fix ACL "control" statement not being available in MIME ACL.
2959 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
2961 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
2962 to test Sieve filters that use "vacation".
2964 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
2965 that changes the way the GnuTLS parameters are stored in the cache file.
2966 The new format can be generated externally. For backward compatibility,
2967 if the data in the cache doesn't make sense, Exim assumes it has read an
2968 old-format file, and it generates new data and writes a new file. This
2969 means that you can't go back to an older release without removing the
2972 PH/03 A redirect router that has both "unseen" and "one_time" set does not
2973 work if there are any delivery delays because "one_time" forces the
2974 parent to be marked "delivered", so its unseen clone is never tried
2975 again. For this reason, Exim now forbids the simultaneous setting of
2978 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
2979 redirected to themselves ("homonym" addresses). Read the long ChangeLog
2980 entry if you want to know the details. The fix, however, neglected to
2981 consider the case when local delivery batching is involved. The test for
2982 "previously delivered" was not happening when checking to see if an
2983 address could be batched with a previous (undelivered) one; under
2984 certain circumstances this could lead to multiple deliveries to the same
2987 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
2988 in its include files, and this causes problems building Exim.
2990 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
2991 header_syntax) but Exim was just ignoring anything given after a slash.
2992 In particular, this caused confusion with an attempt to use "verify =
2993 reverse_host_lookup/defer_ok". An error is now given when options are
2994 supplied for verify items that do not have them. (Maybe reverse_host_
2995 lookup should have a defer_ok option, but that's a different point.)
2997 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
2998 defined by RFC 821) to 2048, because there were problems with some AUTH
2999 commands, and RFC 1869 says the size should be increased for extended
3000 SMTP commands that take arguments.
3002 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
3005 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
3006 "unknown" error; now it says that the functionality isn't in the binary.
3008 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
3009 an address' error message when a string expansion fails (syntax or
3010 whatever). Otherwise the password may appear in the log. Following change
3011 PH/42 below, there is no longer a chance of it appearing in a bounce
3014 PH/11 Installed exipick version 20050225.0 from John Jetmore.
3016 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
3017 of its addresses was ever tried. (Bugzilla bug #2.)
3019 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
3020 the result incorrectly in the debug output. (It correctly added a newline
3021 to what was transported.)
3023 TF/01 Added $received_time.
3025 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
3026 commented out examples of how to interface to a virus scanner and to
3027 SpamAssassin. Also added commented examples of av_scanner and
3028 spamd_address settings.
3030 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
3031 and controls are allowed in which ACLs. There were a couple of minor
3032 errors. Some of the entries in the conditions table (which is a table of
3033 where they are NOT allowed) were getting very unwieldy; rewrote them as a
3034 negation of where the condition IS allowed.
3036 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
3038 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
3039 header file does not have a version number, so I've had to invent a new
3040 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
3041 API. The code is untested by me (my Linux distribution still has 0.3.2 of
3042 radiusclient), but it was contributed by a Radius user.
3044 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
3045 files or directories, for OpenSSL.
3047 PH/19 When an Exim process that is running as root has to create an Exim log
3048 file, it does so in a subprocess that runs as exim:exim so as to get the
3049 ownership right at creation (otherwise, other Exim processes might see
3050 the file with the wrong ownership). There was no test for failure of this
3051 fork() call, which would lead to the process getting stuck as it waited
3052 for a non-existent subprocess. Forks do occasionally fail when resources
3053 run out. I reviewed all the other calls to fork(); they all seem to check
3056 PH/20 When checking for unexpected SMTP input at connect time (before writing
3057 the banner), Exim was not dealing correctly with a non-positive return
3058 from the read() function. If the client had disconnected by this time,
3059 the result was a log entry for a synchronization error with an empty
3060 string after "input=" when read() returned zero. If read() returned -1
3061 (an event I could not check), uninitialized data bytes were printed.
3062 There were reports of junk text (parts of files, etc) appearing after
3065 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
3067 PH/22 Added support for macro redefinition, and (re)definition in between
3068 driver and ACL definitions.
3070 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
3071 forgetting to use the resulting value; it was using the unexpanded value.
3073 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
3074 hadn't been configured. The fix is from Juergen Kreileder, who
3075 understands it better than I do:
3077 "Here's what I see happening with three configured cyrus_sasl
3078 authenticators configured (plain, login, cram-md5):
3080 On startup auth_cyrus_sasl_init() gets called for each of these.
3081 This means three calls to sasl_listmech() without a specified mech_list.
3082 => SASL tests which mechs of all available mechs actually work
3083 => three warnings about OTP not working
3084 => the returned list contains: plain, login, cram-md5, digest-md5, ...
3086 With the patch, sasl_listmech() also gets called three times. But now
3087 SASL's mech_list option is set to the server_mech specified in the the
3088 authenticator. Or in other words, the answer from sasl_listmech()
3089 gets limited to just the mech you're testing for (which is different
3091 => the return list contains just 'plain' or 'login', 'cram-md5' or
3092 nothing depending on the value of ob->server_mech.
3094 I've just tested the patch: Authentication still works fine,
3095 unavailable mechs specified in the exim configuration are still
3096 caught, and the auth.log warnings about OTP are gone."
3098 PH/25 When debugging is enabled, the contents of the command line are added
3099 to the debugging output, even when log_selector=+arguments is not
3102 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
3103 answer is "GNU", and only if the return is "GNU/something" is the answer
3106 PH/27 $acl_verify_message is now set immediately after the failure of a
3107 verification in an ACL, and so is available in subsequent modifiers. In
3108 particular, the message can be preserved by coding like this:
3110 warn !verify = sender
3111 set acl_m0 = $acl_verify_message
3113 Previously, $acl_verify_message was set only while expanding "message"
3114 and "log_message" when a very denied access.
3116 PH/28 Modified OS/os.c-Linux with
3118 -#ifndef OS_LOAD_AVERAGE
3119 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
3121 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
3122 nomenclature these days.)
3124 PH/29 Installed patch from the Sieve maintainer that adds the options
3125 sieve_useraddress and sieve_subaddress to the redirect router.
3127 PH/30 In these circumstances:
3128 . Two addresses routed to the same list of hosts;
3129 . First host does not offer TLS;
3130 . First host accepts first address;
3131 . First host gives temporary error to second address;
3132 . Second host offers TLS and a TLS session is established;
3133 . Second host accepts second address.
3134 Exim incorrectly logged both deliveries with the TLS parameters (cipher
3135 and peerdn, if requested) that were in fact used only for the second
3138 PH/31 When doing a callout as part of verifying an address, Exim was not paying
3139 attention to any local part prefix or suffix that was matched by the
3140 router that accepted the address. It now behaves in the same way as it
3141 does for delivery: the affixes are removed from the local part unless
3142 rcpt_include_affixes is set on the transport.
3144 PH/32 Add the sender address, as F=<...>, to the log line when logging a
3145 timeout during the DATA phase of an incoming message.
3147 PH/33 Sieve envelope tests were broken for match types other than :is. I have
3148 applied a patch sanctioned by the Sieve maintainer.
3150 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
3151 the uid or gid is negative. A case of a negative gid caused this to be
3152 noticed. The fix allows for either to be negative.
3154 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
3155 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
3158 PH/36 The change PH/12 above was broken. Fixed it.
3160 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
3161 the grounds that routing the same address twice would always produce the
3162 same answer. This might have been true once, but it is certainly no
3163 longer true now. Routing a child address may depend on the previous
3164 routing that produced that child. Some complicated redirection strategies
3165 went wrong when messages had multiple recipients, and made Exim's
3166 behaviour dependent on the order in which the addresses were given.
3168 I have moved the duplicate checking until after the routing is complete.
3169 Exim scans the addresses that are assigned to local and remote
3170 transports, and removes any duplicates. This means that more work will be
3171 done, as duplicates will always all be routed, but duplicates are
3172 presumably rare, so I don't expect this is of any significance.
3174 For deliveries to pipes, files, and autoreplies, the duplicate checking
3175 still happens during the routing process, since they are not going to be
3178 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
3179 It corrects a timeout issue with spamd. This is Ian's comment: "The
3180 background is that sometimes spamd either never reads data from a
3181 connection it has accepted, or it never writes response data. The exiscan
3182 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
3183 blindly assumes that writes won't block so it may never time out."
3185 PH/39 Allow G after quota size as well as K and M.
3187 PH/40 The value set for $authenticated_id in an authenticator may not contain
3188 binary zeroes or newlines because the value is written to log lines and
3189 to spool files. There was no check on this. Now the value is run through
3190 the string_printing() function so that such characters are converted to
3191 printable escape sequences.
3193 PH/41 $message_linecount is a new variable that contains the total number of
3194 lines in the message. Compare $body_linecount, which is the count for the
3197 PH/42 Exim no longer gives details of delivery errors for specific addresses in
3198 bounce and delay warning messages, except in certain special cases, which
3201 (a) An SMTP error message from a remote host;
3202 (b) A message specified in a :fail: redirection;
3203 (c) A message specified in a "fail" command in a system filter;
3204 (d) A message specified in a FAIL return from the queryprogram router;
3205 (e) A message specified by the cannot_route_message router option.
3207 In these cases only, Exim does include the error details in bounce and
3208 warning messages. There are also a few cases where bland messages such
3209 as "unrouteable address" or "local delivery error" are given.
3211 PH/43 $value is now also set for the "else" part of a ${run expansion.
3213 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
3214 being worked on, but at least Exim now implements the latest version to
3217 PH/45 In a pipe transport, although a timeout while waiting for the pipe
3218 process to complete was treated as a delivery failure, a timeout while
3219 writing the message to the pipe was logged, but erroneously treated as a
3220 successful delivery. Such timeouts include transport filter timeouts. For
3221 consistency with the overall process timeout, these timeouts are now
3222 treated as errors, giving rise to delivery failures by default. However,
3223 there is now a new Boolean option for the pipe transport called
3224 timeout_defer, which, if set TRUE, converts the failures into defers for
3225 both kinds of timeout. A transport filter timeout is now identified in
3228 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
3229 systems where "make" and "gmake" are different, calling "gmake" at top
3230 level broke things. I've arranged for the value of $(MAKE) to be passed
3231 from the Makefile to this script so that it can call the same version of
3235 A note about Exim versions 4.44 and 4.50
3236 ----------------------------------------
3238 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
3239 changes of various kinds. As a consequence, a big documentation update was
3240 needed. This delayed the release for rather longer than seemed good, especially
3241 in the light of a couple of (minor) security issues. Therefore, the changes
3242 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
3243 release. So 4.44 and 4.50 are in effect two different branches that both start
3246 I have left the 4.50 change log unchanged; it contains all the changes since
3247 4.43. The change log for 4.44 is below; many of its items are identical to
3248 those for 4.50. This seems to be the most sensible way to preserve the
3249 historical information.
3255 1. Minor wording change to the doc/README.SIEVE file.
3257 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
3258 computation of the current number of files was incorrect.
3260 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
3261 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
3264 4. Give more explanation in the error message when the command for a transport
3265 filter fails to execute.
3267 5. There are several places where Exim runs a non-Exim command in a
3268 subprocess. The SIGUSR1 signal should be disabled for these processes. This
3269 was being done only for the command run by the queryprogram router. It is
3270 now done for all such subprocesses. The other cases are: ${run, transport
3271 filters, and the commands run by the lmtp and pipe transports.
3273 6. Added CONFIGURE_GROUP build-time option.
3275 7. Some older OS have a limit of 256 on the maximum number of file
3276 descriptors. Exim was using setrlimit() to set 1000 as a large value
3277 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
3278 systems. I've change it so that if it can't get 1000, it tries for 256.
3280 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
3281 was an oversight, and furthermore, ever since the addition of extra
3282 controls (e.g. 4.43/32), the checks on when to allow different forms of
3283 "control" were broken. There should now be diagnostics for all cases when a
3284 control that does not make sense is encountered.
3286 9. Added the /retain_sender option to "control=submission".
3288 10. $recipients is now available in the predata ACL (oversight).
3290 11. Tidy the search cache before the fork to do a delivery from a message
3291 received from the command line. Otherwise the child will trigger a lookup
3292 failure and thereby defer the delivery if it tries to use (for example) a
3293 cached ldap connection that the parent has called unbind on.
3295 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
3296 of $address_data from the recipient verification was clobbered by the
3297 sender verification.
3299 13. The value of address_data from a sender verification is now available in
3300 $sender_address_data in subsequent conditions in the ACL statement.
3302 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
3304 15. Added a new option "connect=<time>" to callout options, to set a different
3307 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
3308 was its contents. (It was OK if the option was not defined at all.)
3310 17. A "Completed" log line is now written for messages that are removed from
3311 the spool by the -Mrm option.
3313 18. New variables $sender_verify_failure and $recipient_verify_failure contain
3314 information about exactly what failed.
3316 19. Added -dd to debug only the daemon process.
3318 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
3319 handles timeouts, both on the server side and network timeouts. Renamed the
3320 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
3322 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
3323 It is now set to "smtps".
3325 22. $host_address is now set to the target address during the checking of
3326 ignore_target_hosts.
3328 23. When checking ignore_target_hosts for an ipliteral router, no host name was
3329 being passed; this would have caused $sender_host_name to have been used if
3330 matching the list had actually called for a host name (not very likely,
3331 since this list is usually IP addresses). A host name is now passed as
3334 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
3335 code that specifies a non-restarting handler (typically sigaction() in
3336 modern systems) in an attempt to fix a rare and obscure crash bug.
3338 25. Narrowed the window for a race in the daemon that could cause it to ignore
3339 SIGCHLD signals. This is not a major problem, because they are used only to
3340 wake it up if nothing else does.
3342 26. A malformed maildirsize file could cause Exim to calculate negative values
3343 for the mailbox size or file count. Odd effects could occur as a result.
3344 The maildirsize information is now recalculated if the size or filecount
3347 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
3348 support for a long time. Removed HAVE_SYS_VFS_H.
3350 28. Installed the latest version of exipick from John Jetmore.
3352 29. In an address list, if the pattern was not a regular expression, an empty
3353 subject address (from a bounce message) matched only if the pattern was an
3354 empty string. Non-empty patterns were not even tested. This was the wrong
3355 because it is perfectly reasonable to use an empty address as part of a
3356 database query. An empty address is now tested by patterns that are
3357 lookups. However, all the other forms of pattern expect the subject to
3358 contain a local part and a domain, and therefore, for them, an empty
3359 address still always fails if the pattern is not itself empty.
3361 30. Exim went into a mad DNS loop when attempting to do a callout where the
3362 host was specified on an smtp transport, and looking it up yielded more
3363 than one IP address.
3365 31. Re-factored the code for checking spool and log partition space into a
3366 function that finds that data and another that does the check. The former
3367 is then used to implement four new variables: $spool_space, $log_space,
3368 $spool_inodes, and $log_inodes.
3370 32. The RFC2047 encoding function was originally intended for short strings
3371 such as real names; it was not keeping to the 75-character limit for
3372 encoded words that the RFC imposes. It now respects the limit, and
3373 generates multiple encoded words if necessary. To be on the safe side, I
3374 have increased the buffer size for the ${rfc2047: expansion operator from
3377 33. It is now permitted to omit both strings after an "if" condition; if the
3378 condition is true, the result is "true". As before, when the second string
3379 is omitted, a false condition yields an empty string. This makes it less
3380 cumbersome to write custom ACL and router conditions.
3382 34. Failure to deliver a bounce message always caused it to be frozen, even if
3383 there was an errors_to setting on the router. The errors_to setting is now
3386 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
3387 canonical form (fully expanded) before being placed in
3388 $sender_host_address.
3390 36. The table in the code that translates DNS record types into text (T_A to
3391 "A" for instance) was missing entries for NS and CNAME. It is just possible
3392 that this could have caused confusion if both these types were looked up
3393 for the same domain, because the text type is used as part of Exim's
3394 per-process caching. But the chance of anyone hitting this buglet seems
3397 37. The dnsdb lookup has been extended in a number of ways.
3399 (1) There is a new type, "zns", which walks up the domain tree until it
3400 finds some nameserver records. It should be used with care.
3402 (2) There is a new type, "mxh", which is like "mx" except that it returns
3403 just the host names, not the priorities.
3405 (3) It is now possible to give a list of domains (or IP addresses) to be
3406 looked up. The behaviour when one of the lookups defers can be
3407 controlled by a keyword.
3409 (4) It is now possible to specify the separator character for use when
3410 multiple records are returned.
3412 38. The dnslists ACL condition has been extended: it is now possible to supply
3413 a list of IP addresses and/or domains to be looked up in a particular DNS
3416 39. Added log_selector=+queue_time_overall.
3418 40. When running the queue in the test harness, wait just a tad after forking a
3419 delivery process, to get repeatability of debugging output.
3421 41. Include certificate and key file names in error message when GnuTLS fails
3422 to set them up, because the GnuTLS error message doesn't include the name
3423 of the failing file when there is a problem reading it.
3425 42. Allow both -bf and -bF in the same test run.
3427 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
3429 44. The "Exiscan patch" is now merged into the mainline Exim source.
3431 45. Sometimes the final signoff response after QUIT could fail to get
3432 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
3433 < 0 before doing a fflush(). This bug looks as though it goes back to the
3434 introduction of TLS in release 3.20, but "sometimes" must have been rare
3435 because the tests only now provoked it.
3437 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
3438 (this can affect the format of dates).
3440 47. exim_tidydb, when checking for the continued existence of a message for
3441 which it has found a message-specific retry record, was not finding
3442 messages that were in split spool directories. Consequently, it was
3443 deleting retry records that should have stayed in existence.
3445 48. Steve fixed some bugs in eximstats.
3447 49. The SPA authentication driver was not abandoning authentication and moving
3448 on to the next authenticator when an expansion was forced to fail,
3449 contradicting the general specification for all authenticators. Instead it
3450 was generating a temporary error. It now behaves as specified.
3452 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
3453 (the order specifies the preference for clients). The order is now AES256,
3454 AES128, 3DES, ARCFOUR128.
3456 51. Small patch to Sieve code - explicitly set From: when generating an
3459 52. Exim crashed if a remote delivery caused a very long error message to be
3460 recorded - for instance if somebody sent an entire SpamAssassin report back
3461 as a large number of 550 error lines. This bug was coincidentally fixed by
3462 increasing the size of one of Exim's internal buffers (big_buffer) that
3463 happened as part of the Exiscan merge. However, to be on the safe side, I
3464 have made the code more robust (and fixed the comments that describe what
3467 53. Now that there can be additional text after "Completed" in log lines (if
3468 the queue_time_overall log selector is set), a one-byte patch to exigrep
3469 was needed to allow it to recognize "Completed" as not the last thing in
3472 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
3473 patch that reportedly fixes this has been added. I am not expert enough to
3474 create a test for it. This is what the patch creator wrote:
3476 "I found a little strange behaviour of ldap code when working with
3477 Windows 2003 AD Domain, where users was placed in more than one
3478 Organization Units. When I tried to give exim partial DN, the exit code
3479 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
3480 But simultaneously result of request was absolutely normal ldap result,
3481 so I produce this patch..."
3483 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
3484 REFERENCE, so I have modified the code to exclude the patch when that macro
3487 55. Some experimental protocols are using DNS PTR records for new purposes. The
3488 keys for these records are domain names, not reversed IP addresses. The
3489 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
3490 leaves it alone. Component reversal etc. now happens only for IP addresses.
3493 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
3495 57. Double the size of the debug message buffer (to 2048) so that more of very
3496 long debug lines gets shown.
3498 58. The exicyclog utility now does better if the number of log files to keep
3499 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
3501 59. Two changes related to the smtp_active_hostname option:
3503 (1) $smtp_active_hostname is now available as a variable.
3504 (2) The default for smtp_banner uses $smtp_active_hostname instead
3505 of $primary_hostname.
3507 60. The host_aton() function is supposed to be passed a string that is known
3508 to be a valid IP address. However, in the case of IPv6 addresses, it was
3509 not checking this. This is a hostage to fortune. Exim now panics and dies
3510 if the condition is not met. A case was found where this could be provoked
3511 from a dnsdb PTR lookup with an IPv6 address that had more than 8
3512 components; fortuitously, this particular loophole had already been fixed
3513 by change 4.50/55 above.
3515 If there are any other similar loopholes, the new check in host_aton()
3516 itself should stop them being exploited. The report I received stated that
3517 data on the command line could provoke the exploit when Exim was running as
3518 exim, but did not say which command line option was involved. All I could
3519 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
3520 running as the user.
3523 61. There was a buffer overflow vulnerability in the SPA authentication code
3524 (which came originally from the Samba project). I have added a test to the
3525 spa_base64_to_bits() function which I hope fixes it.
3528 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
3529 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
3531 63. The daemon start-up calls getloadavg() while still root for those OS that
3532 need the first call to be done as root, but it missed one case: when
3533 deliver_queue_load_max is set with deliver_drop_privilege. This is
3534 necessary for the benefit of the queue runner, because there is no re-exec
3535 when deliver_drop_privilege is set.
3537 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
3538 This has been fixed.
3540 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
3541 was in use, was not putting the data itself into the right store pool;
3542 consequently, it could be overwritten for a subsequent message in the same
3543 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
3546 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
3548 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
3549 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
3550 the function were treating the return as a boolean value, which happened to
3551 work because 0=false and not-0=true, but is not correct code.
3553 68. The host_aton() function was not handling scoped IPv6 addresses (those
3554 with, for example, "%eth0" on the end) correctly.
3556 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
3557 negated items (that is, ~something) in unsigned ints. Some compilers
3558 apparently mutter when there is no cast.
3560 70. If an address verification called from an ACL failed, and did not produce a
3561 user-specific message (i.e. there was only a "system" message), nothing was
3562 put in $acl_verify_message. In this situation, it now puts the system
3565 71. Change 4.23/11 added synchronization checking at the start of an SMTP
3566 session; change 4.31/43 added the unwanted input to the log line - except
3567 that it did not do this in the start of session case. It now does.
3569 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
3570 This is wrong and can cause the other end to generate a synchronization
3571 error if it is another Exim or anything else that does the synchronization
3572 check. A QUIT command is no longer sent after a timeout.
3574 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
3575 during host lookups.
3577 74. The defer_ok option of callout verification was not working if it was used
3578 when verifying addresses in header lines, that is, for this case:
3580 verify = header_sender/callout=defer_ok
3582 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
3583 those file descriptors could be used for SMTP connections. If anything
3584 wrote to stderr (the example that came up was "warn" in embedded Perl), it
3585 could be sent to the SMTP client, causing chaos. The daemon now opens
3586 stdin, stdout, and stderr to /dev/null when it puts itself into the
3589 76. Arrange for output from Perl's "warn" command to be written to Exim's main
3590 log by default. The user can override this with suitable Perl magic.
3592 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
3593 the log message when discard triggers, was not working for the DATA ACL or
3594 for the non-SMTP ACL.
3596 78. Error message wording change in sieve.c.
3598 79. If smtp_accept_max_per_host was set, the number of connections could be
3599 restricted to fewer than expected, because the daemon was trying to set up
3600 a new connection before checking whether the processes handling previous
3601 connections had finished. The check for completed processes is now done
3602 earlier. On busy systems, this bug wouldn't be noticed because something
3603 else would have woken the daemon, and it would have reaped the completed
3606 80. If a message was submitted locally by a user whose login name contained one
3607 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
3608 It caused a spool format error. I have fixed the spool reading code. A
3609 related problem was that the "from" clause in the Received: line became
3610 illegal because of the space(s). It is now covered by ${quote_local_part.
3612 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
3615 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
3617 83. Patch from Sieve maintainer to fix unterminated string problem in
3618 "vacation" handling.
3620 84. Some minor changes to the Linux configuration files to help with other
3621 OS variants using glibc.
3623 85. One more patch for Sieve to update vacation handling to latest spec.
3626 ----------------------------------------------------
3627 See the note above about the 4.44 and 4.50 releases.
3628 ----------------------------------------------------
3634 1. Change 4.43/35 introduced a bug that caused file counts to be
3635 incorrectly computed when quota_filecount was set in an appendfile
3638 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
3639 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
3642 3. Give more explanation in the error message when the command for a transport
3643 filter fails to execute.
3645 4. There are several places where Exim runs a non-Exim command in a
3646 subprocess. The SIGUSR1 signal should be disabled for these processes. This
3647 was being done only for the command run by the queryprogram router. It is
3648 now done for all such subprocesses. The other cases are: ${run, transport
3649 filters, and the commands run by the lmtp and pipe transports.
3651 5. Some older OS have a limit of 256 on the maximum number of file
3652 descriptors. Exim was using setrlimit() to set 1000 as a large value
3653 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
3654 systems. I've change it so that if it can't get 1000, it tries for 256.
3656 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
3657 was an oversight, and furthermore, ever since the addition of extra
3658 controls (e.g. 4.43/32), the checks on when to allow different forms of
3659 "control" were broken. There should now be diagnostics for all cases when a
3660 control that does not make sense is encountered.
3662 7. $recipients is now available in the predata ACL (oversight).
3664 8. Tidy the search cache before the fork to do a delivery from a message
3665 received from the command line. Otherwise the child will trigger a lookup
3666 failure and thereby defer the delivery if it tries to use (for example) a
3667 cached ldap connection that the parent has called unbind on.
3669 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
3670 of $address_data from the recipient verification was clobbered by the
3671 sender verification.
3673 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
3674 was its contents. (It was OK if the option was not defined at all.)
3676 11. A "Completed" log line is now written for messages that are removed from
3677 the spool by the -Mrm option.
3679 12. $host_address is now set to the target address during the checking of
3680 ignore_target_hosts.
3682 13. When checking ignore_target_hosts for an ipliteral router, no host name was
3683 being passed; this would have caused $sender_host_name to have been used if
3684 matching the list had actually called for a host name (not very likely,
3685 since this list is usually IP addresses). A host name is now passed as
3688 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
3689 code that specifies a non-restarting handler (typically sigaction() in
3690 modern systems) in an attempt to fix a rare and obscure crash bug.
3692 15. Narrowed the window for a race in the daemon that could cause it to ignore
3693 SIGCHLD signals. This is not a major problem, because they are used only to
3694 wake it up if nothing else does.
3696 16. A malformed maildirsize file could cause Exim to calculate negative values
3697 for the mailbox size or file count. Odd effects could occur as a result.
3698 The maildirsize information is now recalculated if the size or filecount
3701 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
3702 support for a long time. Removed HAVE_SYS_VFS_H.
3704 18. Updated exipick to current release from John Jetmore.
3706 19. Allow an empty sender to be matched against a lookup in an address list.
3707 Previously the only cases considered were a regular expression, or an
3710 20. Exim went into a mad DNS lookup loop when doing a callout where the
3711 host was specified on the transport, if the DNS lookup yielded more than
3714 21. The RFC2047 encoding function was originally intended for short strings
3715 such as real names; it was not keeping to the 75-character limit for
3716 encoded words that the RFC imposes. It now respects the limit, and
3717 generates multiple encoded words if necessary. To be on the safe side, I
3718 have increased the buffer size for the ${rfc2047: expansion operator from
3721 22. Failure to deliver a bounce message always caused it to be frozen, even if
3722 there was an errors_to setting on the router. The errors_to setting is now
3725 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
3726 canonical form (fully expanded) before being placed in
3727 $sender_host_address.
3729 24. Updated eximstats to version 1.33
3731 25. Include certificate and key file names in error message when GnuTLS fails
3732 to set them up, because the GnuTLS error message doesn't include the name
3733 of the failing file when there is a problem reading it.
3735 26. Expand error message when OpenSSL has problems setting up cert/key files.
3738 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
3739 (this can affect the format of dates).
3741 28. exim_tidydb, when checking for the continued existence of a message for
3742 which it has found a message-specific retry record, was not finding
3743 messages that were in split spool directories. Consequently, it was
3744 deleting retry records that should have stayed in existence.
3746 29. eximstats updated to version 1.35
3747 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
3748 1.35 - bugfix such that pie charts by volume are generated correctly
3750 30. The SPA authentication driver was not abandoning authentication and moving
3751 on to the next authenticator when an expansion was forced to fail,
3752 contradicting the general specification for all authenticators. Instead it
3753 was generating a temporary error. It now behaves as specified.
3755 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
3756 (the order specifies the preference for clients). The order is now AES256,
3757 AES128, 3DES, ARCFOUR128.
3759 31. Small patch to Sieve code - explicitly set From: when generating an
3762 32. Exim crashed if a remote delivery caused a very long error message to be
3763 recorded - for instance if somebody sent an entire SpamAssassin report back
3764 as a large number of 550 error lines. This bug was coincidentally fixed by
3765 increasing the size of one of Exim's internal buffers (big_buffer) that
3766 happened as part of the Exiscan merge. However, to be on the safe side, I
3767 have made the code more robust (and fixed the comments that describe what
3770 33. Some experimental protocols are using DNS PTR records for new purposes. The
3771 keys for these records are domain names, not reversed IP addresses. The
3772 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
3773 leaves it alone. Component reversal etc. now happens only for IP addresses.
3776 34. The host_aton() function is supposed to be passed a string that is known
3777 to be a valid IP address. However, in the case of IPv6 addresses, it was
3778 not checking this. This is a hostage to fortune. Exim now panics and dies
3779 if the condition is not met. A case was found where this could be provoked
3780 from a dnsdb PTR lookup with an IPv6 address that had more than 8
3781 components; fortuitously, this particular loophole had already been fixed
3782 by change 4.50/55 or 4.44/33 above.
3784 If there are any other similar loopholes, the new check in host_aton()
3785 itself should stop them being exploited. The report I received stated that
3786 data on the command line could provoke the exploit when Exim was running as
3787 exim, but did not say which command line option was involved. All I could
3788 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
3789 running as the user.
3792 35. There was a buffer overflow vulnerability in the SPA authentication code
3793 (which came originally from the Samba project). I have added a test to the
3794 spa_base64_to_bits() function which I hope fixes it.
3797 36. The daemon start-up calls getloadavg() while still root for those OS that
3798 need the first call to be done as root, but it missed one case: when
3799 deliver_queue_load_max is set with deliver_drop_privilege. This is
3800 necessary for the benefit of the queue runner, because there is no re-exec
3801 when deliver_drop_privilege is set.
3803 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
3804 was in use, was not putting the data itself into the right store pool;
3805 consequently, it could be overwritten for a subsequent message in the same
3806 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
3809 38. Sometimes the final signoff response after QUIT could fail to get
3810 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
3811 < 0 before doing a fflush(). This bug looks as though it goes back to the
3812 introduction of TLS in release 3.20, but "sometimes" must have been rare
3813 because the tests only now provoked it.
3819 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
3820 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
3821 Now it returns an integer. A number of calls were still expecting a T/F
3822 return. Fortuitously, in all cases, the tests worked in OK situations,
3823 which is the norm. However, things would have gone wrong on any write
3824 failures on the smtp file descriptor. This function is used when sending
3825 messages over SMTP and also when doing verify callouts.
3827 2. When Exim is called to do synchronous delivery of a locally submitted
3828 message (the -odf or -odi options), it no longer closes stderr before doing
3831 3. Implemented the mua_wrapper option.
3833 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
3835 5. Implemented the functions header_remove(), header_testname(),
3836 header_add_at_position(), and receive_remove_recipient(), and exported them
3839 6. If an ACL "warn" statement specified the addition of headers, Exim already
3840 inserted X-ACL-Warn: at the start if there was no header name. However, it
3841 was not making this test for the second and subsequent header lines if
3842 there were newlines in the string. This meant that an invalid header could
3843 be inserted if Exim was badly configured.
3845 7. Allow an ACL "warn" statement to add header lines at the start or after all
3846 the Received: headers, as well as at the end.
3848 8. Added the rcpt_4xx retry error code.
3850 9. Added postmaster_mailfrom=xxx to callout verification option.
3852 10. Added mailfrom=xxxx to the callout verification option, for verify=
3855 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
3856 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
3858 12. Inserted some casts to stop certain compilers warning when using pointer
3859 differences as field lengths or precisions in printf-type calls (mostly
3860 affecting debugging statements).
3862 13. Added optional readline() support for -be (dynamically loaded).
3864 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
3865 same clock tick as a message's arrival, so that its received time was the
3866 same as the "first fail" time on the retry record, and that message
3867 remained on the queue past the ultimate address timeout, every queue runner
3868 would try a delivery (because it was past the ultimate address timeout) but
3869 after another failure, the ultimate address timeout, which should have then
3870 bounced the address, did not kick in. This was a "< instead of <=" error;
3871 in most cases the first failure would have been in the next clock tick
3872 after the received time, and all would be well.
3874 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
3875 being recognized when the domain list was tested by the match_domain
3876 condition in an expansion string.
3878 16. Added the ${str2b64: operator.
3880 17. Exim was always calling setrlimit() to set a large limit for the number of
3881 processes, without checking whether the existing limit was already
3882 adequate. (It did check for the limit on file descriptors.) Furthermore,
3883 errors from getrlimit() and setrlimit() were being ignored. Now they are
3884 logged to the main and panic logs, but Exim does carry on, to try to do its
3885 job under whatever limits there are.
3887 18. Imported PCRE 5.0.
3889 19. Trivial typo in log message " temporarily refused connection" (the leading
3892 20. If the log selector return_path_on_delivery was set and an address was
3893 redirected to /dev/null, the delivery process crashed because it assumed
3894 that a return path would always be set for a "successful" delivery. In this
3895 case, the whole delivery is bypassed as an optimization, and therefore no
3898 21. Internal re-arrangement: the function for sending a challenge and reading
3899 a response while authentication was assuming a zero-terminated challenge
3900 string. It's now changed to take a pointer and a length, to allow for
3901 binary data in such strings.
3903 22. Added the cyrus_sasl authenticator (code supplied by MBM).
3905 23. Exim was not respecting finduser_retries when seeking the login of the
3906 uid under which it was called; it was always trying 10 times. (The default
3907 setting of finduser_retries is zero.) Also, it was sleeping after the final
3908 failure, which is pointless.
3910 24. Implemented tls_on_connect_ports.
3912 25. Implemented acl_smtp_predata.
3914 26. If the domain in control=submission is set empty, Exim assumes that the
3915 authenticated id is a complete email address when it generates From: or
3916 Sender: header lines.
3918 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
3919 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
3920 chown and chgrp in /bin and hostname in /usr/bin.
3922 28. Exim was keeping the "process log" file open after each use, just as it
3923 does for the main log. This opens the possibility of it remaining open for
3924 long periods when the USR1 signal hits a daemon. Occasional processlog
3925 errors were reported, that could have been caused by this. Anyway, it seems
3926 much more sensible not to leave this file open at all, so that is what now
3929 29. The long-running daemon process does not normally write to the log once it
3930 has entered its main loop, and it closes the log before doing so. This is
3931 so that log files can straightforwardly be renamed and moved. However,
3932 there are a couple of unusual error situations where the daemon does write
3933 log entries, and I had neglected to close the log afterwards.
3935 30. The text of an SMTP error response that was received during a remote
3936 delivery was being truncated at 512 bytes. This is too short for some of
3937 the long messages that one sometimes sees. I've increased the limit to
3940 31. It is now possible to make retry rules that apply only when a message has a
3941 specific sender, in particular, an empty sender.
3943 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
3944 it possible to be selective about when SMTP synchronization is enforced.
3946 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
3948 32. Implemented hosts_connection_nolog.
3950 33. Added an ACL for QUIT.
3952 34. Setting "delay_warning=" to disable warnings was not working; it gave a
3955 35. Added mailbox_size and mailbox_filecount to appendfile.
3957 36. Added control = no_multiline_responses to ACLs.
3959 37. There was a bug in the logic of the code that waits for the clock to tick
3960 in the case where the clock went backwards by a substantial amount such
3961 that the microsecond fraction of "now" was more than the microsecond
3962 fraction of "then" (but the whole seconds number was less).
3964 38. Added support for the libradius Radius client library this is found on
3965 FreeBSD (previously only the radiusclient library was supported).
3971 1. When certain lookups returned multiple values in the form name=value, the
3972 quoting of the values was not always being done properly. Specifically:
3973 (a) If the value started with a double quote, but contained no whitespace,
3975 (b) If the value contained whitespace other than a space character (i.e.
3976 tabs or newlines or carriage returns) it was not quoted.
3977 This fix has been applied to the mysql and pgsql lookups by writing a
3978 separate quoting function and calling it from the lookup code. The fix
3979 should probably also be applied to nisplus, ibase and oracle lookups, but
3980 since I cannot test any of those, I have not disturbed their existing code.
3982 2. A hit in the callout cache for a specific address caused a log line with no
3983 reason for rejecting RCPT. Now it says "Previous (cached) callout
3984 verification failure".
3986 3. There was an off-by-one bug in the queryprogram router. An over-long
3987 return line was truncated at 256 instead of 255 characters, thereby
3988 overflowing its buffer with the terminating zero. As well as fixing this, I
3989 have increased the buffer size to 1024 (and made a note to document this).
3991 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
3992 when Exim is waiting for an SMTP response from a remote server, Exim
3993 restarts its select() call on the socket, thereby resetting its timeout.
3994 This is not a problem when such interrupts are rare. Somebody set up a cron
3995 job to run exiwhat every 2 minutes, which is less than the normal select()
3996 timeout (5 or 10 minutes). This meant that the select() timeout never
3997 kicked in because it was always reset. I have fixed this by comparing the
3998 time when an interrupt arrives with the time at the start of the first call
3999 to select(). If more time than the timeout has elapsed, the interrupt is
4000 treated as a timeout.
4002 5. Some internal re-factoring in preparation for the addition of Sieve
4003 extensions (by MH). In particular, the "personal" test is moved to a
4004 separate function, and given an option for scanning Cc: and Bcc: (which is
4005 not set for Exim filters).
4007 6. When Exim created an email address using the login of the caller as the
4008 local part (e.g. when creating a From: or Sender: header line), it was not
4009 quoting the local part when it contained special characters such as @.
4011 7. Installed new OpenBSD configuration files.
4013 8. Reworded some messages for syntax errors in "and" and "or" conditions to
4014 try to make them clearer.
4016 9. Callout options, other than the timeout value, were being ignored when
4017 verifying sender addresses in header lines. For example, when using
4019 verify = header_sender/callout=no_cache
4021 the cache was (incorrectly) being used.
4023 10. Added a missing instance of ${EXE} to the exim_install script; this affects
4024 only the Cygwin environment.
4026 11. When return_path_on_delivery was set as a log selector, if different remote
4027 addresses in the same message used different return paths and parallel
4028 remote delivery occurred, the wrong values would sometimes be logged.
4029 (Whenever a remote delivery process finished, the return path value from
4030 the most recently started remote delivery process was logged.)
4032 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
4033 lines when AUTH and/or TLS are in use. This is the "received protocol"
4034 field. Exim used to use "asmtp" for authenticated SMTP, without any
4035 indication (in the protocol name) for TLS use. Now it follows the RFC and
4036 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
4037 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
4038 names appear in log lines as well as in Received: header lines.
4040 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
4041 extensions, and comparison tests, and to fix some bugs.
4043 14. Changes to the "personal" filter test:
4045 (1) The test was buggy in that it was just doing the equivalent of
4046 "contains" tests on header lines. For example, if a user's address was
4047 anne@some.where, the "personal" test would incorrectly be true for
4049 To: susanne@some.where
4051 This test is now done by extracting each address from the header in turn,
4052 and checking the entire address. Other tests that are part of "personal"
4053 are now done using regular expressions (for example, to check local parts
4054 of addresses in From: header lines).
4056 (2) The list of non-personal local parts in From: addresses has been
4057 extended to include "listserv", "majordomo", "*-request", and "owner-*",
4058 taken from the Sieve specification recommendations.
4060 (3) If the message contains any header line starting with "List-" it is
4061 treated as non-personal.
4063 (4) The test for "circular" in the Subject: header line has been removed
4064 because it now seems ill-conceived.
4066 15. Minor typos in src/EDITME comments corrected.
4068 16. Installed latest exipick from John Jetmore.
4070 17. If headers_add on a router specified a text string that was too long for
4071 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
4072 of string_sprintf() is now avoided.
4074 18. $message_body_size was not set (it was always zero) when running the DATA
4075 ACL and the local_scan() function.
4077 19. For the "mail" command in an Exim filter, no default was being set for
4078 the once_repeat time, causing a random time value to be used if "once" was
4079 specified. (If the value happened to be <= 0, no repeat happened.) The
4080 default is now 0s, meaning "never repeat". The "vacation" command was OK
4081 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
4082 (I found it when inspecting the code).
4084 20. There is now an overall timeout for performing a callout verification. It
4085 defaults to 4 times the callout timeout, which applies to individual SMTP
4086 commands during the callout. The overall timeout applies when there is more
4087 than one host that can be tried. The timeout is checked before trying the
4088 next host. This prevents very long delays if there are a large number of
4089 hosts and all are timing out (e.g. when the network connections are timing
4090 out). The value of the overall timeout can be changed by specifying an
4091 additional sub-option for "callout", called "maxwait". For example:
4093 verify = sender/callout=5s,maxwait=20s
4095 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
4096 to the end before writing, but this should make it even safer).
4098 22. Exim was forgetting that it had advertised PIPELINING for the second and
4099 subsequent messages on an SMTP connection. It was also not resetting its
4100 memory on STARTTLS and an internal HELO.
4102 23. When Exim logs an SMTP synchronization error within a session, it now
4103 records whether PIPELINING has been advertised or not.
4105 24. Added 3 instances of "(long int)" casts to time_t variables that were being
4106 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
4107 rather than long int.
4109 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
4111 26. Added the never_mail option to autoreply.
4117 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
4118 crash if the getsockname() call failed; this can happen if a connection is
4119 closed very soon after it is established. The problem was simply in the
4120 order in which certain operations were done, causing Exim to try to write
4121 to the SMTP stream before it had set up the file descriptor. The bug has
4122 been fixed by making things happen in the correct order.
4128 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
4129 before the connection was closed, thus losing the rejection response.
4131 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
4132 some early Solaris releases, but causes trouble in current releases where
4133 socklen_t is defined.
4135 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
4138 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
4141 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
4142 that releasing the top of it at the end releases what was used for sub-
4143 expansions (unless the block got too big). However, discard this block if
4144 the first thing is a variable or header, so that we can use its block when
4145 it is dynamic (useful for very large $message_headers, for example).
4147 6. Lookups now cache *every* query, not just the most recent. A new, separate
4148 store pool is used for this. It can be recovered when all lookup caches are
4149 flushed. Lookups now release memory at the end of their result strings.
4150 This has involved some general refactoring of the lookup sources.
4152 7. Some code has been added to the store_xxx() functions to reduce the amount
4153 of flapping under certain conditions.
4155 8. log_incoming_interface used to affect only the <= reception log lines. Now
4156 it causes the local interface and port to be added to several more SMTP log
4157 lines, for example "SMTP connection from", and rejection lines.
4159 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
4161 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
4163 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
4164 could be overwritten at the end of the current message (or the start of a
4165 new message if it was set in a HELO ACL). The value is now preserved for
4166 the duration of the SMTP connection.
4168 12. If a transport had a headers_rewrite setting, and a matching header line
4169 contained an unqualified address, that address was qualified, even if it
4170 did not match any rewriting rules. The underlying bug was that the values
4171 of the flags that permit the existence of unqualified sender and recipient
4172 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
4173 non-local messages, and by -bnq for local messages) were not being
4174 preserved with the message after it was received.
4176 13. When Exim was logging an SMTP synchronization error, it could sometimes log
4177 "next input=" as part of the text comprising the host identity instead of
4178 the correct text. The code was using the same buffer for two different
4179 strings. However, depending on which order the printing function evaluated
4180 its arguments, the bug did not always show up. Under Linux, for example, my
4181 test suite worked just fine.
4183 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
4184 doesn't work with some older versions of Perl. It has been changed to "my",
4185 which in any case is probably the better facility to use.
4187 15. A really picky compiler found some instances of statements for creating
4188 error messages that either had too many or two few arguments for the format
4191 16. The size of the buffer for calls to the DNS resolver has been increased
4192 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
4193 for addresses that have a lot of PTR records. This alleviates a problem; it
4194 does not fully solve it.
4196 17. A dnsdb lookup for PTR records that receives more data than will fit in the
4197 buffer now truncates the list and logs the incident, which is the same
4198 action as happens when Exim is looking up a host name and its aliases.
4199 Previously in this situation something unpredictable would happen;
4200 sometimes it was "internal error: store_reset failed".
4202 18. If a server dropped the connection unexpectedly when an Exim client was
4203 using GnuTLS and trying to read a response, the client delivery process
4204 crashed while trying to generate an error log message.
4206 19. If a "warn" verb in an ACL added multiple headers to a message in a single
4207 string, for example:
4209 warn message = H1: something\nH2: something
4211 the text was added as a single header line from Exim's point of view
4212 though it ended up OK in the delivered message. However, searching for the
4213 second and subsequent header lines using $h_h2: did not work. This has been
4214 fixed. Similarly, if a system filter added multiple headers in this way,
4215 the routers could not see them.
4217 20. Expanded the error message when iplsearch is called with an invalid key to
4218 suggest using net-iplsearch in a host list.
4220 21. When running tests using -bh, any delays imposed by "delay" modifiers in
4221 ACLs are no longer actually imposed (and a message to that effect is
4224 22. If a "gecos" field in a passwd entry contained escaped characters, in
4225 particular, if it contained a \" sequence, Exim got it wrong when building
4226 a From: or a Sender: header from that name. A second bug also caused
4227 incorrect handling when an unquoted " was present following a character
4228 that needed quoting.
4230 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
4231 was not being matched caselessly.
4233 24. Arranged for all hyphens in the exim.8 source to be escaped with
4236 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
4237 properly. Recipient callout cache records were still being keyed to include
4238 the sender, even when use_sender was set false. This led to far more
4239 callouts that were necessary. The sender is no longer included in the key
4240 when use_sender is false.
4242 26. Added "control = submission" modifier to ACLs.
4244 27. Added the ${base62d: operator to decode base 62 numbers.
4246 28. dnsdb lookups can now access SRV records.
4248 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
4249 the configuration file.
4251 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
4252 (-v) mode. This makes the output for a verbose queue run more intelligible.
4254 31. Added a use_postmaster feature to recipient callouts.
4256 32. Added the $body_zerocount variable, containing the number of binary zero
4257 bytes in the message body.
4259 33. The time of last modification of the "new" subdirectory is now used as the
4260 "mailbox time last read" when there is a quota error for a maildir
4263 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
4265 35. Added +ignore_unknown as a special item in host lists.
4267 36. Code for decoding IPv6 addresses in host lists is now included, even if
4268 IPv6 support is not being compiled. This fixes a bug in which an IPv6
4269 address was recognized as an IP address, but was then not correctly decoded
4270 into binary, causing unexpected and incorrect effects when compared with
4277 1. Very minor rewording of debugging text in manualroute to say "list of
4278 hosts" instead of "hostlist".
4280 2. If verify=header_syntax was set, and a header line with an unqualified
4281 address (no domain) and a large number of spaces between the end of the
4282 name and the colon was received, the reception process suffered a buffer
4283 overflow, and (when I tested it) crashed. This was caused by some obsolete
4284 code that should have been removed. The fix is to remove it!
4286 3. When running in the test harness, delay a bit after writing a bounce
4287 message to get a bit more predictability in the log output.
4289 4. Added a call to search_tidyup() just before forking a reception process. In
4290 theory, someone could use a lookup in the expansion of smtp_accept_max_
4291 per_host which, without the tidyup, could leave open a database connection.
4293 5. Added the variables $recipient_data and $sender_data which get set from a
4294 lookup success in an ACL "recipients" or "senders" condition, or a router
4295 "senders" option, similar to $domain_data and $local_part_data.
4297 6. Moved the writing of debug_print from before to after the "senders" test
4300 7. Change 4.31/66 (moving the time when the Received: is generated) caused
4301 problems for message scanning, either using a data ACL, or using
4302 local_scan() because the Received: header was not generated till after they
4303 were called (in order to set the time as the time of reception completion).
4304 I have revised the way this works. The header is now generated after the
4305 body is received, but before the ACL or local_scan() are called. After they
4306 are run, the timestamp in the header is updated.
4312 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
4313 before starting a queue runner without re-exec. This happened only when
4314 deliver_drop_privilege was set or when the Exim user was set to root. The
4315 effect of the bug was that timeouts during subsequent deliveries caused
4316 crashes instead of being properly handled. The handler is now left at its
4317 default (and expected) setting.
4319 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
4320 message, again when deliver_drop_privilege is set or Exim is run as root.
4321 The bug described in (1) was not present in this case, but the tidying up
4322 of the other signals was missing. I have made the two cases consistent.
4324 3. The ignore_target_hosts setting on a manualroute router was being ignored
4325 for hosts that were looked up using the /MX notation.
4327 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
4330 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
4331 operated on the sender address. After changing the $sender_address to <>
4332 for the sender address verify, Exim was re-instated it as the original
4333 (before rewriting) address, but remembering that it had rewritten it, so it
4334 wasn't rewriting it again. This bug also had the effect of breaking the
4335 sender address verification caching when the sender address was rewritten.
4337 6. The ignore_target_hosts option was being ignored by the ipliteral router.
4338 This has been changed so that if the ip literal address matches
4339 ignore_target_hosts, the router declines.
4341 7. Added expansion conditions match_domain, match_address, and match_local_
4342 part (NOT match_host).
4344 8. The placeholder for the Received: header didn't have a length field set.
4346 9. Added code to Exim itself and to exim_lock to test for a specific race
4347 condition that could lead to file corruption when using MBX delivery. The
4348 issue is with the lockfile that is created in /tmp. If this file is removed
4349 after a process has opened it but before that process has acquired a lock,
4350 there is the potential for a second process to recreate the file and also
4351 acquire a lock. This could lead to two Exim processes writing to the file
4352 at the same time. The added code performs the same test as UW imapd; it
4353 checks after acquiring the lock that its file descriptor still refers to
4354 the same named file.
4356 10. The buffer for building added header lines was of fixed size, 8192 bytes.
4357 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
4360 11. Added the smtp_active_hostname option. If used, this will typically be made
4361 to depend on the incoming interface address. Because $interface_address is
4362 not set up until the daemon has forked a reception process, error responses
4363 that can happen earlier (such as "too many connections") no longer contain
4366 12. If an expansion in a condition on a "warn" statement fails because a lookup
4367 defers, the "warn" statement is abandoned, and the next ACL statement is
4368 processed. Previously this caused the whole ACL to be aborted.
4370 13. Added the iplsearch lookup type.
4372 14. Added ident_timeout as a log selector.
4374 15. Added tls_certificate_verified as a log selector.
4376 16. Added a global option tls_require_ciphers (compare the smtp transport
4377 option of the same name). This controls incoming TLS connections.
4379 17. I finally figured out how to make tls_require_ciphers do a similar thing
4380 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
4381 before starting the TLS session.
4383 18. Tabs are now shown as \t in -bP output.
4385 19. If the log selector return_path_on_delivery was set, Exim crashed when
4386 bouncing a message because it had too many Received: header lines.
4388 20. If two routers both had headers_remove settings, and the first one included
4389 a superfluous trailing colon, the final name in the first list and the
4390 first name in the second list were incorrectly joined into one item (with a
4391 colon in the middle).
4397 1. Added -C and -D options to the exinext utility, mainly to make it easier
4398 to include in the automated testing, but these could be helpful when
4399 multiple configurations are in use.
4401 2. The exinext utility was not formatting the output nicely when there was
4402 an alternate port involved in the retry record key, nor when there was a
4403 message id as well (for retries that were specific to a specific message
4404 and a specific host). It was also confused by IPv6 addresses, because of
4405 the additional colons they contain. I have fixed the IPv4 problem, and
4406 patched it up to do a reasonable job for IPv6.
4408 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
4409 delivery, the log line now contains "pipelined" if PIPELINING was used.
4411 4. An SMTP transport process used to panic and die if the bind() call to set
4412 an explicit outgoing interface failed. This has been changed; it is now
4413 treated in the same way as a connect() failure.
4415 5. A reference to $sender_host_name in the part of a conditional expansion
4416 that was being skipped was still causing a DNS lookup. This no longer
4419 6. The def: expansion condition was not recognizing references to header lines
4420 that used bh_ and bheader_.
4422 7. Added the _cache feature to named lists.
4424 8. The code for checking quota_filecount in the appendfile transport was
4425 allowing one more file than it should have been.
4427 9. For compatibility with Sendmail, the command line option
4435 and sets the incoming protocol and host name (for trusted callers). The
4436 host name and its colon can be omitted when only the protocol is to be set.
4437 Note the Exim already has two private options, -pd and -ps, that refer to
4438 embedded Perl. It is therefore impossible to set a protocol value of "d" or
4439 "s", but I don't think that's a major issue.
4441 10. A number of refactoring changes to the code, none of which should affect
4444 (a) The number of logging options was getting close to filling up the
4445 32-bit word that was used as a bit map. I have split them into two classes:
4446 those that are passed in the argument to log_write(), and those that are
4447 only ever tested independently outside of that function. These are now in
4448 separate 32-bit words, so there is plenty of room for expansion again.
4449 There is no change in the user interface or the logging behaviour.
4451 (b) When building, for example, log lines, the code previously used a
4452 macro that called string_cat() twice, in order to add two strings. This is
4453 not really sufficiently general. Furthermore, there was one instance where
4454 it was actually wrong because one of the argument was used twice, and in
4455 one call a function was used. (As it happened, calling the function twice
4456 did not affect the overall behaviour.) The macro has been replaced by a
4457 function that can join an arbitrary number of extra strings onto a growing
4460 (c) The code for expansion conditions now uses a table and a binary chop
4461 instead of a serial search (which was left over from when there were very
4462 few conditions). Also, it now recognizes conditions like "pam" even when
4463 the relevant support is not compiled in: a suitably worded error message is
4464 given if an attempt is made to use such a condition.
4466 11. Added ${time_interval:xxxxx}.
4468 12. A bug was causing one of the ddress fields not to be passed back correctly
4469 from remote delivery subprocesses. The field in question was not being
4470 subsequently used, so this caused to problems in practice.
4472 13. Added new log selectors queue_time and deliver_time.
4474 14. Might have fixed a bug in maildirsizefile handling that threw up
4475 "unexpected character" debug warnings, and recalculated the data
4476 unnecessarily. In any case, I expanded the warning message to give more
4479 15. Added the message "Restricted characters in address" to the statements in
4480 the default ACL that block characters like @ and % in local parts.
4482 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
4483 Three changes have been made:
4485 (a) There was a serious bug; a negative response to MAIL caused the whole
4486 recipient domain to be cached as invalid, thereby blocking all messages
4487 to all local parts at the same domain, from all senders. This bug has
4488 been fixed. The domain is no longer cached after a negative response to
4489 MAIL if the sender used is not empty.
4491 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
4494 (c) A new callout option, "use_sender" has been added for people who want
4495 the modified behaviour.
4501 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
4504 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
4505 indeed breaks things for older releases.
4507 3. Added additional logging to the case where there is a problem reading data
4508 from a filter that is running in a subprocess using a pipe, in order to
4509 try to track down a specific problem.
4511 4. Testing facility fudge: when running in the test harness and attempting
4512 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
4513 getting "No route to host". Convert this to a timeout.
4515 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
4518 6. Some OS don't have socklen_t but use size_t instead. This affects the
4519 fifth argument of getsockopt() amongst other things. This is now
4520 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
4521 can be set for individual OS. I have set it for SunOS5, OSF1, and
4522 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
4523 some earlier ones do not.
4525 7. Change 4.30/15 was not doing the test caselessly.
4527 8. The standard form for an IPv6 address literal was being rejected by address
4528 parsing in, for example, MAIL and RCPT commands. An example of this kind of
4529 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
4530 this, as well as the form without the "IPv6" on the front (but only when
4531 address literals are enabled, of course).
4533 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
4535 10. Exim crashed if a message with an empty sender address specified by -f
4536 encountered a router with an errors_to setting. This could be provoked only
4537 by a command such as
4541 where an empty string was supplied; "<>" did not hit this bug.
4543 11. Installed PCRE release 4.5.
4545 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
4546 remained set. It is now erased.
4548 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
4549 times from message ids (which are base 36 rather than the normal 62).
4551 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
4552 were being counted as actual protocol errors, and logged if the log
4553 selector +smtp_protocol_error was set. One cannot be perfect in this test,
4554 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
4555 and DATA following a set of rejected RCPTs do not count as protocol errors.
4556 In other words, Exim assumes they were pipelined, though this may not
4557 actually be the case. Of course, in all cases the client gets an
4558 appropriate error code.
4560 15. If a lookup fails in an ACL condition, a message about the failure may
4561 be available; it is used if testing the ACL cannot continue, because most
4562 such messages specify what the cause of the deferral is. However, some
4563 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
4564 that caused an old message to be retained and used if a later statement
4565 caused a defer, replacing the real cause of the deferral.
4567 16. If an IP address had so many PTR records that the DNS lookup buffer
4568 was not large enough to hold them, Exim could crash while trying to process
4569 the truncated data. It now detects and logs this case.
4571 17. Further to 4.21/58, another change has been made: if (and only if) the
4572 first line of a message (the first header line) ends with CRLF, a bare LF
4573 in a subsequent header line has a space inserted after it, so as not to
4574 terminate the header.
4576 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
4577 unnecessarily, used atoi() instead of strtol(), and didn't check the
4578 termination when getting file sizes from file names by regex.
4580 19. Completely re-implemented the support for maildirsize files, in the light
4581 of a number of problems with the previous contributed implementation
4582 (4.30/29). In particular:
4584 . If the quota is zero, the maildirsize file is maintained, but no quota is
4587 . If the maildir directory does not exist, it is created before any attempt
4588 to write a maildirsize file.
4590 . The quota value in the file is just a cache; if the quota is changed in
4591 the transport, the new value overrides.
4593 . A regular expression is available for excluding directories from the
4596 20. The autoreply transport checks the characters in options that define the
4597 message's headers; it allows continued headers, but it was checking with
4598 isspace() after an embedded newline instead of explicitly looking for a
4601 21. If all the "regular" hosts to which an address was routed had passed their
4602 expiry times, and had not reached their retry times, the address was
4603 bounced, even if fallback hosts were defined. Now Exim should go on to try
4606 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
4607 equivalent code in the SMTP transport. Some hosts send humungous responses
4608 to HELO/EHLO, more than 1024 it seems.
4610 23. Refactoring: code in filter.c used (void *) for "any old type" but this
4611 gives compiler warnings in some environments. I've now done it "properly",
4614 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
4615 (because of problems with the built-in one) was declared to return uschar *
4616 instead of char *, causing compiler failure.
4618 25. Fixed a file descriptor leak when processing alias/forward files.
4620 26. Fixed a minor format string issue in dbfn.c.
4622 27. Typo in exim.c: ("dmbnz" for "dbmnz").
4624 28. If a filter file refered to $h_xxx or $message_headers, and the headers
4625 contained RFC 2047 "words", Exim's memory could, under certain conditions,
4628 29. When a sender address is verified, it is cached, to save repeating the test
4629 when there is more than one recipient in a message. However, when the
4630 verification involves a callout, it is possible for different callout
4631 options to be set for different recipients. It is too complicated to keep
4632 track of this in the cache, so now Exim always runs a verification when a
4633 callout is required, relying on the callout cache for the optimization.
4634 The overhead is duplication of the address routing, but this should not be
4637 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
4638 to be verified with callout=postmaster, and the main callout worked but the
4639 postmaster check failed, the verification correctly failed. However, if a
4640 subsequent RCPT command asked for sender verification *without* the
4641 postmaster check, incorrect caching caused this verification also to fail,
4644 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
4645 it was not caching the DNS options (qualify_single, search_parents) that
4646 were used when the lookup failed. A subsequent lookup with different
4647 options therefore always gave the same answer, though there were cases
4648 where it should not have. (Example: a "domains = !$mx_any" option on a
4649 dnslookup router: the "domains" option is always processed without any
4650 widening, but the router might have qualify_single set.) Now Exim uses the
4651 cached value only when the same options are set.
4653 32. Added John Jetmore's "exipick" utility to the distribution.
4655 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
4656 than a timeout (e.g. a certificate is required, and is not provided), an
4657 Exim server now closes the connection immediately. Previously it waited for
4658 the client to close - but if the client is SSL, it seems that they each
4659 wait for each other, leading to a delay before one of them times out.
4661 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
4662 maintained 0.8.x compatibility because I don't think many are using it, and
4663 it is clearly obsolete.
4665 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
4668 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
4669 client certificate was expired. A simple patch fixes this, though I don't
4670 understand the full logic of why the verify callback is called multiple
4673 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
4674 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
4675 which causes problems with some clients (such as the Certicom SSL Plus
4676 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
4677 disables the coutermeasure allowing Eudora to connect."
4679 38. Exim was not checking that a write() to a log file succeeded. This could
4680 lead to Bad Things if a log got too big, in particular if it hit a file
4681 size limit. Exim now panics and dies if it cannot write to a log file, just
4682 as it does if it cannot open a log file.
4684 39. Modified OS/Makefile-Linux so that it now contains
4686 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
4688 The two -D definitions ensure that Exim is compiled with large file
4689 support, which makes it possible to handle log files that are bigger than
4692 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
4693 instance) a domain was checked against a named list that involved a lookup,
4694 causing $domain_data to be set, then another domain was checked against the
4695 same list, then the first domain was re-checked, the value of $domain_data
4696 after the final check could be wrong. In particular, if the second check
4697 failed, it could be set empty. This bug probably also applied to
4700 41. The strip_trailing_dot option was not being applied to the address given
4701 with the -f command-line option.
4703 42. The code for reading a message's header from the spool was incrementing
4704 $received_count, but never initializing it. This meant that the value was
4705 incorrect (doubled) while delivering a message in the same process in which
4706 it was received. In the most common configuration of Exim, this never
4707 happens - a fresh exec is done - but it can happen when
4708 deliver_drop_privilege is set.
4710 43. When Exim logs an SMTP synchronization error - client data sent too soon -
4711 it now includes up to 150 characters of the unexpected data in the log
4714 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
4715 and building data strings. The size of both of these buffers was 10 000
4716 bytes - far larger than anybody would *ever* want, thought I. Needless to
4717 say, somebody hit the limit. I have increased the maximum line length to
4718 20 000 and the maximum data length of concatenated lines to 100 000. I have
4719 also fixed two bugs, because there was no checking on these buffers. Tsk,
4720 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
4721 buffer is too small.
4723 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
4724 lsearch lookups. Now it does.
4726 46. When parsing a route_list item in a manualroute router, a fixed-length
4727 buffer was used for the list of hosts. I made this 1024 bytes long,
4728 thinking that nobody would ever have a list of hosts that long. Wrong.
4729 Somebody had a whole pile of complicated expansion conditions, and the
4730 string was silently truncated, leading to an expansion error. It turns out
4731 that it is easier to change to an unlimited length (owing to other changes
4732 that have happened since this code was originally written) than to build
4733 structure for giving a limitation error. The length of the item that
4734 expands into the list of hosts is now unlimited.
4736 47. The lsearch lookup could not handle data where the length of text line was
4737 more than 4095 characters. Such lines were truncated, leading to shortened
4738 data being returned. It should now handle lines of any length.
4740 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
4741 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
4744 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
4745 replacing strings such as BIN_DIRECTORY in a source file: the replacement
4746 no longer happens in comment lines. A list of replacements is now placed
4747 at the head of all of the source files, except those whose only change is
4748 to replace PERL_COMMAND in the very first #! line.
4750 50. Replaced the slow insertion sort in queue.c, for sorting the list of
4751 messages on the queue, with a bottom-up merge sort, using code contributed
4752 by Michael Haardt. This should make operations like -bp somewhat faster on
4753 large queues. It won't affect queue runners, except when queue_run_in_order
4756 51. Installed eximstats 1.31 in the distribution.
4758 52. Added support for SRV lookups to the dnslookup router.
4760 53. If an ACL referred to $message_body or $message_body_end, the value was not
4761 reset for any messages that followed in the same SMTP session.
4763 54. The store-handling optimization for building very long strings was not
4764 differentiating between the different store pools. I don't think this
4765 actually made any difference in practice, but I've tidied it.
4767 55. While running the routers to verify a sender address, $sender_address
4768 was still set to the sender address. This is wrong, because when routing to
4769 send a bounce to the sender, it would be empty. Therefore, I have changed
4770 it so that, while verifying a sender address, $sender_address is set to <>.
4771 (There is no change to what happens when verifying a recipient address.)
4773 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
4774 target A or AAAA records (if not already returned) without resetting the
4775 qualify_single or search_parents options of the DNS resolver. These are
4776 inappropriate in this case because the targets of MX and SRV records must
4777 be FQDNs. A broken DNS record could cause trouble if it happened to have a
4778 target that, when qualified, matched something in the local domain. These
4779 two options are now turned off when doing these lookups.
4781 57. It seems that at least some releases of Reiserfs (which does not have the
4782 concept of a fixed number of inodes) returns zero and not -1 for the
4783 number of available inodes. This interacted badly with check_spool_inodes,
4784 which assumed that -1 was the "no such thing" setting. What I have done is
4785 to check that the total number of inodes is greater than zero before doing
4786 the test of how many are available.
4788 58. When a "warn" ACL statement has a log_message modifier, the message is
4789 remembered, and not repeated. This is to avoid a lot of repetition when a
4790 message has many recipients that cause the same warning to be written.
4791 Howewer, Exim was preserving the list of already written lines for an
4792 entire SMTP session, which doesn't seem right. The memory is now reset if a
4793 new message is started.
4795 59. The "rewrite" debugging flag was not showing the result of rewriting in the
4796 debugging output unless log_rewrite was also set.
4798 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
4799 of (int)(handle) when we know that handle contains (void *)(-1).
4801 61. The Exim daemon panic-logs an error return when it closes the incoming
4802 connection. However "connection reset by peer" seems to be common, and
4803 isn't really an error worthy of noting specially, so that particular error
4806 62. When Exim is trying to find all the local interfaces, it used to panic and
4807 die if the ioctl to get the interface flags failed. However, it seems that
4808 on at least one OS (Solaris 9) it is possible to have an interface that is
4809 included in the list of interfaces, but for which you get a failure error
4810 for this call. This happens when the interface is not "plumbed" into a
4811 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
4812 failure of the "get flags" call assumes that the interface is down.
4814 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
4815 makes life easier for people who are doing arithmetic on fields extracted
4816 from dates, where you often get leading zeros that should not be
4817 interpreted as octal.
4819 64. Added qualify_domain to the redirect router, to override the global
4822 65. If a pathologically long header line contained very many addresses (the
4823 report of this problem mentioned 10 000) and each of them was rewritten,
4824 Exim could use up a very large amount of memory. (It kept on making new
4825 copies of the header line as it rewrote, and never released the old ones.)
4826 At the expense of a bit more processing, the header rewriting function has
4827 been changed so that it no longer eats memory in this way.
4829 66. The generation of the Received: header has been moved from the time that a
4830 message starts to be received, to the time that it finishes. The timestamp
4831 in the Received: header should now be very close to that of the <= log
4832 line. There are two side-effects of this change:
4834 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
4835 logged header lines no longer include the local Received: line, because
4836 it has not yet been created. The same applies to a copy of the message
4837 that is returned to a non-SMTP sender when a message is rejected.
4839 (b) When a filter file is tested using -bf, no additional Received: header
4840 is added to the test message. After some thought, I decided that this
4843 This change does not affect the value of $received_for. It is still set
4844 after address rewriting, but before local_scan() is called.
4846 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
4848 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
4849 gave an unhelpful panic error message, and a defer error. I have managed to
4850 change this behaviour so that it now rejects any supplied certificate,
4851 which seems right, as the list of acceptable certificates is empty.
4853 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
4854 gave an unhelpful defer error. I have not managed to make this reject any
4855 supplied certificates, but the error message it gives is "no certificate
4856 supplied", which is not helpful.
4858 70. exigrep's output now also includes lines that are not associated with any
4859 message, but which match the given pattern. Implemented by a patch from
4860 Martin Sluka, which also tidied up the Perl a bit.
4862 71. Recipient callout verification, like sender verification, was using <> in
4863 the MAIL FROM command. This isn't really the right thing, since the actual
4864 sender may affect whether the remote host accepts the recipient or not. I
4865 have changed it to use the actual sender in the callout; this means that
4866 the cache record is now keyed on a recipient/sender pair, not just the
4867 recipient address. There doesn't seem to be a real danger of callout loops,
4868 since a callout by the remote host to check the sender would use <>.
4869 [SEE ABOVE: changed after hitting problems.]
4871 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
4872 temporary errors. However, in the case of such a code being given after
4873 the end of a data transmission (i.e. after ".") Exim was failing to write
4874 a retry record for the message. (Yes, there was some broken host that was
4875 actually sending 8xx at this point.)
4877 73. An unknown lookup type in a host list could cause Exim to panic-die when
4878 the list was checked. (An example that provoked this was putting <; in the
4879 middle of a list instead of at the start.) If this happened during a DATA
4880 ACL check, a -D file could be left lying around. This kind of configuration
4881 error no longer causes Exim to die; instead it causes a defer errror. The
4882 incident is still logged to the main and panic logs.
4884 74. Buglet left over from Exim 3 conversion. The message "too many messages
4885 in one connection" was written to the rejectlog but not the mainlog, except
4886 when address rewriting (yes!) was being logged.
4888 75. Added write_rejectlog option.
4890 76. When a system filter was run not as root (that is, when system_filter_user
4891 was set), the values of the $n variables were not being returned to the
4892 main process; thus, they were not subsequently available in the $sn
4895 77. Added +return_path_on_delivery log selector.
4897 78. A connection timeout was being treated differently from recipients deferred
4898 when testing hosts_max_try with a message that was older than the host's
4899 retry timeout. (The host should not be counted, thus allowing all hosts to
4900 be tried at least once before bouncing.) This may have been the cause of an
4901 occasionally reported bug whereby a message would remain on the queue
4902 longer than the retry timeout, but would be bounced if a delivery was
4903 forced. I say "may" because I never totally pinned down the problem;
4904 setting up timeout/retry tests is difficult. See also the next item.
4906 79. The ultimate address timeout was not being applied to errors that involved
4907 a combination of host plus message (for example, a timeout on a MAIL
4908 command). When an address resolved to a number of possible hosts, and they
4909 were not all tried for each delivery (e.g. because of hosts_max_try), a
4910 message could remain on the queue longer than the retry timeout.
4912 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
4915 81. Fixed an obscure SMTP outgoing bug which required at least the following
4916 conditions: (a) there was another message waiting for the same server;
4917 (b) the server returned 5xx to all RCPT commands in the first message so
4918 that the message was not completed; (c) the server dropped the connection
4919 or gave a negative response to the RSET that Exim sends to abort the
4920 transaction. The observed case was a dropped connection after DATA that had
4921 been sent in pipelining mode. That is, the server had advertised PIPELINING
4922 but was not implementing it correctly. The effect of the bug was incorrect
4923 behaviour, such as trying another host, and this could lead to a crash.
4929 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
4930 and daemon.c were passed as pointers to ints; they should have been
4931 pointers to socklen_t variables (which are typically unsigned ints).
4933 2. Some signed/unsigned type warnings in the os.c file for Linux have been
4936 3. Fixed a really odd bug that affected only the testing scheme; patching a
4937 certain fixed string in the binary changed the value of another string that
4938 happened to be identical to the end of the original first string.
4940 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
4941 name", it returns that address as the IP address. On some operating
4942 systems (e.g. Solaris), it also passes back the IP address string as the
4943 "host name". However, on others (e.g. Linux), it passes back an empty
4944 string. Exim wasn't checking for this, and was changing the host name to an
4945 empty string, assuming it had been canonicized.
4947 5. Although rare, it is permitted to have more than one PTR record for a given
4948 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
4949 all the names associated with an address, because they do in Solaris.
4950 However, it seems that they do not in Linux for data that comes from the
4951 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
4952 I found this out when I moved to a new Linux workstation and tried to run
4953 the Exim test suite.
4955 To get round this problem I have changed the code so that it now does its
4956 own call to the DNS to look up PTR records when searching for a host name.
4957 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
4958 addresses that are only in /etc/hosts are still found.
4960 This behaviour is, however, controlled by an option called host_lookup_
4961 order, which defaults to "bydns:byaddr". If people want to use the other
4962 order, or indeed, just use one or the other means of lookup, they can
4963 specify it in this variable.
4965 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
4966 some operating systems, this comes back from gethostbyaddr() as an empty
4967 string, and this is what Exim used to test for. However, it seems that in
4968 other systems, "." is yielded. Exim now tests for this case too.
4970 7. The values of check_spool_space and check_log_space are now held internally
4971 as a number of kilobytes instead of an absolute number of bytes. If a
4972 numbers is specified without 'K' or 'M', it is rounded up to the nearest
4973 kilobyte. This means that much larger values can be stored.
4975 8. Exim monitor: an attempt to get the action menu when not actually pointing
4976 at a message produces an empty menu entitled "No message selected". This
4977 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
4978 no entries in it ("Shell widget menu has zero width and/or height"). So I
4979 have added a single, blank menu entry in this case.
4981 9. Added ${quote_local_part.
4983 10. MIME decoding is now applied to the contents of Subject: header lines when
4986 11. Now that a reference to $sender_host_address automatically causes a reverse
4987 lookup to occur if necessary (4.13/18), there is no need to arrange for a
4988 host lookup before query-style lookups in lists that might use this
4989 variable. This has therefore been abolished, and the "net-" prefix is no
4990 longer necessary for query-style lookups.
4992 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
4993 have been a typo for LFLAGS, so it has been changed.
4995 13. The install script calls Exim with "-C /dev/null" in order to find the
4996 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
4997 to be output. Howeve, since Exim outputs its version number before the
4998 error, it didn't break the script. It just looked ugly. I fixed this by
4999 always allowing "-C /dev/null" if the caller is root.
5001 14. Ignore overlarge ACL variable number when reading spool file - insurance
5002 against a later release with more variables having written the file.
5004 15. The standard form for an IPv6 address literal was being rejected by EHLO.
5005 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5006 this, as well as the form without the "IPv6" on the front.
5008 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
5009 OS/Makefile-Darwin file.
5011 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
5012 only to LDAP libraries that do not have LDAP_OPT_DEREF.
5014 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
5015 variable. However, some OS use ints for inodes. I've added cast to long int
5016 to get rid of the compiler warning.
5018 19. I had forgotten to lock out "/../" in configuration file names when
5019 ALT_CONFIG_PREFIX was set.
5021 20. Routers used for verification do not need to specify transports. However,
5022 if such a router generated a host list, and callout was configured, Exim
5023 crashed, because it could not find a port number from the (non-existent)
5024 transport. It now assumes port 25 in this circumstance.
5026 21. Added the -t option to exigrep.
5028 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
5029 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
5030 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
5031 them is set, LOOKUP_LSEARCH is forced.
5033 23. "exim -bV" now outputs a list of lookups that are included in the binary.
5035 24. Added sender and host information to the "rejected by local_scan()" log
5036 line; previously there was no indication of these.
5038 25. Added .include_if_exists.
5040 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
5041 Linux. It turns out that not all file systems support this. Apparently some
5042 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
5043 do it.) To cope with this, the error EINVAL, which means that sync-ing is
5044 not supported on the file descriptor, is now ignored when Exim is trying to
5045 sync a directory. This applies only to Linux.
5047 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
5049 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
5050 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
5053 29. Added support for maildirsize files from supplied patch (modified a bit).
5055 30. The use of :fail: followed by an empty string could lead Exim to respond to
5056 sender verification failures with (e.g.):
5058 550 Verification failed for <xxx>
5059 550 Sender verify failed
5061 where the first response line was missing the '-' that indicates it is not
5062 the final line of the response.
5064 31. The loop for finding the name of the user that called Exim had a hardwired
5065 limit of 10; it now uses the value of finduser_retries, which is used for
5066 all other user lookups.
5068 32. Added $received_count variable, available in data and not_smtp ACLs, and at
5071 33. Exim was neglecting to zero errno before one call of strtol() when
5072 expanding a string and expecting an integer value. On some systems this
5073 resulted in spurious "integer overflow" errors. Also, it was casting the
5074 result into an int without checking.
5076 34. Testing for a connection timeout using "timeout_connect" in the retry rules
5077 did not work. The code looks as if it has *never* worked, though it appears
5078 to have been documented since at least releast 1.62. I have made it work.
5080 35. The "timeout_DNS" error in retry rules, also documented since at least
5081 1.62, also never worked. As it isn't clear exactly what this means, and
5082 clearly it isn't a major issue, I have abolished the feature by treating it
5083 as "timeout", and writing a warning to the main and panic logs.
5085 36. The display of retry rules for -brt wasn't always showing the error code
5088 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
5089 timeout_connect_A, timeout_connect_MX.
5091 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
5092 to the empty sender.
5094 39. The daemon was not analysing the content of -oX till after it had closed
5095 stderr and disconnected from the controlling terminal. This meant that any
5096 syntax errors were only noted on the panic log, and the return code from
5097 the command was 0. By re-arranging the code a little, I've made the
5098 decoding happen first, so such errors now appear on stderr, and the return
5099 code is 1. However, the actual setting up of the sockets still happens in
5100 the disconnected process, so errors there are still only recorded on the
5103 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
5104 connections (as happens on some IP stacks) was logged at start up time as
5105 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
5106 it from "IPv6 and IPv4", which means that two separate sockets are being
5109 41. The debug output for gethostbyname2() or getipnodebyname() failures now
5110 says whether AF_INET or AF_INET6 was passed as an argument.
5112 42. Exiwhat output was messed up when time zones were included in log
5115 43. Exiwhat now gives more information about the daemon's listening ports,
5116 and whether -tls-on-connect was used.
5118 44. The "port" option of the smtp transport is now expanded.
5120 45. A "message" modifier in a "warn" statement in a non-message ACL was being
5121 silently ignored. Now an error message is written to the main and panic
5124 46. There's a new ACL modifier called "logwrite" which writes to a log file
5125 as soon as it is encountered.
5127 47. Added $local_user_uid and $local_user_gid at routing time.
5129 48. Exim crashed when trying to verify a sender address that was being
5132 49. Exim was recognizing only a space character after ".include". It now also
5133 recognizes a tab character.
5135 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
5136 extracting the relevant information from the specification. The man page no
5137 longer contains scrambled data for the -d option, and I've added a section
5138 at the front about calling Exim under different names.
5140 51. Added "extra_headers" argument to the "mail" command in filter files.
5142 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
5145 53. Installed eximstats 1.29.
5147 54. Added transport_filter_timeout as a generic transport option.
5149 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
5150 Cc: header lines. This was required by RFC 822, but it not required by RFC
5153 56. Exim used to add From:, Date:, and Message-Id: header lines to any
5154 incoming messages that did not have them. Now it does so only if the
5155 message originates locally, that is, if there is no associated remote host
5156 address. When Resent- header lines are present, this applies to the Resent-
5157 lines rather than the non-Resent- lines.
5159 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
5160 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
5162 58. Messages for configuration errors now include the name of the main
5163 configuration file - useful now that there may be more than one file in a
5164 list (.included file names were always shown).
5166 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
5167 for those rare installations that do not start the daemon as root or run it
5168 setuid root. I've cut out the call to initgroups() if the daemon is not
5171 60. The Exim user and group can now be bound into the binary as text strings
5172 that are looked up at the start of Exim's processing.
5174 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
5176 62. Added $mailstore_basename variable.
5178 63. Installed patch to sieve.c from Michael Haardt.
5180 64. When Exim failed to open the panic log after failing to open the main log,
5181 the original message it was trying to log was written to stderr and debug
5182 output, but if they were not available (the usual case in production), it
5183 was lost. Now it is written to syslog before the two lines that record the
5184 failures to open the logs.
5186 65. Users' Exim filters run in subprocesses under the user's uid. It is
5187 possible for a "deliver" command or an alias in a "personal" command to
5188 provoke an address rewrite. If logging of address rewriting is configured,
5189 this fails because the process is not running as root or exim. There may be
5190 a better way of dealing with this, but for the moment (because 4.30 needs
5191 to be released), I have disabled address rewrite logging when running a
5192 filter in a non-root, non-exim process.
5198 1. The buildconfig auxiliary program wasn't quoting the value set for
5199 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
5200 not defined. This bug was masked in 4.22 by the effect that was fixed in
5203 2. Some messages that were rejected after a message id was allocated were
5204 shown as "incomplete" by exigrep. It no longer does this for messages that
5205 are rejected by local_scan() or the DATA or non-SMTP ACLs.
5207 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
5208 have allow_domain_literals set, the ID did not get logged in the <= line.
5209 Domain literals are now always recognized in Message-ID: header lines.
5211 4. The first argument for a ${extract expansion item is the key name or field
5212 number. Leading and trailing spaces in this item were not being ignored,
5213 causing some misleading effects.
5215 5. When deliver_drop_privilege was set, single queue runner processes started
5216 manually (i.e. by the command "exim -q") or by the daemon (which uses the
5217 same command in the process it spins off) were not dropping privilege.
5219 6. When the daemon running as "exim" started a queue runner, it always
5220 re-executed Exim in the spun-off process. This is a waste of effort when
5221 deliver_drop_privilege is set. The new process now just calls the
5222 queue-runner function directly.
5228 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
5231 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
5232 ignored. Though the use of -oP was forcing the writing of a pid file, it
5233 was always written to the default place.
5235 3. If the message "no IP address found for host xxxx" is generated during
5236 incoming verification, it is now followed by identification of the incoming
5237 connection (so you can more easily find what provoked it).
5239 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
5241 5. Added some features to "harden" Exim a bit more against certain attacks:
5243 (a) There is now a build-time option called FIXED_NEVER_USERS that can
5244 be put in Local/Makefile. This is like the never_users runtime option,
5245 but it cannot be overridden. The default setting is "root".
5247 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
5248 prefix string with which any file named in a -C command line option
5251 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
5252 is retained for -C and -D only if the caller of Exim is root. Without
5253 it, the exim user may also use -C and -D and retain privilege.
5255 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
5256 command line option is disabled.
5258 6. Macro names set by the -D option must start with an upper case letter, just
5259 like macro names defined in the configuration file.
5261 7. Added "dereference=" facility to LDAP.
5263 8. Two instances of the typo "uknown" in the source files are fixed.
5265 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
5266 the Configure-Makefile script screwed up while processing it.
5268 10. Incorporated PCRE 4.4.
5270 11. The SMTP synchronization check was not operating right at the start of an
5271 SMTP session. For example, it could not catch a HELO sent before the client
5272 waited for the greeting. There is now a check for outstanding input at the
5273 point when the greeting is written. Because of the duplex, asynchronous
5274 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
5275 way, but not yet received, when the check is performed.
5277 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
5278 on TCP/IP sockets, because this apparently causes some broken clients to
5281 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
5282 unchanged) from the Cygwin maintainer.
5284 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
5285 was supported instead of testing for mbx. Effectively a typo.
5287 15. The spa authenticator server code was not checking that the input it
5288 received was valid base64.
5290 16. The debug output line for the "set" modifier in ACLs was not showing the
5291 name of the variable that was being set.
5293 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
5295 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
5296 lookup on its own. Something else was needed to trigger the lookup. For
5297 example, a match in host_lookup or the need for a host name in a host list.
5298 Now, if $sender_host_name is referenced and the host name has not yet been
5299 looked up, a lookup is performed. If the lookup fails, the variable remains
5300 empty, and $host_lookup_failed is set to "1".
5302 19. Added "eqi" as a case-independent comparison operator.
5304 20. The saslauthd authentication condition could segfault if neither service
5305 nor realm was specified.
5307 21. If an overflowing value such as "2048M" was set for message_size_limit, the
5308 error message that was logged was misleading, and incoming SMTP
5309 connections were dropped. The message is now more accurate, and temporary
5310 errors are given to SMTP connections.
5312 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
5313 (except RSET) with a 421 error, until QUIT is received. However, it was
5314 failing to send a response to QUIT.
5316 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
5317 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
5318 ACL is now run after the helo_try_verify_hosts code.
5320 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
5321 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
5322 case-independent, so other case variants are also recognized). Apparently
5323 some systems use these upper case variants.
5325 25. If more than two messages were waiting for the same host, and a transport
5326 filter was specified for the transport, Exim sent two messages over the
5327 same TCP/IP connection, and then failed with "socket operation on non-
5328 socket" when it tried to send the third.
5330 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
5332 27. The extern definition of crypt16() in expand.c was not being excluded when
5333 the OS had its own crypt16() function.
5335 28. Added bounce_return_body as a new option, and bounce_return_size_limit
5336 as a preferred synonym for return_size_limit, both as an option and as an
5339 29. Added LIBS=-liconv to OS/Makefile-OSF1.
5341 30. Changed the default configuration ACL to relax the local part checking rule
5342 for addresses that are not in any local domains. For these addresses,
5343 slashes and pipe symbols are allowed within local parts, but the sequence
5344 /../ is explicitly forbidden.
5346 31. SPA server authentication was not clearing the challenge buffer before
5349 32. log_message in a "warn" ACL statement was writing to the reject log as
5350 well as to the main log, which contradicts the documentation and doesn't
5351 seem right (because no rejection is happening). So I have stopped it.
5353 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
5354 However, I am unable to do any testing of this.
5356 34. Fixed an infelicity in the appendfile transport. When checking directories
5357 for a mailbox, to see if any needed to be created, it was accidentally
5358 using path names with one or more superfluous leading slashes; tracing
5359 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
5361 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
5362 contents are added to the log line that is written for every discarded
5363 recipient. (Previously a log_message setting was ignored.)
5365 36. The ${quote: operator now quotes the string if it is empty.
5367 37. The install script runs exim in order to find its version number. If for
5368 some reason other than non-existence or emptiness, which it checks, it
5369 could not run './exim', it was installing it with an empty version number,
5370 i.e. as "exim-". This error state is now caught, and the installation is
5373 38. An argument was missing from the function that creates an error message
5374 when Exim fails to connect to the socket for saslauthd authentication.
5375 This could cause Exim to crash, or give a corrupted message.
5377 39. Added isip, isip4, and isip6 to ${if conditions.
5379 40. The ACL variables $acl_xx are now saved with the message, and can be
5380 accessed later in routers, transports, and filters.
5382 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
5383 strings in the file are not string-expanded.
5385 42. If a MAIL command specified a SIZE value that was too large to fit into an
5386 int variable, the check against message_size_limit failed. Such values are
5387 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
5388 day this will have to be increased, but I don't think I want to be around
5389 when emails are that large.
5396 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
5397 iconv() is not standard in FreeBSD.
5399 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
5400 IPv6 enabled. The observed symptom was a segmentation fault on return from
5401 the function os_common_find_running_interfaces() in src/os.c.
5403 3. In the check_special_case() function in daemon.c I had used "errno" as an
5404 argument name, which causes warnings on some systems. This was basically a
5405 typo, since it was named "eno" in the comments!
5407 4. The code that waits for the clock to tick (at a resolution of some fraction
5408 of a second) so as to ensure message-id uniqueness was always waiting for
5409 at least one whole tick, when it could have waited for less. [This is
5410 almost certainly not relevant at current processor speeds, where it is
5411 unlikely to ever wait at all. But we try to future-proof.]
5413 5. The function that sleeps for a time interval that includes fractions of a
5414 second contained a race. It did not block SIGALRM between setting the
5415 timer, and suspending (a couple of lines later). If the interval was short
5416 and the sigsuspend() was delayed until after it had expired, the suspension
5417 never ended. On busy systems this could lead to processes getting stuck for
5420 6. Some uncommon configurations may cause a lookup to happen in a queue runner
5421 process, before it forks any delivery processes. The open lookup caching
5422 mechanism meant that the open file or database connection was passed into
5423 the delivery process. The problem was that delivery processes always tidy
5424 up cached lookup data. This could cause a problem for the next delivery
5425 process started by the queue runner, because the external queue runner
5426 process does not know about the closure. So the next delivery process
5427 still has data in the lookup cache. In the case of a file lookup, there was
5428 no problem because closing a file descriptor in a subprocess doesn't affect
5429 the parent. However, if the lookup was caching a connection to a database,
5430 the connection was closed, and the second delivery process was likely to
5431 see errors such as "PGSQL: query failed: server closed the connection
5432 unexpectedly". The problem has been fixed by closing all cached lookups
5433 in a queue runner before running a delivery process.
5435 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
5436 seem to have the "const" qualifier which it has on other OS. I've
5439 8. Change 4.21/2 was too strict. It is only if there are two authenticators
5440 *of the same type* (client or server) with the same public name that an
5441 error should be diagnosed.
5443 9. When Exim looked up a host name for an IP address, but failed to find the
5444 original IP address when looking up the host name (a safety check), it
5445 output the message "<ip address> does not match any IP for NULL", which was
5446 confusing, to say the least. The bug was that the host name should have
5447 appeared instead of "NULL".
5449 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
5450 user that is built into the binary, and the -C or -D options is used, root
5451 privilege is dropped before the configuration file is read. In addition,
5452 logging is switched to stderr instead of the normal log files. If the
5453 configuration then re-defines the Exim user, the unprivileged environment
5454 is probably not what is expected, so Exim logs a panic warning message (but
5457 However, if deliver_drop_privilege is set, the unprivileged state may well
5458 be exactly what is intended, so the warning has been cut out in that case,
5459 and Exim is allowed to try to write to its normal log files.
5465 1. smtp_return_error_details was not giving details for temporary sender
5466 or receiver verification errors.
5468 2. Diagnose a configuration error if two authenticators have the same public
5471 3. Exim used not to create the message log file for a message until the first
5472 delivery attempt. This could be confusing when incoming messages were held
5473 for policy or load reasons. The message log file is now created at the time
5474 the message is received, and an initial "Received" line is written to it.
5476 4. The automatically generated man page for command line options had a minor
5477 bug that caused no ill effects; however, a more serious problem was that
5478 the procedure for building the man page automatically didn't always
5479 operate. Consequently, release 4.20 contains an out-of-date version. This
5480 shouldn't happen again.
5482 5. When building Exim with embedded Perl support, the script that builds the
5483 Makefile was calling 'perl' to find its compile-time parameters, ignoring
5484 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
5486 6. The freeze_tell option was not being used for messages that were frozen on
5487 arrival, either by an ACL or by local_scan().
5489 7. Added the smtp_incomplete_transaction log selector.
5491 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
5492 was accepting AUTH without a new EHLO.
5494 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
5495 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
5498 10. Logging of TCP/IP connections (when configured) now happens in the main
5499 daemon process instead of the child process, so that the TCP/IP connection
5500 count is more accurate (but it can never be perfect).
5502 11. The use of "drop" in a nested ACL was not being handled correctly in the
5503 outer ACL. Now, if condition failure induced by the nested "drop" causes
5504 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
5505 or "require"), the connection is dropped.
5507 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
5508 that yield "discard" can now be used with an "accept" or a "discard" verb,
5509 but an error is generated for any others (because I can't see a useful way
5510 to define what should happen).
5512 13. When an ACL is read dynamically from a file (or anywhere else), the lines
5513 are now processed in the same way as lines in the Exim configuration file.
5514 In particular, continuation lines are supported.
5516 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
5518 15. Added -ti meaning -t -i.
5520 16. Check for letters, digits, hyphens, and dots in the names of dnslist
5521 domains, and warn by logging if others are found.
5523 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
5524 returned from GIFCONF when Exim is trying to find the list of interfaces on
5525 a host. The code in os.c has been modified to copy each ifreq to an aligned
5526 structure in all cases.
5528 Also, in some cases, the returned ifreq's were being copied to a 'struct
5529 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
5530 means the last couple of bytes of an IPv6 address could be chopped if the
5531 ifreq contained only a normal sockaddr (14 bytes storage).
5533 18. Named domain lists were not supported in the hosts_treat_as_local option.
5534 An entry such as +xxxx was not recognized, and was treated as a literal
5537 19. Ensure that header lines added by a DATA ACL are included in the reject log
5538 if the ACL subsequently rejects the message.
5540 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
5541 MD5 (which is deprecated).
5543 21. When testing a filter file using -bf, Exim was writing a message when it
5544 took the sender from a "From " line in the message, but it was not doing so
5545 when it took $return_path from a Return-Path: header line. It now does.
5547 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
5548 with a valid header line field name (a series of printing characters
5549 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
5551 23. Changed "disc" in the source to "disk" to conform to the documentation and
5552 the book and for uniformity.
5554 24. Ignore Sendmail's -Ooption=value command line item.
5556 25. When execve() failed while trying to run a command in a pipe transport,
5557 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
5558 could be confused with a return value of 69 from the command itself. This
5559 has been changed to 127, the value the shell returns if it is asked to run
5560 a non-existent command. The wording for the related log line suggests a
5561 non-existent command as the problem.
5563 26. If received_header_text expands to an empty string, do not add a Received:
5564 header line to the message. (Well, it adds a token one on the spool, but
5565 marks it "old" so that it doesn't get used or transmitted.)
5567 27. Installed eximstats 1.28 (addition of -nt option).
5569 28. There was no check for failure on the call to getsockname() in the daemon
5570 code. This can fail if there is a shortage of resources on the system, with
5571 ENOMEM, for example. A temporary error is now given on failure.
5573 29. Contrary to the C standard, it seems that in some environments, the
5574 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
5575 program. Exim now does this explicitly; it affects the formatting of
5576 timestamps using strftime().
5578 30. If exiqsumm was given junk data, it threw up some uninitialized variable
5579 complaints. I've now initialized all the variables, to avoid this.
5581 32. Header lines added by a system filter were not being "seen" during
5582 transport-time rewrites.
5584 33. The info_callback() function passed to OpenSSL is set up with type void
5585 (*)(SSL *, int, int), as described somewhere. However, when calling the
5586 function (actually a macro) that sets it up, the type void(*)() is
5587 expected. I've put in a cast to prevent warnings from picky compilers.
5589 34. If a DNS black list lookup found a CNAME record, but there were no A
5590 records associated with the domain it pointed at, Exim crashed.
5592 35. If a DNS black list lookup returned more than one A record, Exim ignored
5593 all but the first. It now scans all returned addresses if a particular IP
5594 value is being sought. In this situation, the contents of the
5595 $dnslist_value variable are a list of all the addresses, separated by a
5598 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
5599 to accept a host name and all its aliases if the forward lookup for any of
5600 them yielded the IP address of the incoming connection. Now it accepts only
5601 those names whose forward lookup yields the correct IP address. Any other
5602 names are discarded. This closes a loophole whereby a rogue DNS
5603 administrator could create reverse DNS records to break through a
5604 wildcarded host restriction in an ACL.
5606 37. If a user filter or a system filter that ran in a subprocess used any of
5607 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
5608 the wrong values were passed to the pipe command ($thisaddress had the
5609 value of $0, $0 had the value of $1, etc). This bug was introduced by
5610 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
5612 38. Improved the line breaking for long SMTP error messages from ACLs.
5613 Previously, if there was no break point between 40 and 75 characters, Exim
5614 left the rest of the message alone. Two changes have been made: (a) I've
5615 reduced the minimum length to 35 characters; (b) if it can't find a break
5616 point between 35 and 75 characters, it looks ahead and uses the first one
5617 that it finds. This may give the occasional overlong line, but at least the
5618 remaining text gets split now.
5620 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
5621 file descriptors might be low, and that setting 1000 would always raise it.
5622 It turns out that in some environments, the limit is already over 1000 and
5623 that lowering it causes trouble. So now Exim takes care not to decrease it.
5625 40. When delivering a message, the value of $return_path is set to $sender_
5626 address at the start of routing (routers may change the value). By an
5627 oversight, this default was not being set up when an address was tested by
5628 -bt or -bv, which affected the outcome if any router or filter referred to
5631 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
5632 log only when "message" or "log_message" are set. However, if one of the
5633 conditions was an address verification, or a call to a nested ACL, the
5634 messages generated by the underlying test were being passed through. This
5635 no longer happens. The underlying message is available in $acl_verify_
5636 message for both "message" and "log_message" expansions, so it can be
5637 passed through if needed.
5639 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
5640 new expansion $bh_ to give the encoded byte string without charset
5641 translation. Translation happens only if iconv() is available; HAVE_ICONV
5642 indicates this at build time. HEADERS_CHARSET gives the charset to
5643 translate to; headers_charset can change it in the configuration, and
5644 "headers charset" can change it in an individual filter file.
5646 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
5647 that creates RFC 2047 encoded "words" labels them as that charset instead
5648 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
5649 expansion operator; (ii) when Exim creates a From: line for a local
5650 message; (iii) when a header line is rewritten to include a "phrase" part.
5652 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
5653 buggy, causing it to skip the first lines of messages whose message ID
5654 ended in 'D'. This would not have bitten before Exim release 4.14, because
5655 message IDs were unlikely to end in 'D' before then. The effect was to have
5656 incorrect size information for certain domains.
5658 45. #include "config.h" was missing at the start of the crypt16.c module. This
5659 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
5662 46. If there was a timeout during a "random" callout check, Exim treated it as
5663 a failure of the random address, and carried on sending RSET and the real
5664 address. If the delay was just some slowness somewhere, the response to the
5665 original RCPT would be taken as a response to RSET and so on, causing
5666 mayhem of various kinds.
5668 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
5669 when I implemented it. It didn't allow for the fact that some option values
5670 may legitimatetly be negative (e.g. size_addition), and it didn't even do
5671 the right test for positive values.
5673 48. Domain names in DNS records are case-independent. Exim always looks them up
5674 in lower case. Some resolvers return domain names in exactly the case they
5675 appear in the zone file, that is, they may contain uppercase letters. Not
5676 all resolvers do this - some return always lower case. Exim was treating a
5677 change of case by a resolver as a change of domain, similar to a widening
5678 of a domain abbreviation. This triggered its re-routing code and so it was
5679 trying to route what was effectively the same domain again. This normally
5680 caused routing to fail (because the router wouldn't handle the domain
5681 twice). Now Exim checks for this case specially, and just changes the
5682 casing of the domain that it ultimately uses when it transmits the message
5685 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
5688 50. If a filter generated a file delivery with a non-absolute name (possible if
5689 no home directory exists for the router), the forbid_file option was not
5692 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
5693 the existing equality matching.
5695 52. Exim was using ints instead of ino_t variables in some places where it was
5696 dealing with inode numbers.
5698 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
5699 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
5700 called TMPDIR, and if it finds it is different, it changes its value.
5702 54. The smtp_printf() function is now made available to local_scan() so
5703 additional output lines can be written before returning. There is also an
5704 smtp_fflush() function to enable the detection of a dropped connection.
5705 The variables smtp_input and smtp_batched_input are exported to
5708 55. Changed the default runtime configuration: the message "Unknown user"
5709 has been removed from the ACL, and instead placed on the localuser router,
5710 using the cannot_route_message feature. This means that any verification
5711 failures that generate their own messages won't get overridden. Similarly,
5712 the "Unrouteable address" message that was in the ACL for unverifiable
5713 relay addresses has also been removed.
5715 56. Added hosts_avoid_esmtp to the smtp transport.
5717 57. The exicyclog script was not checking for the esoteric option
5718 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
5719 will work only if exicyclog is run under the appropriate euid.
5721 58. Following a discussion on the list, the rules by which Exim recognises line
5722 endings on incoming messages have been changed. The -dropcr and drop_cr
5723 options are now no-ops, retained only for backwards compatibility. The
5724 following line terminators are recognized: LF CRLF CR. However, special
5725 processing applies to CR:
5727 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
5728 nor a local message in the state where . is a terminator.
5730 (ii) If a bare CR is encountered in a header line, an extra space is added
5731 after the line terminator so as not to end the header. The reasoning
5732 behind this is that bare CRs in header lines are most likely either
5733 to be mistakes, or people trying to play silly games.
5735 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
5736 was being incorrectly given as 18 bytes larger than it should have been.
5737 This is a VOB (very old bug).
5739 60. This may never have affected anything current, but just in case it has:
5740 When the local host is found other than at the start of a list of hosts,
5741 the local host, those with the same MX, and any that follow, are discarded.
5742 When the list in question was part of a longer list of hosts, the following
5743 hosts (not currently being processed) were also being discarded. This no
5744 longer happens. I'm not sure if this situation could ever has previously
5747 61. Added the "/MX" feature to lists of hosts in the manualroute and query
5750 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
5751 header. This is something that is recommended in a new Internet Draft, and
5752 is something that is documented as being done by Sendmail. There are two
5753 possible values. For messages generated by the autoreply transport, Exim
5756 Auto-Submitted: auto-replied
5758 whereas for all other generated messages (e.g. bounces) it adds
5760 Auto-Submitted: auto-generated
5762 63. The "personal" condition in filters now includes a test for the
5763 Auto-Submitted: header. If it contains the string "auto-" the message it
5764 not considered personal.
5766 64. Added rcpt_include_affixes as a generic transport option.
5768 65. Added queue_only_override (default true).
5770 66. Added the syslog_duplication option.
5772 67. If what should have been the first header line of a message consisted of
5773 a space followed by a colon, Exim was mis-interpreting it as a header line.
5774 It isn't of course - it is syntactically invalid and should therefore be
5775 treated as the start of the message body. The misbehaviour could have
5776 caused a number of strange effects, including loss of data in subsequent
5777 header lines, and spool format errors.
5779 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
5780 client host had authenticated. This control can now be exercised by an ACL
5781 for more flexibility.
5783 69. By default, callouts do not happen when testing with -bh. There is now a
5784 variant, -bhc, which does actually run the callout code, including
5785 consulting and updating the callout cache.
5787 70. Added support for saslauthd authentication, courtesy of Alexander
5790 71. If statvfs() failed on the spool or log directories while checking their
5791 size for availability, Exim confusingly gave the error "space shortage".
5792 Furthermore, in debugging mode it crashed with a floating point exception.
5793 These checks are done if check_{spool,log}_{space,inodes} are set, and when
5794 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
5795 serious problem, Exim now writes to the main and panic logs when this
5796 happens, with details of the failure. It then refuses to accept the
5797 incoming message, giving the message "spool directory problem" or "log
5798 directory problem" with a 421 code for SMTP messages.
5800 72. When Exim is about to re-exec itself, it ensures that the file descriptors
5801 0, 1, and 2 exist, because some OS complain for execs without them (see
5802 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
5803 descriptors. However, the code omitted to check that the open succeeded,
5804 causing mysterious errors if for some reason the permissions on /dev/null
5805 got screwed. Now Exim writes a message to the main and panic logs, and
5806 bombs out if it can't open /dev/null.
5808 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
5809 interact so that it is all more flexible. It is supposed to remain
5810 backwards compatible. Also added extra_local_interfaces.
5812 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
5813 to bomb out with an assertion failure - to the client this appears as a
5814 connection drop. This problem occurs in the part of the code that was taken
5815 from the Samba project. Fortunately, the assertion is in a very simple
5816 function, so I have fixed this by reproducing the function inline in the
5817 one place where it is called, and arranging for authentication to fail
5818 instead of killing the process with assert().
5820 75. The SPA client code was not working when the server requested OEM rather
5821 than Unicode encoding.
5823 76. Added code to make require_files with a specific uid setting more usable in
5824 the case where statting the file as root fails - usually a non-root-mounted
5825 NFS file system. When this happens and the failure is EACCES, Exim now
5826 forks a subprocess and does the per-uid checking as the relevant uid.
5828 77. Added process_log_path.
5830 78. If log_file_path was not explicitly set, a setting of check_log_space or
5831 check_log_inodes was ignored.
5833 79. If a space check for the spool or log partitions fails, the incident is now
5834 logged. Of course, in the latter case the data may get lost...
5836 80. Added the %p formatting code to string_format() so that it can be used to
5837 print addresses in debug_print(). Adjusted all the address printing in the
5838 debugging in store.c to use %p rather than %d.
5840 81. There was a concern that a line of code in smtp_in.c could overflow a
5841 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
5842 initially expressed, the concern was not well-founded, because trailing
5843 spaces are removed early. However, if the trailing spaces were followed by
5844 a NULL, they did not get removed, so the overflow was possible. Two fixes
5847 (a) I re-wrote the offending code in a cleaner fashion.
5848 (b) If an incoming SMTP command contains a NULL character, it is rejected
5851 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
5852 runs initgroups(), so that if the Exim user is in any additional groups,
5853 they will be used during message reception.
5859 The change log for 4.20 and earlier releases has been archived.
projects / exim.git / blob