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.
200 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
202 This, or similar/improved, will also be change PP/11 of 4.82.
208 PP/01 Handle short writes when writing local log-files.
209 In practice, only affects FreeBSD (8 onwards).
210 Bugzilla 1053, with thanks to Dmitry Isaikin.
212 NM/01 Bugzilla 949 - Documentation tweak
214 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
217 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
219 PP/02 Implemented gsasl authenticator.
221 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
223 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
224 `pkg-config foo` for cflags/libs.
226 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
227 with rest of GSASL and with heimdal_gssapi.
229 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
230 `pkg-config foo` for cflags/libs for the TLS implementation.
232 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
233 properties get this fed in as external SSF. A number of robustness
234 and debugging improvements to the cyrus_sasl authenticator.
236 PP/08 cyrus_sasl server now expands the server_realm option.
238 PP/09 Bugzilla 1214 - Log authentication information in reject log.
239 Patch by Jeremy Harris.
241 PP/10 Added dbmjz lookup type.
243 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
245 PP/12 MAIL args handles TAB as well as SP, for better interop with
246 non-compliant senders.
247 Analysis and variant patch by Todd Lyons.
249 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
250 Bug report from Lars Müller <lars@samba.org> (via SUSE),
251 Patch from Dirk Mueller <dmueller@suse.com>
253 PP/13 tls_peerdn now print-escaped for spool files.
254 Observed some $tls_peerdn in wild which contained \n, which resulted
255 in spool file corruption.
257 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
258 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
259 or write after TLS renegotiation, which otherwise led to messages
262 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
263 as a tracking header (ie: a signed header comes before the signature).
264 Patch from Wolfgang Breyha.
266 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
267 comma-sep list; embedded commas doubled.
269 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
271 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
273 Report and patch from Dmitry Banschikov.
275 PP/16 Removed "dont_insert_empty_fragments" fron "openssl_options".
276 Removed SSL_clear() after SSL_new() which led to protocol negotiation
277 failures. We appear to now support TLS1.1+ with Exim.
279 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
280 lets Exim select keys and certificates based upon TLS SNI from client.
281 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
282 before an outbound SMTP session. New log_selector, +tls_sni.
284 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
285 NULL dereference. Report and patch from Alun Jones.
287 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
288 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
289 Not seeing resolver debug output on NetBSD, but suspect this is a
290 resolver implementation change.
292 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
293 Left warnings. Added "eximon gdb" invocation mode.
295 PP/21 Defaulting "accept_8bitmime" to true, not false.
297 PP/22 Added -bw for inetd wait mode support.
299 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
300 locate the relevant includes and libraries. Made this the default.
302 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
303 Bugzilla 1246, report and most of solution from Tomasz Kusy.
305 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
306 This may cause build issues on older platforms.
308 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
309 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
310 gnutls_require_mac & gnutls_require_protocols (no longer supported).
311 Added SNI support via GnuTLS too.
312 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
314 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
316 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
317 Applied second patch from Janne, implementing suggestion to default
318 multiple-strings-in-record handling to match SPF spec.
320 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
322 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
323 read-only, out of scope).
324 Patch from Wolfgang Breyha, report from Stuart Northfield.
326 PP/29 Fix three issues highlighted by clang analyser static analysis.
327 Only crash-plausible issue would require the Cambridge-specific
328 iplookup router and a misconfiguration.
329 Report from Marcin Mirosław.
331 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
333 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
334 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
335 As part of this, removing so much warning spew let me fix some minor
336 real issues in debug logging.
338 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
339 assignment on my part. Fixed.
341 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
342 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
343 Janne Snabb (who went above and beyond: thank you).
345 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
346 string otherwise requires a connection and a bunch more work and it's
347 relatively easy to get wrong. Should also expose TLS library linkage
350 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
351 64-bit ${eval} (JH/03).
353 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
354 GNU libc to support some of the 64-bit stuff, should not lead to
355 conflicts. Defined before os.h is pulled in, so if a given platform
356 needs to override this, it can.
358 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
359 protection layer was required, which is not implemented.
360 Bugzilla 1254, patch from Wolfgang Breyha.
362 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
363 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
364 tls_dhparam take prime identifiers. Also unbreak combination of
365 OpenSSL+DH_params+TLSSNI.
367 PP/39 Disable SSLv2 by default in OpenSSL support.
373 PP/01 Solaris build fix for Oracle's LDAP libraries.
374 Bugzilla 1109, patch from Stephen Usher.
376 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
378 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
381 TF/02 Fix a couple more cases where we did not log the error message
382 when unlink() failed. See also change 4.74-TF/03.
384 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
385 lock up or crash if it happened to be inside a call to libc when it
386 got a SIGUSR1 from exiwhat.
388 The SIGUSR1 handler appends the current process status to the process
389 log which is later printed by exiwhat. It used to use the general
390 purpose logging code to do this, but several functions it calls are
391 not safe for signals.
393 The new output code in the SIGUSR1 handler is specific to the process
394 log, and simple enough that it's easy to inspect for signal safety.
395 Removing some special cases also simplifies the general logging code.
396 Removing the spurious timestamps from the process log simplifies
399 TF/04 Improved ratelimit ACL condition.
401 The /noupdate option has been deprecated in favour of /readonly which
402 has clearer semantics. The /leaky, /strict, and /readonly update modes
403 are mutually exclusive. The update mode is no longer included in the
404 database key; it just determines when the database is updated. (This
405 means that when you upgrde Exim will forget old rate measurements.)
407 Exim now checks that the per_* options are used with an update mode that
408 makes sense for the current ACL. For example, when Exim is processing a
409 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
410 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
411 must specify per_mail/readonly. If you omit the update mode it defaults to
412 /leaky where that makes sense (as before) or /readonly where required.
414 The /noupdate option is now undocumented but still supported for
415 backwards compatibility. It is equivalent to /readonly except that in
416 ACLs where /readonly is required you may specify /leaky/noupdate or
417 /strict/noupdate which are treated the same as /readonly.
419 A useful new feature is the /count= option. This is a generalization
420 of the per_byte option, so that you can measure the throughput of other
421 aggregate values. For example, the per_byte option is now equivalent
422 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
424 The per_rcpt option has been generalized using the /count= mechanism
425 (though it's more complicated than the per_byte equivalence). When it is
426 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
427 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
428 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
429 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
430 Note that using per_rcpt with a non-readonly update mode in more than
431 one ACL will cause the recipients to be double-counted. (The per_mail
432 and per_byte options don't have this problem.)
434 The handling of very low rates has changed slightly. If the computed rate
435 is less than the event's count (usually one) then this event is the first
436 after a long gap. In this case the rate is set to the same as this event's
437 count, so that the first message of a spam run is counted properly.
439 The major new feature is a mechanism for counting the rate of unique
440 events. The new per_addr option counts the number of different
441 recipients that someone has sent messages to in the last time period. It
442 behaves like per_rcpt if all the recipient addresses are different, but
443 duplicate recipient addresses do not increase the measured rate. Like
444 the /count= option this is a general mechanism, so the per_addr option
445 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
446 example, measure the rate that a client uses different sender addresses
447 with the options per_mail/unique=$sender_address. There are further
448 details in the main documentation.
450 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
452 TF/06 Removed a few PCRE remnants.
454 TF/07 Automatically extract Exim's version number from tags in the git
455 repository when doing development or release builds.
457 PP/02 Raise smtp_cmd_buffer_size to 16kB.
458 Bugzilla 879. Patch from Paul Fisher.
460 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
461 Heavily based on revision 40f9a89a from Simon Arlott's tree.
464 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
466 PP/05 Variable $av_failed, true if the AV scanner deferred.
467 Bugzilla 1078. Patch from John Horne.
469 PP/06 Stop make process more reliably on build failure.
470 Bugzilla 1087. Patch from Heiko Schlittermann.
472 PP/07 Make maildir_use_size_file an _expandable_ boolean.
473 Bugzilla 1089. Patch from Heiko Schlittermann.
475 PP/08 Handle ${run} returning more data than OS pipe buffer size.
476 Bugzilla 1131. Patch from Holger Weiß.
478 PP/09 Handle IPv6 addresses with SPF.
479 Bugzilla 860. Patch from Wolfgang Breyha.
481 PP/10 GnuTLS: support TLS 1.2 & 1.1.
483 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
486 PP/11 match_* no longer expand right-hand-side by default.
487 New compile-time build option, EXPAND_LISTMATCH_RHS.
488 New expansion conditions, "inlist", "inlisti".
490 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
492 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
494 PP/14 fix log_write() format string regression from TF/03.
495 Bugzilla 1152. Patch from Dmitry Isaikin.
501 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
503 PP/02 Harmonised TLS library version reporting; only show if debugging.
504 Layout now matches that introduced for other libraries in 4.74 PP/03.
506 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
508 PP/04 New "dns_use_edns0" global option.
510 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
513 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
514 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
516 TK/01 Updated PolarSSL code to 0.14.2.
517 Bugzilla 1097. Patch from Andreas Metzler.
519 PP/07 Catch divide-by-zero in ${eval:...}.
522 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
525 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
526 format-string attack -- SECURITY: remote arbitrary code execution.
528 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
529 time unintentionally subject to list matching rules, letting the header
530 cause arbitrary Exim lookups (of items which can occur in lists, *not*
531 arbitrary string expansion). This allowed for information disclosure.
533 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
534 INT_MIN/-1 -- value coerced to INT_MAX.
540 NM/01 Workround for PCRE version dependency in version reporting
543 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
544 This fixes portability to compilers other than gcc, notably
545 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
547 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
548 makefiles for portability to HP-UX and POSIX correctness.
550 PP/01 Permit LOOKUP_foo enabling on the make command-line.
551 Also via indented variable definition in the Makefile.
552 (Debugging by Oliver Heesakkers).
554 PP/02 Restore caching of spamd results with expanded spamd_address.
555 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
557 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
558 Improves build reliability. Fix from: Frank Elsner
560 NM/02 Fix wide character breakage in the rfc2047 coding
561 Fixes bug 1064. Patch from Andrey N. Oktyabrski
563 NM/03 Allow underscore in dnslist lookups
564 Fixes bug 1026. Patch from Graeme Fowler
566 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
567 Code patches from Adam Ciarcinski of NetBSD.
569 NM/04 Fixed exiqgrep to cope with mailq missing size issue
572 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
573 is logged, to avoid truncation. Patch from John Horne.
575 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
576 Patch from Jakob Hirsch.
578 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
579 SQL string expansion failure details.
580 Patch from Andrey Oktyabrski.
582 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
583 Patch from Simon Arlott.
585 PP/09 New lookups functionality failed to compile on old gcc which rejects
586 extern declarations in function scope.
587 Patch from Oliver Fleischmann
589 PP/10 Use sig_atomic_t for flags set from signal handlers.
590 Check getgroups() return and improve debugging.
591 Fixed developed for diagnosis in bug 927 (which turned out to be
594 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
595 Patch from Mark Zealey.
597 PP/12 Bugzilla 1056: Improved spamd server selection.
598 Patch from Mark Zealey.
600 PP/13 Bugzilla 1086: Deal with maildir quota file races.
601 Based on patch from Heiko Schlittermann.
603 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
604 Patch from Uwe Doering, sign-off by Michael Haardt.
606 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
607 variable declaration deep within a block. Bug and patch from
610 PP/15 lookups-Makefile IRIX compatibilty coercion.
612 PP/16 Make DISABLE_DKIM build knob functional.
614 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
615 Patch by Simon Arlott
617 TF/03 Fix valgrind.h portability to C89 compilers that do not support
618 variable argument macros. Our copy now differs from upstream.
624 TF/01 Failure to get a lock on a hints database can have serious
625 consequences so log it to the panic log.
627 TF/02 Log LMTP confirmation messages in the same way as SMTP,
628 controlled using the smtp_confirmation log selector.
630 TF/03 Include the error message when we fail to unlink a spool file.
632 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
633 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
634 for maintaining out-of-tree patches for some time.
636 PP/01 Bugzilla 139: Documentation and portability issues.
637 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
638 Handle per-OS dynamic-module compilation flags.
640 PP/02 Let /dev/null have normal permissions.
641 The 4.73 fixes were a little too stringent and complained about the
642 permissions on /dev/null. Exempt it from some checks.
643 Reported by Andreas M. Kirchwitz.
645 PP/03 Report version information for many libraries, including
646 Exim version information for dynamically loaded libraries. Created
647 version.h, now support a version extension string for distributors
648 who patch heavily. Dynamic module ABI change.
650 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
651 privilege escalation vulnerability whereby the Exim run-time user
652 can cause root to append content of the attacker's choosing to
655 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
658 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
659 If dropping privileges for untrusted macros, we disabled normal logging
660 on the basis that it would fail; for the Exim run-time user, this is not
661 the case, and it resulted in successful deliveries going unlogged.
662 Fixed. Reported by Andreas Metzler.
668 PP/01 Date: & Message-Id: revert to normally being appended to a message,
669 only prepend for the Resent-* case. Fixes regression introduced in
670 Exim 4.70 by NM/22 for Bugzilla 607.
672 PP/02 Include check_rfc2047_length in configure.default because we're seeing
673 increasing numbers of administrators be bitten by this.
675 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
677 PP/03 Bugzilla 994: added openssl_options main configuration option.
679 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
681 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
683 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
685 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
686 without a peer certificate, leading to a segfault because of an
687 assumption that peers always have certificates. Be a little more
688 paranoid. Problem reported by Martin Tscholak.
690 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
691 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
692 NB: ClamAV planning to remove STREAM in "middle of 2010".
693 CL also introduces -bmalware, various -d+acl logging additions and
694 more caution in buffer sizes.
696 PP/09 Implemented reverse_ip expansion operator.
698 PP/10 Bugzilla 937: provide a "debug" ACL control.
700 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
702 PP/12 Bugzilla 973: Implement --version.
704 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
706 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
708 PP/15 Bugzilla 816: support multiple condition rules on Routers.
710 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
711 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
712 ignore trailing whitespace.
714 JJ/02 prevent non-panic DKIM error from being sent to paniclog
716 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
719 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
720 Notification from Dr Andrew Aitchison.
722 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
723 ExtendedDetectionInfo response format.
724 Notification from John Horne.
726 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
729 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
730 XSL and documented dependency on system catalogs, with examples of how
733 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
736 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
737 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
738 configuration file which is writeable by the Exim user or group.
740 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
741 of configuration files to cover files specified with the -C option if
742 they are going to be used with root privileges, not just the default
745 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
746 option (effectively making it always true).
748 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
749 files to be used while preserving root privileges.
751 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
752 that rogue child processes cannot use them.
754 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
755 run-time user, instead of root.
757 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
758 Exim run-time user without dropping privileges.
760 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
761 result string, instead of calling string_vformat() twice with the same
764 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
765 for other users. Others should always drop root privileges if they use
766 -C on the command line, even for a whitelisted configure file.
768 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
770 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
776 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
777 $data_path, and $header_path variables; fixed documentation bugs and
780 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
781 exipick to access non-standard spools, including the "frozen" queue
784 NM/01 Bugzilla 965: Support mysql stored procedures.
785 Patch from Alain Williams
787 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
789 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
790 Patch from Andreas Metzler
792 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
793 Patch from Kirill Miazine
795 NM/05 Bugzilla 671: Added umask to procmail example.
797 JJ/03 installed exipick 20100323.0, fixing doc bug
799 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
800 directory. Notification and patch from Dan Rosenberg.
802 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
804 TK/02 Improve log output when DKIM signing operation fails.
806 MH/01 Treat the transport option dkim_domain as a colon separated
807 list, not as a single string, and sign the message with each element,
808 omitting multiple occurences of the same signer.
810 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
811 Bugzilla 985, 986. Patch by Simon Arlott
813 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
814 Patch by Simon Arlott
816 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
817 MBX locking. Notification from Dan Rosenberg.
823 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
825 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
827 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
829 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
831 NM/04 Bugzilla 915: Flag broken perl installation during build.
837 TK/01 Added patch by Johannes Berg that expands the main option
838 "spamd_address" if it starts with a dollar sign.
840 TK/02 Write list of recipients to X-Envelope-Sender header when building
841 the mbox-format spool file for content scanning (suggested by Jakob
844 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
845 (http://www.dcc-servers.net/) support via dccifd. Activated by
846 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
848 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
849 by Mark Daniel Reidel <mr@df.eu>.
851 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
852 When building exim an external PCRE library is now needed -
853 PCRE is a system library on the majority of modern systems.
854 See entry on PCRE_LIBS in EDITME file.
856 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
857 conversation. Added nologin parameter to request.
858 Patch contributed by Kirill Miazine.
860 TF/01 Do not log submission mode rewrites if they do not change the address.
862 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
864 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
865 log files in place. Contributed by Roberto Lima.
867 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
869 TF/03 Bugzilla 615: When checking the local_parts router precondition
870 after a local_part_suffix or local_part_prefix option, Exim now
871 does not use the address's named list lookup cache, since this
872 contains cached lookups for the whole local part.
874 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
875 Robert Millan. Documentation is in experimental-spec.txt.
877 TF/04 Bugzilla 668: Fix parallel build (make -j).
879 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
881 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
882 Patch provided by Jan Srzednicki.
884 TF/05 Leading white space used to be stripped from $spam_report which
885 wrecked the formatting. Now it is preserved.
887 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
888 that they are available at delivery time.
890 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
892 TF/08 TLS error reporting now respects the incoming_interface and
893 incoming_port log selectors.
895 TF/09 Produce a more useful error message if an SMTP transport's hosts
896 setting expands to an empty string.
898 NM/06 Bugzilla 744: EXPN did not work under TLS.
899 Patch provided by Phil Pennock.
901 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
902 Patch provided by Richard Godbee.
904 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
905 acl_smtp_notquit, added index entry.
907 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
908 Patch provided by Eugene Bujak.
910 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
911 accept(). Patch provided by Maxim Dounin.
913 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
914 Patch provided by Phil Pennock.
916 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
918 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
919 Patch provided by Brad "anomie" Jorsch.
921 NM/14 Bugzilla 622: Added timeout setting to transport filter.
922 Patch provided by Dean Brooks.
924 TK/05 Add native DKIM support (does not depend on external libraries).
926 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
927 Patch provided by Graeme Fowler.
929 NM/16 Bugzilla 851: Documentation example syntax fix.
931 NM/17 Changed NOTICE file to remove references to embedded PCRE.
933 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
936 NM/19 Bugzilla 745: TLS version reporting.
937 Patch provided by Phil Pennock.
939 NM/20 Bugzilla 167: bool: condition support.
940 Patch provided by Phil Pennock.
942 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
943 clients. Patch provided by Phil Pennock.
945 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
946 Patch provided by Brad "anomie" Jorsch.
948 NM/23 Bugzilla 687: Fix misparses in eximstats.
949 Patch provided by Heiko Schlittermann.
951 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
952 Patch provided by Heiko Schlittermann.
954 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
955 plus update to original patch.
957 NM/26 Bugzilla 799: Documentation correction for ratelimit.
959 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
960 Patch provided by David Brownlee.
962 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
964 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
966 NM/30 Bugzilla 888: TLS documentation bugfixes.
968 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
970 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
971 Unlike the original bugzilla I have changed all shell scripts in src tree.
973 NM/33 Bugzilla 898: Transport filter timeout fix.
974 Patch by Todd Rinaldo.
976 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
977 Patch by Serge Demonchaux.
979 NM/35 Bugzilla 39: Base64 decode bug fixes.
980 Patch by Jakob Hirsch.
982 NM/36 Bugzilla 909: Correct connect() call in dcc code.
984 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
986 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
988 NM/39 Bugzilla 911: Fixed MakeLinks build script.
994 TK/01 Add preliminary DKIM support. Currently requires a forked version of
995 ALT-N's libdkim that I have put here:
996 http://duncanthrax.net/exim-experimental/
998 Note to Michael Haardt: I had to rename some vars in sieve.c. They
999 were called 'true' and it seems that C99 defines that as a reserved
1000 keyword to be used with 'bool' variable types. That means you could
1001 not include C99-style headers which use bools without triggering
1002 build errors in sieve.c.
1004 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
1005 as mailq or other aliases. Changed the --help handling significantly
1006 to do whats expected. exim_usage() emits usage/help information.
1008 SC/01 Added the -bylocaldomain option to eximstats.
1010 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
1012 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
1014 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
1020 PH/01 Another patch from the Sieve maintainer.
1022 PH/02 When an IPv6 address is converted to a string for single-key lookup
1023 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
1024 dots are used instead of colons so that keys in lsearch files need not
1025 contain colons. This was done some time before quoting was made available
1026 in lsearch files. However, iplsearch files do require colons in IPv6 keys
1027 (notated using the quote facility) so as to distinguish them from IPv4
1028 keys. This meant that lookups for IP addresses in host lists did not work
1029 for iplsearch lookups.
1031 This has been fixed by arranging for IPv6 addresses to be expressed with
1032 colons if the lookup type is iplsearch. This is not incompatible, because
1033 previously such lookups could never work.
1035 The situation is now rather anomolous, since one *can* have colons in
1036 ordinary lsearch keys. However, making the change in all cases is
1037 incompatible and would probably break a number of configurations.
1039 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
1042 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
1043 conversion specification without a maximum field width, thereby enabling
1044 a rogue spamd server to cause a buffer overflow. While nobody in their
1045 right mind would setup Exim to query an untrusted spamd server, an
1046 attacker that gains access to a server running spamd could potentially
1047 exploit this vulnerability to run arbitrary code as the Exim user.
1049 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
1050 $primary_hostname instead of what libspf2 thinks the hosts name is.
1052 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
1053 a directory entry by the name of the lookup key. Previously, if a
1054 symlink pointed to a non-existing file or a file in a directory that
1055 Exim lacked permissions to read, a lookup for a key matching that
1056 symlink would fail. Now it is enough that a matching directory entry
1057 exists, symlink or not. (Bugzilla 503.)
1059 PH/03 The body_linecount and body_zerocount variables are now exported in the
1062 PH/04 Added the $dnslist_matched variable.
1064 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
1065 This means they are set thereafter only if the connection becomes
1068 PH/06 Added the client_condition to authenticators so that some can be skipped
1069 by clients under certain conditions.
1071 PH/07 The error message for a badly-placed control=no_multiline_responses left
1072 "_responses" off the end of the name.
1074 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
1076 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
1077 (without spaces) instead of just copying the configuration text.
1079 PH/10 Added the /noupdate option to the ratelimit ACL condition.
1081 PH/11 Added $max_received_linelength.
1083 PH/12 Added +ignore_defer and +include_defer to host lists.
1085 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
1086 way in which PCRE > 7.0 is built.
1088 PH/14 Implemented queue_only_load_latch.
1090 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
1091 MAIL command. The effect was to mangle the value on 64-bit systems.
1093 PH/16 Another patch from the Sieve maintainer.
1095 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
1097 PH/18 If a system quota error occurred while trying to create the file for
1098 a maildir delivery, the message "Mailbox is full" was not appended to the
1099 bounce if the delivery eventually timed out. Change 4.67/27 below applied
1100 only to a quota excession during the actual writing of the file.
1102 PH/19 It seems that peer DN values may contain newlines (and other non-printing
1103 characters?) which causes problems in log lines. The DN values are now
1104 passed through string_printing() before being added to log lines.
1106 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
1107 and InterBase are left for another time.)
1109 PH/21 Added message_body_newlines option.
1111 PH/22 Guard against possible overflow in moan_check_errorcopy().
1113 PH/23 POSIX allows open() to be a macro; guard against that.
1115 PH/24 If the recipient of an error message contained an @ in the local part
1116 (suitably quoted, of course), incorrect values were put in $domain and
1117 $local_part during the evaluation of errors_copy.
1123 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
1124 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
1127 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
1128 issue a MAIL command.
1130 PH/02 In an ACL statement such as
1132 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
1134 if a client was not listed at all, or was listed with a value other than
1135 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
1136 the condition was not true (as it should be), so access was not denied.
1137 The bug was that the ! inversion was incorrectly passed on to the second
1138 item. This has been fixed.
1140 PH/03 Added additional dnslists conditions == and =& which are different from
1141 = and & when the dns lookup returns more than one IP address.
1143 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
1144 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
1146 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
1147 FSYNC, which compiles an option called disable_fsync that allows for
1148 bypassing fsync(). The documentation is heavily laced with warnings.
1150 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
1152 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
1153 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
1154 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
1155 including adding "make clean"; (3) Added -fPIC when compiling the test
1156 dynamically loaded module, to get rid of a warning.
1158 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
1159 message fails, move_frozen_messages = true and ignore_bounce_errors_after
1160 = 0s. The bug is otherwise harmless.
1162 PH/07 There was a bug in the dovecot authenticator such that the value of
1163 $auth1 could be overwritten, and so not correctly preserved, after a
1164 successful authentication. This usually meant that the value preserved by
1165 the server_setid option was incorrect.
1167 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
1169 PH/09 Installed PCRE release 7.0.
1171 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
1172 run for batched SMTP input. It is now run at the start of every message
1173 in the batch. While fixing this I discovered that the process information
1174 (output by running exiwhat) was not always getting set for -bs and -bS
1175 input. This is fixed, and it now also says "batched" for BSMTP.
1177 PH/11 Added control=no_pipelining.
1179 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
1180 patch, slightly modified), and move the expansion of helo_data till after
1181 the connection is made in the smtp transport (so it can use these
1184 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
1186 PH/14 Added log_selector = +pid.
1188 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
1190 PH/16 Add ${if forany and ${if forall.
1192 PH/17 Added dsn_from option to vary the From: line in DSNs.
1194 PH/18 Flush SMTP output before performing a callout, unless control =
1195 no_callout_flush is set.
1197 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
1198 was true (the default) a successful delivery failed to delete the retry
1199 item, thus causing premature timeout of the address. The bug is now
1202 PH/20 Added hosts_avoid_pipelining to the smtp transport.
1204 PH/21 Long custom messages for fakedefer and fakereject are now split up
1205 into multiline reponses in the same way that messages for "deny" and
1206 other ACL rejections are.
1208 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
1209 with slight modification.
1211 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
1212 draft, changing the syntax and factoring some duplicate code".
1214 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
1215 for deliveries of the second and subsequent messages over the same SMTP
1218 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
1219 ${reduce, with only minor "tidies".
1221 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
1223 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
1224 expansion side effects.
1226 PH/27 When a message times out after an over-quota error from an Exim-imposed
1227 quota, the bounce message says "mailbox is full". This message was not
1228 being given when it was a system quota that was exceeded. It now should
1231 MH/03 Made $recipients available in local_scan(). local_scan() already has
1232 better access to the recipient list through recipients_list[], but
1233 $recipients can be useful in postmaster-provided expansion strings.
1235 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
1236 in the case of a MAIL command with additional options following the
1237 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
1238 were accidentally chopped off.
1240 PH/29 SMTP synchronization checks are implemented when a command is read -
1241 there is a check that no more input is waiting when there shouldn't be
1242 any. However, for some commands, a delay in an ACL can mean that it is
1243 some time before the response is written. In this time, more input might
1244 arrive, invalidly. So now there are extra checks after an ACL has run for
1245 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
1246 pipelining has not been advertised.
1248 PH/30 MH's patch to allow iscntrl() characters to be list separators.
1250 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
1251 returned in the SMTP response when smtp_return_error_details was false.
1252 This has been fixed.
1254 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
1255 instead of the C I/O that was originally supplied, because problems were
1256 reported on Solaris.
1258 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
1259 Exim which did not show up earlier: it was assuming that a call to
1260 SSL_CTX_set_info_callback() might give an error value. In fact, there is
1261 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
1262 was a macro that became an assignment, so it seemed to work. This has
1263 changed to a proper function call with a void return, hence the compile
1264 error. Exim's code has been fixed.
1266 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
1269 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
1271 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
1273 PH/37 If a message is not accepted after it has had an id assigned (e.g.
1274 because it turns out to be too big or there is a timeout) there is no
1275 "Completed" line in the log. When some messages of this type were
1276 selected by exigrep, they were listed as "not completed". Others were
1277 picked up by some special patterns. I have improved the selection
1278 criteria to be more general.
1280 PH/38 The host_find_failed option in the manualroute router can now be set
1281 to "ignore", to completely ignore a host whose IP address cannot be
1282 found. If all hosts are ignored, the behaviour is controlled by the new
1283 host_all_ignored option.
1285 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
1286 homing or because of multiple MX records with /mx) generated more than
1287 one IP address, and the following item turned out to be the local host,
1288 all the secondary addresses of the first item were incorrectly removed
1289 from the list, along with the local host and any following hosts (which
1290 is what is supposed to happen).
1292 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
1293 whoever called Exim into the -H file. In the case of the daemon it was
1294 behaving confusingly. When first started, it used values for whoever
1295 started the daemon, but after a SIGHUP it used the Exim user (because it
1296 calls itself on a restart). I have changed the code so that it now always
1299 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
1300 message are rejected with the same error (e.g. no authentication or bad
1301 sender address), and a DATA command is nevertheless sent (as can happen
1302 with PIPELINING or a stupid MUA), the error message that was given to the
1303 RCPT commands is included in the rejection of the DATA command. This is
1304 intended to be helpful for MUAs that show only the final error to their
1307 PH/42 Another patch from the Sieve maintainer.
1309 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
1310 Eximstats - Fixed some broken HTML links and added missing column headers
1312 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
1313 columns for Rejects, Temp Rejects, Ham, and Spam rows.
1315 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
1317 PH/43 Yet another patch from the Sieve maintainer.
1319 PH/44 I found a way to check for a TCP/IP connection going away before sending
1320 the response to the final '.' that terminates a message, but only in the
1321 case where the client has not sent further data following the '.'
1322 (unfortunately, this is allowed). However, in many cases there won't be
1323 any further data because there won't be any more messages to send. A call
1324 to select() can be used: if it shows that the input is "ready", there is
1325 either input waiting, or the socket has been closed. An attempt to read
1326 the next input character can distinguish the two cases. Previously, Exim
1327 would have sent an OK response which the client would never have see.
1328 This could lead to message repetition. This fix should cure that, at
1329 least in a lot of common cases.
1331 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
1332 advertised in response to EHLO.
1338 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
1339 fixed by 4.65/MH/01 (is this a record?) are fixed:
1341 (i) An empty string was always treated as zero by the numeric comparison
1342 operators. This behaviour has been restored.
1344 (ii) It is documented that the numeric comparison operators always treat
1345 their arguments as decimal numbers. This was broken in that numbers
1346 starting with 0 were being interpreted as octal.
1348 While fixing these problems I realized that there was another issue that
1349 hadn't been noticed. Values of message_size_limit (both the global option
1350 and the transport option) were treated as octal if they started with 0.
1351 The documentation was vague. These values are now always treated as
1352 decimal, and I will make that clear in the documentation.
1358 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
1359 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
1362 MH/01 Don't check that the operands of numeric comparison operators are
1363 integers when their expansion is in "skipping" mode (fixes bug
1364 introduced by 4.64-PH/07).
1366 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
1367 child addresses, Exim now panics and dies. Previously, because the count
1368 is held in a short int, deliveries were likely to be lost. As such a
1369 large number of recipients for a single message is ridiculous
1370 (performance will be very, very poor), I have chosen to impose a limit
1371 rather than extend the field.
1377 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
1378 leftover -K file (the existence of which was triggered by #402).
1379 While we were at it, introduced process PID as part of the -K
1380 filename. This should rule out race conditions when creating
1383 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
1384 processing considerably. Previous code took too long for large mails,
1385 triggering a timeout which in turn triggers #401.
1387 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
1388 in the DK code in transports.c. sendfile() is not really portable,
1389 hence the _LINUX specificness.
1391 TF/01 In the add_headers option to the mail command in an Exim filter,
1392 there was a bug that Exim would claim a syntax error in any
1393 header after the first one which had an odd number of characters
1396 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
1397 callout verification, Exim cached a "reject" for the entire domain. This
1398 is correct for most verifications, but it is not correct for a recipient
1399 verification with use_sender or use_postmaster set, because in that case
1400 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
1401 case of MAIL FROM:<> rejection from other early rejections (e.g.
1402 rejection of HELO). When verifying a recipient using a non-null MAIL
1403 address, the cache is ignored if it shows MAIL FROM:<> rejection.
1404 Whatever the result of the callout, the value of the domain cache is
1405 left unchanged (for any other kind of callout, getting as far as trying
1406 RCPT means that the domain itself is ok).
1408 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
1411 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
1412 manifest itself as EPIPE rather than ECONNECT. When tidying away a
1413 session, the daemon ignores ECONNECT errors and logs others; it now
1414 ignores EPIPE as well.
1416 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
1417 (quoted-printable decoding).
1419 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
1420 later the small subsequent patch to fix an introduced bug.
1422 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
1424 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
1426 PH/08 An error is now given if message_size_limit is specified negative.
1428 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
1429 to be given (somewhat) arbitrary names.
1431 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
1434 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
1435 miscellaneous code fixes
1437 PH/10 Added the log_reject_target ACL modifier to specify where to log
1440 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
1441 hostname. This is wrong, because it relates to the incoming message (and
1442 probably the interface on which it is arriving) and not to the outgoing
1443 callout (which could be using a different interface). This has been
1444 changed to use the value of the helo_data option from the smtp transport
1445 instead - this is what is used when a message is actually being sent. If
1446 there is no remote transport (possible with a router that sets up host
1447 addresses), $smtp_active_hostname is used.
1449 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
1450 tweaks were necessary in order to get it to work (see also 21 below):
1451 (a) The code assumed that strncpy() returns a negative number on buffer
1452 overflow, which isn't the case. Replaced with Exim's string_format()
1454 (b) There were several signed/unsigned issues. I just did the minimum
1455 hacking in of casts. There is scope for a larger refactoring.
1456 (c) The code used strcasecmp() which is not a standard C function.
1457 Replaced with Exim's strcmpic() function.
1458 (d) The code set only $1; it now sets $auth1 as well.
1459 (e) A simple test gave the error "authentication client didn't specify
1460 service in request". It would seem that Dovecot has changed its
1461 interface. Fortunately there's a specification; I followed it and
1462 changed what the client sends and it appears to be working now.
1464 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
1467 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
1468 address A is aliased to B and C, where B exists and C does not. Without
1469 -v the output is "A verified" because verification stops after a
1470 successful redirection if more than one address is generated. However,
1471 with -v the child addresses are also verified. Exim was outputting "A
1472 failed to verify" and then showing the successful verification for C,
1473 with its parentage. It now outputs "B failed to verify", showing B's
1474 parentage before showing the successful verification of C.
1476 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
1477 look up a TXT record in a specific list after matching in a combined
1480 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
1481 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
1482 they consult the DNS. I had assumed they would set it the way they
1483 wanted; and indeed my experiments on Linux seem to show that in some
1484 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
1485 To be on the safe side, however, I have now made the interface to
1486 host_find_byname() similar to host_find_bydns(), with an argument
1487 containing the DNS resolver options. The host_find_byname() function now
1488 sets these options at its start, just as host_find_bydns() does. The smtp
1489 transport options dns_qualify_single and dns_search_parents are passed to
1490 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
1491 of host_find_byname() use the default settings of RES_DEFNAMES
1492 (qualify_single) but not RES_DNSRCH (search_parents).
1494 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
1495 spool_read_header() do less string testing, by means of a preliminary
1496 switch on the second character of optional "-foo" lines. (This is
1497 overdue, caused by the large number of possibilities that now exist.
1498 Originally there were few.) While I was there, I also converted the
1499 str(n)cmp tests so they don't re-test the leading "-" and the first
1500 character, in the hope this might squeeze out yet more improvement.
1502 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
1503 flag allowing group syntax was set by the header_syntax check but not
1504 turned off, possible causing trouble later; (2) The flag was not being
1505 set at all for the header_verify test, causing "group"-style headers to
1506 be rejected. I have now set it in this case, and also caused header_
1507 verify to ignore an empty address taken from a group. While doing this, I
1508 came across some other cases where the code for allowing group syntax
1509 while scanning a header line wasn't quite right (mostly, not resetting
1510 the flag correctly in the right place). These bugs could have caused
1511 trouble for malformed header lines. I hope it is now all correct.
1513 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
1514 with the "reply" argument non-NULL. The code, however (which originally
1515 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
1516 but it didn't always do it. This confused somebody who was copying the
1517 code for some other use. I have removed all the tests.
1519 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
1520 feature that was used to support insecure browsers during the U.S. crypto
1521 embargo. It requires special client support, and Exim is probably the
1522 only MTA that supported it -- and would never use it because real RSA is
1523 always available. This code has been removed, because it had the bad
1524 effect of slowing Exim down by computing (never used) parameters for the
1525 RSA_EXPORT functionality.
1527 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
1528 authenticator to fail if there's a tab character in the incoming data
1529 (there should never be unless someone is messing about, as it's supposed
1530 to be base64-encoded). Also added, on Timo's advice, the "secured" option
1531 if the connection is using TLS or if the remote IP is the same as the
1532 local IP, and the "valid-client-cert option" if a client certificate has
1535 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
1536 authenticators. This can be used for authorization after authentication
1537 succeeds. (In the case of plaintext, it servers for both authentication
1540 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
1541 if any retry times were supplied.
1543 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
1544 connection, where there is no client IP address to check. In this
1545 situation, the verify now always succeeds.
1547 PH/25 Applied John Jetmore's -Mset patch.
1549 PH/26 Added -bem to be like -Mset, but loading a message from a file.
1551 PH/27 In a string expansion for a processed (not raw) header when multiple
1552 headers of the same name were present, leading whitespace was being
1553 removed from all of them, but trailing whitespace was being removed only
1554 from the last one. Now trailing whitespace is removed from each header
1555 before concatenation. Completely empty headers in a concatenation (as
1556 before) are ignored.
1558 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
1559 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
1561 PH/29 [Removed. This was a change that I later backed out, and forgot to
1562 correct the ChangeLog entry (that I had efficiently created) before
1563 committing the later change.]
1565 PH/30 Exim was sometimes attempting to deliver messages that had suffered
1566 address errors (4xx response to RCPT) over the same connection as other
1567 messages routed to the same hosts. Such deliveries are always "forced",
1568 so retry times are not inspected. This resulted in far too many retries
1569 for the affected addresses. The effect occurred only when there were more
1570 hosts than the hosts_max_try setting in the smtp transport when it had
1571 the 4xx errors. Those hosts that it had tried were not added to the list
1572 of hosts for which the message was waiting, so if all were tried, there
1573 was no problem. Two fixes have been applied:
1575 (i) If there are any address or message errors in an SMTP delivery, none
1576 of the hosts (tried or untried) are now added to the list of hosts
1577 for which the message is waiting, so the message should not be a
1578 candidate for sending over the same connection that was used for a
1579 successful delivery of some other message. This seems entirely
1580 reasonable: after all the message is NOT "waiting for some host".
1581 This is so "obvious" that I'm not sure why it wasn't done
1582 previously. Hope I haven't missed anything, but it can't do any
1583 harm, as the worst effect is to miss an optimization.
1585 (ii) If, despite (i), such a delivery is accidentally attempted, the
1586 routing retry time is respected, so at least it doesn't keep
1587 hammering the server.
1589 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
1590 in ${readsocket because some servers need this prod.
1592 PH/32 Added some extra debug output when updating a wait-xxx database.
1594 PH/33 The hint "could be header name not terminated by colon", which has been
1595 given for certain expansion errors for a long time, was not being given
1596 for the ${if def:h_colon_omitted{... case.
1598 PH/34 The spec says: "With one important exception, whenever a domain list is
1599 being scanned, $domain contains the subject domain." There was at least
1600 one case where this was not true.
1602 PH/35 The error "getsockname() failed: connection reset by peer" was being
1603 written to the panic log as well as the main log, but it isn't really
1604 panic-worthy as it just means the connection died rather early on. I have
1605 removed the panic log writing for the ECONNRESET error when getsockname()
1608 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
1609 runs only) independently of the message's sender address. This meant
1610 that, if the 4xx error was in fact related to the sender, a different
1611 message to the same recipient with a different sender could confuse
1612 things. In particualar, this can happen when sending to a greylisting
1613 server, but other circumstances could also provoke similar problems.
1614 I have changed the default so that the retry time for these errors is now
1615 based a combination of the sender and recipient addresses. This change
1616 can be overridden by setting address_retry_include_sender=false in the
1619 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
1620 remote server are returned as part of bounce messages. This was not
1621 happening for LMTP over a pipe (the lmtp transport), but now it is the
1622 same for both kinds of LMTP.
1624 PH/38 Despite being documented as not happening, Exim was rewriting addresses
1625 in header lines that were in fact CNAMEs. This is no longer the case.
1627 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
1628 and queue runs started by the daemon processed all messages. This has
1629 been fixed so that -R and -S can now usefully be given with -q<time>.
1631 PH/40 Import PCRE release 6.7 (fixes some bugs).
1633 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
1635 PH/42 Give an error if -q is specified more than once.
1637 PH/43 Renamed the variables $interface_address and $interface_port as
1638 $received_ip_address and $received_port, to make it clear that these
1639 values apply to message reception, and not to the outgoing interface when
1640 a message is delivered. (The old names remain recognized, of course.)
1642 PH/44 There was no timeout on the connect() call when using a Unix domain
1643 socket in the ${readsocket expansion. There now is.
1645 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
1646 be meaningful with "accept".
1648 SC/01 Eximstats V1.43
1649 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
1651 SC/02 Eximstats V1.44
1652 Use a glob alias rather than an array ref in the generated
1653 parser. This improves both readability and performance.
1655 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
1656 Collect SpamAssassin and rejection statistics.
1657 Don't display local sender or destination tables unless
1658 there is data to show.
1659 Added average volumes into the top table text output.
1661 SC/04 Eximstats V1.46
1662 Collect data on the number of addresses (recipients)
1663 as well as the number of messages.
1665 SC/05 Eximstats V1.47
1666 Added 'Message too big' to the list of mail rejection
1667 reasons (thanks to Marco Gaiarin).
1669 SC/06 Eximstats V1.48
1670 Mainlog lines which have GMT offsets and are too short to
1671 have a flag are now skipped.
1673 SC/07 Eximstats V1.49 (Alain Williams)
1674 Added the -emptyok flag.
1676 SC/08 Eximstats V1.50
1677 Fixes for obtaining the IP address from reject messages.
1679 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
1680 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
1681 whitesspace changes from 4.64-PH/27
1683 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
1686 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
1687 are found, allow negative numbers in numeric criteria)
1689 JJ/06 exipick.20061117.2, added new $message_body_missing variable
1691 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
1692 to match changes made in 4.64-PH/43
1694 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
1696 PH/47 Put in an explicit test for a DNS lookup of an address record where the
1697 "domain" is actually an IP address, and force a failure. This locks out
1698 those revolvers/nameservers that support "A-for-A" lookups, in
1699 contravention of the specifications.
1701 PH/48 When a host name was looked up from an IP address, and the subsequent
1702 forward lookup of the name timed out, the host name was left in
1703 $sender_host_name, contrary to the specification.
1705 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
1706 restricted to single-key lookups, Exim was not diagnosing an error if
1707 * or *@ was used with a query-style lookup.
1709 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
1711 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
1712 long ago, but noone interested enough thought of it. Let's just say that
1713 the "1.1" means that there are some new functions that weren't there at
1714 some point in the past.
1716 PH/51 Error processing for expansion failure of helo_data from an smtp
1717 transport during callout processing was broken.
1719 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
1720 tested/used via the -bh/-bhc/-bs options.
1722 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
1723 bug, fixed in subsequent PCRE releases).
1725 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
1726 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
1728 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
1734 SC/01 Use a glob alias rather than an array ref in eximstats generated
1735 parser. This improves both readability and performance.
1737 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
1738 Don't display local sender or destination tables in eximstats unless
1739 there is data to show.
1740 Added average volumes into the eximstats top table text output.
1742 SC/03 Collect data on the number of addresses (recipients) as well
1743 as the number of messages in eximstats.
1745 TF/01 Correct an error in the documentation for the redirect router. Exim
1746 does (usually) call initgroups() when daemonizing.
1748 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
1749 with consistent privilege compared to when running as a daemon.
1751 TF/03 Note in the spec that $authenticated_id is not set for local
1752 submissions from trusted users.
1754 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
1755 Thanks to Dean Brooks <dean@iglou.com> for the patch.
1757 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
1758 by adding some example configuration directives to the default
1759 configuration file. A little bit of work is required to uncomment the
1760 directives and define how usernames and passwords are checked, but
1761 there is now a framework to start from.
1763 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
1764 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
1765 without this. I don't know how relevant this is to other LDAP libraries.
1767 PH/02 Add the verb name to the "unknown ACL verb" error.
1769 PH/03 Magnus Holmgren's patch for filter_prepend_home.
1771 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
1773 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
1774 directory not expanded when it should be if an expanded home directory
1775 was set for the address (which is overridden by the transport).
1777 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
1780 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
1781 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
1782 because it is too late at that time, and has no effect.
1784 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
1785 security issue with \' (bugzilla #107). I could not use the
1786 PQescapeStringConn() function, because it needs a PGconn value as one of
1789 PH/08 When testing addresses using -bt, indicate those final addresses that
1790 are duplicates that would not cause an additional delivery. At least one
1791 person was confused, thinking that -bt output corresponded to deliveries.
1792 (Suppressing duplicates isn't a good idea as you lose the information
1793 about possibly different redirections that led to the duplicates.)
1795 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
1796 systems where poll() doesn't work, in particular OS X.
1798 PH/10 Added more information to debugging output for retry time not reached.
1800 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
1801 operations in malware.c.
1803 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
1806 PH/13 If write_rejectlog was set false when logging was sent to syslog with
1807 syslog_duplication set false, log lines that would normally be written
1808 both the the main log and to the reject log were not written to syslog at
1811 PH/14 In the default configuration, change the use of "message" in ACL warn
1812 statements to "add_header".
1814 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
1815 not followed by a command (e.g. "seen endif").
1817 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
1818 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
1821 PH/17 Added extra conditions to the default value of delay_warning_condition
1825 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
1826 { match{$h_precedence:}{(?i)bulk|list|junk} } \
1827 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
1830 The Auto-Submitted: and various List- headers are standardised, whereas I
1831 don't think Precedence: ever was.
1833 PH/18 Refactored debugging code in route_finduser() to show more information,
1834 in particular, the error code if getpwnam() issues one.
1836 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
1837 This is apparently needed in addition to the PH/07 change above to avoid
1838 any possible encoding problems.
1840 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
1841 but not after initializing Perl.
1843 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
1844 output them only if debugging. By default they are written stderr,
1845 apparently, which is not desirable.
1847 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
1850 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
1853 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
1855 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
1856 authenticated or an ident call has been made. Suppress the default
1857 values for $authenticated_id and $authenticated_sender (but permit -oMai
1858 and -oMas) when testing with -bh.
1860 PH/24 Re-jigged the order of the tests in the default configuration so that the
1861 tests for valid domains and recipients precede the DNS black list and CSA
1862 tests, on the grounds that those ones are more expensive.
1864 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
1865 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
1866 This bug exists in every version of Exim that I still have, right back to
1869 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
1870 However, an attempt to turn on case-sensitivity in a regex key by
1871 including (?-i) didn't work because the subject string was already
1872 lowercased, and the effects were non-intuitive. It turns out that a
1873 one-line patch can be used to allow (?-i) to work as expected.
1879 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
1880 other effects) broke the use of negated acl sub-conditions.
1882 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
1885 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
1886 "Deny" causes Exim to reject the incoming connection with a 554 error.
1887 Unfortunately, if there is a major crisis, such as a disk failure,
1888 tcp-wrappers gives "deny", whereas what one would like would be some
1889 kind of temporary error. A kludge has been added to help with this.
1890 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
1891 554 error is used if errno is still zero or contains ENOENT (which occurs
1892 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
1895 PH/03 Add -lutil to the default FreeBSD LIBS setting.
1897 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
1898 errors. Otherwise a message that provokes a temporary error (when other
1899 messages do not) can cause a whole host to time out.
1901 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
1902 addresses were routed directly to files or pipes from a redirect router.
1903 File deliveries just didn't batch; pipe deliveries might have suffered
1906 PH/06 A failure to get a lock for a hints database would erroneously always say
1907 "Failed to get write lock", even when it was really a read lock.
1909 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
1910 of 0600. This has been changed to use the value of the lockfile_mode
1911 option (which defaults to 0600).
1913 PH/08 Applied small patch from the Sieve maintainer.
1915 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
1916 folder from quota calculations, a direct delivery into this folder messed
1917 up the contents of the maildirsize file. This was because the regex was
1918 used only to exclude .Trash (or whatever) when the size of the mailbox
1919 was calculated. There was no check that a delivery was happening into an
1920 excluded directory. This bug has been fixed by ignoring all quota
1921 processing for deliveries into excluded directories.
1923 PH/10 Added the maildirfolder_create_regex option to appendfile.
1929 PH/01 The code for finding all the local interface addresses on a FreeBSD
1930 system running IPv6 was broken. This may well have applied to all BSD
1931 systems, as well as to others that have similar system calls. The broken
1932 code found IPv4 interfaces correctly, but gave incorrect values for the
1933 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
1934 that it would not match correctly against @[] and not recognize the IPv6
1937 PH/02 The ipliteral router was not recognizing addresses of the form user@
1938 [ipv6:....] because it didn't know about the "ipv6:" prefix.
1940 PH/03 Added disable_ipv6.
1942 PH/04 Changed $reply_address to use the raw form of the headers instead of the
1943 decoded form, because it is most often used to construct To: headers
1944 lines in autoreplies, and the decoded form may well be syntactically
1945 invalid. However, $reply_address has leading white space removed, and all
1946 newlines turned into spaces so that the autoreply transport does not
1949 PH/05 If group was specified without a user on a router, and no group or user
1950 was specified on a transport, the group from the router was ignored.
1952 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
1953 for visible compile-time settings that can be used to change these
1954 numbers, for those that want even more. Backwards compatibility with old
1955 spool files has been maintained. However, going back to a previous Exim
1956 release will lost any variables that are in spool files.
1958 PH/07 Two small changes when running in the test harness: increase delay when
1959 passing a TCP/IP connection to a new process, in case the original
1960 process has to generate a bounce, and remove special handling of
1961 127.0.0.2 (sic), which is no longer necessary.
1963 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
1964 be the same on different OS.
1966 PH/09 Moved a debug statement in filter processing to avoid a race problem when
1969 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
1970 whether --show-vars was specified or not
1972 JJ/02 exipick: Added support for new ACL variable spool format introduced
1975 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
1976 syntactically invalid From: or Reply-to: line, and a filter used this to
1977 generate an autoreply, and therefore failed to obtain an address for the
1978 autoreply, Exim could try to deliver to a non-existent relative file
1979 name, causing unrelated and misleading errors. What now happens is that
1980 it logs this as a hard delivery error, but does not attempt to create a
1983 PH/11 The exinext utility has a -C option for testing purposes, but although
1984 the given file was scanned by exinext itself; it wasn't being passed on
1985 when Exim was called.
1987 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
1988 an end-of-file indication when reading a command response.
1990 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
1991 compiled. In many other places in Exim, IPv6 addresses are always
1992 recognized, so I have changed this. It also means that IPv4 domain
1993 literals of the form [IPV4:n.n.n.n] are now always recognized.
1995 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
1996 used if the router is not running as root, for example, when verifying at
1997 ACL time, or when using -bh. The debugging output from this situation was
1998 non-existent - all you got was a failure to exec. I have made two
2001 (a) Failures to set uid/gid, the current directory, or a process leader
2002 in a subprocess such as that created by queryprogram now generate
2003 suitable debugging ouput when -d is set.
2005 (b) The queryprogram router detects when it is not running as root,
2006 outputs suitable debugging information if -d is set, and then runs
2007 the subprocess without attempting to change uid/gid.
2009 PH/15 Minor change to Makefile for building test_host (undocumented testing
2012 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
2013 additional section of a DNS packet that returns MX or SRV records.
2014 Instead, it always explicitly searches for A/AAAA records. This avoids
2015 major problems that occur when a DNS server includes only records of one
2016 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
2017 fixed another bug: if SRV records were looked up and the corresponding
2018 address records were *not* found in the additional section, the port
2019 values from the SRV records were lost.
2021 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
2022 using the correct key (the original address) when searching the retry
2023 rules in order to find which one to use for generating the retry hint.
2025 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
2026 adding the default one. Similarly, if it contains a Reply-To: header, the
2027 errors_reply_to option, if set, is not used.
2029 PH/19 When calculating a retry time, Exim used to measure the "time since
2030 failure" by looking at the "first failed" field in the retry record. Now
2031 it does not use this if it is later than than the arrival time of the
2032 message. Instead it uses the arrival time. This makes for better
2033 behaviour in cases where some deliveries succeed, thus re-setting the
2034 "first failed" field. An example is a quota failure for a huge message
2035 when small messages continue to be delivered. Without this change, the
2036 "time since failure" will always be short, possible causing more frequent
2037 delivery attempts for the huge message than are intended.
2038 [Note: This change was subsequently modified - see PH/04 for 4.62.]
2040 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
2041 $1, $2, $3) because the numerical variables can be reset during some
2042 expansion items (e.g. "match"), thereby losing the authentication data.
2044 PH/21 Make -bV show the size of off_t variables so that the test suite can
2045 decide whether to run tests for quotas > 2G.
2047 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
2048 mailbox_size, and mailbox_filecount in the appendfile transport. If a
2049 filecount value is greater than 2G or if a quota value is greater than 2G
2050 on a system where the size of off_t is not greater than 4, a panic error
2053 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
2054 never match. The debug and -bh output now contains an explicit error
2055 message indicating a malformed IPv4 address or mask.
2057 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
2058 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
2059 PH/23 above applies.
2061 PH/25 Do not write to syslog when running in the test harness. The only
2062 occasion when this arises is a failure to open the main or panic logs
2063 (for which there is an explicit test).
2065 PH/26 Added the /no_tell option to "control=freeze".
2067 PH/27 If a host name lookup failed very early in a connection, for example, if
2068 the IP address matched host_lookup and the reverse lookup yielded a name
2069 that did not have a forward lookup, an error message of the form "no IP
2070 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
2071 could be logged. Now it outputs the IP address instead of "NULL".
2073 PH/28 An enabling patch from MH: add new function child_open_exim2() which
2074 allows the sender and the authenticated sender to be set when
2075 submitting a message from within Exim. Since child_open_exim() is
2076 documented for local_scan(), the new function should be too.
2078 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
2079 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
2080 results in an empty string is now treated as unset.
2082 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
2084 PH/31 Added sender_verify_fail logging option.
2086 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
2087 needed by RFC 822 but not by RFC 2822 was commented out. I have now
2088 tidied the source and removed it altogether.
2090 PH/33 When a queue run was abandoned because the load average was too high, a
2091 log line was always written; now it is written only if the queue_run log
2092 selector is set. In addition, the log line for abandonment now contains
2093 information about the queue run such as the pid. This is always present
2094 in "start" and "stop" lines but was omitted from the "abandon" line.
2096 PH/34 Omit spaces between a header name and the colon in the error message that
2097 is given when verify = headers_syntax fails (if there are lots of them,
2098 the message gets confusing).
2100 PH/35 Change the default for dns_check_names_pattern to allow slashes within
2101 names, as there are now some PTR records that contain slashes. This check
2102 is only to protect against broken name servers that fall over on strange
2103 characters, so the fact that it applies to all lookups doesn't matter.
2105 PH/36 Now that the new test suite is complete, we can remove some of the
2106 special code in Exim that was needed for the old test suite. For example,
2107 sorting DNS records because real resolvers return them in an arbitrary
2108 order. The new test suite's fake resolver always returns records in the
2111 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
2112 bounces) except when queue_only is set, to avoid logging races between
2113 the different processes.
2115 PH/38 Panic-die if .include specifies a non-absolute path.
2117 PH/39 A tweak to the "H" retry rule from its user.
2119 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
2120 a label. They prevented compilation on older perls.
2122 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
2123 a warning to be raised on newish perls.
2125 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
2126 on queue. Changes to match documented behaviour of showing count of
2127 messages matching specified criteria.
2129 PH/40 Changed the default ident timeout from 30s to 5s.
2131 PH/41 Added support for the use of login_cap features, on those BSD systems
2132 that have them, for controlling the resources used by pipe deliveries.
2134 PH/42 The content-scanning code uses fopen() to create files in which to put
2135 message data. Previously it was not paying any attention to the mode of
2136 the files. Exim runs with umask(0) because the rest of the code creates
2137 files with open(), and sets the required mode explicitly. Thus, these
2138 files were ending up world-writeable. This was not a big issue, because,
2139 being within the spool directory, they were not world-accessible. I have
2140 created a function called modefopen, which takes an additional mode
2141 argument. It sets umask(777), creates the file, chmods it to the required
2142 mode, then resets the umask. All the relevant calls to fopen() in the
2143 content scanning code have been changed to use this function.
2145 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
2146 to 24 hours. This avoids potential overflow problems when processing G
2147 and H retry rules. I suspect nobody ever tinkers with this value.
2149 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
2151 PH/45 When the plaintext authenticator is running as a client, the server's
2152 challenges are checked to ensure they are valid base64 strings. By
2153 default, the authentication attempt is cancelled if an invalid string is
2154 received. Setting client_ignore_invalid_base64 true ignores these errors.
2155 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
2156 they are received. Thus, the responses can be made to depend on the
2157 challenges. If an invalid string is ignored, an empty string is placed in
2160 PH/46 Messages that are created by the autoreply transport now contains a
2161 References: header, in accordance with RFCs 2822 and 3834.
2163 PH/47 Added authenticated_sender_force to the smtp transport.
2165 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
2167 PH/49 Installed latest patch from the Sieve maintainer.
2169 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
2170 was also set, the appendfile transport was unnecessarily scanning a
2171 directory of message files (e.g. for maildir delivery) to find the count
2172 of files (along with the size), even though it did not need this
2173 information. It now does the scan only if it needs to find either the
2174 size of the count of files.
2176 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
2178 PH/52 Two bugs concerned with error handling when the smtp transport is
2181 (i) Exim was not creating retry information for temporary errors given
2182 for individual recipients after the DATA command when the smtp transport
2183 was used in LMTP mode. This meant that they could be retried too
2184 frequently, and not timed out correctly.
2186 (ii) Exim was setting the flag that allows error details to be returned
2187 for LMTP errors on RCPT commands, but not for LMTP errors for individual
2188 recipients that were returned after the DATA command.
2190 PH/53 This is related to PH/52, but is more general: for any failing address,
2191 when detailed error information was permitted to be returned to the
2192 sender, but the error was temporary, then after the final timeout, only
2193 "retry timeout exceeded" was returned. Now it returns the full error as
2194 well as "retry timeout exceeded".
2196 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
2197 do this, and (what is worse) MTAs that accept it.
2199 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
2200 will now be deprecated.
2202 PH/56 New os.c-cygwin from the Cygwin maintainer.
2204 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
2205 formats (previously only available in exim formats via -bpr, -bpru,
2206 and -bpra. Now also available in native and exiqgrep formats)
2208 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
2209 with very large, slow to parse queues
2211 JJ/08 exipick: added ! as generic prefix to negate any criteria format
2213 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
2215 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
2216 responses to authentication challenges, though it was showing the
2217 challenges; (ii) I've removed the CR characters from the debug output for
2220 PH/58 Allow for the insertion of a newline as well as a space when a string
2221 is turned into more than one encoded-word during RFC 2047 encoding. The
2222 Sieve code now uses this.
2224 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
2225 data_4xx, lost_connection, tls_required.
2227 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
2228 message was being sent as an SMTP response.
2230 PH/61 Add -l and -k options to exicyclog.
2232 PH/62 When verifying, if an address was redirected to one new address, so that
2233 verification continued, and the new address failed or deferred after
2234 having set something in $address_data, the value of $address_data was not
2235 passed back to the ACL. This was different to the case when no
2236 redirection occurred. The value is now passed back in both cases.
2238 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
2239 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
2240 use login_cap.h, so on its own it isn't the distinguishing feature. The
2241 new name refers directly to the setclassresources() function.
2243 PH/65 Added configuration files for NetBSD3.
2245 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
2247 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
2248 is preferred over IPv4.
2250 PH/68 The bounce_return_message and bounce_return_body options were not being
2251 honoured for bounces generated during the reception of non-SMTP messages.
2252 In particular, this applied to messages rejected by the ACL. This bug has
2253 been fixed. However, if bounce_return_message is true and bounce_return_
2254 body is false, the headers that are returned for a non-SMTP message
2255 include only those that have been read before the error was detected.
2256 (In the case of an ACL rejection, they have all been read.)
2258 PH/69 The HTML version of the specification is now built in a directory called
2259 spec_html instead of spec.html, because the latter looks like a path with
2260 a MIME-type, and this confuses some software.
2262 PH/70 Catch two compiler warnings in sieve.c.
2264 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
2265 function verify_get_ident() calls ip_connect() to connect a socket, but
2266 if the "connect()" function timed out, ip_connect() used to close the
2267 socket. However, verify_get_ident() also closes the socket later, and in
2268 between Exim writes to the log, which may get opened at this point. When
2269 the socket was closed in ip_connect(), the log could get the same file
2270 descriptor number as the socket. This naturally causes chaos. The fix is
2271 not to close the socket in ip_connect(); the socket should be closed by
2272 the function that creates it. There was only one place in the code where
2273 this was missing, in the iplookup router, which I don't think anybody now
2274 uses, but I've fixed it anyway.
2276 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
2277 well as to direct DNS lookups. Otherwise the handling of names in host
2278 lists is inconsistent and therefore confusing.
2284 PH/01 Two changes to the default runtime configuration:
2286 (1) Move the checks for relay_from_hosts and authenticated clients from
2287 after to before the (commented out) DNS black list checks.
2289 (2) Add control=submission to the relay_from_hosts and authenticated
2290 clients checks, on the grounds that messages accepted by these
2291 statements are most likely to be submissions.
2293 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
2295 (1) Generate an error if the third argument for the ${prvs expansion is
2298 (2) Treat a missing third argument of ${prvscheck as if it were an empty
2301 (3) Reset the variables that are obtained from the first argument of
2302 ${prvscheck and used in the second argument before leaving the code,
2303 because their memory is reclaimed, so using them afterwards may do
2306 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
2307 one (it's much easier than Tom thought :-).
2309 (5) Because of (4), we can now allow for the use of $prvscheck_result
2310 inside the third argument.
2312 PH/03 For some reason, the default setting of PATH when running a command from
2313 a pipe transport was just "/usr/bin". I have changed it to
2316 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
2317 anything to be listed in the output from -bV.
2319 PH/05 When a filter generated an autoreply, the entire To: header line was
2320 quoted in the delivery log line, like this:
2322 => >A.N.Other <ano@some.domain> <original@ddress> ...
2324 This has been changed so that it extracts the operative address. There
2325 may be more than one such address. If so, they are comma-separated, like
2328 => >ano@some.domain,ona@other.domain <original@ddress> ...
2330 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
2331 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
2332 not being looked up in the rDNS to get a host name, Exim was showing the
2333 IP address twice in Received: lines, even though the IP addresses were
2334 identical. For example:
2336 Received: from [1.2.3.4] (helo=[1.2.3.4])
2338 However, if the real host name was known, it was omitting the HELO data
2339 if it matched the actual IP address. This has been tidied up so that it
2340 doesn't show the same IP address twice.
2342 PH/07 When both +timestamp and +memory debugging was on, the value given by
2343 $tod_xxx expansions could be wrong, because the tod_stamp() function was
2344 called by the debug printing, thereby overwriting the timestamp buffer.
2345 Debugging no longer uses the tod_stamp() function when +timestamp is set.
2347 PH/08 When the original message was included in an autoreply transport, it
2348 always said "this is a copy of the message, including all the headers",
2349 even if body_only or headers_only was set. It now gives an appropriate
2352 PH/09 Applied a patch from the Sieve maintainer which:
2354 o fixes some comments
2355 o adds the (disabled) notify extension core
2356 o adds some debug output for the result of if/elsif tests
2357 o points to the current vacation draft in the documentation
2358 and documents the missing references header update
2362 o fixes a bug in processing the envelope test (when testing
2363 multiple envelope elements, the last element determinted the
2366 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
2367 Electronic Mail") by including:
2369 Auto-submitted: auto-generated
2371 in the messages that it generates (bounce messages and others, such as
2372 warnings). In the case of bounce messages for non-SMTP mesages, there was
2373 also a typo: it was using "Auto_submitted" (underscore instead of
2374 hyphen). Since every message generated by Exim is necessarily in response
2375 to another message, thes have all been changed to:
2377 Auto-Submitted: auto-replied
2379 in accordance with these statements in the RFC:
2381 The auto-replied keyword:
2383 - SHOULD be used on messages sent in direct response to another
2384 message by an automatic process,
2386 - MUST NOT be used on manually-generated messages,
2388 - MAY be used on Delivery Status Notifications (DSNs) and Message
2389 Disposition Notifications (MDNs),
2391 - MUST NOT be used on messages generated by automatic or periodic
2392 processes, except for messages which are automatic responses to
2395 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
2396 to the default Received: header definition.
2398 PH/12 Added log selector acl_warn_skipped (default on).
2400 PH/13 After a successful wildlsearch lookup, discard the values of numeric
2401 variables because (a) they are in the wrong storage pool and (b) even if
2402 they were copied, it wouldn't work properly because of the caching.
2404 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
2405 checking when decoding. Apparently there are clients that generate
2406 overlong encoded strings. Why am I not surprised?
2408 PH/15 If the first argument of "${if match_address" was not empty, but did not
2409 contain an "@" character, Exim crashed. Now it writes a panic log message
2410 and treats the condition as false.
2412 PH/16 In autoreply, treat an empty string for "once" the same as unset.
2414 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
2415 extension "envelope-auth". The code is finished and in agreement with
2416 other implementations, but there is no documentation so far and in fact,
2417 nobody wrote the draft yet. This extension is currently #undef'ed, thus
2418 not changing the active code.
2420 Print executed "if" and "elsif" statements when debugging is used. This
2421 helps a great deal to understand what a filter does.
2423 Document more things not specified clearly in RFC3028. I had all this
2424 sorted out, when out of a sudden new issues came to my mind. Oops."
2426 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
2429 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
2430 canonical form (as documented). However, after a host name lookup from
2431 the IP address, check_host() was doing a simple string comparison with
2432 addresses acquired from the DNS when checking that the found name did
2433 have the original IP as one of its addresses. Since any found IPv6
2434 addresses are likely to be in abbreviated form, the comparison could
2435 fail. Luckily, there already exists a function for doing the comparison
2436 by converting both addresses to binary, so now that is used instead of
2437 the text comparison.
2439 PH/20 There was another similar case to PH/19, when a complete host name was
2440 given in a host list; looking up its IP address could give an abbreviated
2441 form, whereas the current host's name might or might not be abbreviated.
2442 The same fix has been applied.
2448 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
2449 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
2452 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
2453 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
2455 PH/03 Typo: missing ".o" in src/pcre/Makefile.
2457 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
2458 header line, restrict the check to what is listed in RFCs 2369 and 2929.
2459 Also, for "Auto-Submitted", treat anything other than "no" as
2460 non-personal, in accordance with RFC 3834. (Previously it treated
2461 anything starting "auto-" as non-personal.)
2463 TF/01 The control=submission/name=... option had a problem with syntax
2464 errors if the name included a slash character. The /name= option
2465 now slurps the rest of the string, so it can include any characters
2466 but it must come last in the list of options (after /sender_retain
2469 PH/05 Some modifications to the interface to the fake nameserver for the new
2477 TK/01 Added the "success_on_redirect" address verification option. See
2478 NewStuff for rationale and an example.
2480 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
2482 PH/02 Patch to exigrep to allow it to work on syslog lines.
2484 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
2485 fread() to skip over the body file's header line, because in Cygwin the
2486 header line is locked and is inaccessible.
2488 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
2489 co-exist for some time) to make it clear that it is the Exim ID that is
2490 referenced, not the Message-ID: header line.
2492 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
2493 string_format() function, because snprintf() does not exist on all
2496 PH/06 The use of forbid_filter_existstest now also locks out the use of the
2497 ${stat: expansion item.
2499 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
2500 protocol synchronization error", to keep the pedants happy.
2502 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
2503 well as for IRIX systems, when gcc is being used. See the host.c source
2506 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
2508 PH/10 Named domain lists were not working if used in a queue_smtp_domains
2511 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
2512 transport and to the smtp transport in LMTP mode.
2514 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
2516 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
2517 run a filter in a subprocess. This could lead to confusion in subsequent
2518 lookups in the parent process. There should also be a search_tidyup() at
2519 the end of the subprocess.
2521 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
2522 only if the host matched helo_try_verify_hosts, which caused the
2523 verification to occur when the EHLO/HELO command was issued. The ACL just
2524 tested the remembered result. Now, if a previous verification attempt has
2525 not happened, "verify = helo" does it there and then.
2527 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
2529 TK/03 Fix log output including CR from clamd.
2531 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
2532 exist provoked a memory error which could cause a segfault.
2534 PH/15 Installed PCRE 6.2
2536 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
2538 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
2539 of the problem. Specifically, suggested +O2 rather than +O1 for the
2542 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
2544 PH/20 If a delivery was routed to a non-standard port by means of an SRV
2545 record, the port was not correctly logged when the outgoing_port log
2546 selector was set (it logged the transort's default port).
2548 PH/21 Added support for host-specific ports to manualroute, queryprogram,
2549 fallback_hosts, and "hosts" in the smtp transport.
2551 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
2552 host errors such as "Connection refused".
2554 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
2555 authentication with radiusclient 0.4.9:
2557 - Error returned from rc_read_config was caught wrongly
2558 - Username/password not passed on to radius server due to wrong length.
2560 The presumption is that some radiusclient API changes for 4.51/PH/17
2561 were not taken care of correctly. The code is still untested by me (my
2562 Linux distribution still has 0.3.2 of radiusclient), but it was
2563 contributed by a Radius user.
2565 PH/24 When doing a callout, the value of $domain wasn't set correctly when
2566 expanding the "port" option of the smtp transport.
2568 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
2569 while reading a MIME header. Thanks to Tom Hughes for a patch.
2571 PH/24 Include config.h inside local_scan.h so that configuration settings are
2574 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
2575 that in an ACL for RCPT (for example), you can examine exactly what was
2578 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
2579 commands, but it was not correctly comparing the address with the actual
2580 client host address. Thus, it would show the EHLO address in Received:
2581 header lines when this was not necessary.
2583 PH/27 Added the % operator to ${eval:}.
2585 PH/28 Exim tries to create and chdir to its spool directory when it starts;
2586 it should be ignoring failures (because with -C, for example, it has lost
2587 privilege). It wasn't ignoring creation failures other than "already
2590 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
2593 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
2594 because an input line was too long, either on its own, or by virtue of
2595 too many continuations, the temporary file was not being removed, and the
2596 return code was incorrect.
2598 PH/31 Missing "BOOL" in function definition in filtertest.c.
2600 PH/32 Applied Sieve patches from the maintainer.
2602 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
2604 PH/33 Added "verify = not_blind".
2606 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
2607 Local/Makefile (with some defaults set). These are used in built scripts
2608 such as exicyclog, but they have never been used in the exim_install
2609 script (though there are many overriding facilities there). I have
2610 arranged that the exim_install script now takes note of these two
2613 PH/35 Installed configuration files for Dragonfly.
2615 PH/36 When a locally submitted message by a trusted user did not contain a
2616 From: header, and the sender address was obtained from -f or from an SMTP
2617 MAIL command, and the trusted user did not use -F to supply a sender
2618 name, $originator_name was incorrectly used when constructing a From:
2619 header. Furthermore, $originator_name was used for submission mode
2620 messages from external hosts without From: headers in a similar way,
2621 which is clearly wrong.
2623 PH/37 Added control=suppress_local_fixups.
2625 PH/38 When log_selector = +received_sender was set, and the addition of the
2626 sender made the log line's construction buffer exactly full, or one byte
2627 less than full, an overflow happened when the terminating "\n" was
2630 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
2631 when the result of a list match is failure because a DNS lookup failed.
2633 PH/40 RM_COMMAND is now used in the building process.
2635 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
2636 the "build-* directories that it finds.
2638 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
2639 address in a domain literal was a prefix of an interface address.
2641 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
2642 when verifying a sender address, unless rewrite_headers is false.
2644 PH/44 Wrote a long comment about why errors_to addresses are verified as
2645 recipients, not senders.
2647 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
2648 the ratelimit ACL was added.
2650 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
2652 PH/46 Added extra information about PostgreSQL errors to the error string.
2654 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
2655 avoiding the need to install special zones in a real server. This is
2656 backwards compatible; if it can't find the fake resolver, it drops back.
2657 Thus, both old and new test suites can be run.
2659 TF/02 Added util/ratelimit.pl
2661 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
2662 clock is set back in time.
2664 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
2665 Candler <B.Candler@pobox.com>.
2667 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
2668 OK for addresses that are the result of redirections.
2670 PH/48 A number of further additions for the benefit of the new test suite,
2671 including a fake gethostbyname() that interfaces to the fake DNS resolver
2674 TF/06 The fix for widen_domains has also been applied to qualify_single and
2675 search_parents which are the other dnslookup options that can cause
2678 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
2681 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
2683 TF/07 Exim produced the error message "an SRV record indicated no SMTP
2684 service" if it encountered an MX record with an empty target hostname.
2685 The message is now "an MX or SRV record indicated no SMTP service".
2687 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
2688 if the DNS of the sending site is misconfigured. This is quite a
2689 common situation. This change restores the behaviour of treating a
2690 helo verification defer as a failure.
2692 PH/51 If self=fail was set on a router, the bounce message did not include the
2693 actual error message.
2699 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
2701 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
2702 command itself ended in error, the underlying message about the transport
2703 filter timeout was being overwritten with the pipe command error. Now the
2704 underlying error message should be appended to the second error message.
2706 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
2708 PH/02 Reduce the amount of output that "make" produces by default. Full output
2709 can still be requested.
2711 PH/03 The warning log line about a condition test deferring for a "warn" verb
2712 was being output only once per connection, rather than after each
2713 occurrence (because it was using the same function as for successful
2714 "warn" verbs). This seems wrong, so I have changed it.
2716 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
2717 it should not have, which might have caused a crash in the right
2718 circumstances, but probably never did.
2720 PH/04 Installed a modified version of Tony Finch's patch to make submission
2721 mode fix the return path as well as the Sender: header line, and to
2722 add a /name= option so that you can make the user's friendly name appear
2725 TF/03 Added the control = fakedefer ACL modifier.
2727 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
2728 Mark Lowes for thorough testing.
2730 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
2732 TK/03 Merged latest SRS patch from Miles Wilton.
2734 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
2735 with the definition in sysexits.h (which is #included earlier).
2736 Fortunately, Exim does not actually use EX_OK. The code used to try to
2737 preserve the sysexits.h value, by assumimg that macro definitions were
2738 scanned for macro replacements. I have been disabused of this notion,
2739 so now the code just undefines EX_OK before #including unistd.h.
2741 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
2742 in the smtp transport. When a block could not be written in a single
2743 write() function, the timeout was being re-applied to each part-write.
2744 This seems wrong - if the receiver was accepting one byte at a time it
2745 would take for ever. The timeout is now adjusted when this happens. It
2746 doesn't have to be particularly precise.
2748 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
2749 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
2751 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
2752 without a domain if the check to <postmaster@domain> fails.
2754 SC/01 Eximstats: added -xls and the ability to specify output files
2755 (patch written by Frank Heydlauf).
2757 SC/02 Eximstats: use FileHandles for outputing results.
2759 SC/03 Eximstats: allow any combination of xls, txt, and html output.
2761 SC/04 Eximstats: fixed display of large numbers with -nvr option
2763 SC/05 Eximstats: fixed merging of reports with empty tables.
2765 SC/06 Eximstats: added the -include_original_destination flag
2767 SC/07 Eximstats: removed tabs and trailing whitespace.
2769 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
2771 TK/06 MBOX spool code: Add real "From " MBOX separator line
2772 so the .eml file is really in mbox format (even though
2773 most programs do not really care). Patch from Alex Miller.
2775 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
2776 The latter is generated from $received_to and is only set if the
2777 message has one envelope recipient. SA can use these headers,
2778 obviously out-of-the-box. Patch from Alex Miller.
2780 PH/08 The ${def test on a variable was returning false if the variable's
2781 value was "0", contrary to what the specification has always said!
2782 The result should be true unless the variable is empty.
2784 PH/09 The syntax error of a character other than { following "${if
2785 def:variable_name" (after optional whitespace) was not being diagnosed.
2786 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
2787 accidental colon was present, for example, could give incorrect results.
2789 PH/10 Tidied the code in a number of places where the st_size field of a stat()
2790 result is used (not including appendfile, where other changes are about
2793 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
2794 This involved changing a lot of size variables from int to off_t. It
2795 should work with maildirs and everything.
2797 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
2798 spamd dying while we are connected to it.
2800 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
2803 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
2804 function for BDB 4.3.
2806 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
2808 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
2809 the sources for PCRE have been re-arranged and more files are now
2812 PH/15 The code I had for printing potentially long long variables in PH/11
2813 above was not the best (it lost precision). The length of off_t variables
2814 is now inspected at build time, and an appropriate printing format (%ld
2815 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
2816 to be "long long int" or "long int". This is needed for the internal
2817 formatting function string_vformat().
2819 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
2820 the configuration file to be ":syslog", then the script "guesses" where
2821 the logs files are, rather than using the compiled in default. In our
2822 case the guess is not the same as the compiled default, so the script
2823 suddenly stopped working when I started to use syslog. The patch checks
2824 to see if log_file_path is "". If so, it attempts to read it from exim
2825 with no configuration file to get the compiled in version, before it
2826 falls back to the previous guessing code."
2828 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
2829 implementing BATV in an Exim configuration. See NewStuff for the gory
2832 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
2833 Makefile that are specific to HP-UX.
2835 PH/18 If the "use_postmaster" option was set for a recipient callout together
2836 with the "random" option, the postmaster address was used as the MAIL
2837 FROM address for the random test, but not for the subsequent recipient
2838 test. It is now used for both.
2840 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
2841 patch removes a few documentation additions to RFC 3028, because the
2842 latest draft now contains them. It adds the new en;ascii-case comparator
2843 and a new error check for 8bit text in MIME parts. Comparator and
2844 require names are now matched exactly. I enabled the subaddress
2845 extension, but it is not well tested yet (read: it works for me)."
2847 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
2848 rework some of the code of TK/09 above to avoid the hardwired use of
2849 "%lld" and "long long". Replaced the call to snprintf() with a call to
2852 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
2853 records point to non-existent hosts", "retry timeout exceeded", and
2854 "retry time not reached for any host after a long failure period".
2856 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
2857 experimental DomainKeys support:
2859 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
2860 (2) On an error such as an illegally used "control", the wrong name for
2861 the control was given.
2863 These problems did NOT occur unless DomainKeys support was compiled.
2865 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
2867 PH/24 Added ${if match_ip condition.
2869 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
2870 databases so that it will be absolutely obvious if a crash occurs in the
2871 DB library. This is a regular occurrence (often caused by mis-matched
2874 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
2875 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
2876 picked up on a user's system that detects such things. There doesn't seem
2877 to be a gcc warning option for this - only an attribute that has to be
2878 put on the function's prototype. It seems that in Fedora Core 4 they have
2879 set this on a number of new functions. No doubt there will be more in due
2882 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
2883 specify a transport. However, if an address that was verified by such a
2884 router was the subject of a callout, Exim crashed because it tried to
2885 read the rcpt_include_affixes from the non-existent transport. Now it
2886 just assumes that the setting of that option is false. This bug was
2887 introduced by 4.51/PH/31.
2889 PH/28 Changed -d+all to exclude +memory, because that information is very
2890 rarely of interest, but it makes the output a lot bigger. People tend to
2891 do -d+all out of habit.
2893 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
2894 code in os-type was giving problems when libc.so lives in lib64, like on
2897 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
2898 aren't the modern standard, and it seems that some systems' include files
2899 don't always have them. Exim was already checking for some of the newer
2900 ones like T_AAAA, and defining it itself. I've added checks for all the
2901 record types that Exim uses.
2903 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
2904 not automatically generating a new one, as it is supposed to. This
2905 prevented TLS from working. If the file did exist, but contained invalid
2906 data, a new version was generated, as expected. It was only the case of a
2907 non-existent file that was broken.
2909 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
2910 with a change in libdomainkeys > 0.64.
2912 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
2913 from DNS. If the selector record carries the flag, it now has
2914 precedence over the domain-wide flag.
2916 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
2918 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
2919 the use of a port name that isn't defined in /etc/services) occurred, the
2920 message was deferred as in a normal delivery, and thus remained on the
2921 spool, instead of being failed because of the mua_wrapper setting. This
2922 is now fixed, and I tidied up some of the mua_wrapper messages at the
2925 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
2926 the messages in a hash of arrays rather than using individual hashes.
2927 This is a bit cleaner and results in dramatic memory savings, albeit
2928 at a slight CPU cost.
2930 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
2931 as requested by Marc Sherman.
2933 SC/10 Eximstats: added histograms for user specified patterns as requested
2936 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
2938 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
2939 fopen() in the content-scanning modules that did not already have it.
2945 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
2946 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
2948 TK/02 Fix ACL "control" statement not being available in MIME ACL.
2950 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
2952 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
2953 to test Sieve filters that use "vacation".
2955 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
2956 that changes the way the GnuTLS parameters are stored in the cache file.
2957 The new format can be generated externally. For backward compatibility,
2958 if the data in the cache doesn't make sense, Exim assumes it has read an
2959 old-format file, and it generates new data and writes a new file. This
2960 means that you can't go back to an older release without removing the
2963 PH/03 A redirect router that has both "unseen" and "one_time" set does not
2964 work if there are any delivery delays because "one_time" forces the
2965 parent to be marked "delivered", so its unseen clone is never tried
2966 again. For this reason, Exim now forbids the simultaneous setting of
2969 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
2970 redirected to themselves ("homonym" addresses). Read the long ChangeLog
2971 entry if you want to know the details. The fix, however, neglected to
2972 consider the case when local delivery batching is involved. The test for
2973 "previously delivered" was not happening when checking to see if an
2974 address could be batched with a previous (undelivered) one; under
2975 certain circumstances this could lead to multiple deliveries to the same
2978 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
2979 in its include files, and this causes problems building Exim.
2981 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
2982 header_syntax) but Exim was just ignoring anything given after a slash.
2983 In particular, this caused confusion with an attempt to use "verify =
2984 reverse_host_lookup/defer_ok". An error is now given when options are
2985 supplied for verify items that do not have them. (Maybe reverse_host_
2986 lookup should have a defer_ok option, but that's a different point.)
2988 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
2989 defined by RFC 821) to 2048, because there were problems with some AUTH
2990 commands, and RFC 1869 says the size should be increased for extended
2991 SMTP commands that take arguments.
2993 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
2996 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
2997 "unknown" error; now it says that the functionality isn't in the binary.
2999 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
3000 an address' error message when a string expansion fails (syntax or
3001 whatever). Otherwise the password may appear in the log. Following change
3002 PH/42 below, there is no longer a chance of it appearing in a bounce
3005 PH/11 Installed exipick version 20050225.0 from John Jetmore.
3007 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
3008 of its addresses was ever tried. (Bugzilla bug #2.)
3010 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
3011 the result incorrectly in the debug output. (It correctly added a newline
3012 to what was transported.)
3014 TF/01 Added $received_time.
3016 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
3017 commented out examples of how to interface to a virus scanner and to
3018 SpamAssassin. Also added commented examples of av_scanner and
3019 spamd_address settings.
3021 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
3022 and controls are allowed in which ACLs. There were a couple of minor
3023 errors. Some of the entries in the conditions table (which is a table of
3024 where they are NOT allowed) were getting very unwieldy; rewrote them as a
3025 negation of where the condition IS allowed.
3027 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
3029 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
3030 header file does not have a version number, so I've had to invent a new
3031 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
3032 API. The code is untested by me (my Linux distribution still has 0.3.2 of
3033 radiusclient), but it was contributed by a Radius user.
3035 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
3036 files or directories, for OpenSSL.
3038 PH/19 When an Exim process that is running as root has to create an Exim log
3039 file, it does so in a subprocess that runs as exim:exim so as to get the
3040 ownership right at creation (otherwise, other Exim processes might see
3041 the file with the wrong ownership). There was no test for failure of this
3042 fork() call, which would lead to the process getting stuck as it waited
3043 for a non-existent subprocess. Forks do occasionally fail when resources
3044 run out. I reviewed all the other calls to fork(); they all seem to check
3047 PH/20 When checking for unexpected SMTP input at connect time (before writing
3048 the banner), Exim was not dealing correctly with a non-positive return
3049 from the read() function. If the client had disconnected by this time,
3050 the result was a log entry for a synchronization error with an empty
3051 string after "input=" when read() returned zero. If read() returned -1
3052 (an event I could not check), uninitialized data bytes were printed.
3053 There were reports of junk text (parts of files, etc) appearing after
3056 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
3058 PH/22 Added support for macro redefinition, and (re)definition in between
3059 driver and ACL definitions.
3061 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
3062 forgetting to use the resulting value; it was using the unexpanded value.
3064 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
3065 hadn't been configured. The fix is from Juergen Kreileder, who
3066 understands it better than I do:
3068 "Here's what I see happening with three configured cyrus_sasl
3069 authenticators configured (plain, login, cram-md5):
3071 On startup auth_cyrus_sasl_init() gets called for each of these.
3072 This means three calls to sasl_listmech() without a specified mech_list.
3073 => SASL tests which mechs of all available mechs actually work
3074 => three warnings about OTP not working
3075 => the returned list contains: plain, login, cram-md5, digest-md5, ...
3077 With the patch, sasl_listmech() also gets called three times. But now
3078 SASL's mech_list option is set to the server_mech specified in the the
3079 authenticator. Or in other words, the answer from sasl_listmech()
3080 gets limited to just the mech you're testing for (which is different
3082 => the return list contains just 'plain' or 'login', 'cram-md5' or
3083 nothing depending on the value of ob->server_mech.
3085 I've just tested the patch: Authentication still works fine,
3086 unavailable mechs specified in the exim configuration are still
3087 caught, and the auth.log warnings about OTP are gone."
3089 PH/25 When debugging is enabled, the contents of the command line are added
3090 to the debugging output, even when log_selector=+arguments is not
3093 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
3094 answer is "GNU", and only if the return is "GNU/something" is the answer
3097 PH/27 $acl_verify_message is now set immediately after the failure of a
3098 verification in an ACL, and so is available in subsequent modifiers. In
3099 particular, the message can be preserved by coding like this:
3101 warn !verify = sender
3102 set acl_m0 = $acl_verify_message
3104 Previously, $acl_verify_message was set only while expanding "message"
3105 and "log_message" when a very denied access.
3107 PH/28 Modified OS/os.c-Linux with
3109 -#ifndef OS_LOAD_AVERAGE
3110 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
3112 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
3113 nomenclature these days.)
3115 PH/29 Installed patch from the Sieve maintainer that adds the options
3116 sieve_useraddress and sieve_subaddress to the redirect router.
3118 PH/30 In these circumstances:
3119 . Two addresses routed to the same list of hosts;
3120 . First host does not offer TLS;
3121 . First host accepts first address;
3122 . First host gives temporary error to second address;
3123 . Second host offers TLS and a TLS session is established;
3124 . Second host accepts second address.
3125 Exim incorrectly logged both deliveries with the TLS parameters (cipher
3126 and peerdn, if requested) that were in fact used only for the second
3129 PH/31 When doing a callout as part of verifying an address, Exim was not paying
3130 attention to any local part prefix or suffix that was matched by the
3131 router that accepted the address. It now behaves in the same way as it
3132 does for delivery: the affixes are removed from the local part unless
3133 rcpt_include_affixes is set on the transport.
3135 PH/32 Add the sender address, as F=<...>, to the log line when logging a
3136 timeout during the DATA phase of an incoming message.
3138 PH/33 Sieve envelope tests were broken for match types other than :is. I have
3139 applied a patch sanctioned by the Sieve maintainer.
3141 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
3142 the uid or gid is negative. A case of a negative gid caused this to be
3143 noticed. The fix allows for either to be negative.
3145 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
3146 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
3149 PH/36 The change PH/12 above was broken. Fixed it.
3151 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
3152 the grounds that routing the same address twice would always produce the
3153 same answer. This might have been true once, but it is certainly no
3154 longer true now. Routing a child address may depend on the previous
3155 routing that produced that child. Some complicated redirection strategies
3156 went wrong when messages had multiple recipients, and made Exim's
3157 behaviour dependent on the order in which the addresses were given.
3159 I have moved the duplicate checking until after the routing is complete.
3160 Exim scans the addresses that are assigned to local and remote
3161 transports, and removes any duplicates. This means that more work will be
3162 done, as duplicates will always all be routed, but duplicates are
3163 presumably rare, so I don't expect this is of any significance.
3165 For deliveries to pipes, files, and autoreplies, the duplicate checking
3166 still happens during the routing process, since they are not going to be
3169 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
3170 It corrects a timeout issue with spamd. This is Ian's comment: "The
3171 background is that sometimes spamd either never reads data from a
3172 connection it has accepted, or it never writes response data. The exiscan
3173 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
3174 blindly assumes that writes won't block so it may never time out."
3176 PH/39 Allow G after quota size as well as K and M.
3178 PH/40 The value set for $authenticated_id in an authenticator may not contain
3179 binary zeroes or newlines because the value is written to log lines and
3180 to spool files. There was no check on this. Now the value is run through
3181 the string_printing() function so that such characters are converted to
3182 printable escape sequences.
3184 PH/41 $message_linecount is a new variable that contains the total number of
3185 lines in the message. Compare $body_linecount, which is the count for the
3188 PH/42 Exim no longer gives details of delivery errors for specific addresses in
3189 bounce and delay warning messages, except in certain special cases, which
3192 (a) An SMTP error message from a remote host;
3193 (b) A message specified in a :fail: redirection;
3194 (c) A message specified in a "fail" command in a system filter;
3195 (d) A message specified in a FAIL return from the queryprogram router;
3196 (e) A message specified by the cannot_route_message router option.
3198 In these cases only, Exim does include the error details in bounce and
3199 warning messages. There are also a few cases where bland messages such
3200 as "unrouteable address" or "local delivery error" are given.
3202 PH/43 $value is now also set for the "else" part of a ${run expansion.
3204 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
3205 being worked on, but at least Exim now implements the latest version to
3208 PH/45 In a pipe transport, although a timeout while waiting for the pipe
3209 process to complete was treated as a delivery failure, a timeout while
3210 writing the message to the pipe was logged, but erroneously treated as a
3211 successful delivery. Such timeouts include transport filter timeouts. For
3212 consistency with the overall process timeout, these timeouts are now
3213 treated as errors, giving rise to delivery failures by default. However,
3214 there is now a new Boolean option for the pipe transport called
3215 timeout_defer, which, if set TRUE, converts the failures into defers for
3216 both kinds of timeout. A transport filter timeout is now identified in
3219 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
3220 systems where "make" and "gmake" are different, calling "gmake" at top
3221 level broke things. I've arranged for the value of $(MAKE) to be passed
3222 from the Makefile to this script so that it can call the same version of
3226 A note about Exim versions 4.44 and 4.50
3227 ----------------------------------------
3229 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
3230 changes of various kinds. As a consequence, a big documentation update was
3231 needed. This delayed the release for rather longer than seemed good, especially
3232 in the light of a couple of (minor) security issues. Therefore, the changes
3233 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
3234 release. So 4.44 and 4.50 are in effect two different branches that both start
3237 I have left the 4.50 change log unchanged; it contains all the changes since
3238 4.43. The change log for 4.44 is below; many of its items are identical to
3239 those for 4.50. This seems to be the most sensible way to preserve the
3240 historical information.
3246 1. Minor wording change to the doc/README.SIEVE file.
3248 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
3249 computation of the current number of files was incorrect.
3251 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
3252 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
3255 4. Give more explanation in the error message when the command for a transport
3256 filter fails to execute.
3258 5. There are several places where Exim runs a non-Exim command in a
3259 subprocess. The SIGUSR1 signal should be disabled for these processes. This
3260 was being done only for the command run by the queryprogram router. It is
3261 now done for all such subprocesses. The other cases are: ${run, transport
3262 filters, and the commands run by the lmtp and pipe transports.
3264 6. Added CONFIGURE_GROUP build-time option.
3266 7. Some older OS have a limit of 256 on the maximum number of file
3267 descriptors. Exim was using setrlimit() to set 1000 as a large value
3268 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
3269 systems. I've change it so that if it can't get 1000, it tries for 256.
3271 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
3272 was an oversight, and furthermore, ever since the addition of extra
3273 controls (e.g. 4.43/32), the checks on when to allow different forms of
3274 "control" were broken. There should now be diagnostics for all cases when a
3275 control that does not make sense is encountered.
3277 9. Added the /retain_sender option to "control=submission".
3279 10. $recipients is now available in the predata ACL (oversight).
3281 11. Tidy the search cache before the fork to do a delivery from a message
3282 received from the command line. Otherwise the child will trigger a lookup
3283 failure and thereby defer the delivery if it tries to use (for example) a
3284 cached ldap connection that the parent has called unbind on.
3286 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
3287 of $address_data from the recipient verification was clobbered by the
3288 sender verification.
3290 13. The value of address_data from a sender verification is now available in
3291 $sender_address_data in subsequent conditions in the ACL statement.
3293 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
3295 15. Added a new option "connect=<time>" to callout options, to set a different
3298 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
3299 was its contents. (It was OK if the option was not defined at all.)
3301 17. A "Completed" log line is now written for messages that are removed from
3302 the spool by the -Mrm option.
3304 18. New variables $sender_verify_failure and $recipient_verify_failure contain
3305 information about exactly what failed.
3307 19. Added -dd to debug only the daemon process.
3309 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
3310 handles timeouts, both on the server side and network timeouts. Renamed the
3311 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
3313 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
3314 It is now set to "smtps".
3316 22. $host_address is now set to the target address during the checking of
3317 ignore_target_hosts.
3319 23. When checking ignore_target_hosts for an ipliteral router, no host name was
3320 being passed; this would have caused $sender_host_name to have been used if
3321 matching the list had actually called for a host name (not very likely,
3322 since this list is usually IP addresses). A host name is now passed as
3325 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
3326 code that specifies a non-restarting handler (typically sigaction() in
3327 modern systems) in an attempt to fix a rare and obscure crash bug.
3329 25. Narrowed the window for a race in the daemon that could cause it to ignore
3330 SIGCHLD signals. This is not a major problem, because they are used only to
3331 wake it up if nothing else does.
3333 26. A malformed maildirsize file could cause Exim to calculate negative values
3334 for the mailbox size or file count. Odd effects could occur as a result.
3335 The maildirsize information is now recalculated if the size or filecount
3338 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
3339 support for a long time. Removed HAVE_SYS_VFS_H.
3341 28. Installed the latest version of exipick from John Jetmore.
3343 29. In an address list, if the pattern was not a regular expression, an empty
3344 subject address (from a bounce message) matched only if the pattern was an
3345 empty string. Non-empty patterns were not even tested. This was the wrong
3346 because it is perfectly reasonable to use an empty address as part of a
3347 database query. An empty address is now tested by patterns that are
3348 lookups. However, all the other forms of pattern expect the subject to
3349 contain a local part and a domain, and therefore, for them, an empty
3350 address still always fails if the pattern is not itself empty.
3352 30. Exim went into a mad DNS loop when attempting to do a callout where the
3353 host was specified on an smtp transport, and looking it up yielded more
3354 than one IP address.
3356 31. Re-factored the code for checking spool and log partition space into a
3357 function that finds that data and another that does the check. The former
3358 is then used to implement four new variables: $spool_space, $log_space,
3359 $spool_inodes, and $log_inodes.
3361 32. The RFC2047 encoding function was originally intended for short strings
3362 such as real names; it was not keeping to the 75-character limit for
3363 encoded words that the RFC imposes. It now respects the limit, and
3364 generates multiple encoded words if necessary. To be on the safe side, I
3365 have increased the buffer size for the ${rfc2047: expansion operator from
3368 33. It is now permitted to omit both strings after an "if" condition; if the
3369 condition is true, the result is "true". As before, when the second string
3370 is omitted, a false condition yields an empty string. This makes it less
3371 cumbersome to write custom ACL and router conditions.
3373 34. Failure to deliver a bounce message always caused it to be frozen, even if
3374 there was an errors_to setting on the router. The errors_to setting is now
3377 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
3378 canonical form (fully expanded) before being placed in
3379 $sender_host_address.
3381 36. The table in the code that translates DNS record types into text (T_A to
3382 "A" for instance) was missing entries for NS and CNAME. It is just possible
3383 that this could have caused confusion if both these types were looked up
3384 for the same domain, because the text type is used as part of Exim's
3385 per-process caching. But the chance of anyone hitting this buglet seems
3388 37. The dnsdb lookup has been extended in a number of ways.
3390 (1) There is a new type, "zns", which walks up the domain tree until it
3391 finds some nameserver records. It should be used with care.
3393 (2) There is a new type, "mxh", which is like "mx" except that it returns
3394 just the host names, not the priorities.
3396 (3) It is now possible to give a list of domains (or IP addresses) to be
3397 looked up. The behaviour when one of the lookups defers can be
3398 controlled by a keyword.
3400 (4) It is now possible to specify the separator character for use when
3401 multiple records are returned.
3403 38. The dnslists ACL condition has been extended: it is now possible to supply
3404 a list of IP addresses and/or domains to be looked up in a particular DNS
3407 39. Added log_selector=+queue_time_overall.
3409 40. When running the queue in the test harness, wait just a tad after forking a
3410 delivery process, to get repeatability of debugging output.
3412 41. Include certificate and key file names in error message when GnuTLS fails
3413 to set them up, because the GnuTLS error message doesn't include the name
3414 of the failing file when there is a problem reading it.
3416 42. Allow both -bf and -bF in the same test run.
3418 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
3420 44. The "Exiscan patch" is now merged into the mainline Exim source.
3422 45. Sometimes the final signoff response after QUIT could fail to get
3423 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
3424 < 0 before doing a fflush(). This bug looks as though it goes back to the
3425 introduction of TLS in release 3.20, but "sometimes" must have been rare
3426 because the tests only now provoked it.
3428 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
3429 (this can affect the format of dates).
3431 47. exim_tidydb, when checking for the continued existence of a message for
3432 which it has found a message-specific retry record, was not finding
3433 messages that were in split spool directories. Consequently, it was
3434 deleting retry records that should have stayed in existence.
3436 48. Steve fixed some bugs in eximstats.
3438 49. The SPA authentication driver was not abandoning authentication and moving
3439 on to the next authenticator when an expansion was forced to fail,
3440 contradicting the general specification for all authenticators. Instead it
3441 was generating a temporary error. It now behaves as specified.
3443 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
3444 (the order specifies the preference for clients). The order is now AES256,
3445 AES128, 3DES, ARCFOUR128.
3447 51. Small patch to Sieve code - explicitly set From: when generating an
3450 52. Exim crashed if a remote delivery caused a very long error message to be
3451 recorded - for instance if somebody sent an entire SpamAssassin report back
3452 as a large number of 550 error lines. This bug was coincidentally fixed by
3453 increasing the size of one of Exim's internal buffers (big_buffer) that
3454 happened as part of the Exiscan merge. However, to be on the safe side, I
3455 have made the code more robust (and fixed the comments that describe what
3458 53. Now that there can be additional text after "Completed" in log lines (if
3459 the queue_time_overall log selector is set), a one-byte patch to exigrep
3460 was needed to allow it to recognize "Completed" as not the last thing in
3463 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
3464 patch that reportedly fixes this has been added. I am not expert enough to
3465 create a test for it. This is what the patch creator wrote:
3467 "I found a little strange behaviour of ldap code when working with
3468 Windows 2003 AD Domain, where users was placed in more than one
3469 Organization Units. When I tried to give exim partial DN, the exit code
3470 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
3471 But simultaneously result of request was absolutely normal ldap result,
3472 so I produce this patch..."
3474 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
3475 REFERENCE, so I have modified the code to exclude the patch when that macro
3478 55. Some experimental protocols are using DNS PTR records for new purposes. The
3479 keys for these records are domain names, not reversed IP addresses. The
3480 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
3481 leaves it alone. Component reversal etc. now happens only for IP addresses.
3484 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
3486 57. Double the size of the debug message buffer (to 2048) so that more of very
3487 long debug lines gets shown.
3489 58. The exicyclog utility now does better if the number of log files to keep
3490 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
3492 59. Two changes related to the smtp_active_hostname option:
3494 (1) $smtp_active_hostname is now available as a variable.
3495 (2) The default for smtp_banner uses $smtp_active_hostname instead
3496 of $primary_hostname.
3498 60. The host_aton() function is supposed to be passed a string that is known
3499 to be a valid IP address. However, in the case of IPv6 addresses, it was
3500 not checking this. This is a hostage to fortune. Exim now panics and dies
3501 if the condition is not met. A case was found where this could be provoked
3502 from a dnsdb PTR lookup with an IPv6 address that had more than 8
3503 components; fortuitously, this particular loophole had already been fixed
3504 by change 4.50/55 above.
3506 If there are any other similar loopholes, the new check in host_aton()
3507 itself should stop them being exploited. The report I received stated that
3508 data on the command line could provoke the exploit when Exim was running as
3509 exim, but did not say which command line option was involved. All I could
3510 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
3511 running as the user.
3514 61. There was a buffer overflow vulnerability in the SPA authentication code
3515 (which came originally from the Samba project). I have added a test to the
3516 spa_base64_to_bits() function which I hope fixes it.
3519 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
3520 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
3522 63. The daemon start-up calls getloadavg() while still root for those OS that
3523 need the first call to be done as root, but it missed one case: when
3524 deliver_queue_load_max is set with deliver_drop_privilege. This is
3525 necessary for the benefit of the queue runner, because there is no re-exec
3526 when deliver_drop_privilege is set.
3528 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
3529 This has been fixed.
3531 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
3532 was in use, was not putting the data itself into the right store pool;
3533 consequently, it could be overwritten for a subsequent message in the same
3534 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
3537 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
3539 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
3540 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
3541 the function were treating the return as a boolean value, which happened to
3542 work because 0=false and not-0=true, but is not correct code.
3544 68. The host_aton() function was not handling scoped IPv6 addresses (those
3545 with, for example, "%eth0" on the end) correctly.
3547 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
3548 negated items (that is, ~something) in unsigned ints. Some compilers
3549 apparently mutter when there is no cast.
3551 70. If an address verification called from an ACL failed, and did not produce a
3552 user-specific message (i.e. there was only a "system" message), nothing was
3553 put in $acl_verify_message. In this situation, it now puts the system
3556 71. Change 4.23/11 added synchronization checking at the start of an SMTP
3557 session; change 4.31/43 added the unwanted input to the log line - except
3558 that it did not do this in the start of session case. It now does.
3560 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
3561 This is wrong and can cause the other end to generate a synchronization
3562 error if it is another Exim or anything else that does the synchronization
3563 check. A QUIT command is no longer sent after a timeout.
3565 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
3566 during host lookups.
3568 74. The defer_ok option of callout verification was not working if it was used
3569 when verifying addresses in header lines, that is, for this case:
3571 verify = header_sender/callout=defer_ok
3573 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
3574 those file descriptors could be used for SMTP connections. If anything
3575 wrote to stderr (the example that came up was "warn" in embedded Perl), it
3576 could be sent to the SMTP client, causing chaos. The daemon now opens
3577 stdin, stdout, and stderr to /dev/null when it puts itself into the
3580 76. Arrange for output from Perl's "warn" command to be written to Exim's main
3581 log by default. The user can override this with suitable Perl magic.
3583 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
3584 the log message when discard triggers, was not working for the DATA ACL or
3585 for the non-SMTP ACL.
3587 78. Error message wording change in sieve.c.
3589 79. If smtp_accept_max_per_host was set, the number of connections could be
3590 restricted to fewer than expected, because the daemon was trying to set up
3591 a new connection before checking whether the processes handling previous
3592 connections had finished. The check for completed processes is now done
3593 earlier. On busy systems, this bug wouldn't be noticed because something
3594 else would have woken the daemon, and it would have reaped the completed
3597 80. If a message was submitted locally by a user whose login name contained one
3598 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
3599 It caused a spool format error. I have fixed the spool reading code. A
3600 related problem was that the "from" clause in the Received: line became
3601 illegal because of the space(s). It is now covered by ${quote_local_part.
3603 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
3606 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
3608 83. Patch from Sieve maintainer to fix unterminated string problem in
3609 "vacation" handling.
3611 84. Some minor changes to the Linux configuration files to help with other
3612 OS variants using glibc.
3614 85. One more patch for Sieve to update vacation handling to latest spec.
3617 ----------------------------------------------------
3618 See the note above about the 4.44 and 4.50 releases.
3619 ----------------------------------------------------
3625 1. Change 4.43/35 introduced a bug that caused file counts to be
3626 incorrectly computed when quota_filecount was set in an appendfile
3629 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
3630 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
3633 3. Give more explanation in the error message when the command for a transport
3634 filter fails to execute.
3636 4. There are several places where Exim runs a non-Exim command in a
3637 subprocess. The SIGUSR1 signal should be disabled for these processes. This
3638 was being done only for the command run by the queryprogram router. It is
3639 now done for all such subprocesses. The other cases are: ${run, transport
3640 filters, and the commands run by the lmtp and pipe transports.
3642 5. Some older OS have a limit of 256 on the maximum number of file
3643 descriptors. Exim was using setrlimit() to set 1000 as a large value
3644 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
3645 systems. I've change it so that if it can't get 1000, it tries for 256.
3647 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
3648 was an oversight, and furthermore, ever since the addition of extra
3649 controls (e.g. 4.43/32), the checks on when to allow different forms of
3650 "control" were broken. There should now be diagnostics for all cases when a
3651 control that does not make sense is encountered.
3653 7. $recipients is now available in the predata ACL (oversight).
3655 8. Tidy the search cache before the fork to do a delivery from a message
3656 received from the command line. Otherwise the child will trigger a lookup
3657 failure and thereby defer the delivery if it tries to use (for example) a
3658 cached ldap connection that the parent has called unbind on.
3660 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
3661 of $address_data from the recipient verification was clobbered by the
3662 sender verification.
3664 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
3665 was its contents. (It was OK if the option was not defined at all.)
3667 11. A "Completed" log line is now written for messages that are removed from
3668 the spool by the -Mrm option.
3670 12. $host_address is now set to the target address during the checking of
3671 ignore_target_hosts.
3673 13. When checking ignore_target_hosts for an ipliteral router, no host name was
3674 being passed; this would have caused $sender_host_name to have been used if
3675 matching the list had actually called for a host name (not very likely,
3676 since this list is usually IP addresses). A host name is now passed as
3679 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
3680 code that specifies a non-restarting handler (typically sigaction() in
3681 modern systems) in an attempt to fix a rare and obscure crash bug.
3683 15. Narrowed the window for a race in the daemon that could cause it to ignore
3684 SIGCHLD signals. This is not a major problem, because they are used only to
3685 wake it up if nothing else does.
3687 16. A malformed maildirsize file could cause Exim to calculate negative values
3688 for the mailbox size or file count. Odd effects could occur as a result.
3689 The maildirsize information is now recalculated if the size or filecount
3692 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
3693 support for a long time. Removed HAVE_SYS_VFS_H.
3695 18. Updated exipick to current release from John Jetmore.
3697 19. Allow an empty sender to be matched against a lookup in an address list.
3698 Previously the only cases considered were a regular expression, or an
3701 20. Exim went into a mad DNS lookup loop when doing a callout where the
3702 host was specified on the transport, if the DNS lookup yielded more than
3705 21. The RFC2047 encoding function was originally intended for short strings
3706 such as real names; it was not keeping to the 75-character limit for
3707 encoded words that the RFC imposes. It now respects the limit, and
3708 generates multiple encoded words if necessary. To be on the safe side, I
3709 have increased the buffer size for the ${rfc2047: expansion operator from
3712 22. Failure to deliver a bounce message always caused it to be frozen, even if
3713 there was an errors_to setting on the router. The errors_to setting is now
3716 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
3717 canonical form (fully expanded) before being placed in
3718 $sender_host_address.
3720 24. Updated eximstats to version 1.33
3722 25. Include certificate and key file names in error message when GnuTLS fails
3723 to set them up, because the GnuTLS error message doesn't include the name
3724 of the failing file when there is a problem reading it.
3726 26. Expand error message when OpenSSL has problems setting up cert/key files.
3729 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
3730 (this can affect the format of dates).
3732 28. exim_tidydb, when checking for the continued existence of a message for
3733 which it has found a message-specific retry record, was not finding
3734 messages that were in split spool directories. Consequently, it was
3735 deleting retry records that should have stayed in existence.
3737 29. eximstats updated to version 1.35
3738 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
3739 1.35 - bugfix such that pie charts by volume are generated correctly
3741 30. The SPA authentication driver was not abandoning authentication and moving
3742 on to the next authenticator when an expansion was forced to fail,
3743 contradicting the general specification for all authenticators. Instead it
3744 was generating a temporary error. It now behaves as specified.
3746 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
3747 (the order specifies the preference for clients). The order is now AES256,
3748 AES128, 3DES, ARCFOUR128.
3750 31. Small patch to Sieve code - explicitly set From: when generating an
3753 32. Exim crashed if a remote delivery caused a very long error message to be
3754 recorded - for instance if somebody sent an entire SpamAssassin report back
3755 as a large number of 550 error lines. This bug was coincidentally fixed by
3756 increasing the size of one of Exim's internal buffers (big_buffer) that
3757 happened as part of the Exiscan merge. However, to be on the safe side, I
3758 have made the code more robust (and fixed the comments that describe what
3761 33. Some experimental protocols are using DNS PTR records for new purposes. The
3762 keys for these records are domain names, not reversed IP addresses. The
3763 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
3764 leaves it alone. Component reversal etc. now happens only for IP addresses.
3767 34. The host_aton() function is supposed to be passed a string that is known
3768 to be a valid IP address. However, in the case of IPv6 addresses, it was
3769 not checking this. This is a hostage to fortune. Exim now panics and dies
3770 if the condition is not met. A case was found where this could be provoked
3771 from a dnsdb PTR lookup with an IPv6 address that had more than 8
3772 components; fortuitously, this particular loophole had already been fixed
3773 by change 4.50/55 or 4.44/33 above.
3775 If there are any other similar loopholes, the new check in host_aton()
3776 itself should stop them being exploited. The report I received stated that
3777 data on the command line could provoke the exploit when Exim was running as
3778 exim, but did not say which command line option was involved. All I could
3779 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
3780 running as the user.
3783 35. There was a buffer overflow vulnerability in the SPA authentication code
3784 (which came originally from the Samba project). I have added a test to the
3785 spa_base64_to_bits() function which I hope fixes it.
3788 36. The daemon start-up calls getloadavg() while still root for those OS that
3789 need the first call to be done as root, but it missed one case: when
3790 deliver_queue_load_max is set with deliver_drop_privilege. This is
3791 necessary for the benefit of the queue runner, because there is no re-exec
3792 when deliver_drop_privilege is set.
3794 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
3795 was in use, was not putting the data itself into the right store pool;
3796 consequently, it could be overwritten for a subsequent message in the same
3797 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
3800 38. Sometimes the final signoff response after QUIT could fail to get
3801 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
3802 < 0 before doing a fflush(). This bug looks as though it goes back to the
3803 introduction of TLS in release 3.20, but "sometimes" must have been rare
3804 because the tests only now provoked it.
3810 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
3811 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
3812 Now it returns an integer. A number of calls were still expecting a T/F
3813 return. Fortuitously, in all cases, the tests worked in OK situations,
3814 which is the norm. However, things would have gone wrong on any write
3815 failures on the smtp file descriptor. This function is used when sending
3816 messages over SMTP and also when doing verify callouts.
3818 2. When Exim is called to do synchronous delivery of a locally submitted
3819 message (the -odf or -odi options), it no longer closes stderr before doing
3822 3. Implemented the mua_wrapper option.
3824 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
3826 5. Implemented the functions header_remove(), header_testname(),
3827 header_add_at_position(), and receive_remove_recipient(), and exported them
3830 6. If an ACL "warn" statement specified the addition of headers, Exim already
3831 inserted X-ACL-Warn: at the start if there was no header name. However, it
3832 was not making this test for the second and subsequent header lines if
3833 there were newlines in the string. This meant that an invalid header could
3834 be inserted if Exim was badly configured.
3836 7. Allow an ACL "warn" statement to add header lines at the start or after all
3837 the Received: headers, as well as at the end.
3839 8. Added the rcpt_4xx retry error code.
3841 9. Added postmaster_mailfrom=xxx to callout verification option.
3843 10. Added mailfrom=xxxx to the callout verification option, for verify=
3846 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
3847 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
3849 12. Inserted some casts to stop certain compilers warning when using pointer
3850 differences as field lengths or precisions in printf-type calls (mostly
3851 affecting debugging statements).
3853 13. Added optional readline() support for -be (dynamically loaded).
3855 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
3856 same clock tick as a message's arrival, so that its received time was the
3857 same as the "first fail" time on the retry record, and that message
3858 remained on the queue past the ultimate address timeout, every queue runner
3859 would try a delivery (because it was past the ultimate address timeout) but
3860 after another failure, the ultimate address timeout, which should have then
3861 bounced the address, did not kick in. This was a "< instead of <=" error;
3862 in most cases the first failure would have been in the next clock tick
3863 after the received time, and all would be well.
3865 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
3866 being recognized when the domain list was tested by the match_domain
3867 condition in an expansion string.
3869 16. Added the ${str2b64: operator.
3871 17. Exim was always calling setrlimit() to set a large limit for the number of
3872 processes, without checking whether the existing limit was already
3873 adequate. (It did check for the limit on file descriptors.) Furthermore,
3874 errors from getrlimit() and setrlimit() were being ignored. Now they are
3875 logged to the main and panic logs, but Exim does carry on, to try to do its
3876 job under whatever limits there are.
3878 18. Imported PCRE 5.0.
3880 19. Trivial typo in log message " temporarily refused connection" (the leading
3883 20. If the log selector return_path_on_delivery was set and an address was
3884 redirected to /dev/null, the delivery process crashed because it assumed
3885 that a return path would always be set for a "successful" delivery. In this
3886 case, the whole delivery is bypassed as an optimization, and therefore no
3889 21. Internal re-arrangement: the function for sending a challenge and reading
3890 a response while authentication was assuming a zero-terminated challenge
3891 string. It's now changed to take a pointer and a length, to allow for
3892 binary data in such strings.
3894 22. Added the cyrus_sasl authenticator (code supplied by MBM).
3896 23. Exim was not respecting finduser_retries when seeking the login of the
3897 uid under which it was called; it was always trying 10 times. (The default
3898 setting of finduser_retries is zero.) Also, it was sleeping after the final
3899 failure, which is pointless.
3901 24. Implemented tls_on_connect_ports.
3903 25. Implemented acl_smtp_predata.
3905 26. If the domain in control=submission is set empty, Exim assumes that the
3906 authenticated id is a complete email address when it generates From: or
3907 Sender: header lines.
3909 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
3910 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
3911 chown and chgrp in /bin and hostname in /usr/bin.
3913 28. Exim was keeping the "process log" file open after each use, just as it
3914 does for the main log. This opens the possibility of it remaining open for
3915 long periods when the USR1 signal hits a daemon. Occasional processlog
3916 errors were reported, that could have been caused by this. Anyway, it seems
3917 much more sensible not to leave this file open at all, so that is what now
3920 29. The long-running daemon process does not normally write to the log once it
3921 has entered its main loop, and it closes the log before doing so. This is
3922 so that log files can straightforwardly be renamed and moved. However,
3923 there are a couple of unusual error situations where the daemon does write
3924 log entries, and I had neglected to close the log afterwards.
3926 30. The text of an SMTP error response that was received during a remote
3927 delivery was being truncated at 512 bytes. This is too short for some of
3928 the long messages that one sometimes sees. I've increased the limit to
3931 31. It is now possible to make retry rules that apply only when a message has a
3932 specific sender, in particular, an empty sender.
3934 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
3935 it possible to be selective about when SMTP synchronization is enforced.
3937 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
3939 32. Implemented hosts_connection_nolog.
3941 33. Added an ACL for QUIT.
3943 34. Setting "delay_warning=" to disable warnings was not working; it gave a
3946 35. Added mailbox_size and mailbox_filecount to appendfile.
3948 36. Added control = no_multiline_responses to ACLs.
3950 37. There was a bug in the logic of the code that waits for the clock to tick
3951 in the case where the clock went backwards by a substantial amount such
3952 that the microsecond fraction of "now" was more than the microsecond
3953 fraction of "then" (but the whole seconds number was less).
3955 38. Added support for the libradius Radius client library this is found on
3956 FreeBSD (previously only the radiusclient library was supported).
3962 1. When certain lookups returned multiple values in the form name=value, the
3963 quoting of the values was not always being done properly. Specifically:
3964 (a) If the value started with a double quote, but contained no whitespace,
3966 (b) If the value contained whitespace other than a space character (i.e.
3967 tabs or newlines or carriage returns) it was not quoted.
3968 This fix has been applied to the mysql and pgsql lookups by writing a
3969 separate quoting function and calling it from the lookup code. The fix
3970 should probably also be applied to nisplus, ibase and oracle lookups, but
3971 since I cannot test any of those, I have not disturbed their existing code.
3973 2. A hit in the callout cache for a specific address caused a log line with no
3974 reason for rejecting RCPT. Now it says "Previous (cached) callout
3975 verification failure".
3977 3. There was an off-by-one bug in the queryprogram router. An over-long
3978 return line was truncated at 256 instead of 255 characters, thereby
3979 overflowing its buffer with the terminating zero. As well as fixing this, I
3980 have increased the buffer size to 1024 (and made a note to document this).
3982 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
3983 when Exim is waiting for an SMTP response from a remote server, Exim
3984 restarts its select() call on the socket, thereby resetting its timeout.
3985 This is not a problem when such interrupts are rare. Somebody set up a cron
3986 job to run exiwhat every 2 minutes, which is less than the normal select()
3987 timeout (5 or 10 minutes). This meant that the select() timeout never
3988 kicked in because it was always reset. I have fixed this by comparing the
3989 time when an interrupt arrives with the time at the start of the first call
3990 to select(). If more time than the timeout has elapsed, the interrupt is
3991 treated as a timeout.
3993 5. Some internal re-factoring in preparation for the addition of Sieve
3994 extensions (by MH). In particular, the "personal" test is moved to a
3995 separate function, and given an option for scanning Cc: and Bcc: (which is
3996 not set for Exim filters).
3998 6. When Exim created an email address using the login of the caller as the
3999 local part (e.g. when creating a From: or Sender: header line), it was not
4000 quoting the local part when it contained special characters such as @.
4002 7. Installed new OpenBSD configuration files.
4004 8. Reworded some messages for syntax errors in "and" and "or" conditions to
4005 try to make them clearer.
4007 9. Callout options, other than the timeout value, were being ignored when
4008 verifying sender addresses in header lines. For example, when using
4010 verify = header_sender/callout=no_cache
4012 the cache was (incorrectly) being used.
4014 10. Added a missing instance of ${EXE} to the exim_install script; this affects
4015 only the Cygwin environment.
4017 11. When return_path_on_delivery was set as a log selector, if different remote
4018 addresses in the same message used different return paths and parallel
4019 remote delivery occurred, the wrong values would sometimes be logged.
4020 (Whenever a remote delivery process finished, the return path value from
4021 the most recently started remote delivery process was logged.)
4023 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
4024 lines when AUTH and/or TLS are in use. This is the "received protocol"
4025 field. Exim used to use "asmtp" for authenticated SMTP, without any
4026 indication (in the protocol name) for TLS use. Now it follows the RFC and
4027 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
4028 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
4029 names appear in log lines as well as in Received: header lines.
4031 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
4032 extensions, and comparison tests, and to fix some bugs.
4034 14. Changes to the "personal" filter test:
4036 (1) The test was buggy in that it was just doing the equivalent of
4037 "contains" tests on header lines. For example, if a user's address was
4038 anne@some.where, the "personal" test would incorrectly be true for
4040 To: susanne@some.where
4042 This test is now done by extracting each address from the header in turn,
4043 and checking the entire address. Other tests that are part of "personal"
4044 are now done using regular expressions (for example, to check local parts
4045 of addresses in From: header lines).
4047 (2) The list of non-personal local parts in From: addresses has been
4048 extended to include "listserv", "majordomo", "*-request", and "owner-*",
4049 taken from the Sieve specification recommendations.
4051 (3) If the message contains any header line starting with "List-" it is
4052 treated as non-personal.
4054 (4) The test for "circular" in the Subject: header line has been removed
4055 because it now seems ill-conceived.
4057 15. Minor typos in src/EDITME comments corrected.
4059 16. Installed latest exipick from John Jetmore.
4061 17. If headers_add on a router specified a text string that was too long for
4062 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
4063 of string_sprintf() is now avoided.
4065 18. $message_body_size was not set (it was always zero) when running the DATA
4066 ACL and the local_scan() function.
4068 19. For the "mail" command in an Exim filter, no default was being set for
4069 the once_repeat time, causing a random time value to be used if "once" was
4070 specified. (If the value happened to be <= 0, no repeat happened.) The
4071 default is now 0s, meaning "never repeat". The "vacation" command was OK
4072 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
4073 (I found it when inspecting the code).
4075 20. There is now an overall timeout for performing a callout verification. It
4076 defaults to 4 times the callout timeout, which applies to individual SMTP
4077 commands during the callout. The overall timeout applies when there is more
4078 than one host that can be tried. The timeout is checked before trying the
4079 next host. This prevents very long delays if there are a large number of
4080 hosts and all are timing out (e.g. when the network connections are timing
4081 out). The value of the overall timeout can be changed by specifying an
4082 additional sub-option for "callout", called "maxwait". For example:
4084 verify = sender/callout=5s,maxwait=20s
4086 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
4087 to the end before writing, but this should make it even safer).
4089 22. Exim was forgetting that it had advertised PIPELINING for the second and
4090 subsequent messages on an SMTP connection. It was also not resetting its
4091 memory on STARTTLS and an internal HELO.
4093 23. When Exim logs an SMTP synchronization error within a session, it now
4094 records whether PIPELINING has been advertised or not.
4096 24. Added 3 instances of "(long int)" casts to time_t variables that were being
4097 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
4098 rather than long int.
4100 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
4102 26. Added the never_mail option to autoreply.
4108 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
4109 crash if the getsockname() call failed; this can happen if a connection is
4110 closed very soon after it is established. The problem was simply in the
4111 order in which certain operations were done, causing Exim to try to write
4112 to the SMTP stream before it had set up the file descriptor. The bug has
4113 been fixed by making things happen in the correct order.
4119 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
4120 before the connection was closed, thus losing the rejection response.
4122 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
4123 some early Solaris releases, but causes trouble in current releases where
4124 socklen_t is defined.
4126 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
4129 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
4132 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
4133 that releasing the top of it at the end releases what was used for sub-
4134 expansions (unless the block got too big). However, discard this block if
4135 the first thing is a variable or header, so that we can use its block when
4136 it is dynamic (useful for very large $message_headers, for example).
4138 6. Lookups now cache *every* query, not just the most recent. A new, separate
4139 store pool is used for this. It can be recovered when all lookup caches are
4140 flushed. Lookups now release memory at the end of their result strings.
4141 This has involved some general refactoring of the lookup sources.
4143 7. Some code has been added to the store_xxx() functions to reduce the amount
4144 of flapping under certain conditions.
4146 8. log_incoming_interface used to affect only the <= reception log lines. Now
4147 it causes the local interface and port to be added to several more SMTP log
4148 lines, for example "SMTP connection from", and rejection lines.
4150 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
4152 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
4154 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
4155 could be overwritten at the end of the current message (or the start of a
4156 new message if it was set in a HELO ACL). The value is now preserved for
4157 the duration of the SMTP connection.
4159 12. If a transport had a headers_rewrite setting, and a matching header line
4160 contained an unqualified address, that address was qualified, even if it
4161 did not match any rewriting rules. The underlying bug was that the values
4162 of the flags that permit the existence of unqualified sender and recipient
4163 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
4164 non-local messages, and by -bnq for local messages) were not being
4165 preserved with the message after it was received.
4167 13. When Exim was logging an SMTP synchronization error, it could sometimes log
4168 "next input=" as part of the text comprising the host identity instead of
4169 the correct text. The code was using the same buffer for two different
4170 strings. However, depending on which order the printing function evaluated
4171 its arguments, the bug did not always show up. Under Linux, for example, my
4172 test suite worked just fine.
4174 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
4175 doesn't work with some older versions of Perl. It has been changed to "my",
4176 which in any case is probably the better facility to use.
4178 15. A really picky compiler found some instances of statements for creating
4179 error messages that either had too many or two few arguments for the format
4182 16. The size of the buffer for calls to the DNS resolver has been increased
4183 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
4184 for addresses that have a lot of PTR records. This alleviates a problem; it
4185 does not fully solve it.
4187 17. A dnsdb lookup for PTR records that receives more data than will fit in the
4188 buffer now truncates the list and logs the incident, which is the same
4189 action as happens when Exim is looking up a host name and its aliases.
4190 Previously in this situation something unpredictable would happen;
4191 sometimes it was "internal error: store_reset failed".
4193 18. If a server dropped the connection unexpectedly when an Exim client was
4194 using GnuTLS and trying to read a response, the client delivery process
4195 crashed while trying to generate an error log message.
4197 19. If a "warn" verb in an ACL added multiple headers to a message in a single
4198 string, for example:
4200 warn message = H1: something\nH2: something
4202 the text was added as a single header line from Exim's point of view
4203 though it ended up OK in the delivered message. However, searching for the
4204 second and subsequent header lines using $h_h2: did not work. This has been
4205 fixed. Similarly, if a system filter added multiple headers in this way,
4206 the routers could not see them.
4208 20. Expanded the error message when iplsearch is called with an invalid key to
4209 suggest using net-iplsearch in a host list.
4211 21. When running tests using -bh, any delays imposed by "delay" modifiers in
4212 ACLs are no longer actually imposed (and a message to that effect is
4215 22. If a "gecos" field in a passwd entry contained escaped characters, in
4216 particular, if it contained a \" sequence, Exim got it wrong when building
4217 a From: or a Sender: header from that name. A second bug also caused
4218 incorrect handling when an unquoted " was present following a character
4219 that needed quoting.
4221 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
4222 was not being matched caselessly.
4224 24. Arranged for all hyphens in the exim.8 source to be escaped with
4227 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
4228 properly. Recipient callout cache records were still being keyed to include
4229 the sender, even when use_sender was set false. This led to far more
4230 callouts that were necessary. The sender is no longer included in the key
4231 when use_sender is false.
4233 26. Added "control = submission" modifier to ACLs.
4235 27. Added the ${base62d: operator to decode base 62 numbers.
4237 28. dnsdb lookups can now access SRV records.
4239 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
4240 the configuration file.
4242 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
4243 (-v) mode. This makes the output for a verbose queue run more intelligible.
4245 31. Added a use_postmaster feature to recipient callouts.
4247 32. Added the $body_zerocount variable, containing the number of binary zero
4248 bytes in the message body.
4250 33. The time of last modification of the "new" subdirectory is now used as the
4251 "mailbox time last read" when there is a quota error for a maildir
4254 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
4256 35. Added +ignore_unknown as a special item in host lists.
4258 36. Code for decoding IPv6 addresses in host lists is now included, even if
4259 IPv6 support is not being compiled. This fixes a bug in which an IPv6
4260 address was recognized as an IP address, but was then not correctly decoded
4261 into binary, causing unexpected and incorrect effects when compared with
4268 1. Very minor rewording of debugging text in manualroute to say "list of
4269 hosts" instead of "hostlist".
4271 2. If verify=header_syntax was set, and a header line with an unqualified
4272 address (no domain) and a large number of spaces between the end of the
4273 name and the colon was received, the reception process suffered a buffer
4274 overflow, and (when I tested it) crashed. This was caused by some obsolete
4275 code that should have been removed. The fix is to remove it!
4277 3. When running in the test harness, delay a bit after writing a bounce
4278 message to get a bit more predictability in the log output.
4280 4. Added a call to search_tidyup() just before forking a reception process. In
4281 theory, someone could use a lookup in the expansion of smtp_accept_max_
4282 per_host which, without the tidyup, could leave open a database connection.
4284 5. Added the variables $recipient_data and $sender_data which get set from a
4285 lookup success in an ACL "recipients" or "senders" condition, or a router
4286 "senders" option, similar to $domain_data and $local_part_data.
4288 6. Moved the writing of debug_print from before to after the "senders" test
4291 7. Change 4.31/66 (moving the time when the Received: is generated) caused
4292 problems for message scanning, either using a data ACL, or using
4293 local_scan() because the Received: header was not generated till after they
4294 were called (in order to set the time as the time of reception completion).
4295 I have revised the way this works. The header is now generated after the
4296 body is received, but before the ACL or local_scan() are called. After they
4297 are run, the timestamp in the header is updated.
4303 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
4304 before starting a queue runner without re-exec. This happened only when
4305 deliver_drop_privilege was set or when the Exim user was set to root. The
4306 effect of the bug was that timeouts during subsequent deliveries caused
4307 crashes instead of being properly handled. The handler is now left at its
4308 default (and expected) setting.
4310 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
4311 message, again when deliver_drop_privilege is set or Exim is run as root.
4312 The bug described in (1) was not present in this case, but the tidying up
4313 of the other signals was missing. I have made the two cases consistent.
4315 3. The ignore_target_hosts setting on a manualroute router was being ignored
4316 for hosts that were looked up using the /MX notation.
4318 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
4321 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
4322 operated on the sender address. After changing the $sender_address to <>
4323 for the sender address verify, Exim was re-instated it as the original
4324 (before rewriting) address, but remembering that it had rewritten it, so it
4325 wasn't rewriting it again. This bug also had the effect of breaking the
4326 sender address verification caching when the sender address was rewritten.
4328 6. The ignore_target_hosts option was being ignored by the ipliteral router.
4329 This has been changed so that if the ip literal address matches
4330 ignore_target_hosts, the router declines.
4332 7. Added expansion conditions match_domain, match_address, and match_local_
4333 part (NOT match_host).
4335 8. The placeholder for the Received: header didn't have a length field set.
4337 9. Added code to Exim itself and to exim_lock to test for a specific race
4338 condition that could lead to file corruption when using MBX delivery. The
4339 issue is with the lockfile that is created in /tmp. If this file is removed
4340 after a process has opened it but before that process has acquired a lock,
4341 there is the potential for a second process to recreate the file and also
4342 acquire a lock. This could lead to two Exim processes writing to the file
4343 at the same time. The added code performs the same test as UW imapd; it
4344 checks after acquiring the lock that its file descriptor still refers to
4345 the same named file.
4347 10. The buffer for building added header lines was of fixed size, 8192 bytes.
4348 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
4351 11. Added the smtp_active_hostname option. If used, this will typically be made
4352 to depend on the incoming interface address. Because $interface_address is
4353 not set up until the daemon has forked a reception process, error responses
4354 that can happen earlier (such as "too many connections") no longer contain
4357 12. If an expansion in a condition on a "warn" statement fails because a lookup
4358 defers, the "warn" statement is abandoned, and the next ACL statement is
4359 processed. Previously this caused the whole ACL to be aborted.
4361 13. Added the iplsearch lookup type.
4363 14. Added ident_timeout as a log selector.
4365 15. Added tls_certificate_verified as a log selector.
4367 16. Added a global option tls_require_ciphers (compare the smtp transport
4368 option of the same name). This controls incoming TLS connections.
4370 17. I finally figured out how to make tls_require_ciphers do a similar thing
4371 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
4372 before starting the TLS session.
4374 18. Tabs are now shown as \t in -bP output.
4376 19. If the log selector return_path_on_delivery was set, Exim crashed when
4377 bouncing a message because it had too many Received: header lines.
4379 20. If two routers both had headers_remove settings, and the first one included
4380 a superfluous trailing colon, the final name in the first list and the
4381 first name in the second list were incorrectly joined into one item (with a
4382 colon in the middle).
4388 1. Added -C and -D options to the exinext utility, mainly to make it easier
4389 to include in the automated testing, but these could be helpful when
4390 multiple configurations are in use.
4392 2. The exinext utility was not formatting the output nicely when there was
4393 an alternate port involved in the retry record key, nor when there was a
4394 message id as well (for retries that were specific to a specific message
4395 and a specific host). It was also confused by IPv6 addresses, because of
4396 the additional colons they contain. I have fixed the IPv4 problem, and
4397 patched it up to do a reasonable job for IPv6.
4399 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
4400 delivery, the log line now contains "pipelined" if PIPELINING was used.
4402 4. An SMTP transport process used to panic and die if the bind() call to set
4403 an explicit outgoing interface failed. This has been changed; it is now
4404 treated in the same way as a connect() failure.
4406 5. A reference to $sender_host_name in the part of a conditional expansion
4407 that was being skipped was still causing a DNS lookup. This no longer
4410 6. The def: expansion condition was not recognizing references to header lines
4411 that used bh_ and bheader_.
4413 7. Added the _cache feature to named lists.
4415 8. The code for checking quota_filecount in the appendfile transport was
4416 allowing one more file than it should have been.
4418 9. For compatibility with Sendmail, the command line option
4426 and sets the incoming protocol and host name (for trusted callers). The
4427 host name and its colon can be omitted when only the protocol is to be set.
4428 Note the Exim already has two private options, -pd and -ps, that refer to
4429 embedded Perl. It is therefore impossible to set a protocol value of "d" or
4430 "s", but I don't think that's a major issue.
4432 10. A number of refactoring changes to the code, none of which should affect
4435 (a) The number of logging options was getting close to filling up the
4436 32-bit word that was used as a bit map. I have split them into two classes:
4437 those that are passed in the argument to log_write(), and those that are
4438 only ever tested independently outside of that function. These are now in
4439 separate 32-bit words, so there is plenty of room for expansion again.
4440 There is no change in the user interface or the logging behaviour.
4442 (b) When building, for example, log lines, the code previously used a
4443 macro that called string_cat() twice, in order to add two strings. This is
4444 not really sufficiently general. Furthermore, there was one instance where
4445 it was actually wrong because one of the argument was used twice, and in
4446 one call a function was used. (As it happened, calling the function twice
4447 did not affect the overall behaviour.) The macro has been replaced by a
4448 function that can join an arbitrary number of extra strings onto a growing
4451 (c) The code for expansion conditions now uses a table and a binary chop
4452 instead of a serial search (which was left over from when there were very
4453 few conditions). Also, it now recognizes conditions like "pam" even when
4454 the relevant support is not compiled in: a suitably worded error message is
4455 given if an attempt is made to use such a condition.
4457 11. Added ${time_interval:xxxxx}.
4459 12. A bug was causing one of the ddress fields not to be passed back correctly
4460 from remote delivery subprocesses. The field in question was not being
4461 subsequently used, so this caused to problems in practice.
4463 13. Added new log selectors queue_time and deliver_time.
4465 14. Might have fixed a bug in maildirsizefile handling that threw up
4466 "unexpected character" debug warnings, and recalculated the data
4467 unnecessarily. In any case, I expanded the warning message to give more
4470 15. Added the message "Restricted characters in address" to the statements in
4471 the default ACL that block characters like @ and % in local parts.
4473 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
4474 Three changes have been made:
4476 (a) There was a serious bug; a negative response to MAIL caused the whole
4477 recipient domain to be cached as invalid, thereby blocking all messages
4478 to all local parts at the same domain, from all senders. This bug has
4479 been fixed. The domain is no longer cached after a negative response to
4480 MAIL if the sender used is not empty.
4482 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
4485 (c) A new callout option, "use_sender" has been added for people who want
4486 the modified behaviour.
4492 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
4495 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
4496 indeed breaks things for older releases.
4498 3. Added additional logging to the case where there is a problem reading data
4499 from a filter that is running in a subprocess using a pipe, in order to
4500 try to track down a specific problem.
4502 4. Testing facility fudge: when running in the test harness and attempting
4503 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
4504 getting "No route to host". Convert this to a timeout.
4506 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
4509 6. Some OS don't have socklen_t but use size_t instead. This affects the
4510 fifth argument of getsockopt() amongst other things. This is now
4511 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
4512 can be set for individual OS. I have set it for SunOS5, OSF1, and
4513 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
4514 some earlier ones do not.
4516 7. Change 4.30/15 was not doing the test caselessly.
4518 8. The standard form for an IPv6 address literal was being rejected by address
4519 parsing in, for example, MAIL and RCPT commands. An example of this kind of
4520 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
4521 this, as well as the form without the "IPv6" on the front (but only when
4522 address literals are enabled, of course).
4524 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
4526 10. Exim crashed if a message with an empty sender address specified by -f
4527 encountered a router with an errors_to setting. This could be provoked only
4528 by a command such as
4532 where an empty string was supplied; "<>" did not hit this bug.
4534 11. Installed PCRE release 4.5.
4536 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
4537 remained set. It is now erased.
4539 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
4540 times from message ids (which are base 36 rather than the normal 62).
4542 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
4543 were being counted as actual protocol errors, and logged if the log
4544 selector +smtp_protocol_error was set. One cannot be perfect in this test,
4545 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
4546 and DATA following a set of rejected RCPTs do not count as protocol errors.
4547 In other words, Exim assumes they were pipelined, though this may not
4548 actually be the case. Of course, in all cases the client gets an
4549 appropriate error code.
4551 15. If a lookup fails in an ACL condition, a message about the failure may
4552 be available; it is used if testing the ACL cannot continue, because most
4553 such messages specify what the cause of the deferral is. However, some
4554 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
4555 that caused an old message to be retained and used if a later statement
4556 caused a defer, replacing the real cause of the deferral.
4558 16. If an IP address had so many PTR records that the DNS lookup buffer
4559 was not large enough to hold them, Exim could crash while trying to process
4560 the truncated data. It now detects and logs this case.
4562 17. Further to 4.21/58, another change has been made: if (and only if) the
4563 first line of a message (the first header line) ends with CRLF, a bare LF
4564 in a subsequent header line has a space inserted after it, so as not to
4565 terminate the header.
4567 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
4568 unnecessarily, used atoi() instead of strtol(), and didn't check the
4569 termination when getting file sizes from file names by regex.
4571 19. Completely re-implemented the support for maildirsize files, in the light
4572 of a number of problems with the previous contributed implementation
4573 (4.30/29). In particular:
4575 . If the quota is zero, the maildirsize file is maintained, but no quota is
4578 . If the maildir directory does not exist, it is created before any attempt
4579 to write a maildirsize file.
4581 . The quota value in the file is just a cache; if the quota is changed in
4582 the transport, the new value overrides.
4584 . A regular expression is available for excluding directories from the
4587 20. The autoreply transport checks the characters in options that define the
4588 message's headers; it allows continued headers, but it was checking with
4589 isspace() after an embedded newline instead of explicitly looking for a
4592 21. If all the "regular" hosts to which an address was routed had passed their
4593 expiry times, and had not reached their retry times, the address was
4594 bounced, even if fallback hosts were defined. Now Exim should go on to try
4597 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
4598 equivalent code in the SMTP transport. Some hosts send humungous responses
4599 to HELO/EHLO, more than 1024 it seems.
4601 23. Refactoring: code in filter.c used (void *) for "any old type" but this
4602 gives compiler warnings in some environments. I've now done it "properly",
4605 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
4606 (because of problems with the built-in one) was declared to return uschar *
4607 instead of char *, causing compiler failure.
4609 25. Fixed a file descriptor leak when processing alias/forward files.
4611 26. Fixed a minor format string issue in dbfn.c.
4613 27. Typo in exim.c: ("dmbnz" for "dbmnz").
4615 28. If a filter file refered to $h_xxx or $message_headers, and the headers
4616 contained RFC 2047 "words", Exim's memory could, under certain conditions,
4619 29. When a sender address is verified, it is cached, to save repeating the test
4620 when there is more than one recipient in a message. However, when the
4621 verification involves a callout, it is possible for different callout
4622 options to be set for different recipients. It is too complicated to keep
4623 track of this in the cache, so now Exim always runs a verification when a
4624 callout is required, relying on the callout cache for the optimization.
4625 The overhead is duplication of the address routing, but this should not be
4628 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
4629 to be verified with callout=postmaster, and the main callout worked but the
4630 postmaster check failed, the verification correctly failed. However, if a
4631 subsequent RCPT command asked for sender verification *without* the
4632 postmaster check, incorrect caching caused this verification also to fail,
4635 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
4636 it was not caching the DNS options (qualify_single, search_parents) that
4637 were used when the lookup failed. A subsequent lookup with different
4638 options therefore always gave the same answer, though there were cases
4639 where it should not have. (Example: a "domains = !$mx_any" option on a
4640 dnslookup router: the "domains" option is always processed without any
4641 widening, but the router might have qualify_single set.) Now Exim uses the
4642 cached value only when the same options are set.
4644 32. Added John Jetmore's "exipick" utility to the distribution.
4646 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
4647 than a timeout (e.g. a certificate is required, and is not provided), an
4648 Exim server now closes the connection immediately. Previously it waited for
4649 the client to close - but if the client is SSL, it seems that they each
4650 wait for each other, leading to a delay before one of them times out.
4652 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
4653 maintained 0.8.x compatibility because I don't think many are using it, and
4654 it is clearly obsolete.
4656 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
4659 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
4660 client certificate was expired. A simple patch fixes this, though I don't
4661 understand the full logic of why the verify callback is called multiple
4664 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
4665 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
4666 which causes problems with some clients (such as the Certicom SSL Plus
4667 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
4668 disables the coutermeasure allowing Eudora to connect."
4670 38. Exim was not checking that a write() to a log file succeeded. This could
4671 lead to Bad Things if a log got too big, in particular if it hit a file
4672 size limit. Exim now panics and dies if it cannot write to a log file, just
4673 as it does if it cannot open a log file.
4675 39. Modified OS/Makefile-Linux so that it now contains
4677 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
4679 The two -D definitions ensure that Exim is compiled with large file
4680 support, which makes it possible to handle log files that are bigger than
4683 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
4684 instance) a domain was checked against a named list that involved a lookup,
4685 causing $domain_data to be set, then another domain was checked against the
4686 same list, then the first domain was re-checked, the value of $domain_data
4687 after the final check could be wrong. In particular, if the second check
4688 failed, it could be set empty. This bug probably also applied to
4691 41. The strip_trailing_dot option was not being applied to the address given
4692 with the -f command-line option.
4694 42. The code for reading a message's header from the spool was incrementing
4695 $received_count, but never initializing it. This meant that the value was
4696 incorrect (doubled) while delivering a message in the same process in which
4697 it was received. In the most common configuration of Exim, this never
4698 happens - a fresh exec is done - but it can happen when
4699 deliver_drop_privilege is set.
4701 43. When Exim logs an SMTP synchronization error - client data sent too soon -
4702 it now includes up to 150 characters of the unexpected data in the log
4705 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
4706 and building data strings. The size of both of these buffers was 10 000
4707 bytes - far larger than anybody would *ever* want, thought I. Needless to
4708 say, somebody hit the limit. I have increased the maximum line length to
4709 20 000 and the maximum data length of concatenated lines to 100 000. I have
4710 also fixed two bugs, because there was no checking on these buffers. Tsk,
4711 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
4712 buffer is too small.
4714 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
4715 lsearch lookups. Now it does.
4717 46. When parsing a route_list item in a manualroute router, a fixed-length
4718 buffer was used for the list of hosts. I made this 1024 bytes long,
4719 thinking that nobody would ever have a list of hosts that long. Wrong.
4720 Somebody had a whole pile of complicated expansion conditions, and the
4721 string was silently truncated, leading to an expansion error. It turns out
4722 that it is easier to change to an unlimited length (owing to other changes
4723 that have happened since this code was originally written) than to build
4724 structure for giving a limitation error. The length of the item that
4725 expands into the list of hosts is now unlimited.
4727 47. The lsearch lookup could not handle data where the length of text line was
4728 more than 4095 characters. Such lines were truncated, leading to shortened
4729 data being returned. It should now handle lines of any length.
4731 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
4732 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
4735 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
4736 replacing strings such as BIN_DIRECTORY in a source file: the replacement
4737 no longer happens in comment lines. A list of replacements is now placed
4738 at the head of all of the source files, except those whose only change is
4739 to replace PERL_COMMAND in the very first #! line.
4741 50. Replaced the slow insertion sort in queue.c, for sorting the list of
4742 messages on the queue, with a bottom-up merge sort, using code contributed
4743 by Michael Haardt. This should make operations like -bp somewhat faster on
4744 large queues. It won't affect queue runners, except when queue_run_in_order
4747 51. Installed eximstats 1.31 in the distribution.
4749 52. Added support for SRV lookups to the dnslookup router.
4751 53. If an ACL referred to $message_body or $message_body_end, the value was not
4752 reset for any messages that followed in the same SMTP session.
4754 54. The store-handling optimization for building very long strings was not
4755 differentiating between the different store pools. I don't think this
4756 actually made any difference in practice, but I've tidied it.
4758 55. While running the routers to verify a sender address, $sender_address
4759 was still set to the sender address. This is wrong, because when routing to
4760 send a bounce to the sender, it would be empty. Therefore, I have changed
4761 it so that, while verifying a sender address, $sender_address is set to <>.
4762 (There is no change to what happens when verifying a recipient address.)
4764 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
4765 target A or AAAA records (if not already returned) without resetting the
4766 qualify_single or search_parents options of the DNS resolver. These are
4767 inappropriate in this case because the targets of MX and SRV records must
4768 be FQDNs. A broken DNS record could cause trouble if it happened to have a
4769 target that, when qualified, matched something in the local domain. These
4770 two options are now turned off when doing these lookups.
4772 57. It seems that at least some releases of Reiserfs (which does not have the
4773 concept of a fixed number of inodes) returns zero and not -1 for the
4774 number of available inodes. This interacted badly with check_spool_inodes,
4775 which assumed that -1 was the "no such thing" setting. What I have done is
4776 to check that the total number of inodes is greater than zero before doing
4777 the test of how many are available.
4779 58. When a "warn" ACL statement has a log_message modifier, the message is
4780 remembered, and not repeated. This is to avoid a lot of repetition when a
4781 message has many recipients that cause the same warning to be written.
4782 Howewer, Exim was preserving the list of already written lines for an
4783 entire SMTP session, which doesn't seem right. The memory is now reset if a
4784 new message is started.
4786 59. The "rewrite" debugging flag was not showing the result of rewriting in the
4787 debugging output unless log_rewrite was also set.
4789 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
4790 of (int)(handle) when we know that handle contains (void *)(-1).
4792 61. The Exim daemon panic-logs an error return when it closes the incoming
4793 connection. However "connection reset by peer" seems to be common, and
4794 isn't really an error worthy of noting specially, so that particular error
4797 62. When Exim is trying to find all the local interfaces, it used to panic and
4798 die if the ioctl to get the interface flags failed. However, it seems that
4799 on at least one OS (Solaris 9) it is possible to have an interface that is
4800 included in the list of interfaces, but for which you get a failure error
4801 for this call. This happens when the interface is not "plumbed" into a
4802 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
4803 failure of the "get flags" call assumes that the interface is down.
4805 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
4806 makes life easier for people who are doing arithmetic on fields extracted
4807 from dates, where you often get leading zeros that should not be
4808 interpreted as octal.
4810 64. Added qualify_domain to the redirect router, to override the global
4813 65. If a pathologically long header line contained very many addresses (the
4814 report of this problem mentioned 10 000) and each of them was rewritten,
4815 Exim could use up a very large amount of memory. (It kept on making new
4816 copies of the header line as it rewrote, and never released the old ones.)
4817 At the expense of a bit more processing, the header rewriting function has
4818 been changed so that it no longer eats memory in this way.
4820 66. The generation of the Received: header has been moved from the time that a
4821 message starts to be received, to the time that it finishes. The timestamp
4822 in the Received: header should now be very close to that of the <= log
4823 line. There are two side-effects of this change:
4825 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
4826 logged header lines no longer include the local Received: line, because
4827 it has not yet been created. The same applies to a copy of the message
4828 that is returned to a non-SMTP sender when a message is rejected.
4830 (b) When a filter file is tested using -bf, no additional Received: header
4831 is added to the test message. After some thought, I decided that this
4834 This change does not affect the value of $received_for. It is still set
4835 after address rewriting, but before local_scan() is called.
4837 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
4839 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
4840 gave an unhelpful panic error message, and a defer error. I have managed to
4841 change this behaviour so that it now rejects any supplied certificate,
4842 which seems right, as the list of acceptable certificates is empty.
4844 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
4845 gave an unhelpful defer error. I have not managed to make this reject any
4846 supplied certificates, but the error message it gives is "no certificate
4847 supplied", which is not helpful.
4849 70. exigrep's output now also includes lines that are not associated with any
4850 message, but which match the given pattern. Implemented by a patch from
4851 Martin Sluka, which also tidied up the Perl a bit.
4853 71. Recipient callout verification, like sender verification, was using <> in
4854 the MAIL FROM command. This isn't really the right thing, since the actual
4855 sender may affect whether the remote host accepts the recipient or not. I
4856 have changed it to use the actual sender in the callout; this means that
4857 the cache record is now keyed on a recipient/sender pair, not just the
4858 recipient address. There doesn't seem to be a real danger of callout loops,
4859 since a callout by the remote host to check the sender would use <>.
4860 [SEE ABOVE: changed after hitting problems.]
4862 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
4863 temporary errors. However, in the case of such a code being given after
4864 the end of a data transmission (i.e. after ".") Exim was failing to write
4865 a retry record for the message. (Yes, there was some broken host that was
4866 actually sending 8xx at this point.)
4868 73. An unknown lookup type in a host list could cause Exim to panic-die when
4869 the list was checked. (An example that provoked this was putting <; in the
4870 middle of a list instead of at the start.) If this happened during a DATA
4871 ACL check, a -D file could be left lying around. This kind of configuration
4872 error no longer causes Exim to die; instead it causes a defer errror. The
4873 incident is still logged to the main and panic logs.
4875 74. Buglet left over from Exim 3 conversion. The message "too many messages
4876 in one connection" was written to the rejectlog but not the mainlog, except
4877 when address rewriting (yes!) was being logged.
4879 75. Added write_rejectlog option.
4881 76. When a system filter was run not as root (that is, when system_filter_user
4882 was set), the values of the $n variables were not being returned to the
4883 main process; thus, they were not subsequently available in the $sn
4886 77. Added +return_path_on_delivery log selector.
4888 78. A connection timeout was being treated differently from recipients deferred
4889 when testing hosts_max_try with a message that was older than the host's
4890 retry timeout. (The host should not be counted, thus allowing all hosts to
4891 be tried at least once before bouncing.) This may have been the cause of an
4892 occasionally reported bug whereby a message would remain on the queue
4893 longer than the retry timeout, but would be bounced if a delivery was
4894 forced. I say "may" because I never totally pinned down the problem;
4895 setting up timeout/retry tests is difficult. See also the next item.
4897 79. The ultimate address timeout was not being applied to errors that involved
4898 a combination of host plus message (for example, a timeout on a MAIL
4899 command). When an address resolved to a number of possible hosts, and they
4900 were not all tried for each delivery (e.g. because of hosts_max_try), a
4901 message could remain on the queue longer than the retry timeout.
4903 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
4906 81. Fixed an obscure SMTP outgoing bug which required at least the following
4907 conditions: (a) there was another message waiting for the same server;
4908 (b) the server returned 5xx to all RCPT commands in the first message so
4909 that the message was not completed; (c) the server dropped the connection
4910 or gave a negative response to the RSET that Exim sends to abort the
4911 transaction. The observed case was a dropped connection after DATA that had
4912 been sent in pipelining mode. That is, the server had advertised PIPELINING
4913 but was not implementing it correctly. The effect of the bug was incorrect
4914 behaviour, such as trying another host, and this could lead to a crash.
4920 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
4921 and daemon.c were passed as pointers to ints; they should have been
4922 pointers to socklen_t variables (which are typically unsigned ints).
4924 2. Some signed/unsigned type warnings in the os.c file for Linux have been
4927 3. Fixed a really odd bug that affected only the testing scheme; patching a
4928 certain fixed string in the binary changed the value of another string that
4929 happened to be identical to the end of the original first string.
4931 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
4932 name", it returns that address as the IP address. On some operating
4933 systems (e.g. Solaris), it also passes back the IP address string as the
4934 "host name". However, on others (e.g. Linux), it passes back an empty
4935 string. Exim wasn't checking for this, and was changing the host name to an
4936 empty string, assuming it had been canonicized.
4938 5. Although rare, it is permitted to have more than one PTR record for a given
4939 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
4940 all the names associated with an address, because they do in Solaris.
4941 However, it seems that they do not in Linux for data that comes from the
4942 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
4943 I found this out when I moved to a new Linux workstation and tried to run
4944 the Exim test suite.
4946 To get round this problem I have changed the code so that it now does its
4947 own call to the DNS to look up PTR records when searching for a host name.
4948 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
4949 addresses that are only in /etc/hosts are still found.
4951 This behaviour is, however, controlled by an option called host_lookup_
4952 order, which defaults to "bydns:byaddr". If people want to use the other
4953 order, or indeed, just use one or the other means of lookup, they can
4954 specify it in this variable.
4956 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
4957 some operating systems, this comes back from gethostbyaddr() as an empty
4958 string, and this is what Exim used to test for. However, it seems that in
4959 other systems, "." is yielded. Exim now tests for this case too.
4961 7. The values of check_spool_space and check_log_space are now held internally
4962 as a number of kilobytes instead of an absolute number of bytes. If a
4963 numbers is specified without 'K' or 'M', it is rounded up to the nearest
4964 kilobyte. This means that much larger values can be stored.
4966 8. Exim monitor: an attempt to get the action menu when not actually pointing
4967 at a message produces an empty menu entitled "No message selected". This
4968 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
4969 no entries in it ("Shell widget menu has zero width and/or height"). So I
4970 have added a single, blank menu entry in this case.
4972 9. Added ${quote_local_part.
4974 10. MIME decoding is now applied to the contents of Subject: header lines when
4977 11. Now that a reference to $sender_host_address automatically causes a reverse
4978 lookup to occur if necessary (4.13/18), there is no need to arrange for a
4979 host lookup before query-style lookups in lists that might use this
4980 variable. This has therefore been abolished, and the "net-" prefix is no
4981 longer necessary for query-style lookups.
4983 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
4984 have been a typo for LFLAGS, so it has been changed.
4986 13. The install script calls Exim with "-C /dev/null" in order to find the
4987 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
4988 to be output. Howeve, since Exim outputs its version number before the
4989 error, it didn't break the script. It just looked ugly. I fixed this by
4990 always allowing "-C /dev/null" if the caller is root.
4992 14. Ignore overlarge ACL variable number when reading spool file - insurance
4993 against a later release with more variables having written the file.
4995 15. The standard form for an IPv6 address literal was being rejected by EHLO.
4996 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
4997 this, as well as the form without the "IPv6" on the front.
4999 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
5000 OS/Makefile-Darwin file.
5002 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
5003 only to LDAP libraries that do not have LDAP_OPT_DEREF.
5005 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
5006 variable. However, some OS use ints for inodes. I've added cast to long int
5007 to get rid of the compiler warning.
5009 19. I had forgotten to lock out "/../" in configuration file names when
5010 ALT_CONFIG_PREFIX was set.
5012 20. Routers used for verification do not need to specify transports. However,
5013 if such a router generated a host list, and callout was configured, Exim
5014 crashed, because it could not find a port number from the (non-existent)
5015 transport. It now assumes port 25 in this circumstance.
5017 21. Added the -t option to exigrep.
5019 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
5020 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
5021 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
5022 them is set, LOOKUP_LSEARCH is forced.
5024 23. "exim -bV" now outputs a list of lookups that are included in the binary.
5026 24. Added sender and host information to the "rejected by local_scan()" log
5027 line; previously there was no indication of these.
5029 25. Added .include_if_exists.
5031 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
5032 Linux. It turns out that not all file systems support this. Apparently some
5033 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
5034 do it.) To cope with this, the error EINVAL, which means that sync-ing is
5035 not supported on the file descriptor, is now ignored when Exim is trying to
5036 sync a directory. This applies only to Linux.
5038 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
5040 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
5041 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
5044 29. Added support for maildirsize files from supplied patch (modified a bit).
5046 30. The use of :fail: followed by an empty string could lead Exim to respond to
5047 sender verification failures with (e.g.):
5049 550 Verification failed for <xxx>
5050 550 Sender verify failed
5052 where the first response line was missing the '-' that indicates it is not
5053 the final line of the response.
5055 31. The loop for finding the name of the user that called Exim had a hardwired
5056 limit of 10; it now uses the value of finduser_retries, which is used for
5057 all other user lookups.
5059 32. Added $received_count variable, available in data and not_smtp ACLs, and at
5062 33. Exim was neglecting to zero errno before one call of strtol() when
5063 expanding a string and expecting an integer value. On some systems this
5064 resulted in spurious "integer overflow" errors. Also, it was casting the
5065 result into an int without checking.
5067 34. Testing for a connection timeout using "timeout_connect" in the retry rules
5068 did not work. The code looks as if it has *never* worked, though it appears
5069 to have been documented since at least releast 1.62. I have made it work.
5071 35. The "timeout_DNS" error in retry rules, also documented since at least
5072 1.62, also never worked. As it isn't clear exactly what this means, and
5073 clearly it isn't a major issue, I have abolished the feature by treating it
5074 as "timeout", and writing a warning to the main and panic logs.
5076 36. The display of retry rules for -brt wasn't always showing the error code
5079 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
5080 timeout_connect_A, timeout_connect_MX.
5082 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
5083 to the empty sender.
5085 39. The daemon was not analysing the content of -oX till after it had closed
5086 stderr and disconnected from the controlling terminal. This meant that any
5087 syntax errors were only noted on the panic log, and the return code from
5088 the command was 0. By re-arranging the code a little, I've made the
5089 decoding happen first, so such errors now appear on stderr, and the return
5090 code is 1. However, the actual setting up of the sockets still happens in
5091 the disconnected process, so errors there are still only recorded on the
5094 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
5095 connections (as happens on some IP stacks) was logged at start up time as
5096 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
5097 it from "IPv6 and IPv4", which means that two separate sockets are being
5100 41. The debug output for gethostbyname2() or getipnodebyname() failures now
5101 says whether AF_INET or AF_INET6 was passed as an argument.
5103 42. Exiwhat output was messed up when time zones were included in log
5106 43. Exiwhat now gives more information about the daemon's listening ports,
5107 and whether -tls-on-connect was used.
5109 44. The "port" option of the smtp transport is now expanded.
5111 45. A "message" modifier in a "warn" statement in a non-message ACL was being
5112 silently ignored. Now an error message is written to the main and panic
5115 46. There's a new ACL modifier called "logwrite" which writes to a log file
5116 as soon as it is encountered.
5118 47. Added $local_user_uid and $local_user_gid at routing time.
5120 48. Exim crashed when trying to verify a sender address that was being
5123 49. Exim was recognizing only a space character after ".include". It now also
5124 recognizes a tab character.
5126 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
5127 extracting the relevant information from the specification. The man page no
5128 longer contains scrambled data for the -d option, and I've added a section
5129 at the front about calling Exim under different names.
5131 51. Added "extra_headers" argument to the "mail" command in filter files.
5133 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
5136 53. Installed eximstats 1.29.
5138 54. Added transport_filter_timeout as a generic transport option.
5140 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
5141 Cc: header lines. This was required by RFC 822, but it not required by RFC
5144 56. Exim used to add From:, Date:, and Message-Id: header lines to any
5145 incoming messages that did not have them. Now it does so only if the
5146 message originates locally, that is, if there is no associated remote host
5147 address. When Resent- header lines are present, this applies to the Resent-
5148 lines rather than the non-Resent- lines.
5150 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
5151 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
5153 58. Messages for configuration errors now include the name of the main
5154 configuration file - useful now that there may be more than one file in a
5155 list (.included file names were always shown).
5157 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
5158 for those rare installations that do not start the daemon as root or run it
5159 setuid root. I've cut out the call to initgroups() if the daemon is not
5162 60. The Exim user and group can now be bound into the binary as text strings
5163 that are looked up at the start of Exim's processing.
5165 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
5167 62. Added $mailstore_basename variable.
5169 63. Installed patch to sieve.c from Michael Haardt.
5171 64. When Exim failed to open the panic log after failing to open the main log,
5172 the original message it was trying to log was written to stderr and debug
5173 output, but if they were not available (the usual case in production), it
5174 was lost. Now it is written to syslog before the two lines that record the
5175 failures to open the logs.
5177 65. Users' Exim filters run in subprocesses under the user's uid. It is
5178 possible for a "deliver" command or an alias in a "personal" command to
5179 provoke an address rewrite. If logging of address rewriting is configured,
5180 this fails because the process is not running as root or exim. There may be
5181 a better way of dealing with this, but for the moment (because 4.30 needs
5182 to be released), I have disabled address rewrite logging when running a
5183 filter in a non-root, non-exim process.
5189 1. The buildconfig auxiliary program wasn't quoting the value set for
5190 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
5191 not defined. This bug was masked in 4.22 by the effect that was fixed in
5194 2. Some messages that were rejected after a message id was allocated were
5195 shown as "incomplete" by exigrep. It no longer does this for messages that
5196 are rejected by local_scan() or the DATA or non-SMTP ACLs.
5198 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
5199 have allow_domain_literals set, the ID did not get logged in the <= line.
5200 Domain literals are now always recognized in Message-ID: header lines.
5202 4. The first argument for a ${extract expansion item is the key name or field
5203 number. Leading and trailing spaces in this item were not being ignored,
5204 causing some misleading effects.
5206 5. When deliver_drop_privilege was set, single queue runner processes started
5207 manually (i.e. by the command "exim -q") or by the daemon (which uses the
5208 same command in the process it spins off) were not dropping privilege.
5210 6. When the daemon running as "exim" started a queue runner, it always
5211 re-executed Exim in the spun-off process. This is a waste of effort when
5212 deliver_drop_privilege is set. The new process now just calls the
5213 queue-runner function directly.
5219 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
5222 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
5223 ignored. Though the use of -oP was forcing the writing of a pid file, it
5224 was always written to the default place.
5226 3. If the message "no IP address found for host xxxx" is generated during
5227 incoming verification, it is now followed by identification of the incoming
5228 connection (so you can more easily find what provoked it).
5230 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
5232 5. Added some features to "harden" Exim a bit more against certain attacks:
5234 (a) There is now a build-time option called FIXED_NEVER_USERS that can
5235 be put in Local/Makefile. This is like the never_users runtime option,
5236 but it cannot be overridden. The default setting is "root".
5238 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
5239 prefix string with which any file named in a -C command line option
5242 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
5243 is retained for -C and -D only if the caller of Exim is root. Without
5244 it, the exim user may also use -C and -D and retain privilege.
5246 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
5247 command line option is disabled.
5249 6. Macro names set by the -D option must start with an upper case letter, just
5250 like macro names defined in the configuration file.
5252 7. Added "dereference=" facility to LDAP.
5254 8. Two instances of the typo "uknown" in the source files are fixed.
5256 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
5257 the Configure-Makefile script screwed up while processing it.
5259 10. Incorporated PCRE 4.4.
5261 11. The SMTP synchronization check was not operating right at the start of an
5262 SMTP session. For example, it could not catch a HELO sent before the client
5263 waited for the greeting. There is now a check for outstanding input at the
5264 point when the greeting is written. Because of the duplex, asynchronous
5265 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
5266 way, but not yet received, when the check is performed.
5268 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
5269 on TCP/IP sockets, because this apparently causes some broken clients to
5272 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
5273 unchanged) from the Cygwin maintainer.
5275 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
5276 was supported instead of testing for mbx. Effectively a typo.
5278 15. The spa authenticator server code was not checking that the input it
5279 received was valid base64.
5281 16. The debug output line for the "set" modifier in ACLs was not showing the
5282 name of the variable that was being set.
5284 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
5286 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
5287 lookup on its own. Something else was needed to trigger the lookup. For
5288 example, a match in host_lookup or the need for a host name in a host list.
5289 Now, if $sender_host_name is referenced and the host name has not yet been
5290 looked up, a lookup is performed. If the lookup fails, the variable remains
5291 empty, and $host_lookup_failed is set to "1".
5293 19. Added "eqi" as a case-independent comparison operator.
5295 20. The saslauthd authentication condition could segfault if neither service
5296 nor realm was specified.
5298 21. If an overflowing value such as "2048M" was set for message_size_limit, the
5299 error message that was logged was misleading, and incoming SMTP
5300 connections were dropped. The message is now more accurate, and temporary
5301 errors are given to SMTP connections.
5303 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
5304 (except RSET) with a 421 error, until QUIT is received. However, it was
5305 failing to send a response to QUIT.
5307 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
5308 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
5309 ACL is now run after the helo_try_verify_hosts code.
5311 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
5312 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
5313 case-independent, so other case variants are also recognized). Apparently
5314 some systems use these upper case variants.
5316 25. If more than two messages were waiting for the same host, and a transport
5317 filter was specified for the transport, Exim sent two messages over the
5318 same TCP/IP connection, and then failed with "socket operation on non-
5319 socket" when it tried to send the third.
5321 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
5323 27. The extern definition of crypt16() in expand.c was not being excluded when
5324 the OS had its own crypt16() function.
5326 28. Added bounce_return_body as a new option, and bounce_return_size_limit
5327 as a preferred synonym for return_size_limit, both as an option and as an
5330 29. Added LIBS=-liconv to OS/Makefile-OSF1.
5332 30. Changed the default configuration ACL to relax the local part checking rule
5333 for addresses that are not in any local domains. For these addresses,
5334 slashes and pipe symbols are allowed within local parts, but the sequence
5335 /../ is explicitly forbidden.
5337 31. SPA server authentication was not clearing the challenge buffer before
5340 32. log_message in a "warn" ACL statement was writing to the reject log as
5341 well as to the main log, which contradicts the documentation and doesn't
5342 seem right (because no rejection is happening). So I have stopped it.
5344 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
5345 However, I am unable to do any testing of this.
5347 34. Fixed an infelicity in the appendfile transport. When checking directories
5348 for a mailbox, to see if any needed to be created, it was accidentally
5349 using path names with one or more superfluous leading slashes; tracing
5350 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
5352 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
5353 contents are added to the log line that is written for every discarded
5354 recipient. (Previously a log_message setting was ignored.)
5356 36. The ${quote: operator now quotes the string if it is empty.
5358 37. The install script runs exim in order to find its version number. If for
5359 some reason other than non-existence or emptiness, which it checks, it
5360 could not run './exim', it was installing it with an empty version number,
5361 i.e. as "exim-". This error state is now caught, and the installation is
5364 38. An argument was missing from the function that creates an error message
5365 when Exim fails to connect to the socket for saslauthd authentication.
5366 This could cause Exim to crash, or give a corrupted message.
5368 39. Added isip, isip4, and isip6 to ${if conditions.
5370 40. The ACL variables $acl_xx are now saved with the message, and can be
5371 accessed later in routers, transports, and filters.
5373 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
5374 strings in the file are not string-expanded.
5376 42. If a MAIL command specified a SIZE value that was too large to fit into an
5377 int variable, the check against message_size_limit failed. Such values are
5378 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
5379 day this will have to be increased, but I don't think I want to be around
5380 when emails are that large.
5387 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
5388 iconv() is not standard in FreeBSD.
5390 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
5391 IPv6 enabled. The observed symptom was a segmentation fault on return from
5392 the function os_common_find_running_interfaces() in src/os.c.
5394 3. In the check_special_case() function in daemon.c I had used "errno" as an
5395 argument name, which causes warnings on some systems. This was basically a
5396 typo, since it was named "eno" in the comments!
5398 4. The code that waits for the clock to tick (at a resolution of some fraction
5399 of a second) so as to ensure message-id uniqueness was always waiting for
5400 at least one whole tick, when it could have waited for less. [This is
5401 almost certainly not relevant at current processor speeds, where it is
5402 unlikely to ever wait at all. But we try to future-proof.]
5404 5. The function that sleeps for a time interval that includes fractions of a
5405 second contained a race. It did not block SIGALRM between setting the
5406 timer, and suspending (a couple of lines later). If the interval was short
5407 and the sigsuspend() was delayed until after it had expired, the suspension
5408 never ended. On busy systems this could lead to processes getting stuck for
5411 6. Some uncommon configurations may cause a lookup to happen in a queue runner
5412 process, before it forks any delivery processes. The open lookup caching
5413 mechanism meant that the open file or database connection was passed into
5414 the delivery process. The problem was that delivery processes always tidy
5415 up cached lookup data. This could cause a problem for the next delivery
5416 process started by the queue runner, because the external queue runner
5417 process does not know about the closure. So the next delivery process
5418 still has data in the lookup cache. In the case of a file lookup, there was
5419 no problem because closing a file descriptor in a subprocess doesn't affect
5420 the parent. However, if the lookup was caching a connection to a database,
5421 the connection was closed, and the second delivery process was likely to
5422 see errors such as "PGSQL: query failed: server closed the connection
5423 unexpectedly". The problem has been fixed by closing all cached lookups
5424 in a queue runner before running a delivery process.
5426 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
5427 seem to have the "const" qualifier which it has on other OS. I've
5430 8. Change 4.21/2 was too strict. It is only if there are two authenticators
5431 *of the same type* (client or server) with the same public name that an
5432 error should be diagnosed.
5434 9. When Exim looked up a host name for an IP address, but failed to find the
5435 original IP address when looking up the host name (a safety check), it
5436 output the message "<ip address> does not match any IP for NULL", which was
5437 confusing, to say the least. The bug was that the host name should have
5438 appeared instead of "NULL".
5440 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
5441 user that is built into the binary, and the -C or -D options is used, root
5442 privilege is dropped before the configuration file is read. In addition,
5443 logging is switched to stderr instead of the normal log files. If the
5444 configuration then re-defines the Exim user, the unprivileged environment
5445 is probably not what is expected, so Exim logs a panic warning message (but
5448 However, if deliver_drop_privilege is set, the unprivileged state may well
5449 be exactly what is intended, so the warning has been cut out in that case,
5450 and Exim is allowed to try to write to its normal log files.
5456 1. smtp_return_error_details was not giving details for temporary sender
5457 or receiver verification errors.
5459 2. Diagnose a configuration error if two authenticators have the same public
5462 3. Exim used not to create the message log file for a message until the first
5463 delivery attempt. This could be confusing when incoming messages were held
5464 for policy or load reasons. The message log file is now created at the time
5465 the message is received, and an initial "Received" line is written to it.
5467 4. The automatically generated man page for command line options had a minor
5468 bug that caused no ill effects; however, a more serious problem was that
5469 the procedure for building the man page automatically didn't always
5470 operate. Consequently, release 4.20 contains an out-of-date version. This
5471 shouldn't happen again.
5473 5. When building Exim with embedded Perl support, the script that builds the
5474 Makefile was calling 'perl' to find its compile-time parameters, ignoring
5475 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
5477 6. The freeze_tell option was not being used for messages that were frozen on
5478 arrival, either by an ACL or by local_scan().
5480 7. Added the smtp_incomplete_transaction log selector.
5482 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
5483 was accepting AUTH without a new EHLO.
5485 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
5486 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
5489 10. Logging of TCP/IP connections (when configured) now happens in the main
5490 daemon process instead of the child process, so that the TCP/IP connection
5491 count is more accurate (but it can never be perfect).
5493 11. The use of "drop" in a nested ACL was not being handled correctly in the
5494 outer ACL. Now, if condition failure induced by the nested "drop" causes
5495 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
5496 or "require"), the connection is dropped.
5498 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
5499 that yield "discard" can now be used with an "accept" or a "discard" verb,
5500 but an error is generated for any others (because I can't see a useful way
5501 to define what should happen).
5503 13. When an ACL is read dynamically from a file (or anywhere else), the lines
5504 are now processed in the same way as lines in the Exim configuration file.
5505 In particular, continuation lines are supported.
5507 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
5509 15. Added -ti meaning -t -i.
5511 16. Check for letters, digits, hyphens, and dots in the names of dnslist
5512 domains, and warn by logging if others are found.
5514 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
5515 returned from GIFCONF when Exim is trying to find the list of interfaces on
5516 a host. The code in os.c has been modified to copy each ifreq to an aligned
5517 structure in all cases.
5519 Also, in some cases, the returned ifreq's were being copied to a 'struct
5520 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
5521 means the last couple of bytes of an IPv6 address could be chopped if the
5522 ifreq contained only a normal sockaddr (14 bytes storage).
5524 18. Named domain lists were not supported in the hosts_treat_as_local option.
5525 An entry such as +xxxx was not recognized, and was treated as a literal
5528 19. Ensure that header lines added by a DATA ACL are included in the reject log
5529 if the ACL subsequently rejects the message.
5531 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
5532 MD5 (which is deprecated).
5534 21. When testing a filter file using -bf, Exim was writing a message when it
5535 took the sender from a "From " line in the message, but it was not doing so
5536 when it took $return_path from a Return-Path: header line. It now does.
5538 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
5539 with a valid header line field name (a series of printing characters
5540 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
5542 23. Changed "disc" in the source to "disk" to conform to the documentation and
5543 the book and for uniformity.
5545 24. Ignore Sendmail's -Ooption=value command line item.
5547 25. When execve() failed while trying to run a command in a pipe transport,
5548 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
5549 could be confused with a return value of 69 from the command itself. This
5550 has been changed to 127, the value the shell returns if it is asked to run
5551 a non-existent command. The wording for the related log line suggests a
5552 non-existent command as the problem.
5554 26. If received_header_text expands to an empty string, do not add a Received:
5555 header line to the message. (Well, it adds a token one on the spool, but
5556 marks it "old" so that it doesn't get used or transmitted.)
5558 27. Installed eximstats 1.28 (addition of -nt option).
5560 28. There was no check for failure on the call to getsockname() in the daemon
5561 code. This can fail if there is a shortage of resources on the system, with
5562 ENOMEM, for example. A temporary error is now given on failure.
5564 29. Contrary to the C standard, it seems that in some environments, the
5565 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
5566 program. Exim now does this explicitly; it affects the formatting of
5567 timestamps using strftime().
5569 30. If exiqsumm was given junk data, it threw up some uninitialized variable
5570 complaints. I've now initialized all the variables, to avoid this.
5572 32. Header lines added by a system filter were not being "seen" during
5573 transport-time rewrites.
5575 33. The info_callback() function passed to OpenSSL is set up with type void
5576 (*)(SSL *, int, int), as described somewhere. However, when calling the
5577 function (actually a macro) that sets it up, the type void(*)() is
5578 expected. I've put in a cast to prevent warnings from picky compilers.
5580 34. If a DNS black list lookup found a CNAME record, but there were no A
5581 records associated with the domain it pointed at, Exim crashed.
5583 35. If a DNS black list lookup returned more than one A record, Exim ignored
5584 all but the first. It now scans all returned addresses if a particular IP
5585 value is being sought. In this situation, the contents of the
5586 $dnslist_value variable are a list of all the addresses, separated by a
5589 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
5590 to accept a host name and all its aliases if the forward lookup for any of
5591 them yielded the IP address of the incoming connection. Now it accepts only
5592 those names whose forward lookup yields the correct IP address. Any other
5593 names are discarded. This closes a loophole whereby a rogue DNS
5594 administrator could create reverse DNS records to break through a
5595 wildcarded host restriction in an ACL.
5597 37. If a user filter or a system filter that ran in a subprocess used any of
5598 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
5599 the wrong values were passed to the pipe command ($thisaddress had the
5600 value of $0, $0 had the value of $1, etc). This bug was introduced by
5601 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
5603 38. Improved the line breaking for long SMTP error messages from ACLs.
5604 Previously, if there was no break point between 40 and 75 characters, Exim
5605 left the rest of the message alone. Two changes have been made: (a) I've
5606 reduced the minimum length to 35 characters; (b) if it can't find a break
5607 point between 35 and 75 characters, it looks ahead and uses the first one
5608 that it finds. This may give the occasional overlong line, but at least the
5609 remaining text gets split now.
5611 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
5612 file descriptors might be low, and that setting 1000 would always raise it.
5613 It turns out that in some environments, the limit is already over 1000 and
5614 that lowering it causes trouble. So now Exim takes care not to decrease it.
5616 40. When delivering a message, the value of $return_path is set to $sender_
5617 address at the start of routing (routers may change the value). By an
5618 oversight, this default was not being set up when an address was tested by
5619 -bt or -bv, which affected the outcome if any router or filter referred to
5622 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
5623 log only when "message" or "log_message" are set. However, if one of the
5624 conditions was an address verification, or a call to a nested ACL, the
5625 messages generated by the underlying test were being passed through. This
5626 no longer happens. The underlying message is available in $acl_verify_
5627 message for both "message" and "log_message" expansions, so it can be
5628 passed through if needed.
5630 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
5631 new expansion $bh_ to give the encoded byte string without charset
5632 translation. Translation happens only if iconv() is available; HAVE_ICONV
5633 indicates this at build time. HEADERS_CHARSET gives the charset to
5634 translate to; headers_charset can change it in the configuration, and
5635 "headers charset" can change it in an individual filter file.
5637 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
5638 that creates RFC 2047 encoded "words" labels them as that charset instead
5639 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
5640 expansion operator; (ii) when Exim creates a From: line for a local
5641 message; (iii) when a header line is rewritten to include a "phrase" part.
5643 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
5644 buggy, causing it to skip the first lines of messages whose message ID
5645 ended in 'D'. This would not have bitten before Exim release 4.14, because
5646 message IDs were unlikely to end in 'D' before then. The effect was to have
5647 incorrect size information for certain domains.
5649 45. #include "config.h" was missing at the start of the crypt16.c module. This
5650 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
5653 46. If there was a timeout during a "random" callout check, Exim treated it as
5654 a failure of the random address, and carried on sending RSET and the real
5655 address. If the delay was just some slowness somewhere, the response to the
5656 original RCPT would be taken as a response to RSET and so on, causing
5657 mayhem of various kinds.
5659 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
5660 when I implemented it. It didn't allow for the fact that some option values
5661 may legitimatetly be negative (e.g. size_addition), and it didn't even do
5662 the right test for positive values.
5664 48. Domain names in DNS records are case-independent. Exim always looks them up
5665 in lower case. Some resolvers return domain names in exactly the case they
5666 appear in the zone file, that is, they may contain uppercase letters. Not
5667 all resolvers do this - some return always lower case. Exim was treating a
5668 change of case by a resolver as a change of domain, similar to a widening
5669 of a domain abbreviation. This triggered its re-routing code and so it was
5670 trying to route what was effectively the same domain again. This normally
5671 caused routing to fail (because the router wouldn't handle the domain
5672 twice). Now Exim checks for this case specially, and just changes the
5673 casing of the domain that it ultimately uses when it transmits the message
5676 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
5679 50. If a filter generated a file delivery with a non-absolute name (possible if
5680 no home directory exists for the router), the forbid_file option was not
5683 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
5684 the existing equality matching.
5686 52. Exim was using ints instead of ino_t variables in some places where it was
5687 dealing with inode numbers.
5689 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
5690 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
5691 called TMPDIR, and if it finds it is different, it changes its value.
5693 54. The smtp_printf() function is now made available to local_scan() so
5694 additional output lines can be written before returning. There is also an
5695 smtp_fflush() function to enable the detection of a dropped connection.
5696 The variables smtp_input and smtp_batched_input are exported to
5699 55. Changed the default runtime configuration: the message "Unknown user"
5700 has been removed from the ACL, and instead placed on the localuser router,
5701 using the cannot_route_message feature. This means that any verification
5702 failures that generate their own messages won't get overridden. Similarly,
5703 the "Unrouteable address" message that was in the ACL for unverifiable
5704 relay addresses has also been removed.
5706 56. Added hosts_avoid_esmtp to the smtp transport.
5708 57. The exicyclog script was not checking for the esoteric option
5709 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
5710 will work only if exicyclog is run under the appropriate euid.
5712 58. Following a discussion on the list, the rules by which Exim recognises line
5713 endings on incoming messages have been changed. The -dropcr and drop_cr
5714 options are now no-ops, retained only for backwards compatibility. The
5715 following line terminators are recognized: LF CRLF CR. However, special
5716 processing applies to CR:
5718 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
5719 nor a local message in the state where . is a terminator.
5721 (ii) If a bare CR is encountered in a header line, an extra space is added
5722 after the line terminator so as not to end the header. The reasoning
5723 behind this is that bare CRs in header lines are most likely either
5724 to be mistakes, or people trying to play silly games.
5726 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
5727 was being incorrectly given as 18 bytes larger than it should have been.
5728 This is a VOB (very old bug).
5730 60. This may never have affected anything current, but just in case it has:
5731 When the local host is found other than at the start of a list of hosts,
5732 the local host, those with the same MX, and any that follow, are discarded.
5733 When the list in question was part of a longer list of hosts, the following
5734 hosts (not currently being processed) were also being discarded. This no
5735 longer happens. I'm not sure if this situation could ever has previously
5738 61. Added the "/MX" feature to lists of hosts in the manualroute and query
5741 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
5742 header. This is something that is recommended in a new Internet Draft, and
5743 is something that is documented as being done by Sendmail. There are two
5744 possible values. For messages generated by the autoreply transport, Exim
5747 Auto-Submitted: auto-replied
5749 whereas for all other generated messages (e.g. bounces) it adds
5751 Auto-Submitted: auto-generated
5753 63. The "personal" condition in filters now includes a test for the
5754 Auto-Submitted: header. If it contains the string "auto-" the message it
5755 not considered personal.
5757 64. Added rcpt_include_affixes as a generic transport option.
5759 65. Added queue_only_override (default true).
5761 66. Added the syslog_duplication option.
5763 67. If what should have been the first header line of a message consisted of
5764 a space followed by a colon, Exim was mis-interpreting it as a header line.
5765 It isn't of course - it is syntactically invalid and should therefore be
5766 treated as the start of the message body. The misbehaviour could have
5767 caused a number of strange effects, including loss of data in subsequent
5768 header lines, and spool format errors.
5770 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
5771 client host had authenticated. This control can now be exercised by an ACL
5772 for more flexibility.
5774 69. By default, callouts do not happen when testing with -bh. There is now a
5775 variant, -bhc, which does actually run the callout code, including
5776 consulting and updating the callout cache.
5778 70. Added support for saslauthd authentication, courtesy of Alexander
5781 71. If statvfs() failed on the spool or log directories while checking their
5782 size for availability, Exim confusingly gave the error "space shortage".
5783 Furthermore, in debugging mode it crashed with a floating point exception.
5784 These checks are done if check_{spool,log}_{space,inodes} are set, and when
5785 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
5786 serious problem, Exim now writes to the main and panic logs when this
5787 happens, with details of the failure. It then refuses to accept the
5788 incoming message, giving the message "spool directory problem" or "log
5789 directory problem" with a 421 code for SMTP messages.
5791 72. When Exim is about to re-exec itself, it ensures that the file descriptors
5792 0, 1, and 2 exist, because some OS complain for execs without them (see
5793 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
5794 descriptors. However, the code omitted to check that the open succeeded,
5795 causing mysterious errors if for some reason the permissions on /dev/null
5796 got screwed. Now Exim writes a message to the main and panic logs, and
5797 bombs out if it can't open /dev/null.
5799 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
5800 interact so that it is all more flexible. It is supposed to remain
5801 backwards compatible. Also added extra_local_interfaces.
5803 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
5804 to bomb out with an assertion failure - to the client this appears as a
5805 connection drop. This problem occurs in the part of the code that was taken
5806 from the Samba project. Fortunately, the assertion is in a very simple
5807 function, so I have fixed this by reproducing the function inline in the
5808 one place where it is called, and arranging for authentication to fail
5809 instead of killing the process with assert().
5811 75. The SPA client code was not working when the server requested OEM rather
5812 than Unicode encoding.
5814 76. Added code to make require_files with a specific uid setting more usable in
5815 the case where statting the file as root fails - usually a non-root-mounted
5816 NFS file system. When this happens and the failure is EACCES, Exim now
5817 forks a subprocess and does the per-uid checking as the relevant uid.
5819 77. Added process_log_path.
5821 78. If log_file_path was not explicitly set, a setting of check_log_space or
5822 check_log_inodes was ignored.
5824 79. If a space check for the spool or log partitions fails, the incident is now
5825 logged. Of course, in the latter case the data may get lost...
5827 80. Added the %p formatting code to string_format() so that it can be used to
5828 print addresses in debug_print(). Adjusted all the address printing in the
5829 debugging in store.c to use %p rather than %d.
5831 81. There was a concern that a line of code in smtp_in.c could overflow a
5832 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
5833 initially expressed, the concern was not well-founded, because trailing
5834 spaces are removed early. However, if the trailing spaces were followed by
5835 a NULL, they did not get removed, so the overflow was possible. Two fixes
5838 (a) I re-wrote the offending code in a cleaner fashion.
5839 (b) If an incoming SMTP command contains a NULL character, it is rejected
5842 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
5843 runs initgroups(), so that if the Exim user is in any additional groups,
5844 they will be used during message reception.
5850 The change log for 4.20 and earlier releases has been archived.
projects / exim.git / blob